diff --git "a/checkpoint-35796/trainer_state.json" "b/checkpoint-35796/trainer_state.json" new file mode 100644--- /dev/null +++ "b/checkpoint-35796/trainer_state.json" @@ -0,0 +1,251315 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.999972064698159, + "eval_steps": 500, + "global_step": 35796, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 5.587060368187278e-05, + "grad_norm": Infinity, + "learning_rate": 0.0, + "loss": 14.0294, + "step": 1 + }, + { + "epoch": 0.00011174120736374556, + "grad_norm": 31.15971565246582, + "learning_rate": 1e-05, + "loss": 15.0626, + "step": 2 + }, + { + "epoch": 0.00016761181104561835, + "grad_norm": 30.064119338989258, + "learning_rate": 2e-05, + "loss": 13.9448, + "step": 3 + }, + { + "epoch": 0.00022348241472749112, + "grad_norm": 28.274494171142578, + "learning_rate": 3e-05, + "loss": 13.4444, + "step": 4 + }, + { + "epoch": 0.0002793530184093639, + "grad_norm": 32.294410705566406, + "learning_rate": 4e-05, + "loss": 14.5708, + "step": 5 + }, + { + "epoch": 0.0003352236220912367, + "grad_norm": 31.837663650512695, + "learning_rate": 5e-05, + "loss": 13.5372, + "step": 6 + }, + { + "epoch": 0.0003910942257731095, + "grad_norm": 29.03823471069336, + "learning_rate": 6e-05, + "loss": 13.367, + "step": 7 + }, + { + "epoch": 0.00044696482945498224, + "grad_norm": 31.434101104736328, + "learning_rate": 7.000000000000001e-05, + "loss": 13.6411, + "step": 8 + }, + { + "epoch": 0.000502835433136855, + "grad_norm": 28.008895874023438, + "learning_rate": 8e-05, + "loss": 13.3109, + "step": 9 + }, + { + "epoch": 0.0005587060368187278, + "grad_norm": 33.19596481323242, + "learning_rate": 8.999999999999999e-05, + "loss": 13.2189, + "step": 10 + }, + { + "epoch": 0.0006145766405006006, + "grad_norm": 30.26412010192871, + "learning_rate": 0.0001, + "loss": 12.9284, + "step": 11 + }, + { + "epoch": 0.0006704472441824734, + "grad_norm": 30.577455520629883, + "learning_rate": 0.00011, + "loss": 12.1541, + "step": 12 + }, + { + "epoch": 0.0007263178478643462, + "grad_norm": 26.633588790893555, + "learning_rate": 0.00012, + "loss": 11.0929, + "step": 13 + }, + { + "epoch": 0.000782188451546219, + "grad_norm": 31.039457321166992, + "learning_rate": 0.00013000000000000002, + "loss": 11.9551, + "step": 14 + }, + { + "epoch": 0.0008380590552280918, + "grad_norm": 37.16322326660156, + "learning_rate": 0.00014000000000000001, + "loss": 12.163, + "step": 15 + }, + { + "epoch": 0.0008939296589099645, + "grad_norm": 30.865161895751953, + "learning_rate": 0.00015, + "loss": 11.1355, + "step": 16 + }, + { + "epoch": 0.0009498002625918373, + "grad_norm": 34.7225341796875, + "learning_rate": 0.00016, + "loss": 11.1434, + "step": 17 + }, + { + "epoch": 0.00100567086627371, + "grad_norm": Infinity, + "learning_rate": 0.00016, + "loss": 10.6365, + "step": 18 + }, + { + "epoch": 0.001061541469955583, + "grad_norm": 37.47252655029297, + "learning_rate": 0.00017, + "loss": 10.9009, + "step": 19 + }, + { + "epoch": 0.0011174120736374556, + "grad_norm": Infinity, + "learning_rate": 0.00017, + "loss": 9.5812, + "step": 20 + }, + { + "epoch": 0.0011732826773193283, + "grad_norm": 38.02201461791992, + "learning_rate": 0.00017999999999999998, + "loss": 10.5641, + "step": 21 + }, + { + "epoch": 0.0012291532810012012, + "grad_norm": 37.043426513671875, + "learning_rate": 0.00019, + "loss": 9.6687, + "step": 22 + }, + { + "epoch": 0.001285023884683074, + "grad_norm": 36.53663635253906, + "learning_rate": 0.0002, + "loss": 8.8688, + "step": 23 + }, + { + "epoch": 0.0013408944883649468, + "grad_norm": 34.01047897338867, + "learning_rate": 0.00021, + "loss": 8.236, + "step": 24 + }, + { + "epoch": 0.0013967650920468195, + "grad_norm": 40.686641693115234, + "learning_rate": 0.00022, + "loss": 8.5111, + "step": 25 + }, + { + "epoch": 0.0014526356957286924, + "grad_norm": 36.88578796386719, + "learning_rate": 0.00023, + "loss": 7.2981, + "step": 26 + }, + { + "epoch": 0.001508506299410565, + "grad_norm": 40.65380096435547, + "learning_rate": 0.00024, + "loss": 7.5401, + "step": 27 + }, + { + "epoch": 0.001564376903092438, + "grad_norm": 39.22256088256836, + "learning_rate": 0.00025, + "loss": 7.0077, + "step": 28 + }, + { + "epoch": 0.0016202475067743107, + "grad_norm": 34.54293441772461, + "learning_rate": 0.00026000000000000003, + "loss": 6.1475, + "step": 29 + }, + { + "epoch": 0.0016761181104561836, + "grad_norm": 32.055015563964844, + "learning_rate": 0.00027, + "loss": 5.4821, + "step": 30 + }, + { + "epoch": 0.0017319887141380563, + "grad_norm": 33.87163162231445, + "learning_rate": 0.00028000000000000003, + "loss": 5.4771, + "step": 31 + }, + { + "epoch": 0.001787859317819929, + "grad_norm": 28.76503562927246, + "learning_rate": 0.00029, + "loss": 4.6728, + "step": 32 + }, + { + "epoch": 0.0018437299215018018, + "grad_norm": 22.27045249938965, + "learning_rate": 0.0003, + "loss": 4.1632, + "step": 33 + }, + { + "epoch": 0.0018996005251836745, + "grad_norm": 19.233924865722656, + "learning_rate": 0.00031, + "loss": 3.8299, + "step": 34 + }, + { + "epoch": 0.0019554711288655474, + "grad_norm": 14.80162525177002, + "learning_rate": 0.00032, + "loss": 3.451, + "step": 35 + }, + { + "epoch": 0.00201134173254742, + "grad_norm": 8.414961814880371, + "learning_rate": 0.00033, + "loss": 3.1982, + "step": 36 + }, + { + "epoch": 0.002067212336229293, + "grad_norm": 4.651137351989746, + "learning_rate": 0.00034, + "loss": 3.1148, + "step": 37 + }, + { + "epoch": 0.002123082939911166, + "grad_norm": 2.310208559036255, + "learning_rate": 0.00035, + "loss": 2.9578, + "step": 38 + }, + { + "epoch": 0.0021789535435930386, + "grad_norm": 4.395336151123047, + "learning_rate": 0.00035999999999999997, + "loss": 2.9168, + "step": 39 + }, + { + "epoch": 0.0022348241472749113, + "grad_norm": 5.045258522033691, + "learning_rate": 0.00037, + "loss": 2.9039, + "step": 40 + }, + { + "epoch": 0.002290694750956784, + "grad_norm": 7.06749963760376, + "learning_rate": 0.00038, + "loss": 2.8828, + "step": 41 + }, + { + "epoch": 0.0023465653546386567, + "grad_norm": 6.344650745391846, + "learning_rate": 0.00039000000000000005, + "loss": 2.7999, + "step": 42 + }, + { + "epoch": 0.0024024359583205298, + "grad_norm": 6.877009868621826, + "learning_rate": 0.0004, + "loss": 2.7749, + "step": 43 + }, + { + "epoch": 0.0024583065620024025, + "grad_norm": 4.366193771362305, + "learning_rate": 0.00041, + "loss": 2.7079, + "step": 44 + }, + { + "epoch": 0.002514177165684275, + "grad_norm": 3.483126163482666, + "learning_rate": 0.00042, + "loss": 2.6124, + "step": 45 + }, + { + "epoch": 0.002570047769366148, + "grad_norm": 2.238349199295044, + "learning_rate": 0.00043, + "loss": 2.5352, + "step": 46 + }, + { + "epoch": 0.002625918373048021, + "grad_norm": 2.941171884536743, + "learning_rate": 0.00044, + "loss": 2.4177, + "step": 47 + }, + { + "epoch": 0.0026817889767298936, + "grad_norm": 3.6372733116149902, + "learning_rate": 0.00045000000000000004, + "loss": 2.3239, + "step": 48 + }, + { + "epoch": 0.0027376595804117663, + "grad_norm": 5.599921703338623, + "learning_rate": 0.00046, + "loss": 2.3658, + "step": 49 + }, + { + "epoch": 0.002793530184093639, + "grad_norm": 4.416717052459717, + "learning_rate": 0.00047, + "loss": 2.1377, + "step": 50 + }, + { + "epoch": 0.002849400787775512, + "grad_norm": 3.5336461067199707, + "learning_rate": 0.00048, + "loss": 2.1255, + "step": 51 + }, + { + "epoch": 0.002905271391457385, + "grad_norm": 2.7555551528930664, + "learning_rate": 0.00049, + "loss": 1.8757, + "step": 52 + }, + { + "epoch": 0.0029611419951392575, + "grad_norm": 2.257885694503784, + "learning_rate": 0.0005, + "loss": 1.8612, + "step": 53 + }, + { + "epoch": 0.00301701259882113, + "grad_norm": 2.7476062774658203, + "learning_rate": 0.00051, + "loss": 1.631, + "step": 54 + }, + { + "epoch": 0.003072883202503003, + "grad_norm": 2.4200422763824463, + "learning_rate": 0.0005200000000000001, + "loss": 1.4592, + "step": 55 + }, + { + "epoch": 0.003128753806184876, + "grad_norm": 2.649332284927368, + "learning_rate": 0.0005300000000000001, + "loss": 1.4069, + "step": 56 + }, + { + "epoch": 0.0031846244098667487, + "grad_norm": 1.8735687732696533, + "learning_rate": 0.00054, + "loss": 1.1604, + "step": 57 + }, + { + "epoch": 0.0032404950135486213, + "grad_norm": 1.4942734241485596, + "learning_rate": 0.00055, + "loss": 1.1298, + "step": 58 + }, + { + "epoch": 0.003296365617230494, + "grad_norm": 2.011185884475708, + "learning_rate": 0.0005600000000000001, + "loss": 1.0308, + "step": 59 + }, + { + "epoch": 0.003352236220912367, + "grad_norm": 1.2874828577041626, + "learning_rate": 0.00057, + "loss": 1.0159, + "step": 60 + }, + { + "epoch": 0.00340810682459424, + "grad_norm": 1.1485902070999146, + "learning_rate": 0.00058, + "loss": 0.9802, + "step": 61 + }, + { + "epoch": 0.0034639774282761125, + "grad_norm": 0.8173692226409912, + "learning_rate": 0.00059, + "loss": 0.8533, + "step": 62 + }, + { + "epoch": 0.003519848031957985, + "grad_norm": 0.9755550622940063, + "learning_rate": 0.0006, + "loss": 0.8021, + "step": 63 + }, + { + "epoch": 0.003575718635639858, + "grad_norm": 0.8761739134788513, + "learning_rate": 0.00061, + "loss": 0.7665, + "step": 64 + }, + { + "epoch": 0.003631589239321731, + "grad_norm": 1.611499309539795, + "learning_rate": 0.00062, + "loss": 0.8752, + "step": 65 + }, + { + "epoch": 0.0036874598430036037, + "grad_norm": 1.1592234373092651, + "learning_rate": 0.00063, + "loss": 0.7678, + "step": 66 + }, + { + "epoch": 0.0037433304466854764, + "grad_norm": 0.8209341168403625, + "learning_rate": 0.00064, + "loss": 0.6885, + "step": 67 + }, + { + "epoch": 0.003799201050367349, + "grad_norm": 0.7120041251182556, + "learning_rate": 0.0006500000000000001, + "loss": 0.6465, + "step": 68 + }, + { + "epoch": 0.003855071654049222, + "grad_norm": 0.8481773734092712, + "learning_rate": 0.00066, + "loss": 0.5457, + "step": 69 + }, + { + "epoch": 0.003910942257731095, + "grad_norm": 1.6068261861801147, + "learning_rate": 0.00067, + "loss": 0.823, + "step": 70 + }, + { + "epoch": 0.003966812861412968, + "grad_norm": 0.8877984881401062, + "learning_rate": 0.00068, + "loss": 0.8707, + "step": 71 + }, + { + "epoch": 0.00402268346509484, + "grad_norm": 0.7529939413070679, + "learning_rate": 0.00069, + "loss": 0.6506, + "step": 72 + }, + { + "epoch": 0.004078554068776713, + "grad_norm": 0.7702661156654358, + "learning_rate": 0.0007, + "loss": 0.8459, + "step": 73 + }, + { + "epoch": 0.004134424672458586, + "grad_norm": 1.353736400604248, + "learning_rate": 0.00071, + "loss": 0.6858, + "step": 74 + }, + { + "epoch": 0.004190295276140459, + "grad_norm": 1.048360824584961, + "learning_rate": 0.0007199999999999999, + "loss": 0.7857, + "step": 75 + }, + { + "epoch": 0.004246165879822332, + "grad_norm": 4.6819634437561035, + "learning_rate": 0.00073, + "loss": 0.6741, + "step": 76 + }, + { + "epoch": 0.004302036483504204, + "grad_norm": 0.6726879477500916, + "learning_rate": 0.00074, + "loss": 0.5706, + "step": 77 + }, + { + "epoch": 0.004357907087186077, + "grad_norm": 1.2117947340011597, + "learning_rate": 0.00075, + "loss": 0.6607, + "step": 78 + }, + { + "epoch": 0.0044137776908679495, + "grad_norm": 1.0861023664474487, + "learning_rate": 0.00076, + "loss": 0.5754, + "step": 79 + }, + { + "epoch": 0.004469648294549823, + "grad_norm": 1.0452256202697754, + "learning_rate": 0.0007700000000000001, + "loss": 0.6297, + "step": 80 + }, + { + "epoch": 0.004525518898231696, + "grad_norm": 0.8665221333503723, + "learning_rate": 0.0007800000000000001, + "loss": 0.624, + "step": 81 + }, + { + "epoch": 0.004581389501913568, + "grad_norm": 0.6246809363365173, + "learning_rate": 0.00079, + "loss": 0.5397, + "step": 82 + }, + { + "epoch": 0.004637260105595441, + "grad_norm": 0.8085268139839172, + "learning_rate": 0.0008, + "loss": 0.5705, + "step": 83 + }, + { + "epoch": 0.004693130709277313, + "grad_norm": 0.7800960540771484, + "learning_rate": 0.0008100000000000001, + "loss": 0.6283, + "step": 84 + }, + { + "epoch": 0.0047490013129591864, + "grad_norm": 0.9813174605369568, + "learning_rate": 0.00082, + "loss": 0.5161, + "step": 85 + }, + { + "epoch": 0.0048048719166410596, + "grad_norm": 0.6278838515281677, + "learning_rate": 0.00083, + "loss": 0.5068, + "step": 86 + }, + { + "epoch": 0.004860742520322932, + "grad_norm": 1.2159746885299683, + "learning_rate": 0.00084, + "loss": 0.6959, + "step": 87 + }, + { + "epoch": 0.004916613124004805, + "grad_norm": 0.6308979392051697, + "learning_rate": 0.00085, + "loss": 0.4518, + "step": 88 + }, + { + "epoch": 0.004972483727686678, + "grad_norm": 1.2274696826934814, + "learning_rate": 0.00086, + "loss": 0.7073, + "step": 89 + }, + { + "epoch": 0.00502835433136855, + "grad_norm": 0.7183673977851868, + "learning_rate": 0.00087, + "loss": 0.7146, + "step": 90 + }, + { + "epoch": 0.005084224935050423, + "grad_norm": 7.257446765899658, + "learning_rate": 0.00088, + "loss": 0.7734, + "step": 91 + }, + { + "epoch": 0.005140095538732296, + "grad_norm": 0.8041048645973206, + "learning_rate": 0.0008900000000000001, + "loss": 0.5928, + "step": 92 + }, + { + "epoch": 0.005195966142414169, + "grad_norm": 0.8922842741012573, + "learning_rate": 0.0009000000000000001, + "loss": 0.563, + "step": 93 + }, + { + "epoch": 0.005251836746096042, + "grad_norm": 0.8066903948783875, + "learning_rate": 0.00091, + "loss": 0.6469, + "step": 94 + }, + { + "epoch": 0.005307707349777914, + "grad_norm": 0.7524223327636719, + "learning_rate": 0.00092, + "loss": 0.5798, + "step": 95 + }, + { + "epoch": 0.005363577953459787, + "grad_norm": 1.0560377836227417, + "learning_rate": 0.00093, + "loss": 0.6377, + "step": 96 + }, + { + "epoch": 0.0054194485571416595, + "grad_norm": 0.9373245239257812, + "learning_rate": 0.00094, + "loss": 0.6156, + "step": 97 + }, + { + "epoch": 0.005475319160823533, + "grad_norm": 1.429648518562317, + "learning_rate": 0.00095, + "loss": 0.7202, + "step": 98 + }, + { + "epoch": 0.005531189764505406, + "grad_norm": 1.346039891242981, + "learning_rate": 0.00096, + "loss": 0.745, + "step": 99 + }, + { + "epoch": 0.005587060368187278, + "grad_norm": 0.6590069532394409, + "learning_rate": 0.0009699999999999999, + "loss": 0.5642, + "step": 100 + }, + { + "epoch": 0.005642930971869151, + "grad_norm": 0.7339789867401123, + "learning_rate": 0.00098, + "loss": 0.5656, + "step": 101 + }, + { + "epoch": 0.005698801575551024, + "grad_norm": 0.7111060619354248, + "learning_rate": 0.00099, + "loss": 0.6222, + "step": 102 + }, + { + "epoch": 0.0057546721792328965, + "grad_norm": 0.8675293326377869, + "learning_rate": 0.001, + "loss": 0.493, + "step": 103 + }, + { + "epoch": 0.00581054278291477, + "grad_norm": 1.6767423152923584, + "learning_rate": 0.0009999719856566562, + "loss": 0.6148, + "step": 104 + }, + { + "epoch": 0.005866413386596642, + "grad_norm": 1.4586304426193237, + "learning_rate": 0.0009999439713133124, + "loss": 0.6627, + "step": 105 + }, + { + "epoch": 0.005922283990278515, + "grad_norm": 1.1649280786514282, + "learning_rate": 0.0009999159569699686, + "loss": 0.6704, + "step": 106 + }, + { + "epoch": 0.005978154593960388, + "grad_norm": 0.6497063040733337, + "learning_rate": 0.0009998879426266248, + "loss": 0.551, + "step": 107 + }, + { + "epoch": 0.00603402519764226, + "grad_norm": 0.7460550665855408, + "learning_rate": 0.000999859928283281, + "loss": 0.5696, + "step": 108 + }, + { + "epoch": 0.0060898958013241335, + "grad_norm": 1.0137670040130615, + "learning_rate": 0.0009998319139399373, + "loss": 0.5053, + "step": 109 + }, + { + "epoch": 0.006145766405006006, + "grad_norm": 1.2949976921081543, + "learning_rate": 0.0009998038995965935, + "loss": 0.7657, + "step": 110 + }, + { + "epoch": 0.006201637008687879, + "grad_norm": 1.4768468141555786, + "learning_rate": 0.0009997758852532497, + "loss": 0.5527, + "step": 111 + }, + { + "epoch": 0.006257507612369752, + "grad_norm": 2.4484596252441406, + "learning_rate": 0.0009997478709099059, + "loss": 0.556, + "step": 112 + }, + { + "epoch": 0.006313378216051624, + "grad_norm": 0.7609063386917114, + "learning_rate": 0.000999719856566562, + "loss": 0.5477, + "step": 113 + }, + { + "epoch": 0.006369248819733497, + "grad_norm": 0.9510129690170288, + "learning_rate": 0.0009996918422232183, + "loss": 0.5299, + "step": 114 + }, + { + "epoch": 0.00642511942341537, + "grad_norm": 0.8251046538352966, + "learning_rate": 0.0009996638278798745, + "loss": 0.6567, + "step": 115 + }, + { + "epoch": 0.006480990027097243, + "grad_norm": 0.8193126320838928, + "learning_rate": 0.0009996358135365307, + "loss": 0.5142, + "step": 116 + }, + { + "epoch": 0.006536860630779116, + "grad_norm": 1.595676064491272, + "learning_rate": 0.000999607799193187, + "loss": 0.5903, + "step": 117 + }, + { + "epoch": 0.006592731234460988, + "grad_norm": 0.8393203616142273, + "learning_rate": 0.000999579784849843, + "loss": 0.5849, + "step": 118 + }, + { + "epoch": 0.006648601838142861, + "grad_norm": 2.989581823348999, + "learning_rate": 0.0009995517705064993, + "loss": 0.5863, + "step": 119 + }, + { + "epoch": 0.006704472441824734, + "grad_norm": 0.9709064960479736, + "learning_rate": 0.0009995237561631555, + "loss": 0.6283, + "step": 120 + }, + { + "epoch": 0.0067603430455066066, + "grad_norm": 0.8172082304954529, + "learning_rate": 0.0009994957418198117, + "loss": 0.6997, + "step": 121 + }, + { + "epoch": 0.00681621364918848, + "grad_norm": 1.0192509889602661, + "learning_rate": 0.000999467727476468, + "loss": 0.7992, + "step": 122 + }, + { + "epoch": 0.006872084252870352, + "grad_norm": 1.2695884704589844, + "learning_rate": 0.0009994397131331241, + "loss": 0.4969, + "step": 123 + }, + { + "epoch": 0.006927954856552225, + "grad_norm": 0.6757664084434509, + "learning_rate": 0.0009994116987897803, + "loss": 0.67, + "step": 124 + }, + { + "epoch": 0.006983825460234098, + "grad_norm": 0.6865705847740173, + "learning_rate": 0.0009993836844464365, + "loss": 0.6169, + "step": 125 + }, + { + "epoch": 0.00703969606391597, + "grad_norm": 1.3138227462768555, + "learning_rate": 0.0009993556701030928, + "loss": 0.7572, + "step": 126 + }, + { + "epoch": 0.0070955666675978435, + "grad_norm": 2.8320202827453613, + "learning_rate": 0.000999327655759749, + "loss": 0.695, + "step": 127 + }, + { + "epoch": 0.007151437271279716, + "grad_norm": 1.0458436012268066, + "learning_rate": 0.0009992996414164052, + "loss": 0.5013, + "step": 128 + }, + { + "epoch": 0.007207307874961589, + "grad_norm": 0.6664717793464661, + "learning_rate": 0.0009992716270730614, + "loss": 0.6099, + "step": 129 + }, + { + "epoch": 0.007263178478643462, + "grad_norm": 0.5584750771522522, + "learning_rate": 0.0009992436127297176, + "loss": 0.6457, + "step": 130 + }, + { + "epoch": 0.007319049082325334, + "grad_norm": 0.976669192314148, + "learning_rate": 0.0009992155983863738, + "loss": 0.4133, + "step": 131 + }, + { + "epoch": 0.007374919686007207, + "grad_norm": 0.7198210954666138, + "learning_rate": 0.00099918758404303, + "loss": 0.6265, + "step": 132 + }, + { + "epoch": 0.0074307902896890805, + "grad_norm": 1.0274654626846313, + "learning_rate": 0.0009991595696996862, + "loss": 0.6013, + "step": 133 + }, + { + "epoch": 0.007486660893370953, + "grad_norm": 0.6041138768196106, + "learning_rate": 0.0009991315553563424, + "loss": 0.5762, + "step": 134 + }, + { + "epoch": 0.007542531497052826, + "grad_norm": 0.6494506001472473, + "learning_rate": 0.0009991035410129988, + "loss": 0.4694, + "step": 135 + }, + { + "epoch": 0.007598402100734698, + "grad_norm": 1.2107301950454712, + "learning_rate": 0.0009990755266696548, + "loss": 0.5348, + "step": 136 + }, + { + "epoch": 0.007654272704416571, + "grad_norm": 1.0931322574615479, + "learning_rate": 0.000999047512326311, + "loss": 0.6983, + "step": 137 + }, + { + "epoch": 0.007710143308098444, + "grad_norm": 4.305722713470459, + "learning_rate": 0.0009990194979829672, + "loss": 0.4756, + "step": 138 + }, + { + "epoch": 0.007766013911780317, + "grad_norm": 0.8731628060340881, + "learning_rate": 0.0009989914836396236, + "loss": 0.5748, + "step": 139 + }, + { + "epoch": 0.00782188451546219, + "grad_norm": 1.8492680788040161, + "learning_rate": 0.0009989634692962796, + "loss": 0.5044, + "step": 140 + }, + { + "epoch": 0.007877755119144063, + "grad_norm": 0.6291975975036621, + "learning_rate": 0.0009989354549529358, + "loss": 0.6186, + "step": 141 + }, + { + "epoch": 0.007933625722825936, + "grad_norm": 1.1023533344268799, + "learning_rate": 0.000998907440609592, + "loss": 0.6434, + "step": 142 + }, + { + "epoch": 0.007989496326507807, + "grad_norm": 0.8824913501739502, + "learning_rate": 0.0009988794262662485, + "loss": 0.6231, + "step": 143 + }, + { + "epoch": 0.00804536693018968, + "grad_norm": 1.105514645576477, + "learning_rate": 0.0009988514119229045, + "loss": 0.6069, + "step": 144 + }, + { + "epoch": 0.008101237533871554, + "grad_norm": 0.7954631447792053, + "learning_rate": 0.0009988233975795607, + "loss": 0.8229, + "step": 145 + }, + { + "epoch": 0.008157108137553427, + "grad_norm": 1.503570556640625, + "learning_rate": 0.000998795383236217, + "loss": 0.5088, + "step": 146 + }, + { + "epoch": 0.0082129787412353, + "grad_norm": 1.0187616348266602, + "learning_rate": 0.000998767368892873, + "loss": 0.5962, + "step": 147 + }, + { + "epoch": 0.008268849344917171, + "grad_norm": 0.7943301796913147, + "learning_rate": 0.0009987393545495293, + "loss": 0.4783, + "step": 148 + }, + { + "epoch": 0.008324719948599044, + "grad_norm": 0.6867227554321289, + "learning_rate": 0.0009987113402061855, + "loss": 0.6718, + "step": 149 + }, + { + "epoch": 0.008380590552280917, + "grad_norm": 0.7399408221244812, + "learning_rate": 0.000998683325862842, + "loss": 0.567, + "step": 150 + }, + { + "epoch": 0.00843646115596279, + "grad_norm": 1.346256971359253, + "learning_rate": 0.000998655311519498, + "loss": 0.5806, + "step": 151 + }, + { + "epoch": 0.008492331759644664, + "grad_norm": 0.6399877667427063, + "learning_rate": 0.000998627297176154, + "loss": 0.5179, + "step": 152 + }, + { + "epoch": 0.008548202363326535, + "grad_norm": 1.6485328674316406, + "learning_rate": 0.0009985992828328105, + "loss": 0.5184, + "step": 153 + }, + { + "epoch": 0.008604072967008408, + "grad_norm": 0.5774408578872681, + "learning_rate": 0.0009985712684894667, + "loss": 0.528, + "step": 154 + }, + { + "epoch": 0.008659943570690281, + "grad_norm": 0.7464857697486877, + "learning_rate": 0.0009985432541461227, + "loss": 0.5035, + "step": 155 + }, + { + "epoch": 0.008715814174372154, + "grad_norm": 0.5616545677185059, + "learning_rate": 0.000998515239802779, + "loss": 0.529, + "step": 156 + }, + { + "epoch": 0.008771684778054028, + "grad_norm": 0.7175976037979126, + "learning_rate": 0.0009984872254594354, + "loss": 0.4717, + "step": 157 + }, + { + "epoch": 0.008827555381735899, + "grad_norm": 0.6535531878471375, + "learning_rate": 0.0009984592111160916, + "loss": 0.6204, + "step": 158 + }, + { + "epoch": 0.008883425985417772, + "grad_norm": 3.6171979904174805, + "learning_rate": 0.0009984311967727475, + "loss": 0.5478, + "step": 159 + }, + { + "epoch": 0.008939296589099645, + "grad_norm": 2.9151546955108643, + "learning_rate": 0.000998403182429404, + "loss": 0.5196, + "step": 160 + }, + { + "epoch": 0.008995167192781518, + "grad_norm": 0.8652129173278809, + "learning_rate": 0.0009983751680860602, + "loss": 0.606, + "step": 161 + }, + { + "epoch": 0.009051037796463391, + "grad_norm": 1.096207857131958, + "learning_rate": 0.0009983471537427164, + "loss": 0.734, + "step": 162 + }, + { + "epoch": 0.009106908400145263, + "grad_norm": 0.6264573335647583, + "learning_rate": 0.0009983191393993724, + "loss": 0.6202, + "step": 163 + }, + { + "epoch": 0.009162779003827136, + "grad_norm": 0.8602551817893982, + "learning_rate": 0.0009982911250560288, + "loss": 0.588, + "step": 164 + }, + { + "epoch": 0.009218649607509009, + "grad_norm": 4.818134307861328, + "learning_rate": 0.000998263110712685, + "loss": 0.5832, + "step": 165 + }, + { + "epoch": 0.009274520211190882, + "grad_norm": 1.0116568803787231, + "learning_rate": 0.0009982350963693412, + "loss": 0.554, + "step": 166 + }, + { + "epoch": 0.009330390814872755, + "grad_norm": 0.8113193511962891, + "learning_rate": 0.0009982070820259974, + "loss": 0.5442, + "step": 167 + }, + { + "epoch": 0.009386261418554627, + "grad_norm": 0.7374023199081421, + "learning_rate": 0.0009981790676826536, + "loss": 0.6257, + "step": 168 + }, + { + "epoch": 0.0094421320222365, + "grad_norm": 0.8976988792419434, + "learning_rate": 0.0009981510533393098, + "loss": 0.5686, + "step": 169 + }, + { + "epoch": 0.009498002625918373, + "grad_norm": 1.1638751029968262, + "learning_rate": 0.0009981230389959658, + "loss": 0.4854, + "step": 170 + }, + { + "epoch": 0.009553873229600246, + "grad_norm": 0.8561503291130066, + "learning_rate": 0.0009980950246526222, + "loss": 0.5873, + "step": 171 + }, + { + "epoch": 0.009609743833282119, + "grad_norm": 4.294755458831787, + "learning_rate": 0.0009980670103092784, + "loss": 0.5831, + "step": 172 + }, + { + "epoch": 0.009665614436963992, + "grad_norm": 0.9047147631645203, + "learning_rate": 0.0009980389959659346, + "loss": 0.5426, + "step": 173 + }, + { + "epoch": 0.009721485040645864, + "grad_norm": 1.2661430835723877, + "learning_rate": 0.0009980109816225906, + "loss": 0.5305, + "step": 174 + }, + { + "epoch": 0.009777355644327737, + "grad_norm": 1.1352730989456177, + "learning_rate": 0.000997982967279247, + "loss": 0.7094, + "step": 175 + }, + { + "epoch": 0.00983322624800961, + "grad_norm": 0.8030979633331299, + "learning_rate": 0.0009979549529359033, + "loss": 0.6311, + "step": 176 + }, + { + "epoch": 0.009889096851691483, + "grad_norm": 1.153566598892212, + "learning_rate": 0.0009979269385925595, + "loss": 0.5493, + "step": 177 + }, + { + "epoch": 0.009944967455373356, + "grad_norm": 0.759208083152771, + "learning_rate": 0.0009978989242492157, + "loss": 0.461, + "step": 178 + }, + { + "epoch": 0.010000838059055227, + "grad_norm": 1.2963123321533203, + "learning_rate": 0.0009978709099058719, + "loss": 0.6766, + "step": 179 + }, + { + "epoch": 0.0100567086627371, + "grad_norm": 0.8268970251083374, + "learning_rate": 0.000997842895562528, + "loss": 0.664, + "step": 180 + }, + { + "epoch": 0.010112579266418974, + "grad_norm": 0.6286240220069885, + "learning_rate": 0.0009978148812191843, + "loss": 0.5128, + "step": 181 + }, + { + "epoch": 0.010168449870100847, + "grad_norm": 0.9972626566886902, + "learning_rate": 0.0009977868668758405, + "loss": 0.4637, + "step": 182 + }, + { + "epoch": 0.01022432047378272, + "grad_norm": 0.5816459655761719, + "learning_rate": 0.0009977588525324967, + "loss": 0.5832, + "step": 183 + }, + { + "epoch": 0.010280191077464591, + "grad_norm": 0.7198776602745056, + "learning_rate": 0.000997730838189153, + "loss": 0.4621, + "step": 184 + }, + { + "epoch": 0.010336061681146464, + "grad_norm": 0.8224912881851196, + "learning_rate": 0.0009977028238458091, + "loss": 0.5893, + "step": 185 + }, + { + "epoch": 0.010391932284828338, + "grad_norm": 2.6516220569610596, + "learning_rate": 0.0009976748095024653, + "loss": 0.5009, + "step": 186 + }, + { + "epoch": 0.01044780288851021, + "grad_norm": 0.6114152669906616, + "learning_rate": 0.0009976467951591215, + "loss": 0.4809, + "step": 187 + }, + { + "epoch": 0.010503673492192084, + "grad_norm": 1.4305812120437622, + "learning_rate": 0.0009976187808157777, + "loss": 0.69, + "step": 188 + }, + { + "epoch": 0.010559544095873955, + "grad_norm": 0.6422624588012695, + "learning_rate": 0.000997590766472434, + "loss": 0.4922, + "step": 189 + }, + { + "epoch": 0.010615414699555828, + "grad_norm": 0.6341967582702637, + "learning_rate": 0.0009975627521290901, + "loss": 0.6033, + "step": 190 + }, + { + "epoch": 0.010671285303237701, + "grad_norm": 4.4234466552734375, + "learning_rate": 0.0009975347377857463, + "loss": 0.438, + "step": 191 + }, + { + "epoch": 0.010727155906919575, + "grad_norm": 0.6455267667770386, + "learning_rate": 0.0009975067234424026, + "loss": 0.5439, + "step": 192 + }, + { + "epoch": 0.010783026510601448, + "grad_norm": 0.9678451418876648, + "learning_rate": 0.0009974787090990588, + "loss": 0.5247, + "step": 193 + }, + { + "epoch": 0.010838897114283319, + "grad_norm": 0.9131279587745667, + "learning_rate": 0.000997450694755715, + "loss": 0.5404, + "step": 194 + }, + { + "epoch": 0.010894767717965192, + "grad_norm": 0.5026369690895081, + "learning_rate": 0.0009974226804123712, + "loss": 0.5717, + "step": 195 + }, + { + "epoch": 0.010950638321647065, + "grad_norm": 0.6639294624328613, + "learning_rate": 0.0009973946660690274, + "loss": 0.5912, + "step": 196 + }, + { + "epoch": 0.011006508925328938, + "grad_norm": 0.9167312383651733, + "learning_rate": 0.0009973666517256836, + "loss": 0.619, + "step": 197 + }, + { + "epoch": 0.011062379529010812, + "grad_norm": 0.6164106130599976, + "learning_rate": 0.0009973386373823398, + "loss": 0.5344, + "step": 198 + }, + { + "epoch": 0.011118250132692683, + "grad_norm": 1.3618087768554688, + "learning_rate": 0.000997310623038996, + "loss": 0.5184, + "step": 199 + }, + { + "epoch": 0.011174120736374556, + "grad_norm": 0.9335946440696716, + "learning_rate": 0.0009972826086956522, + "loss": 0.6834, + "step": 200 + }, + { + "epoch": 0.011229991340056429, + "grad_norm": 0.6644759774208069, + "learning_rate": 0.0009972545943523084, + "loss": 0.5255, + "step": 201 + }, + { + "epoch": 0.011285861943738302, + "grad_norm": 0.7159681916236877, + "learning_rate": 0.0009972265800089646, + "loss": 0.6112, + "step": 202 + }, + { + "epoch": 0.011341732547420175, + "grad_norm": 0.982142448425293, + "learning_rate": 0.0009971985656656208, + "loss": 0.5557, + "step": 203 + }, + { + "epoch": 0.011397603151102048, + "grad_norm": 1.1453272104263306, + "learning_rate": 0.000997170551322277, + "loss": 0.7088, + "step": 204 + }, + { + "epoch": 0.01145347375478392, + "grad_norm": 0.9344623684883118, + "learning_rate": 0.0009971425369789332, + "loss": 0.5803, + "step": 205 + }, + { + "epoch": 0.011509344358465793, + "grad_norm": 0.654420793056488, + "learning_rate": 0.0009971145226355894, + "loss": 0.5838, + "step": 206 + }, + { + "epoch": 0.011565214962147666, + "grad_norm": 1.4395933151245117, + "learning_rate": 0.0009970865082922456, + "loss": 0.7311, + "step": 207 + }, + { + "epoch": 0.01162108556582954, + "grad_norm": 1.3957278728485107, + "learning_rate": 0.0009970584939489018, + "loss": 0.5027, + "step": 208 + }, + { + "epoch": 0.011676956169511412, + "grad_norm": 0.968213677406311, + "learning_rate": 0.000997030479605558, + "loss": 0.5836, + "step": 209 + }, + { + "epoch": 0.011732826773193284, + "grad_norm": 0.8348031640052795, + "learning_rate": 0.0009970024652622143, + "loss": 0.6095, + "step": 210 + }, + { + "epoch": 0.011788697376875157, + "grad_norm": 0.9253244400024414, + "learning_rate": 0.0009969744509188705, + "loss": 0.6136, + "step": 211 + }, + { + "epoch": 0.01184456798055703, + "grad_norm": 7.531663417816162, + "learning_rate": 0.0009969464365755267, + "loss": 0.6475, + "step": 212 + }, + { + "epoch": 0.011900438584238903, + "grad_norm": 1.0691826343536377, + "learning_rate": 0.0009969184222321829, + "loss": 0.6441, + "step": 213 + }, + { + "epoch": 0.011956309187920776, + "grad_norm": 1.080286979675293, + "learning_rate": 0.000996890407888839, + "loss": 0.5304, + "step": 214 + }, + { + "epoch": 0.012012179791602648, + "grad_norm": 0.525710940361023, + "learning_rate": 0.0009968623935454953, + "loss": 0.4842, + "step": 215 + }, + { + "epoch": 0.01206805039528452, + "grad_norm": 0.9331747889518738, + "learning_rate": 0.0009968343792021515, + "loss": 0.6858, + "step": 216 + }, + { + "epoch": 0.012123920998966394, + "grad_norm": 4.237966060638428, + "learning_rate": 0.0009968063648588077, + "loss": 0.5871, + "step": 217 + }, + { + "epoch": 0.012179791602648267, + "grad_norm": 0.6103495359420776, + "learning_rate": 0.000996778350515464, + "loss": 0.5181, + "step": 218 + }, + { + "epoch": 0.01223566220633014, + "grad_norm": 0.7770369648933411, + "learning_rate": 0.0009967503361721201, + "loss": 0.6015, + "step": 219 + }, + { + "epoch": 0.012291532810012011, + "grad_norm": 1.623702883720398, + "learning_rate": 0.0009967223218287763, + "loss": 0.51, + "step": 220 + }, + { + "epoch": 0.012347403413693885, + "grad_norm": 0.7091982960700989, + "learning_rate": 0.0009966943074854325, + "loss": 0.6338, + "step": 221 + }, + { + "epoch": 0.012403274017375758, + "grad_norm": 7.936911582946777, + "learning_rate": 0.0009966662931420887, + "loss": 0.7197, + "step": 222 + }, + { + "epoch": 0.01245914462105763, + "grad_norm": 1.292877197265625, + "learning_rate": 0.000996638278798745, + "loss": 0.5284, + "step": 223 + }, + { + "epoch": 0.012515015224739504, + "grad_norm": 5.953118801116943, + "learning_rate": 0.0009966102644554011, + "loss": 0.6565, + "step": 224 + }, + { + "epoch": 0.012570885828421375, + "grad_norm": 0.9460536241531372, + "learning_rate": 0.0009965822501120573, + "loss": 0.6204, + "step": 225 + }, + { + "epoch": 0.012626756432103248, + "grad_norm": 0.6696051955223083, + "learning_rate": 0.0009965542357687136, + "loss": 0.5376, + "step": 226 + }, + { + "epoch": 0.012682627035785122, + "grad_norm": 0.42686882615089417, + "learning_rate": 0.0009965262214253698, + "loss": 0.4124, + "step": 227 + }, + { + "epoch": 0.012738497639466995, + "grad_norm": 0.981127917766571, + "learning_rate": 0.000996498207082026, + "loss": 0.6535, + "step": 228 + }, + { + "epoch": 0.012794368243148868, + "grad_norm": 0.8632869720458984, + "learning_rate": 0.0009964701927386822, + "loss": 0.4834, + "step": 229 + }, + { + "epoch": 0.01285023884683074, + "grad_norm": 0.6714855432510376, + "learning_rate": 0.0009964421783953384, + "loss": 0.497, + "step": 230 + }, + { + "epoch": 0.012906109450512612, + "grad_norm": 0.7276551723480225, + "learning_rate": 0.0009964141640519946, + "loss": 0.5355, + "step": 231 + }, + { + "epoch": 0.012961980054194485, + "grad_norm": 0.777809739112854, + "learning_rate": 0.0009963861497086508, + "loss": 0.5461, + "step": 232 + }, + { + "epoch": 0.013017850657876359, + "grad_norm": 0.9727482795715332, + "learning_rate": 0.000996358135365307, + "loss": 0.5994, + "step": 233 + }, + { + "epoch": 0.013073721261558232, + "grad_norm": 1.1351152658462524, + "learning_rate": 0.0009963301210219632, + "loss": 0.5121, + "step": 234 + }, + { + "epoch": 0.013129591865240105, + "grad_norm": 0.7407101988792419, + "learning_rate": 0.0009963021066786194, + "loss": 0.5373, + "step": 235 + }, + { + "epoch": 0.013185462468921976, + "grad_norm": 0.8484777808189392, + "learning_rate": 0.0009962740923352756, + "loss": 0.6947, + "step": 236 + }, + { + "epoch": 0.01324133307260385, + "grad_norm": 1.792380690574646, + "learning_rate": 0.0009962460779919318, + "loss": 0.6092, + "step": 237 + }, + { + "epoch": 0.013297203676285722, + "grad_norm": 4.273126602172852, + "learning_rate": 0.0009962180636485882, + "loss": 0.5348, + "step": 238 + }, + { + "epoch": 0.013353074279967595, + "grad_norm": 0.956825852394104, + "learning_rate": 0.0009961900493052442, + "loss": 0.5392, + "step": 239 + }, + { + "epoch": 0.013408944883649469, + "grad_norm": 1.4384194612503052, + "learning_rate": 0.0009961620349619004, + "loss": 0.5982, + "step": 240 + }, + { + "epoch": 0.01346481548733134, + "grad_norm": 1.2365484237670898, + "learning_rate": 0.0009961340206185566, + "loss": 0.5187, + "step": 241 + }, + { + "epoch": 0.013520686091013213, + "grad_norm": 0.8781028985977173, + "learning_rate": 0.000996106006275213, + "loss": 0.525, + "step": 242 + }, + { + "epoch": 0.013576556694695086, + "grad_norm": 1.0476173162460327, + "learning_rate": 0.000996077991931869, + "loss": 0.5004, + "step": 243 + }, + { + "epoch": 0.01363242729837696, + "grad_norm": 0.668707549571991, + "learning_rate": 0.0009960499775885253, + "loss": 0.522, + "step": 244 + }, + { + "epoch": 0.013688297902058832, + "grad_norm": 0.9416015148162842, + "learning_rate": 0.0009960219632451817, + "loss": 0.4272, + "step": 245 + }, + { + "epoch": 0.013744168505740704, + "grad_norm": Infinity, + "learning_rate": 0.0009960219632451817, + "loss": 0.5034, + "step": 246 + }, + { + "epoch": 0.013800039109422577, + "grad_norm": 0.8535059094429016, + "learning_rate": 0.0009959939489018379, + "loss": 0.4354, + "step": 247 + }, + { + "epoch": 0.01385590971310445, + "grad_norm": 0.9478965401649475, + "learning_rate": 0.0009959659345584939, + "loss": 0.7323, + "step": 248 + }, + { + "epoch": 0.013911780316786323, + "grad_norm": 2.7022037506103516, + "learning_rate": 0.00099593792021515, + "loss": 0.4983, + "step": 249 + }, + { + "epoch": 0.013967650920468196, + "grad_norm": 1.2693678140640259, + "learning_rate": 0.0009959099058718065, + "loss": 0.4794, + "step": 250 + }, + { + "epoch": 0.014023521524150068, + "grad_norm": 1.4684839248657227, + "learning_rate": 0.0009958818915284627, + "loss": 0.5292, + "step": 251 + }, + { + "epoch": 0.01407939212783194, + "grad_norm": 7.83448600769043, + "learning_rate": 0.0009958538771851187, + "loss": 0.5578, + "step": 252 + }, + { + "epoch": 0.014135262731513814, + "grad_norm": 0.9651688933372498, + "learning_rate": 0.000995825862841775, + "loss": 0.6312, + "step": 253 + }, + { + "epoch": 0.014191133335195687, + "grad_norm": 0.7612367868423462, + "learning_rate": 0.0009957978484984313, + "loss": 0.463, + "step": 254 + }, + { + "epoch": 0.01424700393887756, + "grad_norm": 1.1215522289276123, + "learning_rate": 0.0009957698341550873, + "loss": 0.515, + "step": 255 + }, + { + "epoch": 0.014302874542559432, + "grad_norm": 2.170574188232422, + "learning_rate": 0.0009957418198117435, + "loss": 0.6287, + "step": 256 + }, + { + "epoch": 0.014358745146241305, + "grad_norm": 0.9152688384056091, + "learning_rate": 0.0009957138054684, + "loss": 0.518, + "step": 257 + }, + { + "epoch": 0.014414615749923178, + "grad_norm": 0.7710979580879211, + "learning_rate": 0.0009956857911250562, + "loss": 0.4738, + "step": 258 + }, + { + "epoch": 0.014470486353605051, + "grad_norm": 4.0709099769592285, + "learning_rate": 0.0009956577767817121, + "loss": 0.4637, + "step": 259 + }, + { + "epoch": 0.014526356957286924, + "grad_norm": 1.55964994430542, + "learning_rate": 0.0009956297624383683, + "loss": 0.7035, + "step": 260 + }, + { + "epoch": 0.014582227560968795, + "grad_norm": 1.314137578010559, + "learning_rate": 0.0009956017480950248, + "loss": 0.5678, + "step": 261 + }, + { + "epoch": 0.014638098164650669, + "grad_norm": 0.8742054104804993, + "learning_rate": 0.000995573733751681, + "loss": 0.4658, + "step": 262 + }, + { + "epoch": 0.014693968768332542, + "grad_norm": 1.5565518140792847, + "learning_rate": 0.000995545719408337, + "loss": 0.569, + "step": 263 + }, + { + "epoch": 0.014749839372014415, + "grad_norm": 1.1992205381393433, + "learning_rate": 0.0009955177050649934, + "loss": 0.5458, + "step": 264 + }, + { + "epoch": 0.014805709975696288, + "grad_norm": 2.763195276260376, + "learning_rate": 0.0009954896907216496, + "loss": 0.5302, + "step": 265 + }, + { + "epoch": 0.014861580579378161, + "grad_norm": 0.793287992477417, + "learning_rate": 0.0009954616763783058, + "loss": 0.6315, + "step": 266 + }, + { + "epoch": 0.014917451183060032, + "grad_norm": 0.8853189945220947, + "learning_rate": 0.0009954336620349618, + "loss": 0.5892, + "step": 267 + }, + { + "epoch": 0.014973321786741906, + "grad_norm": 0.7319135665893555, + "learning_rate": 0.0009954056476916182, + "loss": 0.4616, + "step": 268 + }, + { + "epoch": 0.015029192390423779, + "grad_norm": 1.3457797765731812, + "learning_rate": 0.0009953776333482744, + "loss": 0.6295, + "step": 269 + }, + { + "epoch": 0.015085062994105652, + "grad_norm": 3.821444272994995, + "learning_rate": 0.0009953496190049306, + "loss": 0.6652, + "step": 270 + }, + { + "epoch": 0.015140933597787525, + "grad_norm": 0.7695485353469849, + "learning_rate": 0.0009953216046615868, + "loss": 0.6073, + "step": 271 + }, + { + "epoch": 0.015196804201469396, + "grad_norm": 0.8444585800170898, + "learning_rate": 0.000995293590318243, + "loss": 0.4436, + "step": 272 + }, + { + "epoch": 0.01525267480515127, + "grad_norm": 0.6111328601837158, + "learning_rate": 0.0009952655759748992, + "loss": 0.4523, + "step": 273 + }, + { + "epoch": 0.015308545408833142, + "grad_norm": 1.8818416595458984, + "learning_rate": 0.0009952375616315552, + "loss": 0.632, + "step": 274 + }, + { + "epoch": 0.015364416012515016, + "grad_norm": 1.980122447013855, + "learning_rate": 0.0009952095472882116, + "loss": 0.6129, + "step": 275 + }, + { + "epoch": 0.015420286616196889, + "grad_norm": 3.597630023956299, + "learning_rate": 0.0009951815329448679, + "loss": 0.4021, + "step": 276 + }, + { + "epoch": 0.01547615721987876, + "grad_norm": 0.9573484063148499, + "learning_rate": 0.000995153518601524, + "loss": 0.5567, + "step": 277 + }, + { + "epoch": 0.015532027823560633, + "grad_norm": 1.217905044555664, + "learning_rate": 0.0009951255042581803, + "loss": 0.5536, + "step": 278 + }, + { + "epoch": 0.015587898427242506, + "grad_norm": 2.0184216499328613, + "learning_rate": 0.0009950974899148365, + "loss": 0.6804, + "step": 279 + }, + { + "epoch": 0.01564376903092438, + "grad_norm": 0.6289705038070679, + "learning_rate": 0.0009950694755714927, + "loss": 0.5584, + "step": 280 + }, + { + "epoch": 0.01569963963460625, + "grad_norm": 0.7022606730461121, + "learning_rate": 0.0009950414612281489, + "loss": 0.4801, + "step": 281 + }, + { + "epoch": 0.015755510238288126, + "grad_norm": 2.401607036590576, + "learning_rate": 0.000995013446884805, + "loss": 0.6003, + "step": 282 + }, + { + "epoch": 0.015811380841969997, + "grad_norm": 0.7965333461761475, + "learning_rate": 0.0009949854325414613, + "loss": 0.4385, + "step": 283 + }, + { + "epoch": 0.015867251445651872, + "grad_norm": 0.7786020636558533, + "learning_rate": 0.0009949574181981175, + "loss": 0.5219, + "step": 284 + }, + { + "epoch": 0.015923122049333743, + "grad_norm": 1.109107494354248, + "learning_rate": 0.0009949294038547737, + "loss": 0.6919, + "step": 285 + }, + { + "epoch": 0.015978992653015615, + "grad_norm": 0.5793779492378235, + "learning_rate": 0.00099490138951143, + "loss": 0.5706, + "step": 286 + }, + { + "epoch": 0.01603486325669749, + "grad_norm": 1.2447439432144165, + "learning_rate": 0.0009948733751680861, + "loss": 0.7168, + "step": 287 + }, + { + "epoch": 0.01609073386037936, + "grad_norm": 0.7608517408370972, + "learning_rate": 0.0009948453608247423, + "loss": 0.4634, + "step": 288 + }, + { + "epoch": 0.016146604464061236, + "grad_norm": 0.6219472885131836, + "learning_rate": 0.0009948173464813985, + "loss": 0.5957, + "step": 289 + }, + { + "epoch": 0.016202475067743107, + "grad_norm": 0.5357271432876587, + "learning_rate": 0.0009947893321380547, + "loss": 0.4279, + "step": 290 + }, + { + "epoch": 0.01625834567142498, + "grad_norm": 1.0176457166671753, + "learning_rate": 0.000994761317794711, + "loss": 0.5248, + "step": 291 + }, + { + "epoch": 0.016314216275106853, + "grad_norm": 0.6558994054794312, + "learning_rate": 0.0009947333034513671, + "loss": 0.5893, + "step": 292 + }, + { + "epoch": 0.016370086878788725, + "grad_norm": 4.776406764984131, + "learning_rate": 0.0009947052891080234, + "loss": 0.6026, + "step": 293 + }, + { + "epoch": 0.0164259574824706, + "grad_norm": 0.8875206708908081, + "learning_rate": 0.0009946772747646796, + "loss": 0.4935, + "step": 294 + }, + { + "epoch": 0.01648182808615247, + "grad_norm": 0.6755164265632629, + "learning_rate": 0.0009946492604213358, + "loss": 0.5558, + "step": 295 + }, + { + "epoch": 0.016537698689834342, + "grad_norm": 0.6109861731529236, + "learning_rate": 0.000994621246077992, + "loss": 0.5475, + "step": 296 + }, + { + "epoch": 0.016593569293516217, + "grad_norm": 2.507129430770874, + "learning_rate": 0.0009945932317346482, + "loss": 0.5393, + "step": 297 + }, + { + "epoch": 0.01664943989719809, + "grad_norm": 0.5943477749824524, + "learning_rate": 0.0009945652173913044, + "loss": 0.5056, + "step": 298 + }, + { + "epoch": 0.016705310500879963, + "grad_norm": 0.6246666312217712, + "learning_rate": 0.0009945372030479606, + "loss": 0.4129, + "step": 299 + }, + { + "epoch": 0.016761181104561835, + "grad_norm": 0.7372632622718811, + "learning_rate": 0.0009945091887046168, + "loss": 0.4984, + "step": 300 + }, + { + "epoch": 0.016817051708243706, + "grad_norm": 1.3006491661071777, + "learning_rate": 0.000994481174361273, + "loss": 0.5726, + "step": 301 + }, + { + "epoch": 0.01687292231192558, + "grad_norm": 0.7408469319343567, + "learning_rate": 0.0009944531600179292, + "loss": 0.5044, + "step": 302 + }, + { + "epoch": 0.016928792915607453, + "grad_norm": 0.9130346179008484, + "learning_rate": 0.0009944251456745854, + "loss": 0.548, + "step": 303 + }, + { + "epoch": 0.016984663519289327, + "grad_norm": 0.7799953818321228, + "learning_rate": 0.0009943971313312416, + "loss": 0.4926, + "step": 304 + }, + { + "epoch": 0.0170405341229712, + "grad_norm": 0.7420705556869507, + "learning_rate": 0.0009943691169878978, + "loss": 0.5243, + "step": 305 + }, + { + "epoch": 0.01709640472665307, + "grad_norm": 0.621039092540741, + "learning_rate": 0.000994341102644554, + "loss": 0.4938, + "step": 306 + }, + { + "epoch": 0.017152275330334945, + "grad_norm": 0.7581995129585266, + "learning_rate": 0.0009943130883012102, + "loss": 0.631, + "step": 307 + }, + { + "epoch": 0.017208145934016816, + "grad_norm": 0.5527749061584473, + "learning_rate": 0.0009942850739578664, + "loss": 0.4861, + "step": 308 + }, + { + "epoch": 0.01726401653769869, + "grad_norm": 0.7766637206077576, + "learning_rate": 0.0009942570596145226, + "loss": 0.4723, + "step": 309 + }, + { + "epoch": 0.017319887141380563, + "grad_norm": 0.803333580493927, + "learning_rate": 0.0009942290452711789, + "loss": 0.5694, + "step": 310 + }, + { + "epoch": 0.017375757745062434, + "grad_norm": 2.1928653717041016, + "learning_rate": 0.000994201030927835, + "loss": 0.5234, + "step": 311 + }, + { + "epoch": 0.01743162834874431, + "grad_norm": 0.8691964745521545, + "learning_rate": 0.0009941730165844913, + "loss": 0.5403, + "step": 312 + }, + { + "epoch": 0.01748749895242618, + "grad_norm": 2.0495004653930664, + "learning_rate": 0.0009941450022411475, + "loss": 0.7106, + "step": 313 + }, + { + "epoch": 0.017543369556108055, + "grad_norm": 0.6768099069595337, + "learning_rate": 0.0009941169878978037, + "loss": 0.4954, + "step": 314 + }, + { + "epoch": 0.017599240159789926, + "grad_norm": 3.1668097972869873, + "learning_rate": 0.0009940889735544599, + "loss": 0.4457, + "step": 315 + }, + { + "epoch": 0.017655110763471798, + "grad_norm": 0.8995234370231628, + "learning_rate": 0.000994060959211116, + "loss": 0.5588, + "step": 316 + }, + { + "epoch": 0.017710981367153673, + "grad_norm": 0.7006953954696655, + "learning_rate": 0.0009940329448677723, + "loss": 0.4909, + "step": 317 + }, + { + "epoch": 0.017766851970835544, + "grad_norm": 1.1206692457199097, + "learning_rate": 0.0009940049305244285, + "loss": 0.4688, + "step": 318 + }, + { + "epoch": 0.01782272257451742, + "grad_norm": 11.03077220916748, + "learning_rate": 0.0009939769161810847, + "loss": 0.5646, + "step": 319 + }, + { + "epoch": 0.01787859317819929, + "grad_norm": 1.351910948753357, + "learning_rate": 0.000993948901837741, + "loss": 0.503, + "step": 320 + }, + { + "epoch": 0.01793446378188116, + "grad_norm": 0.9659982323646545, + "learning_rate": 0.0009939208874943971, + "loss": 0.6384, + "step": 321 + }, + { + "epoch": 0.017990334385563037, + "grad_norm": 1.247702717781067, + "learning_rate": 0.0009938928731510533, + "loss": 0.7019, + "step": 322 + }, + { + "epoch": 0.018046204989244908, + "grad_norm": 0.8949902653694153, + "learning_rate": 0.0009938648588077095, + "loss": 0.6087, + "step": 323 + }, + { + "epoch": 0.018102075592926783, + "grad_norm": 0.7702025175094604, + "learning_rate": 0.0009938368444643657, + "loss": 0.4487, + "step": 324 + }, + { + "epoch": 0.018157946196608654, + "grad_norm": 1.1744626760482788, + "learning_rate": 0.000993808830121022, + "loss": 0.52, + "step": 325 + }, + { + "epoch": 0.018213816800290526, + "grad_norm": 5.942630290985107, + "learning_rate": 0.0009937808157776781, + "loss": 0.4629, + "step": 326 + }, + { + "epoch": 0.0182696874039724, + "grad_norm": 0.6744368672370911, + "learning_rate": 0.0009937528014343344, + "loss": 0.5328, + "step": 327 + }, + { + "epoch": 0.018325558007654272, + "grad_norm": 1.3294079303741455, + "learning_rate": 0.0009937247870909906, + "loss": 0.659, + "step": 328 + }, + { + "epoch": 0.018381428611336147, + "grad_norm": 0.6858976483345032, + "learning_rate": 0.0009936967727476468, + "loss": 0.6095, + "step": 329 + }, + { + "epoch": 0.018437299215018018, + "grad_norm": 1.0796178579330444, + "learning_rate": 0.000993668758404303, + "loss": 0.6578, + "step": 330 + }, + { + "epoch": 0.01849316981869989, + "grad_norm": 2.2210335731506348, + "learning_rate": 0.0009936407440609592, + "loss": 0.4898, + "step": 331 + }, + { + "epoch": 0.018549040422381764, + "grad_norm": 4.343039035797119, + "learning_rate": 0.0009936127297176154, + "loss": 0.5593, + "step": 332 + }, + { + "epoch": 0.018604911026063636, + "grad_norm": 0.8294679522514343, + "learning_rate": 0.0009935847153742716, + "loss": 0.8311, + "step": 333 + }, + { + "epoch": 0.01866078162974551, + "grad_norm": 0.6621750593185425, + "learning_rate": 0.0009935567010309278, + "loss": 0.4922, + "step": 334 + }, + { + "epoch": 0.018716652233427382, + "grad_norm": 16.380765914916992, + "learning_rate": 0.0009935286866875842, + "loss": 0.5359, + "step": 335 + }, + { + "epoch": 0.018772522837109253, + "grad_norm": 0.7395516037940979, + "learning_rate": 0.0009935006723442402, + "loss": 0.4318, + "step": 336 + }, + { + "epoch": 0.018828393440791128, + "grad_norm": 1.1283190250396729, + "learning_rate": 0.0009934726580008964, + "loss": 0.5202, + "step": 337 + }, + { + "epoch": 0.018884264044473, + "grad_norm": 1.7048271894454956, + "learning_rate": 0.0009934446436575526, + "loss": 0.4498, + "step": 338 + }, + { + "epoch": 0.018940134648154874, + "grad_norm": 10.00355052947998, + "learning_rate": 0.0009934166293142088, + "loss": 0.5875, + "step": 339 + }, + { + "epoch": 0.018996005251836746, + "grad_norm": 6.144944190979004, + "learning_rate": 0.000993388614970865, + "loss": 0.586, + "step": 340 + }, + { + "epoch": 0.01905187585551862, + "grad_norm": 1.1087642908096313, + "learning_rate": 0.0009933606006275212, + "loss": 0.5163, + "step": 341 + }, + { + "epoch": 0.019107746459200492, + "grad_norm": 1.0457431077957153, + "learning_rate": 0.0009933325862841777, + "loss": 0.5936, + "step": 342 + }, + { + "epoch": 0.019163617062882363, + "grad_norm": 0.7320237159729004, + "learning_rate": 0.0009933045719408336, + "loss": 0.5611, + "step": 343 + }, + { + "epoch": 0.019219487666564238, + "grad_norm": 8.197834968566895, + "learning_rate": 0.0009932765575974899, + "loss": 0.4358, + "step": 344 + }, + { + "epoch": 0.01927535827024611, + "grad_norm": 2.2163619995117188, + "learning_rate": 0.000993248543254146, + "loss": 0.7057, + "step": 345 + }, + { + "epoch": 0.019331228873927984, + "grad_norm": 0.9547022581100464, + "learning_rate": 0.0009932205289108025, + "loss": 0.5529, + "step": 346 + }, + { + "epoch": 0.019387099477609856, + "grad_norm": 1.1394741535186768, + "learning_rate": 0.0009931925145674585, + "loss": 0.5451, + "step": 347 + }, + { + "epoch": 0.019442970081291727, + "grad_norm": 0.7439435124397278, + "learning_rate": 0.0009931645002241147, + "loss": 0.4754, + "step": 348 + }, + { + "epoch": 0.019498840684973602, + "grad_norm": 1.2580468654632568, + "learning_rate": 0.000993136485880771, + "loss": 0.6615, + "step": 349 + }, + { + "epoch": 0.019554711288655473, + "grad_norm": 1.0282909870147705, + "learning_rate": 0.0009931084715374273, + "loss": 0.4915, + "step": 350 + }, + { + "epoch": 0.01961058189233735, + "grad_norm": 1.294224500656128, + "learning_rate": 0.0009930804571940833, + "loss": 0.4389, + "step": 351 + }, + { + "epoch": 0.01966645249601922, + "grad_norm": 0.8677000403404236, + "learning_rate": 0.0009930524428507395, + "loss": 0.6936, + "step": 352 + }, + { + "epoch": 0.01972232309970109, + "grad_norm": 0.6173989176750183, + "learning_rate": 0.000993024428507396, + "loss": 0.4747, + "step": 353 + }, + { + "epoch": 0.019778193703382966, + "grad_norm": 1.2623603343963623, + "learning_rate": 0.0009929964141640521, + "loss": 0.5865, + "step": 354 + }, + { + "epoch": 0.019834064307064837, + "grad_norm": 0.8946543335914612, + "learning_rate": 0.0009929683998207081, + "loss": 0.4996, + "step": 355 + }, + { + "epoch": 0.019889934910746712, + "grad_norm": 0.878404438495636, + "learning_rate": 0.0009929403854773643, + "loss": 0.4651, + "step": 356 + }, + { + "epoch": 0.019945805514428584, + "grad_norm": 0.738836407661438, + "learning_rate": 0.0009929123711340207, + "loss": 0.4317, + "step": 357 + }, + { + "epoch": 0.020001676118110455, + "grad_norm": 1.463197946548462, + "learning_rate": 0.000992884356790677, + "loss": 0.6017, + "step": 358 + }, + { + "epoch": 0.02005754672179233, + "grad_norm": 1.265312671661377, + "learning_rate": 0.000992856342447333, + "loss": 0.533, + "step": 359 + }, + { + "epoch": 0.0201134173254742, + "grad_norm": 1.2176283597946167, + "learning_rate": 0.0009928283281039894, + "loss": 0.4931, + "step": 360 + }, + { + "epoch": 0.020169287929156076, + "grad_norm": 0.8092159628868103, + "learning_rate": 0.0009928003137606456, + "loss": 0.5133, + "step": 361 + }, + { + "epoch": 0.020225158532837947, + "grad_norm": 1.8551291227340698, + "learning_rate": 0.0009927722994173016, + "loss": 0.4985, + "step": 362 + }, + { + "epoch": 0.02028102913651982, + "grad_norm": 0.9931104779243469, + "learning_rate": 0.0009927442850739578, + "loss": 0.6966, + "step": 363 + }, + { + "epoch": 0.020336899740201694, + "grad_norm": 1.1260002851486206, + "learning_rate": 0.0009927162707306142, + "loss": 0.4502, + "step": 364 + }, + { + "epoch": 0.020392770343883565, + "grad_norm": 1.06678307056427, + "learning_rate": 0.0009926882563872704, + "loss": 0.5984, + "step": 365 + }, + { + "epoch": 0.02044864094756544, + "grad_norm": 1.278329610824585, + "learning_rate": 0.0009926602420439264, + "loss": 0.5319, + "step": 366 + }, + { + "epoch": 0.02050451155124731, + "grad_norm": 0.7053854465484619, + "learning_rate": 0.0009926322277005828, + "loss": 0.5956, + "step": 367 + }, + { + "epoch": 0.020560382154929183, + "grad_norm": 0.7640781998634338, + "learning_rate": 0.000992604213357239, + "loss": 0.4918, + "step": 368 + }, + { + "epoch": 0.020616252758611057, + "grad_norm": 4.649686813354492, + "learning_rate": 0.0009925761990138952, + "loss": 0.5129, + "step": 369 + }, + { + "epoch": 0.02067212336229293, + "grad_norm": 1.067429542541504, + "learning_rate": 0.0009925481846705512, + "loss": 0.7692, + "step": 370 + }, + { + "epoch": 0.020727993965974804, + "grad_norm": 3.237643003463745, + "learning_rate": 0.0009925201703272076, + "loss": 0.5533, + "step": 371 + }, + { + "epoch": 0.020783864569656675, + "grad_norm": 0.8193700313568115, + "learning_rate": 0.0009924921559838638, + "loss": 0.566, + "step": 372 + }, + { + "epoch": 0.020839735173338546, + "grad_norm": 1.3124401569366455, + "learning_rate": 0.00099246414164052, + "loss": 0.6695, + "step": 373 + }, + { + "epoch": 0.02089560577702042, + "grad_norm": 0.9597424268722534, + "learning_rate": 0.0009924361272971762, + "loss": 0.508, + "step": 374 + }, + { + "epoch": 0.020951476380702293, + "grad_norm": 0.7854158282279968, + "learning_rate": 0.0009924081129538324, + "loss": 0.6181, + "step": 375 + }, + { + "epoch": 0.021007346984384168, + "grad_norm": 0.49362558126449585, + "learning_rate": 0.0009923800986104887, + "loss": 0.5616, + "step": 376 + }, + { + "epoch": 0.02106321758806604, + "grad_norm": 1.3709100484848022, + "learning_rate": 0.0009923520842671449, + "loss": 0.53, + "step": 377 + }, + { + "epoch": 0.02111908819174791, + "grad_norm": 1.007528305053711, + "learning_rate": 0.000992324069923801, + "loss": 0.4298, + "step": 378 + }, + { + "epoch": 0.021174958795429785, + "grad_norm": 1.1620508432388306, + "learning_rate": 0.0009922960555804573, + "loss": 0.6751, + "step": 379 + }, + { + "epoch": 0.021230829399111657, + "grad_norm": 1.2958288192749023, + "learning_rate": 0.0009922680412371135, + "loss": 0.5543, + "step": 380 + }, + { + "epoch": 0.02128670000279353, + "grad_norm": 0.929796040058136, + "learning_rate": 0.0009922400268937697, + "loss": 0.6239, + "step": 381 + }, + { + "epoch": 0.021342570606475403, + "grad_norm": 1.0205943584442139, + "learning_rate": 0.0009922120125504259, + "loss": 0.6479, + "step": 382 + }, + { + "epoch": 0.021398441210157274, + "grad_norm": 0.8800366520881653, + "learning_rate": 0.000992183998207082, + "loss": 0.5343, + "step": 383 + }, + { + "epoch": 0.02145431181383915, + "grad_norm": 1.1886740922927856, + "learning_rate": 0.0009921559838637383, + "loss": 0.5387, + "step": 384 + }, + { + "epoch": 0.02151018241752102, + "grad_norm": 0.7785093784332275, + "learning_rate": 0.0009921279695203945, + "loss": 0.597, + "step": 385 + }, + { + "epoch": 0.021566053021202895, + "grad_norm": 0.9814639091491699, + "learning_rate": 0.0009920999551770507, + "loss": 0.5514, + "step": 386 + }, + { + "epoch": 0.021621923624884767, + "grad_norm": 1.5286850929260254, + "learning_rate": 0.000992071940833707, + "loss": 0.562, + "step": 387 + }, + { + "epoch": 0.021677794228566638, + "grad_norm": 0.7946255207061768, + "learning_rate": 0.0009920439264903631, + "loss": 0.4985, + "step": 388 + }, + { + "epoch": 0.021733664832248513, + "grad_norm": 0.6983494758605957, + "learning_rate": 0.0009920159121470193, + "loss": 0.5985, + "step": 389 + }, + { + "epoch": 0.021789535435930384, + "grad_norm": 0.8431791067123413, + "learning_rate": 0.0009919878978036755, + "loss": 0.6424, + "step": 390 + }, + { + "epoch": 0.02184540603961226, + "grad_norm": 1.0107462406158447, + "learning_rate": 0.0009919598834603317, + "loss": 0.4487, + "step": 391 + }, + { + "epoch": 0.02190127664329413, + "grad_norm": 0.865647554397583, + "learning_rate": 0.000991931869116988, + "loss": 0.643, + "step": 392 + }, + { + "epoch": 0.021957147246976002, + "grad_norm": 0.5327861309051514, + "learning_rate": 0.0009919038547736442, + "loss": 0.5894, + "step": 393 + }, + { + "epoch": 0.022013017850657877, + "grad_norm": 7.441356658935547, + "learning_rate": 0.0009918758404303004, + "loss": 0.558, + "step": 394 + }, + { + "epoch": 0.022068888454339748, + "grad_norm": 0.6429920792579651, + "learning_rate": 0.0009918478260869566, + "loss": 0.6124, + "step": 395 + }, + { + "epoch": 0.022124759058021623, + "grad_norm": 0.7117517590522766, + "learning_rate": 0.0009918198117436128, + "loss": 0.5386, + "step": 396 + }, + { + "epoch": 0.022180629661703494, + "grad_norm": 0.8142665028572083, + "learning_rate": 0.000991791797400269, + "loss": 0.4593, + "step": 397 + }, + { + "epoch": 0.022236500265385366, + "grad_norm": 0.752034068107605, + "learning_rate": 0.0009917637830569252, + "loss": 0.6463, + "step": 398 + }, + { + "epoch": 0.02229237086906724, + "grad_norm": 0.6635132431983948, + "learning_rate": 0.0009917357687135814, + "loss": 0.5597, + "step": 399 + }, + { + "epoch": 0.022348241472749112, + "grad_norm": 1.7643057107925415, + "learning_rate": 0.0009917077543702376, + "loss": 0.4724, + "step": 400 + }, + { + "epoch": 0.022404112076430987, + "grad_norm": 1.499964952468872, + "learning_rate": 0.0009916797400268938, + "loss": 0.5539, + "step": 401 + }, + { + "epoch": 0.022459982680112858, + "grad_norm": 0.7209556102752686, + "learning_rate": 0.00099165172568355, + "loss": 0.5622, + "step": 402 + }, + { + "epoch": 0.022515853283794733, + "grad_norm": 1.381666898727417, + "learning_rate": 0.0009916237113402062, + "loss": 0.4669, + "step": 403 + }, + { + "epoch": 0.022571723887476604, + "grad_norm": 4.490616321563721, + "learning_rate": 0.0009915956969968624, + "loss": 0.4272, + "step": 404 + }, + { + "epoch": 0.022627594491158476, + "grad_norm": 0.828056812286377, + "learning_rate": 0.0009915676826535186, + "loss": 0.5512, + "step": 405 + }, + { + "epoch": 0.02268346509484035, + "grad_norm": 4.167536735534668, + "learning_rate": 0.0009915396683101748, + "loss": 0.4422, + "step": 406 + }, + { + "epoch": 0.022739335698522222, + "grad_norm": 0.662735641002655, + "learning_rate": 0.000991511653966831, + "loss": 0.4676, + "step": 407 + }, + { + "epoch": 0.022795206302204097, + "grad_norm": 0.8873198628425598, + "learning_rate": 0.0009914836396234872, + "loss": 0.4414, + "step": 408 + }, + { + "epoch": 0.02285107690588597, + "grad_norm": 0.8939672112464905, + "learning_rate": 0.0009914556252801434, + "loss": 0.5569, + "step": 409 + }, + { + "epoch": 0.02290694750956784, + "grad_norm": 0.779899001121521, + "learning_rate": 0.0009914276109367997, + "loss": 0.477, + "step": 410 + }, + { + "epoch": 0.022962818113249715, + "grad_norm": 0.7560564875602722, + "learning_rate": 0.0009913995965934559, + "loss": 0.5088, + "step": 411 + }, + { + "epoch": 0.023018688716931586, + "grad_norm": 0.6532883644104004, + "learning_rate": 0.000991371582250112, + "loss": 0.4752, + "step": 412 + }, + { + "epoch": 0.02307455932061346, + "grad_norm": 0.8932944536209106, + "learning_rate": 0.0009913435679067683, + "loss": 0.4863, + "step": 413 + }, + { + "epoch": 0.023130429924295332, + "grad_norm": 0.6040872931480408, + "learning_rate": 0.0009913155535634245, + "loss": 0.5105, + "step": 414 + }, + { + "epoch": 0.023186300527977204, + "grad_norm": 1.1111233234405518, + "learning_rate": 0.0009912875392200807, + "loss": 0.5386, + "step": 415 + }, + { + "epoch": 0.02324217113165908, + "grad_norm": 0.7403465509414673, + "learning_rate": 0.0009912595248767369, + "loss": 0.5566, + "step": 416 + }, + { + "epoch": 0.02329804173534095, + "grad_norm": 0.7970167398452759, + "learning_rate": 0.000991231510533393, + "loss": 0.441, + "step": 417 + }, + { + "epoch": 0.023353912339022825, + "grad_norm": 2.00022554397583, + "learning_rate": 0.0009912034961900493, + "loss": 0.5311, + "step": 418 + }, + { + "epoch": 0.023409782942704696, + "grad_norm": 0.7867471575737, + "learning_rate": 0.0009911754818467055, + "loss": 0.6134, + "step": 419 + }, + { + "epoch": 0.023465653546386567, + "grad_norm": 0.5354019403457642, + "learning_rate": 0.0009911474675033617, + "loss": 0.5491, + "step": 420 + }, + { + "epoch": 0.023521524150068442, + "grad_norm": 0.7504967451095581, + "learning_rate": 0.000991119453160018, + "loss": 0.6183, + "step": 421 + }, + { + "epoch": 0.023577394753750314, + "grad_norm": 0.8208183646202087, + "learning_rate": 0.0009910914388166741, + "loss": 0.548, + "step": 422 + }, + { + "epoch": 0.02363326535743219, + "grad_norm": 4.468395233154297, + "learning_rate": 0.0009910634244733303, + "loss": 0.5213, + "step": 423 + }, + { + "epoch": 0.02368913596111406, + "grad_norm": 0.763007402420044, + "learning_rate": 0.0009910354101299865, + "loss": 0.5413, + "step": 424 + }, + { + "epoch": 0.02374500656479593, + "grad_norm": 0.790843665599823, + "learning_rate": 0.0009910073957866427, + "loss": 0.6547, + "step": 425 + }, + { + "epoch": 0.023800877168477806, + "grad_norm": 0.6620511412620544, + "learning_rate": 0.000990979381443299, + "loss": 0.7021, + "step": 426 + }, + { + "epoch": 0.023856747772159678, + "grad_norm": 1.4064428806304932, + "learning_rate": 0.0009909513670999551, + "loss": 0.6632, + "step": 427 + }, + { + "epoch": 0.023912618375841552, + "grad_norm": 0.7614513039588928, + "learning_rate": 0.0009909233527566114, + "loss": 0.4314, + "step": 428 + }, + { + "epoch": 0.023968488979523424, + "grad_norm": 0.7470812797546387, + "learning_rate": 0.0009908953384132676, + "loss": 0.5311, + "step": 429 + }, + { + "epoch": 0.024024359583205295, + "grad_norm": 0.7772524356842041, + "learning_rate": 0.0009908673240699238, + "loss": 0.4648, + "step": 430 + }, + { + "epoch": 0.02408023018688717, + "grad_norm": 1.6262376308441162, + "learning_rate": 0.00099083930972658, + "loss": 0.6859, + "step": 431 + }, + { + "epoch": 0.02413610079056904, + "grad_norm": 1.0102685689926147, + "learning_rate": 0.0009908112953832362, + "loss": 0.5819, + "step": 432 + }, + { + "epoch": 0.024191971394250916, + "grad_norm": 0.5961595177650452, + "learning_rate": 0.0009907832810398924, + "loss": 0.4765, + "step": 433 + }, + { + "epoch": 0.024247841997932788, + "grad_norm": 0.9459435343742371, + "learning_rate": 0.0009907552666965486, + "loss": 0.5555, + "step": 434 + }, + { + "epoch": 0.02430371260161466, + "grad_norm": 0.6126872301101685, + "learning_rate": 0.0009907272523532048, + "loss": 0.556, + "step": 435 + }, + { + "epoch": 0.024359583205296534, + "grad_norm": 0.6411193013191223, + "learning_rate": 0.000990699238009861, + "loss": 0.5111, + "step": 436 + }, + { + "epoch": 0.024415453808978405, + "grad_norm": 0.5888941884040833, + "learning_rate": 0.0009906712236665172, + "loss": 0.4962, + "step": 437 + }, + { + "epoch": 0.02447132441266028, + "grad_norm": 2.4857442378997803, + "learning_rate": 0.0009906432093231736, + "loss": 0.507, + "step": 438 + }, + { + "epoch": 0.02452719501634215, + "grad_norm": 0.7376643419265747, + "learning_rate": 0.0009906151949798296, + "loss": 0.4503, + "step": 439 + }, + { + "epoch": 0.024583065620024023, + "grad_norm": 0.519430935382843, + "learning_rate": 0.0009905871806364858, + "loss": 0.6127, + "step": 440 + }, + { + "epoch": 0.024638936223705898, + "grad_norm": 0.6996128559112549, + "learning_rate": 0.000990559166293142, + "loss": 0.569, + "step": 441 + }, + { + "epoch": 0.02469480682738777, + "grad_norm": 0.6490321755409241, + "learning_rate": 0.0009905311519497985, + "loss": 0.507, + "step": 442 + }, + { + "epoch": 0.024750677431069644, + "grad_norm": 1.1695075035095215, + "learning_rate": 0.0009905031376064544, + "loss": 0.4559, + "step": 443 + }, + { + "epoch": 0.024806548034751515, + "grad_norm": 1.232068419456482, + "learning_rate": 0.0009904751232631106, + "loss": 0.4182, + "step": 444 + }, + { + "epoch": 0.024862418638433387, + "grad_norm": 1.1061275005340576, + "learning_rate": 0.000990447108919767, + "loss": 0.4732, + "step": 445 + }, + { + "epoch": 0.02491828924211526, + "grad_norm": 0.5274024605751038, + "learning_rate": 0.000990419094576423, + "loss": 0.5174, + "step": 446 + }, + { + "epoch": 0.024974159845797133, + "grad_norm": 0.9064426422119141, + "learning_rate": 0.0009903910802330793, + "loss": 0.632, + "step": 447 + }, + { + "epoch": 0.025030030449479008, + "grad_norm": 0.5154435634613037, + "learning_rate": 0.0009903630658897355, + "loss": 0.4911, + "step": 448 + }, + { + "epoch": 0.02508590105316088, + "grad_norm": 0.7637200951576233, + "learning_rate": 0.000990335051546392, + "loss": 0.5767, + "step": 449 + }, + { + "epoch": 0.02514177165684275, + "grad_norm": 0.5724984407424927, + "learning_rate": 0.0009903070372030479, + "loss": 0.5421, + "step": 450 + }, + { + "epoch": 0.025197642260524625, + "grad_norm": 1.1182059049606323, + "learning_rate": 0.000990279022859704, + "loss": 0.5165, + "step": 451 + }, + { + "epoch": 0.025253512864206497, + "grad_norm": 6.913763999938965, + "learning_rate": 0.0009902510085163605, + "loss": 0.6401, + "step": 452 + }, + { + "epoch": 0.02530938346788837, + "grad_norm": 0.7238820791244507, + "learning_rate": 0.0009902229941730167, + "loss": 0.5162, + "step": 453 + }, + { + "epoch": 0.025365254071570243, + "grad_norm": 0.7434768080711365, + "learning_rate": 0.0009901949798296727, + "loss": 0.4936, + "step": 454 + }, + { + "epoch": 0.025421124675252114, + "grad_norm": 2.833656072616577, + "learning_rate": 0.000990166965486329, + "loss": 0.6008, + "step": 455 + }, + { + "epoch": 0.02547699527893399, + "grad_norm": 0.6603851914405823, + "learning_rate": 0.0009901389511429853, + "loss": 0.599, + "step": 456 + }, + { + "epoch": 0.02553286588261586, + "grad_norm": 0.8363736867904663, + "learning_rate": 0.0009901109367996415, + "loss": 0.4487, + "step": 457 + }, + { + "epoch": 0.025588736486297736, + "grad_norm": 0.9036137461662292, + "learning_rate": 0.0009900829224562975, + "loss": 0.5126, + "step": 458 + }, + { + "epoch": 0.025644607089979607, + "grad_norm": 0.733745813369751, + "learning_rate": 0.000990054908112954, + "loss": 0.5034, + "step": 459 + }, + { + "epoch": 0.02570047769366148, + "grad_norm": 0.6946021914482117, + "learning_rate": 0.0009900268937696102, + "loss": 0.6208, + "step": 460 + }, + { + "epoch": 0.025756348297343353, + "grad_norm": 0.8705863356590271, + "learning_rate": 0.0009899988794262664, + "loss": 0.3802, + "step": 461 + }, + { + "epoch": 0.025812218901025225, + "grad_norm": 0.8276954889297485, + "learning_rate": 0.0009899708650829224, + "loss": 0.5003, + "step": 462 + }, + { + "epoch": 0.0258680895047071, + "grad_norm": 3.32192063331604, + "learning_rate": 0.0009899428507395788, + "loss": 0.4402, + "step": 463 + }, + { + "epoch": 0.02592396010838897, + "grad_norm": 0.8093612194061279, + "learning_rate": 0.000989914836396235, + "loss": 0.4831, + "step": 464 + }, + { + "epoch": 0.025979830712070846, + "grad_norm": 0.9423503279685974, + "learning_rate": 0.0009898868220528912, + "loss": 0.6204, + "step": 465 + }, + { + "epoch": 0.026035701315752717, + "grad_norm": 0.9296507835388184, + "learning_rate": 0.0009898588077095472, + "loss": 0.6327, + "step": 466 + }, + { + "epoch": 0.02609157191943459, + "grad_norm": 0.6413669586181641, + "learning_rate": 0.0009898307933662036, + "loss": 0.4829, + "step": 467 + }, + { + "epoch": 0.026147442523116463, + "grad_norm": 1.2893682718276978, + "learning_rate": 0.0009898027790228598, + "loss": 0.6085, + "step": 468 + }, + { + "epoch": 0.026203313126798335, + "grad_norm": 1.0348926782608032, + "learning_rate": 0.0009897747646795158, + "loss": 0.6296, + "step": 469 + }, + { + "epoch": 0.02625918373048021, + "grad_norm": 1.6568584442138672, + "learning_rate": 0.0009897467503361722, + "loss": 0.4973, + "step": 470 + }, + { + "epoch": 0.02631505433416208, + "grad_norm": 0.5389439463615417, + "learning_rate": 0.0009897187359928284, + "loss": 0.4663, + "step": 471 + }, + { + "epoch": 0.026370924937843952, + "grad_norm": 0.6327576637268066, + "learning_rate": 0.0009896907216494846, + "loss": 0.4531, + "step": 472 + }, + { + "epoch": 0.026426795541525827, + "grad_norm": 0.7864127159118652, + "learning_rate": 0.0009896627073061406, + "loss": 0.4729, + "step": 473 + }, + { + "epoch": 0.0264826661452077, + "grad_norm": 1.2841821908950806, + "learning_rate": 0.000989634692962797, + "loss": 0.5318, + "step": 474 + }, + { + "epoch": 0.026538536748889573, + "grad_norm": 0.7219766974449158, + "learning_rate": 0.0009896066786194532, + "loss": 0.4962, + "step": 475 + }, + { + "epoch": 0.026594407352571445, + "grad_norm": 0.5780172944068909, + "learning_rate": 0.0009895786642761095, + "loss": 0.6065, + "step": 476 + }, + { + "epoch": 0.026650277956253316, + "grad_norm": 0.5112885236740112, + "learning_rate": 0.0009895506499327657, + "loss": 0.4559, + "step": 477 + }, + { + "epoch": 0.02670614855993519, + "grad_norm": 0.5006769299507141, + "learning_rate": 0.0009895226355894219, + "loss": 0.4866, + "step": 478 + }, + { + "epoch": 0.026762019163617062, + "grad_norm": 1.187087893486023, + "learning_rate": 0.000989494621246078, + "loss": 0.5037, + "step": 479 + }, + { + "epoch": 0.026817889767298937, + "grad_norm": 0.703309178352356, + "learning_rate": 0.0009894666069027343, + "loss": 0.5502, + "step": 480 + }, + { + "epoch": 0.02687376037098081, + "grad_norm": 0.9246487617492676, + "learning_rate": 0.0009894385925593905, + "loss": 0.59, + "step": 481 + }, + { + "epoch": 0.02692963097466268, + "grad_norm": 0.572666347026825, + "learning_rate": 0.0009894105782160467, + "loss": 0.4983, + "step": 482 + }, + { + "epoch": 0.026985501578344555, + "grad_norm": 0.8652781248092651, + "learning_rate": 0.000989382563872703, + "loss": 0.5408, + "step": 483 + }, + { + "epoch": 0.027041372182026426, + "grad_norm": 0.8590535521507263, + "learning_rate": 0.000989354549529359, + "loss": 0.574, + "step": 484 + }, + { + "epoch": 0.0270972427857083, + "grad_norm": 3.0314905643463135, + "learning_rate": 0.0009893265351860153, + "loss": 0.4718, + "step": 485 + }, + { + "epoch": 0.027153113389390172, + "grad_norm": 0.9149683713912964, + "learning_rate": 0.0009892985208426715, + "loss": 0.437, + "step": 486 + }, + { + "epoch": 0.027208983993072044, + "grad_norm": 0.7310080528259277, + "learning_rate": 0.0009892705064993277, + "loss": 0.6704, + "step": 487 + }, + { + "epoch": 0.02726485459675392, + "grad_norm": 0.7403769493103027, + "learning_rate": 0.000989242492155984, + "loss": 0.4815, + "step": 488 + }, + { + "epoch": 0.02732072520043579, + "grad_norm": 0.9060088992118835, + "learning_rate": 0.0009892144778126401, + "loss": 0.5178, + "step": 489 + }, + { + "epoch": 0.027376595804117665, + "grad_norm": 1.0428775548934937, + "learning_rate": 0.0009891864634692963, + "loss": 0.6216, + "step": 490 + }, + { + "epoch": 0.027432466407799536, + "grad_norm": 2.348588466644287, + "learning_rate": 0.0009891584491259525, + "loss": 0.4854, + "step": 491 + }, + { + "epoch": 0.027488337011481408, + "grad_norm": 1.0244497060775757, + "learning_rate": 0.0009891304347826087, + "loss": 0.5599, + "step": 492 + }, + { + "epoch": 0.027544207615163283, + "grad_norm": 0.6614739298820496, + "learning_rate": 0.000989102420439265, + "loss": 0.5327, + "step": 493 + }, + { + "epoch": 0.027600078218845154, + "grad_norm": 0.7783992886543274, + "learning_rate": 0.0009890744060959212, + "loss": 0.5485, + "step": 494 + }, + { + "epoch": 0.02765594882252703, + "grad_norm": 0.7415665984153748, + "learning_rate": 0.0009890463917525774, + "loss": 0.6196, + "step": 495 + }, + { + "epoch": 0.0277118194262089, + "grad_norm": 0.8095999956130981, + "learning_rate": 0.0009890183774092336, + "loss": 0.6225, + "step": 496 + }, + { + "epoch": 0.02776769002989077, + "grad_norm": 0.460873544216156, + "learning_rate": 0.0009889903630658898, + "loss": 0.4459, + "step": 497 + }, + { + "epoch": 0.027823560633572646, + "grad_norm": 1.1751947402954102, + "learning_rate": 0.000988962348722546, + "loss": 0.4287, + "step": 498 + }, + { + "epoch": 0.027879431237254518, + "grad_norm": 1.2554341554641724, + "learning_rate": 0.0009889343343792022, + "loss": 0.7465, + "step": 499 + }, + { + "epoch": 0.027935301840936393, + "grad_norm": 1.1855918169021606, + "learning_rate": 0.0009889063200358584, + "loss": 0.5183, + "step": 500 + }, + { + "epoch": 0.027935301840936393, + "eval_cer": 0.10514658549092774, + "eval_loss": 0.3931924104690552, + "eval_runtime": 56.8868, + "eval_samples_per_second": 79.772, + "eval_steps_per_second": 4.992, + "eval_wer": 0.4076143124493729, + "step": 500 + }, + { + "epoch": 0.027991172444618264, + "grad_norm": 0.842576801776886, + "learning_rate": 0.0009888783056925146, + "loss": 0.6563, + "step": 501 + }, + { + "epoch": 0.028047043048300135, + "grad_norm": 1.2841837406158447, + "learning_rate": 0.0009888502913491708, + "loss": 0.538, + "step": 502 + }, + { + "epoch": 0.02810291365198201, + "grad_norm": 1.0915876626968384, + "learning_rate": 0.000988822277005827, + "loss": 0.5889, + "step": 503 + }, + { + "epoch": 0.02815878425566388, + "grad_norm": 0.5407107472419739, + "learning_rate": 0.0009887942626624832, + "loss": 0.5338, + "step": 504 + }, + { + "epoch": 0.028214654859345756, + "grad_norm": 2.611922264099121, + "learning_rate": 0.0009887662483191394, + "loss": 0.4323, + "step": 505 + }, + { + "epoch": 0.028270525463027628, + "grad_norm": 0.9005200862884521, + "learning_rate": 0.0009887382339757956, + "loss": 0.5613, + "step": 506 + }, + { + "epoch": 0.0283263960667095, + "grad_norm": 11.402393341064453, + "learning_rate": 0.0009887102196324518, + "loss": 0.5303, + "step": 507 + }, + { + "epoch": 0.028382266670391374, + "grad_norm": 0.8339524269104004, + "learning_rate": 0.000988682205289108, + "loss": 0.5597, + "step": 508 + }, + { + "epoch": 0.028438137274073245, + "grad_norm": 1.0863631963729858, + "learning_rate": 0.0009886541909457642, + "loss": 0.5897, + "step": 509 + }, + { + "epoch": 0.02849400787775512, + "grad_norm": 0.9381932616233826, + "learning_rate": 0.0009886261766024204, + "loss": 0.6079, + "step": 510 + }, + { + "epoch": 0.02854987848143699, + "grad_norm": 0.6055160760879517, + "learning_rate": 0.0009885981622590767, + "loss": 0.5327, + "step": 511 + }, + { + "epoch": 0.028605749085118863, + "grad_norm": 0.6289492249488831, + "learning_rate": 0.0009885701479157329, + "loss": 0.5302, + "step": 512 + }, + { + "epoch": 0.028661619688800738, + "grad_norm": 0.8781564831733704, + "learning_rate": 0.000988542133572389, + "loss": 0.5028, + "step": 513 + }, + { + "epoch": 0.02871749029248261, + "grad_norm": 0.6210728287696838, + "learning_rate": 0.0009885141192290453, + "loss": 0.6551, + "step": 514 + }, + { + "epoch": 0.028773360896164484, + "grad_norm": 0.7236417531967163, + "learning_rate": 0.0009884861048857015, + "loss": 0.4839, + "step": 515 + }, + { + "epoch": 0.028829231499846356, + "grad_norm": 0.560756504535675, + "learning_rate": 0.0009884580905423577, + "loss": 0.4592, + "step": 516 + }, + { + "epoch": 0.028885102103528227, + "grad_norm": 0.7263710498809814, + "learning_rate": 0.0009884300761990139, + "loss": 0.4656, + "step": 517 + }, + { + "epoch": 0.028940972707210102, + "grad_norm": 0.47448769211769104, + "learning_rate": 0.00098840206185567, + "loss": 0.4824, + "step": 518 + }, + { + "epoch": 0.028996843310891973, + "grad_norm": 1.94016695022583, + "learning_rate": 0.0009883740475123263, + "loss": 0.5508, + "step": 519 + }, + { + "epoch": 0.029052713914573848, + "grad_norm": 0.931218147277832, + "learning_rate": 0.0009883460331689825, + "loss": 0.3945, + "step": 520 + }, + { + "epoch": 0.02910858451825572, + "grad_norm": 0.8405587673187256, + "learning_rate": 0.0009883180188256387, + "loss": 0.5358, + "step": 521 + }, + { + "epoch": 0.02916445512193759, + "grad_norm": 0.7076135277748108, + "learning_rate": 0.000988290004482295, + "loss": 0.645, + "step": 522 + }, + { + "epoch": 0.029220325725619466, + "grad_norm": 1.8663055896759033, + "learning_rate": 0.0009882619901389511, + "loss": 0.6237, + "step": 523 + }, + { + "epoch": 0.029276196329301337, + "grad_norm": 4.486104965209961, + "learning_rate": 0.0009882339757956073, + "loss": 0.5562, + "step": 524 + }, + { + "epoch": 0.029332066932983212, + "grad_norm": 0.8549737334251404, + "learning_rate": 0.0009882059614522635, + "loss": 0.4325, + "step": 525 + }, + { + "epoch": 0.029387937536665083, + "grad_norm": 0.7775002121925354, + "learning_rate": 0.0009881779471089197, + "loss": 0.556, + "step": 526 + }, + { + "epoch": 0.029443808140346958, + "grad_norm": 0.9117982387542725, + "learning_rate": 0.000988149932765576, + "loss": 0.5686, + "step": 527 + }, + { + "epoch": 0.02949967874402883, + "grad_norm": 7.140747547149658, + "learning_rate": 0.0009881219184222322, + "loss": 0.5408, + "step": 528 + }, + { + "epoch": 0.0295555493477107, + "grad_norm": 1.3012076616287231, + "learning_rate": 0.0009880939040788884, + "loss": 0.5647, + "step": 529 + }, + { + "epoch": 0.029611419951392576, + "grad_norm": 1.4608500003814697, + "learning_rate": 0.0009880658897355446, + "loss": 0.701, + "step": 530 + }, + { + "epoch": 0.029667290555074447, + "grad_norm": 0.617730438709259, + "learning_rate": 0.0009880378753922008, + "loss": 0.5954, + "step": 531 + }, + { + "epoch": 0.029723161158756322, + "grad_norm": 0.564232587814331, + "learning_rate": 0.000988009861048857, + "loss": 0.5382, + "step": 532 + }, + { + "epoch": 0.029779031762438193, + "grad_norm": 0.7241876125335693, + "learning_rate": 0.0009879818467055132, + "loss": 0.4669, + "step": 533 + }, + { + "epoch": 0.029834902366120065, + "grad_norm": 1.163967490196228, + "learning_rate": 0.0009879538323621694, + "loss": 0.5469, + "step": 534 + }, + { + "epoch": 0.02989077296980194, + "grad_norm": 0.6273118257522583, + "learning_rate": 0.0009879258180188256, + "loss": 0.5506, + "step": 535 + }, + { + "epoch": 0.02994664357348381, + "grad_norm": 0.6903908252716064, + "learning_rate": 0.0009878978036754818, + "loss": 0.4962, + "step": 536 + }, + { + "epoch": 0.030002514177165686, + "grad_norm": 1.1584815979003906, + "learning_rate": 0.000987869789332138, + "loss": 0.5599, + "step": 537 + }, + { + "epoch": 0.030058384780847557, + "grad_norm": 4.513985633850098, + "learning_rate": 0.0009878417749887942, + "loss": 0.5082, + "step": 538 + }, + { + "epoch": 0.03011425538452943, + "grad_norm": 1.6019222736358643, + "learning_rate": 0.0009878137606454504, + "loss": 0.5464, + "step": 539 + }, + { + "epoch": 0.030170125988211303, + "grad_norm": 0.9352850914001465, + "learning_rate": 0.0009877857463021066, + "loss": 0.4336, + "step": 540 + }, + { + "epoch": 0.030225996591893175, + "grad_norm": 0.5031391382217407, + "learning_rate": 0.000987757731958763, + "loss": 0.4879, + "step": 541 + }, + { + "epoch": 0.03028186719557505, + "grad_norm": 0.7643775343894958, + "learning_rate": 0.000987729717615419, + "loss": 0.5414, + "step": 542 + }, + { + "epoch": 0.03033773779925692, + "grad_norm": 0.6326935291290283, + "learning_rate": 0.0009877017032720752, + "loss": 0.6251, + "step": 543 + }, + { + "epoch": 0.030393608402938792, + "grad_norm": 0.6240907311439514, + "learning_rate": 0.0009876736889287314, + "loss": 0.4691, + "step": 544 + }, + { + "epoch": 0.030449479006620667, + "grad_norm": 1.270589828491211, + "learning_rate": 0.0009876456745853879, + "loss": 0.6843, + "step": 545 + }, + { + "epoch": 0.03050534961030254, + "grad_norm": 0.568419873714447, + "learning_rate": 0.0009876176602420439, + "loss": 0.5505, + "step": 546 + }, + { + "epoch": 0.030561220213984414, + "grad_norm": 1.2199102640151978, + "learning_rate": 0.0009875896458987, + "loss": 0.5195, + "step": 547 + }, + { + "epoch": 0.030617090817666285, + "grad_norm": 0.7663665413856506, + "learning_rate": 0.0009875616315553565, + "loss": 0.5583, + "step": 548 + }, + { + "epoch": 0.030672961421348156, + "grad_norm": 1.6576004028320312, + "learning_rate": 0.0009875336172120127, + "loss": 0.4748, + "step": 549 + }, + { + "epoch": 0.03072883202503003, + "grad_norm": 0.8576673269271851, + "learning_rate": 0.0009875056028686687, + "loss": 0.6405, + "step": 550 + }, + { + "epoch": 0.030784702628711903, + "grad_norm": 0.7192466259002686, + "learning_rate": 0.0009874775885253249, + "loss": 0.5482, + "step": 551 + }, + { + "epoch": 0.030840573232393777, + "grad_norm": 0.6524004936218262, + "learning_rate": 0.0009874495741819813, + "loss": 0.5399, + "step": 552 + }, + { + "epoch": 0.03089644383607565, + "grad_norm": 0.9522773027420044, + "learning_rate": 0.0009874215598386373, + "loss": 0.5315, + "step": 553 + }, + { + "epoch": 0.03095231443975752, + "grad_norm": 0.8997712135314941, + "learning_rate": 0.0009873935454952935, + "loss": 0.6169, + "step": 554 + }, + { + "epoch": 0.031008185043439395, + "grad_norm": 0.6301531791687012, + "learning_rate": 0.00098736553115195, + "loss": 0.4731, + "step": 555 + }, + { + "epoch": 0.031064055647121266, + "grad_norm": 1.1418601274490356, + "learning_rate": 0.0009873375168086061, + "loss": 0.5548, + "step": 556 + }, + { + "epoch": 0.03111992625080314, + "grad_norm": 0.5795314908027649, + "learning_rate": 0.0009873095024652621, + "loss": 0.4842, + "step": 557 + }, + { + "epoch": 0.031175796854485013, + "grad_norm": 1.8990386724472046, + "learning_rate": 0.0009872814881219183, + "loss": 0.5286, + "step": 558 + }, + { + "epoch": 0.031231667458166884, + "grad_norm": 0.7824792861938477, + "learning_rate": 0.0009872534737785748, + "loss": 0.6088, + "step": 559 + }, + { + "epoch": 0.03128753806184876, + "grad_norm": 0.94327712059021, + "learning_rate": 0.000987225459435231, + "loss": 0.6377, + "step": 560 + }, + { + "epoch": 0.031343408665530634, + "grad_norm": 0.7613363862037659, + "learning_rate": 0.000987197445091887, + "loss": 0.4629, + "step": 561 + }, + { + "epoch": 0.0313992792692125, + "grad_norm": 7.961323261260986, + "learning_rate": 0.0009871694307485434, + "loss": 0.4978, + "step": 562 + }, + { + "epoch": 0.031455149872894377, + "grad_norm": 1.000241756439209, + "learning_rate": 0.0009871414164051996, + "loss": 0.6144, + "step": 563 + }, + { + "epoch": 0.03151102047657625, + "grad_norm": 2.7467691898345947, + "learning_rate": 0.0009871134020618558, + "loss": 0.5185, + "step": 564 + }, + { + "epoch": 0.03156689108025812, + "grad_norm": 0.5254828929901123, + "learning_rate": 0.0009870853877185118, + "loss": 0.4154, + "step": 565 + }, + { + "epoch": 0.031622761683939994, + "grad_norm": 1.618910789489746, + "learning_rate": 0.0009870573733751682, + "loss": 0.4915, + "step": 566 + }, + { + "epoch": 0.03167863228762187, + "grad_norm": 1.469223141670227, + "learning_rate": 0.0009870293590318244, + "loss": 0.5752, + "step": 567 + }, + { + "epoch": 0.031734502891303744, + "grad_norm": 0.6660300493240356, + "learning_rate": 0.0009870013446884806, + "loss": 0.4804, + "step": 568 + }, + { + "epoch": 0.03179037349498561, + "grad_norm": 1.1157199144363403, + "learning_rate": 0.0009869733303451366, + "loss": 0.5777, + "step": 569 + }, + { + "epoch": 0.03184624409866749, + "grad_norm": 4.111578464508057, + "learning_rate": 0.000986945316001793, + "loss": 0.5368, + "step": 570 + }, + { + "epoch": 0.03190211470234936, + "grad_norm": 1.917514681816101, + "learning_rate": 0.0009869173016584492, + "loss": 0.5515, + "step": 571 + }, + { + "epoch": 0.03195798530603123, + "grad_norm": 0.7474856972694397, + "learning_rate": 0.0009868892873151052, + "loss": 0.517, + "step": 572 + }, + { + "epoch": 0.032013855909713104, + "grad_norm": 1.0857408046722412, + "learning_rate": 0.0009868612729717616, + "loss": 0.4955, + "step": 573 + }, + { + "epoch": 0.03206972651339498, + "grad_norm": 1.0976957082748413, + "learning_rate": 0.0009868332586284178, + "loss": 0.7505, + "step": 574 + }, + { + "epoch": 0.03212559711707685, + "grad_norm": 0.6001850366592407, + "learning_rate": 0.000986805244285074, + "loss": 0.4399, + "step": 575 + }, + { + "epoch": 0.03218146772075872, + "grad_norm": 1.1638317108154297, + "learning_rate": 0.00098677722994173, + "loss": 0.4824, + "step": 576 + }, + { + "epoch": 0.0322373383244406, + "grad_norm": 1.118512511253357, + "learning_rate": 0.0009867492155983865, + "loss": 0.5636, + "step": 577 + }, + { + "epoch": 0.03229320892812247, + "grad_norm": 0.6995843052864075, + "learning_rate": 0.0009867212012550427, + "loss": 0.6268, + "step": 578 + }, + { + "epoch": 0.03234907953180434, + "grad_norm": 1.910476803779602, + "learning_rate": 0.0009866931869116989, + "loss": 0.467, + "step": 579 + }, + { + "epoch": 0.032404950135486214, + "grad_norm": 0.7815238833427429, + "learning_rate": 0.000986665172568355, + "loss": 0.4812, + "step": 580 + }, + { + "epoch": 0.03246082073916809, + "grad_norm": 0.970793604850769, + "learning_rate": 0.0009866371582250113, + "loss": 0.569, + "step": 581 + }, + { + "epoch": 0.03251669134284996, + "grad_norm": 1.2958917617797852, + "learning_rate": 0.0009866091438816675, + "loss": 0.5685, + "step": 582 + }, + { + "epoch": 0.03257256194653183, + "grad_norm": 0.6104224324226379, + "learning_rate": 0.0009865811295383237, + "loss": 0.6538, + "step": 583 + }, + { + "epoch": 0.03262843255021371, + "grad_norm": 0.698486864566803, + "learning_rate": 0.00098655311519498, + "loss": 0.5195, + "step": 584 + }, + { + "epoch": 0.032684303153895575, + "grad_norm": 1.2138582468032837, + "learning_rate": 0.000986525100851636, + "loss": 0.6175, + "step": 585 + }, + { + "epoch": 0.03274017375757745, + "grad_norm": 12.678753852844238, + "learning_rate": 0.0009864970865082923, + "loss": 0.566, + "step": 586 + }, + { + "epoch": 0.032796044361259324, + "grad_norm": 2.333786964416504, + "learning_rate": 0.0009864690721649485, + "loss": 0.5779, + "step": 587 + }, + { + "epoch": 0.0328519149649412, + "grad_norm": 1.2871074676513672, + "learning_rate": 0.0009864410578216047, + "loss": 0.4153, + "step": 588 + }, + { + "epoch": 0.03290778556862307, + "grad_norm": 0.6042351722717285, + "learning_rate": 0.000986413043478261, + "loss": 0.5, + "step": 589 + }, + { + "epoch": 0.03296365617230494, + "grad_norm": 3.173753023147583, + "learning_rate": 0.0009863850291349171, + "loss": 0.5551, + "step": 590 + }, + { + "epoch": 0.03301952677598682, + "grad_norm": 1.3819231986999512, + "learning_rate": 0.0009863570147915733, + "loss": 0.6111, + "step": 591 + }, + { + "epoch": 0.033075397379668685, + "grad_norm": 0.6694918274879456, + "learning_rate": 0.0009863290004482295, + "loss": 0.7269, + "step": 592 + }, + { + "epoch": 0.03313126798335056, + "grad_norm": 0.6633431315422058, + "learning_rate": 0.0009863009861048857, + "loss": 0.4563, + "step": 593 + }, + { + "epoch": 0.033187138587032435, + "grad_norm": 0.9386100769042969, + "learning_rate": 0.000986272971761542, + "loss": 0.536, + "step": 594 + }, + { + "epoch": 0.0332430091907143, + "grad_norm": 0.93766188621521, + "learning_rate": 0.0009862449574181982, + "loss": 0.6241, + "step": 595 + }, + { + "epoch": 0.03329887979439618, + "grad_norm": 0.4910685122013092, + "learning_rate": 0.0009862169430748544, + "loss": 0.52, + "step": 596 + }, + { + "epoch": 0.03335475039807805, + "grad_norm": 0.5772466659545898, + "learning_rate": 0.0009861889287315106, + "loss": 0.5315, + "step": 597 + }, + { + "epoch": 0.03341062100175993, + "grad_norm": 0.9905041456222534, + "learning_rate": 0.0009861609143881668, + "loss": 0.4999, + "step": 598 + }, + { + "epoch": 0.033466491605441795, + "grad_norm": 0.6076198816299438, + "learning_rate": 0.000986132900044823, + "loss": 0.5281, + "step": 599 + }, + { + "epoch": 0.03352236220912367, + "grad_norm": 0.7989510893821716, + "learning_rate": 0.0009861048857014792, + "loss": 0.6406, + "step": 600 + }, + { + "epoch": 0.033578232812805545, + "grad_norm": 0.8485199213027954, + "learning_rate": 0.0009860768713581354, + "loss": 0.5499, + "step": 601 + }, + { + "epoch": 0.03363410341648741, + "grad_norm": 1.0555579662322998, + "learning_rate": 0.0009860488570147916, + "loss": 0.4938, + "step": 602 + }, + { + "epoch": 0.03368997402016929, + "grad_norm": 0.6512682437896729, + "learning_rate": 0.0009860208426714478, + "loss": 0.5477, + "step": 603 + }, + { + "epoch": 0.03374584462385116, + "grad_norm": 1.121580719947815, + "learning_rate": 0.000985992828328104, + "loss": 0.5494, + "step": 604 + }, + { + "epoch": 0.03380171522753303, + "grad_norm": 0.5266908407211304, + "learning_rate": 0.0009859648139847602, + "loss": 0.4283, + "step": 605 + }, + { + "epoch": 0.033857585831214905, + "grad_norm": 0.8270426988601685, + "learning_rate": 0.0009859367996414164, + "loss": 0.572, + "step": 606 + }, + { + "epoch": 0.03391345643489678, + "grad_norm": 0.6584266424179077, + "learning_rate": 0.0009859087852980726, + "loss": 0.4629, + "step": 607 + }, + { + "epoch": 0.033969327038578655, + "grad_norm": 0.580382764339447, + "learning_rate": 0.0009858807709547288, + "loss": 0.3871, + "step": 608 + }, + { + "epoch": 0.03402519764226052, + "grad_norm": 0.731745183467865, + "learning_rate": 0.000985852756611385, + "loss": 0.3895, + "step": 609 + }, + { + "epoch": 0.0340810682459424, + "grad_norm": 0.6504466533660889, + "learning_rate": 0.0009858247422680412, + "loss": 0.5145, + "step": 610 + }, + { + "epoch": 0.03413693884962427, + "grad_norm": 1.0467389822006226, + "learning_rate": 0.0009857967279246975, + "loss": 0.4776, + "step": 611 + }, + { + "epoch": 0.03419280945330614, + "grad_norm": 3.6105713844299316, + "learning_rate": 0.0009857687135813537, + "loss": 0.5615, + "step": 612 + }, + { + "epoch": 0.034248680056988015, + "grad_norm": 0.7290234565734863, + "learning_rate": 0.0009857406992380099, + "loss": 0.6948, + "step": 613 + }, + { + "epoch": 0.03430455066066989, + "grad_norm": 2.6561849117279053, + "learning_rate": 0.000985712684894666, + "loss": 0.6607, + "step": 614 + }, + { + "epoch": 0.03436042126435176, + "grad_norm": 1.0816316604614258, + "learning_rate": 0.0009856846705513223, + "loss": 0.5326, + "step": 615 + }, + { + "epoch": 0.03441629186803363, + "grad_norm": 1.033316731452942, + "learning_rate": 0.0009856566562079785, + "loss": 0.572, + "step": 616 + }, + { + "epoch": 0.03447216247171551, + "grad_norm": 0.5817947387695312, + "learning_rate": 0.0009856286418646347, + "loss": 0.4693, + "step": 617 + }, + { + "epoch": 0.03452803307539738, + "grad_norm": 0.5991920232772827, + "learning_rate": 0.000985600627521291, + "loss": 0.5119, + "step": 618 + }, + { + "epoch": 0.03458390367907925, + "grad_norm": 0.55665123462677, + "learning_rate": 0.000985572613177947, + "loss": 0.4467, + "step": 619 + }, + { + "epoch": 0.034639774282761125, + "grad_norm": 0.6055389046669006, + "learning_rate": 0.0009855445988346033, + "loss": 0.5236, + "step": 620 + }, + { + "epoch": 0.034695644886443, + "grad_norm": 0.7571671009063721, + "learning_rate": 0.0009855165844912595, + "loss": 0.6651, + "step": 621 + }, + { + "epoch": 0.03475151549012487, + "grad_norm": 0.6363291144371033, + "learning_rate": 0.0009854885701479157, + "loss": 0.4738, + "step": 622 + }, + { + "epoch": 0.03480738609380674, + "grad_norm": 0.7915277481079102, + "learning_rate": 0.000985460555804572, + "loss": 0.459, + "step": 623 + }, + { + "epoch": 0.03486325669748862, + "grad_norm": 0.486939936876297, + "learning_rate": 0.0009854325414612281, + "loss": 0.4536, + "step": 624 + }, + { + "epoch": 0.03491912730117049, + "grad_norm": 0.7784373164176941, + "learning_rate": 0.0009854045271178843, + "loss": 0.5497, + "step": 625 + }, + { + "epoch": 0.03497499790485236, + "grad_norm": 1.731570839881897, + "learning_rate": 0.0009853765127745405, + "loss": 0.567, + "step": 626 + }, + { + "epoch": 0.035030868508534235, + "grad_norm": 1.9897339344024658, + "learning_rate": 0.0009853484984311967, + "loss": 0.5137, + "step": 627 + }, + { + "epoch": 0.03508673911221611, + "grad_norm": 0.8557936549186707, + "learning_rate": 0.000985320484087853, + "loss": 0.5996, + "step": 628 + }, + { + "epoch": 0.03514260971589798, + "grad_norm": 0.4640888571739197, + "learning_rate": 0.0009852924697445092, + "loss": 0.5943, + "step": 629 + }, + { + "epoch": 0.03519848031957985, + "grad_norm": 0.8642069697380066, + "learning_rate": 0.0009852644554011654, + "loss": 0.4665, + "step": 630 + }, + { + "epoch": 0.03525435092326173, + "grad_norm": 1.4095985889434814, + "learning_rate": 0.0009852364410578216, + "loss": 0.6657, + "step": 631 + }, + { + "epoch": 0.035310221526943596, + "grad_norm": 0.6623591780662537, + "learning_rate": 0.0009852084267144778, + "loss": 0.4762, + "step": 632 + }, + { + "epoch": 0.03536609213062547, + "grad_norm": 0.5406720042228699, + "learning_rate": 0.0009851804123711342, + "loss": 0.5103, + "step": 633 + }, + { + "epoch": 0.035421962734307345, + "grad_norm": 1.4970476627349854, + "learning_rate": 0.0009851523980277902, + "loss": 0.5532, + "step": 634 + }, + { + "epoch": 0.03547783333798922, + "grad_norm": 2.7033119201660156, + "learning_rate": 0.0009851243836844464, + "loss": 0.4928, + "step": 635 + }, + { + "epoch": 0.03553370394167109, + "grad_norm": 0.4479169547557831, + "learning_rate": 0.0009850963693411026, + "loss": 0.4585, + "step": 636 + }, + { + "epoch": 0.03558957454535296, + "grad_norm": 0.6719197034835815, + "learning_rate": 0.0009850683549977588, + "loss": 0.5106, + "step": 637 + }, + { + "epoch": 0.03564544514903484, + "grad_norm": 0.9518922567367554, + "learning_rate": 0.000985040340654415, + "loss": 0.6621, + "step": 638 + }, + { + "epoch": 0.035701315752716706, + "grad_norm": 0.535654604434967, + "learning_rate": 0.0009850123263110712, + "loss": 0.467, + "step": 639 + }, + { + "epoch": 0.03575718635639858, + "grad_norm": 0.6412437558174133, + "learning_rate": 0.0009849843119677276, + "loss": 0.5462, + "step": 640 + }, + { + "epoch": 0.035813056960080455, + "grad_norm": 0.8266791105270386, + "learning_rate": 0.0009849562976243836, + "loss": 0.5359, + "step": 641 + }, + { + "epoch": 0.03586892756376232, + "grad_norm": 0.6186512112617493, + "learning_rate": 0.0009849282832810398, + "loss": 0.5381, + "step": 642 + }, + { + "epoch": 0.0359247981674442, + "grad_norm": 0.9557093381881714, + "learning_rate": 0.000984900268937696, + "loss": 0.4615, + "step": 643 + }, + { + "epoch": 0.03598066877112607, + "grad_norm": 2.9302899837493896, + "learning_rate": 0.0009848722545943525, + "loss": 0.4845, + "step": 644 + }, + { + "epoch": 0.03603653937480795, + "grad_norm": 0.7893548011779785, + "learning_rate": 0.0009848442402510085, + "loss": 0.6745, + "step": 645 + }, + { + "epoch": 0.036092409978489816, + "grad_norm": 0.5956614017486572, + "learning_rate": 0.0009848162259076647, + "loss": 0.4633, + "step": 646 + }, + { + "epoch": 0.03614828058217169, + "grad_norm": 1.0337514877319336, + "learning_rate": 0.0009847882115643209, + "loss": 0.4594, + "step": 647 + }, + { + "epoch": 0.036204151185853566, + "grad_norm": 0.7182970643043518, + "learning_rate": 0.0009847601972209773, + "loss": 0.5094, + "step": 648 + }, + { + "epoch": 0.036260021789535433, + "grad_norm": 0.6105180382728577, + "learning_rate": 0.0009847321828776333, + "loss": 0.536, + "step": 649 + }, + { + "epoch": 0.03631589239321731, + "grad_norm": 0.5588029026985168, + "learning_rate": 0.0009847041685342895, + "loss": 0.4846, + "step": 650 + }, + { + "epoch": 0.03637176299689918, + "grad_norm": 0.4398474097251892, + "learning_rate": 0.000984676154190946, + "loss": 0.4312, + "step": 651 + }, + { + "epoch": 0.03642763360058105, + "grad_norm": 24.086254119873047, + "learning_rate": 0.0009846481398476021, + "loss": 0.4817, + "step": 652 + }, + { + "epoch": 0.036483504204262926, + "grad_norm": 6.349640846252441, + "learning_rate": 0.000984620125504258, + "loss": 0.5578, + "step": 653 + }, + { + "epoch": 0.0365393748079448, + "grad_norm": 0.9662876725196838, + "learning_rate": 0.0009845921111609143, + "loss": 0.6038, + "step": 654 + }, + { + "epoch": 0.036595245411626676, + "grad_norm": 0.704684317111969, + "learning_rate": 0.0009845640968175707, + "loss": 0.4993, + "step": 655 + }, + { + "epoch": 0.036651116015308544, + "grad_norm": 0.6935668587684631, + "learning_rate": 0.000984536082474227, + "loss": 0.4846, + "step": 656 + }, + { + "epoch": 0.03670698661899042, + "grad_norm": 0.6569328904151917, + "learning_rate": 0.000984508068130883, + "loss": 0.5809, + "step": 657 + }, + { + "epoch": 0.03676285722267229, + "grad_norm": 3.512237787246704, + "learning_rate": 0.0009844800537875393, + "loss": 0.5208, + "step": 658 + }, + { + "epoch": 0.03681872782635416, + "grad_norm": 0.8686926960945129, + "learning_rate": 0.0009844520394441955, + "loss": 0.5019, + "step": 659 + }, + { + "epoch": 0.036874598430036036, + "grad_norm": 1.4448530673980713, + "learning_rate": 0.0009844240251008515, + "loss": 0.4909, + "step": 660 + }, + { + "epoch": 0.03693046903371791, + "grad_norm": 1.1722660064697266, + "learning_rate": 0.0009843960107575077, + "loss": 0.655, + "step": 661 + }, + { + "epoch": 0.03698633963739978, + "grad_norm": 0.6105414032936096, + "learning_rate": 0.0009843679964141642, + "loss": 0.5826, + "step": 662 + }, + { + "epoch": 0.037042210241081654, + "grad_norm": 1.5178064107894897, + "learning_rate": 0.0009843399820708204, + "loss": 0.5014, + "step": 663 + }, + { + "epoch": 0.03709808084476353, + "grad_norm": 0.7745649218559265, + "learning_rate": 0.0009843119677274764, + "loss": 0.583, + "step": 664 + }, + { + "epoch": 0.0371539514484454, + "grad_norm": 1.6383610963821411, + "learning_rate": 0.0009842839533841328, + "loss": 0.6096, + "step": 665 + }, + { + "epoch": 0.03720982205212727, + "grad_norm": 0.5085414052009583, + "learning_rate": 0.000984255939040789, + "loss": 0.465, + "step": 666 + }, + { + "epoch": 0.037265692655809146, + "grad_norm": 0.570478618144989, + "learning_rate": 0.0009842279246974452, + "loss": 0.3765, + "step": 667 + }, + { + "epoch": 0.03732156325949102, + "grad_norm": 0.5903770923614502, + "learning_rate": 0.0009841999103541012, + "loss": 0.4662, + "step": 668 + }, + { + "epoch": 0.03737743386317289, + "grad_norm": 0.6343687176704407, + "learning_rate": 0.0009841718960107576, + "loss": 0.5031, + "step": 669 + }, + { + "epoch": 0.037433304466854764, + "grad_norm": 1.7174453735351562, + "learning_rate": 0.0009841438816674138, + "loss": 0.5476, + "step": 670 + }, + { + "epoch": 0.03748917507053664, + "grad_norm": 0.8802082538604736, + "learning_rate": 0.00098411586732407, + "loss": 0.5025, + "step": 671 + }, + { + "epoch": 0.037545045674218507, + "grad_norm": 1.1196390390396118, + "learning_rate": 0.0009840878529807262, + "loss": 0.5847, + "step": 672 + }, + { + "epoch": 0.03760091627790038, + "grad_norm": 2.63988995552063, + "learning_rate": 0.0009840598386373824, + "loss": 0.7298, + "step": 673 + }, + { + "epoch": 0.037656786881582256, + "grad_norm": 2.463869094848633, + "learning_rate": 0.0009840318242940386, + "loss": 0.5205, + "step": 674 + }, + { + "epoch": 0.03771265748526413, + "grad_norm": 0.6522949934005737, + "learning_rate": 0.0009840038099506948, + "loss": 0.5852, + "step": 675 + }, + { + "epoch": 0.037768528088946, + "grad_norm": 0.7417572736740112, + "learning_rate": 0.000983975795607351, + "loss": 0.5242, + "step": 676 + }, + { + "epoch": 0.037824398692627874, + "grad_norm": 2.9596893787384033, + "learning_rate": 0.0009839477812640073, + "loss": 0.5636, + "step": 677 + }, + { + "epoch": 0.03788026929630975, + "grad_norm": 3.3078246116638184, + "learning_rate": 0.0009839197669206635, + "loss": 0.53, + "step": 678 + }, + { + "epoch": 0.03793613989999162, + "grad_norm": 2.979210138320923, + "learning_rate": 0.0009838917525773194, + "loss": 0.5723, + "step": 679 + }, + { + "epoch": 0.03799201050367349, + "grad_norm": 2.5099411010742188, + "learning_rate": 0.0009838637382339759, + "loss": 0.5973, + "step": 680 + }, + { + "epoch": 0.038047881107355366, + "grad_norm": 0.9078687429428101, + "learning_rate": 0.000983835723890632, + "loss": 0.4863, + "step": 681 + }, + { + "epoch": 0.03810375171103724, + "grad_norm": 0.8487613201141357, + "learning_rate": 0.0009838077095472883, + "loss": 0.5132, + "step": 682 + }, + { + "epoch": 0.03815962231471911, + "grad_norm": 0.7502951622009277, + "learning_rate": 0.0009837796952039445, + "loss": 0.4443, + "step": 683 + }, + { + "epoch": 0.038215492918400984, + "grad_norm": 1.747426152229309, + "learning_rate": 0.0009837516808606007, + "loss": 0.4493, + "step": 684 + }, + { + "epoch": 0.03827136352208286, + "grad_norm": 0.7013636827468872, + "learning_rate": 0.000983723666517257, + "loss": 0.5012, + "step": 685 + }, + { + "epoch": 0.03832723412576473, + "grad_norm": 1.204539179801941, + "learning_rate": 0.000983695652173913, + "loss": 0.5077, + "step": 686 + }, + { + "epoch": 0.0383831047294466, + "grad_norm": 0.6299290657043457, + "learning_rate": 0.0009836676378305693, + "loss": 0.5265, + "step": 687 + }, + { + "epoch": 0.038438975333128476, + "grad_norm": 7.480317115783691, + "learning_rate": 0.0009836396234872255, + "loss": 0.4649, + "step": 688 + }, + { + "epoch": 0.038494845936810344, + "grad_norm": 0.7401853203773499, + "learning_rate": 0.0009836116091438817, + "loss": 0.4767, + "step": 689 + }, + { + "epoch": 0.03855071654049222, + "grad_norm": 2.0105507373809814, + "learning_rate": 0.000983583594800538, + "loss": 0.4954, + "step": 690 + }, + { + "epoch": 0.038606587144174094, + "grad_norm": 0.6318633556365967, + "learning_rate": 0.0009835555804571941, + "loss": 0.4318, + "step": 691 + }, + { + "epoch": 0.03866245774785597, + "grad_norm": 0.6741903424263, + "learning_rate": 0.0009835275661138503, + "loss": 0.5471, + "step": 692 + }, + { + "epoch": 0.03871832835153784, + "grad_norm": 23.53759765625, + "learning_rate": 0.0009834995517705065, + "loss": 0.6315, + "step": 693 + }, + { + "epoch": 0.03877419895521971, + "grad_norm": 0.8832314610481262, + "learning_rate": 0.0009834715374271628, + "loss": 0.5802, + "step": 694 + }, + { + "epoch": 0.038830069558901587, + "grad_norm": 0.8176652789115906, + "learning_rate": 0.000983443523083819, + "loss": 0.6085, + "step": 695 + }, + { + "epoch": 0.038885940162583454, + "grad_norm": 0.6713727712631226, + "learning_rate": 0.0009834155087404752, + "loss": 0.5495, + "step": 696 + }, + { + "epoch": 0.03894181076626533, + "grad_norm": 0.8862863779067993, + "learning_rate": 0.0009833874943971314, + "loss": 0.5787, + "step": 697 + }, + { + "epoch": 0.038997681369947204, + "grad_norm": 3.3409006595611572, + "learning_rate": 0.0009833594800537876, + "loss": 0.4715, + "step": 698 + }, + { + "epoch": 0.03905355197362907, + "grad_norm": 0.5985406637191772, + "learning_rate": 0.0009833314657104438, + "loss": 0.5209, + "step": 699 + }, + { + "epoch": 0.03910942257731095, + "grad_norm": 0.4946542978286743, + "learning_rate": 0.0009833034513671, + "loss": 0.5146, + "step": 700 + }, + { + "epoch": 0.03916529318099282, + "grad_norm": 1.0103732347488403, + "learning_rate": 0.0009832754370237562, + "loss": 0.5213, + "step": 701 + }, + { + "epoch": 0.0392211637846747, + "grad_norm": 0.6312211155891418, + "learning_rate": 0.0009832474226804124, + "loss": 0.4229, + "step": 702 + }, + { + "epoch": 0.039277034388356565, + "grad_norm": 0.9818792939186096, + "learning_rate": 0.0009832194083370686, + "loss": 0.6269, + "step": 703 + }, + { + "epoch": 0.03933290499203844, + "grad_norm": 0.8461333513259888, + "learning_rate": 0.0009831913939937248, + "loss": 0.5971, + "step": 704 + }, + { + "epoch": 0.039388775595720314, + "grad_norm": 1.0137455463409424, + "learning_rate": 0.000983163379650381, + "loss": 0.5591, + "step": 705 + }, + { + "epoch": 0.03944464619940218, + "grad_norm": 1.1246294975280762, + "learning_rate": 0.0009831353653070372, + "loss": 0.5628, + "step": 706 + }, + { + "epoch": 0.03950051680308406, + "grad_norm": 0.5328571796417236, + "learning_rate": 0.0009831073509636934, + "loss": 0.3898, + "step": 707 + }, + { + "epoch": 0.03955638740676593, + "grad_norm": 1.0333002805709839, + "learning_rate": 0.0009830793366203496, + "loss": 0.6571, + "step": 708 + }, + { + "epoch": 0.0396122580104478, + "grad_norm": 2.8620975017547607, + "learning_rate": 0.0009830513222770058, + "loss": 0.5667, + "step": 709 + }, + { + "epoch": 0.039668128614129675, + "grad_norm": 0.6204800605773926, + "learning_rate": 0.000983023307933662, + "loss": 0.4947, + "step": 710 + }, + { + "epoch": 0.03972399921781155, + "grad_norm": 0.7997000217437744, + "learning_rate": 0.0009829952935903183, + "loss": 0.5414, + "step": 711 + }, + { + "epoch": 0.039779869821493424, + "grad_norm": 1.517962098121643, + "learning_rate": 0.0009829672792469745, + "loss": 0.4888, + "step": 712 + }, + { + "epoch": 0.03983574042517529, + "grad_norm": 0.909848153591156, + "learning_rate": 0.0009829392649036307, + "loss": 0.5376, + "step": 713 + }, + { + "epoch": 0.03989161102885717, + "grad_norm": 1.2070903778076172, + "learning_rate": 0.0009829112505602869, + "loss": 0.3866, + "step": 714 + }, + { + "epoch": 0.03994748163253904, + "grad_norm": 0.790951669216156, + "learning_rate": 0.000982883236216943, + "loss": 0.5312, + "step": 715 + }, + { + "epoch": 0.04000335223622091, + "grad_norm": 0.5431519150733948, + "learning_rate": 0.0009828552218735993, + "loss": 0.4541, + "step": 716 + }, + { + "epoch": 0.040059222839902785, + "grad_norm": 1.949797511100769, + "learning_rate": 0.0009828272075302555, + "loss": 0.5423, + "step": 717 + }, + { + "epoch": 0.04011509344358466, + "grad_norm": 7.877873420715332, + "learning_rate": 0.0009827991931869117, + "loss": 0.5907, + "step": 718 + }, + { + "epoch": 0.04017096404726653, + "grad_norm": 1.864026665687561, + "learning_rate": 0.000982771178843568, + "loss": 0.4525, + "step": 719 + }, + { + "epoch": 0.0402268346509484, + "grad_norm": 0.8437661528587341, + "learning_rate": 0.000982743164500224, + "loss": 0.5373, + "step": 720 + }, + { + "epoch": 0.04028270525463028, + "grad_norm": 0.5720534324645996, + "learning_rate": 0.0009827151501568803, + "loss": 0.551, + "step": 721 + }, + { + "epoch": 0.04033857585831215, + "grad_norm": 0.6950066089630127, + "learning_rate": 0.0009826871358135365, + "loss": 0.4696, + "step": 722 + }, + { + "epoch": 0.04039444646199402, + "grad_norm": 0.8243922591209412, + "learning_rate": 0.0009826591214701927, + "loss": 0.6305, + "step": 723 + }, + { + "epoch": 0.040450317065675895, + "grad_norm": 0.6435898542404175, + "learning_rate": 0.000982631107126849, + "loss": 0.6157, + "step": 724 + }, + { + "epoch": 0.04050618766935777, + "grad_norm": 0.760627031326294, + "learning_rate": 0.0009826030927835051, + "loss": 0.5524, + "step": 725 + }, + { + "epoch": 0.04056205827303964, + "grad_norm": 0.6070602536201477, + "learning_rate": 0.0009825750784401613, + "loss": 0.5391, + "step": 726 + }, + { + "epoch": 0.04061792887672151, + "grad_norm": 0.9351568222045898, + "learning_rate": 0.0009825470640968175, + "loss": 0.6234, + "step": 727 + }, + { + "epoch": 0.04067379948040339, + "grad_norm": 2.279794692993164, + "learning_rate": 0.0009825190497534738, + "loss": 0.5979, + "step": 728 + }, + { + "epoch": 0.040729670084085255, + "grad_norm": 1.0910197496414185, + "learning_rate": 0.00098249103541013, + "loss": 0.6341, + "step": 729 + }, + { + "epoch": 0.04078554068776713, + "grad_norm": 1.9678157567977905, + "learning_rate": 0.0009824630210667862, + "loss": 0.5134, + "step": 730 + }, + { + "epoch": 0.040841411291449005, + "grad_norm": 1.385367751121521, + "learning_rate": 0.0009824350067234424, + "loss": 0.4654, + "step": 731 + }, + { + "epoch": 0.04089728189513088, + "grad_norm": 0.6241648197174072, + "learning_rate": 0.0009824069923800986, + "loss": 0.5878, + "step": 732 + }, + { + "epoch": 0.04095315249881275, + "grad_norm": 1.3406037092208862, + "learning_rate": 0.0009823789780367548, + "loss": 0.5706, + "step": 733 + }, + { + "epoch": 0.04100902310249462, + "grad_norm": 0.9956802725791931, + "learning_rate": 0.000982350963693411, + "loss": 0.7666, + "step": 734 + }, + { + "epoch": 0.0410648937061765, + "grad_norm": 1.0439846515655518, + "learning_rate": 0.0009823229493500672, + "loss": 0.5754, + "step": 735 + }, + { + "epoch": 0.041120764309858365, + "grad_norm": 0.848108172416687, + "learning_rate": 0.0009822949350067236, + "loss": 0.5218, + "step": 736 + }, + { + "epoch": 0.04117663491354024, + "grad_norm": 0.7220278382301331, + "learning_rate": 0.0009822669206633796, + "loss": 0.743, + "step": 737 + }, + { + "epoch": 0.041232505517222115, + "grad_norm": 1.980831265449524, + "learning_rate": 0.0009822389063200358, + "loss": 0.4338, + "step": 738 + }, + { + "epoch": 0.04128837612090398, + "grad_norm": 0.6656761765480042, + "learning_rate": 0.000982210891976692, + "loss": 0.5456, + "step": 739 + }, + { + "epoch": 0.04134424672458586, + "grad_norm": 4.28108024597168, + "learning_rate": 0.0009821828776333484, + "loss": 0.4915, + "step": 740 + }, + { + "epoch": 0.04140011732826773, + "grad_norm": 1.0239499807357788, + "learning_rate": 0.0009821548632900044, + "loss": 0.5371, + "step": 741 + }, + { + "epoch": 0.04145598793194961, + "grad_norm": 1.5214591026306152, + "learning_rate": 0.0009821268489466606, + "loss": 0.5233, + "step": 742 + }, + { + "epoch": 0.041511858535631475, + "grad_norm": 0.8066681623458862, + "learning_rate": 0.000982098834603317, + "loss": 0.5559, + "step": 743 + }, + { + "epoch": 0.04156772913931335, + "grad_norm": 0.7032321691513062, + "learning_rate": 0.000982070820259973, + "loss": 0.5484, + "step": 744 + }, + { + "epoch": 0.041623599742995225, + "grad_norm": 1.7425769567489624, + "learning_rate": 0.0009820428059166292, + "loss": 0.5381, + "step": 745 + }, + { + "epoch": 0.04167947034667709, + "grad_norm": 33.32256317138672, + "learning_rate": 0.0009820147915732855, + "loss": 0.4958, + "step": 746 + }, + { + "epoch": 0.04173534095035897, + "grad_norm": 0.9755686521530151, + "learning_rate": 0.0009819867772299419, + "loss": 0.5218, + "step": 747 + }, + { + "epoch": 0.04179121155404084, + "grad_norm": 0.8219028115272522, + "learning_rate": 0.0009819587628865979, + "loss": 0.5023, + "step": 748 + }, + { + "epoch": 0.04184708215772272, + "grad_norm": 8.895853996276855, + "learning_rate": 0.000981930748543254, + "loss": 0.4203, + "step": 749 + }, + { + "epoch": 0.041902952761404585, + "grad_norm": 0.5859739184379578, + "learning_rate": 0.0009819027341999103, + "loss": 0.4683, + "step": 750 + }, + { + "epoch": 0.04195882336508646, + "grad_norm": 1.0794117450714111, + "learning_rate": 0.0009818747198565667, + "loss": 0.4639, + "step": 751 + }, + { + "epoch": 0.042014693968768335, + "grad_norm": 0.4005800187587738, + "learning_rate": 0.0009818467055132227, + "loss": 0.4537, + "step": 752 + }, + { + "epoch": 0.0420705645724502, + "grad_norm": 3.5574984550476074, + "learning_rate": 0.000981818691169879, + "loss": 0.6942, + "step": 753 + }, + { + "epoch": 0.04212643517613208, + "grad_norm": 0.7732884287834167, + "learning_rate": 0.0009817906768265353, + "loss": 0.5629, + "step": 754 + }, + { + "epoch": 0.04218230577981395, + "grad_norm": 0.8968021273612976, + "learning_rate": 0.0009817626624831915, + "loss": 0.5994, + "step": 755 + }, + { + "epoch": 0.04223817638349582, + "grad_norm": 2.7302043437957764, + "learning_rate": 0.0009817346481398475, + "loss": 0.523, + "step": 756 + }, + { + "epoch": 0.042294046987177696, + "grad_norm": 0.5633674263954163, + "learning_rate": 0.0009817066337965037, + "loss": 0.5101, + "step": 757 + }, + { + "epoch": 0.04234991759085957, + "grad_norm": 0.5443925261497498, + "learning_rate": 0.0009816786194531601, + "loss": 0.515, + "step": 758 + }, + { + "epoch": 0.042405788194541445, + "grad_norm": 0.9148800373077393, + "learning_rate": 0.0009816506051098163, + "loss": 0.4435, + "step": 759 + }, + { + "epoch": 0.04246165879822331, + "grad_norm": 1.6479774713516235, + "learning_rate": 0.0009816225907664723, + "loss": 0.6153, + "step": 760 + }, + { + "epoch": 0.04251752940190519, + "grad_norm": 0.6888164281845093, + "learning_rate": 0.0009815945764231288, + "loss": 0.4845, + "step": 761 + }, + { + "epoch": 0.04257340000558706, + "grad_norm": 2.2958621978759766, + "learning_rate": 0.000981566562079785, + "loss": 0.475, + "step": 762 + }, + { + "epoch": 0.04262927060926893, + "grad_norm": 8.38556957244873, + "learning_rate": 0.000981538547736441, + "loss": 0.5408, + "step": 763 + }, + { + "epoch": 0.042685141212950806, + "grad_norm": 6.907278537750244, + "learning_rate": 0.0009815105333930972, + "loss": 0.4277, + "step": 764 + }, + { + "epoch": 0.04274101181663268, + "grad_norm": 0.934593141078949, + "learning_rate": 0.0009814825190497536, + "loss": 0.619, + "step": 765 + }, + { + "epoch": 0.04279688242031455, + "grad_norm": 1.9661973714828491, + "learning_rate": 0.0009814545047064098, + "loss": 0.601, + "step": 766 + }, + { + "epoch": 0.04285275302399642, + "grad_norm": 0.5971108675003052, + "learning_rate": 0.0009814264903630658, + "loss": 0.4507, + "step": 767 + }, + { + "epoch": 0.0429086236276783, + "grad_norm": 0.6326439380645752, + "learning_rate": 0.0009813984760197222, + "loss": 0.5875, + "step": 768 + }, + { + "epoch": 0.04296449423136017, + "grad_norm": 0.9448563456535339, + "learning_rate": 0.0009813704616763784, + "loss": 0.543, + "step": 769 + }, + { + "epoch": 0.04302036483504204, + "grad_norm": 0.7144125699996948, + "learning_rate": 0.0009813424473330346, + "loss": 0.5107, + "step": 770 + }, + { + "epoch": 0.043076235438723916, + "grad_norm": 0.6923877596855164, + "learning_rate": 0.0009813144329896906, + "loss": 0.4379, + "step": 771 + }, + { + "epoch": 0.04313210604240579, + "grad_norm": 3.2681655883789062, + "learning_rate": 0.000981286418646347, + "loss": 0.4872, + "step": 772 + }, + { + "epoch": 0.04318797664608766, + "grad_norm": 1.0842937231063843, + "learning_rate": 0.0009812584043030032, + "loss": 0.5029, + "step": 773 + }, + { + "epoch": 0.04324384724976953, + "grad_norm": 1.0609720945358276, + "learning_rate": 0.0009812303899596594, + "loss": 0.5177, + "step": 774 + }, + { + "epoch": 0.04329971785345141, + "grad_norm": 0.6644759774208069, + "learning_rate": 0.0009812023756163156, + "loss": 0.6308, + "step": 775 + }, + { + "epoch": 0.043355588457133276, + "grad_norm": 0.6859840154647827, + "learning_rate": 0.0009811743612729718, + "loss": 0.4113, + "step": 776 + }, + { + "epoch": 0.04341145906081515, + "grad_norm": 1.3187384605407715, + "learning_rate": 0.000981146346929628, + "loss": 0.561, + "step": 777 + }, + { + "epoch": 0.043467329664497026, + "grad_norm": 2.753251791000366, + "learning_rate": 0.0009811183325862843, + "loss": 0.53, + "step": 778 + }, + { + "epoch": 0.0435232002681789, + "grad_norm": 0.8931394815444946, + "learning_rate": 0.0009810903182429405, + "loss": 0.5305, + "step": 779 + }, + { + "epoch": 0.04357907087186077, + "grad_norm": 0.7154929041862488, + "learning_rate": 0.0009810623038995967, + "loss": 0.5927, + "step": 780 + }, + { + "epoch": 0.04363494147554264, + "grad_norm": 0.8575743436813354, + "learning_rate": 0.0009810342895562529, + "loss": 0.4946, + "step": 781 + }, + { + "epoch": 0.04369081207922452, + "grad_norm": 1.8491828441619873, + "learning_rate": 0.000981006275212909, + "loss": 0.5338, + "step": 782 + }, + { + "epoch": 0.043746682682906386, + "grad_norm": 0.9369918704032898, + "learning_rate": 0.0009809782608695653, + "loss": 0.3943, + "step": 783 + }, + { + "epoch": 0.04380255328658826, + "grad_norm": 0.7688659429550171, + "learning_rate": 0.0009809502465262215, + "loss": 0.4342, + "step": 784 + }, + { + "epoch": 0.043858423890270136, + "grad_norm": 1.4930638074874878, + "learning_rate": 0.0009809222321828777, + "loss": 0.609, + "step": 785 + }, + { + "epoch": 0.043914294493952004, + "grad_norm": 1.1172701120376587, + "learning_rate": 0.000980894217839534, + "loss": 0.6184, + "step": 786 + }, + { + "epoch": 0.04397016509763388, + "grad_norm": 12.348740577697754, + "learning_rate": 0.0009808662034961901, + "loss": 0.4814, + "step": 787 + }, + { + "epoch": 0.044026035701315754, + "grad_norm": 2.9989047050476074, + "learning_rate": 0.0009808381891528463, + "loss": 0.5113, + "step": 788 + }, + { + "epoch": 0.04408190630499763, + "grad_norm": 1.7649078369140625, + "learning_rate": 0.0009808101748095025, + "loss": 0.4664, + "step": 789 + }, + { + "epoch": 0.044137776908679496, + "grad_norm": 1.096727967262268, + "learning_rate": 0.0009807821604661587, + "loss": 0.5195, + "step": 790 + }, + { + "epoch": 0.04419364751236137, + "grad_norm": 0.803246021270752, + "learning_rate": 0.000980754146122815, + "loss": 0.5657, + "step": 791 + }, + { + "epoch": 0.044249518116043246, + "grad_norm": 0.7011383771896362, + "learning_rate": 0.0009807261317794711, + "loss": 0.6515, + "step": 792 + }, + { + "epoch": 0.044305388719725114, + "grad_norm": 0.9503406882286072, + "learning_rate": 0.0009806981174361273, + "loss": 0.5169, + "step": 793 + }, + { + "epoch": 0.04436125932340699, + "grad_norm": 2.0136280059814453, + "learning_rate": 0.0009806701030927836, + "loss": 0.5624, + "step": 794 + }, + { + "epoch": 0.044417129927088864, + "grad_norm": 0.8789620399475098, + "learning_rate": 0.0009806420887494398, + "loss": 0.51, + "step": 795 + }, + { + "epoch": 0.04447300053077073, + "grad_norm": 0.6043013334274292, + "learning_rate": 0.000980614074406096, + "loss": 0.446, + "step": 796 + }, + { + "epoch": 0.044528871134452606, + "grad_norm": 1.0158456563949585, + "learning_rate": 0.0009805860600627522, + "loss": 0.5681, + "step": 797 + }, + { + "epoch": 0.04458474173813448, + "grad_norm": 0.5511325597763062, + "learning_rate": 0.0009805580457194084, + "loss": 0.5497, + "step": 798 + }, + { + "epoch": 0.044640612341816356, + "grad_norm": 0.590323805809021, + "learning_rate": 0.0009805300313760646, + "loss": 0.4276, + "step": 799 + }, + { + "epoch": 0.044696482945498224, + "grad_norm": 0.8543387651443481, + "learning_rate": 0.0009805020170327208, + "loss": 0.7413, + "step": 800 + }, + { + "epoch": 0.0447523535491801, + "grad_norm": 0.9470694661140442, + "learning_rate": 0.000980474002689377, + "loss": 0.5227, + "step": 801 + }, + { + "epoch": 0.044808224152861974, + "grad_norm": 0.5236680507659912, + "learning_rate": 0.0009804459883460332, + "loss": 0.4271, + "step": 802 + }, + { + "epoch": 0.04486409475654384, + "grad_norm": 0.6104647517204285, + "learning_rate": 0.0009804179740026894, + "loss": 0.5057, + "step": 803 + }, + { + "epoch": 0.044919965360225717, + "grad_norm": 1.2993942499160767, + "learning_rate": 0.0009803899596593456, + "loss": 0.562, + "step": 804 + }, + { + "epoch": 0.04497583596390759, + "grad_norm": 1.0457202196121216, + "learning_rate": 0.0009803619453160018, + "loss": 0.458, + "step": 805 + }, + { + "epoch": 0.045031706567589466, + "grad_norm": 0.7248060703277588, + "learning_rate": 0.000980333930972658, + "loss": 0.4943, + "step": 806 + }, + { + "epoch": 0.045087577171271334, + "grad_norm": 1.0021425485610962, + "learning_rate": 0.0009803059166293142, + "loss": 0.6232, + "step": 807 + }, + { + "epoch": 0.04514344777495321, + "grad_norm": 1.035917043685913, + "learning_rate": 0.0009802779022859704, + "loss": 0.549, + "step": 808 + }, + { + "epoch": 0.045199318378635084, + "grad_norm": 0.7633391618728638, + "learning_rate": 0.0009802498879426266, + "loss": 0.5585, + "step": 809 + }, + { + "epoch": 0.04525518898231695, + "grad_norm": 0.7718755006790161, + "learning_rate": 0.0009802218735992828, + "loss": 0.5268, + "step": 810 + }, + { + "epoch": 0.04531105958599883, + "grad_norm": 0.6716657876968384, + "learning_rate": 0.000980193859255939, + "loss": 0.4606, + "step": 811 + }, + { + "epoch": 0.0453669301896807, + "grad_norm": 0.7175160646438599, + "learning_rate": 0.0009801658449125953, + "loss": 0.5292, + "step": 812 + }, + { + "epoch": 0.04542280079336257, + "grad_norm": 1.2930527925491333, + "learning_rate": 0.0009801378305692515, + "loss": 0.5964, + "step": 813 + }, + { + "epoch": 0.045478671397044444, + "grad_norm": 0.5609349608421326, + "learning_rate": 0.0009801098162259077, + "loss": 0.4623, + "step": 814 + }, + { + "epoch": 0.04553454200072632, + "grad_norm": 0.5269076824188232, + "learning_rate": 0.0009800818018825639, + "loss": 0.4151, + "step": 815 + }, + { + "epoch": 0.045590412604408194, + "grad_norm": 1.6269625425338745, + "learning_rate": 0.00098005378753922, + "loss": 0.5606, + "step": 816 + }, + { + "epoch": 0.04564628320809006, + "grad_norm": 0.6248766779899597, + "learning_rate": 0.0009800257731958763, + "loss": 0.4512, + "step": 817 + }, + { + "epoch": 0.04570215381177194, + "grad_norm": 1.2871336936950684, + "learning_rate": 0.0009799977588525325, + "loss": 0.4435, + "step": 818 + }, + { + "epoch": 0.04575802441545381, + "grad_norm": 0.7572551965713501, + "learning_rate": 0.0009799697445091887, + "loss": 0.6198, + "step": 819 + }, + { + "epoch": 0.04581389501913568, + "grad_norm": 0.5317825078964233, + "learning_rate": 0.000979941730165845, + "loss": 0.3945, + "step": 820 + }, + { + "epoch": 0.045869765622817554, + "grad_norm": 0.7582911849021912, + "learning_rate": 0.000979913715822501, + "loss": 0.5907, + "step": 821 + }, + { + "epoch": 0.04592563622649943, + "grad_norm": 0.6440884470939636, + "learning_rate": 0.0009798857014791573, + "loss": 0.4313, + "step": 822 + }, + { + "epoch": 0.0459815068301813, + "grad_norm": 0.6804897785186768, + "learning_rate": 0.0009798576871358135, + "loss": 0.5073, + "step": 823 + }, + { + "epoch": 0.04603737743386317, + "grad_norm": 1.444077491760254, + "learning_rate": 0.0009798296727924697, + "loss": 0.6111, + "step": 824 + }, + { + "epoch": 0.04609324803754505, + "grad_norm": Infinity, + "learning_rate": 0.0009798296727924697, + "loss": 0.5277, + "step": 825 + }, + { + "epoch": 0.04614911864122692, + "grad_norm": 0.72920823097229, + "learning_rate": 0.000979801658449126, + "loss": 0.4869, + "step": 826 + }, + { + "epoch": 0.04620498924490879, + "grad_norm": 0.740954577922821, + "learning_rate": 0.0009797736441057821, + "loss": 0.5717, + "step": 827 + }, + { + "epoch": 0.046260859848590664, + "grad_norm": 0.8839882612228394, + "learning_rate": 0.0009797456297624383, + "loss": 0.6026, + "step": 828 + }, + { + "epoch": 0.04631673045227254, + "grad_norm": 0.6876107454299927, + "learning_rate": 0.0009797176154190945, + "loss": 0.4046, + "step": 829 + }, + { + "epoch": 0.04637260105595441, + "grad_norm": 0.966625988483429, + "learning_rate": 0.0009796896010757508, + "loss": 0.5729, + "step": 830 + }, + { + "epoch": 0.04642847165963628, + "grad_norm": 0.7008340358734131, + "learning_rate": 0.000979661586732407, + "loss": 0.5041, + "step": 831 + }, + { + "epoch": 0.04648434226331816, + "grad_norm": 0.6344670653343201, + "learning_rate": 0.0009796335723890632, + "loss": 0.5523, + "step": 832 + }, + { + "epoch": 0.046540212867000025, + "grad_norm": 0.6737650036811829, + "learning_rate": 0.0009796055580457194, + "loss": 0.6168, + "step": 833 + }, + { + "epoch": 0.0465960834706819, + "grad_norm": 0.8204056024551392, + "learning_rate": 0.0009795775437023756, + "loss": 0.5609, + "step": 834 + }, + { + "epoch": 0.046651954074363775, + "grad_norm": 1.597615361213684, + "learning_rate": 0.0009795495293590318, + "loss": 0.6744, + "step": 835 + }, + { + "epoch": 0.04670782467804565, + "grad_norm": 0.860788106918335, + "learning_rate": 0.000979521515015688, + "loss": 0.5538, + "step": 836 + }, + { + "epoch": 0.04676369528172752, + "grad_norm": 1.1378577947616577, + "learning_rate": 0.0009794935006723442, + "loss": 0.4317, + "step": 837 + }, + { + "epoch": 0.04681956588540939, + "grad_norm": 0.5424628853797913, + "learning_rate": 0.0009794654863290004, + "loss": 0.4895, + "step": 838 + }, + { + "epoch": 0.04687543648909127, + "grad_norm": 0.4822992980480194, + "learning_rate": 0.0009794374719856566, + "loss": 0.56, + "step": 839 + }, + { + "epoch": 0.046931307092773135, + "grad_norm": 0.7593953609466553, + "learning_rate": 0.000979409457642313, + "loss": 0.5457, + "step": 840 + }, + { + "epoch": 0.04698717769645501, + "grad_norm": 0.701765239238739, + "learning_rate": 0.000979381443298969, + "loss": 0.4049, + "step": 841 + }, + { + "epoch": 0.047043048300136885, + "grad_norm": 0.5492322444915771, + "learning_rate": 0.0009793534289556252, + "loss": 0.4392, + "step": 842 + }, + { + "epoch": 0.04709891890381875, + "grad_norm": 6.409456253051758, + "learning_rate": 0.0009793254146122814, + "loss": 0.5654, + "step": 843 + }, + { + "epoch": 0.04715478950750063, + "grad_norm": 1.7948259115219116, + "learning_rate": 0.0009792974002689379, + "loss": 0.6261, + "step": 844 + }, + { + "epoch": 0.0472106601111825, + "grad_norm": 0.6673821210861206, + "learning_rate": 0.0009792693859255938, + "loss": 0.511, + "step": 845 + }, + { + "epoch": 0.04726653071486438, + "grad_norm": 0.7016047835350037, + "learning_rate": 0.00097924137158225, + "loss": 0.497, + "step": 846 + }, + { + "epoch": 0.047322401318546245, + "grad_norm": 0.6388280391693115, + "learning_rate": 0.0009792133572389065, + "loss": 0.4169, + "step": 847 + }, + { + "epoch": 0.04737827192222812, + "grad_norm": 5.163217067718506, + "learning_rate": 0.0009791853428955627, + "loss": 0.5521, + "step": 848 + }, + { + "epoch": 0.047434142525909995, + "grad_norm": 0.633324384689331, + "learning_rate": 0.0009791573285522187, + "loss": 0.6046, + "step": 849 + }, + { + "epoch": 0.04749001312959186, + "grad_norm": 1.3042035102844238, + "learning_rate": 0.0009791293142088749, + "loss": 0.5696, + "step": 850 + }, + { + "epoch": 0.04754588373327374, + "grad_norm": 0.9735612273216248, + "learning_rate": 0.0009791012998655313, + "loss": 0.5131, + "step": 851 + }, + { + "epoch": 0.04760175433695561, + "grad_norm": 1.3402804136276245, + "learning_rate": 0.0009790732855221873, + "loss": 0.441, + "step": 852 + }, + { + "epoch": 0.04765762494063748, + "grad_norm": 0.6999788880348206, + "learning_rate": 0.0009790452711788435, + "loss": 0.504, + "step": 853 + }, + { + "epoch": 0.047713495544319355, + "grad_norm": 0.8951069712638855, + "learning_rate": 0.0009790172568355, + "loss": 0.6362, + "step": 854 + }, + { + "epoch": 0.04776936614800123, + "grad_norm": 0.4707335829734802, + "learning_rate": 0.0009789892424921561, + "loss": 0.3774, + "step": 855 + }, + { + "epoch": 0.047825236751683105, + "grad_norm": 0.5453493595123291, + "learning_rate": 0.000978961228148812, + "loss": 0.5096, + "step": 856 + }, + { + "epoch": 0.04788110735536497, + "grad_norm": 0.5445558428764343, + "learning_rate": 0.0009789332138054683, + "loss": 0.464, + "step": 857 + }, + { + "epoch": 0.04793697795904685, + "grad_norm": 1.1003377437591553, + "learning_rate": 0.0009789051994621247, + "loss": 0.4959, + "step": 858 + }, + { + "epoch": 0.04799284856272872, + "grad_norm": 0.8389895558357239, + "learning_rate": 0.000978877185118781, + "loss": 0.4975, + "step": 859 + }, + { + "epoch": 0.04804871916641059, + "grad_norm": 0.6628204584121704, + "learning_rate": 0.000978849170775437, + "loss": 0.4535, + "step": 860 + }, + { + "epoch": 0.048104589770092465, + "grad_norm": 1.0718913078308105, + "learning_rate": 0.0009788211564320931, + "loss": 0.6468, + "step": 861 + }, + { + "epoch": 0.04816046037377434, + "grad_norm": 2.0880234241485596, + "learning_rate": 0.0009787931420887496, + "loss": 0.9738, + "step": 862 + }, + { + "epoch": 0.048216330977456215, + "grad_norm": 0.9095547795295715, + "learning_rate": 0.0009787651277454058, + "loss": 0.5082, + "step": 863 + }, + { + "epoch": 0.04827220158113808, + "grad_norm": 0.991227388381958, + "learning_rate": 0.0009787371134020618, + "loss": 0.5475, + "step": 864 + }, + { + "epoch": 0.04832807218481996, + "grad_norm": 0.8409569263458252, + "learning_rate": 0.0009787090990587182, + "loss": 0.4608, + "step": 865 + }, + { + "epoch": 0.04838394278850183, + "grad_norm": 0.8782963752746582, + "learning_rate": 0.0009786810847153744, + "loss": 0.6021, + "step": 866 + }, + { + "epoch": 0.0484398133921837, + "grad_norm": 1.3933281898498535, + "learning_rate": 0.0009786530703720306, + "loss": 0.5974, + "step": 867 + }, + { + "epoch": 0.048495683995865575, + "grad_norm": 7.851535320281982, + "learning_rate": 0.0009786250560286866, + "loss": 0.4722, + "step": 868 + }, + { + "epoch": 0.04855155459954745, + "grad_norm": 3.374971628189087, + "learning_rate": 0.000978597041685343, + "loss": 0.5334, + "step": 869 + }, + { + "epoch": 0.04860742520322932, + "grad_norm": 0.6941614151000977, + "learning_rate": 0.0009785690273419992, + "loss": 0.4273, + "step": 870 + }, + { + "epoch": 0.04866329580691119, + "grad_norm": 0.8112534880638123, + "learning_rate": 0.0009785410129986552, + "loss": 0.4331, + "step": 871 + }, + { + "epoch": 0.04871916641059307, + "grad_norm": 0.9661931991577148, + "learning_rate": 0.0009785129986553116, + "loss": 0.6186, + "step": 872 + }, + { + "epoch": 0.04877503701427494, + "grad_norm": 0.6012800931930542, + "learning_rate": 0.0009784849843119678, + "loss": 0.5016, + "step": 873 + }, + { + "epoch": 0.04883090761795681, + "grad_norm": 1.040717601776123, + "learning_rate": 0.000978456969968624, + "loss": 0.6147, + "step": 874 + }, + { + "epoch": 0.048886778221638685, + "grad_norm": 0.696476399898529, + "learning_rate": 0.00097842895562528, + "loss": 0.4833, + "step": 875 + }, + { + "epoch": 0.04894264882532056, + "grad_norm": 0.5771315097808838, + "learning_rate": 0.0009784009412819364, + "loss": 0.4036, + "step": 876 + }, + { + "epoch": 0.04899851942900243, + "grad_norm": 0.6374980211257935, + "learning_rate": 0.0009783729269385926, + "loss": 0.4536, + "step": 877 + }, + { + "epoch": 0.0490543900326843, + "grad_norm": 0.7906278967857361, + "learning_rate": 0.0009783449125952489, + "loss": 0.7222, + "step": 878 + }, + { + "epoch": 0.04911026063636618, + "grad_norm": 0.7845003008842468, + "learning_rate": 0.000978316898251905, + "loss": 0.4299, + "step": 879 + }, + { + "epoch": 0.049166131240048046, + "grad_norm": 0.6394595503807068, + "learning_rate": 0.0009782888839085613, + "loss": 0.5203, + "step": 880 + }, + { + "epoch": 0.04922200184372992, + "grad_norm": 3.4356772899627686, + "learning_rate": 0.0009782608695652175, + "loss": 0.528, + "step": 881 + }, + { + "epoch": 0.049277872447411795, + "grad_norm": 0.9698755145072937, + "learning_rate": 0.0009782328552218737, + "loss": 0.5494, + "step": 882 + }, + { + "epoch": 0.04933374305109367, + "grad_norm": 0.48002758622169495, + "learning_rate": 0.0009782048408785299, + "loss": 0.5038, + "step": 883 + }, + { + "epoch": 0.04938961365477554, + "grad_norm": 4.47374153137207, + "learning_rate": 0.000978176826535186, + "loss": 0.6252, + "step": 884 + }, + { + "epoch": 0.04944548425845741, + "grad_norm": 0.5357908606529236, + "learning_rate": 0.0009781488121918423, + "loss": 0.5076, + "step": 885 + }, + { + "epoch": 0.04950135486213929, + "grad_norm": 0.5651967525482178, + "learning_rate": 0.0009781207978484985, + "loss": 0.523, + "step": 886 + }, + { + "epoch": 0.049557225465821156, + "grad_norm": 0.6906450986862183, + "learning_rate": 0.0009780927835051547, + "loss": 0.5423, + "step": 887 + }, + { + "epoch": 0.04961309606950303, + "grad_norm": 0.7776225209236145, + "learning_rate": 0.000978064769161811, + "loss": 0.6449, + "step": 888 + }, + { + "epoch": 0.049668966673184906, + "grad_norm": 0.6452109813690186, + "learning_rate": 0.0009780367548184671, + "loss": 0.561, + "step": 889 + }, + { + "epoch": 0.04972483727686677, + "grad_norm": 1.0197460651397705, + "learning_rate": 0.0009780087404751233, + "loss": 0.6516, + "step": 890 + }, + { + "epoch": 0.04978070788054865, + "grad_norm": 1.0949196815490723, + "learning_rate": 0.0009779807261317795, + "loss": 0.5097, + "step": 891 + }, + { + "epoch": 0.04983657848423052, + "grad_norm": 0.581544816493988, + "learning_rate": 0.0009779527117884357, + "loss": 0.5173, + "step": 892 + }, + { + "epoch": 0.0498924490879124, + "grad_norm": 6.4513068199157715, + "learning_rate": 0.000977924697445092, + "loss": 0.5599, + "step": 893 + }, + { + "epoch": 0.049948319691594266, + "grad_norm": 0.9144807457923889, + "learning_rate": 0.0009778966831017481, + "loss": 0.6016, + "step": 894 + }, + { + "epoch": 0.05000419029527614, + "grad_norm": 0.8852243423461914, + "learning_rate": 0.0009778686687584043, + "loss": 0.5715, + "step": 895 + }, + { + "epoch": 0.050060060898958016, + "grad_norm": 0.6177686452865601, + "learning_rate": 0.0009778406544150606, + "loss": 0.4615, + "step": 896 + }, + { + "epoch": 0.050115931502639884, + "grad_norm": 0.5267212390899658, + "learning_rate": 0.0009778126400717168, + "loss": 0.5155, + "step": 897 + }, + { + "epoch": 0.05017180210632176, + "grad_norm": 0.45540744066238403, + "learning_rate": 0.000977784625728373, + "loss": 0.4286, + "step": 898 + }, + { + "epoch": 0.05022767271000363, + "grad_norm": 1.030368447303772, + "learning_rate": 0.0009777566113850292, + "loss": 0.5605, + "step": 899 + }, + { + "epoch": 0.0502835433136855, + "grad_norm": 1.030551552772522, + "learning_rate": 0.0009777285970416854, + "loss": 0.5966, + "step": 900 + }, + { + "epoch": 0.050339413917367376, + "grad_norm": 0.7366147637367249, + "learning_rate": 0.0009777005826983416, + "loss": 0.4379, + "step": 901 + }, + { + "epoch": 0.05039528452104925, + "grad_norm": 2.144909620285034, + "learning_rate": 0.0009776725683549978, + "loss": 0.4919, + "step": 902 + }, + { + "epoch": 0.050451155124731126, + "grad_norm": 1.0029038190841675, + "learning_rate": 0.000977644554011654, + "loss": 0.6251, + "step": 903 + }, + { + "epoch": 0.050507025728412994, + "grad_norm": 0.6969373226165771, + "learning_rate": 0.0009776165396683102, + "loss": 0.4654, + "step": 904 + }, + { + "epoch": 0.05056289633209487, + "grad_norm": 0.7720199823379517, + "learning_rate": 0.0009775885253249664, + "loss": 0.5214, + "step": 905 + }, + { + "epoch": 0.05061876693577674, + "grad_norm": 0.6664357781410217, + "learning_rate": 0.0009775605109816226, + "loss": 0.6063, + "step": 906 + }, + { + "epoch": 0.05067463753945861, + "grad_norm": 0.5272346138954163, + "learning_rate": 0.0009775324966382788, + "loss": 0.477, + "step": 907 + }, + { + "epoch": 0.050730508143140486, + "grad_norm": 0.776414692401886, + "learning_rate": 0.000977504482294935, + "loss": 0.4973, + "step": 908 + }, + { + "epoch": 0.05078637874682236, + "grad_norm": 0.584536612033844, + "learning_rate": 0.0009774764679515912, + "loss": 0.5159, + "step": 909 + }, + { + "epoch": 0.05084224935050423, + "grad_norm": 0.7944260239601135, + "learning_rate": 0.0009774484536082474, + "loss": 0.4176, + "step": 910 + }, + { + "epoch": 0.050898119954186104, + "grad_norm": 0.7821248173713684, + "learning_rate": 0.0009774204392649036, + "loss": 0.6104, + "step": 911 + }, + { + "epoch": 0.05095399055786798, + "grad_norm": 0.5053620338439941, + "learning_rate": 0.0009773924249215598, + "loss": 0.46, + "step": 912 + }, + { + "epoch": 0.05100986116154985, + "grad_norm": 0.6451984643936157, + "learning_rate": 0.000977364410578216, + "loss": 0.5144, + "step": 913 + }, + { + "epoch": 0.05106573176523172, + "grad_norm": 0.6685951352119446, + "learning_rate": 0.0009773363962348723, + "loss": 0.5504, + "step": 914 + }, + { + "epoch": 0.051121602368913596, + "grad_norm": 0.6952266693115234, + "learning_rate": 0.0009773083818915285, + "loss": 0.5274, + "step": 915 + }, + { + "epoch": 0.05117747297259547, + "grad_norm": 0.5194458365440369, + "learning_rate": 0.0009772803675481847, + "loss": 0.4644, + "step": 916 + }, + { + "epoch": 0.05123334357627734, + "grad_norm": 0.8838127851486206, + "learning_rate": 0.0009772523532048409, + "loss": 0.5584, + "step": 917 + }, + { + "epoch": 0.051289214179959214, + "grad_norm": 2.0893125534057617, + "learning_rate": 0.000977224338861497, + "loss": 0.4625, + "step": 918 + }, + { + "epoch": 0.05134508478364109, + "grad_norm": 0.5777636766433716, + "learning_rate": 0.0009771963245181533, + "loss": 0.499, + "step": 919 + }, + { + "epoch": 0.05140095538732296, + "grad_norm": 0.484514445066452, + "learning_rate": 0.0009771683101748095, + "loss": 0.4135, + "step": 920 + }, + { + "epoch": 0.05145682599100483, + "grad_norm": 0.524583637714386, + "learning_rate": 0.0009771402958314657, + "loss": 0.5085, + "step": 921 + }, + { + "epoch": 0.051512696594686706, + "grad_norm": 0.6903277039527893, + "learning_rate": 0.000977112281488122, + "loss": 0.538, + "step": 922 + }, + { + "epoch": 0.05156856719836858, + "grad_norm": 0.6162174344062805, + "learning_rate": 0.0009770842671447781, + "loss": 0.5949, + "step": 923 + }, + { + "epoch": 0.05162443780205045, + "grad_norm": 0.9995852112770081, + "learning_rate": 0.0009770562528014343, + "loss": 0.4428, + "step": 924 + }, + { + "epoch": 0.051680308405732324, + "grad_norm": 0.5183641910552979, + "learning_rate": 0.0009770282384580905, + "loss": 0.4401, + "step": 925 + }, + { + "epoch": 0.0517361790094142, + "grad_norm": 0.5494539141654968, + "learning_rate": 0.0009770002241147467, + "loss": 0.3651, + "step": 926 + }, + { + "epoch": 0.05179204961309607, + "grad_norm": 0.5611405372619629, + "learning_rate": 0.000976972209771403, + "loss": 0.493, + "step": 927 + }, + { + "epoch": 0.05184792021677794, + "grad_norm": 0.662929356098175, + "learning_rate": 0.0009769441954280591, + "loss": 0.5124, + "step": 928 + }, + { + "epoch": 0.051903790820459816, + "grad_norm": 1.6724481582641602, + "learning_rate": 0.0009769161810847153, + "loss": 0.6065, + "step": 929 + }, + { + "epoch": 0.05195966142414169, + "grad_norm": 0.5774344801902771, + "learning_rate": 0.0009768881667413716, + "loss": 0.6705, + "step": 930 + }, + { + "epoch": 0.05201553202782356, + "grad_norm": 0.7646609544754028, + "learning_rate": 0.0009768601523980278, + "loss": 0.5637, + "step": 931 + }, + { + "epoch": 0.052071402631505434, + "grad_norm": 2.2260706424713135, + "learning_rate": 0.000976832138054684, + "loss": 0.7117, + "step": 932 + }, + { + "epoch": 0.05212727323518731, + "grad_norm": 0.811889111995697, + "learning_rate": 0.0009768041237113402, + "loss": 0.5476, + "step": 933 + }, + { + "epoch": 0.05218314383886918, + "grad_norm": 4.457334518432617, + "learning_rate": 0.0009767761093679964, + "loss": 0.562, + "step": 934 + }, + { + "epoch": 0.05223901444255105, + "grad_norm": 0.6585486531257629, + "learning_rate": 0.0009767480950246526, + "loss": 0.4368, + "step": 935 + }, + { + "epoch": 0.052294885046232926, + "grad_norm": 0.9832916259765625, + "learning_rate": 0.0009767200806813088, + "loss": 0.5655, + "step": 936 + }, + { + "epoch": 0.052350755649914794, + "grad_norm": 0.7627753019332886, + "learning_rate": 0.000976692066337965, + "loss": 0.7838, + "step": 937 + }, + { + "epoch": 0.05240662625359667, + "grad_norm": 0.6132301688194275, + "learning_rate": 0.0009766640519946212, + "loss": 0.4433, + "step": 938 + }, + { + "epoch": 0.052462496857278544, + "grad_norm": 0.7402481436729431, + "learning_rate": 0.0009766360376512774, + "loss": 0.4396, + "step": 939 + }, + { + "epoch": 0.05251836746096042, + "grad_norm": 0.9469683170318604, + "learning_rate": 0.0009766080233079336, + "loss": 0.5825, + "step": 940 + }, + { + "epoch": 0.05257423806464229, + "grad_norm": 1.0504471063613892, + "learning_rate": 0.0009765800089645898, + "loss": 0.4845, + "step": 941 + }, + { + "epoch": 0.05263010866832416, + "grad_norm": 0.7283835411071777, + "learning_rate": 0.0009765519946212461, + "loss": 0.4497, + "step": 942 + }, + { + "epoch": 0.05268597927200604, + "grad_norm": 0.8355003595352173, + "learning_rate": 0.0009765239802779023, + "loss": 0.5448, + "step": 943 + }, + { + "epoch": 0.052741849875687905, + "grad_norm": 1.0728886127471924, + "learning_rate": 0.0009764959659345584, + "loss": 0.4716, + "step": 944 + }, + { + "epoch": 0.05279772047936978, + "grad_norm": 1.1489648818969727, + "learning_rate": 0.0009764679515912147, + "loss": 0.4678, + "step": 945 + }, + { + "epoch": 0.052853591083051654, + "grad_norm": 0.6460690498352051, + "learning_rate": 0.000976439937247871, + "loss": 0.56, + "step": 946 + }, + { + "epoch": 0.05290946168673352, + "grad_norm": 1.00322425365448, + "learning_rate": 0.0009764119229045272, + "loss": 0.4724, + "step": 947 + }, + { + "epoch": 0.0529653322904154, + "grad_norm": 6.83618688583374, + "learning_rate": 0.0009763839085611833, + "loss": 0.5362, + "step": 948 + }, + { + "epoch": 0.05302120289409727, + "grad_norm": 1.0731102228164673, + "learning_rate": 0.0009763558942178396, + "loss": 0.5441, + "step": 949 + }, + { + "epoch": 0.05307707349777915, + "grad_norm": 0.7452988624572754, + "learning_rate": 0.0009763278798744958, + "loss": 0.4081, + "step": 950 + }, + { + "epoch": 0.053132944101461015, + "grad_norm": 0.4784471094608307, + "learning_rate": 0.000976299865531152, + "loss": 0.4326, + "step": 951 + }, + { + "epoch": 0.05318881470514289, + "grad_norm": 0.5769765973091125, + "learning_rate": 0.0009762718511878082, + "loss": 0.5234, + "step": 952 + }, + { + "epoch": 0.053244685308824764, + "grad_norm": 1.0114166736602783, + "learning_rate": 0.0009762438368444644, + "loss": 0.5137, + "step": 953 + }, + { + "epoch": 0.05330055591250663, + "grad_norm": 0.8083709478378296, + "learning_rate": 0.0009762158225011206, + "loss": 0.495, + "step": 954 + }, + { + "epoch": 0.05335642651618851, + "grad_norm": 0.6101514101028442, + "learning_rate": 0.0009761878081577767, + "loss": 0.5327, + "step": 955 + }, + { + "epoch": 0.05341229711987038, + "grad_norm": 3.620208501815796, + "learning_rate": 0.000976159793814433, + "loss": 0.4619, + "step": 956 + }, + { + "epoch": 0.05346816772355225, + "grad_norm": 0.7751604914665222, + "learning_rate": 0.0009761317794710892, + "loss": 0.4663, + "step": 957 + }, + { + "epoch": 0.053524038327234125, + "grad_norm": 1.0059489011764526, + "learning_rate": 0.0009761037651277454, + "loss": 0.5969, + "step": 958 + }, + { + "epoch": 0.053579908930916, + "grad_norm": 0.7288175821304321, + "learning_rate": 0.0009760757507844016, + "loss": 0.5219, + "step": 959 + }, + { + "epoch": 0.053635779534597874, + "grad_norm": 0.5954000353813171, + "learning_rate": 0.0009760477364410578, + "loss": 0.5522, + "step": 960 + }, + { + "epoch": 0.05369165013827974, + "grad_norm": 0.5526319146156311, + "learning_rate": 0.000976019722097714, + "loss": 0.3744, + "step": 961 + }, + { + "epoch": 0.05374752074196162, + "grad_norm": 0.3723647892475128, + "learning_rate": 0.0009759917077543702, + "loss": 0.4349, + "step": 962 + }, + { + "epoch": 0.05380339134564349, + "grad_norm": 0.6355952024459839, + "learning_rate": 0.0009759636934110265, + "loss": 0.614, + "step": 963 + }, + { + "epoch": 0.05385926194932536, + "grad_norm": 0.7140101790428162, + "learning_rate": 0.0009759356790676827, + "loss": 0.4856, + "step": 964 + }, + { + "epoch": 0.053915132553007235, + "grad_norm": 0.5996012091636658, + "learning_rate": 0.0009759076647243389, + "loss": 0.5727, + "step": 965 + }, + { + "epoch": 0.05397100315668911, + "grad_norm": 0.6142397522926331, + "learning_rate": 0.0009758796503809952, + "loss": 0.5236, + "step": 966 + }, + { + "epoch": 0.05402687376037098, + "grad_norm": 3.1076486110687256, + "learning_rate": 0.0009758516360376513, + "loss": 0.7234, + "step": 967 + }, + { + "epoch": 0.05408274436405285, + "grad_norm": 49.72739028930664, + "learning_rate": 0.0009758236216943075, + "loss": 0.4028, + "step": 968 + }, + { + "epoch": 0.05413861496773473, + "grad_norm": 1.007879376411438, + "learning_rate": 0.0009757956073509637, + "loss": 0.4754, + "step": 969 + }, + { + "epoch": 0.0541944855714166, + "grad_norm": 0.5384015440940857, + "learning_rate": 0.00097576759300762, + "loss": 0.5368, + "step": 970 + }, + { + "epoch": 0.05425035617509847, + "grad_norm": 0.5629345774650574, + "learning_rate": 0.0009757395786642761, + "loss": 0.5792, + "step": 971 + }, + { + "epoch": 0.054306226778780345, + "grad_norm": 1.3148612976074219, + "learning_rate": 0.0009757115643209323, + "loss": 0.4923, + "step": 972 + }, + { + "epoch": 0.05436209738246222, + "grad_norm": 1.1752134561538696, + "learning_rate": 0.0009756835499775886, + "loss": 0.5602, + "step": 973 + }, + { + "epoch": 0.05441796798614409, + "grad_norm": 0.7510524988174438, + "learning_rate": 0.0009756555356342448, + "loss": 0.4485, + "step": 974 + }, + { + "epoch": 0.05447383858982596, + "grad_norm": 0.5648794770240784, + "learning_rate": 0.0009756275212909009, + "loss": 0.4676, + "step": 975 + }, + { + "epoch": 0.05452970919350784, + "grad_norm": 0.5437530875205994, + "learning_rate": 0.0009755995069475571, + "loss": 0.3786, + "step": 976 + }, + { + "epoch": 0.054585579797189705, + "grad_norm": 0.7478137612342834, + "learning_rate": 0.0009755714926042134, + "loss": 0.5361, + "step": 977 + }, + { + "epoch": 0.05464145040087158, + "grad_norm": 0.46340206265449524, + "learning_rate": 0.0009755434782608695, + "loss": 0.517, + "step": 978 + }, + { + "epoch": 0.054697321004553455, + "grad_norm": 0.5945268869400024, + "learning_rate": 0.0009755154639175257, + "loss": 0.4213, + "step": 979 + }, + { + "epoch": 0.05475319160823533, + "grad_norm": 0.6033769249916077, + "learning_rate": 0.000975487449574182, + "loss": 0.5735, + "step": 980 + }, + { + "epoch": 0.0548090622119172, + "grad_norm": 0.5731101632118225, + "learning_rate": 0.0009754594352308383, + "loss": 0.5435, + "step": 981 + }, + { + "epoch": 0.05486493281559907, + "grad_norm": 0.6571980118751526, + "learning_rate": 0.0009754314208874944, + "loss": 0.5357, + "step": 982 + }, + { + "epoch": 0.05492080341928095, + "grad_norm": 0.49703699350357056, + "learning_rate": 0.0009754034065441506, + "loss": 0.5565, + "step": 983 + }, + { + "epoch": 0.054976674022962815, + "grad_norm": 0.5690906047821045, + "learning_rate": 0.0009753753922008069, + "loss": 0.4089, + "step": 984 + }, + { + "epoch": 0.05503254462664469, + "grad_norm": 0.8740798234939575, + "learning_rate": 0.0009753473778574631, + "loss": 0.4921, + "step": 985 + }, + { + "epoch": 0.055088415230326565, + "grad_norm": 0.7877066731452942, + "learning_rate": 0.0009753193635141192, + "loss": 0.5345, + "step": 986 + }, + { + "epoch": 0.05514428583400844, + "grad_norm": 0.4619380235671997, + "learning_rate": 0.0009752913491707754, + "loss": 0.5176, + "step": 987 + }, + { + "epoch": 0.05520015643769031, + "grad_norm": 0.46911683678627014, + "learning_rate": 0.0009752633348274317, + "loss": 0.4385, + "step": 988 + }, + { + "epoch": 0.05525602704137218, + "grad_norm": 0.6308186054229736, + "learning_rate": 0.0009752353204840879, + "loss": 0.4604, + "step": 989 + }, + { + "epoch": 0.05531189764505406, + "grad_norm": 0.5533828139305115, + "learning_rate": 0.000975207306140744, + "loss": 0.5267, + "step": 990 + }, + { + "epoch": 0.055367768248735925, + "grad_norm": 0.812833309173584, + "learning_rate": 0.0009751792917974003, + "loss": 0.6672, + "step": 991 + }, + { + "epoch": 0.0554236388524178, + "grad_norm": 2.2824389934539795, + "learning_rate": 0.0009751512774540565, + "loss": 0.4423, + "step": 992 + }, + { + "epoch": 0.055479509456099675, + "grad_norm": 0.81022047996521, + "learning_rate": 0.0009751232631107127, + "loss": 0.7088, + "step": 993 + }, + { + "epoch": 0.05553538005978154, + "grad_norm": 0.47881194949150085, + "learning_rate": 0.0009750952487673688, + "loss": 0.4457, + "step": 994 + }, + { + "epoch": 0.05559125066346342, + "grad_norm": 0.477819561958313, + "learning_rate": 0.0009750672344240251, + "loss": 0.5035, + "step": 995 + }, + { + "epoch": 0.05564712126714529, + "grad_norm": 2.2060658931732178, + "learning_rate": 0.0009750392200806814, + "loss": 0.5525, + "step": 996 + }, + { + "epoch": 0.05570299187082717, + "grad_norm": 0.8939707279205322, + "learning_rate": 0.0009750112057373376, + "loss": 0.6125, + "step": 997 + }, + { + "epoch": 0.055758862474509036, + "grad_norm": 0.4530435800552368, + "learning_rate": 0.0009749831913939938, + "loss": 0.457, + "step": 998 + }, + { + "epoch": 0.05581473307819091, + "grad_norm": 0.6206973791122437, + "learning_rate": 0.00097495517705065, + "loss": 0.477, + "step": 999 + }, + { + "epoch": 0.055870603681872785, + "grad_norm": 0.5955625176429749, + "learning_rate": 0.0009749271627073062, + "loss": 0.4379, + "step": 1000 + }, + { + "epoch": 0.055870603681872785, + "eval_cer": 0.10094595921573762, + "eval_loss": 0.37760236859321594, + "eval_runtime": 55.9014, + "eval_samples_per_second": 81.179, + "eval_steps_per_second": 5.08, + "eval_wer": 0.3963576436412391, + "step": 1000 + }, + { + "epoch": 0.05592647428555465, + "grad_norm": 0.5746254920959473, + "learning_rate": 0.0009748991483639623, + "loss": 0.62, + "step": 1001 + }, + { + "epoch": 0.05598234488923653, + "grad_norm": 0.4552527666091919, + "learning_rate": 0.0009748711340206186, + "loss": 0.3876, + "step": 1002 + }, + { + "epoch": 0.0560382154929184, + "grad_norm": 0.8034296631813049, + "learning_rate": 0.0009748431196772748, + "loss": 0.5508, + "step": 1003 + }, + { + "epoch": 0.05609408609660027, + "grad_norm": 3.2844505310058594, + "learning_rate": 0.000974815105333931, + "loss": 0.4879, + "step": 1004 + }, + { + "epoch": 0.056149956700282146, + "grad_norm": 1.6426609754562378, + "learning_rate": 0.0009747870909905872, + "loss": 0.6792, + "step": 1005 + }, + { + "epoch": 0.05620582730396402, + "grad_norm": 0.6991989016532898, + "learning_rate": 0.0009747590766472434, + "loss": 0.5495, + "step": 1006 + }, + { + "epoch": 0.056261697907645895, + "grad_norm": 1.0903160572052002, + "learning_rate": 0.0009747310623038996, + "loss": 0.6356, + "step": 1007 + }, + { + "epoch": 0.05631756851132776, + "grad_norm": 0.668310821056366, + "learning_rate": 0.0009747030479605558, + "loss": 0.5699, + "step": 1008 + }, + { + "epoch": 0.05637343911500964, + "grad_norm": 0.7199364304542542, + "learning_rate": 0.000974675033617212, + "loss": 0.5016, + "step": 1009 + }, + { + "epoch": 0.05642930971869151, + "grad_norm": 1.9603673219680786, + "learning_rate": 0.0009746470192738682, + "loss": 0.5312, + "step": 1010 + }, + { + "epoch": 0.05648518032237338, + "grad_norm": 1.0954893827438354, + "learning_rate": 0.0009746190049305244, + "loss": 0.569, + "step": 1011 + }, + { + "epoch": 0.056541050926055256, + "grad_norm": 1.5825647115707397, + "learning_rate": 0.0009745909905871808, + "loss": 0.5963, + "step": 1012 + }, + { + "epoch": 0.05659692152973713, + "grad_norm": 0.4931645393371582, + "learning_rate": 0.0009745629762438369, + "loss": 0.4437, + "step": 1013 + }, + { + "epoch": 0.056652792133419, + "grad_norm": 0.5798553824424744, + "learning_rate": 0.0009745349619004931, + "loss": 0.5332, + "step": 1014 + }, + { + "epoch": 0.05670866273710087, + "grad_norm": 0.698550283908844, + "learning_rate": 0.0009745069475571493, + "loss": 0.6647, + "step": 1015 + }, + { + "epoch": 0.05676453334078275, + "grad_norm": 1.1300115585327148, + "learning_rate": 0.0009744789332138056, + "loss": 0.5727, + "step": 1016 + }, + { + "epoch": 0.05682040394446462, + "grad_norm": 0.6880732178688049, + "learning_rate": 0.0009744509188704617, + "loss": 0.5886, + "step": 1017 + }, + { + "epoch": 0.05687627454814649, + "grad_norm": 0.596858024597168, + "learning_rate": 0.0009744229045271179, + "loss": 0.6667, + "step": 1018 + }, + { + "epoch": 0.056932145151828366, + "grad_norm": 1.586456060409546, + "learning_rate": 0.0009743948901837741, + "loss": 0.548, + "step": 1019 + }, + { + "epoch": 0.05698801575551024, + "grad_norm": 1.0578091144561768, + "learning_rate": 0.0009743668758404303, + "loss": 0.4743, + "step": 1020 + }, + { + "epoch": 0.05704388635919211, + "grad_norm": 1.8541911840438843, + "learning_rate": 0.0009743388614970865, + "loss": 0.4641, + "step": 1021 + }, + { + "epoch": 0.05709975696287398, + "grad_norm": 1.1144582033157349, + "learning_rate": 0.0009743108471537427, + "loss": 0.5999, + "step": 1022 + }, + { + "epoch": 0.05715562756655586, + "grad_norm": 0.8700639009475708, + "learning_rate": 0.000974282832810399, + "loss": 0.4583, + "step": 1023 + }, + { + "epoch": 0.057211498170237726, + "grad_norm": 6.936180591583252, + "learning_rate": 0.0009742548184670551, + "loss": 0.6048, + "step": 1024 + }, + { + "epoch": 0.0572673687739196, + "grad_norm": 0.7235619425773621, + "learning_rate": 0.0009742268041237113, + "loss": 0.4572, + "step": 1025 + }, + { + "epoch": 0.057323239377601476, + "grad_norm": 0.6703459024429321, + "learning_rate": 0.0009741987897803675, + "loss": 0.5762, + "step": 1026 + }, + { + "epoch": 0.05737910998128335, + "grad_norm": 0.7795577049255371, + "learning_rate": 0.0009741707754370238, + "loss": 0.5773, + "step": 1027 + }, + { + "epoch": 0.05743498058496522, + "grad_norm": 0.5746772885322571, + "learning_rate": 0.0009741427610936799, + "loss": 0.6547, + "step": 1028 + }, + { + "epoch": 0.057490851188647094, + "grad_norm": 0.4863956570625305, + "learning_rate": 0.0009741147467503361, + "loss": 0.5006, + "step": 1029 + }, + { + "epoch": 0.05754672179232897, + "grad_norm": 9.924711227416992, + "learning_rate": 0.0009740867324069925, + "loss": 0.4891, + "step": 1030 + }, + { + "epoch": 0.057602592396010836, + "grad_norm": 0.7525727152824402, + "learning_rate": 0.0009740587180636487, + "loss": 0.5598, + "step": 1031 + }, + { + "epoch": 0.05765846299969271, + "grad_norm": 1.9723610877990723, + "learning_rate": 0.0009740307037203048, + "loss": 0.4483, + "step": 1032 + }, + { + "epoch": 0.057714333603374586, + "grad_norm": 0.5508272051811218, + "learning_rate": 0.000974002689376961, + "loss": 0.4984, + "step": 1033 + }, + { + "epoch": 0.057770204207056454, + "grad_norm": 0.9551763534545898, + "learning_rate": 0.0009739746750336173, + "loss": 0.5009, + "step": 1034 + }, + { + "epoch": 0.05782607481073833, + "grad_norm": 0.6343883872032166, + "learning_rate": 0.0009739466606902735, + "loss": 0.4979, + "step": 1035 + }, + { + "epoch": 0.057881945414420204, + "grad_norm": 0.6532105207443237, + "learning_rate": 0.0009739186463469296, + "loss": 0.5822, + "step": 1036 + }, + { + "epoch": 0.05793781601810208, + "grad_norm": 0.8613942861557007, + "learning_rate": 0.0009738906320035859, + "loss": 0.5165, + "step": 1037 + }, + { + "epoch": 0.057993686621783946, + "grad_norm": 4.688444137573242, + "learning_rate": 0.0009738626176602421, + "loss": 0.4789, + "step": 1038 + }, + { + "epoch": 0.05804955722546582, + "grad_norm": 0.9501351714134216, + "learning_rate": 0.0009738346033168983, + "loss": 0.4239, + "step": 1039 + }, + { + "epoch": 0.058105427829147696, + "grad_norm": 0.520897626876831, + "learning_rate": 0.0009738065889735544, + "loss": 0.5064, + "step": 1040 + }, + { + "epoch": 0.058161298432829564, + "grad_norm": 1.4445416927337646, + "learning_rate": 0.0009737785746302107, + "loss": 0.4871, + "step": 1041 + }, + { + "epoch": 0.05821716903651144, + "grad_norm": 4.147278785705566, + "learning_rate": 0.0009737505602868669, + "loss": 0.5051, + "step": 1042 + }, + { + "epoch": 0.058273039640193314, + "grad_norm": 0.7323347330093384, + "learning_rate": 0.000973722545943523, + "loss": 0.6095, + "step": 1043 + }, + { + "epoch": 0.05832891024387518, + "grad_norm": 0.6222905516624451, + "learning_rate": 0.0009736945316001793, + "loss": 0.5128, + "step": 1044 + }, + { + "epoch": 0.058384780847557056, + "grad_norm": 0.580451488494873, + "learning_rate": 0.0009736665172568355, + "loss": 0.4654, + "step": 1045 + }, + { + "epoch": 0.05844065145123893, + "grad_norm": 1.780754804611206, + "learning_rate": 0.0009736385029134918, + "loss": 0.5755, + "step": 1046 + }, + { + "epoch": 0.058496522054920806, + "grad_norm": 0.5635781288146973, + "learning_rate": 0.0009736104885701479, + "loss": 0.537, + "step": 1047 + }, + { + "epoch": 0.058552392658602674, + "grad_norm": 0.6737411618232727, + "learning_rate": 0.0009735824742268042, + "loss": 0.4061, + "step": 1048 + }, + { + "epoch": 0.05860826326228455, + "grad_norm": 0.6910914778709412, + "learning_rate": 0.0009735544598834604, + "loss": 0.5784, + "step": 1049 + }, + { + "epoch": 0.058664133865966424, + "grad_norm": 0.6480022668838501, + "learning_rate": 0.0009735264455401166, + "loss": 0.5571, + "step": 1050 + }, + { + "epoch": 0.05872000446964829, + "grad_norm": 0.7165181040763855, + "learning_rate": 0.0009734984311967727, + "loss": 0.5023, + "step": 1051 + }, + { + "epoch": 0.05877587507333017, + "grad_norm": 0.5713419914245605, + "learning_rate": 0.000973470416853429, + "loss": 0.4738, + "step": 1052 + }, + { + "epoch": 0.05883174567701204, + "grad_norm": 0.7842467427253723, + "learning_rate": 0.0009734424025100852, + "loss": 0.576, + "step": 1053 + }, + { + "epoch": 0.058887616280693916, + "grad_norm": 0.8829406499862671, + "learning_rate": 0.0009734143881667414, + "loss": 0.5712, + "step": 1054 + }, + { + "epoch": 0.058943486884375784, + "grad_norm": 1.6040854454040527, + "learning_rate": 0.0009733863738233976, + "loss": 0.5176, + "step": 1055 + }, + { + "epoch": 0.05899935748805766, + "grad_norm": 1.3363285064697266, + "learning_rate": 0.0009733583594800538, + "loss": 0.4759, + "step": 1056 + }, + { + "epoch": 0.059055228091739534, + "grad_norm": 0.8210630416870117, + "learning_rate": 0.00097333034513671, + "loss": 0.7796, + "step": 1057 + }, + { + "epoch": 0.0591110986954214, + "grad_norm": 0.6840836405754089, + "learning_rate": 0.0009733023307933662, + "loss": 0.4715, + "step": 1058 + }, + { + "epoch": 0.05916696929910328, + "grad_norm": 0.5459524393081665, + "learning_rate": 0.0009732743164500224, + "loss": 0.4141, + "step": 1059 + }, + { + "epoch": 0.05922283990278515, + "grad_norm": 0.566616952419281, + "learning_rate": 0.0009732463021066786, + "loss": 0.4986, + "step": 1060 + }, + { + "epoch": 0.05927871050646702, + "grad_norm": 0.47836244106292725, + "learning_rate": 0.0009732182877633348, + "loss": 0.4169, + "step": 1061 + }, + { + "epoch": 0.059334581110148894, + "grad_norm": 0.809704065322876, + "learning_rate": 0.000973190273419991, + "loss": 0.5159, + "step": 1062 + }, + { + "epoch": 0.05939045171383077, + "grad_norm": 0.6466063261032104, + "learning_rate": 0.0009731622590766473, + "loss": 0.526, + "step": 1063 + }, + { + "epoch": 0.059446322317512644, + "grad_norm": 0.835321843624115, + "learning_rate": 0.0009731342447333035, + "loss": 0.6012, + "step": 1064 + }, + { + "epoch": 0.05950219292119451, + "grad_norm": 0.5955824851989746, + "learning_rate": 0.0009731062303899597, + "loss": 0.4066, + "step": 1065 + }, + { + "epoch": 0.05955806352487639, + "grad_norm": 0.6803643107414246, + "learning_rate": 0.0009730782160466159, + "loss": 0.4303, + "step": 1066 + }, + { + "epoch": 0.05961393412855826, + "grad_norm": 0.7364829778671265, + "learning_rate": 0.0009730502017032721, + "loss": 0.44, + "step": 1067 + }, + { + "epoch": 0.05966980473224013, + "grad_norm": 1.254533052444458, + "learning_rate": 0.0009730221873599283, + "loss": 0.5532, + "step": 1068 + }, + { + "epoch": 0.059725675335922004, + "grad_norm": 1.0624057054519653, + "learning_rate": 0.0009729941730165846, + "loss": 0.5257, + "step": 1069 + }, + { + "epoch": 0.05978154593960388, + "grad_norm": 1.037156105041504, + "learning_rate": 0.0009729661586732407, + "loss": 0.5358, + "step": 1070 + }, + { + "epoch": 0.05983741654328575, + "grad_norm": 0.7669567465782166, + "learning_rate": 0.0009729381443298969, + "loss": 0.4965, + "step": 1071 + }, + { + "epoch": 0.05989328714696762, + "grad_norm": 0.6366451382637024, + "learning_rate": 0.0009729101299865531, + "loss": 0.5863, + "step": 1072 + }, + { + "epoch": 0.0599491577506495, + "grad_norm": 0.5739390254020691, + "learning_rate": 0.0009728821156432094, + "loss": 0.5313, + "step": 1073 + }, + { + "epoch": 0.06000502835433137, + "grad_norm": 0.590941846370697, + "learning_rate": 0.0009728541012998655, + "loss": 0.4795, + "step": 1074 + }, + { + "epoch": 0.06006089895801324, + "grad_norm": 0.5513612627983093, + "learning_rate": 0.0009728260869565217, + "loss": 0.5558, + "step": 1075 + }, + { + "epoch": 0.060116769561695114, + "grad_norm": 0.778820276260376, + "learning_rate": 0.000972798072613178, + "loss": 0.4781, + "step": 1076 + }, + { + "epoch": 0.06017264016537699, + "grad_norm": 0.5139995813369751, + "learning_rate": 0.0009727700582698342, + "loss": 0.3939, + "step": 1077 + }, + { + "epoch": 0.06022851076905886, + "grad_norm": 1.5144503116607666, + "learning_rate": 0.0009727420439264903, + "loss": 0.6981, + "step": 1078 + }, + { + "epoch": 0.06028438137274073, + "grad_norm": 0.6025422811508179, + "learning_rate": 0.0009727140295831465, + "loss": 0.5265, + "step": 1079 + }, + { + "epoch": 0.06034025197642261, + "grad_norm": 1.1337858438491821, + "learning_rate": 0.0009726860152398029, + "loss": 0.4545, + "step": 1080 + }, + { + "epoch": 0.060396122580104475, + "grad_norm": 0.6767749786376953, + "learning_rate": 0.0009726580008964591, + "loss": 0.6161, + "step": 1081 + }, + { + "epoch": 0.06045199318378635, + "grad_norm": 0.6324553489685059, + "learning_rate": 0.0009726299865531152, + "loss": 0.5021, + "step": 1082 + }, + { + "epoch": 0.060507863787468225, + "grad_norm": 0.5466465353965759, + "learning_rate": 0.0009726019722097715, + "loss": 0.5293, + "step": 1083 + }, + { + "epoch": 0.0605637343911501, + "grad_norm": 0.5013379454612732, + "learning_rate": 0.0009725739578664277, + "loss": 0.448, + "step": 1084 + }, + { + "epoch": 0.06061960499483197, + "grad_norm": 0.5834270119667053, + "learning_rate": 0.0009725459435230838, + "loss": 0.3835, + "step": 1085 + }, + { + "epoch": 0.06067547559851384, + "grad_norm": 1.0084789991378784, + "learning_rate": 0.00097251792917974, + "loss": 0.3961, + "step": 1086 + }, + { + "epoch": 0.06073134620219572, + "grad_norm": 0.5152300000190735, + "learning_rate": 0.0009724899148363963, + "loss": 0.4313, + "step": 1087 + }, + { + "epoch": 0.060787216805877585, + "grad_norm": 0.8065328001976013, + "learning_rate": 0.0009724619004930525, + "loss": 0.6032, + "step": 1088 + }, + { + "epoch": 0.06084308740955946, + "grad_norm": 0.3815320134162903, + "learning_rate": 0.0009724338861497086, + "loss": 0.3795, + "step": 1089 + }, + { + "epoch": 0.060898958013241335, + "grad_norm": 0.6035824418067932, + "learning_rate": 0.0009724058718063648, + "loss": 0.5561, + "step": 1090 + }, + { + "epoch": 0.0609548286169232, + "grad_norm": 0.6146941781044006, + "learning_rate": 0.0009723778574630211, + "loss": 0.4683, + "step": 1091 + }, + { + "epoch": 0.06101069922060508, + "grad_norm": 0.7265572547912598, + "learning_rate": 0.0009723498431196773, + "loss": 0.6106, + "step": 1092 + }, + { + "epoch": 0.06106656982428695, + "grad_norm": 0.979194164276123, + "learning_rate": 0.0009723218287763334, + "loss": 0.3949, + "step": 1093 + }, + { + "epoch": 0.06112244042796883, + "grad_norm": 0.5127589702606201, + "learning_rate": 0.0009722938144329897, + "loss": 0.5179, + "step": 1094 + }, + { + "epoch": 0.061178311031650695, + "grad_norm": 0.598968505859375, + "learning_rate": 0.000972265800089646, + "loss": 0.5791, + "step": 1095 + }, + { + "epoch": 0.06123418163533257, + "grad_norm": 0.6822913885116577, + "learning_rate": 0.0009722377857463022, + "loss": 0.5226, + "step": 1096 + }, + { + "epoch": 0.061290052239014445, + "grad_norm": 1.329878568649292, + "learning_rate": 0.0009722097714029582, + "loss": 0.4954, + "step": 1097 + }, + { + "epoch": 0.06134592284269631, + "grad_norm": 0.6904491782188416, + "learning_rate": 0.0009721817570596146, + "loss": 0.563, + "step": 1098 + }, + { + "epoch": 0.06140179344637819, + "grad_norm": 0.6387484669685364, + "learning_rate": 0.0009721537427162708, + "loss": 0.5517, + "step": 1099 + }, + { + "epoch": 0.06145766405006006, + "grad_norm": 0.9023336172103882, + "learning_rate": 0.000972125728372927, + "loss": 0.4967, + "step": 1100 + }, + { + "epoch": 0.06151353465374193, + "grad_norm": 0.6719636917114258, + "learning_rate": 0.0009720977140295832, + "loss": 0.6642, + "step": 1101 + }, + { + "epoch": 0.061569405257423805, + "grad_norm": 0.5953622460365295, + "learning_rate": 0.0009720696996862394, + "loss": 0.4338, + "step": 1102 + }, + { + "epoch": 0.06162527586110568, + "grad_norm": 0.887468695640564, + "learning_rate": 0.0009720416853428956, + "loss": 0.6183, + "step": 1103 + }, + { + "epoch": 0.061681146464787555, + "grad_norm": 0.481361985206604, + "learning_rate": 0.0009720136709995517, + "loss": 0.3661, + "step": 1104 + }, + { + "epoch": 0.06173701706846942, + "grad_norm": 0.47323089838027954, + "learning_rate": 0.000971985656656208, + "loss": 0.4965, + "step": 1105 + }, + { + "epoch": 0.0617928876721513, + "grad_norm": 0.8812860250473022, + "learning_rate": 0.0009719576423128642, + "loss": 0.6927, + "step": 1106 + }, + { + "epoch": 0.06184875827583317, + "grad_norm": 0.5371952056884766, + "learning_rate": 0.0009719296279695204, + "loss": 0.4427, + "step": 1107 + }, + { + "epoch": 0.06190462887951504, + "grad_norm": 0.7554811835289001, + "learning_rate": 0.0009719016136261766, + "loss": 0.5645, + "step": 1108 + }, + { + "epoch": 0.061960499483196915, + "grad_norm": 0.6391088366508484, + "learning_rate": 0.0009718735992828328, + "loss": 0.5993, + "step": 1109 + }, + { + "epoch": 0.06201637008687879, + "grad_norm": 1.1213533878326416, + "learning_rate": 0.000971845584939489, + "loss": 0.5212, + "step": 1110 + }, + { + "epoch": 0.062072240690560665, + "grad_norm": 1.012993335723877, + "learning_rate": 0.0009718175705961452, + "loss": 0.517, + "step": 1111 + }, + { + "epoch": 0.06212811129424253, + "grad_norm": 2.9934747219085693, + "learning_rate": 0.0009717895562528014, + "loss": 0.4676, + "step": 1112 + }, + { + "epoch": 0.06218398189792441, + "grad_norm": 0.49386894702911377, + "learning_rate": 0.0009717615419094577, + "loss": 0.4301, + "step": 1113 + }, + { + "epoch": 0.06223985250160628, + "grad_norm": 0.6627538204193115, + "learning_rate": 0.0009717335275661139, + "loss": 0.5147, + "step": 1114 + }, + { + "epoch": 0.06229572310528815, + "grad_norm": 0.5267593264579773, + "learning_rate": 0.0009717055132227702, + "loss": 0.4645, + "step": 1115 + }, + { + "epoch": 0.062351593708970025, + "grad_norm": 2.0643975734710693, + "learning_rate": 0.0009716774988794263, + "loss": 0.4926, + "step": 1116 + }, + { + "epoch": 0.0624074643126519, + "grad_norm": 0.6692291498184204, + "learning_rate": 0.0009716494845360825, + "loss": 0.4709, + "step": 1117 + }, + { + "epoch": 0.06246333491633377, + "grad_norm": 0.5622457265853882, + "learning_rate": 0.0009716214701927387, + "loss": 0.5449, + "step": 1118 + }, + { + "epoch": 0.06251920552001565, + "grad_norm": 1.173712968826294, + "learning_rate": 0.000971593455849395, + "loss": 0.5135, + "step": 1119 + }, + { + "epoch": 0.06257507612369752, + "grad_norm": 0.8020060658454895, + "learning_rate": 0.0009715654415060511, + "loss": 0.34, + "step": 1120 + }, + { + "epoch": 0.06263094672737939, + "grad_norm": 0.8978903293609619, + "learning_rate": 0.0009715374271627073, + "loss": 0.5565, + "step": 1121 + }, + { + "epoch": 0.06268681733106127, + "grad_norm": 0.9518550634384155, + "learning_rate": 0.0009715094128193636, + "loss": 0.5309, + "step": 1122 + }, + { + "epoch": 0.06274268793474314, + "grad_norm": 0.9943126440048218, + "learning_rate": 0.0009714813984760198, + "loss": 0.479, + "step": 1123 + }, + { + "epoch": 0.062798558538425, + "grad_norm": 0.4428074359893799, + "learning_rate": 0.0009714533841326759, + "loss": 0.4512, + "step": 1124 + }, + { + "epoch": 0.06285442914210689, + "grad_norm": 2.7266159057617188, + "learning_rate": 0.0009714253697893321, + "loss": 0.4748, + "step": 1125 + }, + { + "epoch": 0.06291029974578875, + "grad_norm": 0.7351318001747131, + "learning_rate": 0.0009713973554459884, + "loss": 0.4678, + "step": 1126 + }, + { + "epoch": 0.06296617034947062, + "grad_norm": 0.8597860336303711, + "learning_rate": 0.0009713693411026445, + "loss": 0.4941, + "step": 1127 + }, + { + "epoch": 0.0630220409531525, + "grad_norm": 0.8028839230537415, + "learning_rate": 0.0009713413267593007, + "loss": 0.5883, + "step": 1128 + }, + { + "epoch": 0.06307791155683437, + "grad_norm": 0.44469234347343445, + "learning_rate": 0.0009713133124159569, + "loss": 0.4157, + "step": 1129 + }, + { + "epoch": 0.06313378216051624, + "grad_norm": 0.7365114688873291, + "learning_rate": 0.0009712852980726133, + "loss": 0.5481, + "step": 1130 + }, + { + "epoch": 0.06318965276419812, + "grad_norm": 0.8004930019378662, + "learning_rate": 0.0009712572837292694, + "loss": 0.5308, + "step": 1131 + }, + { + "epoch": 0.06324552336787999, + "grad_norm": 2.7498955726623535, + "learning_rate": 0.0009712292693859256, + "loss": 0.6472, + "step": 1132 + }, + { + "epoch": 0.06330139397156186, + "grad_norm": 1.0613384246826172, + "learning_rate": 0.0009712012550425819, + "loss": 0.3946, + "step": 1133 + }, + { + "epoch": 0.06335726457524374, + "grad_norm": 0.6960083842277527, + "learning_rate": 0.0009711732406992381, + "loss": 0.739, + "step": 1134 + }, + { + "epoch": 0.0634131351789256, + "grad_norm": 1.1086087226867676, + "learning_rate": 0.0009711452263558942, + "loss": 0.548, + "step": 1135 + }, + { + "epoch": 0.06346900578260749, + "grad_norm": 0.6563412547111511, + "learning_rate": 0.0009711172120125504, + "loss": 0.4572, + "step": 1136 + }, + { + "epoch": 0.06352487638628936, + "grad_norm": 0.7006518840789795, + "learning_rate": 0.0009710891976692067, + "loss": 0.4928, + "step": 1137 + }, + { + "epoch": 0.06358074698997122, + "grad_norm": 0.5440847873687744, + "learning_rate": 0.0009710611833258629, + "loss": 0.4697, + "step": 1138 + }, + { + "epoch": 0.0636366175936531, + "grad_norm": 0.6437940001487732, + "learning_rate": 0.000971033168982519, + "loss": 0.5296, + "step": 1139 + }, + { + "epoch": 0.06369248819733497, + "grad_norm": 0.6402298808097839, + "learning_rate": 0.0009710051546391753, + "loss": 0.4117, + "step": 1140 + }, + { + "epoch": 0.06374835880101684, + "grad_norm": 0.7072002291679382, + "learning_rate": 0.0009709771402958315, + "loss": 0.5065, + "step": 1141 + }, + { + "epoch": 0.06380422940469872, + "grad_norm": 1.007352352142334, + "learning_rate": 0.0009709491259524877, + "loss": 0.6588, + "step": 1142 + }, + { + "epoch": 0.06386010000838059, + "grad_norm": 0.663611114025116, + "learning_rate": 0.0009709211116091438, + "loss": 0.4858, + "step": 1143 + }, + { + "epoch": 0.06391597061206246, + "grad_norm": 2.1629860401153564, + "learning_rate": 0.0009708930972658001, + "loss": 0.4044, + "step": 1144 + }, + { + "epoch": 0.06397184121574434, + "grad_norm": 0.6593952178955078, + "learning_rate": 0.0009708650829224563, + "loss": 0.5337, + "step": 1145 + }, + { + "epoch": 0.06402771181942621, + "grad_norm": 0.8183286786079407, + "learning_rate": 0.0009708370685791126, + "loss": 0.5489, + "step": 1146 + }, + { + "epoch": 0.06408358242310808, + "grad_norm": 3.0181124210357666, + "learning_rate": 0.0009708090542357688, + "loss": 0.4546, + "step": 1147 + }, + { + "epoch": 0.06413945302678996, + "grad_norm": 0.5256533622741699, + "learning_rate": 0.000970781039892425, + "loss": 0.4775, + "step": 1148 + }, + { + "epoch": 0.06419532363047183, + "grad_norm": 0.8322650790214539, + "learning_rate": 0.0009707530255490812, + "loss": 0.5289, + "step": 1149 + }, + { + "epoch": 0.0642511942341537, + "grad_norm": 0.8450261354446411, + "learning_rate": 0.0009707250112057373, + "loss": 0.5505, + "step": 1150 + }, + { + "epoch": 0.06430706483783558, + "grad_norm": 0.8847916722297668, + "learning_rate": 0.0009706969968623936, + "loss": 0.603, + "step": 1151 + }, + { + "epoch": 0.06436293544151744, + "grad_norm": 0.9955880045890808, + "learning_rate": 0.0009706689825190498, + "loss": 0.471, + "step": 1152 + }, + { + "epoch": 0.06441880604519931, + "grad_norm": 3.089514970779419, + "learning_rate": 0.000970640968175706, + "loss": 0.5375, + "step": 1153 + }, + { + "epoch": 0.0644746766488812, + "grad_norm": 0.8168195486068726, + "learning_rate": 0.0009706129538323622, + "loss": 0.4405, + "step": 1154 + }, + { + "epoch": 0.06453054725256306, + "grad_norm": 0.5857439637184143, + "learning_rate": 0.0009705849394890184, + "loss": 0.6646, + "step": 1155 + }, + { + "epoch": 0.06458641785624494, + "grad_norm": 1.1598776578903198, + "learning_rate": 0.0009705569251456746, + "loss": 0.5229, + "step": 1156 + }, + { + "epoch": 0.06464228845992681, + "grad_norm": 0.7277916669845581, + "learning_rate": 0.0009705289108023308, + "loss": 0.4939, + "step": 1157 + }, + { + "epoch": 0.06469815906360868, + "grad_norm": 0.48309823870658875, + "learning_rate": 0.000970500896458987, + "loss": 0.3657, + "step": 1158 + }, + { + "epoch": 0.06475402966729056, + "grad_norm": 1.323746919631958, + "learning_rate": 0.0009704728821156432, + "loss": 0.624, + "step": 1159 + }, + { + "epoch": 0.06480990027097243, + "grad_norm": 0.6194313764572144, + "learning_rate": 0.0009704448677722994, + "loss": 0.4135, + "step": 1160 + }, + { + "epoch": 0.0648657708746543, + "grad_norm": 0.8351618051528931, + "learning_rate": 0.0009704168534289556, + "loss": 0.5919, + "step": 1161 + }, + { + "epoch": 0.06492164147833618, + "grad_norm": 1.249163269996643, + "learning_rate": 0.0009703888390856118, + "loss": 0.6759, + "step": 1162 + }, + { + "epoch": 0.06497751208201805, + "grad_norm": 0.6550889611244202, + "learning_rate": 0.000970360824742268, + "loss": 0.5296, + "step": 1163 + }, + { + "epoch": 0.06503338268569991, + "grad_norm": 0.5061567425727844, + "learning_rate": 0.0009703328103989243, + "loss": 0.4631, + "step": 1164 + }, + { + "epoch": 0.0650892532893818, + "grad_norm": 2.4988107681274414, + "learning_rate": 0.0009703047960555806, + "loss": 0.4812, + "step": 1165 + }, + { + "epoch": 0.06514512389306366, + "grad_norm": 0.9078261256217957, + "learning_rate": 0.0009702767817122367, + "loss": 0.5127, + "step": 1166 + }, + { + "epoch": 0.06520099449674553, + "grad_norm": 0.8619847893714905, + "learning_rate": 0.0009702487673688929, + "loss": 0.5234, + "step": 1167 + }, + { + "epoch": 0.06525686510042741, + "grad_norm": 2.309140205383301, + "learning_rate": 0.0009702207530255491, + "loss": 0.4662, + "step": 1168 + }, + { + "epoch": 0.06531273570410928, + "grad_norm": 0.772172212600708, + "learning_rate": 0.0009701927386822053, + "loss": 0.4619, + "step": 1169 + }, + { + "epoch": 0.06536860630779115, + "grad_norm": 0.5016564130783081, + "learning_rate": 0.0009701647243388615, + "loss": 0.4559, + "step": 1170 + }, + { + "epoch": 0.06542447691147303, + "grad_norm": 0.7117054462432861, + "learning_rate": 0.0009701367099955177, + "loss": 0.5263, + "step": 1171 + }, + { + "epoch": 0.0654803475151549, + "grad_norm": 1.3959567546844482, + "learning_rate": 0.000970108695652174, + "loss": 0.6543, + "step": 1172 + }, + { + "epoch": 0.06553621811883678, + "grad_norm": 0.48802393674850464, + "learning_rate": 0.0009700806813088301, + "loss": 0.514, + "step": 1173 + }, + { + "epoch": 0.06559208872251865, + "grad_norm": 0.6388691067695618, + "learning_rate": 0.0009700526669654863, + "loss": 0.6083, + "step": 1174 + }, + { + "epoch": 0.06564795932620052, + "grad_norm": 0.6451505422592163, + "learning_rate": 0.0009700246526221425, + "loss": 0.4863, + "step": 1175 + }, + { + "epoch": 0.0657038299298824, + "grad_norm": 0.5991710424423218, + "learning_rate": 0.0009699966382787988, + "loss": 0.5122, + "step": 1176 + }, + { + "epoch": 0.06575970053356427, + "grad_norm": 0.522226870059967, + "learning_rate": 0.0009699686239354549, + "loss": 0.4633, + "step": 1177 + }, + { + "epoch": 0.06581557113724613, + "grad_norm": 0.6942064166069031, + "learning_rate": 0.0009699406095921111, + "loss": 0.4613, + "step": 1178 + }, + { + "epoch": 0.06587144174092802, + "grad_norm": 0.6091275811195374, + "learning_rate": 0.0009699125952487675, + "loss": 0.5571, + "step": 1179 + }, + { + "epoch": 0.06592731234460988, + "grad_norm": 0.9745811820030212, + "learning_rate": 0.0009698845809054237, + "loss": 0.4585, + "step": 1180 + }, + { + "epoch": 0.06598318294829175, + "grad_norm": 0.579214334487915, + "learning_rate": 0.0009698565665620798, + "loss": 0.5267, + "step": 1181 + }, + { + "epoch": 0.06603905355197363, + "grad_norm": 1.1098637580871582, + "learning_rate": 0.000969828552218736, + "loss": 0.6237, + "step": 1182 + }, + { + "epoch": 0.0660949241556555, + "grad_norm": 3.7098143100738525, + "learning_rate": 0.0009698005378753923, + "loss": 0.3988, + "step": 1183 + }, + { + "epoch": 0.06615079475933737, + "grad_norm": 1.7550063133239746, + "learning_rate": 0.0009697725235320485, + "loss": 0.5375, + "step": 1184 + }, + { + "epoch": 0.06620666536301925, + "grad_norm": 0.925864577293396, + "learning_rate": 0.0009697445091887046, + "loss": 0.5509, + "step": 1185 + }, + { + "epoch": 0.06626253596670112, + "grad_norm": 0.706131637096405, + "learning_rate": 0.0009697164948453609, + "loss": 0.5471, + "step": 1186 + }, + { + "epoch": 0.06631840657038299, + "grad_norm": 0.6336811184883118, + "learning_rate": 0.0009696884805020171, + "loss": 0.5522, + "step": 1187 + }, + { + "epoch": 0.06637427717406487, + "grad_norm": 2.48801326751709, + "learning_rate": 0.0009696604661586733, + "loss": 0.507, + "step": 1188 + }, + { + "epoch": 0.06643014777774674, + "grad_norm": 0.6308658719062805, + "learning_rate": 0.0009696324518153294, + "loss": 0.452, + "step": 1189 + }, + { + "epoch": 0.0664860183814286, + "grad_norm": 0.9892592430114746, + "learning_rate": 0.0009696044374719857, + "loss": 0.6458, + "step": 1190 + }, + { + "epoch": 0.06654188898511049, + "grad_norm": 0.9263991117477417, + "learning_rate": 0.0009695764231286419, + "loss": 0.6044, + "step": 1191 + }, + { + "epoch": 0.06659775958879235, + "grad_norm": 0.7117300629615784, + "learning_rate": 0.000969548408785298, + "loss": 0.5438, + "step": 1192 + }, + { + "epoch": 0.06665363019247424, + "grad_norm": 1.2078739404678345, + "learning_rate": 0.0009695203944419542, + "loss": 0.6428, + "step": 1193 + }, + { + "epoch": 0.0667095007961561, + "grad_norm": 2.500253438949585, + "learning_rate": 0.0009694923800986105, + "loss": 0.5683, + "step": 1194 + }, + { + "epoch": 0.06676537139983797, + "grad_norm": 0.7970782518386841, + "learning_rate": 0.0009694643657552667, + "loss": 0.6003, + "step": 1195 + }, + { + "epoch": 0.06682124200351985, + "grad_norm": 0.7644243240356445, + "learning_rate": 0.0009694363514119228, + "loss": 0.4923, + "step": 1196 + }, + { + "epoch": 0.06687711260720172, + "grad_norm": 0.8672170639038086, + "learning_rate": 0.0009694083370685792, + "loss": 0.4618, + "step": 1197 + }, + { + "epoch": 0.06693298321088359, + "grad_norm": 0.5016088485717773, + "learning_rate": 0.0009693803227252354, + "loss": 0.4164, + "step": 1198 + }, + { + "epoch": 0.06698885381456547, + "grad_norm": 0.8622342348098755, + "learning_rate": 0.0009693523083818916, + "loss": 0.6488, + "step": 1199 + }, + { + "epoch": 0.06704472441824734, + "grad_norm": 2.4121992588043213, + "learning_rate": 0.0009693242940385477, + "loss": 0.5578, + "step": 1200 + }, + { + "epoch": 0.06710059502192921, + "grad_norm": 8.130661964416504, + "learning_rate": 0.000969296279695204, + "loss": 0.4534, + "step": 1201 + }, + { + "epoch": 0.06715646562561109, + "grad_norm": 0.7272051572799683, + "learning_rate": 0.0009692682653518602, + "loss": 0.5708, + "step": 1202 + }, + { + "epoch": 0.06721233622929296, + "grad_norm": 2.6462111473083496, + "learning_rate": 0.0009692402510085164, + "loss": 0.5122, + "step": 1203 + }, + { + "epoch": 0.06726820683297483, + "grad_norm": 1.1371543407440186, + "learning_rate": 0.0009692122366651726, + "loss": 0.6024, + "step": 1204 + }, + { + "epoch": 0.0673240774366567, + "grad_norm": 0.6209547519683838, + "learning_rate": 0.0009691842223218288, + "loss": 0.4228, + "step": 1205 + }, + { + "epoch": 0.06737994804033857, + "grad_norm": 0.5680100321769714, + "learning_rate": 0.000969156207978485, + "loss": 0.5932, + "step": 1206 + }, + { + "epoch": 0.06743581864402044, + "grad_norm": 0.6599418520927429, + "learning_rate": 0.0009691281936351412, + "loss": 0.556, + "step": 1207 + }, + { + "epoch": 0.06749168924770232, + "grad_norm": 0.5719356536865234, + "learning_rate": 0.0009691001792917974, + "loss": 0.3357, + "step": 1208 + }, + { + "epoch": 0.06754755985138419, + "grad_norm": 0.42119333148002625, + "learning_rate": 0.0009690721649484536, + "loss": 0.5105, + "step": 1209 + }, + { + "epoch": 0.06760343045506606, + "grad_norm": 0.8622509837150574, + "learning_rate": 0.0009690441506051098, + "loss": 0.5419, + "step": 1210 + }, + { + "epoch": 0.06765930105874794, + "grad_norm": 0.5874972939491272, + "learning_rate": 0.000969016136261766, + "loss": 0.5236, + "step": 1211 + }, + { + "epoch": 0.06771517166242981, + "grad_norm": 0.7327354550361633, + "learning_rate": 0.0009689881219184222, + "loss": 0.5397, + "step": 1212 + }, + { + "epoch": 0.06777104226611169, + "grad_norm": 0.846717894077301, + "learning_rate": 0.0009689601075750784, + "loss": 0.4633, + "step": 1213 + }, + { + "epoch": 0.06782691286979356, + "grad_norm": 0.6208289861679077, + "learning_rate": 0.0009689320932317347, + "loss": 0.4489, + "step": 1214 + }, + { + "epoch": 0.06788278347347543, + "grad_norm": 0.3897058665752411, + "learning_rate": 0.0009689040788883909, + "loss": 0.401, + "step": 1215 + }, + { + "epoch": 0.06793865407715731, + "grad_norm": 20.8198184967041, + "learning_rate": 0.0009688760645450471, + "loss": 0.7987, + "step": 1216 + }, + { + "epoch": 0.06799452468083918, + "grad_norm": 0.4787755310535431, + "learning_rate": 0.0009688480502017033, + "loss": 0.4451, + "step": 1217 + }, + { + "epoch": 0.06805039528452105, + "grad_norm": 0.5368200540542603, + "learning_rate": 0.0009688200358583596, + "loss": 0.3796, + "step": 1218 + }, + { + "epoch": 0.06810626588820293, + "grad_norm": 7.289865016937256, + "learning_rate": 0.0009687920215150157, + "loss": 0.5184, + "step": 1219 + }, + { + "epoch": 0.0681621364918848, + "grad_norm": 0.9263502955436707, + "learning_rate": 0.0009687640071716719, + "loss": 0.6686, + "step": 1220 + }, + { + "epoch": 0.06821800709556666, + "grad_norm": 1.039231300354004, + "learning_rate": 0.0009687359928283281, + "loss": 0.5081, + "step": 1221 + }, + { + "epoch": 0.06827387769924854, + "grad_norm": 0.8952227830886841, + "learning_rate": 0.0009687079784849844, + "loss": 0.5286, + "step": 1222 + }, + { + "epoch": 0.06832974830293041, + "grad_norm": 1.48212468624115, + "learning_rate": 0.0009686799641416405, + "loss": 0.4698, + "step": 1223 + }, + { + "epoch": 0.06838561890661228, + "grad_norm": 0.7009533643722534, + "learning_rate": 0.0009686519497982967, + "loss": 0.6543, + "step": 1224 + }, + { + "epoch": 0.06844148951029416, + "grad_norm": 0.6321099996566772, + "learning_rate": 0.000968623935454953, + "loss": 0.5413, + "step": 1225 + }, + { + "epoch": 0.06849736011397603, + "grad_norm": 0.6894535422325134, + "learning_rate": 0.0009685959211116092, + "loss": 0.6345, + "step": 1226 + }, + { + "epoch": 0.0685532307176579, + "grad_norm": 0.42889484763145447, + "learning_rate": 0.0009685679067682653, + "loss": 0.3779, + "step": 1227 + }, + { + "epoch": 0.06860910132133978, + "grad_norm": 0.788731038570404, + "learning_rate": 0.0009685398924249215, + "loss": 0.39, + "step": 1228 + }, + { + "epoch": 0.06866497192502165, + "grad_norm": 1.3431590795516968, + "learning_rate": 0.0009685118780815779, + "loss": 0.8308, + "step": 1229 + }, + { + "epoch": 0.06872084252870352, + "grad_norm": 0.5714833736419678, + "learning_rate": 0.0009684838637382341, + "loss": 0.5093, + "step": 1230 + }, + { + "epoch": 0.0687767131323854, + "grad_norm": 0.9321240186691284, + "learning_rate": 0.0009684558493948902, + "loss": 0.5846, + "step": 1231 + }, + { + "epoch": 0.06883258373606727, + "grad_norm": 0.84440678358078, + "learning_rate": 0.0009684278350515464, + "loss": 0.5013, + "step": 1232 + }, + { + "epoch": 0.06888845433974915, + "grad_norm": 6.071931838989258, + "learning_rate": 0.0009683998207082027, + "loss": 0.3998, + "step": 1233 + }, + { + "epoch": 0.06894432494343102, + "grad_norm": 0.8287359476089478, + "learning_rate": 0.0009683718063648588, + "loss": 0.5303, + "step": 1234 + }, + { + "epoch": 0.06900019554711288, + "grad_norm": 0.6752516031265259, + "learning_rate": 0.000968343792021515, + "loss": 0.4851, + "step": 1235 + }, + { + "epoch": 0.06905606615079476, + "grad_norm": 1.6673821210861206, + "learning_rate": 0.0009683157776781713, + "loss": 0.5359, + "step": 1236 + }, + { + "epoch": 0.06911193675447663, + "grad_norm": 0.5790602564811707, + "learning_rate": 0.0009682877633348275, + "loss": 0.5083, + "step": 1237 + }, + { + "epoch": 0.0691678073581585, + "grad_norm": 8.803149223327637, + "learning_rate": 0.0009682597489914836, + "loss": 0.5225, + "step": 1238 + }, + { + "epoch": 0.06922367796184038, + "grad_norm": 0.8842300772666931, + "learning_rate": 0.0009682317346481398, + "loss": 0.5516, + "step": 1239 + }, + { + "epoch": 0.06927954856552225, + "grad_norm": 0.6785323619842529, + "learning_rate": 0.0009682037203047961, + "loss": 0.4935, + "step": 1240 + }, + { + "epoch": 0.06933541916920412, + "grad_norm": 0.6587780714035034, + "learning_rate": 0.0009681757059614523, + "loss": 0.5336, + "step": 1241 + }, + { + "epoch": 0.069391289772886, + "grad_norm": 0.8125738501548767, + "learning_rate": 0.0009681476916181084, + "loss": 0.4752, + "step": 1242 + }, + { + "epoch": 0.06944716037656787, + "grad_norm": 1.1262879371643066, + "learning_rate": 0.0009681196772747647, + "loss": 0.5611, + "step": 1243 + }, + { + "epoch": 0.06950303098024974, + "grad_norm": 4.192417621612549, + "learning_rate": 0.0009680916629314209, + "loss": 0.4705, + "step": 1244 + }, + { + "epoch": 0.06955890158393162, + "grad_norm": 0.6851063966751099, + "learning_rate": 0.0009680636485880771, + "loss": 0.5234, + "step": 1245 + }, + { + "epoch": 0.06961477218761349, + "grad_norm": 0.7175766229629517, + "learning_rate": 0.0009680356342447332, + "loss": 0.5196, + "step": 1246 + }, + { + "epoch": 0.06967064279129535, + "grad_norm": 0.5043138265609741, + "learning_rate": 0.0009680076199013896, + "loss": 0.5559, + "step": 1247 + }, + { + "epoch": 0.06972651339497724, + "grad_norm": 0.6681637167930603, + "learning_rate": 0.0009679796055580458, + "loss": 0.4515, + "step": 1248 + }, + { + "epoch": 0.0697823839986591, + "grad_norm": 1.2017042636871338, + "learning_rate": 0.000967951591214702, + "loss": 0.38, + "step": 1249 + }, + { + "epoch": 0.06983825460234099, + "grad_norm": 0.9148383736610413, + "learning_rate": 0.0009679235768713582, + "loss": 0.7768, + "step": 1250 + }, + { + "epoch": 0.06989412520602285, + "grad_norm": 0.6964145302772522, + "learning_rate": 0.0009678955625280144, + "loss": 0.5194, + "step": 1251 + }, + { + "epoch": 0.06994999580970472, + "grad_norm": 1.6189056634902954, + "learning_rate": 0.0009678675481846706, + "loss": 0.5738, + "step": 1252 + }, + { + "epoch": 0.0700058664133866, + "grad_norm": 0.6723500490188599, + "learning_rate": 0.0009678395338413267, + "loss": 0.5103, + "step": 1253 + }, + { + "epoch": 0.07006173701706847, + "grad_norm": 2.489980697631836, + "learning_rate": 0.000967811519497983, + "loss": 0.441, + "step": 1254 + }, + { + "epoch": 0.07011760762075034, + "grad_norm": 2.046182632446289, + "learning_rate": 0.0009677835051546392, + "loss": 0.6445, + "step": 1255 + }, + { + "epoch": 0.07017347822443222, + "grad_norm": 0.8560983538627625, + "learning_rate": 0.0009677554908112954, + "loss": 0.6323, + "step": 1256 + }, + { + "epoch": 0.07022934882811409, + "grad_norm": 0.5849862098693848, + "learning_rate": 0.0009677274764679516, + "loss": 0.4718, + "step": 1257 + }, + { + "epoch": 0.07028521943179596, + "grad_norm": 10.700162887573242, + "learning_rate": 0.0009676994621246078, + "loss": 0.5355, + "step": 1258 + }, + { + "epoch": 0.07034109003547784, + "grad_norm": 0.5089840292930603, + "learning_rate": 0.000967671447781264, + "loss": 0.592, + "step": 1259 + }, + { + "epoch": 0.0703969606391597, + "grad_norm": 0.4630272388458252, + "learning_rate": 0.0009676434334379202, + "loss": 0.4896, + "step": 1260 + }, + { + "epoch": 0.07045283124284157, + "grad_norm": 0.7387524843215942, + "learning_rate": 0.0009676154190945764, + "loss": 0.5431, + "step": 1261 + }, + { + "epoch": 0.07050870184652346, + "grad_norm": 0.4200060963630676, + "learning_rate": 0.0009675874047512326, + "loss": 0.4549, + "step": 1262 + }, + { + "epoch": 0.07056457245020532, + "grad_norm": 0.7487926483154297, + "learning_rate": 0.0009675593904078888, + "loss": 0.5624, + "step": 1263 + }, + { + "epoch": 0.07062044305388719, + "grad_norm": 0.8511855602264404, + "learning_rate": 0.0009675313760645452, + "loss": 0.6804, + "step": 1264 + }, + { + "epoch": 0.07067631365756907, + "grad_norm": 0.6215517520904541, + "learning_rate": 0.0009675033617212013, + "loss": 0.4511, + "step": 1265 + }, + { + "epoch": 0.07073218426125094, + "grad_norm": 0.8066599369049072, + "learning_rate": 0.0009674753473778575, + "loss": 0.4563, + "step": 1266 + }, + { + "epoch": 0.07078805486493281, + "grad_norm": 0.4807688295841217, + "learning_rate": 0.0009674473330345137, + "loss": 0.4222, + "step": 1267 + }, + { + "epoch": 0.07084392546861469, + "grad_norm": 0.5622011423110962, + "learning_rate": 0.00096741931869117, + "loss": 0.451, + "step": 1268 + }, + { + "epoch": 0.07089979607229656, + "grad_norm": 0.6550637483596802, + "learning_rate": 0.0009673913043478261, + "loss": 0.497, + "step": 1269 + }, + { + "epoch": 0.07095566667597844, + "grad_norm": 0.847633421421051, + "learning_rate": 0.0009673632900044823, + "loss": 0.6262, + "step": 1270 + }, + { + "epoch": 0.07101153727966031, + "grad_norm": 0.6108511090278625, + "learning_rate": 0.0009673352756611385, + "loss": 0.5073, + "step": 1271 + }, + { + "epoch": 0.07106740788334218, + "grad_norm": 0.8942492604255676, + "learning_rate": 0.0009673072613177948, + "loss": 0.5503, + "step": 1272 + }, + { + "epoch": 0.07112327848702406, + "grad_norm": 0.6260313391685486, + "learning_rate": 0.0009672792469744509, + "loss": 0.5097, + "step": 1273 + }, + { + "epoch": 0.07117914909070593, + "grad_norm": 6.479348659515381, + "learning_rate": 0.0009672512326311071, + "loss": 0.3964, + "step": 1274 + }, + { + "epoch": 0.0712350196943878, + "grad_norm": 0.9336758852005005, + "learning_rate": 0.0009672232182877634, + "loss": 0.6467, + "step": 1275 + }, + { + "epoch": 0.07129089029806968, + "grad_norm": 1.024034857749939, + "learning_rate": 0.0009671952039444195, + "loss": 0.5328, + "step": 1276 + }, + { + "epoch": 0.07134676090175154, + "grad_norm": 0.8046055436134338, + "learning_rate": 0.0009671671896010757, + "loss": 0.6214, + "step": 1277 + }, + { + "epoch": 0.07140263150543341, + "grad_norm": 0.9241152405738831, + "learning_rate": 0.0009671391752577319, + "loss": 0.625, + "step": 1278 + }, + { + "epoch": 0.0714585021091153, + "grad_norm": 1.1843887567520142, + "learning_rate": 0.0009671111609143883, + "loss": 0.5007, + "step": 1279 + }, + { + "epoch": 0.07151437271279716, + "grad_norm": 0.9318219423294067, + "learning_rate": 0.0009670831465710443, + "loss": 0.5044, + "step": 1280 + }, + { + "epoch": 0.07157024331647903, + "grad_norm": 4.944368839263916, + "learning_rate": 0.0009670551322277006, + "loss": 0.4029, + "step": 1281 + }, + { + "epoch": 0.07162611392016091, + "grad_norm": 0.6926432847976685, + "learning_rate": 0.0009670271178843569, + "loss": 0.4747, + "step": 1282 + }, + { + "epoch": 0.07168198452384278, + "grad_norm": 1.047566533088684, + "learning_rate": 0.0009669991035410131, + "loss": 0.633, + "step": 1283 + }, + { + "epoch": 0.07173785512752465, + "grad_norm": 1.0467345714569092, + "learning_rate": 0.0009669710891976692, + "loss": 0.3863, + "step": 1284 + }, + { + "epoch": 0.07179372573120653, + "grad_norm": 0.9288132786750793, + "learning_rate": 0.0009669430748543254, + "loss": 0.4647, + "step": 1285 + }, + { + "epoch": 0.0718495963348884, + "grad_norm": 0.9064891934394836, + "learning_rate": 0.0009669150605109817, + "loss": 0.4831, + "step": 1286 + }, + { + "epoch": 0.07190546693857026, + "grad_norm": 0.46407249569892883, + "learning_rate": 0.0009668870461676379, + "loss": 0.4109, + "step": 1287 + }, + { + "epoch": 0.07196133754225215, + "grad_norm": 0.7585373520851135, + "learning_rate": 0.000966859031824294, + "loss": 0.5672, + "step": 1288 + }, + { + "epoch": 0.07201720814593401, + "grad_norm": 0.587788999080658, + "learning_rate": 0.0009668310174809503, + "loss": 0.4842, + "step": 1289 + }, + { + "epoch": 0.0720730787496159, + "grad_norm": 1.48751699924469, + "learning_rate": 0.0009668030031376065, + "loss": 0.4632, + "step": 1290 + }, + { + "epoch": 0.07212894935329776, + "grad_norm": 1.6349979639053345, + "learning_rate": 0.0009667749887942627, + "loss": 0.5163, + "step": 1291 + }, + { + "epoch": 0.07218481995697963, + "grad_norm": 0.980557918548584, + "learning_rate": 0.0009667469744509188, + "loss": 0.4207, + "step": 1292 + }, + { + "epoch": 0.07224069056066151, + "grad_norm": 0.7738677859306335, + "learning_rate": 0.0009667189601075751, + "loss": 0.5293, + "step": 1293 + }, + { + "epoch": 0.07229656116434338, + "grad_norm": 0.6839911937713623, + "learning_rate": 0.0009666909457642313, + "loss": 0.5595, + "step": 1294 + }, + { + "epoch": 0.07235243176802525, + "grad_norm": 0.6894443035125732, + "learning_rate": 0.0009666629314208874, + "loss": 0.6232, + "step": 1295 + }, + { + "epoch": 0.07240830237170713, + "grad_norm": 0.7277626991271973, + "learning_rate": 0.0009666349170775437, + "loss": 0.5498, + "step": 1296 + }, + { + "epoch": 0.072464172975389, + "grad_norm": 0.6097222566604614, + "learning_rate": 0.0009666069027342, + "loss": 0.4972, + "step": 1297 + }, + { + "epoch": 0.07252004357907087, + "grad_norm": 0.7629479169845581, + "learning_rate": 0.0009665788883908562, + "loss": 0.4634, + "step": 1298 + }, + { + "epoch": 0.07257591418275275, + "grad_norm": 0.9895792603492737, + "learning_rate": 0.0009665508740475123, + "loss": 0.6401, + "step": 1299 + }, + { + "epoch": 0.07263178478643462, + "grad_norm": 0.4240557849407196, + "learning_rate": 0.0009665228597041686, + "loss": 0.4038, + "step": 1300 + }, + { + "epoch": 0.07268765539011648, + "grad_norm": 1.2062939405441284, + "learning_rate": 0.0009664948453608248, + "loss": 0.4244, + "step": 1301 + }, + { + "epoch": 0.07274352599379837, + "grad_norm": 0.4923010468482971, + "learning_rate": 0.000966466831017481, + "loss": 0.5001, + "step": 1302 + }, + { + "epoch": 0.07279939659748023, + "grad_norm": 0.48886317014694214, + "learning_rate": 0.0009664388166741371, + "loss": 0.4004, + "step": 1303 + }, + { + "epoch": 0.0728552672011621, + "grad_norm": 0.6174327731132507, + "learning_rate": 0.0009664108023307934, + "loss": 0.5747, + "step": 1304 + }, + { + "epoch": 0.07291113780484398, + "grad_norm": 0.5241720676422119, + "learning_rate": 0.0009663827879874496, + "loss": 0.4349, + "step": 1305 + }, + { + "epoch": 0.07296700840852585, + "grad_norm": 0.5976858139038086, + "learning_rate": 0.0009663547736441058, + "loss": 0.4744, + "step": 1306 + }, + { + "epoch": 0.07302287901220773, + "grad_norm": 1.8489171266555786, + "learning_rate": 0.000966326759300762, + "loss": 0.5552, + "step": 1307 + }, + { + "epoch": 0.0730787496158896, + "grad_norm": 0.6435372233390808, + "learning_rate": 0.0009662987449574182, + "loss": 0.5385, + "step": 1308 + }, + { + "epoch": 0.07313462021957147, + "grad_norm": 0.42348724603652954, + "learning_rate": 0.0009662707306140744, + "loss": 0.4817, + "step": 1309 + }, + { + "epoch": 0.07319049082325335, + "grad_norm": 1.0893926620483398, + "learning_rate": 0.0009662427162707306, + "loss": 0.6853, + "step": 1310 + }, + { + "epoch": 0.07324636142693522, + "grad_norm": 0.7394664287567139, + "learning_rate": 0.0009662147019273868, + "loss": 0.5692, + "step": 1311 + }, + { + "epoch": 0.07330223203061709, + "grad_norm": 0.7010294795036316, + "learning_rate": 0.000966186687584043, + "loss": 0.4602, + "step": 1312 + }, + { + "epoch": 0.07335810263429897, + "grad_norm": 0.6346744298934937, + "learning_rate": 0.0009661586732406992, + "loss": 0.4864, + "step": 1313 + }, + { + "epoch": 0.07341397323798084, + "grad_norm": 0.44749829173088074, + "learning_rate": 0.0009661306588973556, + "loss": 0.4948, + "step": 1314 + }, + { + "epoch": 0.0734698438416627, + "grad_norm": 0.5430852770805359, + "learning_rate": 0.0009661026445540117, + "loss": 0.5211, + "step": 1315 + }, + { + "epoch": 0.07352571444534459, + "grad_norm": 0.7119840383529663, + "learning_rate": 0.0009660746302106679, + "loss": 0.4491, + "step": 1316 + }, + { + "epoch": 0.07358158504902645, + "grad_norm": 0.5116697549819946, + "learning_rate": 0.0009660466158673241, + "loss": 0.411, + "step": 1317 + }, + { + "epoch": 0.07363745565270832, + "grad_norm": 5.524214267730713, + "learning_rate": 0.0009660186015239803, + "loss": 0.569, + "step": 1318 + }, + { + "epoch": 0.0736933262563902, + "grad_norm": 0.6921236515045166, + "learning_rate": 0.0009659905871806365, + "loss": 0.4611, + "step": 1319 + }, + { + "epoch": 0.07374919686007207, + "grad_norm": 0.44938212633132935, + "learning_rate": 0.0009659625728372927, + "loss": 0.4657, + "step": 1320 + }, + { + "epoch": 0.07380506746375394, + "grad_norm": 0.523338794708252, + "learning_rate": 0.000965934558493949, + "loss": 0.5463, + "step": 1321 + }, + { + "epoch": 0.07386093806743582, + "grad_norm": 0.5819428563117981, + "learning_rate": 0.0009659065441506051, + "loss": 0.5117, + "step": 1322 + }, + { + "epoch": 0.07391680867111769, + "grad_norm": 2.0423853397369385, + "learning_rate": 0.0009658785298072613, + "loss": 0.5076, + "step": 1323 + }, + { + "epoch": 0.07397267927479956, + "grad_norm": 0.9363688826560974, + "learning_rate": 0.0009658505154639175, + "loss": 0.4682, + "step": 1324 + }, + { + "epoch": 0.07402854987848144, + "grad_norm": 0.5844001770019531, + "learning_rate": 0.0009658225011205738, + "loss": 0.4656, + "step": 1325 + }, + { + "epoch": 0.07408442048216331, + "grad_norm": 0.9558589458465576, + "learning_rate": 0.0009657944867772299, + "loss": 0.431, + "step": 1326 + }, + { + "epoch": 0.07414029108584519, + "grad_norm": 0.7594606876373291, + "learning_rate": 0.0009657664724338861, + "loss": 0.4385, + "step": 1327 + }, + { + "epoch": 0.07419616168952706, + "grad_norm": 0.6759605407714844, + "learning_rate": 0.0009657384580905424, + "loss": 0.63, + "step": 1328 + }, + { + "epoch": 0.07425203229320892, + "grad_norm": 1.4086092710494995, + "learning_rate": 0.0009657104437471986, + "loss": 0.5669, + "step": 1329 + }, + { + "epoch": 0.0743079028968908, + "grad_norm": 0.4309212565422058, + "learning_rate": 0.0009656824294038547, + "loss": 0.4939, + "step": 1330 + }, + { + "epoch": 0.07436377350057267, + "grad_norm": 2.4984352588653564, + "learning_rate": 0.000965654415060511, + "loss": 0.4581, + "step": 1331 + }, + { + "epoch": 0.07441964410425454, + "grad_norm": 1.0026131868362427, + "learning_rate": 0.0009656264007171673, + "loss": 0.5504, + "step": 1332 + }, + { + "epoch": 0.07447551470793642, + "grad_norm": 0.5754634141921997, + "learning_rate": 0.0009655983863738235, + "loss": 0.4874, + "step": 1333 + }, + { + "epoch": 0.07453138531161829, + "grad_norm": 1.0713260173797607, + "learning_rate": 0.0009655703720304796, + "loss": 0.5534, + "step": 1334 + }, + { + "epoch": 0.07458725591530016, + "grad_norm": 0.823697566986084, + "learning_rate": 0.0009655423576871359, + "loss": 0.4524, + "step": 1335 + }, + { + "epoch": 0.07464312651898204, + "grad_norm": 0.4016875922679901, + "learning_rate": 0.0009655143433437921, + "loss": 0.4262, + "step": 1336 + }, + { + "epoch": 0.07469899712266391, + "grad_norm": 0.5814744830131531, + "learning_rate": 0.0009654863290004483, + "loss": 0.592, + "step": 1337 + }, + { + "epoch": 0.07475486772634578, + "grad_norm": 0.49748697876930237, + "learning_rate": 0.0009654583146571044, + "loss": 0.4209, + "step": 1338 + }, + { + "epoch": 0.07481073833002766, + "grad_norm": 0.4803832471370697, + "learning_rate": 0.0009654303003137607, + "loss": 0.4715, + "step": 1339 + }, + { + "epoch": 0.07486660893370953, + "grad_norm": 0.6011812686920166, + "learning_rate": 0.0009654022859704169, + "loss": 0.5038, + "step": 1340 + }, + { + "epoch": 0.0749224795373914, + "grad_norm": 0.4870539903640747, + "learning_rate": 0.000965374271627073, + "loss": 0.4128, + "step": 1341 + }, + { + "epoch": 0.07497835014107328, + "grad_norm": 0.4284798800945282, + "learning_rate": 0.0009653462572837292, + "loss": 0.3794, + "step": 1342 + }, + { + "epoch": 0.07503422074475515, + "grad_norm": 0.7192260026931763, + "learning_rate": 0.0009653182429403855, + "loss": 0.5258, + "step": 1343 + }, + { + "epoch": 0.07509009134843701, + "grad_norm": 0.934248685836792, + "learning_rate": 0.0009652902285970417, + "loss": 0.4238, + "step": 1344 + }, + { + "epoch": 0.0751459619521189, + "grad_norm": 0.5323062539100647, + "learning_rate": 0.0009652622142536978, + "loss": 0.4995, + "step": 1345 + }, + { + "epoch": 0.07520183255580076, + "grad_norm": 0.6467271447181702, + "learning_rate": 0.0009652341999103541, + "loss": 0.5907, + "step": 1346 + }, + { + "epoch": 0.07525770315948264, + "grad_norm": 0.6167634725570679, + "learning_rate": 0.0009652061855670104, + "loss": 0.4117, + "step": 1347 + }, + { + "epoch": 0.07531357376316451, + "grad_norm": 1.4055026769638062, + "learning_rate": 0.0009651781712236666, + "loss": 0.4578, + "step": 1348 + }, + { + "epoch": 0.07536944436684638, + "grad_norm": 0.9854038953781128, + "learning_rate": 0.0009651501568803227, + "loss": 0.5926, + "step": 1349 + }, + { + "epoch": 0.07542531497052826, + "grad_norm": 0.5910872220993042, + "learning_rate": 0.000965122142536979, + "loss": 0.4246, + "step": 1350 + }, + { + "epoch": 0.07548118557421013, + "grad_norm": 0.4961700439453125, + "learning_rate": 0.0009650941281936352, + "loss": 0.5265, + "step": 1351 + }, + { + "epoch": 0.075537056177892, + "grad_norm": 0.5632983446121216, + "learning_rate": 0.0009650661138502914, + "loss": 0.3895, + "step": 1352 + }, + { + "epoch": 0.07559292678157388, + "grad_norm": 0.6158227324485779, + "learning_rate": 0.0009650380995069476, + "loss": 0.5178, + "step": 1353 + }, + { + "epoch": 0.07564879738525575, + "grad_norm": 0.5182850956916809, + "learning_rate": 0.0009650100851636038, + "loss": 0.4789, + "step": 1354 + }, + { + "epoch": 0.07570466798893762, + "grad_norm": 0.7551424503326416, + "learning_rate": 0.00096498207082026, + "loss": 0.4135, + "step": 1355 + }, + { + "epoch": 0.0757605385926195, + "grad_norm": 0.6314374208450317, + "learning_rate": 0.0009649540564769162, + "loss": 0.6033, + "step": 1356 + }, + { + "epoch": 0.07581640919630137, + "grad_norm": 0.8133394718170166, + "learning_rate": 0.0009649260421335724, + "loss": 0.4931, + "step": 1357 + }, + { + "epoch": 0.07587227979998323, + "grad_norm": 0.5776365399360657, + "learning_rate": 0.0009648980277902286, + "loss": 0.4897, + "step": 1358 + }, + { + "epoch": 0.07592815040366512, + "grad_norm": 16.084747314453125, + "learning_rate": 0.0009648700134468848, + "loss": 0.4725, + "step": 1359 + }, + { + "epoch": 0.07598402100734698, + "grad_norm": 0.696034848690033, + "learning_rate": 0.000964841999103541, + "loss": 0.531, + "step": 1360 + }, + { + "epoch": 0.07603989161102885, + "grad_norm": 0.7954675555229187, + "learning_rate": 0.0009648139847601972, + "loss": 0.5287, + "step": 1361 + }, + { + "epoch": 0.07609576221471073, + "grad_norm": 0.7137690186500549, + "learning_rate": 0.0009647859704168534, + "loss": 0.4793, + "step": 1362 + }, + { + "epoch": 0.0761516328183926, + "grad_norm": 0.5315437316894531, + "learning_rate": 0.0009647579560735096, + "loss": 0.5379, + "step": 1363 + }, + { + "epoch": 0.07620750342207448, + "grad_norm": 0.7810925841331482, + "learning_rate": 0.0009647299417301659, + "loss": 0.4357, + "step": 1364 + }, + { + "epoch": 0.07626337402575635, + "grad_norm": 5.620739936828613, + "learning_rate": 0.0009647019273868221, + "loss": 0.5188, + "step": 1365 + }, + { + "epoch": 0.07631924462943822, + "grad_norm": 0.885621190071106, + "learning_rate": 0.0009646739130434783, + "loss": 0.4428, + "step": 1366 + }, + { + "epoch": 0.0763751152331201, + "grad_norm": 0.49245819449424744, + "learning_rate": 0.0009646458987001346, + "loss": 0.4801, + "step": 1367 + }, + { + "epoch": 0.07643098583680197, + "grad_norm": 0.917472779750824, + "learning_rate": 0.0009646178843567907, + "loss": 0.4559, + "step": 1368 + }, + { + "epoch": 0.07648685644048384, + "grad_norm": 2.686610221862793, + "learning_rate": 0.0009645898700134469, + "loss": 0.4168, + "step": 1369 + }, + { + "epoch": 0.07654272704416572, + "grad_norm": 0.9760270118713379, + "learning_rate": 0.0009645618556701031, + "loss": 0.5405, + "step": 1370 + }, + { + "epoch": 0.07659859764784759, + "grad_norm": 2.681776285171509, + "learning_rate": 0.0009645338413267594, + "loss": 0.5739, + "step": 1371 + }, + { + "epoch": 0.07665446825152945, + "grad_norm": 0.5609073638916016, + "learning_rate": 0.0009645058269834155, + "loss": 0.4609, + "step": 1372 + }, + { + "epoch": 0.07671033885521134, + "grad_norm": 0.5058500170707703, + "learning_rate": 0.0009644778126400717, + "loss": 0.4116, + "step": 1373 + }, + { + "epoch": 0.0767662094588932, + "grad_norm": 2.946739673614502, + "learning_rate": 0.0009644497982967279, + "loss": 0.4777, + "step": 1374 + }, + { + "epoch": 0.07682208006257507, + "grad_norm": 0.6807973980903625, + "learning_rate": 0.0009644217839533842, + "loss": 0.5116, + "step": 1375 + }, + { + "epoch": 0.07687795066625695, + "grad_norm": 3.994993209838867, + "learning_rate": 0.0009643937696100403, + "loss": 0.4072, + "step": 1376 + }, + { + "epoch": 0.07693382126993882, + "grad_norm": 0.6262413263320923, + "learning_rate": 0.0009643657552666965, + "loss": 0.6187, + "step": 1377 + }, + { + "epoch": 0.07698969187362069, + "grad_norm": 0.6260105967521667, + "learning_rate": 0.0009643377409233528, + "loss": 0.4466, + "step": 1378 + }, + { + "epoch": 0.07704556247730257, + "grad_norm": 0.7468571662902832, + "learning_rate": 0.000964309726580009, + "loss": 0.5639, + "step": 1379 + }, + { + "epoch": 0.07710143308098444, + "grad_norm": 1.5208033323287964, + "learning_rate": 0.0009642817122366651, + "loss": 0.5069, + "step": 1380 + }, + { + "epoch": 0.0771573036846663, + "grad_norm": 0.5963055491447449, + "learning_rate": 0.0009642536978933214, + "loss": 0.6465, + "step": 1381 + }, + { + "epoch": 0.07721317428834819, + "grad_norm": 0.4807388186454773, + "learning_rate": 0.0009642256835499777, + "loss": 0.5121, + "step": 1382 + }, + { + "epoch": 0.07726904489203006, + "grad_norm": 0.5590099692344666, + "learning_rate": 0.0009641976692066338, + "loss": 0.3498, + "step": 1383 + }, + { + "epoch": 0.07732491549571194, + "grad_norm": 0.4971422255039215, + "learning_rate": 0.00096416965486329, + "loss": 0.4872, + "step": 1384 + }, + { + "epoch": 0.0773807860993938, + "grad_norm": 0.5557582378387451, + "learning_rate": 0.0009641416405199463, + "loss": 0.4482, + "step": 1385 + }, + { + "epoch": 0.07743665670307567, + "grad_norm": 0.6781041622161865, + "learning_rate": 0.0009641136261766025, + "loss": 0.5547, + "step": 1386 + }, + { + "epoch": 0.07749252730675756, + "grad_norm": 1.3873612880706787, + "learning_rate": 0.0009640856118332586, + "loss": 0.479, + "step": 1387 + }, + { + "epoch": 0.07754839791043942, + "grad_norm": 0.8447802066802979, + "learning_rate": 0.0009640575974899148, + "loss": 0.7012, + "step": 1388 + }, + { + "epoch": 0.07760426851412129, + "grad_norm": 0.7521064877510071, + "learning_rate": 0.0009640295831465711, + "loss": 0.5051, + "step": 1389 + }, + { + "epoch": 0.07766013911780317, + "grad_norm": 0.7333371639251709, + "learning_rate": 0.0009640015688032273, + "loss": 0.6047, + "step": 1390 + }, + { + "epoch": 0.07771600972148504, + "grad_norm": 0.7882974743843079, + "learning_rate": 0.0009639735544598834, + "loss": 0.5385, + "step": 1391 + }, + { + "epoch": 0.07777188032516691, + "grad_norm": 1.4541822671890259, + "learning_rate": 0.0009639455401165397, + "loss": 0.7965, + "step": 1392 + }, + { + "epoch": 0.07782775092884879, + "grad_norm": 0.884488582611084, + "learning_rate": 0.0009639175257731959, + "loss": 0.7085, + "step": 1393 + }, + { + "epoch": 0.07788362153253066, + "grad_norm": 1.0330965518951416, + "learning_rate": 0.0009638895114298521, + "loss": 0.5441, + "step": 1394 + }, + { + "epoch": 0.07793949213621253, + "grad_norm": 4.929499626159668, + "learning_rate": 0.0009638614970865082, + "loss": 0.4424, + "step": 1395 + }, + { + "epoch": 0.07799536273989441, + "grad_norm": 4.726205825805664, + "learning_rate": 0.0009638334827431645, + "loss": 0.5488, + "step": 1396 + }, + { + "epoch": 0.07805123334357628, + "grad_norm": 0.6969701051712036, + "learning_rate": 0.0009638054683998208, + "loss": 0.5653, + "step": 1397 + }, + { + "epoch": 0.07810710394725814, + "grad_norm": 0.7058079838752747, + "learning_rate": 0.000963777454056477, + "loss": 0.4313, + "step": 1398 + }, + { + "epoch": 0.07816297455094003, + "grad_norm": 0.5591409802436829, + "learning_rate": 0.0009637494397131332, + "loss": 0.4637, + "step": 1399 + }, + { + "epoch": 0.0782188451546219, + "grad_norm": 0.6536149978637695, + "learning_rate": 0.0009637214253697894, + "loss": 0.5135, + "step": 1400 + }, + { + "epoch": 0.07827471575830376, + "grad_norm": 6.4326653480529785, + "learning_rate": 0.0009636934110264456, + "loss": 0.5392, + "step": 1401 + }, + { + "epoch": 0.07833058636198564, + "grad_norm": 0.715863823890686, + "learning_rate": 0.0009636653966831017, + "loss": 0.4962, + "step": 1402 + }, + { + "epoch": 0.07838645696566751, + "grad_norm": 0.6246368288993835, + "learning_rate": 0.000963637382339758, + "loss": 0.5337, + "step": 1403 + }, + { + "epoch": 0.0784423275693494, + "grad_norm": 0.640389084815979, + "learning_rate": 0.0009636093679964142, + "loss": 0.5326, + "step": 1404 + }, + { + "epoch": 0.07849819817303126, + "grad_norm": 1.1883822679519653, + "learning_rate": 0.0009635813536530704, + "loss": 0.5937, + "step": 1405 + }, + { + "epoch": 0.07855406877671313, + "grad_norm": 1.3205671310424805, + "learning_rate": 0.0009635533393097265, + "loss": 0.5032, + "step": 1406 + }, + { + "epoch": 0.07860993938039501, + "grad_norm": 1.0281234979629517, + "learning_rate": 0.0009635253249663828, + "loss": 0.633, + "step": 1407 + }, + { + "epoch": 0.07866580998407688, + "grad_norm": 0.7705580592155457, + "learning_rate": 0.000963497310623039, + "loss": 0.5152, + "step": 1408 + }, + { + "epoch": 0.07872168058775875, + "grad_norm": 1.5164954662322998, + "learning_rate": 0.0009634692962796952, + "loss": 0.5539, + "step": 1409 + }, + { + "epoch": 0.07877755119144063, + "grad_norm": 0.49744343757629395, + "learning_rate": 0.0009634412819363514, + "loss": 0.5389, + "step": 1410 + }, + { + "epoch": 0.0788334217951225, + "grad_norm": 0.7388191223144531, + "learning_rate": 0.0009634132675930076, + "loss": 0.4103, + "step": 1411 + }, + { + "epoch": 0.07888929239880436, + "grad_norm": 0.608299195766449, + "learning_rate": 0.0009633852532496638, + "loss": 0.4068, + "step": 1412 + }, + { + "epoch": 0.07894516300248625, + "grad_norm": 0.6682078242301941, + "learning_rate": 0.00096335723890632, + "loss": 0.4552, + "step": 1413 + }, + { + "epoch": 0.07900103360616811, + "grad_norm": 0.9223785996437073, + "learning_rate": 0.0009633292245629763, + "loss": 0.5034, + "step": 1414 + }, + { + "epoch": 0.07905690420984998, + "grad_norm": 2.39399790763855, + "learning_rate": 0.0009633012102196325, + "loss": 0.4984, + "step": 1415 + }, + { + "epoch": 0.07911277481353186, + "grad_norm": 0.8522182106971741, + "learning_rate": 0.0009632731958762887, + "loss": 0.4984, + "step": 1416 + }, + { + "epoch": 0.07916864541721373, + "grad_norm": 3.276881217956543, + "learning_rate": 0.000963245181532945, + "loss": 0.6917, + "step": 1417 + }, + { + "epoch": 0.0792245160208956, + "grad_norm": 0.8046137094497681, + "learning_rate": 0.0009632171671896011, + "loss": 0.5247, + "step": 1418 + }, + { + "epoch": 0.07928038662457748, + "grad_norm": 1.0491377115249634, + "learning_rate": 0.0009631891528462573, + "loss": 0.5896, + "step": 1419 + }, + { + "epoch": 0.07933625722825935, + "grad_norm": 0.5174822211265564, + "learning_rate": 0.0009631611385029135, + "loss": 0.5198, + "step": 1420 + }, + { + "epoch": 0.07939212783194123, + "grad_norm": 1.5504688024520874, + "learning_rate": 0.0009631331241595698, + "loss": 0.4888, + "step": 1421 + }, + { + "epoch": 0.0794479984356231, + "grad_norm": 0.8341609239578247, + "learning_rate": 0.0009631051098162259, + "loss": 0.4885, + "step": 1422 + }, + { + "epoch": 0.07950386903930497, + "grad_norm": 1.0982465744018555, + "learning_rate": 0.0009630770954728821, + "loss": 0.5509, + "step": 1423 + }, + { + "epoch": 0.07955973964298685, + "grad_norm": 0.47819799184799194, + "learning_rate": 0.0009630490811295384, + "loss": 0.5444, + "step": 1424 + }, + { + "epoch": 0.07961561024666872, + "grad_norm": 0.820235013961792, + "learning_rate": 0.0009630210667861945, + "loss": 0.4354, + "step": 1425 + }, + { + "epoch": 0.07967148085035058, + "grad_norm": 0.7967373132705688, + "learning_rate": 0.0009629930524428507, + "loss": 0.5273, + "step": 1426 + }, + { + "epoch": 0.07972735145403247, + "grad_norm": 0.7325090169906616, + "learning_rate": 0.0009629650380995069, + "loss": 0.4884, + "step": 1427 + }, + { + "epoch": 0.07978322205771433, + "grad_norm": 0.8626403212547302, + "learning_rate": 0.0009629370237561632, + "loss": 0.4541, + "step": 1428 + }, + { + "epoch": 0.0798390926613962, + "grad_norm": 0.523154616355896, + "learning_rate": 0.0009629090094128193, + "loss": 0.5423, + "step": 1429 + }, + { + "epoch": 0.07989496326507808, + "grad_norm": 0.6090967059135437, + "learning_rate": 0.0009628809950694755, + "loss": 0.5127, + "step": 1430 + }, + { + "epoch": 0.07995083386875995, + "grad_norm": 0.5714813470840454, + "learning_rate": 0.0009628529807261319, + "loss": 0.481, + "step": 1431 + }, + { + "epoch": 0.08000670447244182, + "grad_norm": 2.10780930519104, + "learning_rate": 0.0009628249663827881, + "loss": 0.4566, + "step": 1432 + }, + { + "epoch": 0.0800625750761237, + "grad_norm": 7.895953178405762, + "learning_rate": 0.0009627969520394442, + "loss": 0.5284, + "step": 1433 + }, + { + "epoch": 0.08011844567980557, + "grad_norm": 0.6845932006835938, + "learning_rate": 0.0009627689376961004, + "loss": 0.5771, + "step": 1434 + }, + { + "epoch": 0.08017431628348744, + "grad_norm": 0.8416215181350708, + "learning_rate": 0.0009627409233527567, + "loss": 0.5883, + "step": 1435 + }, + { + "epoch": 0.08023018688716932, + "grad_norm": 3.294485092163086, + "learning_rate": 0.0009627129090094129, + "loss": 0.5231, + "step": 1436 + }, + { + "epoch": 0.08028605749085119, + "grad_norm": 0.7762507200241089, + "learning_rate": 0.000962684894666069, + "loss": 0.5524, + "step": 1437 + }, + { + "epoch": 0.08034192809453305, + "grad_norm": 0.4604811370372772, + "learning_rate": 0.0009626568803227253, + "loss": 0.3249, + "step": 1438 + }, + { + "epoch": 0.08039779869821494, + "grad_norm": 0.4450690746307373, + "learning_rate": 0.0009626288659793815, + "loss": 0.5287, + "step": 1439 + }, + { + "epoch": 0.0804536693018968, + "grad_norm": 6.871273994445801, + "learning_rate": 0.0009626008516360377, + "loss": 0.4245, + "step": 1440 + }, + { + "epoch": 0.08050953990557869, + "grad_norm": 0.6615837216377258, + "learning_rate": 0.0009625728372926938, + "loss": 0.4446, + "step": 1441 + }, + { + "epoch": 0.08056541050926055, + "grad_norm": 0.704426646232605, + "learning_rate": 0.0009625448229493501, + "loss": 0.4134, + "step": 1442 + }, + { + "epoch": 0.08062128111294242, + "grad_norm": 1.2423841953277588, + "learning_rate": 0.0009625168086060063, + "loss": 0.4695, + "step": 1443 + }, + { + "epoch": 0.0806771517166243, + "grad_norm": 0.6669187545776367, + "learning_rate": 0.0009624887942626624, + "loss": 0.5122, + "step": 1444 + }, + { + "epoch": 0.08073302232030617, + "grad_norm": 0.5991784930229187, + "learning_rate": 0.0009624607799193186, + "loss": 0.4153, + "step": 1445 + }, + { + "epoch": 0.08078889292398804, + "grad_norm": 1.0039913654327393, + "learning_rate": 0.000962432765575975, + "loss": 0.5162, + "step": 1446 + }, + { + "epoch": 0.08084476352766992, + "grad_norm": 0.5419370532035828, + "learning_rate": 0.0009624047512326312, + "loss": 0.4244, + "step": 1447 + }, + { + "epoch": 0.08090063413135179, + "grad_norm": 0.5997844338417053, + "learning_rate": 0.0009623767368892872, + "loss": 0.4642, + "step": 1448 + }, + { + "epoch": 0.08095650473503366, + "grad_norm": 0.46032610535621643, + "learning_rate": 0.0009623487225459436, + "loss": 0.3524, + "step": 1449 + }, + { + "epoch": 0.08101237533871554, + "grad_norm": 0.6728903651237488, + "learning_rate": 0.0009623207082025998, + "loss": 0.4617, + "step": 1450 + }, + { + "epoch": 0.08106824594239741, + "grad_norm": 0.6810144186019897, + "learning_rate": 0.000962292693859256, + "loss": 0.6325, + "step": 1451 + }, + { + "epoch": 0.08112411654607928, + "grad_norm": 0.7356053590774536, + "learning_rate": 0.0009622646795159121, + "loss": 0.4519, + "step": 1452 + }, + { + "epoch": 0.08117998714976116, + "grad_norm": 0.7296345829963684, + "learning_rate": 0.0009622366651725684, + "loss": 0.4671, + "step": 1453 + }, + { + "epoch": 0.08123585775344302, + "grad_norm": 0.47174689173698425, + "learning_rate": 0.0009622086508292246, + "loss": 0.4546, + "step": 1454 + }, + { + "epoch": 0.08129172835712489, + "grad_norm": 0.9482269883155823, + "learning_rate": 0.0009621806364858808, + "loss": 0.5904, + "step": 1455 + }, + { + "epoch": 0.08134759896080677, + "grad_norm": 0.7149969339370728, + "learning_rate": 0.000962152622142537, + "loss": 0.3862, + "step": 1456 + }, + { + "epoch": 0.08140346956448864, + "grad_norm": 0.7780554294586182, + "learning_rate": 0.0009621246077991932, + "loss": 0.5152, + "step": 1457 + }, + { + "epoch": 0.08145934016817051, + "grad_norm": 0.6226730942726135, + "learning_rate": 0.0009620965934558494, + "loss": 0.6833, + "step": 1458 + }, + { + "epoch": 0.08151521077185239, + "grad_norm": 0.5256891250610352, + "learning_rate": 0.0009620685791125056, + "loss": 0.3955, + "step": 1459 + }, + { + "epoch": 0.08157108137553426, + "grad_norm": 1.058059573173523, + "learning_rate": 0.0009620405647691618, + "loss": 0.4702, + "step": 1460 + }, + { + "epoch": 0.08162695197921614, + "grad_norm": 0.703818678855896, + "learning_rate": 0.000962012550425818, + "loss": 0.6472, + "step": 1461 + }, + { + "epoch": 0.08168282258289801, + "grad_norm": 2.7310357093811035, + "learning_rate": 0.0009619845360824742, + "loss": 0.5397, + "step": 1462 + }, + { + "epoch": 0.08173869318657988, + "grad_norm": 0.48122259974479675, + "learning_rate": 0.0009619565217391306, + "loss": 0.4844, + "step": 1463 + }, + { + "epoch": 0.08179456379026176, + "grad_norm": 0.861792266368866, + "learning_rate": 0.0009619285073957867, + "loss": 0.4311, + "step": 1464 + }, + { + "epoch": 0.08185043439394363, + "grad_norm": 0.741489589214325, + "learning_rate": 0.0009619004930524429, + "loss": 0.5841, + "step": 1465 + }, + { + "epoch": 0.0819063049976255, + "grad_norm": 0.7590195536613464, + "learning_rate": 0.0009618724787090991, + "loss": 0.4628, + "step": 1466 + }, + { + "epoch": 0.08196217560130738, + "grad_norm": 1.0360260009765625, + "learning_rate": 0.0009618444643657553, + "loss": 0.5117, + "step": 1467 + }, + { + "epoch": 0.08201804620498925, + "grad_norm": 0.7268709540367126, + "learning_rate": 0.0009618164500224115, + "loss": 0.5005, + "step": 1468 + }, + { + "epoch": 0.08207391680867111, + "grad_norm": 1.0228300094604492, + "learning_rate": 0.0009617884356790677, + "loss": 0.5201, + "step": 1469 + }, + { + "epoch": 0.082129787412353, + "grad_norm": 0.7765328288078308, + "learning_rate": 0.000961760421335724, + "loss": 0.5283, + "step": 1470 + }, + { + "epoch": 0.08218565801603486, + "grad_norm": 0.519669234752655, + "learning_rate": 0.0009617324069923801, + "loss": 0.4651, + "step": 1471 + }, + { + "epoch": 0.08224152861971673, + "grad_norm": 1.2036023139953613, + "learning_rate": 0.0009617043926490363, + "loss": 0.5742, + "step": 1472 + }, + { + "epoch": 0.08229739922339861, + "grad_norm": 0.6704846620559692, + "learning_rate": 0.0009616763783056925, + "loss": 0.5608, + "step": 1473 + }, + { + "epoch": 0.08235326982708048, + "grad_norm": 0.7023825645446777, + "learning_rate": 0.0009616483639623488, + "loss": 0.6217, + "step": 1474 + }, + { + "epoch": 0.08240914043076235, + "grad_norm": 0.5282848477363586, + "learning_rate": 0.0009616203496190049, + "loss": 0.4899, + "step": 1475 + }, + { + "epoch": 0.08246501103444423, + "grad_norm": 0.5925781726837158, + "learning_rate": 0.0009615923352756611, + "loss": 0.4475, + "step": 1476 + }, + { + "epoch": 0.0825208816381261, + "grad_norm": 0.5560790300369263, + "learning_rate": 0.0009615643209323174, + "loss": 0.4608, + "step": 1477 + }, + { + "epoch": 0.08257675224180797, + "grad_norm": 2.7434537410736084, + "learning_rate": 0.0009615363065889736, + "loss": 0.4804, + "step": 1478 + }, + { + "epoch": 0.08263262284548985, + "grad_norm": 0.48386484384536743, + "learning_rate": 0.0009615082922456297, + "loss": 0.4686, + "step": 1479 + }, + { + "epoch": 0.08268849344917172, + "grad_norm": 3.4349327087402344, + "learning_rate": 0.0009614802779022859, + "loss": 0.3784, + "step": 1480 + }, + { + "epoch": 0.0827443640528536, + "grad_norm": 2.0585591793060303, + "learning_rate": 0.0009614522635589423, + "loss": 0.5736, + "step": 1481 + }, + { + "epoch": 0.08280023465653547, + "grad_norm": 0.7976416349411011, + "learning_rate": 0.0009614242492155985, + "loss": 0.5473, + "step": 1482 + }, + { + "epoch": 0.08285610526021733, + "grad_norm": 1.0097569227218628, + "learning_rate": 0.0009613962348722546, + "loss": 0.51, + "step": 1483 + }, + { + "epoch": 0.08291197586389921, + "grad_norm": 0.7225438356399536, + "learning_rate": 0.0009613682205289108, + "loss": 0.5242, + "step": 1484 + }, + { + "epoch": 0.08296784646758108, + "grad_norm": 0.9528962969779968, + "learning_rate": 0.0009613402061855671, + "loss": 0.5632, + "step": 1485 + }, + { + "epoch": 0.08302371707126295, + "grad_norm": 1.0011579990386963, + "learning_rate": 0.0009613121918422233, + "loss": 0.504, + "step": 1486 + }, + { + "epoch": 0.08307958767494483, + "grad_norm": 0.8716151118278503, + "learning_rate": 0.0009612841774988794, + "loss": 0.4898, + "step": 1487 + }, + { + "epoch": 0.0831354582786267, + "grad_norm": 1.31365168094635, + "learning_rate": 0.0009612561631555357, + "loss": 0.6628, + "step": 1488 + }, + { + "epoch": 0.08319132888230857, + "grad_norm": 0.7998873591423035, + "learning_rate": 0.0009612281488121919, + "loss": 0.5081, + "step": 1489 + }, + { + "epoch": 0.08324719948599045, + "grad_norm": 0.8014880418777466, + "learning_rate": 0.000961200134468848, + "loss": 0.5752, + "step": 1490 + }, + { + "epoch": 0.08330307008967232, + "grad_norm": 5.7616801261901855, + "learning_rate": 0.0009611721201255042, + "loss": 0.5446, + "step": 1491 + }, + { + "epoch": 0.08335894069335419, + "grad_norm": 0.6723014116287231, + "learning_rate": 0.0009611441057821605, + "loss": 0.5208, + "step": 1492 + }, + { + "epoch": 0.08341481129703607, + "grad_norm": 0.5018106698989868, + "learning_rate": 0.0009611160914388167, + "loss": 0.4917, + "step": 1493 + }, + { + "epoch": 0.08347068190071794, + "grad_norm": 0.4959825277328491, + "learning_rate": 0.0009610880770954728, + "loss": 0.4719, + "step": 1494 + }, + { + "epoch": 0.0835265525043998, + "grad_norm": 0.39407995343208313, + "learning_rate": 0.0009610600627521291, + "loss": 0.361, + "step": 1495 + }, + { + "epoch": 0.08358242310808169, + "grad_norm": 0.7668589353561401, + "learning_rate": 0.0009610320484087853, + "loss": 0.4403, + "step": 1496 + }, + { + "epoch": 0.08363829371176355, + "grad_norm": 0.6300973296165466, + "learning_rate": 0.0009610040340654416, + "loss": 0.5308, + "step": 1497 + }, + { + "epoch": 0.08369416431544544, + "grad_norm": 1.226656436920166, + "learning_rate": 0.0009609760197220976, + "loss": 0.4358, + "step": 1498 + }, + { + "epoch": 0.0837500349191273, + "grad_norm": 1.0416011810302734, + "learning_rate": 0.000960948005378754, + "loss": 0.9012, + "step": 1499 + }, + { + "epoch": 0.08380590552280917, + "grad_norm": 0.9477102756500244, + "learning_rate": 0.0009609199910354102, + "loss": 0.5709, + "step": 1500 + }, + { + "epoch": 0.08380590552280917, + "eval_cer": 0.10012765539589538, + "eval_loss": 0.3828731179237366, + "eval_runtime": 55.9473, + "eval_samples_per_second": 81.112, + "eval_steps_per_second": 5.076, + "eval_wer": 0.3934806290327086, + "step": 1500 + }, + { + "epoch": 0.08386177612649105, + "grad_norm": 0.5456610918045044, + "learning_rate": 0.0009608919766920664, + "loss": 0.5078, + "step": 1501 + }, + { + "epoch": 0.08391764673017292, + "grad_norm": 14.573686599731445, + "learning_rate": 0.0009608639623487226, + "loss": 0.5733, + "step": 1502 + }, + { + "epoch": 0.08397351733385479, + "grad_norm": 0.8783748745918274, + "learning_rate": 0.0009608359480053788, + "loss": 0.4625, + "step": 1503 + }, + { + "epoch": 0.08402938793753667, + "grad_norm": 0.921265721321106, + "learning_rate": 0.000960807933662035, + "loss": 0.4727, + "step": 1504 + }, + { + "epoch": 0.08408525854121854, + "grad_norm": 0.789132833480835, + "learning_rate": 0.0009607799193186912, + "loss": 0.5257, + "step": 1505 + }, + { + "epoch": 0.0841411291449004, + "grad_norm": 0.5395810604095459, + "learning_rate": 0.0009607519049753474, + "loss": 0.513, + "step": 1506 + }, + { + "epoch": 0.08419699974858229, + "grad_norm": 1.255974531173706, + "learning_rate": 0.0009607238906320036, + "loss": 0.7278, + "step": 1507 + }, + { + "epoch": 0.08425287035226416, + "grad_norm": 1.5264376401901245, + "learning_rate": 0.0009606958762886598, + "loss": 0.5654, + "step": 1508 + }, + { + "epoch": 0.08430874095594602, + "grad_norm": 0.788278341293335, + "learning_rate": 0.000960667861945316, + "loss": 0.5385, + "step": 1509 + }, + { + "epoch": 0.0843646115596279, + "grad_norm": 0.399993360042572, + "learning_rate": 0.0009606398476019722, + "loss": 0.4272, + "step": 1510 + }, + { + "epoch": 0.08442048216330977, + "grad_norm": 0.7706218957901001, + "learning_rate": 0.0009606118332586284, + "loss": 0.4478, + "step": 1511 + }, + { + "epoch": 0.08447635276699164, + "grad_norm": 1.496484637260437, + "learning_rate": 0.0009605838189152846, + "loss": 0.5297, + "step": 1512 + }, + { + "epoch": 0.08453222337067352, + "grad_norm": 0.5578113794326782, + "learning_rate": 0.0009605558045719408, + "loss": 0.5006, + "step": 1513 + }, + { + "epoch": 0.08458809397435539, + "grad_norm": 1.0529017448425293, + "learning_rate": 0.000960527790228597, + "loss": 0.4886, + "step": 1514 + }, + { + "epoch": 0.08464396457803726, + "grad_norm": 0.5148372054100037, + "learning_rate": 0.0009604997758852533, + "loss": 0.4437, + "step": 1515 + }, + { + "epoch": 0.08469983518171914, + "grad_norm": 0.9611226916313171, + "learning_rate": 0.0009604717615419096, + "loss": 0.5372, + "step": 1516 + }, + { + "epoch": 0.08475570578540101, + "grad_norm": 3.6033782958984375, + "learning_rate": 0.0009604437471985657, + "loss": 0.5062, + "step": 1517 + }, + { + "epoch": 0.08481157638908289, + "grad_norm": 3.4278175830841064, + "learning_rate": 0.0009604157328552219, + "loss": 0.3472, + "step": 1518 + }, + { + "epoch": 0.08486744699276476, + "grad_norm": 0.836211621761322, + "learning_rate": 0.0009603877185118781, + "loss": 0.5106, + "step": 1519 + }, + { + "epoch": 0.08492331759644663, + "grad_norm": 0.602798581123352, + "learning_rate": 0.0009603597041685344, + "loss": 0.4426, + "step": 1520 + }, + { + "epoch": 0.08497918820012851, + "grad_norm": 0.7744857668876648, + "learning_rate": 0.0009603316898251905, + "loss": 0.5259, + "step": 1521 + }, + { + "epoch": 0.08503505880381038, + "grad_norm": 0.49815458059310913, + "learning_rate": 0.0009603036754818467, + "loss": 0.4676, + "step": 1522 + }, + { + "epoch": 0.08509092940749224, + "grad_norm": 0.5999596118927002, + "learning_rate": 0.0009602756611385029, + "loss": 0.5676, + "step": 1523 + }, + { + "epoch": 0.08514680001117413, + "grad_norm": 0.7826496362686157, + "learning_rate": 0.0009602476467951592, + "loss": 0.4395, + "step": 1524 + }, + { + "epoch": 0.085202670614856, + "grad_norm": 0.5523684024810791, + "learning_rate": 0.0009602196324518153, + "loss": 0.436, + "step": 1525 + }, + { + "epoch": 0.08525854121853786, + "grad_norm": 0.7281861305236816, + "learning_rate": 0.0009601916181084715, + "loss": 0.5386, + "step": 1526 + }, + { + "epoch": 0.08531441182221974, + "grad_norm": 3.685054063796997, + "learning_rate": 0.0009601636037651278, + "loss": 0.6117, + "step": 1527 + }, + { + "epoch": 0.08537028242590161, + "grad_norm": 0.9059458374977112, + "learning_rate": 0.000960135589421784, + "loss": 0.5218, + "step": 1528 + }, + { + "epoch": 0.08542615302958348, + "grad_norm": 0.5864368677139282, + "learning_rate": 0.0009601075750784401, + "loss": 0.4421, + "step": 1529 + }, + { + "epoch": 0.08548202363326536, + "grad_norm": 0.7846287488937378, + "learning_rate": 0.0009600795607350963, + "loss": 0.478, + "step": 1530 + }, + { + "epoch": 0.08553789423694723, + "grad_norm": 0.7511133551597595, + "learning_rate": 0.0009600515463917527, + "loss": 0.4788, + "step": 1531 + }, + { + "epoch": 0.0855937648406291, + "grad_norm": 0.5006159543991089, + "learning_rate": 0.0009600235320484088, + "loss": 0.4522, + "step": 1532 + }, + { + "epoch": 0.08564963544431098, + "grad_norm": 0.536539614200592, + "learning_rate": 0.000959995517705065, + "loss": 0.5392, + "step": 1533 + }, + { + "epoch": 0.08570550604799285, + "grad_norm": 1.665384292602539, + "learning_rate": 0.0009599675033617213, + "loss": 0.5305, + "step": 1534 + }, + { + "epoch": 0.08576137665167471, + "grad_norm": 0.8607069849967957, + "learning_rate": 0.0009599394890183775, + "loss": 0.5166, + "step": 1535 + }, + { + "epoch": 0.0858172472553566, + "grad_norm": 0.8729371428489685, + "learning_rate": 0.0009599114746750336, + "loss": 0.4921, + "step": 1536 + }, + { + "epoch": 0.08587311785903846, + "grad_norm": 0.45473599433898926, + "learning_rate": 0.0009598834603316898, + "loss": 0.2811, + "step": 1537 + }, + { + "epoch": 0.08592898846272035, + "grad_norm": 0.49350887537002563, + "learning_rate": 0.0009598554459883461, + "loss": 0.4462, + "step": 1538 + }, + { + "epoch": 0.08598485906640221, + "grad_norm": 1.187887191772461, + "learning_rate": 0.0009598274316450023, + "loss": 0.5929, + "step": 1539 + }, + { + "epoch": 0.08604072967008408, + "grad_norm": 0.709468424320221, + "learning_rate": 0.0009597994173016584, + "loss": 0.4889, + "step": 1540 + }, + { + "epoch": 0.08609660027376596, + "grad_norm": 0.6281852126121521, + "learning_rate": 0.0009597714029583147, + "loss": 0.6743, + "step": 1541 + }, + { + "epoch": 0.08615247087744783, + "grad_norm": 0.5505468249320984, + "learning_rate": 0.0009597433886149709, + "loss": 0.3759, + "step": 1542 + }, + { + "epoch": 0.0862083414811297, + "grad_norm": 1.2373223304748535, + "learning_rate": 0.0009597153742716271, + "loss": 0.4534, + "step": 1543 + }, + { + "epoch": 0.08626421208481158, + "grad_norm": 1.8422657251358032, + "learning_rate": 0.0009596873599282832, + "loss": 0.475, + "step": 1544 + }, + { + "epoch": 0.08632008268849345, + "grad_norm": 0.7255287766456604, + "learning_rate": 0.0009596593455849395, + "loss": 0.4973, + "step": 1545 + }, + { + "epoch": 0.08637595329217532, + "grad_norm": 0.4670320749282837, + "learning_rate": 0.0009596313312415957, + "loss": 0.3868, + "step": 1546 + }, + { + "epoch": 0.0864318238958572, + "grad_norm": 8.504121780395508, + "learning_rate": 0.000959603316898252, + "loss": 0.5877, + "step": 1547 + }, + { + "epoch": 0.08648769449953907, + "grad_norm": 0.44800257682800293, + "learning_rate": 0.0009595753025549082, + "loss": 0.487, + "step": 1548 + }, + { + "epoch": 0.08654356510322093, + "grad_norm": 0.4210573136806488, + "learning_rate": 0.0009595472882115644, + "loss": 0.3845, + "step": 1549 + }, + { + "epoch": 0.08659943570690282, + "grad_norm": 8.782461166381836, + "learning_rate": 0.0009595192738682206, + "loss": 0.423, + "step": 1550 + }, + { + "epoch": 0.08665530631058468, + "grad_norm": 0.46556541323661804, + "learning_rate": 0.0009594912595248767, + "loss": 0.4674, + "step": 1551 + }, + { + "epoch": 0.08671117691426655, + "grad_norm": 1.3812018632888794, + "learning_rate": 0.000959463245181533, + "loss": 0.5154, + "step": 1552 + }, + { + "epoch": 0.08676704751794843, + "grad_norm": 0.5677019357681274, + "learning_rate": 0.0009594352308381892, + "loss": 0.528, + "step": 1553 + }, + { + "epoch": 0.0868229181216303, + "grad_norm": 0.9573871493339539, + "learning_rate": 0.0009594072164948454, + "loss": 0.4676, + "step": 1554 + }, + { + "epoch": 0.08687878872531218, + "grad_norm": 0.9610281586647034, + "learning_rate": 0.0009593792021515015, + "loss": 0.6047, + "step": 1555 + }, + { + "epoch": 0.08693465932899405, + "grad_norm": 1.9297600984573364, + "learning_rate": 0.0009593511878081578, + "loss": 0.5111, + "step": 1556 + }, + { + "epoch": 0.08699052993267592, + "grad_norm": 1.2136298418045044, + "learning_rate": 0.000959323173464814, + "loss": 0.6511, + "step": 1557 + }, + { + "epoch": 0.0870464005363578, + "grad_norm": 0.7997735738754272, + "learning_rate": 0.0009592951591214702, + "loss": 0.7065, + "step": 1558 + }, + { + "epoch": 0.08710227114003967, + "grad_norm": 0.7172945141792297, + "learning_rate": 0.0009592671447781264, + "loss": 0.4836, + "step": 1559 + }, + { + "epoch": 0.08715814174372154, + "grad_norm": 0.4895508289337158, + "learning_rate": 0.0009592391304347826, + "loss": 0.5566, + "step": 1560 + }, + { + "epoch": 0.08721401234740342, + "grad_norm": 1.070980429649353, + "learning_rate": 0.0009592111160914388, + "loss": 0.5016, + "step": 1561 + }, + { + "epoch": 0.08726988295108529, + "grad_norm": 1.1793568134307861, + "learning_rate": 0.000959183101748095, + "loss": 0.5549, + "step": 1562 + }, + { + "epoch": 0.08732575355476715, + "grad_norm": 0.5884143710136414, + "learning_rate": 0.0009591550874047512, + "loss": 0.4412, + "step": 1563 + }, + { + "epoch": 0.08738162415844904, + "grad_norm": 0.6131792068481445, + "learning_rate": 0.0009591270730614074, + "loss": 0.49, + "step": 1564 + }, + { + "epoch": 0.0874374947621309, + "grad_norm": 0.594656229019165, + "learning_rate": 0.0009590990587180637, + "loss": 0.5383, + "step": 1565 + }, + { + "epoch": 0.08749336536581277, + "grad_norm": 0.7630344033241272, + "learning_rate": 0.00095907104437472, + "loss": 0.6036, + "step": 1566 + }, + { + "epoch": 0.08754923596949465, + "grad_norm": 0.47178685665130615, + "learning_rate": 0.0009590430300313761, + "loss": 0.4986, + "step": 1567 + }, + { + "epoch": 0.08760510657317652, + "grad_norm": 0.6409046649932861, + "learning_rate": 0.0009590150156880323, + "loss": 0.3916, + "step": 1568 + }, + { + "epoch": 0.08766097717685839, + "grad_norm": 0.7610702514648438, + "learning_rate": 0.0009589870013446885, + "loss": 0.3821, + "step": 1569 + }, + { + "epoch": 0.08771684778054027, + "grad_norm": 2.7208516597747803, + "learning_rate": 0.0009589589870013448, + "loss": 0.4601, + "step": 1570 + }, + { + "epoch": 0.08777271838422214, + "grad_norm": 1.5150293111801147, + "learning_rate": 0.0009589309726580009, + "loss": 0.7486, + "step": 1571 + }, + { + "epoch": 0.08782858898790401, + "grad_norm": 0.48708483576774597, + "learning_rate": 0.0009589029583146571, + "loss": 0.4459, + "step": 1572 + }, + { + "epoch": 0.08788445959158589, + "grad_norm": 0.7116775512695312, + "learning_rate": 0.0009588749439713134, + "loss": 0.5845, + "step": 1573 + }, + { + "epoch": 0.08794033019526776, + "grad_norm": 0.5046408772468567, + "learning_rate": 0.0009588469296279695, + "loss": 0.5165, + "step": 1574 + }, + { + "epoch": 0.08799620079894964, + "grad_norm": 0.6693805456161499, + "learning_rate": 0.0009588189152846257, + "loss": 0.4857, + "step": 1575 + }, + { + "epoch": 0.08805207140263151, + "grad_norm": 0.6180440783500671, + "learning_rate": 0.0009587909009412819, + "loss": 0.4904, + "step": 1576 + }, + { + "epoch": 0.08810794200631338, + "grad_norm": 0.9309718608856201, + "learning_rate": 0.0009587628865979382, + "loss": 0.5266, + "step": 1577 + }, + { + "epoch": 0.08816381260999526, + "grad_norm": 0.8818928003311157, + "learning_rate": 0.0009587348722545943, + "loss": 0.4884, + "step": 1578 + }, + { + "epoch": 0.08821968321367712, + "grad_norm": 0.4599439799785614, + "learning_rate": 0.0009587068579112505, + "loss": 0.5621, + "step": 1579 + }, + { + "epoch": 0.08827555381735899, + "grad_norm": 0.518778920173645, + "learning_rate": 0.0009586788435679069, + "loss": 0.4984, + "step": 1580 + }, + { + "epoch": 0.08833142442104087, + "grad_norm": 0.7341597676277161, + "learning_rate": 0.0009586508292245631, + "loss": 0.5061, + "step": 1581 + }, + { + "epoch": 0.08838729502472274, + "grad_norm": 0.5114481449127197, + "learning_rate": 0.0009586228148812192, + "loss": 0.3757, + "step": 1582 + }, + { + "epoch": 0.08844316562840461, + "grad_norm": 0.4844626486301422, + "learning_rate": 0.0009585948005378754, + "loss": 0.5481, + "step": 1583 + }, + { + "epoch": 0.08849903623208649, + "grad_norm": 0.5696114897727966, + "learning_rate": 0.0009585667861945317, + "loss": 0.383, + "step": 1584 + }, + { + "epoch": 0.08855490683576836, + "grad_norm": 0.6108005046844482, + "learning_rate": 0.0009585387718511879, + "loss": 0.5844, + "step": 1585 + }, + { + "epoch": 0.08861077743945023, + "grad_norm": 2.1482534408569336, + "learning_rate": 0.000958510757507844, + "loss": 0.5496, + "step": 1586 + }, + { + "epoch": 0.08866664804313211, + "grad_norm": 0.9889042377471924, + "learning_rate": 0.0009584827431645002, + "loss": 0.5135, + "step": 1587 + }, + { + "epoch": 0.08872251864681398, + "grad_norm": 1.5671086311340332, + "learning_rate": 0.0009584547288211565, + "loss": 0.5402, + "step": 1588 + }, + { + "epoch": 0.08877838925049585, + "grad_norm": 0.7464978098869324, + "learning_rate": 0.0009584267144778127, + "loss": 0.5984, + "step": 1589 + }, + { + "epoch": 0.08883425985417773, + "grad_norm": 0.6687489151954651, + "learning_rate": 0.0009583987001344688, + "loss": 0.4572, + "step": 1590 + }, + { + "epoch": 0.0888901304578596, + "grad_norm": 0.9503822922706604, + "learning_rate": 0.0009583706857911251, + "loss": 0.5107, + "step": 1591 + }, + { + "epoch": 0.08894600106154146, + "grad_norm": 1.0355058908462524, + "learning_rate": 0.0009583426714477813, + "loss": 0.5614, + "step": 1592 + }, + { + "epoch": 0.08900187166522334, + "grad_norm": 0.9183691740036011, + "learning_rate": 0.0009583146571044374, + "loss": 0.409, + "step": 1593 + }, + { + "epoch": 0.08905774226890521, + "grad_norm": 0.7922074794769287, + "learning_rate": 0.0009582866427610936, + "loss": 0.5754, + "step": 1594 + }, + { + "epoch": 0.0891136128725871, + "grad_norm": 0.9103812575340271, + "learning_rate": 0.0009582586284177499, + "loss": 0.4721, + "step": 1595 + }, + { + "epoch": 0.08916948347626896, + "grad_norm": 1.1304081678390503, + "learning_rate": 0.0009582306140744061, + "loss": 0.4772, + "step": 1596 + }, + { + "epoch": 0.08922535407995083, + "grad_norm": 0.660851776599884, + "learning_rate": 0.0009582025997310622, + "loss": 0.5961, + "step": 1597 + }, + { + "epoch": 0.08928122468363271, + "grad_norm": 0.5030987858772278, + "learning_rate": 0.0009581745853877186, + "loss": 0.4392, + "step": 1598 + }, + { + "epoch": 0.08933709528731458, + "grad_norm": 1.6843547821044922, + "learning_rate": 0.0009581465710443748, + "loss": 0.6037, + "step": 1599 + }, + { + "epoch": 0.08939296589099645, + "grad_norm": 0.6770682334899902, + "learning_rate": 0.000958118556701031, + "loss": 0.4573, + "step": 1600 + }, + { + "epoch": 0.08944883649467833, + "grad_norm": 0.849312961101532, + "learning_rate": 0.0009580905423576871, + "loss": 0.5474, + "step": 1601 + }, + { + "epoch": 0.0895047070983602, + "grad_norm": 0.9282434582710266, + "learning_rate": 0.0009580625280143434, + "loss": 0.572, + "step": 1602 + }, + { + "epoch": 0.08956057770204207, + "grad_norm": 0.7309882640838623, + "learning_rate": 0.0009580345136709996, + "loss": 0.6267, + "step": 1603 + }, + { + "epoch": 0.08961644830572395, + "grad_norm": 1.2682216167449951, + "learning_rate": 0.0009580064993276558, + "loss": 0.5052, + "step": 1604 + }, + { + "epoch": 0.08967231890940582, + "grad_norm": 0.5698956847190857, + "learning_rate": 0.000957978484984312, + "loss": 0.4906, + "step": 1605 + }, + { + "epoch": 0.08972818951308768, + "grad_norm": 0.8084691166877747, + "learning_rate": 0.0009579504706409682, + "loss": 0.603, + "step": 1606 + }, + { + "epoch": 0.08978406011676957, + "grad_norm": 0.7200801968574524, + "learning_rate": 0.0009579224562976244, + "loss": 0.6294, + "step": 1607 + }, + { + "epoch": 0.08983993072045143, + "grad_norm": 1.8719196319580078, + "learning_rate": 0.0009578944419542806, + "loss": 0.5836, + "step": 1608 + }, + { + "epoch": 0.0898958013241333, + "grad_norm": 0.664040207862854, + "learning_rate": 0.0009578664276109368, + "loss": 0.6947, + "step": 1609 + }, + { + "epoch": 0.08995167192781518, + "grad_norm": 0.5146535038948059, + "learning_rate": 0.000957838413267593, + "loss": 0.3781, + "step": 1610 + }, + { + "epoch": 0.09000754253149705, + "grad_norm": 0.5804701447486877, + "learning_rate": 0.0009578103989242492, + "loss": 0.6765, + "step": 1611 + }, + { + "epoch": 0.09006341313517893, + "grad_norm": 0.564239501953125, + "learning_rate": 0.0009577823845809055, + "loss": 0.5324, + "step": 1612 + }, + { + "epoch": 0.0901192837388608, + "grad_norm": 0.46473726630210876, + "learning_rate": 0.0009577543702375616, + "loss": 0.389, + "step": 1613 + }, + { + "epoch": 0.09017515434254267, + "grad_norm": 0.527326226234436, + "learning_rate": 0.0009577263558942178, + "loss": 0.5752, + "step": 1614 + }, + { + "epoch": 0.09023102494622455, + "grad_norm": 0.617798924446106, + "learning_rate": 0.0009576983415508741, + "loss": 0.5999, + "step": 1615 + }, + { + "epoch": 0.09028689554990642, + "grad_norm": 0.47421929240226746, + "learning_rate": 0.0009576703272075303, + "loss": 0.4201, + "step": 1616 + }, + { + "epoch": 0.09034276615358829, + "grad_norm": 0.44161033630371094, + "learning_rate": 0.0009576423128641865, + "loss": 0.3382, + "step": 1617 + }, + { + "epoch": 0.09039863675727017, + "grad_norm": 0.731928288936615, + "learning_rate": 0.0009576142985208427, + "loss": 0.4662, + "step": 1618 + }, + { + "epoch": 0.09045450736095204, + "grad_norm": 0.8787534236907959, + "learning_rate": 0.000957586284177499, + "loss": 0.4719, + "step": 1619 + }, + { + "epoch": 0.0905103779646339, + "grad_norm": 0.5957030057907104, + "learning_rate": 0.0009575582698341551, + "loss": 0.4342, + "step": 1620 + }, + { + "epoch": 0.09056624856831579, + "grad_norm": 0.49541860818862915, + "learning_rate": 0.0009575302554908113, + "loss": 0.4458, + "step": 1621 + }, + { + "epoch": 0.09062211917199765, + "grad_norm": 0.661164402961731, + "learning_rate": 0.0009575022411474675, + "loss": 0.4928, + "step": 1622 + }, + { + "epoch": 0.09067798977567952, + "grad_norm": 0.7391472458839417, + "learning_rate": 0.0009574742268041238, + "loss": 0.5734, + "step": 1623 + }, + { + "epoch": 0.0907338603793614, + "grad_norm": 0.7559037804603577, + "learning_rate": 0.0009574462124607799, + "loss": 0.6153, + "step": 1624 + }, + { + "epoch": 0.09078973098304327, + "grad_norm": 0.5888568162918091, + "learning_rate": 0.0009574181981174361, + "loss": 0.5682, + "step": 1625 + }, + { + "epoch": 0.09084560158672514, + "grad_norm": 0.7358746528625488, + "learning_rate": 0.0009573901837740923, + "loss": 0.4923, + "step": 1626 + }, + { + "epoch": 0.09090147219040702, + "grad_norm": 0.5687109231948853, + "learning_rate": 0.0009573621694307486, + "loss": 0.5175, + "step": 1627 + }, + { + "epoch": 0.09095734279408889, + "grad_norm": 0.8752040863037109, + "learning_rate": 0.0009573341550874047, + "loss": 0.4889, + "step": 1628 + }, + { + "epoch": 0.09101321339777076, + "grad_norm": 0.613016664981842, + "learning_rate": 0.0009573061407440609, + "loss": 0.4394, + "step": 1629 + }, + { + "epoch": 0.09106908400145264, + "grad_norm": 0.5653653144836426, + "learning_rate": 0.0009572781264007173, + "loss": 0.4927, + "step": 1630 + }, + { + "epoch": 0.0911249546051345, + "grad_norm": 0.7072796821594238, + "learning_rate": 0.0009572501120573735, + "loss": 0.4777, + "step": 1631 + }, + { + "epoch": 0.09118082520881639, + "grad_norm": 6.421507358551025, + "learning_rate": 0.0009572220977140296, + "loss": 0.4716, + "step": 1632 + }, + { + "epoch": 0.09123669581249826, + "grad_norm": 0.4985891580581665, + "learning_rate": 0.0009571940833706858, + "loss": 0.4996, + "step": 1633 + }, + { + "epoch": 0.09129256641618012, + "grad_norm": 1.0108110904693604, + "learning_rate": 0.0009571660690273421, + "loss": 0.503, + "step": 1634 + }, + { + "epoch": 0.091348437019862, + "grad_norm": 2.8976376056671143, + "learning_rate": 0.0009571380546839982, + "loss": 0.5616, + "step": 1635 + }, + { + "epoch": 0.09140430762354387, + "grad_norm": 0.9594224095344543, + "learning_rate": 0.0009571100403406544, + "loss": 0.5685, + "step": 1636 + }, + { + "epoch": 0.09146017822722574, + "grad_norm": 0.7330107688903809, + "learning_rate": 0.0009570820259973107, + "loss": 0.4992, + "step": 1637 + }, + { + "epoch": 0.09151604883090762, + "grad_norm": 0.4827437698841095, + "learning_rate": 0.0009570540116539669, + "loss": 0.385, + "step": 1638 + }, + { + "epoch": 0.09157191943458949, + "grad_norm": 0.497945636510849, + "learning_rate": 0.000957025997310623, + "loss": 0.431, + "step": 1639 + }, + { + "epoch": 0.09162779003827136, + "grad_norm": 0.5826453566551208, + "learning_rate": 0.0009569979829672792, + "loss": 0.5091, + "step": 1640 + }, + { + "epoch": 0.09168366064195324, + "grad_norm": 0.5731382966041565, + "learning_rate": 0.0009569699686239355, + "loss": 0.3804, + "step": 1641 + }, + { + "epoch": 0.09173953124563511, + "grad_norm": 0.5936702489852905, + "learning_rate": 0.0009569419542805917, + "loss": 0.4742, + "step": 1642 + }, + { + "epoch": 0.09179540184931698, + "grad_norm": 0.6041135787963867, + "learning_rate": 0.0009569139399372478, + "loss": 0.4392, + "step": 1643 + }, + { + "epoch": 0.09185127245299886, + "grad_norm": 2.3999183177948, + "learning_rate": 0.0009568859255939041, + "loss": 0.6005, + "step": 1644 + }, + { + "epoch": 0.09190714305668073, + "grad_norm": 9.834634780883789, + "learning_rate": 0.0009568579112505603, + "loss": 0.6007, + "step": 1645 + }, + { + "epoch": 0.0919630136603626, + "grad_norm": 0.4959006905555725, + "learning_rate": 0.0009568298969072165, + "loss": 0.5875, + "step": 1646 + }, + { + "epoch": 0.09201888426404448, + "grad_norm": 0.6738176941871643, + "learning_rate": 0.0009568018825638726, + "loss": 0.5354, + "step": 1647 + }, + { + "epoch": 0.09207475486772634, + "grad_norm": 0.6503636837005615, + "learning_rate": 0.000956773868220529, + "loss": 0.5024, + "step": 1648 + }, + { + "epoch": 0.09213062547140821, + "grad_norm": 0.6488896012306213, + "learning_rate": 0.0009567458538771852, + "loss": 0.4594, + "step": 1649 + }, + { + "epoch": 0.0921864960750901, + "grad_norm": 1.323390245437622, + "learning_rate": 0.0009567178395338414, + "loss": 0.5084, + "step": 1650 + }, + { + "epoch": 0.09224236667877196, + "grad_norm": 0.7127136588096619, + "learning_rate": 0.0009566898251904976, + "loss": 0.493, + "step": 1651 + }, + { + "epoch": 0.09229823728245384, + "grad_norm": 0.9544020295143127, + "learning_rate": 0.0009566618108471538, + "loss": 0.4607, + "step": 1652 + }, + { + "epoch": 0.09235410788613571, + "grad_norm": 0.47326719760894775, + "learning_rate": 0.00095663379650381, + "loss": 0.535, + "step": 1653 + }, + { + "epoch": 0.09240997848981758, + "grad_norm": 1.4693788290023804, + "learning_rate": 0.0009566057821604662, + "loss": 0.6791, + "step": 1654 + }, + { + "epoch": 0.09246584909349946, + "grad_norm": 0.7481864094734192, + "learning_rate": 0.0009565777678171224, + "loss": 0.5707, + "step": 1655 + }, + { + "epoch": 0.09252171969718133, + "grad_norm": 0.5686119794845581, + "learning_rate": 0.0009565497534737786, + "loss": 0.5718, + "step": 1656 + }, + { + "epoch": 0.0925775903008632, + "grad_norm": 0.863732635974884, + "learning_rate": 0.0009565217391304348, + "loss": 0.4943, + "step": 1657 + }, + { + "epoch": 0.09263346090454508, + "grad_norm": 0.5449768900871277, + "learning_rate": 0.0009564937247870909, + "loss": 0.3869, + "step": 1658 + }, + { + "epoch": 0.09268933150822695, + "grad_norm": 0.5411713123321533, + "learning_rate": 0.0009564657104437472, + "loss": 0.5087, + "step": 1659 + }, + { + "epoch": 0.09274520211190881, + "grad_norm": 0.6649173498153687, + "learning_rate": 0.0009564376961004034, + "loss": 0.4983, + "step": 1660 + }, + { + "epoch": 0.0928010727155907, + "grad_norm": 0.7587314248085022, + "learning_rate": 0.0009564096817570596, + "loss": 0.4855, + "step": 1661 + }, + { + "epoch": 0.09285694331927256, + "grad_norm": 22.28857421875, + "learning_rate": 0.0009563816674137158, + "loss": 0.4587, + "step": 1662 + }, + { + "epoch": 0.09291281392295443, + "grad_norm": 0.9370453953742981, + "learning_rate": 0.000956353653070372, + "loss": 0.533, + "step": 1663 + }, + { + "epoch": 0.09296868452663631, + "grad_norm": 1.1891322135925293, + "learning_rate": 0.0009563256387270282, + "loss": 0.4886, + "step": 1664 + }, + { + "epoch": 0.09302455513031818, + "grad_norm": 1.9058458805084229, + "learning_rate": 0.0009562976243836845, + "loss": 0.4509, + "step": 1665 + }, + { + "epoch": 0.09308042573400005, + "grad_norm": 0.5392264723777771, + "learning_rate": 0.0009562696100403407, + "loss": 0.4723, + "step": 1666 + }, + { + "epoch": 0.09313629633768193, + "grad_norm": 0.9007321000099182, + "learning_rate": 0.0009562415956969969, + "loss": 0.5223, + "step": 1667 + }, + { + "epoch": 0.0931921669413638, + "grad_norm": 0.8999434113502502, + "learning_rate": 0.0009562135813536531, + "loss": 0.5931, + "step": 1668 + }, + { + "epoch": 0.09324803754504568, + "grad_norm": 0.4713168442249298, + "learning_rate": 0.0009561855670103094, + "loss": 0.4746, + "step": 1669 + }, + { + "epoch": 0.09330390814872755, + "grad_norm": 0.6415393352508545, + "learning_rate": 0.0009561575526669655, + "loss": 0.6221, + "step": 1670 + }, + { + "epoch": 0.09335977875240942, + "grad_norm": 0.872124195098877, + "learning_rate": 0.0009561295383236217, + "loss": 0.6854, + "step": 1671 + }, + { + "epoch": 0.0934156493560913, + "grad_norm": 0.6027430295944214, + "learning_rate": 0.0009561015239802779, + "loss": 0.396, + "step": 1672 + }, + { + "epoch": 0.09347151995977317, + "grad_norm": 0.7313055992126465, + "learning_rate": 0.0009560735096369342, + "loss": 0.6632, + "step": 1673 + }, + { + "epoch": 0.09352739056345503, + "grad_norm": 0.5191171765327454, + "learning_rate": 0.0009560454952935903, + "loss": 0.4345, + "step": 1674 + }, + { + "epoch": 0.09358326116713692, + "grad_norm": 2.877314805984497, + "learning_rate": 0.0009560174809502465, + "loss": 0.5019, + "step": 1675 + }, + { + "epoch": 0.09363913177081878, + "grad_norm": 2.785935640335083, + "learning_rate": 0.0009559894666069028, + "loss": 0.5515, + "step": 1676 + }, + { + "epoch": 0.09369500237450065, + "grad_norm": 0.7620179057121277, + "learning_rate": 0.000955961452263559, + "loss": 0.4915, + "step": 1677 + }, + { + "epoch": 0.09375087297818253, + "grad_norm": 0.7589294910430908, + "learning_rate": 0.0009559334379202151, + "loss": 0.5524, + "step": 1678 + }, + { + "epoch": 0.0938067435818644, + "grad_norm": 0.8370959162712097, + "learning_rate": 0.0009559054235768713, + "loss": 0.471, + "step": 1679 + }, + { + "epoch": 0.09386261418554627, + "grad_norm": 0.9741494655609131, + "learning_rate": 0.0009558774092335276, + "loss": 0.4739, + "step": 1680 + }, + { + "epoch": 0.09391848478922815, + "grad_norm": 0.6208754181861877, + "learning_rate": 0.0009558493948901837, + "loss": 0.5071, + "step": 1681 + }, + { + "epoch": 0.09397435539291002, + "grad_norm": 0.5007163286209106, + "learning_rate": 0.00095582138054684, + "loss": 0.4559, + "step": 1682 + }, + { + "epoch": 0.09403022599659189, + "grad_norm": 0.7174465656280518, + "learning_rate": 0.0009557933662034963, + "loss": 0.4581, + "step": 1683 + }, + { + "epoch": 0.09408609660027377, + "grad_norm": 1.0325119495391846, + "learning_rate": 0.0009557653518601525, + "loss": 0.4797, + "step": 1684 + }, + { + "epoch": 0.09414196720395564, + "grad_norm": 0.5752109885215759, + "learning_rate": 0.0009557373375168086, + "loss": 0.4369, + "step": 1685 + }, + { + "epoch": 0.0941978378076375, + "grad_norm": 0.8846983313560486, + "learning_rate": 0.0009557093231734648, + "loss": 0.5268, + "step": 1686 + }, + { + "epoch": 0.09425370841131939, + "grad_norm": 1.498984694480896, + "learning_rate": 0.0009556813088301211, + "loss": 0.6166, + "step": 1687 + }, + { + "epoch": 0.09430957901500125, + "grad_norm": 0.9015820622444153, + "learning_rate": 0.0009556532944867773, + "loss": 0.5344, + "step": 1688 + }, + { + "epoch": 0.09436544961868314, + "grad_norm": 0.5844884514808655, + "learning_rate": 0.0009556252801434334, + "loss": 0.465, + "step": 1689 + }, + { + "epoch": 0.094421320222365, + "grad_norm": 0.5039775371551514, + "learning_rate": 0.0009555972658000897, + "loss": 0.4597, + "step": 1690 + }, + { + "epoch": 0.09447719082604687, + "grad_norm": 5.247317790985107, + "learning_rate": 0.0009555692514567459, + "loss": 0.5317, + "step": 1691 + }, + { + "epoch": 0.09453306142972875, + "grad_norm": 12.106361389160156, + "learning_rate": 0.0009555412371134021, + "loss": 0.5116, + "step": 1692 + }, + { + "epoch": 0.09458893203341062, + "grad_norm": 0.6411914825439453, + "learning_rate": 0.0009555132227700582, + "loss": 0.5276, + "step": 1693 + }, + { + "epoch": 0.09464480263709249, + "grad_norm": 0.5287384390830994, + "learning_rate": 0.0009554852084267145, + "loss": 0.4057, + "step": 1694 + }, + { + "epoch": 0.09470067324077437, + "grad_norm": 0.7992839217185974, + "learning_rate": 0.0009554571940833707, + "loss": 0.3579, + "step": 1695 + }, + { + "epoch": 0.09475654384445624, + "grad_norm": 0.6883152723312378, + "learning_rate": 0.0009554291797400269, + "loss": 0.483, + "step": 1696 + }, + { + "epoch": 0.09481241444813811, + "grad_norm": 0.8729891777038574, + "learning_rate": 0.000955401165396683, + "loss": 0.6225, + "step": 1697 + }, + { + "epoch": 0.09486828505181999, + "grad_norm": 0.6426833271980286, + "learning_rate": 0.0009553731510533394, + "loss": 0.4719, + "step": 1698 + }, + { + "epoch": 0.09492415565550186, + "grad_norm": 0.6102331280708313, + "learning_rate": 0.0009553451367099956, + "loss": 0.6723, + "step": 1699 + }, + { + "epoch": 0.09498002625918373, + "grad_norm": 0.6106989979743958, + "learning_rate": 0.0009553171223666517, + "loss": 0.5275, + "step": 1700 + }, + { + "epoch": 0.09503589686286561, + "grad_norm": 0.6960495114326477, + "learning_rate": 0.000955289108023308, + "loss": 0.5927, + "step": 1701 + }, + { + "epoch": 0.09509176746654747, + "grad_norm": 0.8076702356338501, + "learning_rate": 0.0009552610936799642, + "loss": 0.4353, + "step": 1702 + }, + { + "epoch": 0.09514763807022934, + "grad_norm": 0.523046612739563, + "learning_rate": 0.0009552330793366204, + "loss": 0.4458, + "step": 1703 + }, + { + "epoch": 0.09520350867391122, + "grad_norm": 0.6411027908325195, + "learning_rate": 0.0009552050649932765, + "loss": 0.427, + "step": 1704 + }, + { + "epoch": 0.09525937927759309, + "grad_norm": 1.0732743740081787, + "learning_rate": 0.0009551770506499328, + "loss": 0.5054, + "step": 1705 + }, + { + "epoch": 0.09531524988127496, + "grad_norm": 0.5613355040550232, + "learning_rate": 0.000955149036306589, + "loss": 0.462, + "step": 1706 + }, + { + "epoch": 0.09537112048495684, + "grad_norm": 0.3764711618423462, + "learning_rate": 0.0009551210219632452, + "loss": 0.4363, + "step": 1707 + }, + { + "epoch": 0.09542699108863871, + "grad_norm": 0.48602810502052307, + "learning_rate": 0.0009550930076199014, + "loss": 0.4499, + "step": 1708 + }, + { + "epoch": 0.09548286169232059, + "grad_norm": 0.5819774866104126, + "learning_rate": 0.0009550649932765576, + "loss": 0.4821, + "step": 1709 + }, + { + "epoch": 0.09553873229600246, + "grad_norm": 0.6399314403533936, + "learning_rate": 0.0009550369789332138, + "loss": 0.4375, + "step": 1710 + }, + { + "epoch": 0.09559460289968433, + "grad_norm": 0.9159020781517029, + "learning_rate": 0.00095500896458987, + "loss": 0.4683, + "step": 1711 + }, + { + "epoch": 0.09565047350336621, + "grad_norm": 0.882836103439331, + "learning_rate": 0.0009549809502465262, + "loss": 0.4123, + "step": 1712 + }, + { + "epoch": 0.09570634410704808, + "grad_norm": 0.8593130707740784, + "learning_rate": 0.0009549529359031824, + "loss": 0.4172, + "step": 1713 + }, + { + "epoch": 0.09576221471072995, + "grad_norm": 0.880905270576477, + "learning_rate": 0.0009549249215598386, + "loss": 0.5032, + "step": 1714 + }, + { + "epoch": 0.09581808531441183, + "grad_norm": 0.5326239466667175, + "learning_rate": 0.000954896907216495, + "loss": 0.443, + "step": 1715 + }, + { + "epoch": 0.0958739559180937, + "grad_norm": 0.46739739179611206, + "learning_rate": 0.0009548688928731511, + "loss": 0.3919, + "step": 1716 + }, + { + "epoch": 0.09592982652177556, + "grad_norm": 0.4786103665828705, + "learning_rate": 0.0009548408785298073, + "loss": 0.4261, + "step": 1717 + }, + { + "epoch": 0.09598569712545744, + "grad_norm": 0.7867366671562195, + "learning_rate": 0.0009548128641864635, + "loss": 0.501, + "step": 1718 + }, + { + "epoch": 0.09604156772913931, + "grad_norm": 0.6677238345146179, + "learning_rate": 0.0009547848498431198, + "loss": 0.5202, + "step": 1719 + }, + { + "epoch": 0.09609743833282118, + "grad_norm": 0.677115797996521, + "learning_rate": 0.0009547568354997759, + "loss": 0.6568, + "step": 1720 + }, + { + "epoch": 0.09615330893650306, + "grad_norm": 0.9016485810279846, + "learning_rate": 0.0009547288211564321, + "loss": 0.4605, + "step": 1721 + }, + { + "epoch": 0.09620917954018493, + "grad_norm": 3.2021000385284424, + "learning_rate": 0.0009547008068130884, + "loss": 0.4558, + "step": 1722 + }, + { + "epoch": 0.0962650501438668, + "grad_norm": 0.47148841619491577, + "learning_rate": 0.0009546727924697445, + "loss": 0.5108, + "step": 1723 + }, + { + "epoch": 0.09632092074754868, + "grad_norm": 0.5497175455093384, + "learning_rate": 0.0009546447781264007, + "loss": 0.5334, + "step": 1724 + }, + { + "epoch": 0.09637679135123055, + "grad_norm": 1.5560526847839355, + "learning_rate": 0.0009546167637830569, + "loss": 0.5988, + "step": 1725 + }, + { + "epoch": 0.09643266195491243, + "grad_norm": 0.91501384973526, + "learning_rate": 0.0009545887494397132, + "loss": 0.5508, + "step": 1726 + }, + { + "epoch": 0.0964885325585943, + "grad_norm": 0.7299401164054871, + "learning_rate": 0.0009545607350963693, + "loss": 0.4361, + "step": 1727 + }, + { + "epoch": 0.09654440316227617, + "grad_norm": 0.7298431396484375, + "learning_rate": 0.0009545327207530255, + "loss": 0.5672, + "step": 1728 + }, + { + "epoch": 0.09660027376595805, + "grad_norm": 1.0498812198638916, + "learning_rate": 0.0009545047064096818, + "loss": 0.4642, + "step": 1729 + }, + { + "epoch": 0.09665614436963992, + "grad_norm": 0.8376237154006958, + "learning_rate": 0.000954476692066338, + "loss": 0.4105, + "step": 1730 + }, + { + "epoch": 0.09671201497332178, + "grad_norm": 0.5778597593307495, + "learning_rate": 0.0009544486777229941, + "loss": 0.432, + "step": 1731 + }, + { + "epoch": 0.09676788557700366, + "grad_norm": 0.5827915668487549, + "learning_rate": 0.0009544206633796504, + "loss": 0.5479, + "step": 1732 + }, + { + "epoch": 0.09682375618068553, + "grad_norm": 0.47488096356391907, + "learning_rate": 0.0009543926490363067, + "loss": 0.4687, + "step": 1733 + }, + { + "epoch": 0.0968796267843674, + "grad_norm": 1.3137414455413818, + "learning_rate": 0.0009543646346929629, + "loss": 0.473, + "step": 1734 + }, + { + "epoch": 0.09693549738804928, + "grad_norm": 0.7336339354515076, + "learning_rate": 0.000954336620349619, + "loss": 0.6595, + "step": 1735 + }, + { + "epoch": 0.09699136799173115, + "grad_norm": 0.8092811107635498, + "learning_rate": 0.0009543086060062752, + "loss": 0.5302, + "step": 1736 + }, + { + "epoch": 0.09704723859541302, + "grad_norm": 1.7729344367980957, + "learning_rate": 0.0009542805916629315, + "loss": 0.4878, + "step": 1737 + }, + { + "epoch": 0.0971031091990949, + "grad_norm": 1.9526968002319336, + "learning_rate": 0.0009542525773195877, + "loss": 0.5084, + "step": 1738 + }, + { + "epoch": 0.09715897980277677, + "grad_norm": 0.5261747241020203, + "learning_rate": 0.0009542245629762438, + "loss": 0.4123, + "step": 1739 + }, + { + "epoch": 0.09721485040645864, + "grad_norm": 0.8263682126998901, + "learning_rate": 0.0009541965486329001, + "loss": 0.5942, + "step": 1740 + }, + { + "epoch": 0.09727072101014052, + "grad_norm": 0.8589661717414856, + "learning_rate": 0.0009541685342895563, + "loss": 0.5868, + "step": 1741 + }, + { + "epoch": 0.09732659161382239, + "grad_norm": 0.5872436761856079, + "learning_rate": 0.0009541405199462124, + "loss": 0.453, + "step": 1742 + }, + { + "epoch": 0.09738246221750425, + "grad_norm": 0.6542233824729919, + "learning_rate": 0.0009541125056028686, + "loss": 0.556, + "step": 1743 + }, + { + "epoch": 0.09743833282118614, + "grad_norm": 0.5005415678024292, + "learning_rate": 0.0009540844912595249, + "loss": 0.5441, + "step": 1744 + }, + { + "epoch": 0.097494203424868, + "grad_norm": 0.7743383049964905, + "learning_rate": 0.0009540564769161811, + "loss": 0.4656, + "step": 1745 + }, + { + "epoch": 0.09755007402854989, + "grad_norm": 2.573099136352539, + "learning_rate": 0.0009540284625728372, + "loss": 0.4902, + "step": 1746 + }, + { + "epoch": 0.09760594463223175, + "grad_norm": 1.3033016920089722, + "learning_rate": 0.0009540004482294935, + "loss": 0.5367, + "step": 1747 + }, + { + "epoch": 0.09766181523591362, + "grad_norm": 0.5816377997398376, + "learning_rate": 0.0009539724338861498, + "loss": 0.5714, + "step": 1748 + }, + { + "epoch": 0.0977176858395955, + "grad_norm": 1.7851117849349976, + "learning_rate": 0.000953944419542806, + "loss": 0.5456, + "step": 1749 + }, + { + "epoch": 0.09777355644327737, + "grad_norm": 0.5551310181617737, + "learning_rate": 0.0009539164051994621, + "loss": 0.5128, + "step": 1750 + }, + { + "epoch": 0.09782942704695924, + "grad_norm": 0.5493649840354919, + "learning_rate": 0.0009538883908561184, + "loss": 0.7162, + "step": 1751 + }, + { + "epoch": 0.09788529765064112, + "grad_norm": 0.6182380318641663, + "learning_rate": 0.0009538603765127746, + "loss": 0.763, + "step": 1752 + }, + { + "epoch": 0.09794116825432299, + "grad_norm": 0.9493619203567505, + "learning_rate": 0.0009538323621694308, + "loss": 0.5618, + "step": 1753 + }, + { + "epoch": 0.09799703885800486, + "grad_norm": 1.2030988931655884, + "learning_rate": 0.000953804347826087, + "loss": 0.4549, + "step": 1754 + }, + { + "epoch": 0.09805290946168674, + "grad_norm": 2.4514758586883545, + "learning_rate": 0.0009537763334827432, + "loss": 0.4986, + "step": 1755 + }, + { + "epoch": 0.0981087800653686, + "grad_norm": 0.6128562092781067, + "learning_rate": 0.0009537483191393994, + "loss": 0.4902, + "step": 1756 + }, + { + "epoch": 0.09816465066905047, + "grad_norm": 0.9624001979827881, + "learning_rate": 0.0009537203047960556, + "loss": 0.6098, + "step": 1757 + }, + { + "epoch": 0.09822052127273236, + "grad_norm": 1.080871343612671, + "learning_rate": 0.0009536922904527118, + "loss": 0.4653, + "step": 1758 + }, + { + "epoch": 0.09827639187641422, + "grad_norm": 0.5338836312294006, + "learning_rate": 0.000953664276109368, + "loss": 0.4969, + "step": 1759 + }, + { + "epoch": 0.09833226248009609, + "grad_norm": 0.49495428800582886, + "learning_rate": 0.0009536362617660242, + "loss": 0.4597, + "step": 1760 + }, + { + "epoch": 0.09838813308377797, + "grad_norm": 0.6913532018661499, + "learning_rate": 0.0009536082474226805, + "loss": 0.526, + "step": 1761 + }, + { + "epoch": 0.09844400368745984, + "grad_norm": 0.7033137679100037, + "learning_rate": 0.0009535802330793366, + "loss": 0.5631, + "step": 1762 + }, + { + "epoch": 0.09849987429114171, + "grad_norm": 1.6290185451507568, + "learning_rate": 0.0009535522187359928, + "loss": 0.4467, + "step": 1763 + }, + { + "epoch": 0.09855574489482359, + "grad_norm": 0.6314454078674316, + "learning_rate": 0.000953524204392649, + "loss": 0.4686, + "step": 1764 + }, + { + "epoch": 0.09861161549850546, + "grad_norm": 0.5511125326156616, + "learning_rate": 0.0009534961900493053, + "loss": 0.3712, + "step": 1765 + }, + { + "epoch": 0.09866748610218734, + "grad_norm": 0.8877625465393066, + "learning_rate": 0.0009534681757059615, + "loss": 0.495, + "step": 1766 + }, + { + "epoch": 0.09872335670586921, + "grad_norm": 0.8318673968315125, + "learning_rate": 0.0009534401613626177, + "loss": 0.576, + "step": 1767 + }, + { + "epoch": 0.09877922730955108, + "grad_norm": 0.555133581161499, + "learning_rate": 0.0009534121470192739, + "loss": 0.5306, + "step": 1768 + }, + { + "epoch": 0.09883509791323296, + "grad_norm": 1.4647611379623413, + "learning_rate": 0.0009533841326759301, + "loss": 0.5836, + "step": 1769 + }, + { + "epoch": 0.09889096851691483, + "grad_norm": 0.5462441444396973, + "learning_rate": 0.0009533561183325863, + "loss": 0.4526, + "step": 1770 + }, + { + "epoch": 0.0989468391205967, + "grad_norm": 0.5371885895729065, + "learning_rate": 0.0009533281039892425, + "loss": 0.4067, + "step": 1771 + }, + { + "epoch": 0.09900270972427858, + "grad_norm": 3.734924077987671, + "learning_rate": 0.0009533000896458988, + "loss": 0.4393, + "step": 1772 + }, + { + "epoch": 0.09905858032796044, + "grad_norm": 0.6352950930595398, + "learning_rate": 0.0009532720753025549, + "loss": 0.5047, + "step": 1773 + }, + { + "epoch": 0.09911445093164231, + "grad_norm": 0.6068252325057983, + "learning_rate": 0.0009532440609592111, + "loss": 0.464, + "step": 1774 + }, + { + "epoch": 0.0991703215353242, + "grad_norm": 0.4545010030269623, + "learning_rate": 0.0009532160466158673, + "loss": 0.4729, + "step": 1775 + }, + { + "epoch": 0.09922619213900606, + "grad_norm": 0.5594296455383301, + "learning_rate": 0.0009531880322725236, + "loss": 0.5453, + "step": 1776 + }, + { + "epoch": 0.09928206274268793, + "grad_norm": 0.4882468283176422, + "learning_rate": 0.0009531600179291797, + "loss": 0.4671, + "step": 1777 + }, + { + "epoch": 0.09933793334636981, + "grad_norm": 0.48835572600364685, + "learning_rate": 0.0009531320035858359, + "loss": 0.5151, + "step": 1778 + }, + { + "epoch": 0.09939380395005168, + "grad_norm": 0.6018467545509338, + "learning_rate": 0.0009531039892424922, + "loss": 0.4469, + "step": 1779 + }, + { + "epoch": 0.09944967455373355, + "grad_norm": 0.6394640207290649, + "learning_rate": 0.0009530759748991484, + "loss": 0.5517, + "step": 1780 + }, + { + "epoch": 0.09950554515741543, + "grad_norm": 0.4465482234954834, + "learning_rate": 0.0009530479605558045, + "loss": 0.5179, + "step": 1781 + }, + { + "epoch": 0.0995614157610973, + "grad_norm": 0.7507374882698059, + "learning_rate": 0.0009530199462124608, + "loss": 0.4648, + "step": 1782 + }, + { + "epoch": 0.09961728636477916, + "grad_norm": 5.663472652435303, + "learning_rate": 0.0009529919318691171, + "loss": 0.5633, + "step": 1783 + }, + { + "epoch": 0.09967315696846105, + "grad_norm": 0.8286885619163513, + "learning_rate": 0.0009529639175257732, + "loss": 0.5183, + "step": 1784 + }, + { + "epoch": 0.09972902757214291, + "grad_norm": 1.5084742307662964, + "learning_rate": 0.0009529359031824294, + "loss": 0.4597, + "step": 1785 + }, + { + "epoch": 0.0997848981758248, + "grad_norm": 1.5372204780578613, + "learning_rate": 0.0009529078888390857, + "loss": 0.4306, + "step": 1786 + }, + { + "epoch": 0.09984076877950666, + "grad_norm": 0.675079882144928, + "learning_rate": 0.0009528798744957419, + "loss": 0.4691, + "step": 1787 + }, + { + "epoch": 0.09989663938318853, + "grad_norm": 0.6215742230415344, + "learning_rate": 0.000952851860152398, + "loss": 0.4944, + "step": 1788 + }, + { + "epoch": 0.09995250998687041, + "grad_norm": 0.657843291759491, + "learning_rate": 0.0009528238458090542, + "loss": 0.4564, + "step": 1789 + }, + { + "epoch": 0.10000838059055228, + "grad_norm": 1.118149757385254, + "learning_rate": 0.0009527958314657105, + "loss": 0.5164, + "step": 1790 + }, + { + "epoch": 0.10006425119423415, + "grad_norm": 0.8013920783996582, + "learning_rate": 0.0009527678171223667, + "loss": 0.4856, + "step": 1791 + }, + { + "epoch": 0.10012012179791603, + "grad_norm": 3.6593735218048096, + "learning_rate": 0.0009527398027790228, + "loss": 0.5525, + "step": 1792 + }, + { + "epoch": 0.1001759924015979, + "grad_norm": 0.5801917910575867, + "learning_rate": 0.0009527117884356791, + "loss": 0.3724, + "step": 1793 + }, + { + "epoch": 0.10023186300527977, + "grad_norm": 1.0455877780914307, + "learning_rate": 0.0009526837740923353, + "loss": 0.4602, + "step": 1794 + }, + { + "epoch": 0.10028773360896165, + "grad_norm": 1.1388328075408936, + "learning_rate": 0.0009526557597489915, + "loss": 0.5908, + "step": 1795 + }, + { + "epoch": 0.10034360421264352, + "grad_norm": 0.5253142714500427, + "learning_rate": 0.0009526277454056476, + "loss": 0.5279, + "step": 1796 + }, + { + "epoch": 0.10039947481632538, + "grad_norm": 1.1248506307601929, + "learning_rate": 0.000952599731062304, + "loss": 0.5032, + "step": 1797 + }, + { + "epoch": 0.10045534542000727, + "grad_norm": 1.1133630275726318, + "learning_rate": 0.0009525717167189602, + "loss": 0.4241, + "step": 1798 + }, + { + "epoch": 0.10051121602368913, + "grad_norm": 1.1338386535644531, + "learning_rate": 0.0009525437023756164, + "loss": 0.4893, + "step": 1799 + }, + { + "epoch": 0.100567086627371, + "grad_norm": 0.4968947172164917, + "learning_rate": 0.0009525156880322725, + "loss": 0.483, + "step": 1800 + }, + { + "epoch": 0.10062295723105288, + "grad_norm": 1.1110620498657227, + "learning_rate": 0.0009524876736889288, + "loss": 0.5506, + "step": 1801 + }, + { + "epoch": 0.10067882783473475, + "grad_norm": 0.6522441506385803, + "learning_rate": 0.000952459659345585, + "loss": 0.4975, + "step": 1802 + }, + { + "epoch": 0.10073469843841663, + "grad_norm": 1.0056300163269043, + "learning_rate": 0.0009524316450022412, + "loss": 0.4893, + "step": 1803 + }, + { + "epoch": 0.1007905690420985, + "grad_norm": 2.317229986190796, + "learning_rate": 0.0009524036306588974, + "loss": 0.6269, + "step": 1804 + }, + { + "epoch": 0.10084643964578037, + "grad_norm": 1.0169291496276855, + "learning_rate": 0.0009523756163155536, + "loss": 0.5557, + "step": 1805 + }, + { + "epoch": 0.10090231024946225, + "grad_norm": 0.5989283919334412, + "learning_rate": 0.0009523476019722098, + "loss": 0.5587, + "step": 1806 + }, + { + "epoch": 0.10095818085314412, + "grad_norm": 2.7261087894439697, + "learning_rate": 0.0009523195876288659, + "loss": 0.5003, + "step": 1807 + }, + { + "epoch": 0.10101405145682599, + "grad_norm": 0.5368561148643494, + "learning_rate": 0.0009522915732855222, + "loss": 0.4365, + "step": 1808 + }, + { + "epoch": 0.10106992206050787, + "grad_norm": 0.9640926718711853, + "learning_rate": 0.0009522635589421784, + "loss": 0.4987, + "step": 1809 + }, + { + "epoch": 0.10112579266418974, + "grad_norm": 0.9062827825546265, + "learning_rate": 0.0009522355445988346, + "loss": 0.4747, + "step": 1810 + }, + { + "epoch": 0.1011816632678716, + "grad_norm": 0.7110238671302795, + "learning_rate": 0.0009522075302554908, + "loss": 0.5593, + "step": 1811 + }, + { + "epoch": 0.10123753387155349, + "grad_norm": 1.073326587677002, + "learning_rate": 0.000952179515912147, + "loss": 0.5089, + "step": 1812 + }, + { + "epoch": 0.10129340447523535, + "grad_norm": 0.6984613537788391, + "learning_rate": 0.0009521515015688032, + "loss": 0.5007, + "step": 1813 + }, + { + "epoch": 0.10134927507891722, + "grad_norm": 0.5201655626296997, + "learning_rate": 0.0009521234872254594, + "loss": 0.428, + "step": 1814 + }, + { + "epoch": 0.1014051456825991, + "grad_norm": 1.0775914192199707, + "learning_rate": 0.0009520954728821157, + "loss": 0.4994, + "step": 1815 + }, + { + "epoch": 0.10146101628628097, + "grad_norm": 0.5598008632659912, + "learning_rate": 0.0009520674585387719, + "loss": 0.548, + "step": 1816 + }, + { + "epoch": 0.10151688688996284, + "grad_norm": 1.0489352941513062, + "learning_rate": 0.0009520394441954281, + "loss": 0.5783, + "step": 1817 + }, + { + "epoch": 0.10157275749364472, + "grad_norm": 0.5336512923240662, + "learning_rate": 0.0009520114298520844, + "loss": 0.4137, + "step": 1818 + }, + { + "epoch": 0.10162862809732659, + "grad_norm": 0.7232657670974731, + "learning_rate": 0.0009519834155087405, + "loss": 0.5318, + "step": 1819 + }, + { + "epoch": 0.10168449870100846, + "grad_norm": 1.1293115615844727, + "learning_rate": 0.0009519554011653967, + "loss": 0.5389, + "step": 1820 + }, + { + "epoch": 0.10174036930469034, + "grad_norm": 2.8240671157836914, + "learning_rate": 0.0009519273868220529, + "loss": 0.5303, + "step": 1821 + }, + { + "epoch": 0.10179623990837221, + "grad_norm": 0.634150505065918, + "learning_rate": 0.0009518993724787092, + "loss": 0.5055, + "step": 1822 + }, + { + "epoch": 0.10185211051205409, + "grad_norm": 1.006459355354309, + "learning_rate": 0.0009518713581353653, + "loss": 0.5992, + "step": 1823 + }, + { + "epoch": 0.10190798111573596, + "grad_norm": 0.8416535258293152, + "learning_rate": 0.0009518433437920215, + "loss": 0.619, + "step": 1824 + }, + { + "epoch": 0.10196385171941783, + "grad_norm": 0.4263801574707031, + "learning_rate": 0.0009518153294486778, + "loss": 0.4909, + "step": 1825 + }, + { + "epoch": 0.1020197223230997, + "grad_norm": 0.6772006154060364, + "learning_rate": 0.000951787315105334, + "loss": 0.5282, + "step": 1826 + }, + { + "epoch": 0.10207559292678157, + "grad_norm": 0.5484404563903809, + "learning_rate": 0.0009517593007619901, + "loss": 0.4968, + "step": 1827 + }, + { + "epoch": 0.10213146353046344, + "grad_norm": 0.6760662794113159, + "learning_rate": 0.0009517312864186463, + "loss": 0.5579, + "step": 1828 + }, + { + "epoch": 0.10218733413414532, + "grad_norm": 0.5739109516143799, + "learning_rate": 0.0009517032720753026, + "loss": 0.3836, + "step": 1829 + }, + { + "epoch": 0.10224320473782719, + "grad_norm": 1.1255030632019043, + "learning_rate": 0.0009516752577319587, + "loss": 0.503, + "step": 1830 + }, + { + "epoch": 0.10229907534150906, + "grad_norm": 0.4691365957260132, + "learning_rate": 0.0009516472433886149, + "loss": 0.4744, + "step": 1831 + }, + { + "epoch": 0.10235494594519094, + "grad_norm": 0.4942130446434021, + "learning_rate": 0.0009516192290452713, + "loss": 0.511, + "step": 1832 + }, + { + "epoch": 0.10241081654887281, + "grad_norm": 0.5312089323997498, + "learning_rate": 0.0009515912147019275, + "loss": 0.4664, + "step": 1833 + }, + { + "epoch": 0.10246668715255468, + "grad_norm": 1.1877295970916748, + "learning_rate": 0.0009515632003585836, + "loss": 0.4855, + "step": 1834 + }, + { + "epoch": 0.10252255775623656, + "grad_norm": 2.883685350418091, + "learning_rate": 0.0009515351860152398, + "loss": 0.4789, + "step": 1835 + }, + { + "epoch": 0.10257842835991843, + "grad_norm": 0.9810858368873596, + "learning_rate": 0.0009515071716718961, + "loss": 0.6007, + "step": 1836 + }, + { + "epoch": 0.1026342989636003, + "grad_norm": 0.8969569206237793, + "learning_rate": 0.0009514791573285523, + "loss": 0.46, + "step": 1837 + }, + { + "epoch": 0.10269016956728218, + "grad_norm": 1.252123236656189, + "learning_rate": 0.0009514511429852084, + "loss": 0.4417, + "step": 1838 + }, + { + "epoch": 0.10274604017096405, + "grad_norm": 2.4568724632263184, + "learning_rate": 0.0009514231286418646, + "loss": 0.3915, + "step": 1839 + }, + { + "epoch": 0.10280191077464591, + "grad_norm": 0.5393562912940979, + "learning_rate": 0.0009513951142985209, + "loss": 0.4574, + "step": 1840 + }, + { + "epoch": 0.1028577813783278, + "grad_norm": 0.9116827249526978, + "learning_rate": 0.0009513670999551771, + "loss": 0.5374, + "step": 1841 + }, + { + "epoch": 0.10291365198200966, + "grad_norm": 0.4777604341506958, + "learning_rate": 0.0009513390856118332, + "loss": 0.4751, + "step": 1842 + }, + { + "epoch": 0.10296952258569154, + "grad_norm": 7.540298938751221, + "learning_rate": 0.0009513110712684895, + "loss": 0.5925, + "step": 1843 + }, + { + "epoch": 0.10302539318937341, + "grad_norm": 0.5123234391212463, + "learning_rate": 0.0009512830569251457, + "loss": 0.4939, + "step": 1844 + }, + { + "epoch": 0.10308126379305528, + "grad_norm": 0.4342348277568817, + "learning_rate": 0.0009512550425818019, + "loss": 0.3576, + "step": 1845 + }, + { + "epoch": 0.10313713439673716, + "grad_norm": 0.7672308087348938, + "learning_rate": 0.000951227028238458, + "loss": 0.5694, + "step": 1846 + }, + { + "epoch": 0.10319300500041903, + "grad_norm": 2.0108048915863037, + "learning_rate": 0.0009511990138951143, + "loss": 0.4989, + "step": 1847 + }, + { + "epoch": 0.1032488756041009, + "grad_norm": 0.9256094098091125, + "learning_rate": 0.0009511709995517706, + "loss": 0.3753, + "step": 1848 + }, + { + "epoch": 0.10330474620778278, + "grad_norm": 0.5214885473251343, + "learning_rate": 0.0009511429852084266, + "loss": 0.3905, + "step": 1849 + }, + { + "epoch": 0.10336061681146465, + "grad_norm": 0.6119325757026672, + "learning_rate": 0.000951114970865083, + "loss": 0.5621, + "step": 1850 + }, + { + "epoch": 0.10341648741514652, + "grad_norm": 0.9127707481384277, + "learning_rate": 0.0009510869565217392, + "loss": 0.5248, + "step": 1851 + }, + { + "epoch": 0.1034723580188284, + "grad_norm": 0.42444708943367004, + "learning_rate": 0.0009510589421783954, + "loss": 0.4155, + "step": 1852 + }, + { + "epoch": 0.10352822862251027, + "grad_norm": 0.7199243903160095, + "learning_rate": 0.0009510309278350515, + "loss": 0.5126, + "step": 1853 + }, + { + "epoch": 0.10358409922619213, + "grad_norm": 0.8422122001647949, + "learning_rate": 0.0009510029134917078, + "loss": 0.481, + "step": 1854 + }, + { + "epoch": 0.10363996982987402, + "grad_norm": 0.4364738464355469, + "learning_rate": 0.000950974899148364, + "loss": 0.3926, + "step": 1855 + }, + { + "epoch": 0.10369584043355588, + "grad_norm": 0.6307455897331238, + "learning_rate": 0.0009509468848050202, + "loss": 0.3767, + "step": 1856 + }, + { + "epoch": 0.10375171103723775, + "grad_norm": 0.8068780303001404, + "learning_rate": 0.0009509188704616764, + "loss": 0.4685, + "step": 1857 + }, + { + "epoch": 0.10380758164091963, + "grad_norm": 0.5496926307678223, + "learning_rate": 0.0009508908561183326, + "loss": 0.423, + "step": 1858 + }, + { + "epoch": 0.1038634522446015, + "grad_norm": 3.157343864440918, + "learning_rate": 0.0009508628417749888, + "loss": 0.4586, + "step": 1859 + }, + { + "epoch": 0.10391932284828338, + "grad_norm": 1.002434253692627, + "learning_rate": 0.000950834827431645, + "loss": 0.523, + "step": 1860 + }, + { + "epoch": 0.10397519345196525, + "grad_norm": 0.762684166431427, + "learning_rate": 0.0009508068130883012, + "loss": 0.5307, + "step": 1861 + }, + { + "epoch": 0.10403106405564712, + "grad_norm": 0.4304274618625641, + "learning_rate": 0.0009507787987449574, + "loss": 0.4407, + "step": 1862 + }, + { + "epoch": 0.104086934659329, + "grad_norm": 0.5602187514305115, + "learning_rate": 0.0009507507844016136, + "loss": 0.5649, + "step": 1863 + }, + { + "epoch": 0.10414280526301087, + "grad_norm": 1.0350298881530762, + "learning_rate": 0.00095072277005827, + "loss": 0.4557, + "step": 1864 + }, + { + "epoch": 0.10419867586669274, + "grad_norm": 0.6891487836837769, + "learning_rate": 0.000950694755714926, + "loss": 0.4554, + "step": 1865 + }, + { + "epoch": 0.10425454647037462, + "grad_norm": 0.8672835230827332, + "learning_rate": 0.0009506667413715823, + "loss": 0.5153, + "step": 1866 + }, + { + "epoch": 0.10431041707405649, + "grad_norm": 2.0053491592407227, + "learning_rate": 0.0009506387270282385, + "loss": 0.5543, + "step": 1867 + }, + { + "epoch": 0.10436628767773835, + "grad_norm": 0.6168586611747742, + "learning_rate": 0.0009506107126848948, + "loss": 0.5212, + "step": 1868 + }, + { + "epoch": 0.10442215828142024, + "grad_norm": 1.3812037706375122, + "learning_rate": 0.0009505826983415509, + "loss": 0.5302, + "step": 1869 + }, + { + "epoch": 0.1044780288851021, + "grad_norm": 0.8711324334144592, + "learning_rate": 0.0009505546839982071, + "loss": 0.6619, + "step": 1870 + }, + { + "epoch": 0.10453389948878397, + "grad_norm": 0.5571511387825012, + "learning_rate": 0.0009505266696548634, + "loss": 0.5355, + "step": 1871 + }, + { + "epoch": 0.10458977009246585, + "grad_norm": 0.9310008883476257, + "learning_rate": 0.0009504986553115195, + "loss": 0.4341, + "step": 1872 + }, + { + "epoch": 0.10464564069614772, + "grad_norm": 0.4570696949958801, + "learning_rate": 0.0009504706409681757, + "loss": 0.469, + "step": 1873 + }, + { + "epoch": 0.10470151129982959, + "grad_norm": 0.6371240019798279, + "learning_rate": 0.0009504426266248319, + "loss": 0.4849, + "step": 1874 + }, + { + "epoch": 0.10475738190351147, + "grad_norm": 1.2440767288208008, + "learning_rate": 0.0009504146122814882, + "loss": 0.4364, + "step": 1875 + }, + { + "epoch": 0.10481325250719334, + "grad_norm": 0.7217833995819092, + "learning_rate": 0.0009503865979381443, + "loss": 0.5314, + "step": 1876 + }, + { + "epoch": 0.1048691231108752, + "grad_norm": 0.8053521513938904, + "learning_rate": 0.0009503585835948005, + "loss": 0.3985, + "step": 1877 + }, + { + "epoch": 0.10492499371455709, + "grad_norm": 0.6146766543388367, + "learning_rate": 0.0009503305692514567, + "loss": 0.5932, + "step": 1878 + }, + { + "epoch": 0.10498086431823896, + "grad_norm": 0.4758443236351013, + "learning_rate": 0.000950302554908113, + "loss": 0.4071, + "step": 1879 + }, + { + "epoch": 0.10503673492192084, + "grad_norm": 0.8042827844619751, + "learning_rate": 0.0009502745405647691, + "loss": 0.5077, + "step": 1880 + }, + { + "epoch": 0.1050926055256027, + "grad_norm": 0.8773736357688904, + "learning_rate": 0.0009502465262214253, + "loss": 0.4483, + "step": 1881 + }, + { + "epoch": 0.10514847612928457, + "grad_norm": 0.6290498375892639, + "learning_rate": 0.0009502185118780817, + "loss": 0.611, + "step": 1882 + }, + { + "epoch": 0.10520434673296646, + "grad_norm": 0.8071674108505249, + "learning_rate": 0.0009501904975347379, + "loss": 0.471, + "step": 1883 + }, + { + "epoch": 0.10526021733664832, + "grad_norm": 0.4541594982147217, + "learning_rate": 0.000950162483191394, + "loss": 0.3888, + "step": 1884 + }, + { + "epoch": 0.10531608794033019, + "grad_norm": 0.5365697741508484, + "learning_rate": 0.0009501344688480502, + "loss": 0.5255, + "step": 1885 + }, + { + "epoch": 0.10537195854401207, + "grad_norm": 0.8379001021385193, + "learning_rate": 0.0009501064545047065, + "loss": 0.655, + "step": 1886 + }, + { + "epoch": 0.10542782914769394, + "grad_norm": 0.6010827422142029, + "learning_rate": 0.0009500784401613627, + "loss": 0.4916, + "step": 1887 + }, + { + "epoch": 0.10548369975137581, + "grad_norm": 0.7894545793533325, + "learning_rate": 0.0009500504258180188, + "loss": 0.5259, + "step": 1888 + }, + { + "epoch": 0.10553957035505769, + "grad_norm": 2.844721794128418, + "learning_rate": 0.0009500224114746751, + "loss": 0.524, + "step": 1889 + }, + { + "epoch": 0.10559544095873956, + "grad_norm": 0.6844349503517151, + "learning_rate": 0.0009499943971313313, + "loss": 0.5765, + "step": 1890 + }, + { + "epoch": 0.10565131156242143, + "grad_norm": 0.5386525988578796, + "learning_rate": 0.0009499663827879874, + "loss": 0.4328, + "step": 1891 + }, + { + "epoch": 0.10570718216610331, + "grad_norm": 0.5972598791122437, + "learning_rate": 0.0009499383684446436, + "loss": 0.4252, + "step": 1892 + }, + { + "epoch": 0.10576305276978518, + "grad_norm": 0.5130574107170105, + "learning_rate": 0.0009499103541012999, + "loss": 0.467, + "step": 1893 + }, + { + "epoch": 0.10581892337346704, + "grad_norm": 1.1407084465026855, + "learning_rate": 0.0009498823397579561, + "loss": 0.4899, + "step": 1894 + }, + { + "epoch": 0.10587479397714893, + "grad_norm": 0.9882235527038574, + "learning_rate": 0.0009498543254146122, + "loss": 0.6693, + "step": 1895 + }, + { + "epoch": 0.1059306645808308, + "grad_norm": 0.9387367963790894, + "learning_rate": 0.0009498263110712685, + "loss": 0.5701, + "step": 1896 + }, + { + "epoch": 0.10598653518451266, + "grad_norm": 0.601222038269043, + "learning_rate": 0.0009497982967279247, + "loss": 0.4574, + "step": 1897 + }, + { + "epoch": 0.10604240578819454, + "grad_norm": 0.7352948188781738, + "learning_rate": 0.000949770282384581, + "loss": 0.4518, + "step": 1898 + }, + { + "epoch": 0.10609827639187641, + "grad_norm": 0.8132120966911316, + "learning_rate": 0.000949742268041237, + "loss": 0.5796, + "step": 1899 + }, + { + "epoch": 0.1061541469955583, + "grad_norm": 0.44306132197380066, + "learning_rate": 0.0009497142536978934, + "loss": 0.4978, + "step": 1900 + }, + { + "epoch": 0.10621001759924016, + "grad_norm": 0.46563243865966797, + "learning_rate": 0.0009496862393545496, + "loss": 0.5069, + "step": 1901 + }, + { + "epoch": 0.10626588820292203, + "grad_norm": 6.284140110015869, + "learning_rate": 0.0009496582250112058, + "loss": 0.4791, + "step": 1902 + }, + { + "epoch": 0.10632175880660391, + "grad_norm": 2.0974669456481934, + "learning_rate": 0.000949630210667862, + "loss": 0.4975, + "step": 1903 + }, + { + "epoch": 0.10637762941028578, + "grad_norm": 0.7226356863975525, + "learning_rate": 0.0009496021963245182, + "loss": 0.572, + "step": 1904 + }, + { + "epoch": 0.10643350001396765, + "grad_norm": 0.8199789524078369, + "learning_rate": 0.0009495741819811744, + "loss": 0.5167, + "step": 1905 + }, + { + "epoch": 0.10648937061764953, + "grad_norm": 0.8601472973823547, + "learning_rate": 0.0009495461676378306, + "loss": 0.4942, + "step": 1906 + }, + { + "epoch": 0.1065452412213314, + "grad_norm": 2.1813905239105225, + "learning_rate": 0.0009495181532944868, + "loss": 0.4322, + "step": 1907 + }, + { + "epoch": 0.10660111182501326, + "grad_norm": 0.6993409991264343, + "learning_rate": 0.000949490138951143, + "loss": 0.415, + "step": 1908 + }, + { + "epoch": 0.10665698242869515, + "grad_norm": 0.6137135028839111, + "learning_rate": 0.0009494621246077992, + "loss": 0.5906, + "step": 1909 + }, + { + "epoch": 0.10671285303237701, + "grad_norm": 0.5273703932762146, + "learning_rate": 0.0009494341102644555, + "loss": 0.5715, + "step": 1910 + }, + { + "epoch": 0.10676872363605888, + "grad_norm": 0.8477743864059448, + "learning_rate": 0.0009494060959211116, + "loss": 0.3305, + "step": 1911 + }, + { + "epoch": 0.10682459423974076, + "grad_norm": 0.5541388988494873, + "learning_rate": 0.0009493780815777678, + "loss": 0.4707, + "step": 1912 + }, + { + "epoch": 0.10688046484342263, + "grad_norm": 0.7644373178482056, + "learning_rate": 0.000949350067234424, + "loss": 0.6427, + "step": 1913 + }, + { + "epoch": 0.1069363354471045, + "grad_norm": 0.5015829801559448, + "learning_rate": 0.0009493220528910802, + "loss": 0.5629, + "step": 1914 + }, + { + "epoch": 0.10699220605078638, + "grad_norm": 0.5143448710441589, + "learning_rate": 0.0009492940385477364, + "loss": 0.497, + "step": 1915 + }, + { + "epoch": 0.10704807665446825, + "grad_norm": 0.6617125868797302, + "learning_rate": 0.0009492660242043927, + "loss": 0.6499, + "step": 1916 + }, + { + "epoch": 0.10710394725815013, + "grad_norm": 1.7038806676864624, + "learning_rate": 0.0009492380098610489, + "loss": 0.4907, + "step": 1917 + }, + { + "epoch": 0.107159817861832, + "grad_norm": 0.8535299897193909, + "learning_rate": 0.0009492099955177051, + "loss": 0.5476, + "step": 1918 + }, + { + "epoch": 0.10721568846551387, + "grad_norm": 0.6506881713867188, + "learning_rate": 0.0009491819811743613, + "loss": 0.4202, + "step": 1919 + }, + { + "epoch": 0.10727155906919575, + "grad_norm": 0.586366593837738, + "learning_rate": 0.0009491539668310175, + "loss": 0.4899, + "step": 1920 + }, + { + "epoch": 0.10732742967287762, + "grad_norm": 0.6719419956207275, + "learning_rate": 0.0009491259524876738, + "loss": 0.4443, + "step": 1921 + }, + { + "epoch": 0.10738330027655948, + "grad_norm": 2.1756560802459717, + "learning_rate": 0.0009490979381443299, + "loss": 0.4988, + "step": 1922 + }, + { + "epoch": 0.10743917088024137, + "grad_norm": 0.627149760723114, + "learning_rate": 0.0009490699238009861, + "loss": 0.6102, + "step": 1923 + }, + { + "epoch": 0.10749504148392323, + "grad_norm": 1.4481992721557617, + "learning_rate": 0.0009490419094576423, + "loss": 0.5465, + "step": 1924 + }, + { + "epoch": 0.1075509120876051, + "grad_norm": 0.6017656326293945, + "learning_rate": 0.0009490138951142986, + "loss": 0.4926, + "step": 1925 + }, + { + "epoch": 0.10760678269128698, + "grad_norm": 0.5886608958244324, + "learning_rate": 0.0009489858807709547, + "loss": 0.4631, + "step": 1926 + }, + { + "epoch": 0.10766265329496885, + "grad_norm": 0.7403014302253723, + "learning_rate": 0.0009489578664276109, + "loss": 0.4747, + "step": 1927 + }, + { + "epoch": 0.10771852389865072, + "grad_norm": 0.799184262752533, + "learning_rate": 0.0009489298520842672, + "loss": 0.5434, + "step": 1928 + }, + { + "epoch": 0.1077743945023326, + "grad_norm": 1.5089305639266968, + "learning_rate": 0.0009489018377409234, + "loss": 0.5042, + "step": 1929 + }, + { + "epoch": 0.10783026510601447, + "grad_norm": 0.6355103254318237, + "learning_rate": 0.0009488738233975795, + "loss": 0.5632, + "step": 1930 + }, + { + "epoch": 0.10788613570969634, + "grad_norm": 0.5459015369415283, + "learning_rate": 0.0009488458090542357, + "loss": 0.5488, + "step": 1931 + }, + { + "epoch": 0.10794200631337822, + "grad_norm": 0.4808809459209442, + "learning_rate": 0.0009488177947108921, + "loss": 0.488, + "step": 1932 + }, + { + "epoch": 0.10799787691706009, + "grad_norm": 0.7002121806144714, + "learning_rate": 0.0009487897803675482, + "loss": 0.61, + "step": 1933 + }, + { + "epoch": 0.10805374752074196, + "grad_norm": 0.4634250998497009, + "learning_rate": 0.0009487617660242044, + "loss": 0.4016, + "step": 1934 + }, + { + "epoch": 0.10810961812442384, + "grad_norm": 3.473952293395996, + "learning_rate": 0.0009487337516808607, + "loss": 0.4622, + "step": 1935 + }, + { + "epoch": 0.1081654887281057, + "grad_norm": 0.4732334017753601, + "learning_rate": 0.0009487057373375169, + "loss": 0.5168, + "step": 1936 + }, + { + "epoch": 0.10822135933178759, + "grad_norm": 2.032881498336792, + "learning_rate": 0.000948677722994173, + "loss": 0.4457, + "step": 1937 + }, + { + "epoch": 0.10827722993546945, + "grad_norm": 0.8290843367576599, + "learning_rate": 0.0009486497086508292, + "loss": 0.5574, + "step": 1938 + }, + { + "epoch": 0.10833310053915132, + "grad_norm": 1.6392165422439575, + "learning_rate": 0.0009486216943074855, + "loss": 0.8935, + "step": 1939 + }, + { + "epoch": 0.1083889711428332, + "grad_norm": 2.342341899871826, + "learning_rate": 0.0009485936799641417, + "loss": 0.4304, + "step": 1940 + }, + { + "epoch": 0.10844484174651507, + "grad_norm": 0.5193524360656738, + "learning_rate": 0.0009485656656207978, + "loss": 0.4875, + "step": 1941 + }, + { + "epoch": 0.10850071235019694, + "grad_norm": 0.5474240779876709, + "learning_rate": 0.0009485376512774541, + "loss": 0.4916, + "step": 1942 + }, + { + "epoch": 0.10855658295387882, + "grad_norm": 0.5389048457145691, + "learning_rate": 0.0009485096369341103, + "loss": 0.4698, + "step": 1943 + }, + { + "epoch": 0.10861245355756069, + "grad_norm": 0.632021963596344, + "learning_rate": 0.0009484816225907665, + "loss": 0.618, + "step": 1944 + }, + { + "epoch": 0.10866832416124256, + "grad_norm": 1.005350947380066, + "learning_rate": 0.0009484536082474226, + "loss": 0.6136, + "step": 1945 + }, + { + "epoch": 0.10872419476492444, + "grad_norm": 0.426517516374588, + "learning_rate": 0.0009484255939040789, + "loss": 0.4423, + "step": 1946 + }, + { + "epoch": 0.10878006536860631, + "grad_norm": 1.1932344436645508, + "learning_rate": 0.0009483975795607351, + "loss": 0.7346, + "step": 1947 + }, + { + "epoch": 0.10883593597228818, + "grad_norm": 0.44669416546821594, + "learning_rate": 0.0009483695652173914, + "loss": 0.4194, + "step": 1948 + }, + { + "epoch": 0.10889180657597006, + "grad_norm": 1.122280478477478, + "learning_rate": 0.0009483415508740474, + "loss": 0.6062, + "step": 1949 + }, + { + "epoch": 0.10894767717965192, + "grad_norm": 0.6865066289901733, + "learning_rate": 0.0009483135365307038, + "loss": 0.5031, + "step": 1950 + }, + { + "epoch": 0.10900354778333379, + "grad_norm": 1.00994074344635, + "learning_rate": 0.00094828552218736, + "loss": 0.4638, + "step": 1951 + }, + { + "epoch": 0.10905941838701567, + "grad_norm": 0.5694208145141602, + "learning_rate": 0.0009482575078440162, + "loss": 0.5165, + "step": 1952 + }, + { + "epoch": 0.10911528899069754, + "grad_norm": 0.5136293768882751, + "learning_rate": 0.0009482294935006724, + "loss": 0.4931, + "step": 1953 + }, + { + "epoch": 0.10917115959437941, + "grad_norm": 0.6891731023788452, + "learning_rate": 0.0009482014791573286, + "loss": 0.4358, + "step": 1954 + }, + { + "epoch": 0.10922703019806129, + "grad_norm": 0.5973116755485535, + "learning_rate": 0.0009481734648139848, + "loss": 0.4061, + "step": 1955 + }, + { + "epoch": 0.10928290080174316, + "grad_norm": 0.9311191439628601, + "learning_rate": 0.0009481454504706409, + "loss": 0.5055, + "step": 1956 + }, + { + "epoch": 0.10933877140542504, + "grad_norm": 3.724738121032715, + "learning_rate": 0.0009481174361272972, + "loss": 0.5712, + "step": 1957 + }, + { + "epoch": 0.10939464200910691, + "grad_norm": 0.6356360912322998, + "learning_rate": 0.0009480894217839534, + "loss": 0.47, + "step": 1958 + }, + { + "epoch": 0.10945051261278878, + "grad_norm": 1.118715763092041, + "learning_rate": 0.0009480614074406096, + "loss": 0.5199, + "step": 1959 + }, + { + "epoch": 0.10950638321647066, + "grad_norm": 1.1779310703277588, + "learning_rate": 0.0009480333930972658, + "loss": 0.6001, + "step": 1960 + }, + { + "epoch": 0.10956225382015253, + "grad_norm": 0.8611148595809937, + "learning_rate": 0.000948005378753922, + "loss": 0.5395, + "step": 1961 + }, + { + "epoch": 0.1096181244238344, + "grad_norm": 1.273004412651062, + "learning_rate": 0.0009479773644105782, + "loss": 0.7232, + "step": 1962 + }, + { + "epoch": 0.10967399502751628, + "grad_norm": 0.7096453309059143, + "learning_rate": 0.0009479493500672344, + "loss": 0.4104, + "step": 1963 + }, + { + "epoch": 0.10972986563119815, + "grad_norm": 0.7990739941596985, + "learning_rate": 0.0009479213357238906, + "loss": 0.5483, + "step": 1964 + }, + { + "epoch": 0.10978573623488001, + "grad_norm": 0.7700029611587524, + "learning_rate": 0.0009478933213805468, + "loss": 0.503, + "step": 1965 + }, + { + "epoch": 0.1098416068385619, + "grad_norm": 0.5771376490592957, + "learning_rate": 0.0009478653070372031, + "loss": 0.6048, + "step": 1966 + }, + { + "epoch": 0.10989747744224376, + "grad_norm": 0.5989543795585632, + "learning_rate": 0.0009478372926938594, + "loss": 0.5832, + "step": 1967 + }, + { + "epoch": 0.10995334804592563, + "grad_norm": 0.588057279586792, + "learning_rate": 0.0009478092783505155, + "loss": 0.5803, + "step": 1968 + }, + { + "epoch": 0.11000921864960751, + "grad_norm": 0.8072620034217834, + "learning_rate": 0.0009477812640071717, + "loss": 0.5024, + "step": 1969 + }, + { + "epoch": 0.11006508925328938, + "grad_norm": 0.7584587335586548, + "learning_rate": 0.0009477532496638279, + "loss": 0.5576, + "step": 1970 + }, + { + "epoch": 0.11012095985697125, + "grad_norm": 0.5190718770027161, + "learning_rate": 0.0009477252353204842, + "loss": 0.4683, + "step": 1971 + }, + { + "epoch": 0.11017683046065313, + "grad_norm": 0.6593866944313049, + "learning_rate": 0.0009476972209771403, + "loss": 0.514, + "step": 1972 + }, + { + "epoch": 0.110232701064335, + "grad_norm": 0.7128916382789612, + "learning_rate": 0.0009476692066337965, + "loss": 0.5104, + "step": 1973 + }, + { + "epoch": 0.11028857166801688, + "grad_norm": 0.8403809070587158, + "learning_rate": 0.0009476411922904528, + "loss": 0.5481, + "step": 1974 + }, + { + "epoch": 0.11034444227169875, + "grad_norm": 0.55734783411026, + "learning_rate": 0.0009476131779471089, + "loss": 0.4892, + "step": 1975 + }, + { + "epoch": 0.11040031287538062, + "grad_norm": 0.6262481212615967, + "learning_rate": 0.0009475851636037651, + "loss": 0.6007, + "step": 1976 + }, + { + "epoch": 0.1104561834790625, + "grad_norm": 0.6474262475967407, + "learning_rate": 0.0009475571492604213, + "loss": 0.5397, + "step": 1977 + }, + { + "epoch": 0.11051205408274437, + "grad_norm": 0.9546395540237427, + "learning_rate": 0.0009475291349170776, + "loss": 0.4909, + "step": 1978 + }, + { + "epoch": 0.11056792468642623, + "grad_norm": 1.211961030960083, + "learning_rate": 0.0009475011205737337, + "loss": 0.4354, + "step": 1979 + }, + { + "epoch": 0.11062379529010812, + "grad_norm": 3.478574752807617, + "learning_rate": 0.0009474731062303899, + "loss": 0.5704, + "step": 1980 + }, + { + "epoch": 0.11067966589378998, + "grad_norm": 1.0694595575332642, + "learning_rate": 0.0009474450918870461, + "loss": 0.445, + "step": 1981 + }, + { + "epoch": 0.11073553649747185, + "grad_norm": 0.7023395895957947, + "learning_rate": 0.0009474170775437025, + "loss": 0.5871, + "step": 1982 + }, + { + "epoch": 0.11079140710115373, + "grad_norm": 0.6397503614425659, + "learning_rate": 0.0009473890632003586, + "loss": 0.5052, + "step": 1983 + }, + { + "epoch": 0.1108472777048356, + "grad_norm": 0.6123632192611694, + "learning_rate": 0.0009473610488570148, + "loss": 0.3892, + "step": 1984 + }, + { + "epoch": 0.11090314830851747, + "grad_norm": 0.39251452684402466, + "learning_rate": 0.0009473330345136711, + "loss": 0.4211, + "step": 1985 + }, + { + "epoch": 0.11095901891219935, + "grad_norm": 0.5877288579940796, + "learning_rate": 0.0009473050201703273, + "loss": 0.5047, + "step": 1986 + }, + { + "epoch": 0.11101488951588122, + "grad_norm": 0.5071821808815002, + "learning_rate": 0.0009472770058269834, + "loss": 0.4783, + "step": 1987 + }, + { + "epoch": 0.11107076011956309, + "grad_norm": 0.5443317890167236, + "learning_rate": 0.0009472489914836396, + "loss": 0.5829, + "step": 1988 + }, + { + "epoch": 0.11112663072324497, + "grad_norm": 3.9766952991485596, + "learning_rate": 0.0009472209771402959, + "loss": 0.5135, + "step": 1989 + }, + { + "epoch": 0.11118250132692684, + "grad_norm": 1.36964750289917, + "learning_rate": 0.0009471929627969521, + "loss": 0.4811, + "step": 1990 + }, + { + "epoch": 0.1112383719306087, + "grad_norm": 0.6804949045181274, + "learning_rate": 0.0009471649484536082, + "loss": 0.6065, + "step": 1991 + }, + { + "epoch": 0.11129424253429059, + "grad_norm": 1.0986756086349487, + "learning_rate": 0.0009471369341102645, + "loss": 0.5733, + "step": 1992 + }, + { + "epoch": 0.11135011313797245, + "grad_norm": 1.430835247039795, + "learning_rate": 0.0009471089197669207, + "loss": 0.4121, + "step": 1993 + }, + { + "epoch": 0.11140598374165434, + "grad_norm": 0.5699262022972107, + "learning_rate": 0.0009470809054235769, + "loss": 0.4112, + "step": 1994 + }, + { + "epoch": 0.1114618543453362, + "grad_norm": 0.6627857685089111, + "learning_rate": 0.000947052891080233, + "loss": 0.5284, + "step": 1995 + }, + { + "epoch": 0.11151772494901807, + "grad_norm": 0.7134384512901306, + "learning_rate": 0.0009470248767368893, + "loss": 0.4567, + "step": 1996 + }, + { + "epoch": 0.11157359555269995, + "grad_norm": 0.506657063961029, + "learning_rate": 0.0009469968623935455, + "loss": 0.4083, + "step": 1997 + }, + { + "epoch": 0.11162946615638182, + "grad_norm": 1.3553966283798218, + "learning_rate": 0.0009469688480502016, + "loss": 0.4629, + "step": 1998 + }, + { + "epoch": 0.11168533676006369, + "grad_norm": 0.5635818243026733, + "learning_rate": 0.000946940833706858, + "loss": 0.5149, + "step": 1999 + }, + { + "epoch": 0.11174120736374557, + "grad_norm": 0.9314305186271667, + "learning_rate": 0.0009469128193635142, + "loss": 0.6257, + "step": 2000 + }, + { + "epoch": 0.11174120736374557, + "eval_cer": 0.09944573554602686, + "eval_loss": 0.37101906538009644, + "eval_runtime": 55.9202, + "eval_samples_per_second": 81.151, + "eval_steps_per_second": 5.079, + "eval_wer": 0.39303371414206306, + "step": 2000 + }, + { + "epoch": 0.11179707796742744, + "grad_norm": 0.5191553235054016, + "learning_rate": 0.0009468848050201704, + "loss": 0.4812, + "step": 2001 + }, + { + "epoch": 0.1118529485711093, + "grad_norm": 13.674315452575684, + "learning_rate": 0.0009468567906768265, + "loss": 0.5499, + "step": 2002 + }, + { + "epoch": 0.11190881917479119, + "grad_norm": 0.7816175222396851, + "learning_rate": 0.0009468287763334828, + "loss": 0.5728, + "step": 2003 + }, + { + "epoch": 0.11196468977847306, + "grad_norm": 0.602145791053772, + "learning_rate": 0.000946800761990139, + "loss": 0.5228, + "step": 2004 + }, + { + "epoch": 0.11202056038215492, + "grad_norm": 3.2499842643737793, + "learning_rate": 0.0009467727476467952, + "loss": 0.4916, + "step": 2005 + }, + { + "epoch": 0.1120764309858368, + "grad_norm": 0.7166106700897217, + "learning_rate": 0.0009467447333034514, + "loss": 0.4471, + "step": 2006 + }, + { + "epoch": 0.11213230158951867, + "grad_norm": 1.5544142723083496, + "learning_rate": 0.0009467167189601076, + "loss": 0.4821, + "step": 2007 + }, + { + "epoch": 0.11218817219320054, + "grad_norm": 0.513493001461029, + "learning_rate": 0.0009466887046167638, + "loss": 0.4745, + "step": 2008 + }, + { + "epoch": 0.11224404279688242, + "grad_norm": 3.4371232986450195, + "learning_rate": 0.00094666069027342, + "loss": 0.4651, + "step": 2009 + }, + { + "epoch": 0.11229991340056429, + "grad_norm": 1.2598001956939697, + "learning_rate": 0.0009466326759300762, + "loss": 0.5645, + "step": 2010 + }, + { + "epoch": 0.11235578400424616, + "grad_norm": 0.5782309770584106, + "learning_rate": 0.0009466046615867324, + "loss": 0.4455, + "step": 2011 + }, + { + "epoch": 0.11241165460792804, + "grad_norm": 0.5934144258499146, + "learning_rate": 0.0009465766472433886, + "loss": 0.4599, + "step": 2012 + }, + { + "epoch": 0.11246752521160991, + "grad_norm": 1.001348853111267, + "learning_rate": 0.0009465486329000449, + "loss": 0.4652, + "step": 2013 + }, + { + "epoch": 0.11252339581529179, + "grad_norm": 0.8479374051094055, + "learning_rate": 0.000946520618556701, + "loss": 0.457, + "step": 2014 + }, + { + "epoch": 0.11257926641897366, + "grad_norm": 0.5280236601829529, + "learning_rate": 0.0009464926042133572, + "loss": 0.3594, + "step": 2015 + }, + { + "epoch": 0.11263513702265553, + "grad_norm": 6.638137340545654, + "learning_rate": 0.0009464645898700135, + "loss": 0.553, + "step": 2016 + }, + { + "epoch": 0.11269100762633741, + "grad_norm": 0.5598524808883667, + "learning_rate": 0.0009464365755266698, + "loss": 0.4583, + "step": 2017 + }, + { + "epoch": 0.11274687823001928, + "grad_norm": 2.2101151943206787, + "learning_rate": 0.0009464085611833259, + "loss": 0.4275, + "step": 2018 + }, + { + "epoch": 0.11280274883370114, + "grad_norm": 0.7216609120368958, + "learning_rate": 0.0009463805468399821, + "loss": 0.5524, + "step": 2019 + }, + { + "epoch": 0.11285861943738303, + "grad_norm": 3.3117897510528564, + "learning_rate": 0.0009463525324966383, + "loss": 0.5135, + "step": 2020 + }, + { + "epoch": 0.1129144900410649, + "grad_norm": 0.8048688769340515, + "learning_rate": 0.0009463245181532945, + "loss": 0.5594, + "step": 2021 + }, + { + "epoch": 0.11297036064474676, + "grad_norm": 2.1196467876434326, + "learning_rate": 0.0009462965038099507, + "loss": 0.8114, + "step": 2022 + }, + { + "epoch": 0.11302623124842864, + "grad_norm": 0.8233445882797241, + "learning_rate": 0.0009462684894666069, + "loss": 0.5078, + "step": 2023 + }, + { + "epoch": 0.11308210185211051, + "grad_norm": 0.5339500308036804, + "learning_rate": 0.0009462404751232632, + "loss": 0.5399, + "step": 2024 + }, + { + "epoch": 0.11313797245579238, + "grad_norm": 0.5323941111564636, + "learning_rate": 0.0009462124607799193, + "loss": 0.4304, + "step": 2025 + }, + { + "epoch": 0.11319384305947426, + "grad_norm": 0.4987124502658844, + "learning_rate": 0.0009461844464365755, + "loss": 0.5424, + "step": 2026 + }, + { + "epoch": 0.11324971366315613, + "grad_norm": 0.6799033284187317, + "learning_rate": 0.0009461564320932317, + "loss": 0.6264, + "step": 2027 + }, + { + "epoch": 0.113305584266838, + "grad_norm": 0.5243008732795715, + "learning_rate": 0.000946128417749888, + "loss": 0.4178, + "step": 2028 + }, + { + "epoch": 0.11336145487051988, + "grad_norm": 0.4481528401374817, + "learning_rate": 0.0009461004034065441, + "loss": 0.4603, + "step": 2029 + }, + { + "epoch": 0.11341732547420175, + "grad_norm": 0.5991536974906921, + "learning_rate": 0.0009460723890632003, + "loss": 0.3843, + "step": 2030 + }, + { + "epoch": 0.11347319607788361, + "grad_norm": 0.8581655025482178, + "learning_rate": 0.0009460443747198566, + "loss": 0.6438, + "step": 2031 + }, + { + "epoch": 0.1135290666815655, + "grad_norm": 0.5735349655151367, + "learning_rate": 0.0009460163603765129, + "loss": 0.5311, + "step": 2032 + }, + { + "epoch": 0.11358493728524736, + "grad_norm": 19.750274658203125, + "learning_rate": 0.000945988346033169, + "loss": 0.5812, + "step": 2033 + }, + { + "epoch": 0.11364080788892925, + "grad_norm": 0.5416042804718018, + "learning_rate": 0.0009459603316898252, + "loss": 0.523, + "step": 2034 + }, + { + "epoch": 0.11369667849261111, + "grad_norm": 0.9479570984840393, + "learning_rate": 0.0009459323173464815, + "loss": 0.3744, + "step": 2035 + }, + { + "epoch": 0.11375254909629298, + "grad_norm": 0.5689051151275635, + "learning_rate": 0.0009459043030031377, + "loss": 0.4827, + "step": 2036 + }, + { + "epoch": 0.11380841969997486, + "grad_norm": 0.5921659469604492, + "learning_rate": 0.0009458762886597938, + "loss": 0.507, + "step": 2037 + }, + { + "epoch": 0.11386429030365673, + "grad_norm": 0.4852313697338104, + "learning_rate": 0.0009458482743164501, + "loss": 0.5147, + "step": 2038 + }, + { + "epoch": 0.1139201609073386, + "grad_norm": 0.8255274891853333, + "learning_rate": 0.0009458202599731063, + "loss": 0.4353, + "step": 2039 + }, + { + "epoch": 0.11397603151102048, + "grad_norm": 0.6877469420433044, + "learning_rate": 0.0009457922456297624, + "loss": 0.445, + "step": 2040 + }, + { + "epoch": 0.11403190211470235, + "grad_norm": 1.8083528280258179, + "learning_rate": 0.0009457642312864186, + "loss": 0.4914, + "step": 2041 + }, + { + "epoch": 0.11408777271838422, + "grad_norm": 0.6493225693702698, + "learning_rate": 0.0009457362169430749, + "loss": 0.4937, + "step": 2042 + }, + { + "epoch": 0.1141436433220661, + "grad_norm": 0.4428676664829254, + "learning_rate": 0.0009457082025997311, + "loss": 0.3373, + "step": 2043 + }, + { + "epoch": 0.11419951392574797, + "grad_norm": 0.9638345837593079, + "learning_rate": 0.0009456801882563872, + "loss": 0.4597, + "step": 2044 + }, + { + "epoch": 0.11425538452942983, + "grad_norm": 0.8265095353126526, + "learning_rate": 0.0009456521739130435, + "loss": 0.4339, + "step": 2045 + }, + { + "epoch": 0.11431125513311172, + "grad_norm": 0.4673021733760834, + "learning_rate": 0.0009456241595696997, + "loss": 0.4958, + "step": 2046 + }, + { + "epoch": 0.11436712573679358, + "grad_norm": 0.44778406620025635, + "learning_rate": 0.0009455961452263559, + "loss": 0.5012, + "step": 2047 + }, + { + "epoch": 0.11442299634047545, + "grad_norm": 0.47487860918045044, + "learning_rate": 0.000945568130883012, + "loss": 0.4185, + "step": 2048 + }, + { + "epoch": 0.11447886694415733, + "grad_norm": 0.7937703132629395, + "learning_rate": 0.0009455401165396684, + "loss": 0.5575, + "step": 2049 + }, + { + "epoch": 0.1145347375478392, + "grad_norm": 0.9465872049331665, + "learning_rate": 0.0009455121021963246, + "loss": 0.7356, + "step": 2050 + }, + { + "epoch": 0.11459060815152108, + "grad_norm": 0.7032338380813599, + "learning_rate": 0.0009454840878529808, + "loss": 0.5327, + "step": 2051 + }, + { + "epoch": 0.11464647875520295, + "grad_norm": 0.7516316175460815, + "learning_rate": 0.0009454560735096369, + "loss": 0.5364, + "step": 2052 + }, + { + "epoch": 0.11470234935888482, + "grad_norm": 1.2842832803726196, + "learning_rate": 0.0009454280591662932, + "loss": 0.5345, + "step": 2053 + }, + { + "epoch": 0.1147582199625667, + "grad_norm": 1.1659198999404907, + "learning_rate": 0.0009454000448229494, + "loss": 0.4242, + "step": 2054 + }, + { + "epoch": 0.11481409056624857, + "grad_norm": 0.6989535689353943, + "learning_rate": 0.0009453720304796056, + "loss": 0.5323, + "step": 2055 + }, + { + "epoch": 0.11486996116993044, + "grad_norm": 0.6187009811401367, + "learning_rate": 0.0009453440161362618, + "loss": 0.505, + "step": 2056 + }, + { + "epoch": 0.11492583177361232, + "grad_norm": 0.830905556678772, + "learning_rate": 0.000945316001792918, + "loss": 0.5779, + "step": 2057 + }, + { + "epoch": 0.11498170237729419, + "grad_norm": 1.0308338403701782, + "learning_rate": 0.0009452879874495742, + "loss": 0.4314, + "step": 2058 + }, + { + "epoch": 0.11503757298097605, + "grad_norm": 0.961164653301239, + "learning_rate": 0.0009452599731062304, + "loss": 0.4822, + "step": 2059 + }, + { + "epoch": 0.11509344358465794, + "grad_norm": 0.8403220176696777, + "learning_rate": 0.0009452319587628866, + "loss": 0.7769, + "step": 2060 + }, + { + "epoch": 0.1151493141883398, + "grad_norm": 0.626973569393158, + "learning_rate": 0.0009452039444195428, + "loss": 0.4334, + "step": 2061 + }, + { + "epoch": 0.11520518479202167, + "grad_norm": 0.8463736176490784, + "learning_rate": 0.000945175930076199, + "loss": 0.4235, + "step": 2062 + }, + { + "epoch": 0.11526105539570355, + "grad_norm": 1.025352120399475, + "learning_rate": 0.0009451479157328552, + "loss": 0.4114, + "step": 2063 + }, + { + "epoch": 0.11531692599938542, + "grad_norm": 0.5526437163352966, + "learning_rate": 0.0009451199013895114, + "loss": 0.3699, + "step": 2064 + }, + { + "epoch": 0.11537279660306729, + "grad_norm": 6.289414882659912, + "learning_rate": 0.0009450918870461676, + "loss": 0.5741, + "step": 2065 + }, + { + "epoch": 0.11542866720674917, + "grad_norm": 1.0387017726898193, + "learning_rate": 0.0009450638727028239, + "loss": 0.5492, + "step": 2066 + }, + { + "epoch": 0.11548453781043104, + "grad_norm": 0.6817046403884888, + "learning_rate": 0.0009450358583594801, + "loss": 0.552, + "step": 2067 + }, + { + "epoch": 0.11554040841411291, + "grad_norm": 0.8701254725456238, + "learning_rate": 0.0009450078440161363, + "loss": 0.5612, + "step": 2068 + }, + { + "epoch": 0.11559627901779479, + "grad_norm": 1.299547791481018, + "learning_rate": 0.0009449798296727925, + "loss": 0.4987, + "step": 2069 + }, + { + "epoch": 0.11565214962147666, + "grad_norm": 1.2439982891082764, + "learning_rate": 0.0009449518153294488, + "loss": 0.4745, + "step": 2070 + }, + { + "epoch": 0.11570802022515854, + "grad_norm": 2.379472017288208, + "learning_rate": 0.0009449238009861049, + "loss": 0.4662, + "step": 2071 + }, + { + "epoch": 0.11576389082884041, + "grad_norm": 1.0365045070648193, + "learning_rate": 0.0009448957866427611, + "loss": 0.5348, + "step": 2072 + }, + { + "epoch": 0.11581976143252228, + "grad_norm": 1.182061791419983, + "learning_rate": 0.0009448677722994173, + "loss": 0.5066, + "step": 2073 + }, + { + "epoch": 0.11587563203620416, + "grad_norm": 0.5962015986442566, + "learning_rate": 0.0009448397579560736, + "loss": 0.544, + "step": 2074 + }, + { + "epoch": 0.11593150263988602, + "grad_norm": 2.0832889080047607, + "learning_rate": 0.0009448117436127297, + "loss": 0.6408, + "step": 2075 + }, + { + "epoch": 0.11598737324356789, + "grad_norm": 0.5351213216781616, + "learning_rate": 0.0009447837292693859, + "loss": 0.5441, + "step": 2076 + }, + { + "epoch": 0.11604324384724977, + "grad_norm": 0.66936856508255, + "learning_rate": 0.0009447557149260422, + "loss": 0.6215, + "step": 2077 + }, + { + "epoch": 0.11609911445093164, + "grad_norm": 0.6291682720184326, + "learning_rate": 0.0009447277005826984, + "loss": 0.5963, + "step": 2078 + }, + { + "epoch": 0.11615498505461351, + "grad_norm": 0.7051401138305664, + "learning_rate": 0.0009446996862393545, + "loss": 0.5599, + "step": 2079 + }, + { + "epoch": 0.11621085565829539, + "grad_norm": 0.5302199125289917, + "learning_rate": 0.0009446716718960107, + "loss": 0.525, + "step": 2080 + }, + { + "epoch": 0.11626672626197726, + "grad_norm": 0.4427800178527832, + "learning_rate": 0.000944643657552667, + "loss": 0.4234, + "step": 2081 + }, + { + "epoch": 0.11632259686565913, + "grad_norm": 0.568162202835083, + "learning_rate": 0.0009446156432093231, + "loss": 0.5596, + "step": 2082 + }, + { + "epoch": 0.11637846746934101, + "grad_norm": 0.5092861652374268, + "learning_rate": 0.0009445876288659794, + "loss": 0.6718, + "step": 2083 + }, + { + "epoch": 0.11643433807302288, + "grad_norm": 0.9525302052497864, + "learning_rate": 0.0009445596145226357, + "loss": 0.4894, + "step": 2084 + }, + { + "epoch": 0.11649020867670475, + "grad_norm": 0.8299450874328613, + "learning_rate": 0.0009445316001792919, + "loss": 0.4871, + "step": 2085 + }, + { + "epoch": 0.11654607928038663, + "grad_norm": 0.7669451236724854, + "learning_rate": 0.000944503585835948, + "loss": 0.5536, + "step": 2086 + }, + { + "epoch": 0.1166019498840685, + "grad_norm": 0.46633195877075195, + "learning_rate": 0.0009444755714926042, + "loss": 0.365, + "step": 2087 + }, + { + "epoch": 0.11665782048775036, + "grad_norm": 0.5965064167976379, + "learning_rate": 0.0009444475571492605, + "loss": 0.453, + "step": 2088 + }, + { + "epoch": 0.11671369109143225, + "grad_norm": 0.7006927132606506, + "learning_rate": 0.0009444195428059167, + "loss": 0.5595, + "step": 2089 + }, + { + "epoch": 0.11676956169511411, + "grad_norm": 0.6959309577941895, + "learning_rate": 0.0009443915284625728, + "loss": 0.4649, + "step": 2090 + }, + { + "epoch": 0.116825432298796, + "grad_norm": 0.7438554763793945, + "learning_rate": 0.000944363514119229, + "loss": 0.6485, + "step": 2091 + }, + { + "epoch": 0.11688130290247786, + "grad_norm": 0.5389924049377441, + "learning_rate": 0.0009443354997758853, + "loss": 0.5414, + "step": 2092 + }, + { + "epoch": 0.11693717350615973, + "grad_norm": 0.5817840099334717, + "learning_rate": 0.0009443074854325415, + "loss": 0.4734, + "step": 2093 + }, + { + "epoch": 0.11699304410984161, + "grad_norm": 1.643972396850586, + "learning_rate": 0.0009442794710891976, + "loss": 0.5486, + "step": 2094 + }, + { + "epoch": 0.11704891471352348, + "grad_norm": 0.5776335597038269, + "learning_rate": 0.0009442514567458539, + "loss": 0.5818, + "step": 2095 + }, + { + "epoch": 0.11710478531720535, + "grad_norm": 0.5946539640426636, + "learning_rate": 0.0009442234424025101, + "loss": 0.5203, + "step": 2096 + }, + { + "epoch": 0.11716065592088723, + "grad_norm": 0.4671609103679657, + "learning_rate": 0.0009441954280591663, + "loss": 0.4818, + "step": 2097 + }, + { + "epoch": 0.1172165265245691, + "grad_norm": 0.6474834084510803, + "learning_rate": 0.0009441674137158224, + "loss": 0.5167, + "step": 2098 + }, + { + "epoch": 0.11727239712825097, + "grad_norm": 1.9330421686172485, + "learning_rate": 0.0009441393993724788, + "loss": 0.5367, + "step": 2099 + }, + { + "epoch": 0.11732826773193285, + "grad_norm": 1.4110362529754639, + "learning_rate": 0.000944111385029135, + "loss": 0.564, + "step": 2100 + }, + { + "epoch": 0.11738413833561472, + "grad_norm": 0.5877069234848022, + "learning_rate": 0.0009440833706857912, + "loss": 0.4898, + "step": 2101 + }, + { + "epoch": 0.11744000893929658, + "grad_norm": 0.5535922646522522, + "learning_rate": 0.0009440553563424474, + "loss": 0.4532, + "step": 2102 + }, + { + "epoch": 0.11749587954297847, + "grad_norm": 0.6080106496810913, + "learning_rate": 0.0009440273419991036, + "loss": 0.4882, + "step": 2103 + }, + { + "epoch": 0.11755175014666033, + "grad_norm": 0.4476223587989807, + "learning_rate": 0.0009439993276557598, + "loss": 0.5153, + "step": 2104 + }, + { + "epoch": 0.1176076207503422, + "grad_norm": 0.6140780448913574, + "learning_rate": 0.0009439713133124159, + "loss": 0.4492, + "step": 2105 + }, + { + "epoch": 0.11766349135402408, + "grad_norm": 0.6752381920814514, + "learning_rate": 0.0009439432989690722, + "loss": 0.5643, + "step": 2106 + }, + { + "epoch": 0.11771936195770595, + "grad_norm": 0.7019727230072021, + "learning_rate": 0.0009439152846257284, + "loss": 0.4755, + "step": 2107 + }, + { + "epoch": 0.11777523256138783, + "grad_norm": 0.5702913999557495, + "learning_rate": 0.0009438872702823846, + "loss": 0.4536, + "step": 2108 + }, + { + "epoch": 0.1178311031650697, + "grad_norm": 0.5470852255821228, + "learning_rate": 0.0009438592559390408, + "loss": 0.4597, + "step": 2109 + }, + { + "epoch": 0.11788697376875157, + "grad_norm": 0.779044508934021, + "learning_rate": 0.000943831241595697, + "loss": 0.3712, + "step": 2110 + }, + { + "epoch": 0.11794284437243345, + "grad_norm": 0.6225262880325317, + "learning_rate": 0.0009438032272523532, + "loss": 0.4239, + "step": 2111 + }, + { + "epoch": 0.11799871497611532, + "grad_norm": 1.1106367111206055, + "learning_rate": 0.0009437752129090094, + "loss": 0.793, + "step": 2112 + }, + { + "epoch": 0.11805458557979719, + "grad_norm": 0.5389660000801086, + "learning_rate": 0.0009437471985656656, + "loss": 0.5417, + "step": 2113 + }, + { + "epoch": 0.11811045618347907, + "grad_norm": 0.503609299659729, + "learning_rate": 0.0009437191842223218, + "loss": 0.4107, + "step": 2114 + }, + { + "epoch": 0.11816632678716094, + "grad_norm": 0.9092045426368713, + "learning_rate": 0.000943691169878978, + "loss": 0.5374, + "step": 2115 + }, + { + "epoch": 0.1182221973908428, + "grad_norm": 0.7034417390823364, + "learning_rate": 0.0009436631555356344, + "loss": 0.5696, + "step": 2116 + }, + { + "epoch": 0.11827806799452469, + "grad_norm": 0.6084075570106506, + "learning_rate": 0.0009436351411922905, + "loss": 0.4512, + "step": 2117 + }, + { + "epoch": 0.11833393859820655, + "grad_norm": 0.5520884394645691, + "learning_rate": 0.0009436071268489467, + "loss": 0.4371, + "step": 2118 + }, + { + "epoch": 0.11838980920188842, + "grad_norm": 0.6036223769187927, + "learning_rate": 0.0009435791125056029, + "loss": 0.5593, + "step": 2119 + }, + { + "epoch": 0.1184456798055703, + "grad_norm": 0.6342809200286865, + "learning_rate": 0.0009435510981622592, + "loss": 0.4232, + "step": 2120 + }, + { + "epoch": 0.11850155040925217, + "grad_norm": 0.7616103887557983, + "learning_rate": 0.0009435230838189153, + "loss": 0.5867, + "step": 2121 + }, + { + "epoch": 0.11855742101293404, + "grad_norm": 5.5267157554626465, + "learning_rate": 0.0009434950694755715, + "loss": 0.4879, + "step": 2122 + }, + { + "epoch": 0.11861329161661592, + "grad_norm": 0.6034570336341858, + "learning_rate": 0.0009434670551322278, + "loss": 0.3758, + "step": 2123 + }, + { + "epoch": 0.11866916222029779, + "grad_norm": 1.0404324531555176, + "learning_rate": 0.0009434390407888839, + "loss": 0.4608, + "step": 2124 + }, + { + "epoch": 0.11872503282397966, + "grad_norm": 1.545246958732605, + "learning_rate": 0.0009434110264455401, + "loss": 0.5257, + "step": 2125 + }, + { + "epoch": 0.11878090342766154, + "grad_norm": 1.0296686887741089, + "learning_rate": 0.0009433830121021963, + "loss": 0.4862, + "step": 2126 + }, + { + "epoch": 0.1188367740313434, + "grad_norm": 0.7403366565704346, + "learning_rate": 0.0009433549977588526, + "loss": 0.6331, + "step": 2127 + }, + { + "epoch": 0.11889264463502529, + "grad_norm": 0.6498391032218933, + "learning_rate": 0.0009433269834155087, + "loss": 0.5547, + "step": 2128 + }, + { + "epoch": 0.11894851523870716, + "grad_norm": 0.48558107018470764, + "learning_rate": 0.0009432989690721649, + "loss": 0.4707, + "step": 2129 + }, + { + "epoch": 0.11900438584238902, + "grad_norm": 1.011030912399292, + "learning_rate": 0.0009432709547288211, + "loss": 0.5571, + "step": 2130 + }, + { + "epoch": 0.1190602564460709, + "grad_norm": 0.7757394909858704, + "learning_rate": 0.0009432429403854774, + "loss": 0.3863, + "step": 2131 + }, + { + "epoch": 0.11911612704975277, + "grad_norm": 0.4734877943992615, + "learning_rate": 0.0009432149260421335, + "loss": 0.4779, + "step": 2132 + }, + { + "epoch": 0.11917199765343464, + "grad_norm": 0.8350332975387573, + "learning_rate": 0.0009431869116987898, + "loss": 0.4804, + "step": 2133 + }, + { + "epoch": 0.11922786825711652, + "grad_norm": 0.5999109745025635, + "learning_rate": 0.0009431588973554461, + "loss": 0.4484, + "step": 2134 + }, + { + "epoch": 0.11928373886079839, + "grad_norm": 0.8907409906387329, + "learning_rate": 0.0009431308830121023, + "loss": 0.6812, + "step": 2135 + }, + { + "epoch": 0.11933960946448026, + "grad_norm": 0.8605990409851074, + "learning_rate": 0.0009431028686687584, + "loss": 0.4652, + "step": 2136 + }, + { + "epoch": 0.11939548006816214, + "grad_norm": 0.6015204787254333, + "learning_rate": 0.0009430748543254146, + "loss": 0.5171, + "step": 2137 + }, + { + "epoch": 0.11945135067184401, + "grad_norm": 3.6220145225524902, + "learning_rate": 0.0009430468399820709, + "loss": 0.4609, + "step": 2138 + }, + { + "epoch": 0.11950722127552588, + "grad_norm": 0.619138777256012, + "learning_rate": 0.0009430188256387271, + "loss": 0.5642, + "step": 2139 + }, + { + "epoch": 0.11956309187920776, + "grad_norm": 0.4321158230304718, + "learning_rate": 0.0009429908112953832, + "loss": 0.4612, + "step": 2140 + }, + { + "epoch": 0.11961896248288963, + "grad_norm": 0.6307247877120972, + "learning_rate": 0.0009429627969520395, + "loss": 0.5437, + "step": 2141 + }, + { + "epoch": 0.1196748330865715, + "grad_norm": 0.6279172897338867, + "learning_rate": 0.0009429347826086957, + "loss": 0.4984, + "step": 2142 + }, + { + "epoch": 0.11973070369025338, + "grad_norm": 0.7359809279441833, + "learning_rate": 0.0009429067682653519, + "loss": 0.4997, + "step": 2143 + }, + { + "epoch": 0.11978657429393524, + "grad_norm": 0.6475677490234375, + "learning_rate": 0.000942878753922008, + "loss": 0.4455, + "step": 2144 + }, + { + "epoch": 0.11984244489761711, + "grad_norm": 0.6707870364189148, + "learning_rate": 0.0009428507395786643, + "loss": 0.4895, + "step": 2145 + }, + { + "epoch": 0.119898315501299, + "grad_norm": 17.45414161682129, + "learning_rate": 0.0009428227252353205, + "loss": 0.4452, + "step": 2146 + }, + { + "epoch": 0.11995418610498086, + "grad_norm": 4.69363260269165, + "learning_rate": 0.0009427947108919766, + "loss": 0.5852, + "step": 2147 + }, + { + "epoch": 0.12001005670866274, + "grad_norm": 0.6800097227096558, + "learning_rate": 0.000942766696548633, + "loss": 0.5625, + "step": 2148 + }, + { + "epoch": 0.12006592731234461, + "grad_norm": 1.599409580230713, + "learning_rate": 0.0009427386822052892, + "loss": 0.4902, + "step": 2149 + }, + { + "epoch": 0.12012179791602648, + "grad_norm": 0.6347957849502563, + "learning_rate": 0.0009427106678619454, + "loss": 0.4745, + "step": 2150 + }, + { + "epoch": 0.12017766851970836, + "grad_norm": 0.7527361512184143, + "learning_rate": 0.0009426826535186015, + "loss": 0.621, + "step": 2151 + }, + { + "epoch": 0.12023353912339023, + "grad_norm": 0.5423722863197327, + "learning_rate": 0.0009426546391752578, + "loss": 0.4971, + "step": 2152 + }, + { + "epoch": 0.1202894097270721, + "grad_norm": 0.5632123947143555, + "learning_rate": 0.000942626624831914, + "loss": 0.525, + "step": 2153 + }, + { + "epoch": 0.12034528033075398, + "grad_norm": 0.542081892490387, + "learning_rate": 0.0009425986104885702, + "loss": 0.4343, + "step": 2154 + }, + { + "epoch": 0.12040115093443585, + "grad_norm": 0.7899245023727417, + "learning_rate": 0.0009425705961452264, + "loss": 0.4889, + "step": 2155 + }, + { + "epoch": 0.12045702153811771, + "grad_norm": 0.511922299861908, + "learning_rate": 0.0009425425818018826, + "loss": 0.4613, + "step": 2156 + }, + { + "epoch": 0.1205128921417996, + "grad_norm": 0.5903447270393372, + "learning_rate": 0.0009425145674585388, + "loss": 0.4832, + "step": 2157 + }, + { + "epoch": 0.12056876274548146, + "grad_norm": 0.6432642936706543, + "learning_rate": 0.000942486553115195, + "loss": 0.5348, + "step": 2158 + }, + { + "epoch": 0.12062463334916333, + "grad_norm": 0.6075130105018616, + "learning_rate": 0.0009424585387718512, + "loss": 0.5174, + "step": 2159 + }, + { + "epoch": 0.12068050395284521, + "grad_norm": 0.8147237300872803, + "learning_rate": 0.0009424305244285074, + "loss": 0.493, + "step": 2160 + }, + { + "epoch": 0.12073637455652708, + "grad_norm": 2.191859722137451, + "learning_rate": 0.0009424025100851636, + "loss": 0.4178, + "step": 2161 + }, + { + "epoch": 0.12079224516020895, + "grad_norm": 0.7643881440162659, + "learning_rate": 0.0009423744957418198, + "loss": 0.521, + "step": 2162 + }, + { + "epoch": 0.12084811576389083, + "grad_norm": 2.313981771469116, + "learning_rate": 0.000942346481398476, + "loss": 0.5469, + "step": 2163 + }, + { + "epoch": 0.1209039863675727, + "grad_norm": 0.7794030904769897, + "learning_rate": 0.0009423184670551322, + "loss": 0.6193, + "step": 2164 + }, + { + "epoch": 0.12095985697125458, + "grad_norm": 1.4916082620620728, + "learning_rate": 0.0009422904527117884, + "loss": 0.4133, + "step": 2165 + }, + { + "epoch": 0.12101572757493645, + "grad_norm": 0.591500461101532, + "learning_rate": 0.0009422624383684448, + "loss": 0.5684, + "step": 2166 + }, + { + "epoch": 0.12107159817861832, + "grad_norm": 0.5895897746086121, + "learning_rate": 0.0009422344240251009, + "loss": 0.501, + "step": 2167 + }, + { + "epoch": 0.1211274687823002, + "grad_norm": 0.7481902241706848, + "learning_rate": 0.0009422064096817571, + "loss": 0.3888, + "step": 2168 + }, + { + "epoch": 0.12118333938598207, + "grad_norm": 0.5855405330657959, + "learning_rate": 0.0009421783953384133, + "loss": 0.4885, + "step": 2169 + }, + { + "epoch": 0.12123920998966393, + "grad_norm": 0.5942724347114563, + "learning_rate": 0.0009421503809950695, + "loss": 0.5082, + "step": 2170 + }, + { + "epoch": 0.12129508059334582, + "grad_norm": 0.7993726134300232, + "learning_rate": 0.0009421223666517257, + "loss": 0.4155, + "step": 2171 + }, + { + "epoch": 0.12135095119702768, + "grad_norm": 0.6160216331481934, + "learning_rate": 0.0009420943523083819, + "loss": 0.5369, + "step": 2172 + }, + { + "epoch": 0.12140682180070955, + "grad_norm": 0.616572380065918, + "learning_rate": 0.0009420663379650382, + "loss": 0.5634, + "step": 2173 + }, + { + "epoch": 0.12146269240439143, + "grad_norm": 0.3965007960796356, + "learning_rate": 0.0009420383236216943, + "loss": 0.385, + "step": 2174 + }, + { + "epoch": 0.1215185630080733, + "grad_norm": 1.1658257246017456, + "learning_rate": 0.0009420103092783505, + "loss": 0.5654, + "step": 2175 + }, + { + "epoch": 0.12157443361175517, + "grad_norm": 0.8873440027236938, + "learning_rate": 0.0009419822949350067, + "loss": 0.5189, + "step": 2176 + }, + { + "epoch": 0.12163030421543705, + "grad_norm": 0.5085958242416382, + "learning_rate": 0.000941954280591663, + "loss": 0.4288, + "step": 2177 + }, + { + "epoch": 0.12168617481911892, + "grad_norm": 0.5420334339141846, + "learning_rate": 0.0009419262662483191, + "loss": 0.4999, + "step": 2178 + }, + { + "epoch": 0.12174204542280079, + "grad_norm": 0.59429532289505, + "learning_rate": 0.0009418982519049753, + "loss": 0.599, + "step": 2179 + }, + { + "epoch": 0.12179791602648267, + "grad_norm": 0.8385865688323975, + "learning_rate": 0.0009418702375616316, + "loss": 0.4893, + "step": 2180 + }, + { + "epoch": 0.12185378663016454, + "grad_norm": 0.6262706518173218, + "learning_rate": 0.0009418422232182878, + "loss": 0.4698, + "step": 2181 + }, + { + "epoch": 0.1219096572338464, + "grad_norm": 0.8843130469322205, + "learning_rate": 0.0009418142088749439, + "loss": 0.5084, + "step": 2182 + }, + { + "epoch": 0.12196552783752829, + "grad_norm": 0.5063118934631348, + "learning_rate": 0.0009417861945316002, + "loss": 0.4666, + "step": 2183 + }, + { + "epoch": 0.12202139844121015, + "grad_norm": 0.8885238766670227, + "learning_rate": 0.0009417581801882565, + "loss": 0.4299, + "step": 2184 + }, + { + "epoch": 0.12207726904489204, + "grad_norm": 0.9437589645385742, + "learning_rate": 0.0009417301658449127, + "loss": 0.5717, + "step": 2185 + }, + { + "epoch": 0.1221331396485739, + "grad_norm": 1.4695696830749512, + "learning_rate": 0.0009417021515015688, + "loss": 0.475, + "step": 2186 + }, + { + "epoch": 0.12218901025225577, + "grad_norm": 0.9108664989471436, + "learning_rate": 0.0009416741371582251, + "loss": 0.5613, + "step": 2187 + }, + { + "epoch": 0.12224488085593765, + "grad_norm": 0.9732390642166138, + "learning_rate": 0.0009416461228148813, + "loss": 0.6524, + "step": 2188 + }, + { + "epoch": 0.12230075145961952, + "grad_norm": 0.5163931846618652, + "learning_rate": 0.0009416181084715374, + "loss": 0.4922, + "step": 2189 + }, + { + "epoch": 0.12235662206330139, + "grad_norm": 2.608489990234375, + "learning_rate": 0.0009415900941281936, + "loss": 0.3734, + "step": 2190 + }, + { + "epoch": 0.12241249266698327, + "grad_norm": 0.4591454565525055, + "learning_rate": 0.0009415620797848499, + "loss": 0.419, + "step": 2191 + }, + { + "epoch": 0.12246836327066514, + "grad_norm": 0.74837726354599, + "learning_rate": 0.0009415340654415061, + "loss": 0.516, + "step": 2192 + }, + { + "epoch": 0.12252423387434701, + "grad_norm": 0.6531306505203247, + "learning_rate": 0.0009415060510981622, + "loss": 0.4973, + "step": 2193 + }, + { + "epoch": 0.12258010447802889, + "grad_norm": 0.6258606314659119, + "learning_rate": 0.0009414780367548184, + "loss": 0.5864, + "step": 2194 + }, + { + "epoch": 0.12263597508171076, + "grad_norm": 0.454781711101532, + "learning_rate": 0.0009414500224114747, + "loss": 0.444, + "step": 2195 + }, + { + "epoch": 0.12269184568539263, + "grad_norm": 0.5555285215377808, + "learning_rate": 0.0009414220080681309, + "loss": 0.4093, + "step": 2196 + }, + { + "epoch": 0.12274771628907451, + "grad_norm": 0.5018996000289917, + "learning_rate": 0.000941393993724787, + "loss": 0.441, + "step": 2197 + }, + { + "epoch": 0.12280358689275638, + "grad_norm": 0.5245765447616577, + "learning_rate": 0.0009413659793814433, + "loss": 0.4683, + "step": 2198 + }, + { + "epoch": 0.12285945749643824, + "grad_norm": 2.6002445220947266, + "learning_rate": 0.0009413379650380996, + "loss": 0.5076, + "step": 2199 + }, + { + "epoch": 0.12291532810012012, + "grad_norm": 0.4664488136768341, + "learning_rate": 0.0009413099506947558, + "loss": 0.3791, + "step": 2200 + }, + { + "epoch": 0.12297119870380199, + "grad_norm": 2.2923786640167236, + "learning_rate": 0.0009412819363514119, + "loss": 0.4969, + "step": 2201 + }, + { + "epoch": 0.12302706930748386, + "grad_norm": 0.7816728353500366, + "learning_rate": 0.0009412539220080682, + "loss": 0.5199, + "step": 2202 + }, + { + "epoch": 0.12308293991116574, + "grad_norm": 1.3513637781143188, + "learning_rate": 0.0009412259076647244, + "loss": 0.5362, + "step": 2203 + }, + { + "epoch": 0.12313881051484761, + "grad_norm": 0.8744179010391235, + "learning_rate": 0.0009411978933213806, + "loss": 0.4832, + "step": 2204 + }, + { + "epoch": 0.12319468111852949, + "grad_norm": 0.8423755764961243, + "learning_rate": 0.0009411698789780368, + "loss": 0.5231, + "step": 2205 + }, + { + "epoch": 0.12325055172221136, + "grad_norm": 0.5851179957389832, + "learning_rate": 0.000941141864634693, + "loss": 0.4356, + "step": 2206 + }, + { + "epoch": 0.12330642232589323, + "grad_norm": 0.5133440494537354, + "learning_rate": 0.0009411138502913492, + "loss": 0.5393, + "step": 2207 + }, + { + "epoch": 0.12336229292957511, + "grad_norm": 0.6359091997146606, + "learning_rate": 0.0009410858359480054, + "loss": 0.4107, + "step": 2208 + }, + { + "epoch": 0.12341816353325698, + "grad_norm": 0.67991703748703, + "learning_rate": 0.0009410578216046616, + "loss": 0.7375, + "step": 2209 + }, + { + "epoch": 0.12347403413693885, + "grad_norm": 0.5096514821052551, + "learning_rate": 0.0009410298072613178, + "loss": 0.5468, + "step": 2210 + }, + { + "epoch": 0.12352990474062073, + "grad_norm": 1.1547318696975708, + "learning_rate": 0.000941001792917974, + "loss": 0.5737, + "step": 2211 + }, + { + "epoch": 0.1235857753443026, + "grad_norm": 0.6450685262680054, + "learning_rate": 0.0009409737785746302, + "loss": 0.4134, + "step": 2212 + }, + { + "epoch": 0.12364164594798446, + "grad_norm": 0.5509928464889526, + "learning_rate": 0.0009409457642312864, + "loss": 0.5046, + "step": 2213 + }, + { + "epoch": 0.12369751655166634, + "grad_norm": 0.47758427262306213, + "learning_rate": 0.0009409177498879426, + "loss": 0.4084, + "step": 2214 + }, + { + "epoch": 0.12375338715534821, + "grad_norm": 0.6636220812797546, + "learning_rate": 0.0009408897355445988, + "loss": 0.4567, + "step": 2215 + }, + { + "epoch": 0.12380925775903008, + "grad_norm": 1.002748966217041, + "learning_rate": 0.000940861721201255, + "loss": 0.5207, + "step": 2216 + }, + { + "epoch": 0.12386512836271196, + "grad_norm": 0.6602751612663269, + "learning_rate": 0.0009408337068579113, + "loss": 0.4858, + "step": 2217 + }, + { + "epoch": 0.12392099896639383, + "grad_norm": 0.5695028901100159, + "learning_rate": 0.0009408056925145675, + "loss": 0.5222, + "step": 2218 + }, + { + "epoch": 0.1239768695700757, + "grad_norm": 0.8996002078056335, + "learning_rate": 0.0009407776781712238, + "loss": 0.5894, + "step": 2219 + }, + { + "epoch": 0.12403274017375758, + "grad_norm": 0.8456904888153076, + "learning_rate": 0.0009407496638278799, + "loss": 0.7115, + "step": 2220 + }, + { + "epoch": 0.12408861077743945, + "grad_norm": 1.1676087379455566, + "learning_rate": 0.0009407216494845361, + "loss": 0.513, + "step": 2221 + }, + { + "epoch": 0.12414448138112133, + "grad_norm": 1.076745867729187, + "learning_rate": 0.0009406936351411923, + "loss": 0.4987, + "step": 2222 + }, + { + "epoch": 0.1242003519848032, + "grad_norm": 1.3907880783081055, + "learning_rate": 0.0009406656207978486, + "loss": 0.5384, + "step": 2223 + }, + { + "epoch": 0.12425622258848507, + "grad_norm": 0.4558967053890228, + "learning_rate": 0.0009406376064545047, + "loss": 0.5124, + "step": 2224 + }, + { + "epoch": 0.12431209319216695, + "grad_norm": 0.6085279583930969, + "learning_rate": 0.0009406095921111609, + "loss": 0.4347, + "step": 2225 + }, + { + "epoch": 0.12436796379584882, + "grad_norm": 0.5871221423149109, + "learning_rate": 0.0009405815777678172, + "loss": 0.6111, + "step": 2226 + }, + { + "epoch": 0.12442383439953068, + "grad_norm": 0.613803505897522, + "learning_rate": 0.0009405535634244734, + "loss": 0.5354, + "step": 2227 + }, + { + "epoch": 0.12447970500321257, + "grad_norm": 0.7752054333686829, + "learning_rate": 0.0009405255490811295, + "loss": 0.467, + "step": 2228 + }, + { + "epoch": 0.12453557560689443, + "grad_norm": 0.6396591067314148, + "learning_rate": 0.0009404975347377857, + "loss": 0.5107, + "step": 2229 + }, + { + "epoch": 0.1245914462105763, + "grad_norm": 0.6726458072662354, + "learning_rate": 0.000940469520394442, + "loss": 0.6119, + "step": 2230 + }, + { + "epoch": 0.12464731681425818, + "grad_norm": 0.6210963129997253, + "learning_rate": 0.0009404415060510981, + "loss": 0.399, + "step": 2231 + }, + { + "epoch": 0.12470318741794005, + "grad_norm": 0.6717049479484558, + "learning_rate": 0.0009404134917077543, + "loss": 0.542, + "step": 2232 + }, + { + "epoch": 0.12475905802162192, + "grad_norm": 0.7349048256874084, + "learning_rate": 0.0009403854773644105, + "loss": 0.5717, + "step": 2233 + }, + { + "epoch": 0.1248149286253038, + "grad_norm": 0.7059723138809204, + "learning_rate": 0.0009403574630210669, + "loss": 0.4614, + "step": 2234 + }, + { + "epoch": 0.12487079922898567, + "grad_norm": 0.7385600209236145, + "learning_rate": 0.000940329448677723, + "loss": 0.4155, + "step": 2235 + }, + { + "epoch": 0.12492666983266754, + "grad_norm": 1.6761624813079834, + "learning_rate": 0.0009403014343343792, + "loss": 0.5566, + "step": 2236 + }, + { + "epoch": 0.12498254043634942, + "grad_norm": 0.7138071656227112, + "learning_rate": 0.0009402734199910355, + "loss": 0.5157, + "step": 2237 + }, + { + "epoch": 0.1250384110400313, + "grad_norm": 1.2630904912948608, + "learning_rate": 0.0009402454056476917, + "loss": 0.5155, + "step": 2238 + }, + { + "epoch": 0.12509428164371317, + "grad_norm": 1.0752350091934204, + "learning_rate": 0.0009402173913043478, + "loss": 0.5884, + "step": 2239 + }, + { + "epoch": 0.12515015224739504, + "grad_norm": 0.5227307081222534, + "learning_rate": 0.000940189376961004, + "loss": 0.5006, + "step": 2240 + }, + { + "epoch": 0.1252060228510769, + "grad_norm": 0.9550474882125854, + "learning_rate": 0.0009401613626176603, + "loss": 0.4995, + "step": 2241 + }, + { + "epoch": 0.12526189345475877, + "grad_norm": 0.6650369167327881, + "learning_rate": 0.0009401333482743165, + "loss": 0.5182, + "step": 2242 + }, + { + "epoch": 0.12531776405844064, + "grad_norm": 0.6472020745277405, + "learning_rate": 0.0009401053339309726, + "loss": 0.5637, + "step": 2243 + }, + { + "epoch": 0.12537363466212254, + "grad_norm": 1.0955619812011719, + "learning_rate": 0.0009400773195876289, + "loss": 0.5202, + "step": 2244 + }, + { + "epoch": 0.1254295052658044, + "grad_norm": 0.6509215831756592, + "learning_rate": 0.0009400493052442851, + "loss": 0.4556, + "step": 2245 + }, + { + "epoch": 0.12548537586948627, + "grad_norm": 1.4495781660079956, + "learning_rate": 0.0009400212909009413, + "loss": 0.539, + "step": 2246 + }, + { + "epoch": 0.12554124647316814, + "grad_norm": 0.7062668800354004, + "learning_rate": 0.0009399932765575974, + "loss": 0.6374, + "step": 2247 + }, + { + "epoch": 0.12559711707685, + "grad_norm": 3.9377245903015137, + "learning_rate": 0.0009399652622142537, + "loss": 0.3953, + "step": 2248 + }, + { + "epoch": 0.12565298768053187, + "grad_norm": 0.7876891493797302, + "learning_rate": 0.00093993724787091, + "loss": 0.7186, + "step": 2249 + }, + { + "epoch": 0.12570885828421377, + "grad_norm": 0.5549075603485107, + "learning_rate": 0.0009399092335275662, + "loss": 0.5215, + "step": 2250 + }, + { + "epoch": 0.12576472888789564, + "grad_norm": 0.4843216836452484, + "learning_rate": 0.0009398812191842224, + "loss": 0.4969, + "step": 2251 + }, + { + "epoch": 0.1258205994915775, + "grad_norm": 0.6186768412590027, + "learning_rate": 0.0009398532048408786, + "loss": 0.3518, + "step": 2252 + }, + { + "epoch": 0.12587647009525937, + "grad_norm": 3.8802034854888916, + "learning_rate": 0.0009398251904975348, + "loss": 0.5356, + "step": 2253 + }, + { + "epoch": 0.12593234069894124, + "grad_norm": 1.8890042304992676, + "learning_rate": 0.0009397971761541909, + "loss": 0.5715, + "step": 2254 + }, + { + "epoch": 0.12598821130262314, + "grad_norm": 0.6475428342819214, + "learning_rate": 0.0009397691618108472, + "loss": 0.5952, + "step": 2255 + }, + { + "epoch": 0.126044081906305, + "grad_norm": 3.9783735275268555, + "learning_rate": 0.0009397411474675034, + "loss": 0.609, + "step": 2256 + }, + { + "epoch": 0.12609995250998687, + "grad_norm": 0.6721128225326538, + "learning_rate": 0.0009397131331241596, + "loss": 0.4762, + "step": 2257 + }, + { + "epoch": 0.12615582311366874, + "grad_norm": 0.5434451699256897, + "learning_rate": 0.0009396851187808158, + "loss": 0.469, + "step": 2258 + }, + { + "epoch": 0.1262116937173506, + "grad_norm": 0.6085752248764038, + "learning_rate": 0.000939657104437472, + "loss": 0.4786, + "step": 2259 + }, + { + "epoch": 0.12626756432103248, + "grad_norm": 0.7501881122589111, + "learning_rate": 0.0009396290900941282, + "loss": 0.5964, + "step": 2260 + }, + { + "epoch": 0.12632343492471437, + "grad_norm": 0.5339687466621399, + "learning_rate": 0.0009396010757507844, + "loss": 0.5631, + "step": 2261 + }, + { + "epoch": 0.12637930552839624, + "grad_norm": 0.5395646691322327, + "learning_rate": 0.0009395730614074406, + "loss": 0.454, + "step": 2262 + }, + { + "epoch": 0.1264351761320781, + "grad_norm": 0.8476229310035706, + "learning_rate": 0.0009395450470640968, + "loss": 0.4392, + "step": 2263 + }, + { + "epoch": 0.12649104673575998, + "grad_norm": 0.7199327945709229, + "learning_rate": 0.000939517032720753, + "loss": 0.6044, + "step": 2264 + }, + { + "epoch": 0.12654691733944184, + "grad_norm": 0.6828478574752808, + "learning_rate": 0.0009394890183774094, + "loss": 0.4849, + "step": 2265 + }, + { + "epoch": 0.1266027879431237, + "grad_norm": 0.5172474384307861, + "learning_rate": 0.0009394610040340654, + "loss": 0.4906, + "step": 2266 + }, + { + "epoch": 0.1266586585468056, + "grad_norm": 0.9520291686058044, + "learning_rate": 0.0009394329896907217, + "loss": 0.5084, + "step": 2267 + }, + { + "epoch": 0.12671452915048748, + "grad_norm": 12.211310386657715, + "learning_rate": 0.0009394049753473779, + "loss": 0.4408, + "step": 2268 + }, + { + "epoch": 0.12677039975416934, + "grad_norm": 0.4198701083660126, + "learning_rate": 0.0009393769610040342, + "loss": 0.5085, + "step": 2269 + }, + { + "epoch": 0.1268262703578512, + "grad_norm": 0.6523088216781616, + "learning_rate": 0.0009393489466606903, + "loss": 0.5739, + "step": 2270 + }, + { + "epoch": 0.12688214096153308, + "grad_norm": 6.618338584899902, + "learning_rate": 0.0009393209323173465, + "loss": 0.5091, + "step": 2271 + }, + { + "epoch": 0.12693801156521498, + "grad_norm": 0.5739362835884094, + "learning_rate": 0.0009392929179740027, + "loss": 0.4933, + "step": 2272 + }, + { + "epoch": 0.12699388216889684, + "grad_norm": 0.6995110511779785, + "learning_rate": 0.0009392649036306589, + "loss": 0.6443, + "step": 2273 + }, + { + "epoch": 0.1270497527725787, + "grad_norm": 0.8331899642944336, + "learning_rate": 0.0009392368892873151, + "loss": 0.7766, + "step": 2274 + }, + { + "epoch": 0.12710562337626058, + "grad_norm": 0.5776708722114563, + "learning_rate": 0.0009392088749439713, + "loss": 0.3913, + "step": 2275 + }, + { + "epoch": 0.12716149397994245, + "grad_norm": 1.0144821405410767, + "learning_rate": 0.0009391808606006276, + "loss": 0.5063, + "step": 2276 + }, + { + "epoch": 0.12721736458362431, + "grad_norm": 2.541971206665039, + "learning_rate": 0.0009391528462572837, + "loss": 0.5166, + "step": 2277 + }, + { + "epoch": 0.1272732351873062, + "grad_norm": 0.6198235750198364, + "learning_rate": 0.0009391248319139399, + "loss": 0.6059, + "step": 2278 + }, + { + "epoch": 0.12732910579098808, + "grad_norm": 0.562153160572052, + "learning_rate": 0.0009390968175705961, + "loss": 0.5116, + "step": 2279 + }, + { + "epoch": 0.12738497639466995, + "grad_norm": 0.4938313663005829, + "learning_rate": 0.0009390688032272524, + "loss": 0.4272, + "step": 2280 + }, + { + "epoch": 0.12744084699835181, + "grad_norm": 0.7928466796875, + "learning_rate": 0.0009390407888839085, + "loss": 0.6112, + "step": 2281 + }, + { + "epoch": 0.12749671760203368, + "grad_norm": 0.6822344660758972, + "learning_rate": 0.0009390127745405647, + "loss": 0.4825, + "step": 2282 + }, + { + "epoch": 0.12755258820571555, + "grad_norm": 0.8719252943992615, + "learning_rate": 0.0009389847601972211, + "loss": 0.4947, + "step": 2283 + }, + { + "epoch": 0.12760845880939745, + "grad_norm": 0.8298627734184265, + "learning_rate": 0.0009389567458538773, + "loss": 0.5959, + "step": 2284 + }, + { + "epoch": 0.1276643294130793, + "grad_norm": 0.777157723903656, + "learning_rate": 0.0009389287315105334, + "loss": 0.4311, + "step": 2285 + }, + { + "epoch": 0.12772020001676118, + "grad_norm": 1.577745795249939, + "learning_rate": 0.0009389007171671896, + "loss": 0.6344, + "step": 2286 + }, + { + "epoch": 0.12777607062044305, + "grad_norm": 0.7122716903686523, + "learning_rate": 0.0009388727028238459, + "loss": 0.4092, + "step": 2287 + }, + { + "epoch": 0.12783194122412492, + "grad_norm": 0.7324351072311401, + "learning_rate": 0.0009388446884805021, + "loss": 0.4721, + "step": 2288 + }, + { + "epoch": 0.1278878118278068, + "grad_norm": 0.6810046434402466, + "learning_rate": 0.0009388166741371582, + "loss": 0.4489, + "step": 2289 + }, + { + "epoch": 0.12794368243148868, + "grad_norm": 7.6004252433776855, + "learning_rate": 0.0009387886597938145, + "loss": 0.5087, + "step": 2290 + }, + { + "epoch": 0.12799955303517055, + "grad_norm": 0.7739803791046143, + "learning_rate": 0.0009387606454504707, + "loss": 0.6177, + "step": 2291 + }, + { + "epoch": 0.12805542363885242, + "grad_norm": 0.6685127019882202, + "learning_rate": 0.0009387326311071269, + "loss": 0.4717, + "step": 2292 + }, + { + "epoch": 0.12811129424253428, + "grad_norm": 6.477637767791748, + "learning_rate": 0.000938704616763783, + "loss": 0.5412, + "step": 2293 + }, + { + "epoch": 0.12816716484621615, + "grad_norm": 0.8689881563186646, + "learning_rate": 0.0009386766024204393, + "loss": 0.5753, + "step": 2294 + }, + { + "epoch": 0.12822303544989805, + "grad_norm": 2.160341739654541, + "learning_rate": 0.0009386485880770955, + "loss": 0.5041, + "step": 2295 + }, + { + "epoch": 0.12827890605357992, + "grad_norm": 0.9332179427146912, + "learning_rate": 0.0009386205737337516, + "loss": 0.496, + "step": 2296 + }, + { + "epoch": 0.12833477665726178, + "grad_norm": 0.9899423718452454, + "learning_rate": 0.0009385925593904079, + "loss": 0.4679, + "step": 2297 + }, + { + "epoch": 0.12839064726094365, + "grad_norm": 1.820859432220459, + "learning_rate": 0.0009385645450470641, + "loss": 0.462, + "step": 2298 + }, + { + "epoch": 0.12844651786462552, + "grad_norm": 0.7449340224266052, + "learning_rate": 0.0009385365307037204, + "loss": 0.4764, + "step": 2299 + }, + { + "epoch": 0.1285023884683074, + "grad_norm": 0.7752236127853394, + "learning_rate": 0.0009385085163603764, + "loss": 0.497, + "step": 2300 + }, + { + "epoch": 0.12855825907198928, + "grad_norm": 0.8111915588378906, + "learning_rate": 0.0009384805020170328, + "loss": 0.5314, + "step": 2301 + }, + { + "epoch": 0.12861412967567115, + "grad_norm": 0.5429022312164307, + "learning_rate": 0.000938452487673689, + "loss": 0.4929, + "step": 2302 + }, + { + "epoch": 0.12867000027935302, + "grad_norm": 0.6009039282798767, + "learning_rate": 0.0009384244733303452, + "loss": 0.5945, + "step": 2303 + }, + { + "epoch": 0.1287258708830349, + "grad_norm": 0.6619232892990112, + "learning_rate": 0.0009383964589870013, + "loss": 0.5391, + "step": 2304 + }, + { + "epoch": 0.12878174148671676, + "grad_norm": 0.6577353477478027, + "learning_rate": 0.0009383684446436576, + "loss": 0.6022, + "step": 2305 + }, + { + "epoch": 0.12883761209039862, + "grad_norm": 0.5239747762680054, + "learning_rate": 0.0009383404303003138, + "loss": 0.5059, + "step": 2306 + }, + { + "epoch": 0.12889348269408052, + "grad_norm": 0.5149891376495361, + "learning_rate": 0.00093831241595697, + "loss": 0.4667, + "step": 2307 + }, + { + "epoch": 0.1289493532977624, + "grad_norm": 0.503126323223114, + "learning_rate": 0.0009382844016136262, + "loss": 0.514, + "step": 2308 + }, + { + "epoch": 0.12900522390144425, + "grad_norm": 0.5035586357116699, + "learning_rate": 0.0009382563872702824, + "loss": 0.4438, + "step": 2309 + }, + { + "epoch": 0.12906109450512612, + "grad_norm": 0.6301321387290955, + "learning_rate": 0.0009382283729269386, + "loss": 0.5225, + "step": 2310 + }, + { + "epoch": 0.129116965108808, + "grad_norm": 1.2934800386428833, + "learning_rate": 0.0009382003585835948, + "loss": 0.3918, + "step": 2311 + }, + { + "epoch": 0.1291728357124899, + "grad_norm": 5.1211323738098145, + "learning_rate": 0.000938172344240251, + "loss": 0.4743, + "step": 2312 + }, + { + "epoch": 0.12922870631617175, + "grad_norm": 0.47122734785079956, + "learning_rate": 0.0009381443298969072, + "loss": 0.4452, + "step": 2313 + }, + { + "epoch": 0.12928457691985362, + "grad_norm": 0.6090487241744995, + "learning_rate": 0.0009381163155535634, + "loss": 0.4249, + "step": 2314 + }, + { + "epoch": 0.1293404475235355, + "grad_norm": 0.6140914559364319, + "learning_rate": 0.0009380883012102196, + "loss": 0.4734, + "step": 2315 + }, + { + "epoch": 0.12939631812721736, + "grad_norm": 0.560208261013031, + "learning_rate": 0.0009380602868668758, + "loss": 0.6755, + "step": 2316 + }, + { + "epoch": 0.12945218873089923, + "grad_norm": 0.548759400844574, + "learning_rate": 0.0009380322725235321, + "loss": 0.4876, + "step": 2317 + }, + { + "epoch": 0.12950805933458112, + "grad_norm": 1.3840906620025635, + "learning_rate": 0.0009380042581801883, + "loss": 0.5911, + "step": 2318 + }, + { + "epoch": 0.129563929938263, + "grad_norm": 0.47081857919692993, + "learning_rate": 0.0009379762438368445, + "loss": 0.5278, + "step": 2319 + }, + { + "epoch": 0.12961980054194486, + "grad_norm": 0.7722494006156921, + "learning_rate": 0.0009379482294935007, + "loss": 0.4519, + "step": 2320 + }, + { + "epoch": 0.12967567114562673, + "grad_norm": 0.5467429757118225, + "learning_rate": 0.0009379202151501569, + "loss": 0.5228, + "step": 2321 + }, + { + "epoch": 0.1297315417493086, + "grad_norm": 0.5769525170326233, + "learning_rate": 0.0009378922008068132, + "loss": 0.5172, + "step": 2322 + }, + { + "epoch": 0.12978741235299046, + "grad_norm": 1.8228737115859985, + "learning_rate": 0.0009378641864634693, + "loss": 0.7726, + "step": 2323 + }, + { + "epoch": 0.12984328295667236, + "grad_norm": 0.5499016046524048, + "learning_rate": 0.0009378361721201255, + "loss": 0.5098, + "step": 2324 + }, + { + "epoch": 0.12989915356035422, + "grad_norm": 5.43359899520874, + "learning_rate": 0.0009378081577767817, + "loss": 0.4895, + "step": 2325 + }, + { + "epoch": 0.1299550241640361, + "grad_norm": 0.8051378726959229, + "learning_rate": 0.000937780143433438, + "loss": 0.6973, + "step": 2326 + }, + { + "epoch": 0.13001089476771796, + "grad_norm": 5.88642692565918, + "learning_rate": 0.0009377521290900941, + "loss": 0.5055, + "step": 2327 + }, + { + "epoch": 0.13006676537139983, + "grad_norm": 0.5498276948928833, + "learning_rate": 0.0009377241147467503, + "loss": 0.5082, + "step": 2328 + }, + { + "epoch": 0.13012263597508172, + "grad_norm": 0.580733060836792, + "learning_rate": 0.0009376961004034066, + "loss": 0.4925, + "step": 2329 + }, + { + "epoch": 0.1301785065787636, + "grad_norm": 0.767695426940918, + "learning_rate": 0.0009376680860600628, + "loss": 0.4908, + "step": 2330 + }, + { + "epoch": 0.13023437718244546, + "grad_norm": 0.824196994304657, + "learning_rate": 0.0009376400717167189, + "loss": 0.4251, + "step": 2331 + }, + { + "epoch": 0.13029024778612733, + "grad_norm": 0.6772290468215942, + "learning_rate": 0.0009376120573733751, + "loss": 0.5702, + "step": 2332 + }, + { + "epoch": 0.1303461183898092, + "grad_norm": 0.7272119522094727, + "learning_rate": 0.0009375840430300315, + "loss": 0.4612, + "step": 2333 + }, + { + "epoch": 0.13040198899349106, + "grad_norm": 0.7801129221916199, + "learning_rate": 0.0009375560286866877, + "loss": 0.4989, + "step": 2334 + }, + { + "epoch": 0.13045785959717296, + "grad_norm": 0.7754640579223633, + "learning_rate": 0.0009375280143433438, + "loss": 0.6673, + "step": 2335 + }, + { + "epoch": 0.13051373020085483, + "grad_norm": 1.771153450012207, + "learning_rate": 0.0009375, + "loss": 0.4896, + "step": 2336 + }, + { + "epoch": 0.1305696008045367, + "grad_norm": 0.6792982816696167, + "learning_rate": 0.0009374719856566563, + "loss": 0.4427, + "step": 2337 + }, + { + "epoch": 0.13062547140821856, + "grad_norm": 0.7984562516212463, + "learning_rate": 0.0009374439713133124, + "loss": 0.6544, + "step": 2338 + }, + { + "epoch": 0.13068134201190043, + "grad_norm": 3.379286766052246, + "learning_rate": 0.0009374159569699686, + "loss": 0.5219, + "step": 2339 + }, + { + "epoch": 0.1307372126155823, + "grad_norm": 0.6189900040626526, + "learning_rate": 0.0009373879426266249, + "loss": 0.8706, + "step": 2340 + }, + { + "epoch": 0.1307930832192642, + "grad_norm": 2.5658459663391113, + "learning_rate": 0.0009373599282832811, + "loss": 0.5184, + "step": 2341 + }, + { + "epoch": 0.13084895382294606, + "grad_norm": 0.43486592173576355, + "learning_rate": 0.0009373319139399372, + "loss": 0.4032, + "step": 2342 + }, + { + "epoch": 0.13090482442662793, + "grad_norm": 0.8091822862625122, + "learning_rate": 0.0009373038995965934, + "loss": 0.5097, + "step": 2343 + }, + { + "epoch": 0.1309606950303098, + "grad_norm": 0.5604686141014099, + "learning_rate": 0.0009372758852532497, + "loss": 0.5143, + "step": 2344 + }, + { + "epoch": 0.13101656563399167, + "grad_norm": 1.225691318511963, + "learning_rate": 0.0009372478709099059, + "loss": 0.4694, + "step": 2345 + }, + { + "epoch": 0.13107243623767356, + "grad_norm": 0.6802666783332825, + "learning_rate": 0.000937219856566562, + "loss": 0.4366, + "step": 2346 + }, + { + "epoch": 0.13112830684135543, + "grad_norm": 0.5615556836128235, + "learning_rate": 0.0009371918422232183, + "loss": 0.4779, + "step": 2347 + }, + { + "epoch": 0.1311841774450373, + "grad_norm": 0.5355339050292969, + "learning_rate": 0.0009371638278798745, + "loss": 0.4906, + "step": 2348 + }, + { + "epoch": 0.13124004804871917, + "grad_norm": 0.653611421585083, + "learning_rate": 0.0009371358135365307, + "loss": 0.4288, + "step": 2349 + }, + { + "epoch": 0.13129591865240103, + "grad_norm": 0.5288708806037903, + "learning_rate": 0.0009371077991931868, + "loss": 0.4596, + "step": 2350 + }, + { + "epoch": 0.1313517892560829, + "grad_norm": 0.5487123727798462, + "learning_rate": 0.0009370797848498432, + "loss": 0.5085, + "step": 2351 + }, + { + "epoch": 0.1314076598597648, + "grad_norm": 0.7549780607223511, + "learning_rate": 0.0009370517705064994, + "loss": 0.531, + "step": 2352 + }, + { + "epoch": 0.13146353046344667, + "grad_norm": 0.7310227751731873, + "learning_rate": 0.0009370237561631556, + "loss": 0.4782, + "step": 2353 + }, + { + "epoch": 0.13151940106712853, + "grad_norm": 1.8481508493423462, + "learning_rate": 0.0009369957418198118, + "loss": 0.4987, + "step": 2354 + }, + { + "epoch": 0.1315752716708104, + "grad_norm": 0.6918179988861084, + "learning_rate": 0.000936967727476468, + "loss": 0.5439, + "step": 2355 + }, + { + "epoch": 0.13163114227449227, + "grad_norm": 0.512470543384552, + "learning_rate": 0.0009369397131331242, + "loss": 0.4778, + "step": 2356 + }, + { + "epoch": 0.13168701287817414, + "grad_norm": 0.7718348503112793, + "learning_rate": 0.0009369116987897804, + "loss": 0.5013, + "step": 2357 + }, + { + "epoch": 0.13174288348185603, + "grad_norm": 0.7974266409873962, + "learning_rate": 0.0009368836844464366, + "loss": 0.4584, + "step": 2358 + }, + { + "epoch": 0.1317987540855379, + "grad_norm": 0.5043377876281738, + "learning_rate": 0.0009368556701030928, + "loss": 0.4397, + "step": 2359 + }, + { + "epoch": 0.13185462468921977, + "grad_norm": 1.0162396430969238, + "learning_rate": 0.000936827655759749, + "loss": 0.3751, + "step": 2360 + }, + { + "epoch": 0.13191049529290164, + "grad_norm": 1.0201570987701416, + "learning_rate": 0.0009367996414164052, + "loss": 0.5135, + "step": 2361 + }, + { + "epoch": 0.1319663658965835, + "grad_norm": 0.828855574131012, + "learning_rate": 0.0009367716270730614, + "loss": 0.5987, + "step": 2362 + }, + { + "epoch": 0.13202223650026537, + "grad_norm": 0.5529502034187317, + "learning_rate": 0.0009367436127297176, + "loss": 0.4284, + "step": 2363 + }, + { + "epoch": 0.13207810710394727, + "grad_norm": 0.6278518438339233, + "learning_rate": 0.0009367155983863738, + "loss": 0.452, + "step": 2364 + }, + { + "epoch": 0.13213397770762914, + "grad_norm": 0.4900380074977875, + "learning_rate": 0.00093668758404303, + "loss": 0.4496, + "step": 2365 + }, + { + "epoch": 0.132189848311311, + "grad_norm": 2.770153284072876, + "learning_rate": 0.0009366595696996862, + "loss": 0.5474, + "step": 2366 + }, + { + "epoch": 0.13224571891499287, + "grad_norm": 0.5616536140441895, + "learning_rate": 0.0009366315553563425, + "loss": 0.4874, + "step": 2367 + }, + { + "epoch": 0.13230158951867474, + "grad_norm": 0.4610631465911865, + "learning_rate": 0.0009366035410129988, + "loss": 0.4944, + "step": 2368 + }, + { + "epoch": 0.13235746012235663, + "grad_norm": 0.8243130445480347, + "learning_rate": 0.0009365755266696549, + "loss": 0.4782, + "step": 2369 + }, + { + "epoch": 0.1324133307260385, + "grad_norm": 0.6543483734130859, + "learning_rate": 0.0009365475123263111, + "loss": 0.4235, + "step": 2370 + }, + { + "epoch": 0.13246920132972037, + "grad_norm": 0.5078830718994141, + "learning_rate": 0.0009365194979829673, + "loss": 0.5492, + "step": 2371 + }, + { + "epoch": 0.13252507193340224, + "grad_norm": 0.4537014067173004, + "learning_rate": 0.0009364914836396236, + "loss": 0.4551, + "step": 2372 + }, + { + "epoch": 0.1325809425370841, + "grad_norm": 0.6426098942756653, + "learning_rate": 0.0009364634692962797, + "loss": 0.4757, + "step": 2373 + }, + { + "epoch": 0.13263681314076597, + "grad_norm": 0.6287769079208374, + "learning_rate": 0.0009364354549529359, + "loss": 0.4483, + "step": 2374 + }, + { + "epoch": 0.13269268374444787, + "grad_norm": 1.0234981775283813, + "learning_rate": 0.0009364074406095921, + "loss": 0.4639, + "step": 2375 + }, + { + "epoch": 0.13274855434812974, + "grad_norm": 0.9784273505210876, + "learning_rate": 0.0009363794262662484, + "loss": 0.5852, + "step": 2376 + }, + { + "epoch": 0.1328044249518116, + "grad_norm": 0.5492579340934753, + "learning_rate": 0.0009363514119229045, + "loss": 0.4289, + "step": 2377 + }, + { + "epoch": 0.13286029555549347, + "grad_norm": 0.5434505343437195, + "learning_rate": 0.0009363233975795607, + "loss": 0.478, + "step": 2378 + }, + { + "epoch": 0.13291616615917534, + "grad_norm": 0.7076989412307739, + "learning_rate": 0.000936295383236217, + "loss": 0.629, + "step": 2379 + }, + { + "epoch": 0.1329720367628572, + "grad_norm": 0.6039761304855347, + "learning_rate": 0.0009362673688928731, + "loss": 0.5247, + "step": 2380 + }, + { + "epoch": 0.1330279073665391, + "grad_norm": 0.901278555393219, + "learning_rate": 0.0009362393545495293, + "loss": 0.5762, + "step": 2381 + }, + { + "epoch": 0.13308377797022097, + "grad_norm": 2.2320969104766846, + "learning_rate": 0.0009362113402061855, + "loss": 0.4778, + "step": 2382 + }, + { + "epoch": 0.13313964857390284, + "grad_norm": 0.6742126941680908, + "learning_rate": 0.0009361833258628419, + "loss": 0.4888, + "step": 2383 + }, + { + "epoch": 0.1331955191775847, + "grad_norm": 0.6253450512886047, + "learning_rate": 0.000936155311519498, + "loss": 0.4997, + "step": 2384 + }, + { + "epoch": 0.13325138978126658, + "grad_norm": 0.6299285888671875, + "learning_rate": 0.0009361272971761542, + "loss": 0.4173, + "step": 2385 + }, + { + "epoch": 0.13330726038494847, + "grad_norm": 1.8023881912231445, + "learning_rate": 0.0009360992828328105, + "loss": 0.4727, + "step": 2386 + }, + { + "epoch": 0.13336313098863034, + "grad_norm": 0.5397359728813171, + "learning_rate": 0.0009360712684894667, + "loss": 0.6326, + "step": 2387 + }, + { + "epoch": 0.1334190015923122, + "grad_norm": 0.5429871082305908, + "learning_rate": 0.0009360432541461228, + "loss": 0.5279, + "step": 2388 + }, + { + "epoch": 0.13347487219599408, + "grad_norm": 0.5359370708465576, + "learning_rate": 0.000936015239802779, + "loss": 0.4203, + "step": 2389 + }, + { + "epoch": 0.13353074279967594, + "grad_norm": 0.9029973745346069, + "learning_rate": 0.0009359872254594353, + "loss": 0.5331, + "step": 2390 + }, + { + "epoch": 0.1335866134033578, + "grad_norm": 0.5609516501426697, + "learning_rate": 0.0009359592111160915, + "loss": 0.4815, + "step": 2391 + }, + { + "epoch": 0.1336424840070397, + "grad_norm": 0.5003063082695007, + "learning_rate": 0.0009359311967727476, + "loss": 0.5154, + "step": 2392 + }, + { + "epoch": 0.13369835461072158, + "grad_norm": 0.793759286403656, + "learning_rate": 0.0009359031824294039, + "loss": 0.5121, + "step": 2393 + }, + { + "epoch": 0.13375422521440344, + "grad_norm": 0.7004041075706482, + "learning_rate": 0.0009358751680860601, + "loss": 0.4642, + "step": 2394 + }, + { + "epoch": 0.1338100958180853, + "grad_norm": 0.961705207824707, + "learning_rate": 0.0009358471537427163, + "loss": 0.5687, + "step": 2395 + }, + { + "epoch": 0.13386596642176718, + "grad_norm": 0.689286470413208, + "learning_rate": 0.0009358191393993724, + "loss": 0.457, + "step": 2396 + }, + { + "epoch": 0.13392183702544905, + "grad_norm": 0.48053938150405884, + "learning_rate": 0.0009357911250560287, + "loss": 0.4761, + "step": 2397 + }, + { + "epoch": 0.13397770762913094, + "grad_norm": 0.6129217743873596, + "learning_rate": 0.0009357631107126849, + "loss": 0.5069, + "step": 2398 + }, + { + "epoch": 0.1340335782328128, + "grad_norm": 0.767074465751648, + "learning_rate": 0.0009357350963693411, + "loss": 0.4456, + "step": 2399 + }, + { + "epoch": 0.13408944883649468, + "grad_norm": 1.0963823795318604, + "learning_rate": 0.0009357070820259974, + "loss": 0.4845, + "step": 2400 + }, + { + "epoch": 0.13414531944017655, + "grad_norm": 0.6370484828948975, + "learning_rate": 0.0009356790676826536, + "loss": 0.4398, + "step": 2401 + }, + { + "epoch": 0.13420119004385841, + "grad_norm": 2.5470218658447266, + "learning_rate": 0.0009356510533393098, + "loss": 0.5572, + "step": 2402 + }, + { + "epoch": 0.1342570606475403, + "grad_norm": 0.9465886950492859, + "learning_rate": 0.0009356230389959659, + "loss": 0.5843, + "step": 2403 + }, + { + "epoch": 0.13431293125122218, + "grad_norm": 0.9283004999160767, + "learning_rate": 0.0009355950246526222, + "loss": 0.5537, + "step": 2404 + }, + { + "epoch": 0.13436880185490405, + "grad_norm": 1.1027721166610718, + "learning_rate": 0.0009355670103092784, + "loss": 0.3978, + "step": 2405 + }, + { + "epoch": 0.13442467245858591, + "grad_norm": 1.1196120977401733, + "learning_rate": 0.0009355389959659346, + "loss": 0.4353, + "step": 2406 + }, + { + "epoch": 0.13448054306226778, + "grad_norm": 0.7123436331748962, + "learning_rate": 0.0009355109816225907, + "loss": 0.4193, + "step": 2407 + }, + { + "epoch": 0.13453641366594965, + "grad_norm": 0.5865284204483032, + "learning_rate": 0.000935482967279247, + "loss": 0.4681, + "step": 2408 + }, + { + "epoch": 0.13459228426963155, + "grad_norm": 0.6923812031745911, + "learning_rate": 0.0009354549529359032, + "loss": 0.4467, + "step": 2409 + }, + { + "epoch": 0.1346481548733134, + "grad_norm": 0.5380160808563232, + "learning_rate": 0.0009354269385925594, + "loss": 0.4796, + "step": 2410 + }, + { + "epoch": 0.13470402547699528, + "grad_norm": 0.5595297813415527, + "learning_rate": 0.0009353989242492156, + "loss": 0.4203, + "step": 2411 + }, + { + "epoch": 0.13475989608067715, + "grad_norm": 0.7681472301483154, + "learning_rate": 0.0009353709099058718, + "loss": 0.618, + "step": 2412 + }, + { + "epoch": 0.13481576668435902, + "grad_norm": 0.48107534646987915, + "learning_rate": 0.000935342895562528, + "loss": 0.5247, + "step": 2413 + }, + { + "epoch": 0.13487163728804089, + "grad_norm": 0.4898092746734619, + "learning_rate": 0.0009353148812191842, + "loss": 0.469, + "step": 2414 + }, + { + "epoch": 0.13492750789172278, + "grad_norm": 1.4108259677886963, + "learning_rate": 0.0009352868668758404, + "loss": 0.4519, + "step": 2415 + }, + { + "epoch": 0.13498337849540465, + "grad_norm": 0.719360888004303, + "learning_rate": 0.0009352588525324966, + "loss": 0.3708, + "step": 2416 + }, + { + "epoch": 0.13503924909908652, + "grad_norm": 0.5191084146499634, + "learning_rate": 0.0009352308381891529, + "loss": 0.4277, + "step": 2417 + }, + { + "epoch": 0.13509511970276838, + "grad_norm": 0.870060384273529, + "learning_rate": 0.0009352028238458092, + "loss": 0.5794, + "step": 2418 + }, + { + "epoch": 0.13515099030645025, + "grad_norm": 0.6279982924461365, + "learning_rate": 0.0009351748095024653, + "loss": 0.5062, + "step": 2419 + }, + { + "epoch": 0.13520686091013212, + "grad_norm": 0.6179054379463196, + "learning_rate": 0.0009351467951591215, + "loss": 0.4373, + "step": 2420 + }, + { + "epoch": 0.13526273151381402, + "grad_norm": 0.78786700963974, + "learning_rate": 0.0009351187808157777, + "loss": 0.5111, + "step": 2421 + }, + { + "epoch": 0.13531860211749588, + "grad_norm": 0.5138941407203674, + "learning_rate": 0.0009350907664724339, + "loss": 0.4652, + "step": 2422 + }, + { + "epoch": 0.13537447272117775, + "grad_norm": 0.7723894119262695, + "learning_rate": 0.0009350627521290901, + "loss": 0.7173, + "step": 2423 + }, + { + "epoch": 0.13543034332485962, + "grad_norm": 0.694516658782959, + "learning_rate": 0.0009350347377857463, + "loss": 0.5352, + "step": 2424 + }, + { + "epoch": 0.1354862139285415, + "grad_norm": 0.5925862193107605, + "learning_rate": 0.0009350067234424026, + "loss": 0.5172, + "step": 2425 + }, + { + "epoch": 0.13554208453222338, + "grad_norm": 1.1658717393875122, + "learning_rate": 0.0009349787090990587, + "loss": 0.5079, + "step": 2426 + }, + { + "epoch": 0.13559795513590525, + "grad_norm": 0.5712295174598694, + "learning_rate": 0.0009349506947557149, + "loss": 0.4774, + "step": 2427 + }, + { + "epoch": 0.13565382573958712, + "grad_norm": 0.6435292363166809, + "learning_rate": 0.0009349226804123711, + "loss": 0.6122, + "step": 2428 + }, + { + "epoch": 0.135709696343269, + "grad_norm": 0.9454666972160339, + "learning_rate": 0.0009348946660690274, + "loss": 0.4496, + "step": 2429 + }, + { + "epoch": 0.13576556694695086, + "grad_norm": 0.43431735038757324, + "learning_rate": 0.0009348666517256835, + "loss": 0.4183, + "step": 2430 + }, + { + "epoch": 0.13582143755063272, + "grad_norm": 0.520236074924469, + "learning_rate": 0.0009348386373823397, + "loss": 0.5267, + "step": 2431 + }, + { + "epoch": 0.13587730815431462, + "grad_norm": 0.8118947744369507, + "learning_rate": 0.000934810623038996, + "loss": 0.4785, + "step": 2432 + }, + { + "epoch": 0.1359331787579965, + "grad_norm": 0.7891973853111267, + "learning_rate": 0.0009347826086956523, + "loss": 0.4671, + "step": 2433 + }, + { + "epoch": 0.13598904936167835, + "grad_norm": 0.6550859808921814, + "learning_rate": 0.0009347545943523084, + "loss": 0.4443, + "step": 2434 + }, + { + "epoch": 0.13604491996536022, + "grad_norm": 0.44750410318374634, + "learning_rate": 0.0009347265800089646, + "loss": 0.4713, + "step": 2435 + }, + { + "epoch": 0.1361007905690421, + "grad_norm": 0.6525522470474243, + "learning_rate": 0.0009346985656656209, + "loss": 0.4774, + "step": 2436 + }, + { + "epoch": 0.13615666117272396, + "grad_norm": 0.8258258104324341, + "learning_rate": 0.0009346705513222771, + "loss": 0.4829, + "step": 2437 + }, + { + "epoch": 0.13621253177640585, + "grad_norm": 1.0558582544326782, + "learning_rate": 0.0009346425369789332, + "loss": 0.6199, + "step": 2438 + }, + { + "epoch": 0.13626840238008772, + "grad_norm": 0.5674573183059692, + "learning_rate": 0.0009346145226355895, + "loss": 0.5377, + "step": 2439 + }, + { + "epoch": 0.1363242729837696, + "grad_norm": 0.4739237129688263, + "learning_rate": 0.0009345865082922457, + "loss": 0.3788, + "step": 2440 + }, + { + "epoch": 0.13638014358745146, + "grad_norm": 0.5413340330123901, + "learning_rate": 0.0009345584939489019, + "loss": 0.4782, + "step": 2441 + }, + { + "epoch": 0.13643601419113333, + "grad_norm": 1.4440451860427856, + "learning_rate": 0.000934530479605558, + "loss": 0.4717, + "step": 2442 + }, + { + "epoch": 0.13649188479481522, + "grad_norm": 0.6521468758583069, + "learning_rate": 0.0009345024652622143, + "loss": 0.4567, + "step": 2443 + }, + { + "epoch": 0.1365477553984971, + "grad_norm": 1.4619220495224, + "learning_rate": 0.0009344744509188705, + "loss": 0.4871, + "step": 2444 + }, + { + "epoch": 0.13660362600217896, + "grad_norm": 0.5928704142570496, + "learning_rate": 0.0009344464365755266, + "loss": 0.4477, + "step": 2445 + }, + { + "epoch": 0.13665949660586083, + "grad_norm": 0.952032208442688, + "learning_rate": 0.0009344184222321828, + "loss": 0.499, + "step": 2446 + }, + { + "epoch": 0.1367153672095427, + "grad_norm": 0.5560368895530701, + "learning_rate": 0.0009343904078888391, + "loss": 0.4466, + "step": 2447 + }, + { + "epoch": 0.13677123781322456, + "grad_norm": 4.875126361846924, + "learning_rate": 0.0009343623935454953, + "loss": 0.5796, + "step": 2448 + }, + { + "epoch": 0.13682710841690646, + "grad_norm": 0.5593515634536743, + "learning_rate": 0.0009343343792021514, + "loss": 0.5174, + "step": 2449 + }, + { + "epoch": 0.13688297902058832, + "grad_norm": 0.5454943776130676, + "learning_rate": 0.0009343063648588078, + "loss": 0.5029, + "step": 2450 + }, + { + "epoch": 0.1369388496242702, + "grad_norm": 4.492527484893799, + "learning_rate": 0.000934278350515464, + "loss": 0.5037, + "step": 2451 + }, + { + "epoch": 0.13699472022795206, + "grad_norm": 2.568387985229492, + "learning_rate": 0.0009342503361721202, + "loss": 0.7036, + "step": 2452 + }, + { + "epoch": 0.13705059083163393, + "grad_norm": 4.55485200881958, + "learning_rate": 0.0009342223218287763, + "loss": 0.3912, + "step": 2453 + }, + { + "epoch": 0.1371064614353158, + "grad_norm": 0.9727553129196167, + "learning_rate": 0.0009341943074854326, + "loss": 0.4855, + "step": 2454 + }, + { + "epoch": 0.1371623320389977, + "grad_norm": 0.5304064750671387, + "learning_rate": 0.0009341662931420888, + "loss": 0.4976, + "step": 2455 + }, + { + "epoch": 0.13721820264267956, + "grad_norm": 1.8803430795669556, + "learning_rate": 0.000934138278798745, + "loss": 0.5492, + "step": 2456 + }, + { + "epoch": 0.13727407324636143, + "grad_norm": 0.5615953207015991, + "learning_rate": 0.0009341102644554012, + "loss": 0.4891, + "step": 2457 + }, + { + "epoch": 0.1373299438500433, + "grad_norm": 0.5044034123420715, + "learning_rate": 0.0009340822501120574, + "loss": 0.433, + "step": 2458 + }, + { + "epoch": 0.13738581445372516, + "grad_norm": 0.6585912704467773, + "learning_rate": 0.0009340542357687136, + "loss": 0.4319, + "step": 2459 + }, + { + "epoch": 0.13744168505740703, + "grad_norm": 0.6582136750221252, + "learning_rate": 0.0009340262214253698, + "loss": 0.3893, + "step": 2460 + }, + { + "epoch": 0.13749755566108893, + "grad_norm": 0.6968470215797424, + "learning_rate": 0.000933998207082026, + "loss": 0.5002, + "step": 2461 + }, + { + "epoch": 0.1375534262647708, + "grad_norm": 0.6057218909263611, + "learning_rate": 0.0009339701927386822, + "loss": 0.5937, + "step": 2462 + }, + { + "epoch": 0.13760929686845266, + "grad_norm": 0.7825283408164978, + "learning_rate": 0.0009339421783953384, + "loss": 0.6306, + "step": 2463 + }, + { + "epoch": 0.13766516747213453, + "grad_norm": 0.8319100737571716, + "learning_rate": 0.0009339141640519946, + "loss": 0.5016, + "step": 2464 + }, + { + "epoch": 0.1377210380758164, + "grad_norm": 0.6460074782371521, + "learning_rate": 0.0009338861497086508, + "loss": 0.5389, + "step": 2465 + }, + { + "epoch": 0.1377769086794983, + "grad_norm": 0.7123767137527466, + "learning_rate": 0.000933858135365307, + "loss": 0.4392, + "step": 2466 + }, + { + "epoch": 0.13783277928318016, + "grad_norm": 0.6445744037628174, + "learning_rate": 0.0009338301210219633, + "loss": 0.4573, + "step": 2467 + }, + { + "epoch": 0.13788864988686203, + "grad_norm": 1.190752387046814, + "learning_rate": 0.0009338021066786195, + "loss": 0.3996, + "step": 2468 + }, + { + "epoch": 0.1379445204905439, + "grad_norm": 0.6908252239227295, + "learning_rate": 0.0009337740923352757, + "loss": 0.6015, + "step": 2469 + }, + { + "epoch": 0.13800039109422577, + "grad_norm": 0.7207625508308411, + "learning_rate": 0.0009337460779919319, + "loss": 0.4795, + "step": 2470 + }, + { + "epoch": 0.13805626169790763, + "grad_norm": 2.88751482963562, + "learning_rate": 0.0009337180636485882, + "loss": 0.3907, + "step": 2471 + }, + { + "epoch": 0.13811213230158953, + "grad_norm": 0.7122745513916016, + "learning_rate": 0.0009336900493052443, + "loss": 0.5685, + "step": 2472 + }, + { + "epoch": 0.1381680029052714, + "grad_norm": 0.6525972485542297, + "learning_rate": 0.0009336620349619005, + "loss": 0.5084, + "step": 2473 + }, + { + "epoch": 0.13822387350895327, + "grad_norm": 1.1809091567993164, + "learning_rate": 0.0009336340206185567, + "loss": 0.66, + "step": 2474 + }, + { + "epoch": 0.13827974411263513, + "grad_norm": 0.507585346698761, + "learning_rate": 0.000933606006275213, + "loss": 0.5009, + "step": 2475 + }, + { + "epoch": 0.138335614716317, + "grad_norm": 0.9773241281509399, + "learning_rate": 0.0009335779919318691, + "loss": 0.4824, + "step": 2476 + }, + { + "epoch": 0.13839148531999887, + "grad_norm": 1.5625030994415283, + "learning_rate": 0.0009335499775885253, + "loss": 0.5828, + "step": 2477 + }, + { + "epoch": 0.13844735592368076, + "grad_norm": 6.245696067810059, + "learning_rate": 0.0009335219632451816, + "loss": 0.5479, + "step": 2478 + }, + { + "epoch": 0.13850322652736263, + "grad_norm": 0.7169110774993896, + "learning_rate": 0.0009334939489018378, + "loss": 0.5374, + "step": 2479 + }, + { + "epoch": 0.1385590971310445, + "grad_norm": 16.486364364624023, + "learning_rate": 0.0009334659345584939, + "loss": 0.6291, + "step": 2480 + }, + { + "epoch": 0.13861496773472637, + "grad_norm": 1.3078811168670654, + "learning_rate": 0.0009334379202151501, + "loss": 0.5297, + "step": 2481 + }, + { + "epoch": 0.13867083833840824, + "grad_norm": 1.3116075992584229, + "learning_rate": 0.0009334099058718064, + "loss": 0.663, + "step": 2482 + }, + { + "epoch": 0.13872670894209013, + "grad_norm": 0.5865618586540222, + "learning_rate": 0.0009333818915284627, + "loss": 0.5496, + "step": 2483 + }, + { + "epoch": 0.138782579545772, + "grad_norm": 0.6319894194602966, + "learning_rate": 0.0009333538771851188, + "loss": 0.5217, + "step": 2484 + }, + { + "epoch": 0.13883845014945387, + "grad_norm": 1.1971015930175781, + "learning_rate": 0.000933325862841775, + "loss": 0.407, + "step": 2485 + }, + { + "epoch": 0.13889432075313574, + "grad_norm": 3.0417113304138184, + "learning_rate": 0.0009332978484984313, + "loss": 0.6032, + "step": 2486 + }, + { + "epoch": 0.1389501913568176, + "grad_norm": 0.9076368808746338, + "learning_rate": 0.0009332698341550874, + "loss": 0.4767, + "step": 2487 + }, + { + "epoch": 0.13900606196049947, + "grad_norm": 0.7165853977203369, + "learning_rate": 0.0009332418198117436, + "loss": 0.4309, + "step": 2488 + }, + { + "epoch": 0.13906193256418137, + "grad_norm": 0.6923040151596069, + "learning_rate": 0.0009332138054683999, + "loss": 0.503, + "step": 2489 + }, + { + "epoch": 0.13911780316786324, + "grad_norm": 0.6640058159828186, + "learning_rate": 0.0009331857911250561, + "loss": 0.454, + "step": 2490 + }, + { + "epoch": 0.1391736737715451, + "grad_norm": 0.5700077414512634, + "learning_rate": 0.0009331577767817122, + "loss": 0.4195, + "step": 2491 + }, + { + "epoch": 0.13922954437522697, + "grad_norm": 1.9293936491012573, + "learning_rate": 0.0009331297624383684, + "loss": 0.5048, + "step": 2492 + }, + { + "epoch": 0.13928541497890884, + "grad_norm": 0.6100828647613525, + "learning_rate": 0.0009331017480950247, + "loss": 0.3709, + "step": 2493 + }, + { + "epoch": 0.1393412855825907, + "grad_norm": 4.568056106567383, + "learning_rate": 0.0009330737337516809, + "loss": 0.4812, + "step": 2494 + }, + { + "epoch": 0.1393971561862726, + "grad_norm": 1.6525168418884277, + "learning_rate": 0.000933045719408337, + "loss": 0.5657, + "step": 2495 + }, + { + "epoch": 0.13945302678995447, + "grad_norm": 0.6785444617271423, + "learning_rate": 0.0009330177050649933, + "loss": 0.5597, + "step": 2496 + }, + { + "epoch": 0.13950889739363634, + "grad_norm": 0.6084670424461365, + "learning_rate": 0.0009329896907216495, + "loss": 0.4588, + "step": 2497 + }, + { + "epoch": 0.1395647679973182, + "grad_norm": 1.31582510471344, + "learning_rate": 0.0009329616763783057, + "loss": 0.6769, + "step": 2498 + }, + { + "epoch": 0.13962063860100007, + "grad_norm": 0.9281377196311951, + "learning_rate": 0.0009329336620349618, + "loss": 0.4521, + "step": 2499 + }, + { + "epoch": 0.13967650920468197, + "grad_norm": 0.8159233331680298, + "learning_rate": 0.0009329056476916182, + "loss": 0.5065, + "step": 2500 + }, + { + "epoch": 0.13967650920468197, + "eval_cer": 0.10084776275735655, + "eval_loss": 0.3796367943286896, + "eval_runtime": 55.9179, + "eval_samples_per_second": 81.155, + "eval_steps_per_second": 5.079, + "eval_wer": 0.39373201865869667, + "step": 2500 + }, + { + "epoch": 0.13973237980836384, + "grad_norm": 0.42360323667526245, + "learning_rate": 0.0009328776333482744, + "loss": 0.3916, + "step": 2501 + }, + { + "epoch": 0.1397882504120457, + "grad_norm": 0.6667183637619019, + "learning_rate": 0.0009328496190049306, + "loss": 0.4945, + "step": 2502 + }, + { + "epoch": 0.13984412101572757, + "grad_norm": 0.567635715007782, + "learning_rate": 0.0009328216046615868, + "loss": 0.5408, + "step": 2503 + }, + { + "epoch": 0.13989999161940944, + "grad_norm": 0.8093522191047668, + "learning_rate": 0.000932793590318243, + "loss": 0.5798, + "step": 2504 + }, + { + "epoch": 0.1399558622230913, + "grad_norm": 0.7080284357070923, + "learning_rate": 0.0009327655759748992, + "loss": 0.5125, + "step": 2505 + }, + { + "epoch": 0.1400117328267732, + "grad_norm": 0.5052550435066223, + "learning_rate": 0.0009327375616315553, + "loss": 0.5169, + "step": 2506 + }, + { + "epoch": 0.14006760343045507, + "grad_norm": 0.5504215955734253, + "learning_rate": 0.0009327095472882116, + "loss": 0.4869, + "step": 2507 + }, + { + "epoch": 0.14012347403413694, + "grad_norm": 0.8864015340805054, + "learning_rate": 0.0009326815329448678, + "loss": 0.4546, + "step": 2508 + }, + { + "epoch": 0.1401793446378188, + "grad_norm": 0.9367356300354004, + "learning_rate": 0.000932653518601524, + "loss": 0.4449, + "step": 2509 + }, + { + "epoch": 0.14023521524150068, + "grad_norm": 0.8201373219490051, + "learning_rate": 0.0009326255042581802, + "loss": 0.5098, + "step": 2510 + }, + { + "epoch": 0.14029108584518254, + "grad_norm": 2.680048942565918, + "learning_rate": 0.0009325974899148364, + "loss": 0.5783, + "step": 2511 + }, + { + "epoch": 0.14034695644886444, + "grad_norm": 0.6723512411117554, + "learning_rate": 0.0009325694755714926, + "loss": 0.4304, + "step": 2512 + }, + { + "epoch": 0.1404028270525463, + "grad_norm": 1.1711534261703491, + "learning_rate": 0.0009325414612281488, + "loss": 0.5121, + "step": 2513 + }, + { + "epoch": 0.14045869765622818, + "grad_norm": 0.5774499773979187, + "learning_rate": 0.000932513446884805, + "loss": 0.4001, + "step": 2514 + }, + { + "epoch": 0.14051456825991004, + "grad_norm": 0.5288100838661194, + "learning_rate": 0.0009324854325414612, + "loss": 0.4257, + "step": 2515 + }, + { + "epoch": 0.1405704388635919, + "grad_norm": 0.42888545989990234, + "learning_rate": 0.0009324574181981174, + "loss": 0.4337, + "step": 2516 + }, + { + "epoch": 0.14062630946727378, + "grad_norm": 0.909542977809906, + "learning_rate": 0.0009324294038547737, + "loss": 0.6389, + "step": 2517 + }, + { + "epoch": 0.14068218007095568, + "grad_norm": 1.5811307430267334, + "learning_rate": 0.0009324013895114299, + "loss": 0.654, + "step": 2518 + }, + { + "epoch": 0.14073805067463754, + "grad_norm": 0.817436933517456, + "learning_rate": 0.0009323733751680861, + "loss": 0.4575, + "step": 2519 + }, + { + "epoch": 0.1407939212783194, + "grad_norm": 0.46136197447776794, + "learning_rate": 0.0009323453608247423, + "loss": 0.5045, + "step": 2520 + }, + { + "epoch": 0.14084979188200128, + "grad_norm": 0.687739372253418, + "learning_rate": 0.0009323173464813986, + "loss": 0.4399, + "step": 2521 + }, + { + "epoch": 0.14090566248568315, + "grad_norm": 0.5013472437858582, + "learning_rate": 0.0009322893321380547, + "loss": 0.4228, + "step": 2522 + }, + { + "epoch": 0.14096153308936504, + "grad_norm": 0.6458079218864441, + "learning_rate": 0.0009322613177947109, + "loss": 0.4042, + "step": 2523 + }, + { + "epoch": 0.1410174036930469, + "grad_norm": 2.2879559993743896, + "learning_rate": 0.0009322333034513671, + "loss": 0.4177, + "step": 2524 + }, + { + "epoch": 0.14107327429672878, + "grad_norm": 0.9168562889099121, + "learning_rate": 0.0009322052891080234, + "loss": 0.4507, + "step": 2525 + }, + { + "epoch": 0.14112914490041065, + "grad_norm": 0.8930234313011169, + "learning_rate": 0.0009321772747646795, + "loss": 0.5474, + "step": 2526 + }, + { + "epoch": 0.14118501550409251, + "grad_norm": 0.44630077481269836, + "learning_rate": 0.0009321492604213357, + "loss": 0.3823, + "step": 2527 + }, + { + "epoch": 0.14124088610777438, + "grad_norm": 1.0088295936584473, + "learning_rate": 0.000932121246077992, + "loss": 0.5528, + "step": 2528 + }, + { + "epoch": 0.14129675671145628, + "grad_norm": 0.6344780325889587, + "learning_rate": 0.0009320932317346481, + "loss": 0.5224, + "step": 2529 + }, + { + "epoch": 0.14135262731513815, + "grad_norm": 0.77131587266922, + "learning_rate": 0.0009320652173913043, + "loss": 0.4077, + "step": 2530 + }, + { + "epoch": 0.14140849791882001, + "grad_norm": 0.6915258169174194, + "learning_rate": 0.0009320372030479605, + "loss": 0.4914, + "step": 2531 + }, + { + "epoch": 0.14146436852250188, + "grad_norm": 4.294975280761719, + "learning_rate": 0.0009320091887046168, + "loss": 0.4446, + "step": 2532 + }, + { + "epoch": 0.14152023912618375, + "grad_norm": 0.6202715635299683, + "learning_rate": 0.0009319811743612729, + "loss": 0.5189, + "step": 2533 + }, + { + "epoch": 0.14157610972986562, + "grad_norm": 0.49865883588790894, + "learning_rate": 0.0009319531600179292, + "loss": 0.5379, + "step": 2534 + }, + { + "epoch": 0.1416319803335475, + "grad_norm": 0.659381628036499, + "learning_rate": 0.0009319251456745855, + "loss": 0.5545, + "step": 2535 + }, + { + "epoch": 0.14168785093722938, + "grad_norm": 0.7195130586624146, + "learning_rate": 0.0009318971313312417, + "loss": 0.4621, + "step": 2536 + }, + { + "epoch": 0.14174372154091125, + "grad_norm": 1.03648042678833, + "learning_rate": 0.0009318691169878978, + "loss": 0.4426, + "step": 2537 + }, + { + "epoch": 0.14179959214459312, + "grad_norm": 2.748897075653076, + "learning_rate": 0.000931841102644554, + "loss": 0.5606, + "step": 2538 + }, + { + "epoch": 0.14185546274827499, + "grad_norm": 0.3928808569908142, + "learning_rate": 0.0009318130883012103, + "loss": 0.3879, + "step": 2539 + }, + { + "epoch": 0.14191133335195688, + "grad_norm": 1.4371105432510376, + "learning_rate": 0.0009317850739578665, + "loss": 0.5985, + "step": 2540 + }, + { + "epoch": 0.14196720395563875, + "grad_norm": 0.775253415107727, + "learning_rate": 0.0009317570596145226, + "loss": 0.5096, + "step": 2541 + }, + { + "epoch": 0.14202307455932062, + "grad_norm": 0.46555808186531067, + "learning_rate": 0.0009317290452711789, + "loss": 0.5514, + "step": 2542 + }, + { + "epoch": 0.14207894516300248, + "grad_norm": 0.6411169171333313, + "learning_rate": 0.0009317010309278351, + "loss": 0.4321, + "step": 2543 + }, + { + "epoch": 0.14213481576668435, + "grad_norm": 0.5415703058242798, + "learning_rate": 0.0009316730165844913, + "loss": 0.5138, + "step": 2544 + }, + { + "epoch": 0.14219068637036622, + "grad_norm": 0.8139814138412476, + "learning_rate": 0.0009316450022411474, + "loss": 0.4603, + "step": 2545 + }, + { + "epoch": 0.14224655697404812, + "grad_norm": 0.7348641157150269, + "learning_rate": 0.0009316169878978037, + "loss": 0.6052, + "step": 2546 + }, + { + "epoch": 0.14230242757772998, + "grad_norm": 2.093893527984619, + "learning_rate": 0.0009315889735544599, + "loss": 0.508, + "step": 2547 + }, + { + "epoch": 0.14235829818141185, + "grad_norm": 0.7186673283576965, + "learning_rate": 0.0009315609592111161, + "loss": 0.5689, + "step": 2548 + }, + { + "epoch": 0.14241416878509372, + "grad_norm": 0.4833289086818695, + "learning_rate": 0.0009315329448677722, + "loss": 0.4548, + "step": 2549 + }, + { + "epoch": 0.1424700393887756, + "grad_norm": 0.42815327644348145, + "learning_rate": 0.0009315049305244286, + "loss": 0.5123, + "step": 2550 + }, + { + "epoch": 0.14252590999245746, + "grad_norm": 0.5422353744506836, + "learning_rate": 0.0009314769161810848, + "loss": 0.5242, + "step": 2551 + }, + { + "epoch": 0.14258178059613935, + "grad_norm": 0.5827128887176514, + "learning_rate": 0.0009314489018377409, + "loss": 0.4851, + "step": 2552 + }, + { + "epoch": 0.14263765119982122, + "grad_norm": 0.6805925369262695, + "learning_rate": 0.0009314208874943972, + "loss": 0.5232, + "step": 2553 + }, + { + "epoch": 0.1426935218035031, + "grad_norm": 0.7072283625602722, + "learning_rate": 0.0009313928731510534, + "loss": 0.6303, + "step": 2554 + }, + { + "epoch": 0.14274939240718496, + "grad_norm": 1.4853183031082153, + "learning_rate": 0.0009313648588077096, + "loss": 0.6193, + "step": 2555 + }, + { + "epoch": 0.14280526301086682, + "grad_norm": 2.314371347427368, + "learning_rate": 0.0009313368444643657, + "loss": 0.5936, + "step": 2556 + }, + { + "epoch": 0.14286113361454872, + "grad_norm": 0.5001146197319031, + "learning_rate": 0.000931308830121022, + "loss": 0.5538, + "step": 2557 + }, + { + "epoch": 0.1429170042182306, + "grad_norm": 3.902667284011841, + "learning_rate": 0.0009312808157776782, + "loss": 0.5635, + "step": 2558 + }, + { + "epoch": 0.14297287482191245, + "grad_norm": 0.5832546353340149, + "learning_rate": 0.0009312528014343344, + "loss": 0.5576, + "step": 2559 + }, + { + "epoch": 0.14302874542559432, + "grad_norm": 0.771173894405365, + "learning_rate": 0.0009312247870909906, + "loss": 0.4155, + "step": 2560 + }, + { + "epoch": 0.1430846160292762, + "grad_norm": 1.9446437358856201, + "learning_rate": 0.0009311967727476468, + "loss": 0.4434, + "step": 2561 + }, + { + "epoch": 0.14314048663295806, + "grad_norm": 0.5480421781539917, + "learning_rate": 0.000931168758404303, + "loss": 0.5525, + "step": 2562 + }, + { + "epoch": 0.14319635723663995, + "grad_norm": 5.431925296783447, + "learning_rate": 0.0009311407440609592, + "loss": 0.4964, + "step": 2563 + }, + { + "epoch": 0.14325222784032182, + "grad_norm": 0.5309224724769592, + "learning_rate": 0.0009311127297176154, + "loss": 0.4364, + "step": 2564 + }, + { + "epoch": 0.1433080984440037, + "grad_norm": 0.7984029650688171, + "learning_rate": 0.0009310847153742716, + "loss": 0.4687, + "step": 2565 + }, + { + "epoch": 0.14336396904768556, + "grad_norm": 0.5771551728248596, + "learning_rate": 0.0009310567010309278, + "loss": 0.4619, + "step": 2566 + }, + { + "epoch": 0.14341983965136743, + "grad_norm": 0.5826465487480164, + "learning_rate": 0.0009310286866875842, + "loss": 0.6153, + "step": 2567 + }, + { + "epoch": 0.1434757102550493, + "grad_norm": 4.1241254806518555, + "learning_rate": 0.0009310006723442403, + "loss": 0.5589, + "step": 2568 + }, + { + "epoch": 0.1435315808587312, + "grad_norm": 0.8903993368148804, + "learning_rate": 0.0009309726580008965, + "loss": 0.6232, + "step": 2569 + }, + { + "epoch": 0.14358745146241306, + "grad_norm": 0.46794942021369934, + "learning_rate": 0.0009309446436575527, + "loss": 0.4035, + "step": 2570 + }, + { + "epoch": 0.14364332206609493, + "grad_norm": 0.5597148537635803, + "learning_rate": 0.0009309166293142089, + "loss": 0.4631, + "step": 2571 + }, + { + "epoch": 0.1436991926697768, + "grad_norm": 1.1420543193817139, + "learning_rate": 0.0009308886149708651, + "loss": 0.6142, + "step": 2572 + }, + { + "epoch": 0.14375506327345866, + "grad_norm": 0.556740403175354, + "learning_rate": 0.0009308606006275213, + "loss": 0.4829, + "step": 2573 + }, + { + "epoch": 0.14381093387714053, + "grad_norm": 0.7975441813468933, + "learning_rate": 0.0009308325862841776, + "loss": 0.4665, + "step": 2574 + }, + { + "epoch": 0.14386680448082242, + "grad_norm": 6.180104732513428, + "learning_rate": 0.0009308045719408337, + "loss": 0.6175, + "step": 2575 + }, + { + "epoch": 0.1439226750845043, + "grad_norm": 0.7764971256256104, + "learning_rate": 0.0009307765575974899, + "loss": 0.4773, + "step": 2576 + }, + { + "epoch": 0.14397854568818616, + "grad_norm": 0.8303751349449158, + "learning_rate": 0.0009307485432541461, + "loss": 0.6614, + "step": 2577 + }, + { + "epoch": 0.14403441629186803, + "grad_norm": 0.7853827476501465, + "learning_rate": 0.0009307205289108024, + "loss": 0.62, + "step": 2578 + }, + { + "epoch": 0.1440902868955499, + "grad_norm": 0.6165528893470764, + "learning_rate": 0.0009306925145674585, + "loss": 0.5347, + "step": 2579 + }, + { + "epoch": 0.1441461574992318, + "grad_norm": 0.6020848155021667, + "learning_rate": 0.0009306645002241147, + "loss": 0.5235, + "step": 2580 + }, + { + "epoch": 0.14420202810291366, + "grad_norm": 0.712619423866272, + "learning_rate": 0.000930636485880771, + "loss": 0.4503, + "step": 2581 + }, + { + "epoch": 0.14425789870659553, + "grad_norm": 0.5719115734100342, + "learning_rate": 0.0009306084715374272, + "loss": 0.4698, + "step": 2582 + }, + { + "epoch": 0.1443137693102774, + "grad_norm": 0.5748353004455566, + "learning_rate": 0.0009305804571940833, + "loss": 0.5922, + "step": 2583 + }, + { + "epoch": 0.14436963991395926, + "grad_norm": 0.5952667593955994, + "learning_rate": 0.0009305524428507395, + "loss": 0.469, + "step": 2584 + }, + { + "epoch": 0.14442551051764113, + "grad_norm": 0.6945236921310425, + "learning_rate": 0.0009305244285073959, + "loss": 0.6551, + "step": 2585 + }, + { + "epoch": 0.14448138112132303, + "grad_norm": 0.7934621572494507, + "learning_rate": 0.0009304964141640521, + "loss": 0.4493, + "step": 2586 + }, + { + "epoch": 0.1445372517250049, + "grad_norm": 0.49064695835113525, + "learning_rate": 0.0009304683998207082, + "loss": 0.4314, + "step": 2587 + }, + { + "epoch": 0.14459312232868676, + "grad_norm": 1.3204824924468994, + "learning_rate": 0.0009304403854773644, + "loss": 0.5605, + "step": 2588 + }, + { + "epoch": 0.14464899293236863, + "grad_norm": 0.7447421550750732, + "learning_rate": 0.0009304123711340207, + "loss": 0.6351, + "step": 2589 + }, + { + "epoch": 0.1447048635360505, + "grad_norm": 0.6576716303825378, + "learning_rate": 0.0009303843567906769, + "loss": 0.5406, + "step": 2590 + }, + { + "epoch": 0.14476073413973237, + "grad_norm": 0.6636106371879578, + "learning_rate": 0.000930356342447333, + "loss": 0.5799, + "step": 2591 + }, + { + "epoch": 0.14481660474341426, + "grad_norm": 0.8832970857620239, + "learning_rate": 0.0009303283281039893, + "loss": 0.5103, + "step": 2592 + }, + { + "epoch": 0.14487247534709613, + "grad_norm": 0.49574288725852966, + "learning_rate": 0.0009303003137606455, + "loss": 0.4584, + "step": 2593 + }, + { + "epoch": 0.144928345950778, + "grad_norm": 1.1693203449249268, + "learning_rate": 0.0009302722994173016, + "loss": 0.3992, + "step": 2594 + }, + { + "epoch": 0.14498421655445987, + "grad_norm": 0.7095686197280884, + "learning_rate": 0.0009302442850739578, + "loss": 0.5216, + "step": 2595 + }, + { + "epoch": 0.14504008715814173, + "grad_norm": 0.5746054649353027, + "learning_rate": 0.0009302162707306141, + "loss": 0.4578, + "step": 2596 + }, + { + "epoch": 0.14509595776182363, + "grad_norm": 0.4660589098930359, + "learning_rate": 0.0009301882563872703, + "loss": 0.4072, + "step": 2597 + }, + { + "epoch": 0.1451518283655055, + "grad_norm": 6.34377908706665, + "learning_rate": 0.0009301602420439264, + "loss": 0.4875, + "step": 2598 + }, + { + "epoch": 0.14520769896918737, + "grad_norm": 1.0275946855545044, + "learning_rate": 0.0009301322277005827, + "loss": 0.5705, + "step": 2599 + }, + { + "epoch": 0.14526356957286923, + "grad_norm": 0.7125551700592041, + "learning_rate": 0.000930104213357239, + "loss": 0.6163, + "step": 2600 + }, + { + "epoch": 0.1453194401765511, + "grad_norm": 0.6042466163635254, + "learning_rate": 0.0009300761990138952, + "loss": 0.5836, + "step": 2601 + }, + { + "epoch": 0.14537531078023297, + "grad_norm": 0.9524787664413452, + "learning_rate": 0.0009300481846705513, + "loss": 0.4258, + "step": 2602 + }, + { + "epoch": 0.14543118138391486, + "grad_norm": 1.0709465742111206, + "learning_rate": 0.0009300201703272076, + "loss": 0.502, + "step": 2603 + }, + { + "epoch": 0.14548705198759673, + "grad_norm": 1.0216082334518433, + "learning_rate": 0.0009299921559838638, + "loss": 0.5646, + "step": 2604 + }, + { + "epoch": 0.1455429225912786, + "grad_norm": 0.46253690123558044, + "learning_rate": 0.00092996414164052, + "loss": 0.3753, + "step": 2605 + }, + { + "epoch": 0.14559879319496047, + "grad_norm": 1.9312692880630493, + "learning_rate": 0.0009299361272971762, + "loss": 0.4112, + "step": 2606 + }, + { + "epoch": 0.14565466379864234, + "grad_norm": 0.5686039924621582, + "learning_rate": 0.0009299081129538324, + "loss": 0.4422, + "step": 2607 + }, + { + "epoch": 0.1457105344023242, + "grad_norm": 0.5558215975761414, + "learning_rate": 0.0009298800986104886, + "loss": 0.4268, + "step": 2608 + }, + { + "epoch": 0.1457664050060061, + "grad_norm": 1.9876872301101685, + "learning_rate": 0.0009298520842671448, + "loss": 0.5002, + "step": 2609 + }, + { + "epoch": 0.14582227560968797, + "grad_norm": 0.5739237666130066, + "learning_rate": 0.000929824069923801, + "loss": 0.4363, + "step": 2610 + }, + { + "epoch": 0.14587814621336984, + "grad_norm": 0.639873206615448, + "learning_rate": 0.0009297960555804572, + "loss": 0.4748, + "step": 2611 + }, + { + "epoch": 0.1459340168170517, + "grad_norm": 0.7696776986122131, + "learning_rate": 0.0009297680412371134, + "loss": 0.7065, + "step": 2612 + }, + { + "epoch": 0.14598988742073357, + "grad_norm": 0.4580008387565613, + "learning_rate": 0.0009297400268937696, + "loss": 0.4294, + "step": 2613 + }, + { + "epoch": 0.14604575802441547, + "grad_norm": 0.608950674533844, + "learning_rate": 0.0009297120125504258, + "loss": 0.5229, + "step": 2614 + }, + { + "epoch": 0.14610162862809734, + "grad_norm": 0.6460857391357422, + "learning_rate": 0.000929683998207082, + "loss": 0.5381, + "step": 2615 + }, + { + "epoch": 0.1461574992317792, + "grad_norm": 0.762181282043457, + "learning_rate": 0.0009296559838637382, + "loss": 0.3484, + "step": 2616 + }, + { + "epoch": 0.14621336983546107, + "grad_norm": 0.7781669497489929, + "learning_rate": 0.0009296279695203944, + "loss": 0.664, + "step": 2617 + }, + { + "epoch": 0.14626924043914294, + "grad_norm": 0.4056796431541443, + "learning_rate": 0.0009295999551770507, + "loss": 0.4392, + "step": 2618 + }, + { + "epoch": 0.1463251110428248, + "grad_norm": 0.5402657985687256, + "learning_rate": 0.0009295719408337069, + "loss": 0.5627, + "step": 2619 + }, + { + "epoch": 0.1463809816465067, + "grad_norm": 0.5659785270690918, + "learning_rate": 0.0009295439264903632, + "loss": 0.5178, + "step": 2620 + }, + { + "epoch": 0.14643685225018857, + "grad_norm": 0.44199323654174805, + "learning_rate": 0.0009295159121470193, + "loss": 0.3403, + "step": 2621 + }, + { + "epoch": 0.14649272285387044, + "grad_norm": 0.917273998260498, + "learning_rate": 0.0009294878978036755, + "loss": 0.6191, + "step": 2622 + }, + { + "epoch": 0.1465485934575523, + "grad_norm": 0.454668253660202, + "learning_rate": 0.0009294598834603317, + "loss": 0.4559, + "step": 2623 + }, + { + "epoch": 0.14660446406123417, + "grad_norm": 1.0395748615264893, + "learning_rate": 0.000929431869116988, + "loss": 0.6088, + "step": 2624 + }, + { + "epoch": 0.14666033466491604, + "grad_norm": 2.63779354095459, + "learning_rate": 0.0009294038547736441, + "loss": 0.5487, + "step": 2625 + }, + { + "epoch": 0.14671620526859794, + "grad_norm": 0.8605122566223145, + "learning_rate": 0.0009293758404303003, + "loss": 0.4436, + "step": 2626 + }, + { + "epoch": 0.1467720758722798, + "grad_norm": 0.9000152945518494, + "learning_rate": 0.0009293478260869565, + "loss": 0.5456, + "step": 2627 + }, + { + "epoch": 0.14682794647596167, + "grad_norm": 0.6718060970306396, + "learning_rate": 0.0009293198117436128, + "loss": 0.4509, + "step": 2628 + }, + { + "epoch": 0.14688381707964354, + "grad_norm": 0.5090430378913879, + "learning_rate": 0.0009292917974002689, + "loss": 0.505, + "step": 2629 + }, + { + "epoch": 0.1469396876833254, + "grad_norm": 1.0891247987747192, + "learning_rate": 0.0009292637830569251, + "loss": 0.491, + "step": 2630 + }, + { + "epoch": 0.14699555828700728, + "grad_norm": 0.4779854416847229, + "learning_rate": 0.0009292357687135814, + "loss": 0.5133, + "step": 2631 + }, + { + "epoch": 0.14705142889068917, + "grad_norm": 0.6611591577529907, + "learning_rate": 0.0009292077543702376, + "loss": 0.4977, + "step": 2632 + }, + { + "epoch": 0.14710729949437104, + "grad_norm": 1.675977349281311, + "learning_rate": 0.0009291797400268937, + "loss": 0.566, + "step": 2633 + }, + { + "epoch": 0.1471631700980529, + "grad_norm": 0.43685588240623474, + "learning_rate": 0.00092915172568355, + "loss": 0.4829, + "step": 2634 + }, + { + "epoch": 0.14721904070173478, + "grad_norm": 0.4947856664657593, + "learning_rate": 0.0009291237113402063, + "loss": 0.4591, + "step": 2635 + }, + { + "epoch": 0.14727491130541664, + "grad_norm": 0.62579745054245, + "learning_rate": 0.0009290956969968624, + "loss": 0.4314, + "step": 2636 + }, + { + "epoch": 0.14733078190909854, + "grad_norm": 0.8555086255073547, + "learning_rate": 0.0009290676826535186, + "loss": 0.4797, + "step": 2637 + }, + { + "epoch": 0.1473866525127804, + "grad_norm": 0.578436553478241, + "learning_rate": 0.0009290396683101749, + "loss": 0.4637, + "step": 2638 + }, + { + "epoch": 0.14744252311646228, + "grad_norm": 0.6934192776679993, + "learning_rate": 0.0009290116539668311, + "loss": 0.4321, + "step": 2639 + }, + { + "epoch": 0.14749839372014414, + "grad_norm": 1.3540377616882324, + "learning_rate": 0.0009289836396234872, + "loss": 0.5081, + "step": 2640 + }, + { + "epoch": 0.147554264323826, + "grad_norm": 2.0626964569091797, + "learning_rate": 0.0009289556252801434, + "loss": 0.6046, + "step": 2641 + }, + { + "epoch": 0.14761013492750788, + "grad_norm": 0.5651251673698425, + "learning_rate": 0.0009289276109367997, + "loss": 0.4087, + "step": 2642 + }, + { + "epoch": 0.14766600553118978, + "grad_norm": 0.797877848148346, + "learning_rate": 0.0009288995965934559, + "loss": 0.6021, + "step": 2643 + }, + { + "epoch": 0.14772187613487164, + "grad_norm": 2.3150289058685303, + "learning_rate": 0.000928871582250112, + "loss": 0.4766, + "step": 2644 + }, + { + "epoch": 0.1477777467385535, + "grad_norm": 0.49206408858299255, + "learning_rate": 0.0009288435679067683, + "loss": 0.3943, + "step": 2645 + }, + { + "epoch": 0.14783361734223538, + "grad_norm": 1.3967119455337524, + "learning_rate": 0.0009288155535634245, + "loss": 0.5349, + "step": 2646 + }, + { + "epoch": 0.14788948794591725, + "grad_norm": 0.5251957774162292, + "learning_rate": 0.0009287875392200807, + "loss": 0.4638, + "step": 2647 + }, + { + "epoch": 0.14794535854959912, + "grad_norm": 0.6593282222747803, + "learning_rate": 0.0009287595248767368, + "loss": 0.4776, + "step": 2648 + }, + { + "epoch": 0.148001229153281, + "grad_norm": 0.6378556489944458, + "learning_rate": 0.0009287315105333931, + "loss": 0.3735, + "step": 2649 + }, + { + "epoch": 0.14805709975696288, + "grad_norm": 1.079982042312622, + "learning_rate": 0.0009287034961900494, + "loss": 0.5511, + "step": 2650 + }, + { + "epoch": 0.14811297036064475, + "grad_norm": 0.7436265349388123, + "learning_rate": 0.0009286754818467056, + "loss": 0.531, + "step": 2651 + }, + { + "epoch": 0.14816884096432661, + "grad_norm": 0.5329391956329346, + "learning_rate": 0.0009286474675033618, + "loss": 0.4743, + "step": 2652 + }, + { + "epoch": 0.14822471156800848, + "grad_norm": 3.962742805480957, + "learning_rate": 0.000928619453160018, + "loss": 0.5249, + "step": 2653 + }, + { + "epoch": 0.14828058217169038, + "grad_norm": 0.5119608044624329, + "learning_rate": 0.0009285914388166742, + "loss": 0.5689, + "step": 2654 + }, + { + "epoch": 0.14833645277537225, + "grad_norm": 1.3363020420074463, + "learning_rate": 0.0009285634244733303, + "loss": 0.4465, + "step": 2655 + }, + { + "epoch": 0.14839232337905411, + "grad_norm": 0.5160219669342041, + "learning_rate": 0.0009285354101299866, + "loss": 0.5154, + "step": 2656 + }, + { + "epoch": 0.14844819398273598, + "grad_norm": 0.7308619022369385, + "learning_rate": 0.0009285073957866428, + "loss": 0.5896, + "step": 2657 + }, + { + "epoch": 0.14850406458641785, + "grad_norm": 0.5925816893577576, + "learning_rate": 0.000928479381443299, + "loss": 0.4772, + "step": 2658 + }, + { + "epoch": 0.14855993519009972, + "grad_norm": 0.5689205527305603, + "learning_rate": 0.0009284513670999551, + "loss": 0.5026, + "step": 2659 + }, + { + "epoch": 0.1486158057937816, + "grad_norm": 0.880480945110321, + "learning_rate": 0.0009284233527566114, + "loss": 0.4844, + "step": 2660 + }, + { + "epoch": 0.14867167639746348, + "grad_norm": 0.46987292170524597, + "learning_rate": 0.0009283953384132676, + "loss": 0.4757, + "step": 2661 + }, + { + "epoch": 0.14872754700114535, + "grad_norm": 0.5107002854347229, + "learning_rate": 0.0009283673240699238, + "loss": 0.4553, + "step": 2662 + }, + { + "epoch": 0.14878341760482722, + "grad_norm": 0.5597710609436035, + "learning_rate": 0.00092833930972658, + "loss": 0.5218, + "step": 2663 + }, + { + "epoch": 0.14883928820850909, + "grad_norm": 0.776808500289917, + "learning_rate": 0.0009283112953832362, + "loss": 0.6204, + "step": 2664 + }, + { + "epoch": 0.14889515881219095, + "grad_norm": 2.6580770015716553, + "learning_rate": 0.0009282832810398924, + "loss": 0.5437, + "step": 2665 + }, + { + "epoch": 0.14895102941587285, + "grad_norm": 0.6379950642585754, + "learning_rate": 0.0009282552666965486, + "loss": 0.4281, + "step": 2666 + }, + { + "epoch": 0.14900690001955472, + "grad_norm": 0.7101867198944092, + "learning_rate": 0.0009282272523532048, + "loss": 0.6826, + "step": 2667 + }, + { + "epoch": 0.14906277062323658, + "grad_norm": 0.6609296202659607, + "learning_rate": 0.0009281992380098611, + "loss": 0.5769, + "step": 2668 + }, + { + "epoch": 0.14911864122691845, + "grad_norm": 0.45090430974960327, + "learning_rate": 0.0009281712236665173, + "loss": 0.4092, + "step": 2669 + }, + { + "epoch": 0.14917451183060032, + "grad_norm": 0.8072538375854492, + "learning_rate": 0.0009281432093231736, + "loss": 0.5536, + "step": 2670 + }, + { + "epoch": 0.14923038243428222, + "grad_norm": 1.241591453552246, + "learning_rate": 0.0009281151949798297, + "loss": 0.5488, + "step": 2671 + }, + { + "epoch": 0.14928625303796408, + "grad_norm": 0.5530011653900146, + "learning_rate": 0.0009280871806364859, + "loss": 0.581, + "step": 2672 + }, + { + "epoch": 0.14934212364164595, + "grad_norm": 2.6944785118103027, + "learning_rate": 0.0009280591662931421, + "loss": 0.5722, + "step": 2673 + }, + { + "epoch": 0.14939799424532782, + "grad_norm": 0.7433595657348633, + "learning_rate": 0.0009280311519497984, + "loss": 0.4941, + "step": 2674 + }, + { + "epoch": 0.1494538648490097, + "grad_norm": 0.6053360104560852, + "learning_rate": 0.0009280031376064545, + "loss": 0.5692, + "step": 2675 + }, + { + "epoch": 0.14950973545269156, + "grad_norm": 0.4846705496311188, + "learning_rate": 0.0009279751232631107, + "loss": 0.5402, + "step": 2676 + }, + { + "epoch": 0.14956560605637345, + "grad_norm": 0.736153244972229, + "learning_rate": 0.000927947108919767, + "loss": 0.5608, + "step": 2677 + }, + { + "epoch": 0.14962147666005532, + "grad_norm": 3.3773982524871826, + "learning_rate": 0.0009279190945764231, + "loss": 0.4538, + "step": 2678 + }, + { + "epoch": 0.1496773472637372, + "grad_norm": 0.7538591027259827, + "learning_rate": 0.0009278910802330793, + "loss": 0.4631, + "step": 2679 + }, + { + "epoch": 0.14973321786741906, + "grad_norm": 0.6352667212486267, + "learning_rate": 0.0009278630658897355, + "loss": 0.45, + "step": 2680 + }, + { + "epoch": 0.14978908847110092, + "grad_norm": 0.6433882713317871, + "learning_rate": 0.0009278350515463918, + "loss": 0.6247, + "step": 2681 + }, + { + "epoch": 0.1498449590747828, + "grad_norm": 0.528546154499054, + "learning_rate": 0.0009278070372030479, + "loss": 0.4334, + "step": 2682 + }, + { + "epoch": 0.1499008296784647, + "grad_norm": 0.6029179096221924, + "learning_rate": 0.0009277790228597041, + "loss": 0.4777, + "step": 2683 + }, + { + "epoch": 0.14995670028214655, + "grad_norm": 0.7406453490257263, + "learning_rate": 0.0009277510085163605, + "loss": 0.4624, + "step": 2684 + }, + { + "epoch": 0.15001257088582842, + "grad_norm": 5.882260799407959, + "learning_rate": 0.0009277229941730167, + "loss": 0.681, + "step": 2685 + }, + { + "epoch": 0.1500684414895103, + "grad_norm": 0.6008306741714478, + "learning_rate": 0.0009276949798296728, + "loss": 0.475, + "step": 2686 + }, + { + "epoch": 0.15012431209319216, + "grad_norm": 0.5073904991149902, + "learning_rate": 0.000927666965486329, + "loss": 0.638, + "step": 2687 + }, + { + "epoch": 0.15018018269687403, + "grad_norm": 0.7879272699356079, + "learning_rate": 0.0009276389511429853, + "loss": 0.4674, + "step": 2688 + }, + { + "epoch": 0.15023605330055592, + "grad_norm": 0.9717096090316772, + "learning_rate": 0.0009276109367996415, + "loss": 0.4335, + "step": 2689 + }, + { + "epoch": 0.1502919239042378, + "grad_norm": 0.6255761384963989, + "learning_rate": 0.0009275829224562976, + "loss": 0.4537, + "step": 2690 + }, + { + "epoch": 0.15034779450791966, + "grad_norm": 0.6170027852058411, + "learning_rate": 0.0009275549081129539, + "loss": 0.4667, + "step": 2691 + }, + { + "epoch": 0.15040366511160153, + "grad_norm": 0.4852813482284546, + "learning_rate": 0.0009275268937696101, + "loss": 0.4539, + "step": 2692 + }, + { + "epoch": 0.1504595357152834, + "grad_norm": 1.1137597560882568, + "learning_rate": 0.0009274988794262663, + "loss": 0.4634, + "step": 2693 + }, + { + "epoch": 0.1505154063189653, + "grad_norm": 0.5261353850364685, + "learning_rate": 0.0009274708650829224, + "loss": 0.5577, + "step": 2694 + }, + { + "epoch": 0.15057127692264716, + "grad_norm": 1.6274616718292236, + "learning_rate": 0.0009274428507395787, + "loss": 0.682, + "step": 2695 + }, + { + "epoch": 0.15062714752632902, + "grad_norm": 0.4199308753013611, + "learning_rate": 0.0009274148363962349, + "loss": 0.4666, + "step": 2696 + }, + { + "epoch": 0.1506830181300109, + "grad_norm": 0.6567255854606628, + "learning_rate": 0.0009273868220528911, + "loss": 0.5256, + "step": 2697 + }, + { + "epoch": 0.15073888873369276, + "grad_norm": 0.7289178371429443, + "learning_rate": 0.0009273588077095472, + "loss": 0.495, + "step": 2698 + }, + { + "epoch": 0.15079475933737463, + "grad_norm": 0.9651411771774292, + "learning_rate": 0.0009273307933662035, + "loss": 0.4971, + "step": 2699 + }, + { + "epoch": 0.15085062994105652, + "grad_norm": 0.5107855796813965, + "learning_rate": 0.0009273027790228597, + "loss": 0.452, + "step": 2700 + }, + { + "epoch": 0.1509065005447384, + "grad_norm": 0.4463571012020111, + "learning_rate": 0.0009272747646795158, + "loss": 0.4906, + "step": 2701 + }, + { + "epoch": 0.15096237114842026, + "grad_norm": 0.591081976890564, + "learning_rate": 0.0009272467503361722, + "loss": 0.414, + "step": 2702 + }, + { + "epoch": 0.15101824175210213, + "grad_norm": 0.620652973651886, + "learning_rate": 0.0009272187359928284, + "loss": 0.4842, + "step": 2703 + }, + { + "epoch": 0.151074112355784, + "grad_norm": 1.0185794830322266, + "learning_rate": 0.0009271907216494846, + "loss": 0.4985, + "step": 2704 + }, + { + "epoch": 0.15112998295946586, + "grad_norm": 0.9040365219116211, + "learning_rate": 0.0009271627073061407, + "loss": 0.8006, + "step": 2705 + }, + { + "epoch": 0.15118585356314776, + "grad_norm": 0.389040470123291, + "learning_rate": 0.000927134692962797, + "loss": 0.362, + "step": 2706 + }, + { + "epoch": 0.15124172416682963, + "grad_norm": 0.4721764624118805, + "learning_rate": 0.0009271066786194532, + "loss": 0.4367, + "step": 2707 + }, + { + "epoch": 0.1512975947705115, + "grad_norm": 0.6150627136230469, + "learning_rate": 0.0009270786642761094, + "loss": 0.5426, + "step": 2708 + }, + { + "epoch": 0.15135346537419336, + "grad_norm": 0.4268595576286316, + "learning_rate": 0.0009270506499327656, + "loss": 0.4016, + "step": 2709 + }, + { + "epoch": 0.15140933597787523, + "grad_norm": 0.4047594368457794, + "learning_rate": 0.0009270226355894218, + "loss": 0.4694, + "step": 2710 + }, + { + "epoch": 0.15146520658155713, + "grad_norm": 1.0212794542312622, + "learning_rate": 0.000926994621246078, + "loss": 0.4807, + "step": 2711 + }, + { + "epoch": 0.151521077185239, + "grad_norm": 0.9949529767036438, + "learning_rate": 0.0009269666069027342, + "loss": 0.5263, + "step": 2712 + }, + { + "epoch": 0.15157694778892086, + "grad_norm": 7.446710109710693, + "learning_rate": 0.0009269385925593904, + "loss": 0.6107, + "step": 2713 + }, + { + "epoch": 0.15163281839260273, + "grad_norm": 0.9215781688690186, + "learning_rate": 0.0009269105782160466, + "loss": 0.5036, + "step": 2714 + }, + { + "epoch": 0.1516886889962846, + "grad_norm": 0.6141756772994995, + "learning_rate": 0.0009268825638727028, + "loss": 0.5241, + "step": 2715 + }, + { + "epoch": 0.15174455959996647, + "grad_norm": 0.7401598691940308, + "learning_rate": 0.0009268545495293592, + "loss": 0.5225, + "step": 2716 + }, + { + "epoch": 0.15180043020364836, + "grad_norm": 0.7860554456710815, + "learning_rate": 0.0009268265351860152, + "loss": 0.5127, + "step": 2717 + }, + { + "epoch": 0.15185630080733023, + "grad_norm": 0.5747883915901184, + "learning_rate": 0.0009267985208426715, + "loss": 0.4696, + "step": 2718 + }, + { + "epoch": 0.1519121714110121, + "grad_norm": 37.808902740478516, + "learning_rate": 0.0009267705064993277, + "loss": 0.4393, + "step": 2719 + }, + { + "epoch": 0.15196804201469397, + "grad_norm": 1.6135320663452148, + "learning_rate": 0.0009267424921559839, + "loss": 0.4653, + "step": 2720 + }, + { + "epoch": 0.15202391261837583, + "grad_norm": 1.2312575578689575, + "learning_rate": 0.0009267144778126401, + "loss": 0.3821, + "step": 2721 + }, + { + "epoch": 0.1520797832220577, + "grad_norm": 0.9121775031089783, + "learning_rate": 0.0009266864634692963, + "loss": 0.5697, + "step": 2722 + }, + { + "epoch": 0.1521356538257396, + "grad_norm": 1.3149700164794922, + "learning_rate": 0.0009266584491259526, + "loss": 0.4978, + "step": 2723 + }, + { + "epoch": 0.15219152442942147, + "grad_norm": 0.760952353477478, + "learning_rate": 0.0009266304347826087, + "loss": 0.5931, + "step": 2724 + }, + { + "epoch": 0.15224739503310333, + "grad_norm": 0.6883904337882996, + "learning_rate": 0.0009266024204392649, + "loss": 0.4967, + "step": 2725 + }, + { + "epoch": 0.1523032656367852, + "grad_norm": 2.355473518371582, + "learning_rate": 0.0009265744060959211, + "loss": 0.4827, + "step": 2726 + }, + { + "epoch": 0.15235913624046707, + "grad_norm": 0.6225693225860596, + "learning_rate": 0.0009265463917525774, + "loss": 0.4938, + "step": 2727 + }, + { + "epoch": 0.15241500684414896, + "grad_norm": 0.7114613056182861, + "learning_rate": 0.0009265183774092335, + "loss": 0.5497, + "step": 2728 + }, + { + "epoch": 0.15247087744783083, + "grad_norm": 0.5098021626472473, + "learning_rate": 0.0009264903630658897, + "loss": 0.4728, + "step": 2729 + }, + { + "epoch": 0.1525267480515127, + "grad_norm": 0.9305760860443115, + "learning_rate": 0.0009264623487225459, + "loss": 0.5192, + "step": 2730 + }, + { + "epoch": 0.15258261865519457, + "grad_norm": 0.5959597826004028, + "learning_rate": 0.0009264343343792022, + "loss": 0.48, + "step": 2731 + }, + { + "epoch": 0.15263848925887644, + "grad_norm": 0.5001663565635681, + "learning_rate": 0.0009264063200358583, + "loss": 0.4772, + "step": 2732 + }, + { + "epoch": 0.1526943598625583, + "grad_norm": 0.5440109968185425, + "learning_rate": 0.0009263783056925145, + "loss": 0.38, + "step": 2733 + }, + { + "epoch": 0.1527502304662402, + "grad_norm": 0.6639846563339233, + "learning_rate": 0.0009263502913491709, + "loss": 0.5472, + "step": 2734 + }, + { + "epoch": 0.15280610106992207, + "grad_norm": 2.1323044300079346, + "learning_rate": 0.0009263222770058271, + "loss": 0.4941, + "step": 2735 + }, + { + "epoch": 0.15286197167360394, + "grad_norm": 0.595000147819519, + "learning_rate": 0.0009262942626624832, + "loss": 0.4087, + "step": 2736 + }, + { + "epoch": 0.1529178422772858, + "grad_norm": 0.6552373170852661, + "learning_rate": 0.0009262662483191394, + "loss": 0.6999, + "step": 2737 + }, + { + "epoch": 0.15297371288096767, + "grad_norm": 0.5951988101005554, + "learning_rate": 0.0009262382339757957, + "loss": 0.3948, + "step": 2738 + }, + { + "epoch": 0.15302958348464954, + "grad_norm": 1.5530757904052734, + "learning_rate": 0.0009262102196324519, + "loss": 0.3703, + "step": 2739 + }, + { + "epoch": 0.15308545408833144, + "grad_norm": 0.579536497592926, + "learning_rate": 0.000926182205289108, + "loss": 0.4334, + "step": 2740 + }, + { + "epoch": 0.1531413246920133, + "grad_norm": 0.6887215375900269, + "learning_rate": 0.0009261541909457643, + "loss": 0.6171, + "step": 2741 + }, + { + "epoch": 0.15319719529569517, + "grad_norm": 0.49227288365364075, + "learning_rate": 0.0009261261766024205, + "loss": 0.4296, + "step": 2742 + }, + { + "epoch": 0.15325306589937704, + "grad_norm": 0.8016662001609802, + "learning_rate": 0.0009260981622590766, + "loss": 0.3828, + "step": 2743 + }, + { + "epoch": 0.1533089365030589, + "grad_norm": 0.818570613861084, + "learning_rate": 0.0009260701479157328, + "loss": 0.5101, + "step": 2744 + }, + { + "epoch": 0.15336480710674077, + "grad_norm": 0.5985581278800964, + "learning_rate": 0.0009260421335723891, + "loss": 0.5611, + "step": 2745 + }, + { + "epoch": 0.15342067771042267, + "grad_norm": 0.6170482635498047, + "learning_rate": 0.0009260141192290453, + "loss": 0.5069, + "step": 2746 + }, + { + "epoch": 0.15347654831410454, + "grad_norm": 0.6525503993034363, + "learning_rate": 0.0009259861048857014, + "loss": 0.4093, + "step": 2747 + }, + { + "epoch": 0.1535324189177864, + "grad_norm": 0.8485221266746521, + "learning_rate": 0.0009259580905423577, + "loss": 0.4821, + "step": 2748 + }, + { + "epoch": 0.15358828952146827, + "grad_norm": 0.6231567859649658, + "learning_rate": 0.0009259300761990139, + "loss": 0.6086, + "step": 2749 + }, + { + "epoch": 0.15364416012515014, + "grad_norm": 1.0230515003204346, + "learning_rate": 0.0009259020618556701, + "loss": 0.6217, + "step": 2750 + }, + { + "epoch": 0.15370003072883204, + "grad_norm": 0.8746201395988464, + "learning_rate": 0.0009258740475123262, + "loss": 0.6499, + "step": 2751 + }, + { + "epoch": 0.1537559013325139, + "grad_norm": 2.110921859741211, + "learning_rate": 0.0009258460331689826, + "loss": 0.4557, + "step": 2752 + }, + { + "epoch": 0.15381177193619577, + "grad_norm": 0.6953683495521545, + "learning_rate": 0.0009258180188256388, + "loss": 0.5786, + "step": 2753 + }, + { + "epoch": 0.15386764253987764, + "grad_norm": 0.5209238529205322, + "learning_rate": 0.000925790004482295, + "loss": 0.4906, + "step": 2754 + }, + { + "epoch": 0.1539235131435595, + "grad_norm": 0.9265364408493042, + "learning_rate": 0.0009257619901389512, + "loss": 0.562, + "step": 2755 + }, + { + "epoch": 0.15397938374724138, + "grad_norm": 1.9063770771026611, + "learning_rate": 0.0009257339757956074, + "loss": 0.6209, + "step": 2756 + }, + { + "epoch": 0.15403525435092327, + "grad_norm": 0.7473033666610718, + "learning_rate": 0.0009257059614522636, + "loss": 0.4785, + "step": 2757 + }, + { + "epoch": 0.15409112495460514, + "grad_norm": 0.6800265312194824, + "learning_rate": 0.0009256779471089198, + "loss": 0.4952, + "step": 2758 + }, + { + "epoch": 0.154146995558287, + "grad_norm": 0.926862359046936, + "learning_rate": 0.000925649932765576, + "loss": 0.5624, + "step": 2759 + }, + { + "epoch": 0.15420286616196888, + "grad_norm": 0.5958459377288818, + "learning_rate": 0.0009256219184222322, + "loss": 0.6017, + "step": 2760 + }, + { + "epoch": 0.15425873676565074, + "grad_norm": 0.5426074862480164, + "learning_rate": 0.0009255939040788884, + "loss": 0.4358, + "step": 2761 + }, + { + "epoch": 0.1543146073693326, + "grad_norm": 1.3651551008224487, + "learning_rate": 0.0009255658897355445, + "loss": 0.4816, + "step": 2762 + }, + { + "epoch": 0.1543704779730145, + "grad_norm": 0.9924950003623962, + "learning_rate": 0.0009255378753922008, + "loss": 0.5663, + "step": 2763 + }, + { + "epoch": 0.15442634857669638, + "grad_norm": 0.5120459198951721, + "learning_rate": 0.000925509861048857, + "loss": 0.3703, + "step": 2764 + }, + { + "epoch": 0.15448221918037824, + "grad_norm": 0.9256970286369324, + "learning_rate": 0.0009254818467055132, + "loss": 0.4647, + "step": 2765 + }, + { + "epoch": 0.1545380897840601, + "grad_norm": 0.647300660610199, + "learning_rate": 0.0009254538323621694, + "loss": 0.6205, + "step": 2766 + }, + { + "epoch": 0.15459396038774198, + "grad_norm": 0.8255757093429565, + "learning_rate": 0.0009254258180188256, + "loss": 0.6183, + "step": 2767 + }, + { + "epoch": 0.15464983099142388, + "grad_norm": 1.6922404766082764, + "learning_rate": 0.0009253978036754819, + "loss": 0.457, + "step": 2768 + }, + { + "epoch": 0.15470570159510574, + "grad_norm": 1.022546648979187, + "learning_rate": 0.0009253697893321381, + "loss": 0.4911, + "step": 2769 + }, + { + "epoch": 0.1547615721987876, + "grad_norm": 0.600975751876831, + "learning_rate": 0.0009253417749887943, + "loss": 0.5545, + "step": 2770 + }, + { + "epoch": 0.15481744280246948, + "grad_norm": 0.4750097990036011, + "learning_rate": 0.0009253137606454505, + "loss": 0.4262, + "step": 2771 + }, + { + "epoch": 0.15487331340615135, + "grad_norm": 0.8419404029846191, + "learning_rate": 0.0009252857463021067, + "loss": 0.3668, + "step": 2772 + }, + { + "epoch": 0.15492918400983322, + "grad_norm": 1.0870410203933716, + "learning_rate": 0.000925257731958763, + "loss": 0.4559, + "step": 2773 + }, + { + "epoch": 0.1549850546135151, + "grad_norm": 1.0200614929199219, + "learning_rate": 0.0009252297176154191, + "loss": 0.567, + "step": 2774 + }, + { + "epoch": 0.15504092521719698, + "grad_norm": 0.6524651050567627, + "learning_rate": 0.0009252017032720753, + "loss": 0.4729, + "step": 2775 + }, + { + "epoch": 0.15509679582087885, + "grad_norm": 0.8620721101760864, + "learning_rate": 0.0009251736889287315, + "loss": 0.4669, + "step": 2776 + }, + { + "epoch": 0.15515266642456071, + "grad_norm": 1.0041366815567017, + "learning_rate": 0.0009251456745853878, + "loss": 0.3748, + "step": 2777 + }, + { + "epoch": 0.15520853702824258, + "grad_norm": 1.2549293041229248, + "learning_rate": 0.0009251176602420439, + "loss": 0.4932, + "step": 2778 + }, + { + "epoch": 0.15526440763192445, + "grad_norm": 0.6902197003364563, + "learning_rate": 0.0009250896458987001, + "loss": 0.4966, + "step": 2779 + }, + { + "epoch": 0.15532027823560635, + "grad_norm": 1.146175503730774, + "learning_rate": 0.0009250616315553564, + "loss": 0.7343, + "step": 2780 + }, + { + "epoch": 0.15537614883928821, + "grad_norm": 0.6019544005393982, + "learning_rate": 0.0009250336172120126, + "loss": 0.6385, + "step": 2781 + }, + { + "epoch": 0.15543201944297008, + "grad_norm": 5.045692443847656, + "learning_rate": 0.0009250056028686687, + "loss": 0.5718, + "step": 2782 + }, + { + "epoch": 0.15548789004665195, + "grad_norm": 0.9089503288269043, + "learning_rate": 0.0009249775885253249, + "loss": 0.4232, + "step": 2783 + }, + { + "epoch": 0.15554376065033382, + "grad_norm": 0.9016115069389343, + "learning_rate": 0.0009249495741819813, + "loss": 0.7699, + "step": 2784 + }, + { + "epoch": 0.1555996312540157, + "grad_norm": 3.453972578048706, + "learning_rate": 0.0009249215598386374, + "loss": 0.4729, + "step": 2785 + }, + { + "epoch": 0.15565550185769758, + "grad_norm": 15.58098030090332, + "learning_rate": 0.0009248935454952936, + "loss": 0.5069, + "step": 2786 + }, + { + "epoch": 0.15571137246137945, + "grad_norm": 2.801409959793091, + "learning_rate": 0.0009248655311519499, + "loss": 0.4679, + "step": 2787 + }, + { + "epoch": 0.15576724306506132, + "grad_norm": 7.833441257476807, + "learning_rate": 0.0009248375168086061, + "loss": 0.5149, + "step": 2788 + }, + { + "epoch": 0.15582311366874319, + "grad_norm": 1.0044400691986084, + "learning_rate": 0.0009248095024652622, + "loss": 0.5874, + "step": 2789 + }, + { + "epoch": 0.15587898427242505, + "grad_norm": 0.5010474324226379, + "learning_rate": 0.0009247814881219184, + "loss": 0.4123, + "step": 2790 + }, + { + "epoch": 0.15593485487610695, + "grad_norm": 0.6666485071182251, + "learning_rate": 0.0009247534737785747, + "loss": 0.4476, + "step": 2791 + }, + { + "epoch": 0.15599072547978882, + "grad_norm": 0.5263987183570862, + "learning_rate": 0.0009247254594352309, + "loss": 0.5719, + "step": 2792 + }, + { + "epoch": 0.15604659608347068, + "grad_norm": 0.776013970375061, + "learning_rate": 0.000924697445091887, + "loss": 0.4736, + "step": 2793 + }, + { + "epoch": 0.15610246668715255, + "grad_norm": 0.6157413125038147, + "learning_rate": 0.0009246694307485433, + "loss": 0.3974, + "step": 2794 + }, + { + "epoch": 0.15615833729083442, + "grad_norm": 0.5957254767417908, + "learning_rate": 0.0009246414164051995, + "loss": 0.4849, + "step": 2795 + }, + { + "epoch": 0.1562142078945163, + "grad_norm": 0.9577169418334961, + "learning_rate": 0.0009246134020618557, + "loss": 0.5727, + "step": 2796 + }, + { + "epoch": 0.15627007849819818, + "grad_norm": 0.7453992962837219, + "learning_rate": 0.0009245853877185118, + "loss": 0.4651, + "step": 2797 + }, + { + "epoch": 0.15632594910188005, + "grad_norm": 0.7105717658996582, + "learning_rate": 0.0009245573733751681, + "loss": 0.4247, + "step": 2798 + }, + { + "epoch": 0.15638181970556192, + "grad_norm": 0.5346956253051758, + "learning_rate": 0.0009245293590318243, + "loss": 0.4306, + "step": 2799 + }, + { + "epoch": 0.1564376903092438, + "grad_norm": 0.563231885433197, + "learning_rate": 0.0009245013446884805, + "loss": 0.5177, + "step": 2800 + }, + { + "epoch": 0.15649356091292566, + "grad_norm": 0.5116049647331238, + "learning_rate": 0.0009244733303451366, + "loss": 0.4714, + "step": 2801 + }, + { + "epoch": 0.15654943151660752, + "grad_norm": 0.7895889282226562, + "learning_rate": 0.000924445316001793, + "loss": 0.4216, + "step": 2802 + }, + { + "epoch": 0.15660530212028942, + "grad_norm": 0.4833142161369324, + "learning_rate": 0.0009244173016584492, + "loss": 0.4022, + "step": 2803 + }, + { + "epoch": 0.1566611727239713, + "grad_norm": 0.5871998071670532, + "learning_rate": 0.0009243892873151053, + "loss": 0.6082, + "step": 2804 + }, + { + "epoch": 0.15671704332765315, + "grad_norm": 0.48700323700904846, + "learning_rate": 0.0009243612729717616, + "loss": 0.4128, + "step": 2805 + }, + { + "epoch": 0.15677291393133502, + "grad_norm": 0.6454588174819946, + "learning_rate": 0.0009243332586284178, + "loss": 0.52, + "step": 2806 + }, + { + "epoch": 0.1568287845350169, + "grad_norm": 0.5520349144935608, + "learning_rate": 0.000924305244285074, + "loss": 0.4532, + "step": 2807 + }, + { + "epoch": 0.1568846551386988, + "grad_norm": 0.5029950737953186, + "learning_rate": 0.0009242772299417301, + "loss": 0.4538, + "step": 2808 + }, + { + "epoch": 0.15694052574238065, + "grad_norm": 1.060746431350708, + "learning_rate": 0.0009242492155983864, + "loss": 0.5263, + "step": 2809 + }, + { + "epoch": 0.15699639634606252, + "grad_norm": 0.5071749091148376, + "learning_rate": 0.0009242212012550426, + "loss": 0.4754, + "step": 2810 + }, + { + "epoch": 0.1570522669497444, + "grad_norm": 1.2927393913269043, + "learning_rate": 0.0009241931869116988, + "loss": 0.4938, + "step": 2811 + }, + { + "epoch": 0.15710813755342626, + "grad_norm": 3.2002995014190674, + "learning_rate": 0.000924165172568355, + "loss": 0.4895, + "step": 2812 + }, + { + "epoch": 0.15716400815710813, + "grad_norm": 0.5589813590049744, + "learning_rate": 0.0009241371582250112, + "loss": 0.5905, + "step": 2813 + }, + { + "epoch": 0.15721987876079002, + "grad_norm": 1.668705701828003, + "learning_rate": 0.0009241091438816674, + "loss": 0.4918, + "step": 2814 + }, + { + "epoch": 0.1572757493644719, + "grad_norm": 0.6944273114204407, + "learning_rate": 0.0009240811295383236, + "loss": 0.4496, + "step": 2815 + }, + { + "epoch": 0.15733161996815376, + "grad_norm": 2.9336435794830322, + "learning_rate": 0.0009240531151949798, + "loss": 0.4555, + "step": 2816 + }, + { + "epoch": 0.15738749057183563, + "grad_norm": 0.5915203094482422, + "learning_rate": 0.000924025100851636, + "loss": 0.4705, + "step": 2817 + }, + { + "epoch": 0.1574433611755175, + "grad_norm": 0.47126927971839905, + "learning_rate": 0.0009239970865082923, + "loss": 0.4714, + "step": 2818 + }, + { + "epoch": 0.15749923177919936, + "grad_norm": 0.5103738307952881, + "learning_rate": 0.0009239690721649486, + "loss": 0.4263, + "step": 2819 + }, + { + "epoch": 0.15755510238288126, + "grad_norm": 0.7324618101119995, + "learning_rate": 0.0009239410578216047, + "loss": 0.4608, + "step": 2820 + }, + { + "epoch": 0.15761097298656312, + "grad_norm": 4.666616439819336, + "learning_rate": 0.0009239130434782609, + "loss": 0.4623, + "step": 2821 + }, + { + "epoch": 0.157666843590245, + "grad_norm": 0.5811932682991028, + "learning_rate": 0.0009238850291349171, + "loss": 0.5072, + "step": 2822 + }, + { + "epoch": 0.15772271419392686, + "grad_norm": 0.6854141354560852, + "learning_rate": 0.0009238570147915734, + "loss": 0.4934, + "step": 2823 + }, + { + "epoch": 0.15777858479760873, + "grad_norm": 0.6018785238265991, + "learning_rate": 0.0009238290004482295, + "loss": 0.5519, + "step": 2824 + }, + { + "epoch": 0.15783445540129062, + "grad_norm": 0.4834415912628174, + "learning_rate": 0.0009238009861048857, + "loss": 0.4873, + "step": 2825 + }, + { + "epoch": 0.1578903260049725, + "grad_norm": 0.5917794108390808, + "learning_rate": 0.000923772971761542, + "loss": 0.475, + "step": 2826 + }, + { + "epoch": 0.15794619660865436, + "grad_norm": 0.5333681702613831, + "learning_rate": 0.0009237449574181981, + "loss": 0.5243, + "step": 2827 + }, + { + "epoch": 0.15800206721233623, + "grad_norm": 7.103582382202148, + "learning_rate": 0.0009237169430748543, + "loss": 0.5368, + "step": 2828 + }, + { + "epoch": 0.1580579378160181, + "grad_norm": 0.484635591506958, + "learning_rate": 0.0009236889287315105, + "loss": 0.48, + "step": 2829 + }, + { + "epoch": 0.15811380841969996, + "grad_norm": 0.8063584566116333, + "learning_rate": 0.0009236609143881668, + "loss": 0.4955, + "step": 2830 + }, + { + "epoch": 0.15816967902338186, + "grad_norm": 0.9674395322799683, + "learning_rate": 0.0009236329000448229, + "loss": 0.5505, + "step": 2831 + }, + { + "epoch": 0.15822554962706373, + "grad_norm": 0.7566810846328735, + "learning_rate": 0.0009236048857014791, + "loss": 0.5393, + "step": 2832 + }, + { + "epoch": 0.1582814202307456, + "grad_norm": 0.7948341965675354, + "learning_rate": 0.0009235768713581354, + "loss": 0.6386, + "step": 2833 + }, + { + "epoch": 0.15833729083442746, + "grad_norm": 0.604433536529541, + "learning_rate": 0.0009235488570147917, + "loss": 0.5981, + "step": 2834 + }, + { + "epoch": 0.15839316143810933, + "grad_norm": 0.7601238489151001, + "learning_rate": 0.0009235208426714478, + "loss": 0.4586, + "step": 2835 + }, + { + "epoch": 0.1584490320417912, + "grad_norm": 5.308541297912598, + "learning_rate": 0.000923492828328104, + "loss": 0.457, + "step": 2836 + }, + { + "epoch": 0.1585049026454731, + "grad_norm": 11.049656867980957, + "learning_rate": 0.0009234648139847603, + "loss": 0.4678, + "step": 2837 + }, + { + "epoch": 0.15856077324915496, + "grad_norm": 0.7719625234603882, + "learning_rate": 0.0009234367996414165, + "loss": 0.6728, + "step": 2838 + }, + { + "epoch": 0.15861664385283683, + "grad_norm": 0.43567532300949097, + "learning_rate": 0.0009234087852980726, + "loss": 0.464, + "step": 2839 + }, + { + "epoch": 0.1586725144565187, + "grad_norm": 0.6049202084541321, + "learning_rate": 0.0009233807709547288, + "loss": 0.5113, + "step": 2840 + }, + { + "epoch": 0.15872838506020057, + "grad_norm": 0.8588513135910034, + "learning_rate": 0.0009233527566113851, + "loss": 0.4825, + "step": 2841 + }, + { + "epoch": 0.15878425566388246, + "grad_norm": 0.6631292104721069, + "learning_rate": 0.0009233247422680413, + "loss": 0.5165, + "step": 2842 + }, + { + "epoch": 0.15884012626756433, + "grad_norm": 0.9781851172447205, + "learning_rate": 0.0009232967279246974, + "loss": 0.6458, + "step": 2843 + }, + { + "epoch": 0.1588959968712462, + "grad_norm": 1.1852104663848877, + "learning_rate": 0.0009232687135813537, + "loss": 0.5544, + "step": 2844 + }, + { + "epoch": 0.15895186747492807, + "grad_norm": 0.5930797457695007, + "learning_rate": 0.0009232406992380099, + "loss": 0.4591, + "step": 2845 + }, + { + "epoch": 0.15900773807860993, + "grad_norm": 0.9984377026557922, + "learning_rate": 0.000923212684894666, + "loss": 0.4558, + "step": 2846 + }, + { + "epoch": 0.1590636086822918, + "grad_norm": 0.8516427278518677, + "learning_rate": 0.0009231846705513222, + "loss": 0.6173, + "step": 2847 + }, + { + "epoch": 0.1591194792859737, + "grad_norm": 0.6722419857978821, + "learning_rate": 0.0009231566562079785, + "loss": 0.5307, + "step": 2848 + }, + { + "epoch": 0.15917534988965557, + "grad_norm": 0.681698203086853, + "learning_rate": 0.0009231286418646347, + "loss": 0.5011, + "step": 2849 + }, + { + "epoch": 0.15923122049333743, + "grad_norm": 0.9817609786987305, + "learning_rate": 0.0009231006275212908, + "loss": 0.6298, + "step": 2850 + }, + { + "epoch": 0.1592870910970193, + "grad_norm": 0.491059273481369, + "learning_rate": 0.0009230726131779472, + "loss": 0.4779, + "step": 2851 + }, + { + "epoch": 0.15934296170070117, + "grad_norm": 0.5822526216506958, + "learning_rate": 0.0009230445988346034, + "loss": 0.4867, + "step": 2852 + }, + { + "epoch": 0.15939883230438304, + "grad_norm": 1.0650060176849365, + "learning_rate": 0.0009230165844912596, + "loss": 0.5791, + "step": 2853 + }, + { + "epoch": 0.15945470290806493, + "grad_norm": 0.9338141083717346, + "learning_rate": 0.0009229885701479157, + "loss": 0.4407, + "step": 2854 + }, + { + "epoch": 0.1595105735117468, + "grad_norm": 3.5389037132263184, + "learning_rate": 0.000922960555804572, + "loss": 0.6493, + "step": 2855 + }, + { + "epoch": 0.15956644411542867, + "grad_norm": 0.6928831338882446, + "learning_rate": 0.0009229325414612282, + "loss": 0.4612, + "step": 2856 + }, + { + "epoch": 0.15962231471911054, + "grad_norm": 2.700698137283325, + "learning_rate": 0.0009229045271178844, + "loss": 0.587, + "step": 2857 + }, + { + "epoch": 0.1596781853227924, + "grad_norm": 0.8587356209754944, + "learning_rate": 0.0009228765127745406, + "loss": 0.4671, + "step": 2858 + }, + { + "epoch": 0.15973405592647427, + "grad_norm": 2.7162528038024902, + "learning_rate": 0.0009228484984311968, + "loss": 0.4353, + "step": 2859 + }, + { + "epoch": 0.15978992653015617, + "grad_norm": 1.1174883842468262, + "learning_rate": 0.000922820484087853, + "loss": 0.5062, + "step": 2860 + }, + { + "epoch": 0.15984579713383804, + "grad_norm": 1.349967360496521, + "learning_rate": 0.0009227924697445092, + "loss": 0.6287, + "step": 2861 + }, + { + "epoch": 0.1599016677375199, + "grad_norm": 0.8962681889533997, + "learning_rate": 0.0009227644554011654, + "loss": 0.64, + "step": 2862 + }, + { + "epoch": 0.15995753834120177, + "grad_norm": 0.6431578993797302, + "learning_rate": 0.0009227364410578216, + "loss": 0.5359, + "step": 2863 + }, + { + "epoch": 0.16001340894488364, + "grad_norm": 0.6147413849830627, + "learning_rate": 0.0009227084267144778, + "loss": 0.4943, + "step": 2864 + }, + { + "epoch": 0.16006927954856554, + "grad_norm": 0.6523920893669128, + "learning_rate": 0.0009226804123711341, + "loss": 0.5644, + "step": 2865 + }, + { + "epoch": 0.1601251501522474, + "grad_norm": 0.7837699055671692, + "learning_rate": 0.0009226523980277902, + "loss": 0.5234, + "step": 2866 + }, + { + "epoch": 0.16018102075592927, + "grad_norm": 0.7677902579307556, + "learning_rate": 0.0009226243836844464, + "loss": 0.4897, + "step": 2867 + }, + { + "epoch": 0.16023689135961114, + "grad_norm": 0.9971684217453003, + "learning_rate": 0.0009225963693411027, + "loss": 0.5265, + "step": 2868 + }, + { + "epoch": 0.160292761963293, + "grad_norm": 0.7514908313751221, + "learning_rate": 0.0009225683549977589, + "loss": 0.5985, + "step": 2869 + }, + { + "epoch": 0.16034863256697487, + "grad_norm": 0.7579535245895386, + "learning_rate": 0.0009225403406544151, + "loss": 0.4736, + "step": 2870 + }, + { + "epoch": 0.16040450317065677, + "grad_norm": 0.48046672344207764, + "learning_rate": 0.0009225123263110713, + "loss": 0.528, + "step": 2871 + }, + { + "epoch": 0.16046037377433864, + "grad_norm": 0.7468258738517761, + "learning_rate": 0.0009224843119677276, + "loss": 0.4868, + "step": 2872 + }, + { + "epoch": 0.1605162443780205, + "grad_norm": 0.6573393940925598, + "learning_rate": 0.0009224562976243837, + "loss": 0.506, + "step": 2873 + }, + { + "epoch": 0.16057211498170237, + "grad_norm": 1.2053605318069458, + "learning_rate": 0.0009224282832810399, + "loss": 0.6571, + "step": 2874 + }, + { + "epoch": 0.16062798558538424, + "grad_norm": 0.8957147002220154, + "learning_rate": 0.0009224002689376961, + "loss": 0.5006, + "step": 2875 + }, + { + "epoch": 0.1606838561890661, + "grad_norm": 2.669316291809082, + "learning_rate": 0.0009223722545943524, + "loss": 0.4771, + "step": 2876 + }, + { + "epoch": 0.160739726792748, + "grad_norm": 0.8547194004058838, + "learning_rate": 0.0009223442402510085, + "loss": 0.4848, + "step": 2877 + }, + { + "epoch": 0.16079559739642987, + "grad_norm": 0.9180343151092529, + "learning_rate": 0.0009223162259076647, + "loss": 0.3971, + "step": 2878 + }, + { + "epoch": 0.16085146800011174, + "grad_norm": 1.0123227834701538, + "learning_rate": 0.0009222882115643209, + "loss": 0.4357, + "step": 2879 + }, + { + "epoch": 0.1609073386037936, + "grad_norm": 0.47982335090637207, + "learning_rate": 0.0009222601972209772, + "loss": 0.5482, + "step": 2880 + }, + { + "epoch": 0.16096320920747548, + "grad_norm": 0.4351023733615875, + "learning_rate": 0.0009222321828776333, + "loss": 0.4891, + "step": 2881 + }, + { + "epoch": 0.16101907981115737, + "grad_norm": 0.8333148956298828, + "learning_rate": 0.0009222041685342895, + "loss": 0.5311, + "step": 2882 + }, + { + "epoch": 0.16107495041483924, + "grad_norm": 1.2017695903778076, + "learning_rate": 0.0009221761541909458, + "loss": 0.5521, + "step": 2883 + }, + { + "epoch": 0.1611308210185211, + "grad_norm": 0.6929494142532349, + "learning_rate": 0.000922148139847602, + "loss": 0.3863, + "step": 2884 + }, + { + "epoch": 0.16118669162220298, + "grad_norm": 0.8215415477752686, + "learning_rate": 0.0009221201255042582, + "loss": 0.6197, + "step": 2885 + }, + { + "epoch": 0.16124256222588484, + "grad_norm": 0.5138126015663147, + "learning_rate": 0.0009220921111609144, + "loss": 0.3835, + "step": 2886 + }, + { + "epoch": 0.1612984328295667, + "grad_norm": 0.7895022630691528, + "learning_rate": 0.0009220640968175707, + "loss": 0.4828, + "step": 2887 + }, + { + "epoch": 0.1613543034332486, + "grad_norm": 0.4749007821083069, + "learning_rate": 0.0009220360824742269, + "loss": 0.4486, + "step": 2888 + }, + { + "epoch": 0.16141017403693048, + "grad_norm": 0.7198782563209534, + "learning_rate": 0.000922008068130883, + "loss": 0.4649, + "step": 2889 + }, + { + "epoch": 0.16146604464061234, + "grad_norm": 0.5590113997459412, + "learning_rate": 0.0009219800537875393, + "loss": 0.4988, + "step": 2890 + }, + { + "epoch": 0.1615219152442942, + "grad_norm": 0.5365570187568665, + "learning_rate": 0.0009219520394441955, + "loss": 0.3732, + "step": 2891 + }, + { + "epoch": 0.16157778584797608, + "grad_norm": 1.6891059875488281, + "learning_rate": 0.0009219240251008516, + "loss": 0.4558, + "step": 2892 + }, + { + "epoch": 0.16163365645165795, + "grad_norm": 0.623045027256012, + "learning_rate": 0.0009218960107575078, + "loss": 0.5143, + "step": 2893 + }, + { + "epoch": 0.16168952705533984, + "grad_norm": 0.5371109247207642, + "learning_rate": 0.0009218679964141641, + "loss": 0.364, + "step": 2894 + }, + { + "epoch": 0.1617453976590217, + "grad_norm": 1.1466373205184937, + "learning_rate": 0.0009218399820708203, + "loss": 0.5526, + "step": 2895 + }, + { + "epoch": 0.16180126826270358, + "grad_norm": 0.7934058904647827, + "learning_rate": 0.0009218119677274764, + "loss": 0.5374, + "step": 2896 + }, + { + "epoch": 0.16185713886638545, + "grad_norm": 0.6979703307151794, + "learning_rate": 0.0009217839533841327, + "loss": 0.5468, + "step": 2897 + }, + { + "epoch": 0.16191300947006732, + "grad_norm": 0.6033763289451599, + "learning_rate": 0.0009217559390407889, + "loss": 0.4617, + "step": 2898 + }, + { + "epoch": 0.1619688800737492, + "grad_norm": 1.0116925239562988, + "learning_rate": 0.0009217279246974451, + "loss": 0.4733, + "step": 2899 + }, + { + "epoch": 0.16202475067743108, + "grad_norm": 1.212654709815979, + "learning_rate": 0.0009216999103541012, + "loss": 0.5094, + "step": 2900 + }, + { + "epoch": 0.16208062128111295, + "grad_norm": 0.47805866599082947, + "learning_rate": 0.0009216718960107576, + "loss": 0.543, + "step": 2901 + }, + { + "epoch": 0.16213649188479481, + "grad_norm": 0.7479064464569092, + "learning_rate": 0.0009216438816674138, + "loss": 0.5191, + "step": 2902 + }, + { + "epoch": 0.16219236248847668, + "grad_norm": 0.726576566696167, + "learning_rate": 0.00092161586732407, + "loss": 0.5736, + "step": 2903 + }, + { + "epoch": 0.16224823309215855, + "grad_norm": 0.4731319844722748, + "learning_rate": 0.0009215878529807262, + "loss": 0.5085, + "step": 2904 + }, + { + "epoch": 0.16230410369584045, + "grad_norm": 0.4865752160549164, + "learning_rate": 0.0009215598386373824, + "loss": 0.6006, + "step": 2905 + }, + { + "epoch": 0.1623599742995223, + "grad_norm": 0.5418570041656494, + "learning_rate": 0.0009215318242940386, + "loss": 0.5103, + "step": 2906 + }, + { + "epoch": 0.16241584490320418, + "grad_norm": 0.5391319990158081, + "learning_rate": 0.0009215038099506948, + "loss": 0.602, + "step": 2907 + }, + { + "epoch": 0.16247171550688605, + "grad_norm": 0.5497490167617798, + "learning_rate": 0.000921475795607351, + "loss": 0.4872, + "step": 2908 + }, + { + "epoch": 0.16252758611056792, + "grad_norm": 0.5558151006698608, + "learning_rate": 0.0009214477812640072, + "loss": 0.4938, + "step": 2909 + }, + { + "epoch": 0.16258345671424979, + "grad_norm": 0.6539651155471802, + "learning_rate": 0.0009214197669206634, + "loss": 0.4965, + "step": 2910 + }, + { + "epoch": 0.16263932731793168, + "grad_norm": 1.3866524696350098, + "learning_rate": 0.0009213917525773195, + "loss": 0.5895, + "step": 2911 + }, + { + "epoch": 0.16269519792161355, + "grad_norm": 0.6529863476753235, + "learning_rate": 0.0009213637382339758, + "loss": 0.6305, + "step": 2912 + }, + { + "epoch": 0.16275106852529542, + "grad_norm": 0.693434476852417, + "learning_rate": 0.000921335723890632, + "loss": 0.5803, + "step": 2913 + }, + { + "epoch": 0.16280693912897728, + "grad_norm": 0.6006316542625427, + "learning_rate": 0.0009213077095472882, + "loss": 0.483, + "step": 2914 + }, + { + "epoch": 0.16286280973265915, + "grad_norm": 0.6436848044395447, + "learning_rate": 0.0009212796952039444, + "loss": 0.4902, + "step": 2915 + }, + { + "epoch": 0.16291868033634102, + "grad_norm": 0.48094379901885986, + "learning_rate": 0.0009212516808606006, + "loss": 0.3792, + "step": 2916 + }, + { + "epoch": 0.16297455094002292, + "grad_norm": 0.5497828722000122, + "learning_rate": 0.0009212236665172568, + "loss": 0.3622, + "step": 2917 + }, + { + "epoch": 0.16303042154370478, + "grad_norm": 0.4628417193889618, + "learning_rate": 0.000921195652173913, + "loss": 0.3545, + "step": 2918 + }, + { + "epoch": 0.16308629214738665, + "grad_norm": 0.7704576253890991, + "learning_rate": 0.0009211676378305693, + "loss": 0.452, + "step": 2919 + }, + { + "epoch": 0.16314216275106852, + "grad_norm": 0.5392871499061584, + "learning_rate": 0.0009211396234872255, + "loss": 0.41, + "step": 2920 + }, + { + "epoch": 0.1631980333547504, + "grad_norm": 0.48902904987335205, + "learning_rate": 0.0009211116091438817, + "loss": 0.5367, + "step": 2921 + }, + { + "epoch": 0.16325390395843228, + "grad_norm": 1.138019323348999, + "learning_rate": 0.000921083594800538, + "loss": 0.62, + "step": 2922 + }, + { + "epoch": 0.16330977456211415, + "grad_norm": 0.6904631853103638, + "learning_rate": 0.0009210555804571941, + "loss": 0.5553, + "step": 2923 + }, + { + "epoch": 0.16336564516579602, + "grad_norm": 2.0372533798217773, + "learning_rate": 0.0009210275661138503, + "loss": 0.4976, + "step": 2924 + }, + { + "epoch": 0.1634215157694779, + "grad_norm": 1.0321234464645386, + "learning_rate": 0.0009209995517705065, + "loss": 0.4942, + "step": 2925 + }, + { + "epoch": 0.16347738637315976, + "grad_norm": 0.598773181438446, + "learning_rate": 0.0009209715374271628, + "loss": 0.4865, + "step": 2926 + }, + { + "epoch": 0.16353325697684162, + "grad_norm": 0.7538778185844421, + "learning_rate": 0.0009209435230838189, + "loss": 0.5345, + "step": 2927 + }, + { + "epoch": 0.16358912758052352, + "grad_norm": 0.6870383024215698, + "learning_rate": 0.0009209155087404751, + "loss": 0.5206, + "step": 2928 + }, + { + "epoch": 0.1636449981842054, + "grad_norm": 0.5862061977386475, + "learning_rate": 0.0009208874943971314, + "loss": 0.5196, + "step": 2929 + }, + { + "epoch": 0.16370086878788725, + "grad_norm": 0.4904618263244629, + "learning_rate": 0.0009208594800537876, + "loss": 0.4716, + "step": 2930 + }, + { + "epoch": 0.16375673939156912, + "grad_norm": 0.715714693069458, + "learning_rate": 0.0009208314657104437, + "loss": 0.4995, + "step": 2931 + }, + { + "epoch": 0.163812609995251, + "grad_norm": 0.9581141471862793, + "learning_rate": 0.0009208034513670999, + "loss": 0.6201, + "step": 2932 + }, + { + "epoch": 0.16386848059893286, + "grad_norm": 0.6825672388076782, + "learning_rate": 0.0009207754370237562, + "loss": 0.5419, + "step": 2933 + }, + { + "epoch": 0.16392435120261475, + "grad_norm": 0.7303814888000488, + "learning_rate": 0.0009207474226804123, + "loss": 0.3829, + "step": 2934 + }, + { + "epoch": 0.16398022180629662, + "grad_norm": 1.000221610069275, + "learning_rate": 0.0009207194083370685, + "loss": 0.5709, + "step": 2935 + }, + { + "epoch": 0.1640360924099785, + "grad_norm": 0.6487939953804016, + "learning_rate": 0.0009206913939937249, + "loss": 0.5738, + "step": 2936 + }, + { + "epoch": 0.16409196301366036, + "grad_norm": 0.7571101784706116, + "learning_rate": 0.0009206633796503811, + "loss": 0.3678, + "step": 2937 + }, + { + "epoch": 0.16414783361734223, + "grad_norm": 0.6976777911186218, + "learning_rate": 0.0009206353653070372, + "loss": 0.4845, + "step": 2938 + }, + { + "epoch": 0.16420370422102412, + "grad_norm": 0.5157960653305054, + "learning_rate": 0.0009206073509636934, + "loss": 0.4626, + "step": 2939 + }, + { + "epoch": 0.164259574824706, + "grad_norm": 0.5080830454826355, + "learning_rate": 0.0009205793366203497, + "loss": 0.5028, + "step": 2940 + }, + { + "epoch": 0.16431544542838786, + "grad_norm": 0.6012086868286133, + "learning_rate": 0.0009205513222770059, + "loss": 0.519, + "step": 2941 + }, + { + "epoch": 0.16437131603206973, + "grad_norm": 0.5531623363494873, + "learning_rate": 0.000920523307933662, + "loss": 0.4876, + "step": 2942 + }, + { + "epoch": 0.1644271866357516, + "grad_norm": 34.15959930419922, + "learning_rate": 0.0009204952935903182, + "loss": 0.464, + "step": 2943 + }, + { + "epoch": 0.16448305723943346, + "grad_norm": 0.47535914182662964, + "learning_rate": 0.0009204672792469745, + "loss": 0.4227, + "step": 2944 + }, + { + "epoch": 0.16453892784311536, + "grad_norm": 0.5730754733085632, + "learning_rate": 0.0009204392649036307, + "loss": 0.402, + "step": 2945 + }, + { + "epoch": 0.16459479844679722, + "grad_norm": 1.8807287216186523, + "learning_rate": 0.0009204112505602868, + "loss": 0.5679, + "step": 2946 + }, + { + "epoch": 0.1646506690504791, + "grad_norm": 0.5375819802284241, + "learning_rate": 0.0009203832362169431, + "loss": 0.4524, + "step": 2947 + }, + { + "epoch": 0.16470653965416096, + "grad_norm": 0.6772319078445435, + "learning_rate": 0.0009203552218735993, + "loss": 0.5074, + "step": 2948 + }, + { + "epoch": 0.16476241025784283, + "grad_norm": 1.0583676099777222, + "learning_rate": 0.0009203272075302555, + "loss": 0.4803, + "step": 2949 + }, + { + "epoch": 0.1648182808615247, + "grad_norm": 3.802293062210083, + "learning_rate": 0.0009202991931869116, + "loss": 0.5627, + "step": 2950 + }, + { + "epoch": 0.1648741514652066, + "grad_norm": 0.6039202809333801, + "learning_rate": 0.000920271178843568, + "loss": 0.4858, + "step": 2951 + }, + { + "epoch": 0.16493002206888846, + "grad_norm": 0.6299573183059692, + "learning_rate": 0.0009202431645002242, + "loss": 0.4865, + "step": 2952 + }, + { + "epoch": 0.16498589267257033, + "grad_norm": 0.8051428198814392, + "learning_rate": 0.0009202151501568803, + "loss": 0.482, + "step": 2953 + }, + { + "epoch": 0.1650417632762522, + "grad_norm": 0.8834247589111328, + "learning_rate": 0.0009201871358135366, + "loss": 0.5003, + "step": 2954 + }, + { + "epoch": 0.16509763387993406, + "grad_norm": 0.6654444932937622, + "learning_rate": 0.0009201591214701928, + "loss": 0.517, + "step": 2955 + }, + { + "epoch": 0.16515350448361593, + "grad_norm": 0.7218305468559265, + "learning_rate": 0.000920131107126849, + "loss": 0.5394, + "step": 2956 + }, + { + "epoch": 0.16520937508729783, + "grad_norm": 0.49676766991615295, + "learning_rate": 0.0009201030927835051, + "loss": 0.5238, + "step": 2957 + }, + { + "epoch": 0.1652652456909797, + "grad_norm": 0.7906652688980103, + "learning_rate": 0.0009200750784401614, + "loss": 0.685, + "step": 2958 + }, + { + "epoch": 0.16532111629466156, + "grad_norm": 0.7057422399520874, + "learning_rate": 0.0009200470640968176, + "loss": 0.4983, + "step": 2959 + }, + { + "epoch": 0.16537698689834343, + "grad_norm": 0.950018048286438, + "learning_rate": 0.0009200190497534738, + "loss": 0.5059, + "step": 2960 + }, + { + "epoch": 0.1654328575020253, + "grad_norm": 1.3410903215408325, + "learning_rate": 0.00091999103541013, + "loss": 0.4612, + "step": 2961 + }, + { + "epoch": 0.1654887281057072, + "grad_norm": 0.5885089039802551, + "learning_rate": 0.0009199630210667862, + "loss": 0.5718, + "step": 2962 + }, + { + "epoch": 0.16554459870938906, + "grad_norm": 0.5323497653007507, + "learning_rate": 0.0009199350067234424, + "loss": 0.4566, + "step": 2963 + }, + { + "epoch": 0.16560046931307093, + "grad_norm": 1.0536043643951416, + "learning_rate": 0.0009199069923800986, + "loss": 0.4703, + "step": 2964 + }, + { + "epoch": 0.1656563399167528, + "grad_norm": 0.5344094038009644, + "learning_rate": 0.0009198789780367548, + "loss": 0.5194, + "step": 2965 + }, + { + "epoch": 0.16571221052043467, + "grad_norm": 0.8497459292411804, + "learning_rate": 0.000919850963693411, + "loss": 0.5759, + "step": 2966 + }, + { + "epoch": 0.16576808112411653, + "grad_norm": 0.467883437871933, + "learning_rate": 0.0009198229493500672, + "loss": 0.4351, + "step": 2967 + }, + { + "epoch": 0.16582395172779843, + "grad_norm": 0.6904860734939575, + "learning_rate": 0.0009197949350067236, + "loss": 0.584, + "step": 2968 + }, + { + "epoch": 0.1658798223314803, + "grad_norm": 0.7303336262702942, + "learning_rate": 0.0009197669206633797, + "loss": 0.5057, + "step": 2969 + }, + { + "epoch": 0.16593569293516217, + "grad_norm": 0.7487726807594299, + "learning_rate": 0.0009197389063200359, + "loss": 0.441, + "step": 2970 + }, + { + "epoch": 0.16599156353884403, + "grad_norm": 0.677872359752655, + "learning_rate": 0.0009197108919766921, + "loss": 0.456, + "step": 2971 + }, + { + "epoch": 0.1660474341425259, + "grad_norm": 3.8834011554718018, + "learning_rate": 0.0009196828776333484, + "loss": 0.4839, + "step": 2972 + }, + { + "epoch": 0.16610330474620777, + "grad_norm": 0.7055858969688416, + "learning_rate": 0.0009196548632900045, + "loss": 0.4733, + "step": 2973 + }, + { + "epoch": 0.16615917534988967, + "grad_norm": 2.9706506729125977, + "learning_rate": 0.0009196268489466607, + "loss": 0.4107, + "step": 2974 + }, + { + "epoch": 0.16621504595357153, + "grad_norm": 0.6332435011863708, + "learning_rate": 0.000919598834603317, + "loss": 0.533, + "step": 2975 + }, + { + "epoch": 0.1662709165572534, + "grad_norm": 0.5895186066627502, + "learning_rate": 0.0009195708202599731, + "loss": 0.4692, + "step": 2976 + }, + { + "epoch": 0.16632678716093527, + "grad_norm": 0.6181948184967041, + "learning_rate": 0.0009195428059166293, + "loss": 0.4285, + "step": 2977 + }, + { + "epoch": 0.16638265776461714, + "grad_norm": 0.8411385416984558, + "learning_rate": 0.0009195147915732855, + "loss": 0.5542, + "step": 2978 + }, + { + "epoch": 0.16643852836829903, + "grad_norm": 0.6046685576438904, + "learning_rate": 0.0009194867772299418, + "loss": 0.4309, + "step": 2979 + }, + { + "epoch": 0.1664943989719809, + "grad_norm": 0.7958986759185791, + "learning_rate": 0.0009194587628865979, + "loss": 0.4836, + "step": 2980 + }, + { + "epoch": 0.16655026957566277, + "grad_norm": 0.4791279733181, + "learning_rate": 0.0009194307485432541, + "loss": 0.4589, + "step": 2981 + }, + { + "epoch": 0.16660614017934464, + "grad_norm": 0.525902271270752, + "learning_rate": 0.0009194027341999103, + "loss": 0.5214, + "step": 2982 + }, + { + "epoch": 0.1666620107830265, + "grad_norm": 0.5200362205505371, + "learning_rate": 0.0009193747198565666, + "loss": 0.4062, + "step": 2983 + }, + { + "epoch": 0.16671788138670837, + "grad_norm": 0.48273965716362, + "learning_rate": 0.0009193467055132227, + "loss": 0.4212, + "step": 2984 + }, + { + "epoch": 0.16677375199039027, + "grad_norm": 0.4546259343624115, + "learning_rate": 0.000919318691169879, + "loss": 0.4023, + "step": 2985 + }, + { + "epoch": 0.16682962259407214, + "grad_norm": 0.5690484642982483, + "learning_rate": 0.0009192906768265353, + "loss": 0.4781, + "step": 2986 + }, + { + "epoch": 0.166885493197754, + "grad_norm": 0.5440406799316406, + "learning_rate": 0.0009192626624831915, + "loss": 0.3979, + "step": 2987 + }, + { + "epoch": 0.16694136380143587, + "grad_norm": 0.7004792094230652, + "learning_rate": 0.0009192346481398476, + "loss": 0.432, + "step": 2988 + }, + { + "epoch": 0.16699723440511774, + "grad_norm": 0.5999491214752197, + "learning_rate": 0.0009192066337965038, + "loss": 0.4981, + "step": 2989 + }, + { + "epoch": 0.1670531050087996, + "grad_norm": 1.4643707275390625, + "learning_rate": 0.0009191786194531601, + "loss": 0.4327, + "step": 2990 + }, + { + "epoch": 0.1671089756124815, + "grad_norm": 0.7047449946403503, + "learning_rate": 0.0009191506051098163, + "loss": 0.4809, + "step": 2991 + }, + { + "epoch": 0.16716484621616337, + "grad_norm": 0.5643956661224365, + "learning_rate": 0.0009191225907664724, + "loss": 0.4858, + "step": 2992 + }, + { + "epoch": 0.16722071681984524, + "grad_norm": 0.5514084696769714, + "learning_rate": 0.0009190945764231287, + "loss": 0.5025, + "step": 2993 + }, + { + "epoch": 0.1672765874235271, + "grad_norm": 0.7304065227508545, + "learning_rate": 0.0009190665620797849, + "loss": 0.692, + "step": 2994 + }, + { + "epoch": 0.16733245802720897, + "grad_norm": 1.3913012742996216, + "learning_rate": 0.000919038547736441, + "loss": 0.4362, + "step": 2995 + }, + { + "epoch": 0.16738832863089087, + "grad_norm": 0.8254660367965698, + "learning_rate": 0.0009190105333930972, + "loss": 0.4322, + "step": 2996 + }, + { + "epoch": 0.16744419923457274, + "grad_norm": 0.5317649841308594, + "learning_rate": 0.0009189825190497535, + "loss": 0.5252, + "step": 2997 + }, + { + "epoch": 0.1675000698382546, + "grad_norm": 0.6153241395950317, + "learning_rate": 0.0009189545047064097, + "loss": 0.6743, + "step": 2998 + }, + { + "epoch": 0.16755594044193647, + "grad_norm": 0.7857419848442078, + "learning_rate": 0.0009189264903630658, + "loss": 0.4347, + "step": 2999 + }, + { + "epoch": 0.16761181104561834, + "grad_norm": 0.5217227339744568, + "learning_rate": 0.0009188984760197221, + "loss": 0.4613, + "step": 3000 + }, + { + "epoch": 0.16761181104561834, + "eval_cer": 0.09769456537156448, + "eval_loss": 0.363749235868454, + "eval_runtime": 55.7329, + "eval_samples_per_second": 81.424, + "eval_steps_per_second": 5.096, + "eval_wer": 0.3849054495684478, + "step": 3000 + }, + { + "epoch": 0.1676676816493002, + "grad_norm": 0.42212188243865967, + "learning_rate": 0.0009188704616763784, + "loss": 0.5, + "step": 3001 + }, + { + "epoch": 0.1677235522529821, + "grad_norm": 1.3326336145401, + "learning_rate": 0.0009188424473330346, + "loss": 0.4423, + "step": 3002 + }, + { + "epoch": 0.16777942285666397, + "grad_norm": 0.7589061260223389, + "learning_rate": 0.0009188144329896907, + "loss": 0.5383, + "step": 3003 + }, + { + "epoch": 0.16783529346034584, + "grad_norm": 0.8087379336357117, + "learning_rate": 0.000918786418646347, + "loss": 0.6438, + "step": 3004 + }, + { + "epoch": 0.1678911640640277, + "grad_norm": 0.6463863849639893, + "learning_rate": 0.0009187584043030032, + "loss": 0.621, + "step": 3005 + }, + { + "epoch": 0.16794703466770958, + "grad_norm": 6.584296226501465, + "learning_rate": 0.0009187303899596594, + "loss": 0.4674, + "step": 3006 + }, + { + "epoch": 0.16800290527139145, + "grad_norm": 0.5086818933486938, + "learning_rate": 0.0009187023756163156, + "loss": 0.5039, + "step": 3007 + }, + { + "epoch": 0.16805877587507334, + "grad_norm": 3.950380802154541, + "learning_rate": 0.0009186743612729718, + "loss": 0.4682, + "step": 3008 + }, + { + "epoch": 0.1681146464787552, + "grad_norm": 0.864630401134491, + "learning_rate": 0.000918646346929628, + "loss": 0.576, + "step": 3009 + }, + { + "epoch": 0.16817051708243708, + "grad_norm": 0.6413453221321106, + "learning_rate": 0.0009186183325862842, + "loss": 0.5559, + "step": 3010 + }, + { + "epoch": 0.16822638768611894, + "grad_norm": 0.7002838253974915, + "learning_rate": 0.0009185903182429404, + "loss": 0.634, + "step": 3011 + }, + { + "epoch": 0.1682822582898008, + "grad_norm": 0.5080361366271973, + "learning_rate": 0.0009185623038995966, + "loss": 0.504, + "step": 3012 + }, + { + "epoch": 0.16833812889348268, + "grad_norm": 0.6141003370285034, + "learning_rate": 0.0009185342895562528, + "loss": 0.5203, + "step": 3013 + }, + { + "epoch": 0.16839399949716458, + "grad_norm": 0.7424399256706238, + "learning_rate": 0.0009185062752129091, + "loss": 0.468, + "step": 3014 + }, + { + "epoch": 0.16844987010084644, + "grad_norm": 0.6062166094779968, + "learning_rate": 0.0009184782608695652, + "loss": 0.4889, + "step": 3015 + }, + { + "epoch": 0.1685057407045283, + "grad_norm": 1.9199438095092773, + "learning_rate": 0.0009184502465262214, + "loss": 0.4747, + "step": 3016 + }, + { + "epoch": 0.16856161130821018, + "grad_norm": 0.4913809597492218, + "learning_rate": 0.0009184222321828776, + "loss": 0.4528, + "step": 3017 + }, + { + "epoch": 0.16861748191189205, + "grad_norm": 0.8706223368644714, + "learning_rate": 0.0009183942178395338, + "loss": 0.5238, + "step": 3018 + }, + { + "epoch": 0.16867335251557394, + "grad_norm": 0.7703204154968262, + "learning_rate": 0.0009183662034961901, + "loss": 0.5252, + "step": 3019 + }, + { + "epoch": 0.1687292231192558, + "grad_norm": 0.657082200050354, + "learning_rate": 0.0009183381891528463, + "loss": 0.5545, + "step": 3020 + }, + { + "epoch": 0.16878509372293768, + "grad_norm": 0.5840556621551514, + "learning_rate": 0.0009183101748095025, + "loss": 0.4083, + "step": 3021 + }, + { + "epoch": 0.16884096432661955, + "grad_norm": 0.5065975189208984, + "learning_rate": 0.0009182821604661587, + "loss": 0.5191, + "step": 3022 + }, + { + "epoch": 0.16889683493030141, + "grad_norm": 0.9118289947509766, + "learning_rate": 0.0009182541461228149, + "loss": 0.5331, + "step": 3023 + }, + { + "epoch": 0.16895270553398328, + "grad_norm": 1.6419117450714111, + "learning_rate": 0.0009182261317794711, + "loss": 0.4249, + "step": 3024 + }, + { + "epoch": 0.16900857613766518, + "grad_norm": 2.6154470443725586, + "learning_rate": 0.0009181981174361274, + "loss": 0.8564, + "step": 3025 + }, + { + "epoch": 0.16906444674134705, + "grad_norm": 1.2075742483139038, + "learning_rate": 0.0009181701030927835, + "loss": 0.4515, + "step": 3026 + }, + { + "epoch": 0.16912031734502891, + "grad_norm": 0.625933051109314, + "learning_rate": 0.0009181420887494397, + "loss": 0.4622, + "step": 3027 + }, + { + "epoch": 0.16917618794871078, + "grad_norm": 0.6257807612419128, + "learning_rate": 0.0009181140744060959, + "loss": 0.4312, + "step": 3028 + }, + { + "epoch": 0.16923205855239265, + "grad_norm": 1.0303670167922974, + "learning_rate": 0.0009180860600627522, + "loss": 0.4892, + "step": 3029 + }, + { + "epoch": 0.16928792915607452, + "grad_norm": 0.8254990577697754, + "learning_rate": 0.0009180580457194083, + "loss": 0.4695, + "step": 3030 + }, + { + "epoch": 0.1693437997597564, + "grad_norm": 0.5202803611755371, + "learning_rate": 0.0009180300313760645, + "loss": 0.5137, + "step": 3031 + }, + { + "epoch": 0.16939967036343828, + "grad_norm": 0.5910012722015381, + "learning_rate": 0.0009180020170327208, + "loss": 0.4396, + "step": 3032 + }, + { + "epoch": 0.16945554096712015, + "grad_norm": 0.78654545545578, + "learning_rate": 0.000917974002689377, + "loss": 0.413, + "step": 3033 + }, + { + "epoch": 0.16951141157080202, + "grad_norm": 0.630510151386261, + "learning_rate": 0.0009179459883460331, + "loss": 0.4977, + "step": 3034 + }, + { + "epoch": 0.16956728217448389, + "grad_norm": 3.150639295578003, + "learning_rate": 0.0009179179740026893, + "loss": 0.5695, + "step": 3035 + }, + { + "epoch": 0.16962315277816578, + "grad_norm": 0.47996777296066284, + "learning_rate": 0.0009178899596593457, + "loss": 0.3851, + "step": 3036 + }, + { + "epoch": 0.16967902338184765, + "grad_norm": 0.8576956391334534, + "learning_rate": 0.0009178619453160019, + "loss": 0.501, + "step": 3037 + }, + { + "epoch": 0.16973489398552952, + "grad_norm": 0.6556514501571655, + "learning_rate": 0.000917833930972658, + "loss": 0.5157, + "step": 3038 + }, + { + "epoch": 0.16979076458921138, + "grad_norm": 0.5240998864173889, + "learning_rate": 0.0009178059166293143, + "loss": 0.4176, + "step": 3039 + }, + { + "epoch": 0.16984663519289325, + "grad_norm": 0.8503729701042175, + "learning_rate": 0.0009177779022859705, + "loss": 0.4572, + "step": 3040 + }, + { + "epoch": 0.16990250579657512, + "grad_norm": 0.8590173125267029, + "learning_rate": 0.0009177498879426266, + "loss": 0.4509, + "step": 3041 + }, + { + "epoch": 0.16995837640025702, + "grad_norm": 0.38075995445251465, + "learning_rate": 0.0009177218735992828, + "loss": 0.3707, + "step": 3042 + }, + { + "epoch": 0.17001424700393888, + "grad_norm": 2.2734808921813965, + "learning_rate": 0.0009176938592559391, + "loss": 0.7086, + "step": 3043 + }, + { + "epoch": 0.17007011760762075, + "grad_norm": 2.1208863258361816, + "learning_rate": 0.0009176658449125953, + "loss": 0.5552, + "step": 3044 + }, + { + "epoch": 0.17012598821130262, + "grad_norm": 0.5567620992660522, + "learning_rate": 0.0009176378305692514, + "loss": 0.5076, + "step": 3045 + }, + { + "epoch": 0.1701818588149845, + "grad_norm": 0.5879213809967041, + "learning_rate": 0.0009176098162259077, + "loss": 0.3715, + "step": 3046 + }, + { + "epoch": 0.17023772941866636, + "grad_norm": 0.5651243329048157, + "learning_rate": 0.0009175818018825639, + "loss": 0.4111, + "step": 3047 + }, + { + "epoch": 0.17029360002234825, + "grad_norm": 0.9518117308616638, + "learning_rate": 0.0009175537875392201, + "loss": 0.5168, + "step": 3048 + }, + { + "epoch": 0.17034947062603012, + "grad_norm": 0.9429636597633362, + "learning_rate": 0.0009175257731958762, + "loss": 0.4669, + "step": 3049 + }, + { + "epoch": 0.170405341229712, + "grad_norm": 0.7809973359107971, + "learning_rate": 0.0009174977588525325, + "loss": 0.5018, + "step": 3050 + }, + { + "epoch": 0.17046121183339386, + "grad_norm": 0.8313552141189575, + "learning_rate": 0.0009174697445091887, + "loss": 0.656, + "step": 3051 + }, + { + "epoch": 0.17051708243707572, + "grad_norm": 0.7720755338668823, + "learning_rate": 0.000917441730165845, + "loss": 0.5573, + "step": 3052 + }, + { + "epoch": 0.17057295304075762, + "grad_norm": 0.7742499709129333, + "learning_rate": 0.000917413715822501, + "loss": 0.5445, + "step": 3053 + }, + { + "epoch": 0.1706288236444395, + "grad_norm": 0.48344215750694275, + "learning_rate": 0.0009173857014791574, + "loss": 0.5429, + "step": 3054 + }, + { + "epoch": 0.17068469424812135, + "grad_norm": 1.1149572134017944, + "learning_rate": 0.0009173576871358136, + "loss": 0.5572, + "step": 3055 + }, + { + "epoch": 0.17074056485180322, + "grad_norm": 0.771146833896637, + "learning_rate": 0.0009173296727924698, + "loss": 0.436, + "step": 3056 + }, + { + "epoch": 0.1707964354554851, + "grad_norm": 0.705645740032196, + "learning_rate": 0.000917301658449126, + "loss": 0.6625, + "step": 3057 + }, + { + "epoch": 0.17085230605916696, + "grad_norm": 0.7425920963287354, + "learning_rate": 0.0009172736441057822, + "loss": 0.5601, + "step": 3058 + }, + { + "epoch": 0.17090817666284885, + "grad_norm": 0.5220244526863098, + "learning_rate": 0.0009172456297624384, + "loss": 0.524, + "step": 3059 + }, + { + "epoch": 0.17096404726653072, + "grad_norm": 0.5673323273658752, + "learning_rate": 0.0009172176154190945, + "loss": 0.5638, + "step": 3060 + }, + { + "epoch": 0.1710199178702126, + "grad_norm": 1.409318447113037, + "learning_rate": 0.0009171896010757508, + "loss": 0.4927, + "step": 3061 + }, + { + "epoch": 0.17107578847389446, + "grad_norm": 0.7394165396690369, + "learning_rate": 0.000917161586732407, + "loss": 0.4359, + "step": 3062 + }, + { + "epoch": 0.17113165907757633, + "grad_norm": 0.5745514035224915, + "learning_rate": 0.0009171335723890632, + "loss": 0.5011, + "step": 3063 + }, + { + "epoch": 0.1711875296812582, + "grad_norm": 0.6176352500915527, + "learning_rate": 0.0009171055580457194, + "loss": 0.5136, + "step": 3064 + }, + { + "epoch": 0.1712434002849401, + "grad_norm": 0.7928115129470825, + "learning_rate": 0.0009170775437023756, + "loss": 0.5886, + "step": 3065 + }, + { + "epoch": 0.17129927088862196, + "grad_norm": 0.9043012857437134, + "learning_rate": 0.0009170495293590318, + "loss": 0.4628, + "step": 3066 + }, + { + "epoch": 0.17135514149230383, + "grad_norm": 0.5250157713890076, + "learning_rate": 0.000917021515015688, + "loss": 0.4204, + "step": 3067 + }, + { + "epoch": 0.1714110120959857, + "grad_norm": 0.6880174279212952, + "learning_rate": 0.0009169935006723442, + "loss": 0.5616, + "step": 3068 + }, + { + "epoch": 0.17146688269966756, + "grad_norm": 1.0695061683654785, + "learning_rate": 0.0009169654863290005, + "loss": 0.4726, + "step": 3069 + }, + { + "epoch": 0.17152275330334943, + "grad_norm": 1.8268007040023804, + "learning_rate": 0.0009169374719856567, + "loss": 0.4891, + "step": 3070 + }, + { + "epoch": 0.17157862390703132, + "grad_norm": 0.8457609415054321, + "learning_rate": 0.000916909457642313, + "loss": 0.5229, + "step": 3071 + }, + { + "epoch": 0.1716344945107132, + "grad_norm": 0.6061376333236694, + "learning_rate": 0.0009168814432989691, + "loss": 0.4632, + "step": 3072 + }, + { + "epoch": 0.17169036511439506, + "grad_norm": 0.5620182156562805, + "learning_rate": 0.0009168534289556253, + "loss": 0.5218, + "step": 3073 + }, + { + "epoch": 0.17174623571807693, + "grad_norm": 0.5241831541061401, + "learning_rate": 0.0009168254146122815, + "loss": 0.5545, + "step": 3074 + }, + { + "epoch": 0.1718021063217588, + "grad_norm": 0.788925290107727, + "learning_rate": 0.0009167974002689378, + "loss": 0.4344, + "step": 3075 + }, + { + "epoch": 0.1718579769254407, + "grad_norm": 0.6518178582191467, + "learning_rate": 0.0009167693859255939, + "loss": 0.429, + "step": 3076 + }, + { + "epoch": 0.17191384752912256, + "grad_norm": 0.5516464114189148, + "learning_rate": 0.0009167413715822501, + "loss": 0.4048, + "step": 3077 + }, + { + "epoch": 0.17196971813280443, + "grad_norm": 0.5297027826309204, + "learning_rate": 0.0009167133572389064, + "loss": 0.4929, + "step": 3078 + }, + { + "epoch": 0.1720255887364863, + "grad_norm": 0.45310527086257935, + "learning_rate": 0.0009166853428955626, + "loss": 0.4147, + "step": 3079 + }, + { + "epoch": 0.17208145934016816, + "grad_norm": 1.1511369943618774, + "learning_rate": 0.0009166573285522187, + "loss": 0.5283, + "step": 3080 + }, + { + "epoch": 0.17213732994385003, + "grad_norm": 0.9251008629798889, + "learning_rate": 0.0009166293142088749, + "loss": 0.5712, + "step": 3081 + }, + { + "epoch": 0.17219320054753193, + "grad_norm": 0.49326953291893005, + "learning_rate": 0.0009166012998655312, + "loss": 0.3958, + "step": 3082 + }, + { + "epoch": 0.1722490711512138, + "grad_norm": 0.5591871738433838, + "learning_rate": 0.0009165732855221873, + "loss": 0.5294, + "step": 3083 + }, + { + "epoch": 0.17230494175489566, + "grad_norm": 0.4764940142631531, + "learning_rate": 0.0009165452711788435, + "loss": 0.4854, + "step": 3084 + }, + { + "epoch": 0.17236081235857753, + "grad_norm": 1.060171127319336, + "learning_rate": 0.0009165172568354999, + "loss": 0.7596, + "step": 3085 + }, + { + "epoch": 0.1724166829622594, + "grad_norm": 0.7144439816474915, + "learning_rate": 0.0009164892424921561, + "loss": 0.5054, + "step": 3086 + }, + { + "epoch": 0.17247255356594127, + "grad_norm": 0.8566113710403442, + "learning_rate": 0.0009164612281488122, + "loss": 0.5809, + "step": 3087 + }, + { + "epoch": 0.17252842416962316, + "grad_norm": 0.9833007454872131, + "learning_rate": 0.0009164332138054684, + "loss": 0.6775, + "step": 3088 + }, + { + "epoch": 0.17258429477330503, + "grad_norm": 0.5888664722442627, + "learning_rate": 0.0009164051994621247, + "loss": 0.4904, + "step": 3089 + }, + { + "epoch": 0.1726401653769869, + "grad_norm": 0.7733088731765747, + "learning_rate": 0.0009163771851187809, + "loss": 0.4027, + "step": 3090 + }, + { + "epoch": 0.17269603598066877, + "grad_norm": 0.793528139591217, + "learning_rate": 0.000916349170775437, + "loss": 0.6066, + "step": 3091 + }, + { + "epoch": 0.17275190658435063, + "grad_norm": 0.5176665782928467, + "learning_rate": 0.0009163211564320932, + "loss": 0.3918, + "step": 3092 + }, + { + "epoch": 0.17280777718803253, + "grad_norm": 0.8907937407493591, + "learning_rate": 0.0009162931420887495, + "loss": 0.5682, + "step": 3093 + }, + { + "epoch": 0.1728636477917144, + "grad_norm": 0.6836316585540771, + "learning_rate": 0.0009162651277454057, + "loss": 0.4696, + "step": 3094 + }, + { + "epoch": 0.17291951839539627, + "grad_norm": 0.9902024865150452, + "learning_rate": 0.0009162371134020618, + "loss": 0.4544, + "step": 3095 + }, + { + "epoch": 0.17297538899907813, + "grad_norm": 0.5486159324645996, + "learning_rate": 0.0009162090990587181, + "loss": 0.5174, + "step": 3096 + }, + { + "epoch": 0.17303125960276, + "grad_norm": 0.624129056930542, + "learning_rate": 0.0009161810847153743, + "loss": 0.4536, + "step": 3097 + }, + { + "epoch": 0.17308713020644187, + "grad_norm": 0.685325562953949, + "learning_rate": 0.0009161530703720305, + "loss": 0.4365, + "step": 3098 + }, + { + "epoch": 0.17314300081012376, + "grad_norm": 1.0395479202270508, + "learning_rate": 0.0009161250560286866, + "loss": 0.459, + "step": 3099 + }, + { + "epoch": 0.17319887141380563, + "grad_norm": 0.7181414365768433, + "learning_rate": 0.0009160970416853429, + "loss": 0.6961, + "step": 3100 + }, + { + "epoch": 0.1732547420174875, + "grad_norm": 0.6323641538619995, + "learning_rate": 0.0009160690273419991, + "loss": 0.5149, + "step": 3101 + }, + { + "epoch": 0.17331061262116937, + "grad_norm": 0.65785151720047, + "learning_rate": 0.0009160410129986552, + "loss": 0.4845, + "step": 3102 + }, + { + "epoch": 0.17336648322485124, + "grad_norm": 0.7085111737251282, + "learning_rate": 0.0009160129986553116, + "loss": 0.4793, + "step": 3103 + }, + { + "epoch": 0.1734223538285331, + "grad_norm": 0.6875640153884888, + "learning_rate": 0.0009159849843119678, + "loss": 0.5398, + "step": 3104 + }, + { + "epoch": 0.173478224432215, + "grad_norm": 0.6683698892593384, + "learning_rate": 0.000915956969968624, + "loss": 0.379, + "step": 3105 + }, + { + "epoch": 0.17353409503589687, + "grad_norm": 0.41732051968574524, + "learning_rate": 0.0009159289556252801, + "loss": 0.3874, + "step": 3106 + }, + { + "epoch": 0.17358996563957874, + "grad_norm": 1.2118172645568848, + "learning_rate": 0.0009159009412819364, + "loss": 0.7846, + "step": 3107 + }, + { + "epoch": 0.1736458362432606, + "grad_norm": 1.2887234687805176, + "learning_rate": 0.0009158729269385926, + "loss": 0.3643, + "step": 3108 + }, + { + "epoch": 0.17370170684694247, + "grad_norm": 7.230725288391113, + "learning_rate": 0.0009158449125952488, + "loss": 0.5143, + "step": 3109 + }, + { + "epoch": 0.17375757745062437, + "grad_norm": 0.38708606362342834, + "learning_rate": 0.000915816898251905, + "loss": 0.4086, + "step": 3110 + }, + { + "epoch": 0.17381344805430624, + "grad_norm": 0.43327343463897705, + "learning_rate": 0.0009157888839085612, + "loss": 0.4455, + "step": 3111 + }, + { + "epoch": 0.1738693186579881, + "grad_norm": 6.430209159851074, + "learning_rate": 0.0009157608695652174, + "loss": 0.5199, + "step": 3112 + }, + { + "epoch": 0.17392518926166997, + "grad_norm": 0.4630264639854431, + "learning_rate": 0.0009157328552218736, + "loss": 0.4489, + "step": 3113 + }, + { + "epoch": 0.17398105986535184, + "grad_norm": 1.2416386604309082, + "learning_rate": 0.0009157048408785298, + "loss": 0.5109, + "step": 3114 + }, + { + "epoch": 0.1740369304690337, + "grad_norm": 0.9067546725273132, + "learning_rate": 0.000915676826535186, + "loss": 0.4053, + "step": 3115 + }, + { + "epoch": 0.1740928010727156, + "grad_norm": 0.6269712448120117, + "learning_rate": 0.0009156488121918422, + "loss": 0.5456, + "step": 3116 + }, + { + "epoch": 0.17414867167639747, + "grad_norm": 1.060382604598999, + "learning_rate": 0.0009156207978484986, + "loss": 0.5851, + "step": 3117 + }, + { + "epoch": 0.17420454228007934, + "grad_norm": 0.7031071782112122, + "learning_rate": 0.0009155927835051546, + "loss": 0.4491, + "step": 3118 + }, + { + "epoch": 0.1742604128837612, + "grad_norm": 0.5694454908370972, + "learning_rate": 0.0009155647691618109, + "loss": 0.5631, + "step": 3119 + }, + { + "epoch": 0.17431628348744307, + "grad_norm": 0.5311488509178162, + "learning_rate": 0.0009155367548184671, + "loss": 0.4836, + "step": 3120 + }, + { + "epoch": 0.17437215409112494, + "grad_norm": 1.4300131797790527, + "learning_rate": 0.0009155087404751234, + "loss": 0.5237, + "step": 3121 + }, + { + "epoch": 0.17442802469480684, + "grad_norm": 0.7304103374481201, + "learning_rate": 0.0009154807261317795, + "loss": 0.4808, + "step": 3122 + }, + { + "epoch": 0.1744838952984887, + "grad_norm": 0.6167106628417969, + "learning_rate": 0.0009154527117884357, + "loss": 0.4739, + "step": 3123 + }, + { + "epoch": 0.17453976590217057, + "grad_norm": 1.2434625625610352, + "learning_rate": 0.0009154246974450919, + "loss": 0.491, + "step": 3124 + }, + { + "epoch": 0.17459563650585244, + "grad_norm": 0.9072832465171814, + "learning_rate": 0.0009153966831017481, + "loss": 0.6192, + "step": 3125 + }, + { + "epoch": 0.1746515071095343, + "grad_norm": 0.45731571316719055, + "learning_rate": 0.0009153686687584043, + "loss": 0.3834, + "step": 3126 + }, + { + "epoch": 0.17470737771321618, + "grad_norm": 0.5295108556747437, + "learning_rate": 0.0009153406544150605, + "loss": 0.5056, + "step": 3127 + }, + { + "epoch": 0.17476324831689807, + "grad_norm": 0.9724134802818298, + "learning_rate": 0.0009153126400717168, + "loss": 0.4864, + "step": 3128 + }, + { + "epoch": 0.17481911892057994, + "grad_norm": 0.6094204783439636, + "learning_rate": 0.0009152846257283729, + "loss": 0.4335, + "step": 3129 + }, + { + "epoch": 0.1748749895242618, + "grad_norm": 0.40716758370399475, + "learning_rate": 0.0009152566113850291, + "loss": 0.3794, + "step": 3130 + }, + { + "epoch": 0.17493086012794368, + "grad_norm": 0.4735812842845917, + "learning_rate": 0.0009152285970416853, + "loss": 0.4872, + "step": 3131 + }, + { + "epoch": 0.17498673073162554, + "grad_norm": 0.8964188098907471, + "learning_rate": 0.0009152005826983416, + "loss": 0.5568, + "step": 3132 + }, + { + "epoch": 0.17504260133530744, + "grad_norm": 2.904324531555176, + "learning_rate": 0.0009151725683549977, + "loss": 0.4381, + "step": 3133 + }, + { + "epoch": 0.1750984719389893, + "grad_norm": 0.9221100211143494, + "learning_rate": 0.0009151445540116539, + "loss": 0.5165, + "step": 3134 + }, + { + "epoch": 0.17515434254267118, + "grad_norm": 1.87740957736969, + "learning_rate": 0.0009151165396683103, + "loss": 0.4126, + "step": 3135 + }, + { + "epoch": 0.17521021314635304, + "grad_norm": 1.2971426248550415, + "learning_rate": 0.0009150885253249665, + "loss": 0.4234, + "step": 3136 + }, + { + "epoch": 0.1752660837500349, + "grad_norm": 0.6354167461395264, + "learning_rate": 0.0009150605109816226, + "loss": 0.6093, + "step": 3137 + }, + { + "epoch": 0.17532195435371678, + "grad_norm": 0.790918231010437, + "learning_rate": 0.0009150324966382788, + "loss": 0.6204, + "step": 3138 + }, + { + "epoch": 0.17537782495739868, + "grad_norm": 1.2524199485778809, + "learning_rate": 0.0009150044822949351, + "loss": 0.5961, + "step": 3139 + }, + { + "epoch": 0.17543369556108054, + "grad_norm": 0.8077495098114014, + "learning_rate": 0.0009149764679515913, + "loss": 0.4794, + "step": 3140 + }, + { + "epoch": 0.1754895661647624, + "grad_norm": 2.7079081535339355, + "learning_rate": 0.0009149484536082474, + "loss": 0.4725, + "step": 3141 + }, + { + "epoch": 0.17554543676844428, + "grad_norm": 0.7630857825279236, + "learning_rate": 0.0009149204392649037, + "loss": 0.6111, + "step": 3142 + }, + { + "epoch": 0.17560130737212615, + "grad_norm": 0.5395917892456055, + "learning_rate": 0.0009148924249215599, + "loss": 0.5804, + "step": 3143 + }, + { + "epoch": 0.17565717797580802, + "grad_norm": 1.9205433130264282, + "learning_rate": 0.000914864410578216, + "loss": 0.3869, + "step": 3144 + }, + { + "epoch": 0.1757130485794899, + "grad_norm": 0.9752199649810791, + "learning_rate": 0.0009148363962348722, + "loss": 0.4086, + "step": 3145 + }, + { + "epoch": 0.17576891918317178, + "grad_norm": 0.6667129993438721, + "learning_rate": 0.0009148083818915285, + "loss": 0.5325, + "step": 3146 + }, + { + "epoch": 0.17582478978685365, + "grad_norm": 3.454848051071167, + "learning_rate": 0.0009147803675481847, + "loss": 0.4924, + "step": 3147 + }, + { + "epoch": 0.17588066039053551, + "grad_norm": 0.590476393699646, + "learning_rate": 0.0009147523532048408, + "loss": 0.4452, + "step": 3148 + }, + { + "epoch": 0.17593653099421738, + "grad_norm": 0.49714866280555725, + "learning_rate": 0.0009147243388614971, + "loss": 0.385, + "step": 3149 + }, + { + "epoch": 0.17599240159789928, + "grad_norm": 0.5889528393745422, + "learning_rate": 0.0009146963245181533, + "loss": 0.4358, + "step": 3150 + }, + { + "epoch": 0.17604827220158115, + "grad_norm": 1.0172359943389893, + "learning_rate": 0.0009146683101748095, + "loss": 0.4986, + "step": 3151 + }, + { + "epoch": 0.17610414280526301, + "grad_norm": 1.0529673099517822, + "learning_rate": 0.0009146402958314656, + "loss": 0.4279, + "step": 3152 + }, + { + "epoch": 0.17616001340894488, + "grad_norm": 1.2631455659866333, + "learning_rate": 0.000914612281488122, + "loss": 0.6346, + "step": 3153 + }, + { + "epoch": 0.17621588401262675, + "grad_norm": 0.5541217923164368, + "learning_rate": 0.0009145842671447782, + "loss": 0.423, + "step": 3154 + }, + { + "epoch": 0.17627175461630862, + "grad_norm": 0.4946008324623108, + "learning_rate": 0.0009145562528014344, + "loss": 0.5008, + "step": 3155 + }, + { + "epoch": 0.1763276252199905, + "grad_norm": 0.8072817325592041, + "learning_rate": 0.0009145282384580905, + "loss": 0.463, + "step": 3156 + }, + { + "epoch": 0.17638349582367238, + "grad_norm": 0.5313062071800232, + "learning_rate": 0.0009145002241147468, + "loss": 0.3914, + "step": 3157 + }, + { + "epoch": 0.17643936642735425, + "grad_norm": 3.1839725971221924, + "learning_rate": 0.000914472209771403, + "loss": 0.4301, + "step": 3158 + }, + { + "epoch": 0.17649523703103612, + "grad_norm": 0.655853271484375, + "learning_rate": 0.0009144441954280592, + "loss": 0.3803, + "step": 3159 + }, + { + "epoch": 0.17655110763471799, + "grad_norm": 0.48929905891418457, + "learning_rate": 0.0009144161810847154, + "loss": 0.4237, + "step": 3160 + }, + { + "epoch": 0.17660697823839985, + "grad_norm": 0.5070779323577881, + "learning_rate": 0.0009143881667413716, + "loss": 0.5758, + "step": 3161 + }, + { + "epoch": 0.17666284884208175, + "grad_norm": 0.7746042609214783, + "learning_rate": 0.0009143601523980278, + "loss": 0.6002, + "step": 3162 + }, + { + "epoch": 0.17671871944576362, + "grad_norm": 0.8474822044372559, + "learning_rate": 0.000914332138054684, + "loss": 0.389, + "step": 3163 + }, + { + "epoch": 0.17677459004944548, + "grad_norm": 0.9823592901229858, + "learning_rate": 0.0009143041237113402, + "loss": 0.5921, + "step": 3164 + }, + { + "epoch": 0.17683046065312735, + "grad_norm": 0.45890775322914124, + "learning_rate": 0.0009142761093679964, + "loss": 0.4864, + "step": 3165 + }, + { + "epoch": 0.17688633125680922, + "grad_norm": 0.4844875931739807, + "learning_rate": 0.0009142480950246526, + "loss": 0.4387, + "step": 3166 + }, + { + "epoch": 0.17694220186049112, + "grad_norm": 0.5198652744293213, + "learning_rate": 0.0009142200806813088, + "loss": 0.5214, + "step": 3167 + }, + { + "epoch": 0.17699807246417298, + "grad_norm": 4.8306145668029785, + "learning_rate": 0.000914192066337965, + "loss": 0.4684, + "step": 3168 + }, + { + "epoch": 0.17705394306785485, + "grad_norm": 0.5256745219230652, + "learning_rate": 0.0009141640519946213, + "loss": 0.4691, + "step": 3169 + }, + { + "epoch": 0.17710981367153672, + "grad_norm": 0.5314294099807739, + "learning_rate": 0.0009141360376512775, + "loss": 0.472, + "step": 3170 + }, + { + "epoch": 0.1771656842752186, + "grad_norm": 0.6808104515075684, + "learning_rate": 0.0009141080233079337, + "loss": 0.4651, + "step": 3171 + }, + { + "epoch": 0.17722155487890046, + "grad_norm": 2.808093309402466, + "learning_rate": 0.0009140800089645899, + "loss": 0.5755, + "step": 3172 + }, + { + "epoch": 0.17727742548258235, + "grad_norm": 0.7999137043952942, + "learning_rate": 0.0009140519946212461, + "loss": 0.4263, + "step": 3173 + }, + { + "epoch": 0.17733329608626422, + "grad_norm": 0.5510773062705994, + "learning_rate": 0.0009140239802779024, + "loss": 0.3618, + "step": 3174 + }, + { + "epoch": 0.1773891666899461, + "grad_norm": 0.7560214996337891, + "learning_rate": 0.0009139959659345585, + "loss": 0.4412, + "step": 3175 + }, + { + "epoch": 0.17744503729362796, + "grad_norm": 1.2376866340637207, + "learning_rate": 0.0009139679515912147, + "loss": 0.5553, + "step": 3176 + }, + { + "epoch": 0.17750090789730982, + "grad_norm": 1.3640494346618652, + "learning_rate": 0.0009139399372478709, + "loss": 0.4374, + "step": 3177 + }, + { + "epoch": 0.1775567785009917, + "grad_norm": 0.6274778842926025, + "learning_rate": 0.0009139119229045272, + "loss": 0.5852, + "step": 3178 + }, + { + "epoch": 0.1776126491046736, + "grad_norm": 2.0200607776641846, + "learning_rate": 0.0009138839085611833, + "loss": 0.5178, + "step": 3179 + }, + { + "epoch": 0.17766851970835545, + "grad_norm": 0.7359070777893066, + "learning_rate": 0.0009138558942178395, + "loss": 0.4797, + "step": 3180 + }, + { + "epoch": 0.17772439031203732, + "grad_norm": 2.802586793899536, + "learning_rate": 0.0009138278798744958, + "loss": 0.7395, + "step": 3181 + }, + { + "epoch": 0.1777802609157192, + "grad_norm": 0.4713343381881714, + "learning_rate": 0.000913799865531152, + "loss": 0.4702, + "step": 3182 + }, + { + "epoch": 0.17783613151940106, + "grad_norm": 0.6889004111289978, + "learning_rate": 0.0009137718511878081, + "loss": 0.4173, + "step": 3183 + }, + { + "epoch": 0.17789200212308293, + "grad_norm": 0.5345036387443542, + "learning_rate": 0.0009137438368444643, + "loss": 0.4065, + "step": 3184 + }, + { + "epoch": 0.17794787272676482, + "grad_norm": 0.667104184627533, + "learning_rate": 0.0009137158225011207, + "loss": 0.4046, + "step": 3185 + }, + { + "epoch": 0.1780037433304467, + "grad_norm": 6.631842136383057, + "learning_rate": 0.0009136878081577768, + "loss": 0.4267, + "step": 3186 + }, + { + "epoch": 0.17805961393412856, + "grad_norm": 1.2375357151031494, + "learning_rate": 0.000913659793814433, + "loss": 0.5047, + "step": 3187 + }, + { + "epoch": 0.17811548453781043, + "grad_norm": 0.7432243227958679, + "learning_rate": 0.0009136317794710893, + "loss": 0.4947, + "step": 3188 + }, + { + "epoch": 0.1781713551414923, + "grad_norm": 0.7305024266242981, + "learning_rate": 0.0009136037651277455, + "loss": 0.4347, + "step": 3189 + }, + { + "epoch": 0.1782272257451742, + "grad_norm": 0.5812109112739563, + "learning_rate": 0.0009135757507844016, + "loss": 0.562, + "step": 3190 + }, + { + "epoch": 0.17828309634885606, + "grad_norm": 0.8705951571464539, + "learning_rate": 0.0009135477364410578, + "loss": 0.8116, + "step": 3191 + }, + { + "epoch": 0.17833896695253793, + "grad_norm": 0.6702172756195068, + "learning_rate": 0.0009135197220977141, + "loss": 0.4378, + "step": 3192 + }, + { + "epoch": 0.1783948375562198, + "grad_norm": 0.4294123649597168, + "learning_rate": 0.0009134917077543703, + "loss": 0.4003, + "step": 3193 + }, + { + "epoch": 0.17845070815990166, + "grad_norm": 0.7481146454811096, + "learning_rate": 0.0009134636934110264, + "loss": 0.4322, + "step": 3194 + }, + { + "epoch": 0.17850657876358353, + "grad_norm": 1.2789626121520996, + "learning_rate": 0.0009134356790676826, + "loss": 0.389, + "step": 3195 + }, + { + "epoch": 0.17856244936726542, + "grad_norm": 0.7844642996788025, + "learning_rate": 0.0009134076647243389, + "loss": 0.5219, + "step": 3196 + }, + { + "epoch": 0.1786183199709473, + "grad_norm": 14.529525756835938, + "learning_rate": 0.0009133796503809951, + "loss": 0.3906, + "step": 3197 + }, + { + "epoch": 0.17867419057462916, + "grad_norm": 0.5129237174987793, + "learning_rate": 0.0009133516360376512, + "loss": 0.4456, + "step": 3198 + }, + { + "epoch": 0.17873006117831103, + "grad_norm": 0.5015367269515991, + "learning_rate": 0.0009133236216943075, + "loss": 0.4954, + "step": 3199 + }, + { + "epoch": 0.1787859317819929, + "grad_norm": 0.6601790189743042, + "learning_rate": 0.0009132956073509637, + "loss": 0.536, + "step": 3200 + }, + { + "epoch": 0.17884180238567476, + "grad_norm": 4.3263773918151855, + "learning_rate": 0.00091326759300762, + "loss": 0.4469, + "step": 3201 + }, + { + "epoch": 0.17889767298935666, + "grad_norm": 1.182437539100647, + "learning_rate": 0.000913239578664276, + "loss": 0.9002, + "step": 3202 + }, + { + "epoch": 0.17895354359303853, + "grad_norm": 0.38856711983680725, + "learning_rate": 0.0009132115643209324, + "loss": 0.3807, + "step": 3203 + }, + { + "epoch": 0.1790094141967204, + "grad_norm": 0.5731101036071777, + "learning_rate": 0.0009131835499775886, + "loss": 0.49, + "step": 3204 + }, + { + "epoch": 0.17906528480040226, + "grad_norm": 0.6795342564582825, + "learning_rate": 0.0009131555356342448, + "loss": 0.4195, + "step": 3205 + }, + { + "epoch": 0.17912115540408413, + "grad_norm": 0.9043541550636292, + "learning_rate": 0.000913127521290901, + "loss": 0.4784, + "step": 3206 + }, + { + "epoch": 0.17917702600776603, + "grad_norm": 0.9208500981330872, + "learning_rate": 0.0009130995069475572, + "loss": 0.5043, + "step": 3207 + }, + { + "epoch": 0.1792328966114479, + "grad_norm": 3.9519219398498535, + "learning_rate": 0.0009130714926042134, + "loss": 0.5507, + "step": 3208 + }, + { + "epoch": 0.17928876721512976, + "grad_norm": 0.5324614644050598, + "learning_rate": 0.0009130434782608695, + "loss": 0.4938, + "step": 3209 + }, + { + "epoch": 0.17934463781881163, + "grad_norm": 0.6885451078414917, + "learning_rate": 0.0009130154639175258, + "loss": 0.4607, + "step": 3210 + }, + { + "epoch": 0.1794005084224935, + "grad_norm": 0.47876593470573425, + "learning_rate": 0.000912987449574182, + "loss": 0.4349, + "step": 3211 + }, + { + "epoch": 0.17945637902617537, + "grad_norm": 1.615561604499817, + "learning_rate": 0.0009129594352308382, + "loss": 0.688, + "step": 3212 + }, + { + "epoch": 0.17951224962985726, + "grad_norm": 0.569739580154419, + "learning_rate": 0.0009129314208874944, + "loss": 0.5874, + "step": 3213 + }, + { + "epoch": 0.17956812023353913, + "grad_norm": 0.7025525569915771, + "learning_rate": 0.0009129034065441506, + "loss": 0.4164, + "step": 3214 + }, + { + "epoch": 0.179623990837221, + "grad_norm": 0.6531718373298645, + "learning_rate": 0.0009128753922008068, + "loss": 0.4064, + "step": 3215 + }, + { + "epoch": 0.17967986144090287, + "grad_norm": 0.5471765398979187, + "learning_rate": 0.000912847377857463, + "loss": 0.4664, + "step": 3216 + }, + { + "epoch": 0.17973573204458473, + "grad_norm": 1.5987255573272705, + "learning_rate": 0.0009128193635141192, + "loss": 0.5471, + "step": 3217 + }, + { + "epoch": 0.1797916026482666, + "grad_norm": 0.7898545265197754, + "learning_rate": 0.0009127913491707754, + "loss": 0.513, + "step": 3218 + }, + { + "epoch": 0.1798474732519485, + "grad_norm": 0.9465287923812866, + "learning_rate": 0.0009127633348274317, + "loss": 0.3902, + "step": 3219 + }, + { + "epoch": 0.17990334385563037, + "grad_norm": 0.7045376896858215, + "learning_rate": 0.000912735320484088, + "loss": 0.565, + "step": 3220 + }, + { + "epoch": 0.17995921445931223, + "grad_norm": 0.6421300172805786, + "learning_rate": 0.0009127073061407441, + "loss": 0.5416, + "step": 3221 + }, + { + "epoch": 0.1800150850629941, + "grad_norm": 0.5252315402030945, + "learning_rate": 0.0009126792917974003, + "loss": 0.4797, + "step": 3222 + }, + { + "epoch": 0.18007095566667597, + "grad_norm": 0.5433999300003052, + "learning_rate": 0.0009126512774540565, + "loss": 0.5885, + "step": 3223 + }, + { + "epoch": 0.18012682627035786, + "grad_norm": 1.1242016553878784, + "learning_rate": 0.0009126232631107128, + "loss": 0.6832, + "step": 3224 + }, + { + "epoch": 0.18018269687403973, + "grad_norm": 0.7803924679756165, + "learning_rate": 0.0009125952487673689, + "loss": 0.3913, + "step": 3225 + }, + { + "epoch": 0.1802385674777216, + "grad_norm": 1.322400689125061, + "learning_rate": 0.0009125672344240251, + "loss": 0.4884, + "step": 3226 + }, + { + "epoch": 0.18029443808140347, + "grad_norm": 0.6658082008361816, + "learning_rate": 0.0009125392200806814, + "loss": 0.5651, + "step": 3227 + }, + { + "epoch": 0.18035030868508534, + "grad_norm": 0.5535721778869629, + "learning_rate": 0.0009125112057373376, + "loss": 0.4719, + "step": 3228 + }, + { + "epoch": 0.1804061792887672, + "grad_norm": 0.6475862264633179, + "learning_rate": 0.0009124831913939937, + "loss": 0.6671, + "step": 3229 + }, + { + "epoch": 0.1804620498924491, + "grad_norm": 3.7221691608428955, + "learning_rate": 0.0009124551770506499, + "loss": 0.6865, + "step": 3230 + }, + { + "epoch": 0.18051792049613097, + "grad_norm": 1.3835757970809937, + "learning_rate": 0.0009124271627073062, + "loss": 0.4656, + "step": 3231 + }, + { + "epoch": 0.18057379109981284, + "grad_norm": 0.7288638353347778, + "learning_rate": 0.0009123991483639623, + "loss": 0.4898, + "step": 3232 + }, + { + "epoch": 0.1806296617034947, + "grad_norm": 0.5597330927848816, + "learning_rate": 0.0009123711340206185, + "loss": 0.4547, + "step": 3233 + }, + { + "epoch": 0.18068553230717657, + "grad_norm": 0.5052541494369507, + "learning_rate": 0.0009123431196772747, + "loss": 0.4737, + "step": 3234 + }, + { + "epoch": 0.18074140291085844, + "grad_norm": 0.4704730808734894, + "learning_rate": 0.000912315105333931, + "loss": 0.433, + "step": 3235 + }, + { + "epoch": 0.18079727351454034, + "grad_norm": 0.7136178016662598, + "learning_rate": 0.0009122870909905872, + "loss": 0.5567, + "step": 3236 + }, + { + "epoch": 0.1808531441182222, + "grad_norm": 0.7960693836212158, + "learning_rate": 0.0009122590766472434, + "loss": 0.4932, + "step": 3237 + }, + { + "epoch": 0.18090901472190407, + "grad_norm": 0.48540249466896057, + "learning_rate": 0.0009122310623038997, + "loss": 0.4431, + "step": 3238 + }, + { + "epoch": 0.18096488532558594, + "grad_norm": 0.5126280188560486, + "learning_rate": 0.0009122030479605559, + "loss": 0.4397, + "step": 3239 + }, + { + "epoch": 0.1810207559292678, + "grad_norm": 0.6612403392791748, + "learning_rate": 0.000912175033617212, + "loss": 0.4631, + "step": 3240 + }, + { + "epoch": 0.18107662653294967, + "grad_norm": 0.4474509358406067, + "learning_rate": 0.0009121470192738682, + "loss": 0.3371, + "step": 3241 + }, + { + "epoch": 0.18113249713663157, + "grad_norm": 0.47827214002609253, + "learning_rate": 0.0009121190049305245, + "loss": 0.4306, + "step": 3242 + }, + { + "epoch": 0.18118836774031344, + "grad_norm": 0.6153408288955688, + "learning_rate": 0.0009120909905871807, + "loss": 0.4588, + "step": 3243 + }, + { + "epoch": 0.1812442383439953, + "grad_norm": 0.5550630688667297, + "learning_rate": 0.0009120629762438368, + "loss": 0.473, + "step": 3244 + }, + { + "epoch": 0.18130010894767717, + "grad_norm": 0.6195087432861328, + "learning_rate": 0.0009120349619004931, + "loss": 0.4248, + "step": 3245 + }, + { + "epoch": 0.18135597955135904, + "grad_norm": 0.5990931391716003, + "learning_rate": 0.0009120069475571493, + "loss": 0.4969, + "step": 3246 + }, + { + "epoch": 0.18141185015504094, + "grad_norm": 1.6270360946655273, + "learning_rate": 0.0009119789332138055, + "loss": 0.4262, + "step": 3247 + }, + { + "epoch": 0.1814677207587228, + "grad_norm": 0.4994128346443176, + "learning_rate": 0.0009119509188704616, + "loss": 0.4426, + "step": 3248 + }, + { + "epoch": 0.18152359136240467, + "grad_norm": 0.759335458278656, + "learning_rate": 0.0009119229045271179, + "loss": 0.5353, + "step": 3249 + }, + { + "epoch": 0.18157946196608654, + "grad_norm": 1.014723777770996, + "learning_rate": 0.0009118948901837741, + "loss": 0.4807, + "step": 3250 + }, + { + "epoch": 0.1816353325697684, + "grad_norm": 1.0633926391601562, + "learning_rate": 0.0009118668758404302, + "loss": 0.3135, + "step": 3251 + }, + { + "epoch": 0.18169120317345028, + "grad_norm": 0.6327760219573975, + "learning_rate": 0.0009118388614970866, + "loss": 0.3801, + "step": 3252 + }, + { + "epoch": 0.18174707377713217, + "grad_norm": 1.0403140783309937, + "learning_rate": 0.0009118108471537428, + "loss": 0.6232, + "step": 3253 + }, + { + "epoch": 0.18180294438081404, + "grad_norm": 0.6405489444732666, + "learning_rate": 0.000911782832810399, + "loss": 0.4641, + "step": 3254 + }, + { + "epoch": 0.1818588149844959, + "grad_norm": 0.5029010772705078, + "learning_rate": 0.0009117548184670551, + "loss": 0.4819, + "step": 3255 + }, + { + "epoch": 0.18191468558817778, + "grad_norm": 0.7612084746360779, + "learning_rate": 0.0009117268041237114, + "loss": 0.4357, + "step": 3256 + }, + { + "epoch": 0.18197055619185964, + "grad_norm": 0.7984657883644104, + "learning_rate": 0.0009116987897803676, + "loss": 0.4332, + "step": 3257 + }, + { + "epoch": 0.1820264267955415, + "grad_norm": 0.8375059962272644, + "learning_rate": 0.0009116707754370238, + "loss": 0.5706, + "step": 3258 + }, + { + "epoch": 0.1820822973992234, + "grad_norm": 0.8893278241157532, + "learning_rate": 0.00091164276109368, + "loss": 0.4427, + "step": 3259 + }, + { + "epoch": 0.18213816800290528, + "grad_norm": 0.8220874071121216, + "learning_rate": 0.0009116147467503362, + "loss": 0.493, + "step": 3260 + }, + { + "epoch": 0.18219403860658714, + "grad_norm": 0.9383020401000977, + "learning_rate": 0.0009115867324069924, + "loss": 0.4841, + "step": 3261 + }, + { + "epoch": 0.182249909210269, + "grad_norm": 0.6462740898132324, + "learning_rate": 0.0009115587180636486, + "loss": 0.514, + "step": 3262 + }, + { + "epoch": 0.18230577981395088, + "grad_norm": 0.4991518259048462, + "learning_rate": 0.0009115307037203048, + "loss": 0.4544, + "step": 3263 + }, + { + "epoch": 0.18236165041763278, + "grad_norm": 0.649513304233551, + "learning_rate": 0.000911502689376961, + "loss": 0.6028, + "step": 3264 + }, + { + "epoch": 0.18241752102131464, + "grad_norm": 0.6755000948905945, + "learning_rate": 0.0009114746750336172, + "loss": 0.4238, + "step": 3265 + }, + { + "epoch": 0.1824733916249965, + "grad_norm": 0.8998586535453796, + "learning_rate": 0.0009114466606902735, + "loss": 0.5607, + "step": 3266 + }, + { + "epoch": 0.18252926222867838, + "grad_norm": 1.962577223777771, + "learning_rate": 0.0009114186463469296, + "loss": 0.377, + "step": 3267 + }, + { + "epoch": 0.18258513283236025, + "grad_norm": 0.7597375512123108, + "learning_rate": 0.0009113906320035858, + "loss": 0.5112, + "step": 3268 + }, + { + "epoch": 0.18264100343604212, + "grad_norm": 0.456163614988327, + "learning_rate": 0.000911362617660242, + "loss": 0.4033, + "step": 3269 + }, + { + "epoch": 0.182696874039724, + "grad_norm": 0.6165705323219299, + "learning_rate": 0.0009113346033168984, + "loss": 0.4666, + "step": 3270 + }, + { + "epoch": 0.18275274464340588, + "grad_norm": 0.5977029800415039, + "learning_rate": 0.0009113065889735545, + "loss": 0.4968, + "step": 3271 + }, + { + "epoch": 0.18280861524708775, + "grad_norm": 0.45970308780670166, + "learning_rate": 0.0009112785746302107, + "loss": 0.4395, + "step": 3272 + }, + { + "epoch": 0.18286448585076961, + "grad_norm": 0.37348729372024536, + "learning_rate": 0.0009112505602868669, + "loss": 0.3788, + "step": 3273 + }, + { + "epoch": 0.18292035645445148, + "grad_norm": 0.5500589609146118, + "learning_rate": 0.0009112225459435231, + "loss": 0.4854, + "step": 3274 + }, + { + "epoch": 0.18297622705813335, + "grad_norm": 0.5475522875785828, + "learning_rate": 0.0009111945316001793, + "loss": 0.5381, + "step": 3275 + }, + { + "epoch": 0.18303209766181525, + "grad_norm": 0.899115800857544, + "learning_rate": 0.0009111665172568355, + "loss": 0.4565, + "step": 3276 + }, + { + "epoch": 0.18308796826549711, + "grad_norm": 0.5856340527534485, + "learning_rate": 0.0009111385029134918, + "loss": 0.4595, + "step": 3277 + }, + { + "epoch": 0.18314383886917898, + "grad_norm": 0.9175402522087097, + "learning_rate": 0.0009111104885701479, + "loss": 0.5016, + "step": 3278 + }, + { + "epoch": 0.18319970947286085, + "grad_norm": 0.47064724564552307, + "learning_rate": 0.0009110824742268041, + "loss": 0.5008, + "step": 3279 + }, + { + "epoch": 0.18325558007654272, + "grad_norm": 2.5907483100891113, + "learning_rate": 0.0009110544598834603, + "loss": 0.5359, + "step": 3280 + }, + { + "epoch": 0.1833114506802246, + "grad_norm": 0.7434201836585999, + "learning_rate": 0.0009110264455401166, + "loss": 0.4747, + "step": 3281 + }, + { + "epoch": 0.18336732128390648, + "grad_norm": 0.9901055097579956, + "learning_rate": 0.0009109984311967727, + "loss": 0.6939, + "step": 3282 + }, + { + "epoch": 0.18342319188758835, + "grad_norm": 0.7393150329589844, + "learning_rate": 0.0009109704168534289, + "loss": 0.4919, + "step": 3283 + }, + { + "epoch": 0.18347906249127022, + "grad_norm": 0.7317896485328674, + "learning_rate": 0.0009109424025100852, + "loss": 0.512, + "step": 3284 + }, + { + "epoch": 0.18353493309495209, + "grad_norm": 1.2751895189285278, + "learning_rate": 0.0009109143881667415, + "loss": 0.5602, + "step": 3285 + }, + { + "epoch": 0.18359080369863395, + "grad_norm": 1.1691102981567383, + "learning_rate": 0.0009108863738233975, + "loss": 0.4298, + "step": 3286 + }, + { + "epoch": 0.18364667430231585, + "grad_norm": 0.6030598282814026, + "learning_rate": 0.0009108583594800538, + "loss": 0.4349, + "step": 3287 + }, + { + "epoch": 0.18370254490599772, + "grad_norm": 0.4514387249946594, + "learning_rate": 0.0009108303451367101, + "loss": 0.3797, + "step": 3288 + }, + { + "epoch": 0.18375841550967958, + "grad_norm": 1.5449548959732056, + "learning_rate": 0.0009108023307933663, + "loss": 0.5098, + "step": 3289 + }, + { + "epoch": 0.18381428611336145, + "grad_norm": 4.418604850769043, + "learning_rate": 0.0009107743164500224, + "loss": 0.5082, + "step": 3290 + }, + { + "epoch": 0.18387015671704332, + "grad_norm": 5.144406318664551, + "learning_rate": 0.0009107463021066787, + "loss": 0.362, + "step": 3291 + }, + { + "epoch": 0.1839260273207252, + "grad_norm": 0.6655681133270264, + "learning_rate": 0.0009107182877633349, + "loss": 0.6338, + "step": 3292 + }, + { + "epoch": 0.18398189792440708, + "grad_norm": 0.9282097220420837, + "learning_rate": 0.000910690273419991, + "loss": 0.7506, + "step": 3293 + }, + { + "epoch": 0.18403776852808895, + "grad_norm": 4.131616592407227, + "learning_rate": 0.0009106622590766472, + "loss": 0.4004, + "step": 3294 + }, + { + "epoch": 0.18409363913177082, + "grad_norm": 0.9418845772743225, + "learning_rate": 0.0009106342447333035, + "loss": 0.6244, + "step": 3295 + }, + { + "epoch": 0.1841495097354527, + "grad_norm": 0.7499305605888367, + "learning_rate": 0.0009106062303899597, + "loss": 0.4448, + "step": 3296 + }, + { + "epoch": 0.18420538033913456, + "grad_norm": 0.4821210205554962, + "learning_rate": 0.0009105782160466158, + "loss": 0.3905, + "step": 3297 + }, + { + "epoch": 0.18426125094281642, + "grad_norm": 1.038218379020691, + "learning_rate": 0.0009105502017032721, + "loss": 0.4821, + "step": 3298 + }, + { + "epoch": 0.18431712154649832, + "grad_norm": 0.7028923630714417, + "learning_rate": 0.0009105221873599283, + "loss": 0.4301, + "step": 3299 + }, + { + "epoch": 0.1843729921501802, + "grad_norm": 0.5253230333328247, + "learning_rate": 0.0009104941730165845, + "loss": 0.4947, + "step": 3300 + }, + { + "epoch": 0.18442886275386206, + "grad_norm": 0.608197033405304, + "learning_rate": 0.0009104661586732406, + "loss": 0.5263, + "step": 3301 + }, + { + "epoch": 0.18448473335754392, + "grad_norm": Infinity, + "learning_rate": 0.0009104661586732406, + "loss": 0.5853, + "step": 3302 + }, + { + "epoch": 0.1845406039612258, + "grad_norm": 0.5942262411117554, + "learning_rate": 0.000910438144329897, + "loss": 0.4473, + "step": 3303 + }, + { + "epoch": 0.1845964745649077, + "grad_norm": 0.8028691411018372, + "learning_rate": 0.0009104101299865532, + "loss": 0.4467, + "step": 3304 + }, + { + "epoch": 0.18465234516858955, + "grad_norm": 0.6163011193275452, + "learning_rate": 0.0009103821156432094, + "loss": 0.4911, + "step": 3305 + }, + { + "epoch": 0.18470821577227142, + "grad_norm": 0.6428738832473755, + "learning_rate": 0.0009103541012998655, + "loss": 0.5585, + "step": 3306 + }, + { + "epoch": 0.1847640863759533, + "grad_norm": 0.8852730393409729, + "learning_rate": 0.0009103260869565218, + "loss": 0.5651, + "step": 3307 + }, + { + "epoch": 0.18481995697963516, + "grad_norm": 0.6504756808280945, + "learning_rate": 0.000910298072613178, + "loss": 0.4807, + "step": 3308 + }, + { + "epoch": 0.18487582758331703, + "grad_norm": 0.740881621837616, + "learning_rate": 0.0009102700582698342, + "loss": 0.5636, + "step": 3309 + }, + { + "epoch": 0.18493169818699892, + "grad_norm": 0.4179275333881378, + "learning_rate": 0.0009102420439264904, + "loss": 0.591, + "step": 3310 + }, + { + "epoch": 0.1849875687906808, + "grad_norm": 0.48489660024642944, + "learning_rate": 0.0009102140295831466, + "loss": 0.4444, + "step": 3311 + }, + { + "epoch": 0.18504343939436266, + "grad_norm": 0.6403635740280151, + "learning_rate": 0.0009101860152398028, + "loss": 0.4281, + "step": 3312 + }, + { + "epoch": 0.18509930999804453, + "grad_norm": 0.9520873427391052, + "learning_rate": 0.000910158000896459, + "loss": 0.4387, + "step": 3313 + }, + { + "epoch": 0.1851551806017264, + "grad_norm": 5.272343635559082, + "learning_rate": 0.0009101299865531152, + "loss": 0.4936, + "step": 3314 + }, + { + "epoch": 0.18521105120540826, + "grad_norm": 0.6673287749290466, + "learning_rate": 0.0009101019722097714, + "loss": 0.4167, + "step": 3315 + }, + { + "epoch": 0.18526692180909016, + "grad_norm": 0.8553990125656128, + "learning_rate": 0.0009100739578664276, + "loss": 0.4144, + "step": 3316 + }, + { + "epoch": 0.18532279241277202, + "grad_norm": 0.6990291476249695, + "learning_rate": 0.0009100459435230838, + "loss": 0.6392, + "step": 3317 + }, + { + "epoch": 0.1853786630164539, + "grad_norm": 0.6664606928825378, + "learning_rate": 0.00091001792917974, + "loss": 0.4745, + "step": 3318 + }, + { + "epoch": 0.18543453362013576, + "grad_norm": 0.47338104248046875, + "learning_rate": 0.0009099899148363962, + "loss": 0.4678, + "step": 3319 + }, + { + "epoch": 0.18549040422381763, + "grad_norm": 0.6400175094604492, + "learning_rate": 0.0009099619004930525, + "loss": 0.5829, + "step": 3320 + }, + { + "epoch": 0.18554627482749952, + "grad_norm": 0.9702564477920532, + "learning_rate": 0.0009099338861497087, + "loss": 0.5725, + "step": 3321 + }, + { + "epoch": 0.1856021454311814, + "grad_norm": 0.7059080600738525, + "learning_rate": 0.0009099058718063649, + "loss": 0.4952, + "step": 3322 + }, + { + "epoch": 0.18565801603486326, + "grad_norm": 0.6894605755805969, + "learning_rate": 0.0009098778574630211, + "loss": 0.5356, + "step": 3323 + }, + { + "epoch": 0.18571388663854513, + "grad_norm": 0.6558418869972229, + "learning_rate": 0.0009098498431196774, + "loss": 0.4896, + "step": 3324 + }, + { + "epoch": 0.185769757242227, + "grad_norm": 0.5089638829231262, + "learning_rate": 0.0009098218287763335, + "loss": 0.5839, + "step": 3325 + }, + { + "epoch": 0.18582562784590886, + "grad_norm": 1.8952771425247192, + "learning_rate": 0.0009097938144329897, + "loss": 0.6008, + "step": 3326 + }, + { + "epoch": 0.18588149844959076, + "grad_norm": 0.5059298872947693, + "learning_rate": 0.0009097658000896459, + "loss": 0.5794, + "step": 3327 + }, + { + "epoch": 0.18593736905327263, + "grad_norm": 1.0685532093048096, + "learning_rate": 0.0009097377857463022, + "loss": 0.5712, + "step": 3328 + }, + { + "epoch": 0.1859932396569545, + "grad_norm": 0.6401758790016174, + "learning_rate": 0.0009097097714029583, + "loss": 0.4651, + "step": 3329 + }, + { + "epoch": 0.18604911026063636, + "grad_norm": 0.49053457379341125, + "learning_rate": 0.0009096817570596145, + "loss": 0.3884, + "step": 3330 + }, + { + "epoch": 0.18610498086431823, + "grad_norm": 0.83258455991745, + "learning_rate": 0.0009096537427162708, + "loss": 0.5107, + "step": 3331 + }, + { + "epoch": 0.1861608514680001, + "grad_norm": 0.4602666199207306, + "learning_rate": 0.000909625728372927, + "loss": 0.5379, + "step": 3332 + }, + { + "epoch": 0.186216722071682, + "grad_norm": 0.8814432621002197, + "learning_rate": 0.0009095977140295831, + "loss": 0.5017, + "step": 3333 + }, + { + "epoch": 0.18627259267536386, + "grad_norm": 0.5462316274642944, + "learning_rate": 0.0009095696996862393, + "loss": 0.4603, + "step": 3334 + }, + { + "epoch": 0.18632846327904573, + "grad_norm": 0.8106711506843567, + "learning_rate": 0.0009095416853428956, + "loss": 0.5156, + "step": 3335 + }, + { + "epoch": 0.1863843338827276, + "grad_norm": 0.49242377281188965, + "learning_rate": 0.0009095136709995517, + "loss": 0.4095, + "step": 3336 + }, + { + "epoch": 0.18644020448640947, + "grad_norm": 0.6854003667831421, + "learning_rate": 0.000909485656656208, + "loss": 0.5267, + "step": 3337 + }, + { + "epoch": 0.18649607509009136, + "grad_norm": 0.7229044437408447, + "learning_rate": 0.0009094576423128642, + "loss": 0.6576, + "step": 3338 + }, + { + "epoch": 0.18655194569377323, + "grad_norm": 0.6781089305877686, + "learning_rate": 0.0009094296279695205, + "loss": 0.5432, + "step": 3339 + }, + { + "epoch": 0.1866078162974551, + "grad_norm": 0.6528939604759216, + "learning_rate": 0.0009094016136261766, + "loss": 0.6265, + "step": 3340 + }, + { + "epoch": 0.18666368690113697, + "grad_norm": 3.8742380142211914, + "learning_rate": 0.0009093735992828328, + "loss": 0.5018, + "step": 3341 + }, + { + "epoch": 0.18671955750481883, + "grad_norm": 0.6768704652786255, + "learning_rate": 0.0009093455849394891, + "loss": 0.4991, + "step": 3342 + }, + { + "epoch": 0.1867754281085007, + "grad_norm": 0.5993204116821289, + "learning_rate": 0.0009093175705961453, + "loss": 0.4965, + "step": 3343 + }, + { + "epoch": 0.1868312987121826, + "grad_norm": 0.7310294508934021, + "learning_rate": 0.0009092895562528014, + "loss": 0.423, + "step": 3344 + }, + { + "epoch": 0.18688716931586447, + "grad_norm": 0.45503008365631104, + "learning_rate": 0.0009092615419094576, + "loss": 0.5303, + "step": 3345 + }, + { + "epoch": 0.18694303991954633, + "grad_norm": 1.2170852422714233, + "learning_rate": 0.0009092335275661139, + "loss": 0.5037, + "step": 3346 + }, + { + "epoch": 0.1869989105232282, + "grad_norm": 0.5183893442153931, + "learning_rate": 0.0009092055132227701, + "loss": 0.4058, + "step": 3347 + }, + { + "epoch": 0.18705478112691007, + "grad_norm": 0.6611664891242981, + "learning_rate": 0.0009091774988794262, + "loss": 0.4511, + "step": 3348 + }, + { + "epoch": 0.18711065173059194, + "grad_norm": 0.5041138529777527, + "learning_rate": 0.0009091494845360825, + "loss": 0.4822, + "step": 3349 + }, + { + "epoch": 0.18716652233427383, + "grad_norm": 0.630629301071167, + "learning_rate": 0.0009091214701927387, + "loss": 0.43, + "step": 3350 + }, + { + "epoch": 0.1872223929379557, + "grad_norm": 0.513480007648468, + "learning_rate": 0.0009090934558493949, + "loss": 0.5409, + "step": 3351 + }, + { + "epoch": 0.18727826354163757, + "grad_norm": 0.45948415994644165, + "learning_rate": 0.000909065441506051, + "loss": 0.4148, + "step": 3352 + }, + { + "epoch": 0.18733413414531944, + "grad_norm": 0.5706478357315063, + "learning_rate": 0.0009090374271627074, + "loss": 0.475, + "step": 3353 + }, + { + "epoch": 0.1873900047490013, + "grad_norm": 0.5666611194610596, + "learning_rate": 0.0009090094128193636, + "loss": 0.5309, + "step": 3354 + }, + { + "epoch": 0.18744587535268317, + "grad_norm": 3.154940605163574, + "learning_rate": 0.0009089813984760198, + "loss": 0.5229, + "step": 3355 + }, + { + "epoch": 0.18750174595636507, + "grad_norm": 0.8829767107963562, + "learning_rate": 0.000908953384132676, + "loss": 0.5119, + "step": 3356 + }, + { + "epoch": 0.18755761656004694, + "grad_norm": 0.8963273167610168, + "learning_rate": 0.0009089253697893322, + "loss": 0.4502, + "step": 3357 + }, + { + "epoch": 0.1876134871637288, + "grad_norm": 0.6722939014434814, + "learning_rate": 0.0009088973554459884, + "loss": 0.4537, + "step": 3358 + }, + { + "epoch": 0.18766935776741067, + "grad_norm": 0.7148534655570984, + "learning_rate": 0.0009088693411026445, + "loss": 0.4591, + "step": 3359 + }, + { + "epoch": 0.18772522837109254, + "grad_norm": 1.0010405778884888, + "learning_rate": 0.0009088413267593008, + "loss": 0.5357, + "step": 3360 + }, + { + "epoch": 0.18778109897477444, + "grad_norm": 0.5154001712799072, + "learning_rate": 0.000908813312415957, + "loss": 0.5021, + "step": 3361 + }, + { + "epoch": 0.1878369695784563, + "grad_norm": 2.019460439682007, + "learning_rate": 0.0009087852980726132, + "loss": 0.4105, + "step": 3362 + }, + { + "epoch": 0.18789284018213817, + "grad_norm": 0.7014985084533691, + "learning_rate": 0.0009087572837292694, + "loss": 0.6079, + "step": 3363 + }, + { + "epoch": 0.18794871078582004, + "grad_norm": 0.5529154539108276, + "learning_rate": 0.0009087292693859256, + "loss": 0.4157, + "step": 3364 + }, + { + "epoch": 0.1880045813895019, + "grad_norm": 0.8485970497131348, + "learning_rate": 0.0009087012550425818, + "loss": 0.4375, + "step": 3365 + }, + { + "epoch": 0.18806045199318377, + "grad_norm": 0.8746246695518494, + "learning_rate": 0.000908673240699238, + "loss": 0.4538, + "step": 3366 + }, + { + "epoch": 0.18811632259686567, + "grad_norm": 2.051299810409546, + "learning_rate": 0.0009086452263558942, + "loss": 0.6339, + "step": 3367 + }, + { + "epoch": 0.18817219320054754, + "grad_norm": 0.9032443165779114, + "learning_rate": 0.0009086172120125504, + "loss": 0.7125, + "step": 3368 + }, + { + "epoch": 0.1882280638042294, + "grad_norm": 0.6318488121032715, + "learning_rate": 0.0009085891976692066, + "loss": 0.4437, + "step": 3369 + }, + { + "epoch": 0.18828393440791127, + "grad_norm": 0.5517737865447998, + "learning_rate": 0.000908561183325863, + "loss": 0.4596, + "step": 3370 + }, + { + "epoch": 0.18833980501159314, + "grad_norm": 0.5358610153198242, + "learning_rate": 0.0009085331689825191, + "loss": 0.5026, + "step": 3371 + }, + { + "epoch": 0.188395675615275, + "grad_norm": 0.913818895816803, + "learning_rate": 0.0009085051546391753, + "loss": 0.4617, + "step": 3372 + }, + { + "epoch": 0.1884515462189569, + "grad_norm": 0.6401166319847107, + "learning_rate": 0.0009084771402958315, + "loss": 0.5978, + "step": 3373 + }, + { + "epoch": 0.18850741682263877, + "grad_norm": 0.5660611987113953, + "learning_rate": 0.0009084491259524878, + "loss": 0.5003, + "step": 3374 + }, + { + "epoch": 0.18856328742632064, + "grad_norm": 0.5682265162467957, + "learning_rate": 0.0009084211116091439, + "loss": 0.4292, + "step": 3375 + }, + { + "epoch": 0.1886191580300025, + "grad_norm": 0.5448540449142456, + "learning_rate": 0.0009083930972658001, + "loss": 0.5384, + "step": 3376 + }, + { + "epoch": 0.18867502863368438, + "grad_norm": 0.46684953570365906, + "learning_rate": 0.0009083650829224563, + "loss": 0.5064, + "step": 3377 + }, + { + "epoch": 0.18873089923736627, + "grad_norm": 1.4197136163711548, + "learning_rate": 0.0009083370685791126, + "loss": 0.575, + "step": 3378 + }, + { + "epoch": 0.18878676984104814, + "grad_norm": 0.7055583000183105, + "learning_rate": 0.0009083090542357687, + "loss": 0.503, + "step": 3379 + }, + { + "epoch": 0.18884264044473, + "grad_norm": 1.7543485164642334, + "learning_rate": 0.0009082810398924249, + "loss": 0.471, + "step": 3380 + }, + { + "epoch": 0.18889851104841188, + "grad_norm": 0.7002995014190674, + "learning_rate": 0.0009082530255490812, + "loss": 0.4454, + "step": 3381 + }, + { + "epoch": 0.18895438165209374, + "grad_norm": 0.67257159948349, + "learning_rate": 0.0009082250112057373, + "loss": 0.4519, + "step": 3382 + }, + { + "epoch": 0.1890102522557756, + "grad_norm": 0.4257473647594452, + "learning_rate": 0.0009081969968623935, + "loss": 0.3966, + "step": 3383 + }, + { + "epoch": 0.1890661228594575, + "grad_norm": 4.648218631744385, + "learning_rate": 0.0009081689825190497, + "loss": 0.4237, + "step": 3384 + }, + { + "epoch": 0.18912199346313938, + "grad_norm": 0.7169526815414429, + "learning_rate": 0.000908140968175706, + "loss": 0.4474, + "step": 3385 + }, + { + "epoch": 0.18917786406682124, + "grad_norm": 0.5232897400856018, + "learning_rate": 0.0009081129538323621, + "loss": 0.4765, + "step": 3386 + }, + { + "epoch": 0.1892337346705031, + "grad_norm": 0.8710025548934937, + "learning_rate": 0.0009080849394890183, + "loss": 0.6688, + "step": 3387 + }, + { + "epoch": 0.18928960527418498, + "grad_norm": 0.7977496385574341, + "learning_rate": 0.0009080569251456747, + "loss": 0.4748, + "step": 3388 + }, + { + "epoch": 0.18934547587786685, + "grad_norm": 0.8349243998527527, + "learning_rate": 0.0009080289108023309, + "loss": 0.4681, + "step": 3389 + }, + { + "epoch": 0.18940134648154874, + "grad_norm": 0.5444942712783813, + "learning_rate": 0.000908000896458987, + "loss": 0.561, + "step": 3390 + }, + { + "epoch": 0.1894572170852306, + "grad_norm": 0.879895031452179, + "learning_rate": 0.0009079728821156432, + "loss": 0.4314, + "step": 3391 + }, + { + "epoch": 0.18951308768891248, + "grad_norm": 0.6728169918060303, + "learning_rate": 0.0009079448677722995, + "loss": 0.4507, + "step": 3392 + }, + { + "epoch": 0.18956895829259435, + "grad_norm": 1.1526408195495605, + "learning_rate": 0.0009079168534289557, + "loss": 0.471, + "step": 3393 + }, + { + "epoch": 0.18962482889627622, + "grad_norm": 0.4240914285182953, + "learning_rate": 0.0009078888390856118, + "loss": 0.4173, + "step": 3394 + }, + { + "epoch": 0.1896806994999581, + "grad_norm": 0.8078109622001648, + "learning_rate": 0.0009078608247422681, + "loss": 0.424, + "step": 3395 + }, + { + "epoch": 0.18973657010363998, + "grad_norm": 0.6191677451133728, + "learning_rate": 0.0009078328103989243, + "loss": 0.5082, + "step": 3396 + }, + { + "epoch": 0.18979244070732185, + "grad_norm": 0.584273099899292, + "learning_rate": 0.0009078047960555805, + "loss": 0.5497, + "step": 3397 + }, + { + "epoch": 0.18984831131100371, + "grad_norm": 0.5061673521995544, + "learning_rate": 0.0009077767817122366, + "loss": 0.4782, + "step": 3398 + }, + { + "epoch": 0.18990418191468558, + "grad_norm": 0.6307739615440369, + "learning_rate": 0.0009077487673688929, + "loss": 0.5489, + "step": 3399 + }, + { + "epoch": 0.18996005251836745, + "grad_norm": 0.551802933216095, + "learning_rate": 0.0009077207530255491, + "loss": 0.3992, + "step": 3400 + }, + { + "epoch": 0.19001592312204935, + "grad_norm": 0.6044238805770874, + "learning_rate": 0.0009076927386822052, + "loss": 0.4689, + "step": 3401 + }, + { + "epoch": 0.19007179372573121, + "grad_norm": 1.2252622842788696, + "learning_rate": 0.0009076647243388615, + "loss": 0.6255, + "step": 3402 + }, + { + "epoch": 0.19012766432941308, + "grad_norm": 0.6713767647743225, + "learning_rate": 0.0009076367099955177, + "loss": 0.4359, + "step": 3403 + }, + { + "epoch": 0.19018353493309495, + "grad_norm": 0.5145004391670227, + "learning_rate": 0.000907608695652174, + "loss": 0.4863, + "step": 3404 + }, + { + "epoch": 0.19023940553677682, + "grad_norm": 0.5176512598991394, + "learning_rate": 0.00090758068130883, + "loss": 0.5874, + "step": 3405 + }, + { + "epoch": 0.19029527614045869, + "grad_norm": 0.6093581914901733, + "learning_rate": 0.0009075526669654864, + "loss": 0.5589, + "step": 3406 + }, + { + "epoch": 0.19035114674414058, + "grad_norm": 0.7026590704917908, + "learning_rate": 0.0009075246526221426, + "loss": 0.5129, + "step": 3407 + }, + { + "epoch": 0.19040701734782245, + "grad_norm": 0.8545080423355103, + "learning_rate": 0.0009074966382787988, + "loss": 0.5218, + "step": 3408 + }, + { + "epoch": 0.19046288795150432, + "grad_norm": 0.7119805812835693, + "learning_rate": 0.0009074686239354549, + "loss": 0.4122, + "step": 3409 + }, + { + "epoch": 0.19051875855518619, + "grad_norm": 1.1662635803222656, + "learning_rate": 0.0009074406095921112, + "loss": 0.4782, + "step": 3410 + }, + { + "epoch": 0.19057462915886805, + "grad_norm": 0.9559227228164673, + "learning_rate": 0.0009074125952487674, + "loss": 0.4834, + "step": 3411 + }, + { + "epoch": 0.19063049976254992, + "grad_norm": 4.0002360343933105, + "learning_rate": 0.0009073845809054236, + "loss": 0.452, + "step": 3412 + }, + { + "epoch": 0.19068637036623182, + "grad_norm": 1.2418535947799683, + "learning_rate": 0.0009073565665620798, + "loss": 0.5468, + "step": 3413 + }, + { + "epoch": 0.19074224096991368, + "grad_norm": 0.6499142050743103, + "learning_rate": 0.000907328552218736, + "loss": 0.4394, + "step": 3414 + }, + { + "epoch": 0.19079811157359555, + "grad_norm": 1.269054651260376, + "learning_rate": 0.0009073005378753922, + "loss": 0.5024, + "step": 3415 + }, + { + "epoch": 0.19085398217727742, + "grad_norm": 0.9319174885749817, + "learning_rate": 0.0009072725235320484, + "loss": 0.48, + "step": 3416 + }, + { + "epoch": 0.1909098527809593, + "grad_norm": 0.5049799084663391, + "learning_rate": 0.0009072445091887046, + "loss": 0.6361, + "step": 3417 + }, + { + "epoch": 0.19096572338464118, + "grad_norm": 0.5761445164680481, + "learning_rate": 0.0009072164948453608, + "loss": 0.4502, + "step": 3418 + }, + { + "epoch": 0.19102159398832305, + "grad_norm": 0.8635989427566528, + "learning_rate": 0.000907188480502017, + "loss": 0.4513, + "step": 3419 + }, + { + "epoch": 0.19107746459200492, + "grad_norm": 1.6966564655303955, + "learning_rate": 0.0009071604661586734, + "loss": 0.5046, + "step": 3420 + }, + { + "epoch": 0.1911333351956868, + "grad_norm": 4.119761943817139, + "learning_rate": 0.0009071324518153295, + "loss": 0.5076, + "step": 3421 + }, + { + "epoch": 0.19118920579936866, + "grad_norm": 0.7383440732955933, + "learning_rate": 0.0009071044374719857, + "loss": 0.5495, + "step": 3422 + }, + { + "epoch": 0.19124507640305052, + "grad_norm": 0.4985031187534332, + "learning_rate": 0.0009070764231286419, + "loss": 0.4304, + "step": 3423 + }, + { + "epoch": 0.19130094700673242, + "grad_norm": Infinity, + "learning_rate": 0.0009070764231286419, + "loss": 0.3802, + "step": 3424 + }, + { + "epoch": 0.1913568176104143, + "grad_norm": 15.070128440856934, + "learning_rate": 0.0009070484087852981, + "loss": 0.3875, + "step": 3425 + }, + { + "epoch": 0.19141268821409615, + "grad_norm": 0.9775311350822449, + "learning_rate": 0.0009070203944419543, + "loss": 0.4643, + "step": 3426 + }, + { + "epoch": 0.19146855881777802, + "grad_norm": 0.9591082334518433, + "learning_rate": 0.0009069923800986105, + "loss": 0.5848, + "step": 3427 + }, + { + "epoch": 0.1915244294214599, + "grad_norm": 0.6319851875305176, + "learning_rate": 0.0009069643657552668, + "loss": 0.4787, + "step": 3428 + }, + { + "epoch": 0.19158030002514176, + "grad_norm": 0.5997673273086548, + "learning_rate": 0.0009069363514119229, + "loss": 0.5686, + "step": 3429 + }, + { + "epoch": 0.19163617062882365, + "grad_norm": 0.7031666040420532, + "learning_rate": 0.0009069083370685791, + "loss": 0.4976, + "step": 3430 + }, + { + "epoch": 0.19169204123250552, + "grad_norm": 1.0295140743255615, + "learning_rate": 0.0009068803227252353, + "loss": 0.4558, + "step": 3431 + }, + { + "epoch": 0.1917479118361874, + "grad_norm": 0.5224736332893372, + "learning_rate": 0.0009068523083818916, + "loss": 0.3925, + "step": 3432 + }, + { + "epoch": 0.19180378243986926, + "grad_norm": 0.5192094445228577, + "learning_rate": 0.0009068242940385477, + "loss": 0.4428, + "step": 3433 + }, + { + "epoch": 0.19185965304355113, + "grad_norm": 0.9556205868721008, + "learning_rate": 0.0009067962796952039, + "loss": 0.4719, + "step": 3434 + }, + { + "epoch": 0.19191552364723302, + "grad_norm": 0.5464856624603271, + "learning_rate": 0.0009067682653518602, + "loss": 0.5219, + "step": 3435 + }, + { + "epoch": 0.1919713942509149, + "grad_norm": 0.5340002775192261, + "learning_rate": 0.0009067402510085164, + "loss": 0.5351, + "step": 3436 + }, + { + "epoch": 0.19202726485459676, + "grad_norm": 1.0901850461959839, + "learning_rate": 0.0009067122366651725, + "loss": 0.6043, + "step": 3437 + }, + { + "epoch": 0.19208313545827863, + "grad_norm": 0.5335768461227417, + "learning_rate": 0.0009066842223218287, + "loss": 0.3244, + "step": 3438 + }, + { + "epoch": 0.1921390060619605, + "grad_norm": 0.6579670906066895, + "learning_rate": 0.0009066562079784851, + "loss": 0.5547, + "step": 3439 + }, + { + "epoch": 0.19219487666564236, + "grad_norm": 0.9836254119873047, + "learning_rate": 0.0009066281936351413, + "loss": 0.5211, + "step": 3440 + }, + { + "epoch": 0.19225074726932426, + "grad_norm": 0.6513175368309021, + "learning_rate": 0.0009066001792917974, + "loss": 0.4162, + "step": 3441 + }, + { + "epoch": 0.19230661787300612, + "grad_norm": 0.718776524066925, + "learning_rate": 0.0009065721649484537, + "loss": 0.5426, + "step": 3442 + }, + { + "epoch": 0.192362488476688, + "grad_norm": 0.470436692237854, + "learning_rate": 0.0009065441506051099, + "loss": 0.4266, + "step": 3443 + }, + { + "epoch": 0.19241835908036986, + "grad_norm": 0.7222057580947876, + "learning_rate": 0.000906516136261766, + "loss": 0.4711, + "step": 3444 + }, + { + "epoch": 0.19247422968405173, + "grad_norm": 0.6594362258911133, + "learning_rate": 0.0009064881219184222, + "loss": 0.5445, + "step": 3445 + }, + { + "epoch": 0.1925301002877336, + "grad_norm": 1.392532467842102, + "learning_rate": 0.0009064601075750785, + "loss": 0.5001, + "step": 3446 + }, + { + "epoch": 0.1925859708914155, + "grad_norm": 1.0443798303604126, + "learning_rate": 0.0009064320932317347, + "loss": 0.7385, + "step": 3447 + }, + { + "epoch": 0.19264184149509736, + "grad_norm": 1.6132087707519531, + "learning_rate": 0.0009064040788883908, + "loss": 0.5654, + "step": 3448 + }, + { + "epoch": 0.19269771209877923, + "grad_norm": 0.6596794128417969, + "learning_rate": 0.000906376064545047, + "loss": 0.4265, + "step": 3449 + }, + { + "epoch": 0.1927535827024611, + "grad_norm": 1.4326192140579224, + "learning_rate": 0.0009063480502017033, + "loss": 0.5256, + "step": 3450 + }, + { + "epoch": 0.19280945330614296, + "grad_norm": 0.7755450010299683, + "learning_rate": 0.0009063200358583595, + "loss": 0.469, + "step": 3451 + }, + { + "epoch": 0.19286532390982486, + "grad_norm": 0.6799972653388977, + "learning_rate": 0.0009062920215150156, + "loss": 0.4766, + "step": 3452 + }, + { + "epoch": 0.19292119451350673, + "grad_norm": 0.471942663192749, + "learning_rate": 0.0009062640071716719, + "loss": 0.4731, + "step": 3453 + }, + { + "epoch": 0.1929770651171886, + "grad_norm": 0.5123503804206848, + "learning_rate": 0.0009062359928283281, + "loss": 0.519, + "step": 3454 + }, + { + "epoch": 0.19303293572087046, + "grad_norm": 0.5173944234848022, + "learning_rate": 0.0009062079784849844, + "loss": 0.5259, + "step": 3455 + }, + { + "epoch": 0.19308880632455233, + "grad_norm": 0.6632512211799622, + "learning_rate": 0.0009061799641416405, + "loss": 0.4862, + "step": 3456 + }, + { + "epoch": 0.1931446769282342, + "grad_norm": 0.5951452851295471, + "learning_rate": 0.0009061519497982968, + "loss": 0.4444, + "step": 3457 + }, + { + "epoch": 0.1932005475319161, + "grad_norm": 0.6630435585975647, + "learning_rate": 0.000906123935454953, + "loss": 0.4487, + "step": 3458 + }, + { + "epoch": 0.19325641813559796, + "grad_norm": 0.5834782123565674, + "learning_rate": 0.0009060959211116092, + "loss": 0.5091, + "step": 3459 + }, + { + "epoch": 0.19331228873927983, + "grad_norm": 0.7597706317901611, + "learning_rate": 0.0009060679067682654, + "loss": 0.5055, + "step": 3460 + }, + { + "epoch": 0.1933681593429617, + "grad_norm": 0.6109134554862976, + "learning_rate": 0.0009060398924249216, + "loss": 0.4471, + "step": 3461 + }, + { + "epoch": 0.19342402994664357, + "grad_norm": 0.6219416260719299, + "learning_rate": 0.0009060118780815778, + "loss": 0.3874, + "step": 3462 + }, + { + "epoch": 0.19347990055032543, + "grad_norm": 0.6352624297142029, + "learning_rate": 0.000905983863738234, + "loss": 0.505, + "step": 3463 + }, + { + "epoch": 0.19353577115400733, + "grad_norm": 0.7415608763694763, + "learning_rate": 0.0009059558493948902, + "loss": 0.5289, + "step": 3464 + }, + { + "epoch": 0.1935916417576892, + "grad_norm": 6.9586663246154785, + "learning_rate": 0.0009059278350515464, + "loss": 0.6327, + "step": 3465 + }, + { + "epoch": 0.19364751236137107, + "grad_norm": 0.6010996699333191, + "learning_rate": 0.0009058998207082026, + "loss": 0.4621, + "step": 3466 + }, + { + "epoch": 0.19370338296505293, + "grad_norm": 0.9033005237579346, + "learning_rate": 0.0009058718063648588, + "loss": 0.5564, + "step": 3467 + }, + { + "epoch": 0.1937592535687348, + "grad_norm": 1.6992417573928833, + "learning_rate": 0.000905843792021515, + "loss": 0.4963, + "step": 3468 + }, + { + "epoch": 0.19381512417241667, + "grad_norm": 0.6380214095115662, + "learning_rate": 0.0009058157776781712, + "loss": 0.6466, + "step": 3469 + }, + { + "epoch": 0.19387099477609857, + "grad_norm": 1.8342453241348267, + "learning_rate": 0.0009057877633348274, + "loss": 0.5224, + "step": 3470 + }, + { + "epoch": 0.19392686537978043, + "grad_norm": 0.5540059804916382, + "learning_rate": 0.0009057597489914836, + "loss": 0.4686, + "step": 3471 + }, + { + "epoch": 0.1939827359834623, + "grad_norm": 3.044827461242676, + "learning_rate": 0.0009057317346481399, + "loss": 0.5709, + "step": 3472 + }, + { + "epoch": 0.19403860658714417, + "grad_norm": 0.8391507267951965, + "learning_rate": 0.0009057037203047961, + "loss": 0.4924, + "step": 3473 + }, + { + "epoch": 0.19409447719082604, + "grad_norm": 0.5878692269325256, + "learning_rate": 0.0009056757059614524, + "loss": 0.4412, + "step": 3474 + }, + { + "epoch": 0.19415034779450793, + "grad_norm": 0.8442159295082092, + "learning_rate": 0.0009056476916181085, + "loss": 0.5942, + "step": 3475 + }, + { + "epoch": 0.1942062183981898, + "grad_norm": 1.5450928211212158, + "learning_rate": 0.0009056196772747647, + "loss": 0.6311, + "step": 3476 + }, + { + "epoch": 0.19426208900187167, + "grad_norm": 1.4520169496536255, + "learning_rate": 0.0009055916629314209, + "loss": 0.6604, + "step": 3477 + }, + { + "epoch": 0.19431795960555354, + "grad_norm": 2.0298750400543213, + "learning_rate": 0.0009055636485880772, + "loss": 0.4979, + "step": 3478 + }, + { + "epoch": 0.1943738302092354, + "grad_norm": 0.9795212149620056, + "learning_rate": 0.0009055356342447333, + "loss": 0.5004, + "step": 3479 + }, + { + "epoch": 0.19442970081291727, + "grad_norm": 0.562337338924408, + "learning_rate": 0.0009055076199013895, + "loss": 0.4861, + "step": 3480 + }, + { + "epoch": 0.19448557141659917, + "grad_norm": 0.9057343006134033, + "learning_rate": 0.0009054796055580458, + "loss": 0.4483, + "step": 3481 + }, + { + "epoch": 0.19454144202028104, + "grad_norm": 0.640274703502655, + "learning_rate": 0.000905451591214702, + "loss": 0.4191, + "step": 3482 + }, + { + "epoch": 0.1945973126239629, + "grad_norm": 1.1509721279144287, + "learning_rate": 0.0009054235768713581, + "loss": 0.4299, + "step": 3483 + }, + { + "epoch": 0.19465318322764477, + "grad_norm": 1.0606777667999268, + "learning_rate": 0.0009053955625280143, + "loss": 0.556, + "step": 3484 + }, + { + "epoch": 0.19470905383132664, + "grad_norm": 0.5955864191055298, + "learning_rate": 0.0009053675481846706, + "loss": 0.7352, + "step": 3485 + }, + { + "epoch": 0.1947649244350085, + "grad_norm": 0.5603527426719666, + "learning_rate": 0.0009053395338413267, + "loss": 0.3824, + "step": 3486 + }, + { + "epoch": 0.1948207950386904, + "grad_norm": 1.0059882402420044, + "learning_rate": 0.0009053115194979829, + "loss": 0.5749, + "step": 3487 + }, + { + "epoch": 0.19487666564237227, + "grad_norm": 0.6974728107452393, + "learning_rate": 0.0009052835051546391, + "loss": 0.5076, + "step": 3488 + }, + { + "epoch": 0.19493253624605414, + "grad_norm": 0.4875369369983673, + "learning_rate": 0.0009052554908112955, + "loss": 0.4305, + "step": 3489 + }, + { + "epoch": 0.194988406849736, + "grad_norm": 0.95201176404953, + "learning_rate": 0.0009052274764679516, + "loss": 0.454, + "step": 3490 + }, + { + "epoch": 0.19504427745341787, + "grad_norm": 0.9427027106285095, + "learning_rate": 0.0009051994621246078, + "loss": 0.441, + "step": 3491 + }, + { + "epoch": 0.19510014805709977, + "grad_norm": 0.6674656271934509, + "learning_rate": 0.0009051714477812641, + "loss": 0.4623, + "step": 3492 + }, + { + "epoch": 0.19515601866078164, + "grad_norm": 0.5177181959152222, + "learning_rate": 0.0009051434334379203, + "loss": 0.4133, + "step": 3493 + }, + { + "epoch": 0.1952118892644635, + "grad_norm": 0.5465467572212219, + "learning_rate": 0.0009051154190945764, + "loss": 0.4043, + "step": 3494 + }, + { + "epoch": 0.19526775986814537, + "grad_norm": 0.7806662321090698, + "learning_rate": 0.0009050874047512326, + "loss": 0.4795, + "step": 3495 + }, + { + "epoch": 0.19532363047182724, + "grad_norm": 0.7427379488945007, + "learning_rate": 0.0009050593904078889, + "loss": 0.4636, + "step": 3496 + }, + { + "epoch": 0.1953795010755091, + "grad_norm": 0.5675991773605347, + "learning_rate": 0.0009050313760645451, + "loss": 0.4609, + "step": 3497 + }, + { + "epoch": 0.195435371679191, + "grad_norm": 0.5532568097114563, + "learning_rate": 0.0009050033617212012, + "loss": 0.4542, + "step": 3498 + }, + { + "epoch": 0.19549124228287287, + "grad_norm": 0.7623684406280518, + "learning_rate": 0.0009049753473778575, + "loss": 0.4298, + "step": 3499 + }, + { + "epoch": 0.19554711288655474, + "grad_norm": 0.6991674304008484, + "learning_rate": 0.0009049473330345137, + "loss": 0.5305, + "step": 3500 + }, + { + "epoch": 0.19554711288655474, + "eval_cer": 0.09714902949166966, + "eval_loss": 0.3674440383911133, + "eval_runtime": 56.1033, + "eval_samples_per_second": 80.887, + "eval_steps_per_second": 5.062, + "eval_wer": 0.3800731823133432, + "step": 3500 + }, + { + "epoch": 0.1956029834902366, + "grad_norm": 0.7770458459854126, + "learning_rate": 0.0009049193186911699, + "loss": 0.5821, + "step": 3501 + }, + { + "epoch": 0.19565885409391848, + "grad_norm": 0.6847448945045471, + "learning_rate": 0.000904891304347826, + "loss": 0.4415, + "step": 3502 + }, + { + "epoch": 0.19571472469760035, + "grad_norm": 0.8752630949020386, + "learning_rate": 0.0009048632900044823, + "loss": 0.5132, + "step": 3503 + }, + { + "epoch": 0.19577059530128224, + "grad_norm": 0.8012353777885437, + "learning_rate": 0.0009048352756611385, + "loss": 0.4702, + "step": 3504 + }, + { + "epoch": 0.1958264659049641, + "grad_norm": 1.5591967105865479, + "learning_rate": 0.0009048072613177948, + "loss": 0.452, + "step": 3505 + }, + { + "epoch": 0.19588233650864598, + "grad_norm": 0.9578115344047546, + "learning_rate": 0.000904779246974451, + "loss": 0.5266, + "step": 3506 + }, + { + "epoch": 0.19593820711232784, + "grad_norm": 0.765874445438385, + "learning_rate": 0.0009047512326311072, + "loss": 0.4961, + "step": 3507 + }, + { + "epoch": 0.1959940777160097, + "grad_norm": 1.2692182064056396, + "learning_rate": 0.0009047232182877634, + "loss": 0.4631, + "step": 3508 + }, + { + "epoch": 0.19604994831969158, + "grad_norm": 0.7260220646858215, + "learning_rate": 0.0009046952039444195, + "loss": 0.4022, + "step": 3509 + }, + { + "epoch": 0.19610581892337348, + "grad_norm": 1.4838045835494995, + "learning_rate": 0.0009046671896010758, + "loss": 0.5635, + "step": 3510 + }, + { + "epoch": 0.19616168952705534, + "grad_norm": 1.0155529975891113, + "learning_rate": 0.000904639175257732, + "loss": 0.4542, + "step": 3511 + }, + { + "epoch": 0.1962175601307372, + "grad_norm": 1.019705891609192, + "learning_rate": 0.0009046111609143882, + "loss": 0.4683, + "step": 3512 + }, + { + "epoch": 0.19627343073441908, + "grad_norm": 0.7242581844329834, + "learning_rate": 0.0009045831465710444, + "loss": 0.4642, + "step": 3513 + }, + { + "epoch": 0.19632930133810095, + "grad_norm": 1.3675050735473633, + "learning_rate": 0.0009045551322277006, + "loss": 0.4396, + "step": 3514 + }, + { + "epoch": 0.19638517194178284, + "grad_norm": 0.6249293088912964, + "learning_rate": 0.0009045271178843568, + "loss": 0.485, + "step": 3515 + }, + { + "epoch": 0.1964410425454647, + "grad_norm": 0.766784131526947, + "learning_rate": 0.000904499103541013, + "loss": 0.4401, + "step": 3516 + }, + { + "epoch": 0.19649691314914658, + "grad_norm": 0.6216867566108704, + "learning_rate": 0.0009044710891976692, + "loss": 0.4867, + "step": 3517 + }, + { + "epoch": 0.19655278375282845, + "grad_norm": 0.575286328792572, + "learning_rate": 0.0009044430748543254, + "loss": 0.4368, + "step": 3518 + }, + { + "epoch": 0.19660865435651032, + "grad_norm": 1.1054067611694336, + "learning_rate": 0.0009044150605109816, + "loss": 0.5665, + "step": 3519 + }, + { + "epoch": 0.19666452496019218, + "grad_norm": 0.742592990398407, + "learning_rate": 0.0009043870461676378, + "loss": 0.4888, + "step": 3520 + }, + { + "epoch": 0.19672039556387408, + "grad_norm": 0.5025954246520996, + "learning_rate": 0.000904359031824294, + "loss": 0.3831, + "step": 3521 + }, + { + "epoch": 0.19677626616755595, + "grad_norm": 0.7456044554710388, + "learning_rate": 0.0009043310174809503, + "loss": 0.5278, + "step": 3522 + }, + { + "epoch": 0.19683213677123781, + "grad_norm": 0.6059399247169495, + "learning_rate": 0.0009043030031376065, + "loss": 0.5511, + "step": 3523 + }, + { + "epoch": 0.19688800737491968, + "grad_norm": 0.9072201251983643, + "learning_rate": 0.0009042749887942628, + "loss": 0.4299, + "step": 3524 + }, + { + "epoch": 0.19694387797860155, + "grad_norm": 0.9296449422836304, + "learning_rate": 0.0009042469744509189, + "loss": 0.4913, + "step": 3525 + }, + { + "epoch": 0.19699974858228342, + "grad_norm": 0.5233350992202759, + "learning_rate": 0.0009042189601075751, + "loss": 0.4368, + "step": 3526 + }, + { + "epoch": 0.19705561918596531, + "grad_norm": 0.7396813631057739, + "learning_rate": 0.0009041909457642313, + "loss": 0.5327, + "step": 3527 + }, + { + "epoch": 0.19711148978964718, + "grad_norm": 0.4618791937828064, + "learning_rate": 0.0009041629314208875, + "loss": 0.4027, + "step": 3528 + }, + { + "epoch": 0.19716736039332905, + "grad_norm": 0.7634966969490051, + "learning_rate": 0.0009041349170775437, + "loss": 0.531, + "step": 3529 + }, + { + "epoch": 0.19722323099701092, + "grad_norm": 0.8934791684150696, + "learning_rate": 0.0009041069027341999, + "loss": 0.4678, + "step": 3530 + }, + { + "epoch": 0.19727910160069279, + "grad_norm": 0.530190646648407, + "learning_rate": 0.0009040788883908562, + "loss": 0.4541, + "step": 3531 + }, + { + "epoch": 0.19733497220437468, + "grad_norm": 0.6410190463066101, + "learning_rate": 0.0009040508740475123, + "loss": 0.5153, + "step": 3532 + }, + { + "epoch": 0.19739084280805655, + "grad_norm": 0.5680357217788696, + "learning_rate": 0.0009040228597041685, + "loss": 0.4805, + "step": 3533 + }, + { + "epoch": 0.19744671341173842, + "grad_norm": 0.7376148104667664, + "learning_rate": 0.0009039948453608247, + "loss": 0.4407, + "step": 3534 + }, + { + "epoch": 0.19750258401542028, + "grad_norm": 0.8737553358078003, + "learning_rate": 0.000903966831017481, + "loss": 0.5486, + "step": 3535 + }, + { + "epoch": 0.19755845461910215, + "grad_norm": 0.5502943396568298, + "learning_rate": 0.0009039388166741371, + "loss": 0.4418, + "step": 3536 + }, + { + "epoch": 0.19761432522278402, + "grad_norm": 1.3910448551177979, + "learning_rate": 0.0009039108023307933, + "loss": 0.4241, + "step": 3537 + }, + { + "epoch": 0.19767019582646592, + "grad_norm": 0.37981894612312317, + "learning_rate": 0.0009038827879874497, + "loss": 0.4492, + "step": 3538 + }, + { + "epoch": 0.19772606643014778, + "grad_norm": 1.0041446685791016, + "learning_rate": 0.0009038547736441059, + "loss": 0.4587, + "step": 3539 + }, + { + "epoch": 0.19778193703382965, + "grad_norm": 0.8270252346992493, + "learning_rate": 0.000903826759300762, + "loss": 0.3529, + "step": 3540 + }, + { + "epoch": 0.19783780763751152, + "grad_norm": 0.5568301677703857, + "learning_rate": 0.0009037987449574182, + "loss": 0.4732, + "step": 3541 + }, + { + "epoch": 0.1978936782411934, + "grad_norm": 0.6066697239875793, + "learning_rate": 0.0009037707306140745, + "loss": 0.4706, + "step": 3542 + }, + { + "epoch": 0.19794954884487526, + "grad_norm": 0.6082010269165039, + "learning_rate": 0.0009037427162707307, + "loss": 0.492, + "step": 3543 + }, + { + "epoch": 0.19800541944855715, + "grad_norm": 0.4333319067955017, + "learning_rate": 0.0009037147019273868, + "loss": 0.4449, + "step": 3544 + }, + { + "epoch": 0.19806129005223902, + "grad_norm": 0.6380464434623718, + "learning_rate": 0.0009036866875840431, + "loss": 0.3751, + "step": 3545 + }, + { + "epoch": 0.1981171606559209, + "grad_norm": 0.6893977522850037, + "learning_rate": 0.0009036586732406993, + "loss": 0.4559, + "step": 3546 + }, + { + "epoch": 0.19817303125960276, + "grad_norm": 0.5229524374008179, + "learning_rate": 0.0009036306588973555, + "loss": 0.4568, + "step": 3547 + }, + { + "epoch": 0.19822890186328462, + "grad_norm": 0.6136829257011414, + "learning_rate": 0.0009036026445540116, + "loss": 0.6797, + "step": 3548 + }, + { + "epoch": 0.19828477246696652, + "grad_norm": 1.0470588207244873, + "learning_rate": 0.0009035746302106679, + "loss": 0.6305, + "step": 3549 + }, + { + "epoch": 0.1983406430706484, + "grad_norm": 2.3831872940063477, + "learning_rate": 0.0009035466158673241, + "loss": 0.4493, + "step": 3550 + }, + { + "epoch": 0.19839651367433025, + "grad_norm": 1.1083189249038696, + "learning_rate": 0.0009035186015239802, + "loss": 0.4795, + "step": 3551 + }, + { + "epoch": 0.19845238427801212, + "grad_norm": 0.6626293063163757, + "learning_rate": 0.0009034905871806364, + "loss": 0.5363, + "step": 3552 + }, + { + "epoch": 0.198508254881694, + "grad_norm": 0.5469528436660767, + "learning_rate": 0.0009034625728372927, + "loss": 0.3905, + "step": 3553 + }, + { + "epoch": 0.19856412548537586, + "grad_norm": 0.8348404169082642, + "learning_rate": 0.000903434558493949, + "loss": 0.583, + "step": 3554 + }, + { + "epoch": 0.19861999608905775, + "grad_norm": 0.7322965264320374, + "learning_rate": 0.000903406544150605, + "loss": 0.4736, + "step": 3555 + }, + { + "epoch": 0.19867586669273962, + "grad_norm": 36.61244201660156, + "learning_rate": 0.0009033785298072614, + "loss": 0.4399, + "step": 3556 + }, + { + "epoch": 0.1987317372964215, + "grad_norm": 18.7752742767334, + "learning_rate": 0.0009033505154639176, + "loss": 0.4447, + "step": 3557 + }, + { + "epoch": 0.19878760790010336, + "grad_norm": 0.6046849489212036, + "learning_rate": 0.0009033225011205738, + "loss": 0.5013, + "step": 3558 + }, + { + "epoch": 0.19884347850378523, + "grad_norm": 0.7321492433547974, + "learning_rate": 0.0009032944867772299, + "loss": 0.4806, + "step": 3559 + }, + { + "epoch": 0.1988993491074671, + "grad_norm": 0.5534999370574951, + "learning_rate": 0.0009032664724338862, + "loss": 0.5382, + "step": 3560 + }, + { + "epoch": 0.198955219711149, + "grad_norm": 0.9596459865570068, + "learning_rate": 0.0009032384580905424, + "loss": 0.4839, + "step": 3561 + }, + { + "epoch": 0.19901109031483086, + "grad_norm": 0.6110329627990723, + "learning_rate": 0.0009032104437471986, + "loss": 0.5348, + "step": 3562 + }, + { + "epoch": 0.19906696091851273, + "grad_norm": 1.5610177516937256, + "learning_rate": 0.0009031824294038548, + "loss": 0.4043, + "step": 3563 + }, + { + "epoch": 0.1991228315221946, + "grad_norm": 0.7088295817375183, + "learning_rate": 0.000903154415060511, + "loss": 0.4939, + "step": 3564 + }, + { + "epoch": 0.19917870212587646, + "grad_norm": 1.5162221193313599, + "learning_rate": 0.0009031264007171672, + "loss": 0.5611, + "step": 3565 + }, + { + "epoch": 0.19923457272955833, + "grad_norm": 0.6697241067886353, + "learning_rate": 0.0009030983863738234, + "loss": 0.4918, + "step": 3566 + }, + { + "epoch": 0.19929044333324022, + "grad_norm": 0.888247549533844, + "learning_rate": 0.0009030703720304796, + "loss": 0.5006, + "step": 3567 + }, + { + "epoch": 0.1993463139369221, + "grad_norm": 0.6804181337356567, + "learning_rate": 0.0009030423576871358, + "loss": 0.391, + "step": 3568 + }, + { + "epoch": 0.19940218454060396, + "grad_norm": 0.7184508442878723, + "learning_rate": 0.000903014343343792, + "loss": 0.6003, + "step": 3569 + }, + { + "epoch": 0.19945805514428583, + "grad_norm": 0.6997994780540466, + "learning_rate": 0.0009029863290004483, + "loss": 0.4936, + "step": 3570 + }, + { + "epoch": 0.1995139257479677, + "grad_norm": 0.3804865777492523, + "learning_rate": 0.0009029583146571044, + "loss": 0.4367, + "step": 3571 + }, + { + "epoch": 0.1995697963516496, + "grad_norm": 0.4841291308403015, + "learning_rate": 0.0009029303003137607, + "loss": 0.4787, + "step": 3572 + }, + { + "epoch": 0.19962566695533146, + "grad_norm": 1.1336332559585571, + "learning_rate": 0.0009029022859704169, + "loss": 0.5083, + "step": 3573 + }, + { + "epoch": 0.19968153755901333, + "grad_norm": 0.7600587010383606, + "learning_rate": 0.0009028742716270731, + "loss": 0.4202, + "step": 3574 + }, + { + "epoch": 0.1997374081626952, + "grad_norm": 1.8934845924377441, + "learning_rate": 0.0009028462572837293, + "loss": 0.5065, + "step": 3575 + }, + { + "epoch": 0.19979327876637706, + "grad_norm": 0.9098082184791565, + "learning_rate": 0.0009028182429403855, + "loss": 0.441, + "step": 3576 + }, + { + "epoch": 0.19984914937005893, + "grad_norm": 0.6218282580375671, + "learning_rate": 0.0009027902285970418, + "loss": 0.4956, + "step": 3577 + }, + { + "epoch": 0.19990501997374083, + "grad_norm": 0.6869965195655823, + "learning_rate": 0.0009027622142536979, + "loss": 0.6166, + "step": 3578 + }, + { + "epoch": 0.1999608905774227, + "grad_norm": 0.6225520372390747, + "learning_rate": 0.0009027341999103541, + "loss": 0.5388, + "step": 3579 + }, + { + "epoch": 0.20001676118110456, + "grad_norm": 1.0198191404342651, + "learning_rate": 0.0009027061855670103, + "loss": 0.4621, + "step": 3580 + }, + { + "epoch": 0.20007263178478643, + "grad_norm": 0.6235898733139038, + "learning_rate": 0.0009026781712236666, + "loss": 0.4912, + "step": 3581 + }, + { + "epoch": 0.2001285023884683, + "grad_norm": 8.533157348632812, + "learning_rate": 0.0009026501568803227, + "loss": 0.4576, + "step": 3582 + }, + { + "epoch": 0.20018437299215017, + "grad_norm": 2.080754041671753, + "learning_rate": 0.0009026221425369789, + "loss": 0.4078, + "step": 3583 + }, + { + "epoch": 0.20024024359583206, + "grad_norm": 0.9668788313865662, + "learning_rate": 0.0009025941281936352, + "loss": 0.5101, + "step": 3584 + }, + { + "epoch": 0.20029611419951393, + "grad_norm": 5.297525405883789, + "learning_rate": 0.0009025661138502914, + "loss": 0.4945, + "step": 3585 + }, + { + "epoch": 0.2003519848031958, + "grad_norm": 1.1066324710845947, + "learning_rate": 0.0009025380995069475, + "loss": 0.4467, + "step": 3586 + }, + { + "epoch": 0.20040785540687767, + "grad_norm": 0.7696866989135742, + "learning_rate": 0.0009025100851636037, + "loss": 0.5354, + "step": 3587 + }, + { + "epoch": 0.20046372601055953, + "grad_norm": 0.9501246213912964, + "learning_rate": 0.00090248207082026, + "loss": 0.463, + "step": 3588 + }, + { + "epoch": 0.20051959661424143, + "grad_norm": 1.820833683013916, + "learning_rate": 0.0009024540564769163, + "loss": 0.5476, + "step": 3589 + }, + { + "epoch": 0.2005754672179233, + "grad_norm": 1.8860077857971191, + "learning_rate": 0.0009024260421335724, + "loss": 0.6032, + "step": 3590 + }, + { + "epoch": 0.20063133782160517, + "grad_norm": 1.4408416748046875, + "learning_rate": 0.0009023980277902286, + "loss": 0.5113, + "step": 3591 + }, + { + "epoch": 0.20068720842528703, + "grad_norm": 0.7070097327232361, + "learning_rate": 0.0009023700134468849, + "loss": 0.5089, + "step": 3592 + }, + { + "epoch": 0.2007430790289689, + "grad_norm": 0.4896397590637207, + "learning_rate": 0.000902341999103541, + "loss": 0.4443, + "step": 3593 + }, + { + "epoch": 0.20079894963265077, + "grad_norm": 1.143958330154419, + "learning_rate": 0.0009023139847601972, + "loss": 0.5602, + "step": 3594 + }, + { + "epoch": 0.20085482023633267, + "grad_norm": 3.0072648525238037, + "learning_rate": 0.0009022859704168535, + "loss": 0.4412, + "step": 3595 + }, + { + "epoch": 0.20091069084001453, + "grad_norm": 0.9834116101264954, + "learning_rate": 0.0009022579560735097, + "loss": 0.4507, + "step": 3596 + }, + { + "epoch": 0.2009665614436964, + "grad_norm": 0.9961720705032349, + "learning_rate": 0.0009022299417301658, + "loss": 0.5818, + "step": 3597 + }, + { + "epoch": 0.20102243204737827, + "grad_norm": 4.496664524078369, + "learning_rate": 0.000902201927386822, + "loss": 0.5676, + "step": 3598 + }, + { + "epoch": 0.20107830265106014, + "grad_norm": 0.793182373046875, + "learning_rate": 0.0009021739130434783, + "loss": 0.5735, + "step": 3599 + }, + { + "epoch": 0.201134173254742, + "grad_norm": 0.5504240393638611, + "learning_rate": 0.0009021458987001345, + "loss": 0.5034, + "step": 3600 + }, + { + "epoch": 0.2011900438584239, + "grad_norm": 0.6414144039154053, + "learning_rate": 0.0009021178843567906, + "loss": 0.5099, + "step": 3601 + }, + { + "epoch": 0.20124591446210577, + "grad_norm": 7.485801696777344, + "learning_rate": 0.0009020898700134469, + "loss": 0.5306, + "step": 3602 + }, + { + "epoch": 0.20130178506578764, + "grad_norm": 1.0554862022399902, + "learning_rate": 0.0009020618556701031, + "loss": 0.4583, + "step": 3603 + }, + { + "epoch": 0.2013576556694695, + "grad_norm": 0.5271552801132202, + "learning_rate": 0.0009020338413267593, + "loss": 0.4272, + "step": 3604 + }, + { + "epoch": 0.20141352627315137, + "grad_norm": 2.8193604946136475, + "learning_rate": 0.0009020058269834154, + "loss": 0.5259, + "step": 3605 + }, + { + "epoch": 0.20146939687683327, + "grad_norm": 0.4069269001483917, + "learning_rate": 0.0009019778126400718, + "loss": 0.4231, + "step": 3606 + }, + { + "epoch": 0.20152526748051514, + "grad_norm": 0.7664451599121094, + "learning_rate": 0.000901949798296728, + "loss": 0.47, + "step": 3607 + }, + { + "epoch": 0.201581138084197, + "grad_norm": 0.9415386915206909, + "learning_rate": 0.0009019217839533842, + "loss": 0.4149, + "step": 3608 + }, + { + "epoch": 0.20163700868787887, + "grad_norm": 1.8283052444458008, + "learning_rate": 0.0009018937696100404, + "loss": 0.4482, + "step": 3609 + }, + { + "epoch": 0.20169287929156074, + "grad_norm": 0.9990724325180054, + "learning_rate": 0.0009018657552666966, + "loss": 0.6836, + "step": 3610 + }, + { + "epoch": 0.2017487498952426, + "grad_norm": 0.537653923034668, + "learning_rate": 0.0009018377409233528, + "loss": 0.3973, + "step": 3611 + }, + { + "epoch": 0.2018046204989245, + "grad_norm": 0.5545988082885742, + "learning_rate": 0.000901809726580009, + "loss": 0.4835, + "step": 3612 + }, + { + "epoch": 0.20186049110260637, + "grad_norm": 11.169219970703125, + "learning_rate": 0.0009017817122366652, + "loss": 0.4879, + "step": 3613 + }, + { + "epoch": 0.20191636170628824, + "grad_norm": 0.9393489956855774, + "learning_rate": 0.0009017536978933214, + "loss": 0.5627, + "step": 3614 + }, + { + "epoch": 0.2019722323099701, + "grad_norm": 0.6782003045082092, + "learning_rate": 0.0009017256835499776, + "loss": 0.5577, + "step": 3615 + }, + { + "epoch": 0.20202810291365197, + "grad_norm": 0.8406596779823303, + "learning_rate": 0.0009016976692066338, + "loss": 0.5358, + "step": 3616 + }, + { + "epoch": 0.20208397351733384, + "grad_norm": 0.6405194997787476, + "learning_rate": 0.00090166965486329, + "loss": 0.4735, + "step": 3617 + }, + { + "epoch": 0.20213984412101574, + "grad_norm": 0.8123499155044556, + "learning_rate": 0.0009016416405199462, + "loss": 0.454, + "step": 3618 + }, + { + "epoch": 0.2021957147246976, + "grad_norm": 0.5154933929443359, + "learning_rate": 0.0009016136261766024, + "loss": 0.4323, + "step": 3619 + }, + { + "epoch": 0.20225158532837947, + "grad_norm": 0.594066858291626, + "learning_rate": 0.0009015856118332586, + "loss": 0.5118, + "step": 3620 + }, + { + "epoch": 0.20230745593206134, + "grad_norm": 0.914502739906311, + "learning_rate": 0.0009015575974899148, + "loss": 0.6628, + "step": 3621 + }, + { + "epoch": 0.2023633265357432, + "grad_norm": 0.7711836695671082, + "learning_rate": 0.000901529583146571, + "loss": 0.477, + "step": 3622 + }, + { + "epoch": 0.20241919713942508, + "grad_norm": 0.7035220861434937, + "learning_rate": 0.0009015015688032274, + "loss": 0.5071, + "step": 3623 + }, + { + "epoch": 0.20247506774310697, + "grad_norm": 0.717041552066803, + "learning_rate": 0.0009014735544598835, + "loss": 0.4751, + "step": 3624 + }, + { + "epoch": 0.20253093834678884, + "grad_norm": 0.49735721945762634, + "learning_rate": 0.0009014455401165397, + "loss": 0.4733, + "step": 3625 + }, + { + "epoch": 0.2025868089504707, + "grad_norm": 0.4256522059440613, + "learning_rate": 0.0009014175257731959, + "loss": 0.3879, + "step": 3626 + }, + { + "epoch": 0.20264267955415258, + "grad_norm": 4.370864391326904, + "learning_rate": 0.0009013895114298522, + "loss": 0.4711, + "step": 3627 + }, + { + "epoch": 0.20269855015783445, + "grad_norm": 0.6374932527542114, + "learning_rate": 0.0009013614970865083, + "loss": 0.5544, + "step": 3628 + }, + { + "epoch": 0.20275442076151634, + "grad_norm": 2.538144111633301, + "learning_rate": 0.0009013334827431645, + "loss": 0.5947, + "step": 3629 + }, + { + "epoch": 0.2028102913651982, + "grad_norm": 0.46030375361442566, + "learning_rate": 0.0009013054683998207, + "loss": 0.4265, + "step": 3630 + }, + { + "epoch": 0.20286616196888008, + "grad_norm": 0.8169751167297363, + "learning_rate": 0.000901277454056477, + "loss": 0.5624, + "step": 3631 + }, + { + "epoch": 0.20292203257256194, + "grad_norm": 0.5733919739723206, + "learning_rate": 0.0009012494397131331, + "loss": 0.4562, + "step": 3632 + }, + { + "epoch": 0.2029779031762438, + "grad_norm": 0.5828341245651245, + "learning_rate": 0.0009012214253697893, + "loss": 0.4687, + "step": 3633 + }, + { + "epoch": 0.20303377377992568, + "grad_norm": 0.7024948000907898, + "learning_rate": 0.0009011934110264456, + "loss": 0.5382, + "step": 3634 + }, + { + "epoch": 0.20308964438360758, + "grad_norm": 0.7504926323890686, + "learning_rate": 0.0009011653966831017, + "loss": 0.3492, + "step": 3635 + }, + { + "epoch": 0.20314551498728944, + "grad_norm": 0.6261107325553894, + "learning_rate": 0.0009011373823397579, + "loss": 0.4538, + "step": 3636 + }, + { + "epoch": 0.2032013855909713, + "grad_norm": 0.710516631603241, + "learning_rate": 0.0009011093679964141, + "loss": 0.5075, + "step": 3637 + }, + { + "epoch": 0.20325725619465318, + "grad_norm": 0.6439149379730225, + "learning_rate": 0.0009010813536530705, + "loss": 0.3835, + "step": 3638 + }, + { + "epoch": 0.20331312679833505, + "grad_norm": 0.649797797203064, + "learning_rate": 0.0009010533393097265, + "loss": 0.4852, + "step": 3639 + }, + { + "epoch": 0.20336899740201692, + "grad_norm": 0.6192416548728943, + "learning_rate": 0.0009010253249663828, + "loss": 0.4685, + "step": 3640 + }, + { + "epoch": 0.2034248680056988, + "grad_norm": 1.248913288116455, + "learning_rate": 0.0009009973106230391, + "loss": 0.5126, + "step": 3641 + }, + { + "epoch": 0.20348073860938068, + "grad_norm": 0.8286505341529846, + "learning_rate": 0.0009009692962796953, + "loss": 0.6425, + "step": 3642 + }, + { + "epoch": 0.20353660921306255, + "grad_norm": 0.7176082730293274, + "learning_rate": 0.0009009412819363514, + "loss": 0.5205, + "step": 3643 + }, + { + "epoch": 0.20359247981674441, + "grad_norm": 0.7132717967033386, + "learning_rate": 0.0009009132675930076, + "loss": 0.4837, + "step": 3644 + }, + { + "epoch": 0.20364835042042628, + "grad_norm": 1.5763846635818481, + "learning_rate": 0.0009008852532496639, + "loss": 0.5638, + "step": 3645 + }, + { + "epoch": 0.20370422102410818, + "grad_norm": 1.0540639162063599, + "learning_rate": 0.0009008572389063201, + "loss": 0.3997, + "step": 3646 + }, + { + "epoch": 0.20376009162779005, + "grad_norm": 0.8735400438308716, + "learning_rate": 0.0009008292245629762, + "loss": 0.544, + "step": 3647 + }, + { + "epoch": 0.20381596223147191, + "grad_norm": 0.8207980394363403, + "learning_rate": 0.0009008012102196325, + "loss": 0.3498, + "step": 3648 + }, + { + "epoch": 0.20387183283515378, + "grad_norm": 0.5362919569015503, + "learning_rate": 0.0009007731958762887, + "loss": 0.4296, + "step": 3649 + }, + { + "epoch": 0.20392770343883565, + "grad_norm": 0.7567645907402039, + "learning_rate": 0.0009007451815329449, + "loss": 0.5397, + "step": 3650 + }, + { + "epoch": 0.20398357404251752, + "grad_norm": 0.49839794635772705, + "learning_rate": 0.000900717167189601, + "loss": 0.5056, + "step": 3651 + }, + { + "epoch": 0.2040394446461994, + "grad_norm": 0.6365333795547485, + "learning_rate": 0.0009006891528462573, + "loss": 0.4668, + "step": 3652 + }, + { + "epoch": 0.20409531524988128, + "grad_norm": 0.5628067255020142, + "learning_rate": 0.0009006611385029135, + "loss": 0.367, + "step": 3653 + }, + { + "epoch": 0.20415118585356315, + "grad_norm": 1.182703971862793, + "learning_rate": 0.0009006331241595697, + "loss": 0.5643, + "step": 3654 + }, + { + "epoch": 0.20420705645724502, + "grad_norm": 0.86634761095047, + "learning_rate": 0.000900605109816226, + "loss": 0.5918, + "step": 3655 + }, + { + "epoch": 0.20426292706092689, + "grad_norm": 0.863488495349884, + "learning_rate": 0.0009005770954728822, + "loss": 0.5991, + "step": 3656 + }, + { + "epoch": 0.20431879766460875, + "grad_norm": 1.162996530532837, + "learning_rate": 0.0009005490811295384, + "loss": 0.489, + "step": 3657 + }, + { + "epoch": 0.20437466826829065, + "grad_norm": 0.6431537866592407, + "learning_rate": 0.0009005210667861945, + "loss": 0.4612, + "step": 3658 + }, + { + "epoch": 0.20443053887197252, + "grad_norm": 0.6305604577064514, + "learning_rate": 0.0009004930524428508, + "loss": 0.4077, + "step": 3659 + }, + { + "epoch": 0.20448640947565438, + "grad_norm": 0.5274586081504822, + "learning_rate": 0.000900465038099507, + "loss": 0.6065, + "step": 3660 + }, + { + "epoch": 0.20454228007933625, + "grad_norm": 0.8412783741950989, + "learning_rate": 0.0009004370237561632, + "loss": 0.4537, + "step": 3661 + }, + { + "epoch": 0.20459815068301812, + "grad_norm": 4.662811756134033, + "learning_rate": 0.0009004090094128193, + "loss": 0.4886, + "step": 3662 + }, + { + "epoch": 0.20465402128670002, + "grad_norm": 1.4199450016021729, + "learning_rate": 0.0009003809950694756, + "loss": 0.5879, + "step": 3663 + }, + { + "epoch": 0.20470989189038188, + "grad_norm": 1.1043227910995483, + "learning_rate": 0.0009003529807261318, + "loss": 0.486, + "step": 3664 + }, + { + "epoch": 0.20476576249406375, + "grad_norm": 2.5059285163879395, + "learning_rate": 0.000900324966382788, + "loss": 0.5187, + "step": 3665 + }, + { + "epoch": 0.20482163309774562, + "grad_norm": 0.5092585682868958, + "learning_rate": 0.0009002969520394442, + "loss": 0.4777, + "step": 3666 + }, + { + "epoch": 0.2048775037014275, + "grad_norm": 2.851189136505127, + "learning_rate": 0.0009002689376961004, + "loss": 0.386, + "step": 3667 + }, + { + "epoch": 0.20493337430510936, + "grad_norm": 0.8775118589401245, + "learning_rate": 0.0009002409233527566, + "loss": 0.5823, + "step": 3668 + }, + { + "epoch": 0.20498924490879125, + "grad_norm": 0.8073964715003967, + "learning_rate": 0.0009002129090094128, + "loss": 0.541, + "step": 3669 + }, + { + "epoch": 0.20504511551247312, + "grad_norm": 1.7558910846710205, + "learning_rate": 0.000900184894666069, + "loss": 0.6962, + "step": 3670 + }, + { + "epoch": 0.205100986116155, + "grad_norm": 0.6841502785682678, + "learning_rate": 0.0009001568803227252, + "loss": 0.4617, + "step": 3671 + }, + { + "epoch": 0.20515685671983686, + "grad_norm": 1.2318612337112427, + "learning_rate": 0.0009001288659793815, + "loss": 0.6659, + "step": 3672 + }, + { + "epoch": 0.20521272732351872, + "grad_norm": 0.5853779911994934, + "learning_rate": 0.0009001008516360378, + "loss": 0.4861, + "step": 3673 + }, + { + "epoch": 0.2052685979272006, + "grad_norm": 0.4444562792778015, + "learning_rate": 0.0009000728372926939, + "loss": 0.5203, + "step": 3674 + }, + { + "epoch": 0.2053244685308825, + "grad_norm": 0.9464027881622314, + "learning_rate": 0.0009000448229493501, + "loss": 0.4763, + "step": 3675 + }, + { + "epoch": 0.20538033913456435, + "grad_norm": 0.6396483182907104, + "learning_rate": 0.0009000168086060063, + "loss": 0.6049, + "step": 3676 + }, + { + "epoch": 0.20543620973824622, + "grad_norm": 0.5218935012817383, + "learning_rate": 0.0008999887942626625, + "loss": 0.3899, + "step": 3677 + }, + { + "epoch": 0.2054920803419281, + "grad_norm": 0.40790942311286926, + "learning_rate": 0.0008999607799193187, + "loss": 0.3476, + "step": 3678 + }, + { + "epoch": 0.20554795094560996, + "grad_norm": 0.5944613814353943, + "learning_rate": 0.0008999327655759749, + "loss": 0.5042, + "step": 3679 + }, + { + "epoch": 0.20560382154929183, + "grad_norm": 3.880249261856079, + "learning_rate": 0.0008999047512326312, + "loss": 0.448, + "step": 3680 + }, + { + "epoch": 0.20565969215297372, + "grad_norm": 0.7706067562103271, + "learning_rate": 0.0008998767368892873, + "loss": 0.4191, + "step": 3681 + }, + { + "epoch": 0.2057155627566556, + "grad_norm": 4.115269660949707, + "learning_rate": 0.0008998487225459435, + "loss": 0.4392, + "step": 3682 + }, + { + "epoch": 0.20577143336033746, + "grad_norm": 0.5015891194343567, + "learning_rate": 0.0008998207082025997, + "loss": 0.4455, + "step": 3683 + }, + { + "epoch": 0.20582730396401933, + "grad_norm": 0.8186747431755066, + "learning_rate": 0.000899792693859256, + "loss": 0.5554, + "step": 3684 + }, + { + "epoch": 0.2058831745677012, + "grad_norm": 0.5838703513145447, + "learning_rate": 0.0008997646795159121, + "loss": 0.4629, + "step": 3685 + }, + { + "epoch": 0.2059390451713831, + "grad_norm": 0.976266086101532, + "learning_rate": 0.0008997366651725683, + "loss": 0.4872, + "step": 3686 + }, + { + "epoch": 0.20599491577506496, + "grad_norm": 0.908288836479187, + "learning_rate": 0.0008997086508292246, + "loss": 0.4223, + "step": 3687 + }, + { + "epoch": 0.20605078637874683, + "grad_norm": 0.6786814332008362, + "learning_rate": 0.0008996806364858809, + "loss": 0.5098, + "step": 3688 + }, + { + "epoch": 0.2061066569824287, + "grad_norm": 0.450008362531662, + "learning_rate": 0.000899652622142537, + "loss": 0.4006, + "step": 3689 + }, + { + "epoch": 0.20616252758611056, + "grad_norm": 0.7517444491386414, + "learning_rate": 0.0008996246077991932, + "loss": 0.5248, + "step": 3690 + }, + { + "epoch": 0.20621839818979243, + "grad_norm": 0.5932966470718384, + "learning_rate": 0.0008995965934558495, + "loss": 0.4946, + "step": 3691 + }, + { + "epoch": 0.20627426879347432, + "grad_norm": 0.5805421471595764, + "learning_rate": 0.0008995685791125057, + "loss": 0.5077, + "step": 3692 + }, + { + "epoch": 0.2063301393971562, + "grad_norm": 0.8588129878044128, + "learning_rate": 0.0008995405647691618, + "loss": 0.423, + "step": 3693 + }, + { + "epoch": 0.20638601000083806, + "grad_norm": 1.3981451988220215, + "learning_rate": 0.0008995125504258181, + "loss": 0.6515, + "step": 3694 + }, + { + "epoch": 0.20644188060451993, + "grad_norm": 1.0960360765457153, + "learning_rate": 0.0008994845360824743, + "loss": 0.5519, + "step": 3695 + }, + { + "epoch": 0.2064977512082018, + "grad_norm": 0.9362337589263916, + "learning_rate": 0.0008994565217391305, + "loss": 0.7277, + "step": 3696 + }, + { + "epoch": 0.20655362181188366, + "grad_norm": 0.7404981851577759, + "learning_rate": 0.0008994285073957866, + "loss": 0.4373, + "step": 3697 + }, + { + "epoch": 0.20660949241556556, + "grad_norm": 0.5967028141021729, + "learning_rate": 0.0008994004930524429, + "loss": 0.4208, + "step": 3698 + }, + { + "epoch": 0.20666536301924743, + "grad_norm": 0.6012880802154541, + "learning_rate": 0.0008993724787090991, + "loss": 0.4249, + "step": 3699 + }, + { + "epoch": 0.2067212336229293, + "grad_norm": 0.9663028120994568, + "learning_rate": 0.0008993444643657552, + "loss": 0.487, + "step": 3700 + }, + { + "epoch": 0.20677710422661116, + "grad_norm": 0.9490858912467957, + "learning_rate": 0.0008993164500224114, + "loss": 0.5581, + "step": 3701 + }, + { + "epoch": 0.20683297483029303, + "grad_norm": 0.4999719560146332, + "learning_rate": 0.0008992884356790677, + "loss": 0.4088, + "step": 3702 + }, + { + "epoch": 0.20688884543397493, + "grad_norm": 2.886892318725586, + "learning_rate": 0.0008992604213357239, + "loss": 0.486, + "step": 3703 + }, + { + "epoch": 0.2069447160376568, + "grad_norm": 0.8166035413742065, + "learning_rate": 0.00089923240699238, + "loss": 0.4691, + "step": 3704 + }, + { + "epoch": 0.20700058664133866, + "grad_norm": 1.4518768787384033, + "learning_rate": 0.0008992043926490364, + "loss": 0.5263, + "step": 3705 + }, + { + "epoch": 0.20705645724502053, + "grad_norm": 0.5526158213615417, + "learning_rate": 0.0008991763783056926, + "loss": 0.4574, + "step": 3706 + }, + { + "epoch": 0.2071123278487024, + "grad_norm": 0.7177834510803223, + "learning_rate": 0.0008991483639623488, + "loss": 0.5411, + "step": 3707 + }, + { + "epoch": 0.20716819845238427, + "grad_norm": 0.5037501454353333, + "learning_rate": 0.0008991203496190049, + "loss": 0.5284, + "step": 3708 + }, + { + "epoch": 0.20722406905606616, + "grad_norm": 0.7060887813568115, + "learning_rate": 0.0008990923352756612, + "loss": 0.5343, + "step": 3709 + }, + { + "epoch": 0.20727993965974803, + "grad_norm": 0.8158430457115173, + "learning_rate": 0.0008990643209323174, + "loss": 0.517, + "step": 3710 + }, + { + "epoch": 0.2073358102634299, + "grad_norm": 0.6996387243270874, + "learning_rate": 0.0008990363065889736, + "loss": 0.6378, + "step": 3711 + }, + { + "epoch": 0.20739168086711177, + "grad_norm": 5.725865840911865, + "learning_rate": 0.0008990082922456298, + "loss": 0.4535, + "step": 3712 + }, + { + "epoch": 0.20744755147079363, + "grad_norm": 0.9549941420555115, + "learning_rate": 0.000898980277902286, + "loss": 0.6314, + "step": 3713 + }, + { + "epoch": 0.2075034220744755, + "grad_norm": 0.9611788392066956, + "learning_rate": 0.0008989522635589422, + "loss": 0.4213, + "step": 3714 + }, + { + "epoch": 0.2075592926781574, + "grad_norm": 0.4928436279296875, + "learning_rate": 0.0008989242492155984, + "loss": 0.4581, + "step": 3715 + }, + { + "epoch": 0.20761516328183927, + "grad_norm": 1.0227296352386475, + "learning_rate": 0.0008988962348722546, + "loss": 0.4836, + "step": 3716 + }, + { + "epoch": 0.20767103388552113, + "grad_norm": 0.6972116231918335, + "learning_rate": 0.0008988682205289108, + "loss": 0.4358, + "step": 3717 + }, + { + "epoch": 0.207726904489203, + "grad_norm": 3.121626853942871, + "learning_rate": 0.000898840206185567, + "loss": 0.4553, + "step": 3718 + }, + { + "epoch": 0.20778277509288487, + "grad_norm": 0.7048122882843018, + "learning_rate": 0.0008988121918422233, + "loss": 0.4793, + "step": 3719 + }, + { + "epoch": 0.20783864569656677, + "grad_norm": 0.5191512107849121, + "learning_rate": 0.0008987841774988794, + "loss": 0.5275, + "step": 3720 + }, + { + "epoch": 0.20789451630024863, + "grad_norm": 0.6167207360267639, + "learning_rate": 0.0008987561631555356, + "loss": 0.4146, + "step": 3721 + }, + { + "epoch": 0.2079503869039305, + "grad_norm": 0.4742671251296997, + "learning_rate": 0.0008987281488121918, + "loss": 0.4528, + "step": 3722 + }, + { + "epoch": 0.20800625750761237, + "grad_norm": 2.072347402572632, + "learning_rate": 0.0008987001344688481, + "loss": 0.5652, + "step": 3723 + }, + { + "epoch": 0.20806212811129424, + "grad_norm": 0.8131927847862244, + "learning_rate": 0.0008986721201255043, + "loss": 0.5557, + "step": 3724 + }, + { + "epoch": 0.2081179987149761, + "grad_norm": 1.2258285284042358, + "learning_rate": 0.0008986441057821605, + "loss": 0.5446, + "step": 3725 + }, + { + "epoch": 0.208173869318658, + "grad_norm": 3.3557844161987305, + "learning_rate": 0.0008986160914388168, + "loss": 0.4605, + "step": 3726 + }, + { + "epoch": 0.20822973992233987, + "grad_norm": 0.5310865640640259, + "learning_rate": 0.0008985880770954729, + "loss": 0.439, + "step": 3727 + }, + { + "epoch": 0.20828561052602174, + "grad_norm": 0.5129857659339905, + "learning_rate": 0.0008985600627521291, + "loss": 0.3944, + "step": 3728 + }, + { + "epoch": 0.2083414811297036, + "grad_norm": 0.48191145062446594, + "learning_rate": 0.0008985320484087853, + "loss": 0.4908, + "step": 3729 + }, + { + "epoch": 0.20839735173338547, + "grad_norm": 0.7548501491546631, + "learning_rate": 0.0008985040340654416, + "loss": 0.4187, + "step": 3730 + }, + { + "epoch": 0.20845322233706734, + "grad_norm": 0.6063972115516663, + "learning_rate": 0.0008984760197220977, + "loss": 0.4217, + "step": 3731 + }, + { + "epoch": 0.20850909294074924, + "grad_norm": 0.431465744972229, + "learning_rate": 0.0008984480053787539, + "loss": 0.416, + "step": 3732 + }, + { + "epoch": 0.2085649635444311, + "grad_norm": 4.261516094207764, + "learning_rate": 0.0008984199910354101, + "loss": 0.4552, + "step": 3733 + }, + { + "epoch": 0.20862083414811297, + "grad_norm": 1.4569470882415771, + "learning_rate": 0.0008983919766920664, + "loss": 0.3573, + "step": 3734 + }, + { + "epoch": 0.20867670475179484, + "grad_norm": 0.9126104116439819, + "learning_rate": 0.0008983639623487225, + "loss": 0.4192, + "step": 3735 + }, + { + "epoch": 0.2087325753554767, + "grad_norm": 2.740112543106079, + "learning_rate": 0.0008983359480053787, + "loss": 0.4821, + "step": 3736 + }, + { + "epoch": 0.20878844595915858, + "grad_norm": 0.9895162582397461, + "learning_rate": 0.000898307933662035, + "loss": 0.5361, + "step": 3737 + }, + { + "epoch": 0.20884431656284047, + "grad_norm": 0.8092295527458191, + "learning_rate": 0.0008982799193186913, + "loss": 0.4776, + "step": 3738 + }, + { + "epoch": 0.20890018716652234, + "grad_norm": 0.7118065357208252, + "learning_rate": 0.0008982519049753473, + "loss": 0.5678, + "step": 3739 + }, + { + "epoch": 0.2089560577702042, + "grad_norm": 0.5253150463104248, + "learning_rate": 0.0008982238906320036, + "loss": 0.4655, + "step": 3740 + }, + { + "epoch": 0.20901192837388607, + "grad_norm": 0.7770140767097473, + "learning_rate": 0.0008981958762886599, + "loss": 0.4416, + "step": 3741 + }, + { + "epoch": 0.20906779897756794, + "grad_norm": 0.5377078652381897, + "learning_rate": 0.000898167861945316, + "loss": 0.3383, + "step": 3742 + }, + { + "epoch": 0.20912366958124984, + "grad_norm": 0.9398568868637085, + "learning_rate": 0.0008981398476019722, + "loss": 0.4057, + "step": 3743 + }, + { + "epoch": 0.2091795401849317, + "grad_norm": 0.5038449764251709, + "learning_rate": 0.0008981118332586285, + "loss": 0.4516, + "step": 3744 + }, + { + "epoch": 0.20923541078861357, + "grad_norm": 0.6047619581222534, + "learning_rate": 0.0008980838189152847, + "loss": 0.5108, + "step": 3745 + }, + { + "epoch": 0.20929128139229544, + "grad_norm": 0.5708025097846985, + "learning_rate": 0.0008980558045719408, + "loss": 0.5545, + "step": 3746 + }, + { + "epoch": 0.2093471519959773, + "grad_norm": 0.8094294667243958, + "learning_rate": 0.000898027790228597, + "loss": 0.4443, + "step": 3747 + }, + { + "epoch": 0.20940302259965918, + "grad_norm": 1.3261032104492188, + "learning_rate": 0.0008979997758852533, + "loss": 0.4884, + "step": 3748 + }, + { + "epoch": 0.20945889320334107, + "grad_norm": 1.9844321012496948, + "learning_rate": 0.0008979717615419095, + "loss": 0.5318, + "step": 3749 + }, + { + "epoch": 0.20951476380702294, + "grad_norm": 0.4113161265850067, + "learning_rate": 0.0008979437471985656, + "loss": 0.386, + "step": 3750 + }, + { + "epoch": 0.2095706344107048, + "grad_norm": 0.8617240190505981, + "learning_rate": 0.0008979157328552219, + "loss": 0.4872, + "step": 3751 + }, + { + "epoch": 0.20962650501438668, + "grad_norm": 0.5311805605888367, + "learning_rate": 0.0008978877185118781, + "loss": 0.4026, + "step": 3752 + }, + { + "epoch": 0.20968237561806854, + "grad_norm": 0.44844868779182434, + "learning_rate": 0.0008978597041685343, + "loss": 0.469, + "step": 3753 + }, + { + "epoch": 0.2097382462217504, + "grad_norm": 0.6522153615951538, + "learning_rate": 0.0008978316898251904, + "loss": 0.5169, + "step": 3754 + }, + { + "epoch": 0.2097941168254323, + "grad_norm": 0.57787024974823, + "learning_rate": 0.0008978036754818467, + "loss": 0.5283, + "step": 3755 + }, + { + "epoch": 0.20984998742911418, + "grad_norm": 1.4191255569458008, + "learning_rate": 0.000897775661138503, + "loss": 0.5888, + "step": 3756 + }, + { + "epoch": 0.20990585803279604, + "grad_norm": 0.4914593994617462, + "learning_rate": 0.0008977476467951592, + "loss": 0.396, + "step": 3757 + }, + { + "epoch": 0.2099617286364779, + "grad_norm": 1.3051931858062744, + "learning_rate": 0.0008977196324518154, + "loss": 0.5933, + "step": 3758 + }, + { + "epoch": 0.21001759924015978, + "grad_norm": 0.7936989068984985, + "learning_rate": 0.0008976916181084716, + "loss": 0.4331, + "step": 3759 + }, + { + "epoch": 0.21007346984384168, + "grad_norm": 0.8687162399291992, + "learning_rate": 0.0008976636037651278, + "loss": 0.4872, + "step": 3760 + }, + { + "epoch": 0.21012934044752354, + "grad_norm": 0.7042644023895264, + "learning_rate": 0.000897635589421784, + "loss": 0.4636, + "step": 3761 + }, + { + "epoch": 0.2101852110512054, + "grad_norm": 0.6725339293479919, + "learning_rate": 0.0008976075750784402, + "loss": 0.4712, + "step": 3762 + }, + { + "epoch": 0.21024108165488728, + "grad_norm": 0.3838529586791992, + "learning_rate": 0.0008975795607350964, + "loss": 0.3775, + "step": 3763 + }, + { + "epoch": 0.21029695225856915, + "grad_norm": 5.241764545440674, + "learning_rate": 0.0008975515463917526, + "loss": 0.4869, + "step": 3764 + }, + { + "epoch": 0.21035282286225102, + "grad_norm": 0.7683026194572449, + "learning_rate": 0.0008975235320484087, + "loss": 0.4555, + "step": 3765 + }, + { + "epoch": 0.2104086934659329, + "grad_norm": 0.9060300588607788, + "learning_rate": 0.000897495517705065, + "loss": 0.5858, + "step": 3766 + }, + { + "epoch": 0.21046456406961478, + "grad_norm": 0.48656606674194336, + "learning_rate": 0.0008974675033617212, + "loss": 0.4688, + "step": 3767 + }, + { + "epoch": 0.21052043467329665, + "grad_norm": 0.6615592241287231, + "learning_rate": 0.0008974394890183774, + "loss": 0.5476, + "step": 3768 + }, + { + "epoch": 0.21057630527697851, + "grad_norm": 0.6112557053565979, + "learning_rate": 0.0008974114746750336, + "loss": 0.5241, + "step": 3769 + }, + { + "epoch": 0.21063217588066038, + "grad_norm": 0.5809392929077148, + "learning_rate": 0.0008973834603316898, + "loss": 0.4097, + "step": 3770 + }, + { + "epoch": 0.21068804648434225, + "grad_norm": 0.7532441020011902, + "learning_rate": 0.000897355445988346, + "loss": 0.378, + "step": 3771 + }, + { + "epoch": 0.21074391708802415, + "grad_norm": 10.93654727935791, + "learning_rate": 0.0008973274316450022, + "loss": 0.4941, + "step": 3772 + }, + { + "epoch": 0.21079978769170601, + "grad_norm": 0.7417576909065247, + "learning_rate": 0.0008972994173016585, + "loss": 0.6091, + "step": 3773 + }, + { + "epoch": 0.21085565829538788, + "grad_norm": 0.6769590973854065, + "learning_rate": 0.0008972714029583147, + "loss": 0.4975, + "step": 3774 + }, + { + "epoch": 0.21091152889906975, + "grad_norm": 0.7837998270988464, + "learning_rate": 0.0008972433886149709, + "loss": 0.5312, + "step": 3775 + }, + { + "epoch": 0.21096739950275162, + "grad_norm": 0.6330733895301819, + "learning_rate": 0.0008972153742716272, + "loss": 0.5115, + "step": 3776 + }, + { + "epoch": 0.2110232701064335, + "grad_norm": 0.6158359050750732, + "learning_rate": 0.0008971873599282833, + "loss": 0.5134, + "step": 3777 + }, + { + "epoch": 0.21107914071011538, + "grad_norm": 0.5248937010765076, + "learning_rate": 0.0008971593455849395, + "loss": 0.4691, + "step": 3778 + }, + { + "epoch": 0.21113501131379725, + "grad_norm": 0.9401174187660217, + "learning_rate": 0.0008971313312415957, + "loss": 0.4642, + "step": 3779 + }, + { + "epoch": 0.21119088191747912, + "grad_norm": 2.616041421890259, + "learning_rate": 0.000897103316898252, + "loss": 0.532, + "step": 3780 + }, + { + "epoch": 0.21124675252116099, + "grad_norm": 8.550025939941406, + "learning_rate": 0.0008970753025549081, + "loss": 0.35, + "step": 3781 + }, + { + "epoch": 0.21130262312484285, + "grad_norm": 0.622586727142334, + "learning_rate": 0.0008970472882115643, + "loss": 0.503, + "step": 3782 + }, + { + "epoch": 0.21135849372852475, + "grad_norm": 0.7314617037773132, + "learning_rate": 0.0008970192738682206, + "loss": 0.4719, + "step": 3783 + }, + { + "epoch": 0.21141436433220662, + "grad_norm": 5.004934310913086, + "learning_rate": 0.0008969912595248767, + "loss": 0.5185, + "step": 3784 + }, + { + "epoch": 0.21147023493588848, + "grad_norm": 0.5641790628433228, + "learning_rate": 0.0008969632451815329, + "loss": 0.4346, + "step": 3785 + }, + { + "epoch": 0.21152610553957035, + "grad_norm": 0.7786459922790527, + "learning_rate": 0.0008969352308381891, + "loss": 0.5301, + "step": 3786 + }, + { + "epoch": 0.21158197614325222, + "grad_norm": 0.8518033623695374, + "learning_rate": 0.0008969072164948454, + "loss": 0.4069, + "step": 3787 + }, + { + "epoch": 0.2116378467469341, + "grad_norm": 0.6151695251464844, + "learning_rate": 0.0008968792021515015, + "loss": 0.4397, + "step": 3788 + }, + { + "epoch": 0.21169371735061598, + "grad_norm": 2.0726122856140137, + "learning_rate": 0.0008968511878081577, + "loss": 0.4793, + "step": 3789 + }, + { + "epoch": 0.21174958795429785, + "grad_norm": 0.3994410037994385, + "learning_rate": 0.0008968231734648141, + "loss": 0.3955, + "step": 3790 + }, + { + "epoch": 0.21180545855797972, + "grad_norm": 0.8450800776481628, + "learning_rate": 0.0008967951591214703, + "loss": 0.6328, + "step": 3791 + }, + { + "epoch": 0.2118613291616616, + "grad_norm": 0.9152820706367493, + "learning_rate": 0.0008967671447781264, + "loss": 0.5393, + "step": 3792 + }, + { + "epoch": 0.21191719976534346, + "grad_norm": 0.5171294808387756, + "learning_rate": 0.0008967391304347826, + "loss": 0.503, + "step": 3793 + }, + { + "epoch": 0.21197307036902532, + "grad_norm": 0.7821773886680603, + "learning_rate": 0.0008967111160914389, + "loss": 0.5141, + "step": 3794 + }, + { + "epoch": 0.21202894097270722, + "grad_norm": 0.7875365018844604, + "learning_rate": 0.0008966831017480951, + "loss": 0.5057, + "step": 3795 + }, + { + "epoch": 0.2120848115763891, + "grad_norm": 0.9026669263839722, + "learning_rate": 0.0008966550874047512, + "loss": 0.4969, + "step": 3796 + }, + { + "epoch": 0.21214068218007096, + "grad_norm": 1.118860125541687, + "learning_rate": 0.0008966270730614075, + "loss": 0.5686, + "step": 3797 + }, + { + "epoch": 0.21219655278375282, + "grad_norm": 0.5238868594169617, + "learning_rate": 0.0008965990587180637, + "loss": 0.471, + "step": 3798 + }, + { + "epoch": 0.2122524233874347, + "grad_norm": 0.6659799218177795, + "learning_rate": 0.0008965710443747199, + "loss": 0.5542, + "step": 3799 + }, + { + "epoch": 0.2123082939911166, + "grad_norm": 0.8771379590034485, + "learning_rate": 0.000896543030031376, + "loss": 0.5354, + "step": 3800 + }, + { + "epoch": 0.21236416459479845, + "grad_norm": 0.8020469546318054, + "learning_rate": 0.0008965150156880323, + "loss": 0.4526, + "step": 3801 + }, + { + "epoch": 0.21242003519848032, + "grad_norm": 3.2212743759155273, + "learning_rate": 0.0008964870013446885, + "loss": 0.5721, + "step": 3802 + }, + { + "epoch": 0.2124759058021622, + "grad_norm": 0.6716476082801819, + "learning_rate": 0.0008964589870013447, + "loss": 0.3927, + "step": 3803 + }, + { + "epoch": 0.21253177640584406, + "grad_norm": 0.7657524943351746, + "learning_rate": 0.0008964309726580008, + "loss": 0.444, + "step": 3804 + }, + { + "epoch": 0.21258764700952593, + "grad_norm": 0.6472284197807312, + "learning_rate": 0.0008964029583146571, + "loss": 0.4469, + "step": 3805 + }, + { + "epoch": 0.21264351761320782, + "grad_norm": 1.2406967878341675, + "learning_rate": 0.0008963749439713134, + "loss": 0.4559, + "step": 3806 + }, + { + "epoch": 0.2126993882168897, + "grad_norm": 0.46128907799720764, + "learning_rate": 0.0008963469296279695, + "loss": 0.3746, + "step": 3807 + }, + { + "epoch": 0.21275525882057156, + "grad_norm": 2.1250674724578857, + "learning_rate": 0.0008963189152846258, + "loss": 0.6162, + "step": 3808 + }, + { + "epoch": 0.21281112942425343, + "grad_norm": 0.5622358322143555, + "learning_rate": 0.000896290900941282, + "loss": 0.4796, + "step": 3809 + }, + { + "epoch": 0.2128670000279353, + "grad_norm": 0.6851847171783447, + "learning_rate": 0.0008962628865979382, + "loss": 0.4135, + "step": 3810 + }, + { + "epoch": 0.21292287063161716, + "grad_norm": 0.8282010555267334, + "learning_rate": 0.0008962348722545943, + "loss": 0.3885, + "step": 3811 + }, + { + "epoch": 0.21297874123529906, + "grad_norm": 0.4507001042366028, + "learning_rate": 0.0008962068579112506, + "loss": 0.3966, + "step": 3812 + }, + { + "epoch": 0.21303461183898093, + "grad_norm": 0.6374524831771851, + "learning_rate": 0.0008961788435679068, + "loss": 0.5376, + "step": 3813 + }, + { + "epoch": 0.2130904824426628, + "grad_norm": 1.3030565977096558, + "learning_rate": 0.000896150829224563, + "loss": 0.4307, + "step": 3814 + }, + { + "epoch": 0.21314635304634466, + "grad_norm": 1.0831102132797241, + "learning_rate": 0.0008961228148812192, + "loss": 0.6221, + "step": 3815 + }, + { + "epoch": 0.21320222365002653, + "grad_norm": 0.7258475422859192, + "learning_rate": 0.0008960948005378754, + "loss": 0.4829, + "step": 3816 + }, + { + "epoch": 0.21325809425370842, + "grad_norm": 0.9131131172180176, + "learning_rate": 0.0008960667861945316, + "loss": 0.3723, + "step": 3817 + }, + { + "epoch": 0.2133139648573903, + "grad_norm": 16.63396453857422, + "learning_rate": 0.0008960387718511878, + "loss": 0.4506, + "step": 3818 + }, + { + "epoch": 0.21336983546107216, + "grad_norm": 1.0513525009155273, + "learning_rate": 0.000896010757507844, + "loss": 0.5262, + "step": 3819 + }, + { + "epoch": 0.21342570606475403, + "grad_norm": 0.918570876121521, + "learning_rate": 0.0008959827431645002, + "loss": 0.4365, + "step": 3820 + }, + { + "epoch": 0.2134815766684359, + "grad_norm": 0.7982017397880554, + "learning_rate": 0.0008959547288211564, + "loss": 0.4788, + "step": 3821 + }, + { + "epoch": 0.21353744727211776, + "grad_norm": 0.8860021233558655, + "learning_rate": 0.0008959267144778128, + "loss": 0.4044, + "step": 3822 + }, + { + "epoch": 0.21359331787579966, + "grad_norm": 1.7084189653396606, + "learning_rate": 0.0008958987001344689, + "loss": 0.3842, + "step": 3823 + }, + { + "epoch": 0.21364918847948153, + "grad_norm": 0.7001113295555115, + "learning_rate": 0.0008958706857911251, + "loss": 0.3723, + "step": 3824 + }, + { + "epoch": 0.2137050590831634, + "grad_norm": 0.6935416460037231, + "learning_rate": 0.0008958426714477813, + "loss": 0.4551, + "step": 3825 + }, + { + "epoch": 0.21376092968684526, + "grad_norm": 0.768629252910614, + "learning_rate": 0.0008958146571044375, + "loss": 0.4494, + "step": 3826 + }, + { + "epoch": 0.21381680029052713, + "grad_norm": 0.9107675552368164, + "learning_rate": 0.0008957866427610937, + "loss": 0.4643, + "step": 3827 + }, + { + "epoch": 0.213872670894209, + "grad_norm": 0.48932817578315735, + "learning_rate": 0.0008957586284177499, + "loss": 0.4312, + "step": 3828 + }, + { + "epoch": 0.2139285414978909, + "grad_norm": 0.7728615999221802, + "learning_rate": 0.0008957306140744062, + "loss": 0.5168, + "step": 3829 + }, + { + "epoch": 0.21398441210157276, + "grad_norm": 0.7624552249908447, + "learning_rate": 0.0008957025997310623, + "loss": 0.5952, + "step": 3830 + }, + { + "epoch": 0.21404028270525463, + "grad_norm": 0.5440824031829834, + "learning_rate": 0.0008956745853877185, + "loss": 0.5101, + "step": 3831 + }, + { + "epoch": 0.2140961533089365, + "grad_norm": 0.760299563407898, + "learning_rate": 0.0008956465710443747, + "loss": 0.5181, + "step": 3832 + }, + { + "epoch": 0.21415202391261837, + "grad_norm": 1.2422231435775757, + "learning_rate": 0.000895618556701031, + "loss": 0.4097, + "step": 3833 + }, + { + "epoch": 0.21420789451630026, + "grad_norm": 0.8446205854415894, + "learning_rate": 0.0008955905423576871, + "loss": 0.5094, + "step": 3834 + }, + { + "epoch": 0.21426376511998213, + "grad_norm": 0.5309578776359558, + "learning_rate": 0.0008955625280143433, + "loss": 0.4761, + "step": 3835 + }, + { + "epoch": 0.214319635723664, + "grad_norm": 0.5860316157341003, + "learning_rate": 0.0008955345136709996, + "loss": 0.4377, + "step": 3836 + }, + { + "epoch": 0.21437550632734587, + "grad_norm": 0.5627002120018005, + "learning_rate": 0.0008955064993276558, + "loss": 0.5688, + "step": 3837 + }, + { + "epoch": 0.21443137693102773, + "grad_norm": 10.268046379089355, + "learning_rate": 0.0008954784849843119, + "loss": 0.496, + "step": 3838 + }, + { + "epoch": 0.2144872475347096, + "grad_norm": 0.7407131791114807, + "learning_rate": 0.0008954504706409681, + "loss": 0.56, + "step": 3839 + }, + { + "epoch": 0.2145431181383915, + "grad_norm": 13.633955955505371, + "learning_rate": 0.0008954224562976245, + "loss": 0.5426, + "step": 3840 + }, + { + "epoch": 0.21459898874207337, + "grad_norm": 0.8580158352851868, + "learning_rate": 0.0008953944419542807, + "loss": 0.4831, + "step": 3841 + }, + { + "epoch": 0.21465485934575523, + "grad_norm": 1.686303973197937, + "learning_rate": 0.0008953664276109368, + "loss": 0.5271, + "step": 3842 + }, + { + "epoch": 0.2147107299494371, + "grad_norm": 0.8238720893859863, + "learning_rate": 0.000895338413267593, + "loss": 0.5058, + "step": 3843 + }, + { + "epoch": 0.21476660055311897, + "grad_norm": 0.6019569635391235, + "learning_rate": 0.0008953103989242493, + "loss": 0.5683, + "step": 3844 + }, + { + "epoch": 0.21482247115680084, + "grad_norm": 0.5272091627120972, + "learning_rate": 0.0008952823845809055, + "loss": 0.5335, + "step": 3845 + }, + { + "epoch": 0.21487834176048273, + "grad_norm": 0.4811709225177765, + "learning_rate": 0.0008952543702375616, + "loss": 0.4926, + "step": 3846 + }, + { + "epoch": 0.2149342123641646, + "grad_norm": 0.4874570965766907, + "learning_rate": 0.0008952263558942179, + "loss": 0.3897, + "step": 3847 + }, + { + "epoch": 0.21499008296784647, + "grad_norm": 0.5404701828956604, + "learning_rate": 0.0008951983415508741, + "loss": 0.4498, + "step": 3848 + }, + { + "epoch": 0.21504595357152834, + "grad_norm": 0.5904034972190857, + "learning_rate": 0.0008951703272075302, + "loss": 0.4686, + "step": 3849 + }, + { + "epoch": 0.2151018241752102, + "grad_norm": 0.7663411498069763, + "learning_rate": 0.0008951423128641864, + "loss": 0.4899, + "step": 3850 + }, + { + "epoch": 0.21515769477889207, + "grad_norm": 0.5876378417015076, + "learning_rate": 0.0008951142985208427, + "loss": 0.5256, + "step": 3851 + }, + { + "epoch": 0.21521356538257397, + "grad_norm": 1.0562478303909302, + "learning_rate": 0.0008950862841774989, + "loss": 0.4491, + "step": 3852 + }, + { + "epoch": 0.21526943598625584, + "grad_norm": 0.5922881960868835, + "learning_rate": 0.000895058269834155, + "loss": 0.4465, + "step": 3853 + }, + { + "epoch": 0.2153253065899377, + "grad_norm": 1.202791690826416, + "learning_rate": 0.0008950302554908113, + "loss": 0.5189, + "step": 3854 + }, + { + "epoch": 0.21538117719361957, + "grad_norm": 0.7051172256469727, + "learning_rate": 0.0008950022411474675, + "loss": 0.4878, + "step": 3855 + }, + { + "epoch": 0.21543704779730144, + "grad_norm": 0.4516621232032776, + "learning_rate": 0.0008949742268041238, + "loss": 0.4171, + "step": 3856 + }, + { + "epoch": 0.21549291840098334, + "grad_norm": 0.6120790243148804, + "learning_rate": 0.0008949462124607799, + "loss": 0.61, + "step": 3857 + }, + { + "epoch": 0.2155487890046652, + "grad_norm": 2.017521619796753, + "learning_rate": 0.0008949181981174362, + "loss": 0.4692, + "step": 3858 + }, + { + "epoch": 0.21560465960834707, + "grad_norm": 0.5733019709587097, + "learning_rate": 0.0008948901837740924, + "loss": 0.4962, + "step": 3859 + }, + { + "epoch": 0.21566053021202894, + "grad_norm": 1.3965150117874146, + "learning_rate": 0.0008948621694307486, + "loss": 0.5517, + "step": 3860 + }, + { + "epoch": 0.2157164008157108, + "grad_norm": 1.4238839149475098, + "learning_rate": 0.0008948341550874048, + "loss": 0.4992, + "step": 3861 + }, + { + "epoch": 0.21577227141939267, + "grad_norm": 1.6210429668426514, + "learning_rate": 0.000894806140744061, + "loss": 0.372, + "step": 3862 + }, + { + "epoch": 0.21582814202307457, + "grad_norm": 0.5021910071372986, + "learning_rate": 0.0008947781264007172, + "loss": 0.3598, + "step": 3863 + }, + { + "epoch": 0.21588401262675644, + "grad_norm": 0.5159035921096802, + "learning_rate": 0.0008947501120573734, + "loss": 0.4508, + "step": 3864 + }, + { + "epoch": 0.2159398832304383, + "grad_norm": 0.41909360885620117, + "learning_rate": 0.0008947220977140296, + "loss": 0.4075, + "step": 3865 + }, + { + "epoch": 0.21599575383412017, + "grad_norm": 0.45368146896362305, + "learning_rate": 0.0008946940833706858, + "loss": 0.3985, + "step": 3866 + }, + { + "epoch": 0.21605162443780204, + "grad_norm": 0.5540871024131775, + "learning_rate": 0.000894666069027342, + "loss": 0.5051, + "step": 3867 + }, + { + "epoch": 0.2161074950414839, + "grad_norm": 0.7100427150726318, + "learning_rate": 0.0008946380546839982, + "loss": 0.5108, + "step": 3868 + }, + { + "epoch": 0.2161633656451658, + "grad_norm": 0.6388814449310303, + "learning_rate": 0.0008946100403406544, + "loss": 0.4171, + "step": 3869 + }, + { + "epoch": 0.21621923624884767, + "grad_norm": 0.44408515095710754, + "learning_rate": 0.0008945820259973106, + "loss": 0.4511, + "step": 3870 + }, + { + "epoch": 0.21627510685252954, + "grad_norm": 0.4026928246021271, + "learning_rate": 0.0008945540116539668, + "loss": 0.4463, + "step": 3871 + }, + { + "epoch": 0.2163309774562114, + "grad_norm": 1.345652461051941, + "learning_rate": 0.000894525997310623, + "loss": 0.4972, + "step": 3872 + }, + { + "epoch": 0.21638684805989328, + "grad_norm": 0.46807926893234253, + "learning_rate": 0.0008944979829672793, + "loss": 0.4629, + "step": 3873 + }, + { + "epoch": 0.21644271866357517, + "grad_norm": 0.4371721148490906, + "learning_rate": 0.0008944699686239355, + "loss": 0.3893, + "step": 3874 + }, + { + "epoch": 0.21649858926725704, + "grad_norm": 0.6310511231422424, + "learning_rate": 0.0008944419542805917, + "loss": 0.4809, + "step": 3875 + }, + { + "epoch": 0.2165544598709389, + "grad_norm": 0.5707883834838867, + "learning_rate": 0.0008944139399372479, + "loss": 0.466, + "step": 3876 + }, + { + "epoch": 0.21661033047462078, + "grad_norm": 0.5343812108039856, + "learning_rate": 0.0008943859255939041, + "loss": 0.5045, + "step": 3877 + }, + { + "epoch": 0.21666620107830264, + "grad_norm": 0.5613197088241577, + "learning_rate": 0.0008943579112505603, + "loss": 0.5446, + "step": 3878 + }, + { + "epoch": 0.2167220716819845, + "grad_norm": 0.5343694090843201, + "learning_rate": 0.0008943298969072166, + "loss": 0.4972, + "step": 3879 + }, + { + "epoch": 0.2167779422856664, + "grad_norm": 0.6170094609260559, + "learning_rate": 0.0008943018825638727, + "loss": 0.5306, + "step": 3880 + }, + { + "epoch": 0.21683381288934828, + "grad_norm": 1.1212972402572632, + "learning_rate": 0.0008942738682205289, + "loss": 0.5258, + "step": 3881 + }, + { + "epoch": 0.21688968349303014, + "grad_norm": 0.8429740071296692, + "learning_rate": 0.0008942458538771851, + "loss": 0.5517, + "step": 3882 + }, + { + "epoch": 0.216945554096712, + "grad_norm": 12.091228485107422, + "learning_rate": 0.0008942178395338414, + "loss": 0.4263, + "step": 3883 + }, + { + "epoch": 0.21700142470039388, + "grad_norm": 0.586048424243927, + "learning_rate": 0.0008941898251904975, + "loss": 0.474, + "step": 3884 + }, + { + "epoch": 0.21705729530407575, + "grad_norm": 1.7051026821136475, + "learning_rate": 0.0008941618108471537, + "loss": 0.4735, + "step": 3885 + }, + { + "epoch": 0.21711316590775764, + "grad_norm": 0.8701267838478088, + "learning_rate": 0.00089413379650381, + "loss": 0.446, + "step": 3886 + }, + { + "epoch": 0.2171690365114395, + "grad_norm": 4.476706027984619, + "learning_rate": 0.0008941057821604662, + "loss": 0.5378, + "step": 3887 + }, + { + "epoch": 0.21722490711512138, + "grad_norm": 0.7064699530601501, + "learning_rate": 0.0008940777678171223, + "loss": 0.3956, + "step": 3888 + }, + { + "epoch": 0.21728077771880325, + "grad_norm": 1.5636115074157715, + "learning_rate": 0.0008940497534737785, + "loss": 0.432, + "step": 3889 + }, + { + "epoch": 0.21733664832248512, + "grad_norm": 1.6558887958526611, + "learning_rate": 0.0008940217391304349, + "loss": 0.6477, + "step": 3890 + }, + { + "epoch": 0.217392518926167, + "grad_norm": 0.7441420555114746, + "learning_rate": 0.000893993724787091, + "loss": 0.4711, + "step": 3891 + }, + { + "epoch": 0.21744838952984888, + "grad_norm": 0.4970442056655884, + "learning_rate": 0.0008939657104437472, + "loss": 0.4306, + "step": 3892 + }, + { + "epoch": 0.21750426013353075, + "grad_norm": 3.099379777908325, + "learning_rate": 0.0008939376961004035, + "loss": 0.4792, + "step": 3893 + }, + { + "epoch": 0.21756013073721261, + "grad_norm": 0.5549399256706238, + "learning_rate": 0.0008939096817570597, + "loss": 0.5309, + "step": 3894 + }, + { + "epoch": 0.21761600134089448, + "grad_norm": 0.5294151306152344, + "learning_rate": 0.0008938816674137158, + "loss": 0.4681, + "step": 3895 + }, + { + "epoch": 0.21767187194457635, + "grad_norm": 3.640198230743408, + "learning_rate": 0.000893853653070372, + "loss": 0.439, + "step": 3896 + }, + { + "epoch": 0.21772774254825825, + "grad_norm": 0.5310273766517639, + "learning_rate": 0.0008938256387270283, + "loss": 0.402, + "step": 3897 + }, + { + "epoch": 0.21778361315194011, + "grad_norm": 0.4757938086986542, + "learning_rate": 0.0008937976243836845, + "loss": 0.4277, + "step": 3898 + }, + { + "epoch": 0.21783948375562198, + "grad_norm": 0.43462255597114563, + "learning_rate": 0.0008937696100403406, + "loss": 0.4111, + "step": 3899 + }, + { + "epoch": 0.21789535435930385, + "grad_norm": 0.7563605904579163, + "learning_rate": 0.0008937415956969969, + "loss": 0.5354, + "step": 3900 + }, + { + "epoch": 0.21795122496298572, + "grad_norm": 0.5849786400794983, + "learning_rate": 0.0008937135813536531, + "loss": 0.5526, + "step": 3901 + }, + { + "epoch": 0.21800709556666759, + "grad_norm": 0.43472397327423096, + "learning_rate": 0.0008936855670103093, + "loss": 0.4591, + "step": 3902 + }, + { + "epoch": 0.21806296617034948, + "grad_norm": 0.6649121642112732, + "learning_rate": 0.0008936575526669654, + "loss": 0.4375, + "step": 3903 + }, + { + "epoch": 0.21811883677403135, + "grad_norm": 10.136550903320312, + "learning_rate": 0.0008936295383236217, + "loss": 0.4843, + "step": 3904 + }, + { + "epoch": 0.21817470737771322, + "grad_norm": 0.6417964100837708, + "learning_rate": 0.000893601523980278, + "loss": 0.4961, + "step": 3905 + }, + { + "epoch": 0.21823057798139509, + "grad_norm": 0.554856538772583, + "learning_rate": 0.0008935735096369342, + "loss": 0.5045, + "step": 3906 + }, + { + "epoch": 0.21828644858507695, + "grad_norm": 0.5607935786247253, + "learning_rate": 0.0008935454952935903, + "loss": 0.5152, + "step": 3907 + }, + { + "epoch": 0.21834231918875882, + "grad_norm": 0.8672791123390198, + "learning_rate": 0.0008935174809502466, + "loss": 0.5617, + "step": 3908 + }, + { + "epoch": 0.21839818979244072, + "grad_norm": 1.3835567235946655, + "learning_rate": 0.0008934894666069028, + "loss": 0.3903, + "step": 3909 + }, + { + "epoch": 0.21845406039612258, + "grad_norm": 0.5252801179885864, + "learning_rate": 0.000893461452263559, + "loss": 0.5006, + "step": 3910 + }, + { + "epoch": 0.21850993099980445, + "grad_norm": 1.113996982574463, + "learning_rate": 0.0008934334379202152, + "loss": 0.5386, + "step": 3911 + }, + { + "epoch": 0.21856580160348632, + "grad_norm": 0.7158941030502319, + "learning_rate": 0.0008934054235768714, + "loss": 0.4238, + "step": 3912 + }, + { + "epoch": 0.2186216722071682, + "grad_norm": 0.5502124428749084, + "learning_rate": 0.0008933774092335276, + "loss": 0.4474, + "step": 3913 + }, + { + "epoch": 0.21867754281085008, + "grad_norm": 0.7130098938941956, + "learning_rate": 0.0008933493948901837, + "loss": 0.475, + "step": 3914 + }, + { + "epoch": 0.21873341341453195, + "grad_norm": 0.6645488142967224, + "learning_rate": 0.00089332138054684, + "loss": 0.6547, + "step": 3915 + }, + { + "epoch": 0.21878928401821382, + "grad_norm": 0.4206792116165161, + "learning_rate": 0.0008932933662034962, + "loss": 0.4275, + "step": 3916 + }, + { + "epoch": 0.2188451546218957, + "grad_norm": 0.45420965552330017, + "learning_rate": 0.0008932653518601524, + "loss": 0.5236, + "step": 3917 + }, + { + "epoch": 0.21890102522557756, + "grad_norm": 0.5523491501808167, + "learning_rate": 0.0008932373375168086, + "loss": 0.4419, + "step": 3918 + }, + { + "epoch": 0.21895689582925942, + "grad_norm": 0.6626412272453308, + "learning_rate": 0.0008932093231734648, + "loss": 0.4989, + "step": 3919 + }, + { + "epoch": 0.21901276643294132, + "grad_norm": 0.5199128985404968, + "learning_rate": 0.000893181308830121, + "loss": 0.4584, + "step": 3920 + }, + { + "epoch": 0.2190686370366232, + "grad_norm": 0.48792564868927, + "learning_rate": 0.0008931532944867772, + "loss": 0.4218, + "step": 3921 + }, + { + "epoch": 0.21912450764030506, + "grad_norm": 0.5987926125526428, + "learning_rate": 0.0008931252801434334, + "loss": 0.5029, + "step": 3922 + }, + { + "epoch": 0.21918037824398692, + "grad_norm": 0.8852396011352539, + "learning_rate": 0.0008930972658000897, + "loss": 0.4605, + "step": 3923 + }, + { + "epoch": 0.2192362488476688, + "grad_norm": 0.6552949547767639, + "learning_rate": 0.0008930692514567459, + "loss": 0.655, + "step": 3924 + }, + { + "epoch": 0.21929211945135066, + "grad_norm": 1.0978001356124878, + "learning_rate": 0.0008930412371134022, + "loss": 0.4285, + "step": 3925 + }, + { + "epoch": 0.21934799005503255, + "grad_norm": 0.44010767340660095, + "learning_rate": 0.0008930132227700583, + "loss": 0.4977, + "step": 3926 + }, + { + "epoch": 0.21940386065871442, + "grad_norm": 0.4398822784423828, + "learning_rate": 0.0008929852084267145, + "loss": 0.5483, + "step": 3927 + }, + { + "epoch": 0.2194597312623963, + "grad_norm": 0.6711917519569397, + "learning_rate": 0.0008929571940833707, + "loss": 0.5019, + "step": 3928 + }, + { + "epoch": 0.21951560186607816, + "grad_norm": 0.497565358877182, + "learning_rate": 0.000892929179740027, + "loss": 0.4383, + "step": 3929 + }, + { + "epoch": 0.21957147246976003, + "grad_norm": 0.5020688772201538, + "learning_rate": 0.0008929011653966831, + "loss": 0.5027, + "step": 3930 + }, + { + "epoch": 0.21962734307344192, + "grad_norm": 0.4561457931995392, + "learning_rate": 0.0008928731510533393, + "loss": 0.5233, + "step": 3931 + }, + { + "epoch": 0.2196832136771238, + "grad_norm": 0.7286627888679504, + "learning_rate": 0.0008928451367099956, + "loss": 0.4942, + "step": 3932 + }, + { + "epoch": 0.21973908428080566, + "grad_norm": 0.832646906375885, + "learning_rate": 0.0008928171223666517, + "loss": 0.5113, + "step": 3933 + }, + { + "epoch": 0.21979495488448753, + "grad_norm": 0.5741322040557861, + "learning_rate": 0.0008927891080233079, + "loss": 0.4468, + "step": 3934 + }, + { + "epoch": 0.2198508254881694, + "grad_norm": 0.658847451210022, + "learning_rate": 0.0008927610936799641, + "loss": 0.3979, + "step": 3935 + }, + { + "epoch": 0.21990669609185126, + "grad_norm": 0.5909132957458496, + "learning_rate": 0.0008927330793366204, + "loss": 0.4463, + "step": 3936 + }, + { + "epoch": 0.21996256669553316, + "grad_norm": 0.49905481934547424, + "learning_rate": 0.0008927050649932765, + "loss": 0.6695, + "step": 3937 + }, + { + "epoch": 0.22001843729921503, + "grad_norm": 0.31959378719329834, + "learning_rate": 0.0008926770506499327, + "loss": 0.3452, + "step": 3938 + }, + { + "epoch": 0.2200743079028969, + "grad_norm": 2.3500967025756836, + "learning_rate": 0.000892649036306589, + "loss": 0.489, + "step": 3939 + }, + { + "epoch": 0.22013017850657876, + "grad_norm": 0.6528761982917786, + "learning_rate": 0.0008926210219632453, + "loss": 0.4985, + "step": 3940 + }, + { + "epoch": 0.22018604911026063, + "grad_norm": 0.68022620677948, + "learning_rate": 0.0008925930076199014, + "loss": 0.6075, + "step": 3941 + }, + { + "epoch": 0.2202419197139425, + "grad_norm": 0.7997224926948547, + "learning_rate": 0.0008925649932765576, + "loss": 0.4856, + "step": 3942 + }, + { + "epoch": 0.2202977903176244, + "grad_norm": 0.6331867575645447, + "learning_rate": 0.0008925369789332139, + "loss": 0.4857, + "step": 3943 + }, + { + "epoch": 0.22035366092130626, + "grad_norm": 0.7228105068206787, + "learning_rate": 0.0008925089645898701, + "loss": 0.6824, + "step": 3944 + }, + { + "epoch": 0.22040953152498813, + "grad_norm": 0.3645518124103546, + "learning_rate": 0.0008924809502465262, + "loss": 0.4314, + "step": 3945 + }, + { + "epoch": 0.22046540212867, + "grad_norm": 0.47368374466896057, + "learning_rate": 0.0008924529359031824, + "loss": 0.4562, + "step": 3946 + }, + { + "epoch": 0.22052127273235186, + "grad_norm": 1.1678330898284912, + "learning_rate": 0.0008924249215598387, + "loss": 0.5465, + "step": 3947 + }, + { + "epoch": 0.22057714333603376, + "grad_norm": 1.7244764566421509, + "learning_rate": 0.0008923969072164949, + "loss": 0.4226, + "step": 3948 + }, + { + "epoch": 0.22063301393971563, + "grad_norm": 2.118135929107666, + "learning_rate": 0.000892368892873151, + "loss": 0.4802, + "step": 3949 + }, + { + "epoch": 0.2206888845433975, + "grad_norm": 1.3387562036514282, + "learning_rate": 0.0008923408785298073, + "loss": 0.5342, + "step": 3950 + }, + { + "epoch": 0.22074475514707936, + "grad_norm": 0.4732717275619507, + "learning_rate": 0.0008923128641864635, + "loss": 0.445, + "step": 3951 + }, + { + "epoch": 0.22080062575076123, + "grad_norm": 0.442640095949173, + "learning_rate": 0.0008922848498431197, + "loss": 0.5075, + "step": 3952 + }, + { + "epoch": 0.2208564963544431, + "grad_norm": 0.7812895774841309, + "learning_rate": 0.0008922568354997758, + "loss": 0.4742, + "step": 3953 + }, + { + "epoch": 0.220912366958125, + "grad_norm": 0.6620772480964661, + "learning_rate": 0.0008922288211564321, + "loss": 0.424, + "step": 3954 + }, + { + "epoch": 0.22096823756180686, + "grad_norm": 0.470889687538147, + "learning_rate": 0.0008922008068130883, + "loss": 0.4511, + "step": 3955 + }, + { + "epoch": 0.22102410816548873, + "grad_norm": 3.684222459793091, + "learning_rate": 0.0008921727924697444, + "loss": 0.5531, + "step": 3956 + }, + { + "epoch": 0.2210799787691706, + "grad_norm": 1.3747622966766357, + "learning_rate": 0.0008921447781264008, + "loss": 0.4099, + "step": 3957 + }, + { + "epoch": 0.22113584937285247, + "grad_norm": 1.7534475326538086, + "learning_rate": 0.000892116763783057, + "loss": 0.4816, + "step": 3958 + }, + { + "epoch": 0.22119171997653433, + "grad_norm": 0.6581915616989136, + "learning_rate": 0.0008920887494397132, + "loss": 0.4638, + "step": 3959 + }, + { + "epoch": 0.22124759058021623, + "grad_norm": 0.6816015243530273, + "learning_rate": 0.0008920607350963693, + "loss": 0.4944, + "step": 3960 + }, + { + "epoch": 0.2213034611838981, + "grad_norm": 0.5764141082763672, + "learning_rate": 0.0008920327207530256, + "loss": 0.5357, + "step": 3961 + }, + { + "epoch": 0.22135933178757997, + "grad_norm": 0.608946681022644, + "learning_rate": 0.0008920047064096818, + "loss": 0.4812, + "step": 3962 + }, + { + "epoch": 0.22141520239126183, + "grad_norm": 1.2516587972640991, + "learning_rate": 0.000891976692066338, + "loss": 0.5301, + "step": 3963 + }, + { + "epoch": 0.2214710729949437, + "grad_norm": 0.5165690779685974, + "learning_rate": 0.0008919486777229942, + "loss": 0.4562, + "step": 3964 + }, + { + "epoch": 0.22152694359862557, + "grad_norm": 37.83469009399414, + "learning_rate": 0.0008919206633796504, + "loss": 0.4594, + "step": 3965 + }, + { + "epoch": 0.22158281420230747, + "grad_norm": 0.5175660848617554, + "learning_rate": 0.0008918926490363066, + "loss": 0.5544, + "step": 3966 + }, + { + "epoch": 0.22163868480598933, + "grad_norm": 0.5487615466117859, + "learning_rate": 0.0008918646346929628, + "loss": 0.5065, + "step": 3967 + }, + { + "epoch": 0.2216945554096712, + "grad_norm": 0.5601195096969604, + "learning_rate": 0.000891836620349619, + "loss": 0.6016, + "step": 3968 + }, + { + "epoch": 0.22175042601335307, + "grad_norm": 0.6917957067489624, + "learning_rate": 0.0008918086060062752, + "loss": 0.5484, + "step": 3969 + }, + { + "epoch": 0.22180629661703494, + "grad_norm": 0.5843915343284607, + "learning_rate": 0.0008917805916629314, + "loss": 0.4305, + "step": 3970 + }, + { + "epoch": 0.22186216722071683, + "grad_norm": 0.588407576084137, + "learning_rate": 0.0008917525773195877, + "loss": 0.4207, + "step": 3971 + }, + { + "epoch": 0.2219180378243987, + "grad_norm": 1.9209349155426025, + "learning_rate": 0.0008917245629762438, + "loss": 0.5234, + "step": 3972 + }, + { + "epoch": 0.22197390842808057, + "grad_norm": 0.8092542886734009, + "learning_rate": 0.0008916965486329, + "loss": 0.5727, + "step": 3973 + }, + { + "epoch": 0.22202977903176244, + "grad_norm": 0.5815432667732239, + "learning_rate": 0.0008916685342895563, + "loss": 0.4097, + "step": 3974 + }, + { + "epoch": 0.2220856496354443, + "grad_norm": 0.6238686442375183, + "learning_rate": 0.0008916405199462125, + "loss": 0.4063, + "step": 3975 + }, + { + "epoch": 0.22214152023912617, + "grad_norm": 0.9054069519042969, + "learning_rate": 0.0008916125056028687, + "loss": 0.456, + "step": 3976 + }, + { + "epoch": 0.22219739084280807, + "grad_norm": 0.4318905472755432, + "learning_rate": 0.0008915844912595249, + "loss": 0.5123, + "step": 3977 + }, + { + "epoch": 0.22225326144648994, + "grad_norm": 0.8128601908683777, + "learning_rate": 0.0008915564769161812, + "loss": 0.4531, + "step": 3978 + }, + { + "epoch": 0.2223091320501718, + "grad_norm": 1.1233323812484741, + "learning_rate": 0.0008915284625728373, + "loss": 0.5325, + "step": 3979 + }, + { + "epoch": 0.22236500265385367, + "grad_norm": 0.7407247424125671, + "learning_rate": 0.0008915004482294935, + "loss": 0.6075, + "step": 3980 + }, + { + "epoch": 0.22242087325753554, + "grad_norm": 17.21967315673828, + "learning_rate": 0.0008914724338861497, + "loss": 0.4533, + "step": 3981 + }, + { + "epoch": 0.2224767438612174, + "grad_norm": 0.5168680548667908, + "learning_rate": 0.000891444419542806, + "loss": 0.62, + "step": 3982 + }, + { + "epoch": 0.2225326144648993, + "grad_norm": 0.5772609710693359, + "learning_rate": 0.0008914164051994621, + "loss": 0.4898, + "step": 3983 + }, + { + "epoch": 0.22258848506858117, + "grad_norm": 2.0520472526550293, + "learning_rate": 0.0008913883908561183, + "loss": 0.4806, + "step": 3984 + }, + { + "epoch": 0.22264435567226304, + "grad_norm": 0.509787380695343, + "learning_rate": 0.0008913603765127745, + "loss": 0.3898, + "step": 3985 + }, + { + "epoch": 0.2227002262759449, + "grad_norm": 0.4516289234161377, + "learning_rate": 0.0008913323621694308, + "loss": 0.4247, + "step": 3986 + }, + { + "epoch": 0.22275609687962677, + "grad_norm": 0.6648383140563965, + "learning_rate": 0.0008913043478260869, + "loss": 0.3738, + "step": 3987 + }, + { + "epoch": 0.22281196748330867, + "grad_norm": 0.436013400554657, + "learning_rate": 0.0008912763334827431, + "loss": 0.3798, + "step": 3988 + }, + { + "epoch": 0.22286783808699054, + "grad_norm": 2.674609899520874, + "learning_rate": 0.0008912483191393995, + "loss": 0.4987, + "step": 3989 + }, + { + "epoch": 0.2229237086906724, + "grad_norm": 4.7928147315979, + "learning_rate": 0.0008912203047960557, + "loss": 0.5059, + "step": 3990 + }, + { + "epoch": 0.22297957929435427, + "grad_norm": 0.8411749005317688, + "learning_rate": 0.0008911922904527118, + "loss": 0.483, + "step": 3991 + }, + { + "epoch": 0.22303544989803614, + "grad_norm": 0.6769095063209534, + "learning_rate": 0.000891164276109368, + "loss": 0.5143, + "step": 3992 + }, + { + "epoch": 0.223091320501718, + "grad_norm": 1.0033080577850342, + "learning_rate": 0.0008911362617660243, + "loss": 0.5062, + "step": 3993 + }, + { + "epoch": 0.2231471911053999, + "grad_norm": 1.6229004859924316, + "learning_rate": 0.0008911082474226805, + "loss": 0.4502, + "step": 3994 + }, + { + "epoch": 0.22320306170908177, + "grad_norm": 0.6522353291511536, + "learning_rate": 0.0008910802330793366, + "loss": 0.4458, + "step": 3995 + }, + { + "epoch": 0.22325893231276364, + "grad_norm": 4.084201812744141, + "learning_rate": 0.0008910522187359929, + "loss": 0.5433, + "step": 3996 + }, + { + "epoch": 0.2233148029164455, + "grad_norm": 4.081396579742432, + "learning_rate": 0.0008910242043926491, + "loss": 0.4546, + "step": 3997 + }, + { + "epoch": 0.22337067352012738, + "grad_norm": 0.5890430808067322, + "learning_rate": 0.0008909961900493052, + "loss": 0.5691, + "step": 3998 + }, + { + "epoch": 0.22342654412380925, + "grad_norm": 0.6269248723983765, + "learning_rate": 0.0008909681757059614, + "loss": 0.5888, + "step": 3999 + }, + { + "epoch": 0.22348241472749114, + "grad_norm": 0.6138067245483398, + "learning_rate": 0.0008909401613626177, + "loss": 0.4303, + "step": 4000 + }, + { + "epoch": 0.22348241472749114, + "eval_cer": 0.09798369938790874, + "eval_loss": 0.36631831526756287, + "eval_runtime": 56.1465, + "eval_samples_per_second": 80.824, + "eval_steps_per_second": 5.058, + "eval_wer": 0.3875310745509902, + "step": 4000 + }, + { + "epoch": 0.223538285331173, + "grad_norm": 0.570866048336029, + "learning_rate": 0.0008909121470192739, + "loss": 0.5311, + "step": 4001 + }, + { + "epoch": 0.22359415593485488, + "grad_norm": 0.5556325912475586, + "learning_rate": 0.00089088413267593, + "loss": 0.3696, + "step": 4002 + }, + { + "epoch": 0.22365002653853674, + "grad_norm": 0.5995553731918335, + "learning_rate": 0.0008908561183325863, + "loss": 0.3584, + "step": 4003 + }, + { + "epoch": 0.2237058971422186, + "grad_norm": 1.2446351051330566, + "learning_rate": 0.0008908281039892425, + "loss": 0.4912, + "step": 4004 + }, + { + "epoch": 0.2237617677459005, + "grad_norm": 0.4915957450866699, + "learning_rate": 0.0008908000896458987, + "loss": 0.4525, + "step": 4005 + }, + { + "epoch": 0.22381763834958238, + "grad_norm": 1.3055627346038818, + "learning_rate": 0.0008907720753025548, + "loss": 0.4891, + "step": 4006 + }, + { + "epoch": 0.22387350895326424, + "grad_norm": 0.6530665159225464, + "learning_rate": 0.0008907440609592112, + "loss": 0.6663, + "step": 4007 + }, + { + "epoch": 0.2239293795569461, + "grad_norm": 0.7777358293533325, + "learning_rate": 0.0008907160466158674, + "loss": 0.4864, + "step": 4008 + }, + { + "epoch": 0.22398525016062798, + "grad_norm": 0.6184050440788269, + "learning_rate": 0.0008906880322725236, + "loss": 0.3986, + "step": 4009 + }, + { + "epoch": 0.22404112076430985, + "grad_norm": 0.7062439322471619, + "learning_rate": 0.0008906600179291798, + "loss": 0.4682, + "step": 4010 + }, + { + "epoch": 0.22409699136799174, + "grad_norm": 0.6164241433143616, + "learning_rate": 0.000890632003585836, + "loss": 0.6316, + "step": 4011 + }, + { + "epoch": 0.2241528619716736, + "grad_norm": 1.1855337619781494, + "learning_rate": 0.0008906039892424922, + "loss": 0.4638, + "step": 4012 + }, + { + "epoch": 0.22420873257535548, + "grad_norm": 0.45490655303001404, + "learning_rate": 0.0008905759748991484, + "loss": 0.3536, + "step": 4013 + }, + { + "epoch": 0.22426460317903735, + "grad_norm": 1.538182020187378, + "learning_rate": 0.0008905479605558046, + "loss": 0.4883, + "step": 4014 + }, + { + "epoch": 0.22432047378271922, + "grad_norm": 0.9693992137908936, + "learning_rate": 0.0008905199462124608, + "loss": 0.543, + "step": 4015 + }, + { + "epoch": 0.22437634438640108, + "grad_norm": 1.293018102645874, + "learning_rate": 0.000890491931869117, + "loss": 0.447, + "step": 4016 + }, + { + "epoch": 0.22443221499008298, + "grad_norm": 0.4427696466445923, + "learning_rate": 0.0008904639175257731, + "loss": 0.4596, + "step": 4017 + }, + { + "epoch": 0.22448808559376485, + "grad_norm": 0.45733705163002014, + "learning_rate": 0.0008904359031824294, + "loss": 0.4763, + "step": 4018 + }, + { + "epoch": 0.22454395619744671, + "grad_norm": 2.1486289501190186, + "learning_rate": 0.0008904078888390856, + "loss": 0.4708, + "step": 4019 + }, + { + "epoch": 0.22459982680112858, + "grad_norm": 0.48881077766418457, + "learning_rate": 0.0008903798744957418, + "loss": 0.4161, + "step": 4020 + }, + { + "epoch": 0.22465569740481045, + "grad_norm": 0.5633781552314758, + "learning_rate": 0.000890351860152398, + "loss": 0.4804, + "step": 4021 + }, + { + "epoch": 0.22471156800849232, + "grad_norm": 0.576551079750061, + "learning_rate": 0.0008903238458090542, + "loss": 0.4881, + "step": 4022 + }, + { + "epoch": 0.22476743861217421, + "grad_norm": 0.8512468934059143, + "learning_rate": 0.0008902958314657105, + "loss": 0.5119, + "step": 4023 + }, + { + "epoch": 0.22482330921585608, + "grad_norm": 9.923932075500488, + "learning_rate": 0.0008902678171223667, + "loss": 0.4237, + "step": 4024 + }, + { + "epoch": 0.22487917981953795, + "grad_norm": 0.6624336242675781, + "learning_rate": 0.0008902398027790229, + "loss": 0.4265, + "step": 4025 + }, + { + "epoch": 0.22493505042321982, + "grad_norm": 0.6714197397232056, + "learning_rate": 0.0008902117884356791, + "loss": 0.4598, + "step": 4026 + }, + { + "epoch": 0.22499092102690169, + "grad_norm": 0.5006936192512512, + "learning_rate": 0.0008901837740923353, + "loss": 0.3684, + "step": 4027 + }, + { + "epoch": 0.22504679163058358, + "grad_norm": 0.47265687584877014, + "learning_rate": 0.0008901557597489916, + "loss": 0.5085, + "step": 4028 + }, + { + "epoch": 0.22510266223426545, + "grad_norm": 1.910417914390564, + "learning_rate": 0.0008901277454056477, + "loss": 0.6873, + "step": 4029 + }, + { + "epoch": 0.22515853283794732, + "grad_norm": 0.8129180669784546, + "learning_rate": 0.0008900997310623039, + "loss": 0.4897, + "step": 4030 + }, + { + "epoch": 0.22521440344162919, + "grad_norm": 0.46618661284446716, + "learning_rate": 0.0008900717167189601, + "loss": 0.4722, + "step": 4031 + }, + { + "epoch": 0.22527027404531105, + "grad_norm": 2.5965871810913086, + "learning_rate": 0.0008900437023756164, + "loss": 0.5127, + "step": 4032 + }, + { + "epoch": 0.22532614464899292, + "grad_norm": 0.5975320339202881, + "learning_rate": 0.0008900156880322725, + "loss": 0.407, + "step": 4033 + }, + { + "epoch": 0.22538201525267482, + "grad_norm": 0.8633830547332764, + "learning_rate": 0.0008899876736889287, + "loss": 0.5462, + "step": 4034 + }, + { + "epoch": 0.22543788585635668, + "grad_norm": 0.9786322712898254, + "learning_rate": 0.000889959659345585, + "loss": 0.5983, + "step": 4035 + }, + { + "epoch": 0.22549375646003855, + "grad_norm": 3.9991743564605713, + "learning_rate": 0.0008899316450022412, + "loss": 0.5059, + "step": 4036 + }, + { + "epoch": 0.22554962706372042, + "grad_norm": 4.2744364738464355, + "learning_rate": 0.0008899036306588973, + "loss": 0.6465, + "step": 4037 + }, + { + "epoch": 0.2256054976674023, + "grad_norm": 0.7274045944213867, + "learning_rate": 0.0008898756163155535, + "loss": 0.4999, + "step": 4038 + }, + { + "epoch": 0.22566136827108416, + "grad_norm": 0.9889260530471802, + "learning_rate": 0.0008898476019722099, + "loss": 0.7601, + "step": 4039 + }, + { + "epoch": 0.22571723887476605, + "grad_norm": 0.6077877879142761, + "learning_rate": 0.000889819587628866, + "loss": 0.4829, + "step": 4040 + }, + { + "epoch": 0.22577310947844792, + "grad_norm": 0.7499828338623047, + "learning_rate": 0.0008897915732855222, + "loss": 0.4684, + "step": 4041 + }, + { + "epoch": 0.2258289800821298, + "grad_norm": 0.6942073702812195, + "learning_rate": 0.0008897635589421785, + "loss": 0.4108, + "step": 4042 + }, + { + "epoch": 0.22588485068581166, + "grad_norm": 1.4965434074401855, + "learning_rate": 0.0008897355445988347, + "loss": 0.7168, + "step": 4043 + }, + { + "epoch": 0.22594072128949352, + "grad_norm": 0.6210461258888245, + "learning_rate": 0.0008897075302554908, + "loss": 0.6233, + "step": 4044 + }, + { + "epoch": 0.22599659189317542, + "grad_norm": 0.6624723076820374, + "learning_rate": 0.000889679515912147, + "loss": 0.4949, + "step": 4045 + }, + { + "epoch": 0.2260524624968573, + "grad_norm": 0.5725747346878052, + "learning_rate": 0.0008896515015688033, + "loss": 0.457, + "step": 4046 + }, + { + "epoch": 0.22610833310053916, + "grad_norm": 0.6206610202789307, + "learning_rate": 0.0008896234872254595, + "loss": 0.4992, + "step": 4047 + }, + { + "epoch": 0.22616420370422102, + "grad_norm": 0.6793126463890076, + "learning_rate": 0.0008895954728821156, + "loss": 0.5807, + "step": 4048 + }, + { + "epoch": 0.2262200743079029, + "grad_norm": 0.5154039859771729, + "learning_rate": 0.0008895674585387719, + "loss": 0.4782, + "step": 4049 + }, + { + "epoch": 0.22627594491158476, + "grad_norm": 0.5950642228126526, + "learning_rate": 0.0008895394441954281, + "loss": 0.5713, + "step": 4050 + }, + { + "epoch": 0.22633181551526665, + "grad_norm": 3.38019061088562, + "learning_rate": 0.0008895114298520843, + "loss": 0.6231, + "step": 4051 + }, + { + "epoch": 0.22638768611894852, + "grad_norm": 0.774253785610199, + "learning_rate": 0.0008894834155087404, + "loss": 0.5225, + "step": 4052 + }, + { + "epoch": 0.2264435567226304, + "grad_norm": 0.7485145330429077, + "learning_rate": 0.0008894554011653967, + "loss": 0.5263, + "step": 4053 + }, + { + "epoch": 0.22649942732631226, + "grad_norm": 1.0582728385925293, + "learning_rate": 0.0008894273868220529, + "loss": 0.5241, + "step": 4054 + }, + { + "epoch": 0.22655529792999413, + "grad_norm": 0.544423520565033, + "learning_rate": 0.0008893993724787091, + "loss": 0.3934, + "step": 4055 + }, + { + "epoch": 0.226611168533676, + "grad_norm": 0.7137669920921326, + "learning_rate": 0.0008893713581353652, + "loss": 0.4375, + "step": 4056 + }, + { + "epoch": 0.2266670391373579, + "grad_norm": 1.6917755603790283, + "learning_rate": 0.0008893433437920216, + "loss": 0.4541, + "step": 4057 + }, + { + "epoch": 0.22672290974103976, + "grad_norm": 0.7437692880630493, + "learning_rate": 0.0008893153294486778, + "loss": 0.4831, + "step": 4058 + }, + { + "epoch": 0.22677878034472163, + "grad_norm": 0.9113633036613464, + "learning_rate": 0.000889287315105334, + "loss": 0.5394, + "step": 4059 + }, + { + "epoch": 0.2268346509484035, + "grad_norm": 0.6214584112167358, + "learning_rate": 0.0008892593007619902, + "loss": 0.3643, + "step": 4060 + }, + { + "epoch": 0.22689052155208536, + "grad_norm": 0.5684372782707214, + "learning_rate": 0.0008892312864186464, + "loss": 0.5175, + "step": 4061 + }, + { + "epoch": 0.22694639215576723, + "grad_norm": 0.6084681153297424, + "learning_rate": 0.0008892032720753026, + "loss": 0.4731, + "step": 4062 + }, + { + "epoch": 0.22700226275944912, + "grad_norm": 0.4884560704231262, + "learning_rate": 0.0008891752577319587, + "loss": 0.419, + "step": 4063 + }, + { + "epoch": 0.227058133363131, + "grad_norm": 0.6482440233230591, + "learning_rate": 0.000889147243388615, + "loss": 0.4816, + "step": 4064 + }, + { + "epoch": 0.22711400396681286, + "grad_norm": 0.7099177837371826, + "learning_rate": 0.0008891192290452712, + "loss": 0.5382, + "step": 4065 + }, + { + "epoch": 0.22716987457049473, + "grad_norm": 1.0833303928375244, + "learning_rate": 0.0008890912147019274, + "loss": 0.3719, + "step": 4066 + }, + { + "epoch": 0.2272257451741766, + "grad_norm": 0.6549798846244812, + "learning_rate": 0.0008890632003585836, + "loss": 0.6366, + "step": 4067 + }, + { + "epoch": 0.2272816157778585, + "grad_norm": 0.8159138560295105, + "learning_rate": 0.0008890351860152398, + "loss": 0.5927, + "step": 4068 + }, + { + "epoch": 0.22733748638154036, + "grad_norm": 0.815945565700531, + "learning_rate": 0.000889007171671896, + "loss": 0.5223, + "step": 4069 + }, + { + "epoch": 0.22739335698522223, + "grad_norm": 0.5902562737464905, + "learning_rate": 0.0008889791573285522, + "loss": 0.4464, + "step": 4070 + }, + { + "epoch": 0.2274492275889041, + "grad_norm": 0.4283751845359802, + "learning_rate": 0.0008889511429852084, + "loss": 0.4585, + "step": 4071 + }, + { + "epoch": 0.22750509819258596, + "grad_norm": 0.5742055177688599, + "learning_rate": 0.0008889231286418646, + "loss": 0.4336, + "step": 4072 + }, + { + "epoch": 0.22756096879626783, + "grad_norm": 0.8261416554450989, + "learning_rate": 0.0008888951142985208, + "loss": 0.4182, + "step": 4073 + }, + { + "epoch": 0.22761683939994973, + "grad_norm": 1.053876519203186, + "learning_rate": 0.0008888670999551772, + "loss": 0.7973, + "step": 4074 + }, + { + "epoch": 0.2276727100036316, + "grad_norm": 2.603222131729126, + "learning_rate": 0.0008888390856118333, + "loss": 0.6624, + "step": 4075 + }, + { + "epoch": 0.22772858060731346, + "grad_norm": 0.5482119917869568, + "learning_rate": 0.0008888110712684895, + "loss": 0.5296, + "step": 4076 + }, + { + "epoch": 0.22778445121099533, + "grad_norm": 0.6303643584251404, + "learning_rate": 0.0008887830569251457, + "loss": 0.4781, + "step": 4077 + }, + { + "epoch": 0.2278403218146772, + "grad_norm": 17.615766525268555, + "learning_rate": 0.000888755042581802, + "loss": 0.5824, + "step": 4078 + }, + { + "epoch": 0.22789619241835907, + "grad_norm": 0.932546079158783, + "learning_rate": 0.0008887270282384581, + "loss": 0.5729, + "step": 4079 + }, + { + "epoch": 0.22795206302204096, + "grad_norm": 0.7653948664665222, + "learning_rate": 0.0008886990138951143, + "loss": 0.4299, + "step": 4080 + }, + { + "epoch": 0.22800793362572283, + "grad_norm": 0.9158506393432617, + "learning_rate": 0.0008886709995517706, + "loss": 0.4343, + "step": 4081 + }, + { + "epoch": 0.2280638042294047, + "grad_norm": 0.8838110566139221, + "learning_rate": 0.0008886429852084267, + "loss": 0.4899, + "step": 4082 + }, + { + "epoch": 0.22811967483308657, + "grad_norm": 1.2689201831817627, + "learning_rate": 0.0008886149708650829, + "loss": 0.4987, + "step": 4083 + }, + { + "epoch": 0.22817554543676843, + "grad_norm": 0.7487786412239075, + "learning_rate": 0.0008885869565217391, + "loss": 0.5045, + "step": 4084 + }, + { + "epoch": 0.22823141604045033, + "grad_norm": 0.8310326933860779, + "learning_rate": 0.0008885589421783954, + "loss": 0.4779, + "step": 4085 + }, + { + "epoch": 0.2282872866441322, + "grad_norm": 0.6724143028259277, + "learning_rate": 0.0008885309278350515, + "loss": 0.4754, + "step": 4086 + }, + { + "epoch": 0.22834315724781407, + "grad_norm": 0.9616978764533997, + "learning_rate": 0.0008885029134917077, + "loss": 0.6152, + "step": 4087 + }, + { + "epoch": 0.22839902785149593, + "grad_norm": 0.5692036151885986, + "learning_rate": 0.0008884748991483639, + "loss": 0.4544, + "step": 4088 + }, + { + "epoch": 0.2284548984551778, + "grad_norm": 0.6460684537887573, + "learning_rate": 0.0008884468848050203, + "loss": 0.5559, + "step": 4089 + }, + { + "epoch": 0.22851076905885967, + "grad_norm": 0.59904545545578, + "learning_rate": 0.0008884188704616763, + "loss": 0.6051, + "step": 4090 + }, + { + "epoch": 0.22856663966254157, + "grad_norm": 0.49216458201408386, + "learning_rate": 0.0008883908561183326, + "loss": 0.3992, + "step": 4091 + }, + { + "epoch": 0.22862251026622343, + "grad_norm": 0.4944162368774414, + "learning_rate": 0.0008883628417749889, + "loss": 0.5661, + "step": 4092 + }, + { + "epoch": 0.2286783808699053, + "grad_norm": 0.8962874412536621, + "learning_rate": 0.0008883348274316451, + "loss": 0.4075, + "step": 4093 + }, + { + "epoch": 0.22873425147358717, + "grad_norm": 0.7753958702087402, + "learning_rate": 0.0008883068130883012, + "loss": 0.4234, + "step": 4094 + }, + { + "epoch": 0.22879012207726904, + "grad_norm": 0.5497835278511047, + "learning_rate": 0.0008882787987449574, + "loss": 0.5146, + "step": 4095 + }, + { + "epoch": 0.2288459926809509, + "grad_norm": 3.014810562133789, + "learning_rate": 0.0008882507844016137, + "loss": 0.4927, + "step": 4096 + }, + { + "epoch": 0.2289018632846328, + "grad_norm": 0.7672441601753235, + "learning_rate": 0.0008882227700582699, + "loss": 0.6169, + "step": 4097 + }, + { + "epoch": 0.22895773388831467, + "grad_norm": 0.6990808844566345, + "learning_rate": 0.000888194755714926, + "loss": 0.5549, + "step": 4098 + }, + { + "epoch": 0.22901360449199654, + "grad_norm": 1.5984786748886108, + "learning_rate": 0.0008881667413715823, + "loss": 0.5376, + "step": 4099 + }, + { + "epoch": 0.2290694750956784, + "grad_norm": 0.6477362513542175, + "learning_rate": 0.0008881387270282385, + "loss": 0.4009, + "step": 4100 + }, + { + "epoch": 0.22912534569936027, + "grad_norm": 4.063788414001465, + "learning_rate": 0.0008881107126848947, + "loss": 0.4992, + "step": 4101 + }, + { + "epoch": 0.22918121630304217, + "grad_norm": 0.7975984811782837, + "learning_rate": 0.0008880826983415508, + "loss": 0.4372, + "step": 4102 + }, + { + "epoch": 0.22923708690672404, + "grad_norm": 0.6561374068260193, + "learning_rate": 0.0008880546839982071, + "loss": 0.5344, + "step": 4103 + }, + { + "epoch": 0.2292929575104059, + "grad_norm": 0.6760945320129395, + "learning_rate": 0.0008880266696548633, + "loss": 0.4841, + "step": 4104 + }, + { + "epoch": 0.22934882811408777, + "grad_norm": 0.72145015001297, + "learning_rate": 0.0008879986553115194, + "loss": 0.445, + "step": 4105 + }, + { + "epoch": 0.22940469871776964, + "grad_norm": 0.649454653263092, + "learning_rate": 0.0008879706409681757, + "loss": 0.4093, + "step": 4106 + }, + { + "epoch": 0.2294605693214515, + "grad_norm": 14.74445629119873, + "learning_rate": 0.000887942626624832, + "loss": 0.5048, + "step": 4107 + }, + { + "epoch": 0.2295164399251334, + "grad_norm": 0.9731113314628601, + "learning_rate": 0.0008879146122814882, + "loss": 0.484, + "step": 4108 + }, + { + "epoch": 0.22957231052881527, + "grad_norm": 1.872645378112793, + "learning_rate": 0.0008878865979381443, + "loss": 0.4591, + "step": 4109 + }, + { + "epoch": 0.22962818113249714, + "grad_norm": 2.003380060195923, + "learning_rate": 0.0008878585835948006, + "loss": 0.5134, + "step": 4110 + }, + { + "epoch": 0.229684051736179, + "grad_norm": 0.5488346815109253, + "learning_rate": 0.0008878305692514568, + "loss": 0.5776, + "step": 4111 + }, + { + "epoch": 0.22973992233986087, + "grad_norm": 4.012121677398682, + "learning_rate": 0.000887802554908113, + "loss": 0.4473, + "step": 4112 + }, + { + "epoch": 0.22979579294354274, + "grad_norm": 0.7532994747161865, + "learning_rate": 0.0008877745405647692, + "loss": 0.5257, + "step": 4113 + }, + { + "epoch": 0.22985166354722464, + "grad_norm": 2.989384412765503, + "learning_rate": 0.0008877465262214254, + "loss": 0.5658, + "step": 4114 + }, + { + "epoch": 0.2299075341509065, + "grad_norm": 0.9988793730735779, + "learning_rate": 0.0008877185118780816, + "loss": 0.5473, + "step": 4115 + }, + { + "epoch": 0.22996340475458837, + "grad_norm": 1.4696277379989624, + "learning_rate": 0.0008876904975347378, + "loss": 0.4301, + "step": 4116 + }, + { + "epoch": 0.23001927535827024, + "grad_norm": 1.1142463684082031, + "learning_rate": 0.000887662483191394, + "loss": 0.5017, + "step": 4117 + }, + { + "epoch": 0.2300751459619521, + "grad_norm": 1.6017396450042725, + "learning_rate": 0.0008876344688480502, + "loss": 0.3322, + "step": 4118 + }, + { + "epoch": 0.23013101656563398, + "grad_norm": 8.469249725341797, + "learning_rate": 0.0008876064545047064, + "loss": 0.5739, + "step": 4119 + }, + { + "epoch": 0.23018688716931587, + "grad_norm": 0.49803560972213745, + "learning_rate": 0.0008875784401613627, + "loss": 0.4506, + "step": 4120 + }, + { + "epoch": 0.23024275777299774, + "grad_norm": 0.8713529706001282, + "learning_rate": 0.0008875504258180188, + "loss": 0.5481, + "step": 4121 + }, + { + "epoch": 0.2302986283766796, + "grad_norm": 0.8964894413948059, + "learning_rate": 0.000887522411474675, + "loss": 0.4914, + "step": 4122 + }, + { + "epoch": 0.23035449898036148, + "grad_norm": 0.6333115100860596, + "learning_rate": 0.0008874943971313312, + "loss": 0.4589, + "step": 4123 + }, + { + "epoch": 0.23041036958404335, + "grad_norm": 0.674534261226654, + "learning_rate": 0.0008874663827879875, + "loss": 0.4321, + "step": 4124 + }, + { + "epoch": 0.23046624018772524, + "grad_norm": 0.7373489737510681, + "learning_rate": 0.0008874383684446437, + "loss": 0.4979, + "step": 4125 + }, + { + "epoch": 0.2305221107914071, + "grad_norm": 0.946284294128418, + "learning_rate": 0.0008874103541012999, + "loss": 0.4782, + "step": 4126 + }, + { + "epoch": 0.23057798139508898, + "grad_norm": 0.602643609046936, + "learning_rate": 0.0008873823397579561, + "loss": 0.4129, + "step": 4127 + }, + { + "epoch": 0.23063385199877084, + "grad_norm": 0.6372336149215698, + "learning_rate": 0.0008873543254146123, + "loss": 0.6265, + "step": 4128 + }, + { + "epoch": 0.2306897226024527, + "grad_norm": 0.7090263962745667, + "learning_rate": 0.0008873263110712685, + "loss": 0.3985, + "step": 4129 + }, + { + "epoch": 0.23074559320613458, + "grad_norm": 2.037280797958374, + "learning_rate": 0.0008872982967279247, + "loss": 0.4967, + "step": 4130 + }, + { + "epoch": 0.23080146380981648, + "grad_norm": 1.0498453378677368, + "learning_rate": 0.000887270282384581, + "loss": 0.5379, + "step": 4131 + }, + { + "epoch": 0.23085733441349834, + "grad_norm": 0.8330334424972534, + "learning_rate": 0.0008872422680412371, + "loss": 0.5894, + "step": 4132 + }, + { + "epoch": 0.2309132050171802, + "grad_norm": 0.5115959644317627, + "learning_rate": 0.0008872142536978933, + "loss": 0.4027, + "step": 4133 + }, + { + "epoch": 0.23096907562086208, + "grad_norm": 0.7372562289237976, + "learning_rate": 0.0008871862393545495, + "loss": 0.5132, + "step": 4134 + }, + { + "epoch": 0.23102494622454395, + "grad_norm": 0.757757306098938, + "learning_rate": 0.0008871582250112058, + "loss": 0.4629, + "step": 4135 + }, + { + "epoch": 0.23108081682822582, + "grad_norm": 0.4416324198246002, + "learning_rate": 0.0008871302106678619, + "loss": 0.3974, + "step": 4136 + }, + { + "epoch": 0.2311366874319077, + "grad_norm": 0.779175341129303, + "learning_rate": 0.0008871021963245181, + "loss": 0.4259, + "step": 4137 + }, + { + "epoch": 0.23119255803558958, + "grad_norm": 1.0031583309173584, + "learning_rate": 0.0008870741819811744, + "loss": 0.4478, + "step": 4138 + }, + { + "epoch": 0.23124842863927145, + "grad_norm": 0.8423380255699158, + "learning_rate": 0.0008870461676378307, + "loss": 0.4142, + "step": 4139 + }, + { + "epoch": 0.23130429924295332, + "grad_norm": 1.0636897087097168, + "learning_rate": 0.0008870181532944867, + "loss": 0.5253, + "step": 4140 + }, + { + "epoch": 0.23136016984663518, + "grad_norm": 0.88477623462677, + "learning_rate": 0.000886990138951143, + "loss": 0.4615, + "step": 4141 + }, + { + "epoch": 0.23141604045031708, + "grad_norm": 0.5518140196800232, + "learning_rate": 0.0008869621246077993, + "loss": 0.4873, + "step": 4142 + }, + { + "epoch": 0.23147191105399895, + "grad_norm": 1.412856936454773, + "learning_rate": 0.0008869341102644555, + "loss": 0.524, + "step": 4143 + }, + { + "epoch": 0.23152778165768081, + "grad_norm": 2.0048749446868896, + "learning_rate": 0.0008869060959211116, + "loss": 0.4086, + "step": 4144 + }, + { + "epoch": 0.23158365226136268, + "grad_norm": 0.6785237789154053, + "learning_rate": 0.0008868780815777679, + "loss": 0.5417, + "step": 4145 + }, + { + "epoch": 0.23163952286504455, + "grad_norm": 0.5273226499557495, + "learning_rate": 0.0008868500672344241, + "loss": 0.3489, + "step": 4146 + }, + { + "epoch": 0.23169539346872642, + "grad_norm": 0.6882335543632507, + "learning_rate": 0.0008868220528910802, + "loss": 0.549, + "step": 4147 + }, + { + "epoch": 0.23175126407240831, + "grad_norm": 0.4936909079551697, + "learning_rate": 0.0008867940385477364, + "loss": 0.3936, + "step": 4148 + }, + { + "epoch": 0.23180713467609018, + "grad_norm": 0.5566964149475098, + "learning_rate": 0.0008867660242043927, + "loss": 0.4845, + "step": 4149 + }, + { + "epoch": 0.23186300527977205, + "grad_norm": 0.8627514243125916, + "learning_rate": 0.0008867380098610489, + "loss": 0.4092, + "step": 4150 + }, + { + "epoch": 0.23191887588345392, + "grad_norm": 0.5589913725852966, + "learning_rate": 0.000886709995517705, + "loss": 0.5083, + "step": 4151 + }, + { + "epoch": 0.23197474648713579, + "grad_norm": 0.5308307409286499, + "learning_rate": 0.0008866819811743613, + "loss": 0.4023, + "step": 4152 + }, + { + "epoch": 0.23203061709081765, + "grad_norm": 0.553930401802063, + "learning_rate": 0.0008866539668310175, + "loss": 0.467, + "step": 4153 + }, + { + "epoch": 0.23208648769449955, + "grad_norm": 0.5052673816680908, + "learning_rate": 0.0008866259524876737, + "loss": 0.3992, + "step": 4154 + }, + { + "epoch": 0.23214235829818142, + "grad_norm": 0.44688311219215393, + "learning_rate": 0.0008865979381443298, + "loss": 0.3519, + "step": 4155 + }, + { + "epoch": 0.23219822890186329, + "grad_norm": 0.7223708033561707, + "learning_rate": 0.0008865699238009861, + "loss": 0.4072, + "step": 4156 + }, + { + "epoch": 0.23225409950554515, + "grad_norm": 0.6906889081001282, + "learning_rate": 0.0008865419094576424, + "loss": 0.675, + "step": 4157 + }, + { + "epoch": 0.23230997010922702, + "grad_norm": 3.141852378845215, + "learning_rate": 0.0008865138951142986, + "loss": 0.4269, + "step": 4158 + }, + { + "epoch": 0.23236584071290892, + "grad_norm": 0.5158798694610596, + "learning_rate": 0.0008864858807709547, + "loss": 0.5571, + "step": 4159 + }, + { + "epoch": 0.23242171131659078, + "grad_norm": 0.4623003900051117, + "learning_rate": 0.000886457866427611, + "loss": 0.498, + "step": 4160 + }, + { + "epoch": 0.23247758192027265, + "grad_norm": 0.6572248339653015, + "learning_rate": 0.0008864298520842672, + "loss": 0.5037, + "step": 4161 + }, + { + "epoch": 0.23253345252395452, + "grad_norm": 0.623054027557373, + "learning_rate": 0.0008864018377409234, + "loss": 0.4116, + "step": 4162 + }, + { + "epoch": 0.2325893231276364, + "grad_norm": 0.5163794159889221, + "learning_rate": 0.0008863738233975796, + "loss": 0.3655, + "step": 4163 + }, + { + "epoch": 0.23264519373131826, + "grad_norm": 10.138554573059082, + "learning_rate": 0.0008863458090542358, + "loss": 0.4396, + "step": 4164 + }, + { + "epoch": 0.23270106433500015, + "grad_norm": 0.648532509803772, + "learning_rate": 0.000886317794710892, + "loss": 0.3925, + "step": 4165 + }, + { + "epoch": 0.23275693493868202, + "grad_norm": 0.7129004597663879, + "learning_rate": 0.0008862897803675481, + "loss": 0.587, + "step": 4166 + }, + { + "epoch": 0.2328128055423639, + "grad_norm": 0.4831683039665222, + "learning_rate": 0.0008862617660242044, + "loss": 0.4683, + "step": 4167 + }, + { + "epoch": 0.23286867614604576, + "grad_norm": 0.9249421954154968, + "learning_rate": 0.0008862337516808606, + "loss": 0.4526, + "step": 4168 + }, + { + "epoch": 0.23292454674972762, + "grad_norm": 1.3727049827575684, + "learning_rate": 0.0008862057373375168, + "loss": 0.4895, + "step": 4169 + }, + { + "epoch": 0.2329804173534095, + "grad_norm": 9.471634864807129, + "learning_rate": 0.000886177722994173, + "loss": 0.4479, + "step": 4170 + }, + { + "epoch": 0.2330362879570914, + "grad_norm": 0.5847155451774597, + "learning_rate": 0.0008861497086508292, + "loss": 0.4633, + "step": 4171 + }, + { + "epoch": 0.23309215856077325, + "grad_norm": Infinity, + "learning_rate": 0.0008861497086508292, + "loss": 0.4257, + "step": 4172 + }, + { + "epoch": 0.23314802916445512, + "grad_norm": 0.7210089564323425, + "learning_rate": 0.0008861216943074854, + "loss": 0.525, + "step": 4173 + }, + { + "epoch": 0.233203899768137, + "grad_norm": 0.5260524749755859, + "learning_rate": 0.0008860936799641416, + "loss": 0.5151, + "step": 4174 + }, + { + "epoch": 0.23325977037181886, + "grad_norm": 0.5985652804374695, + "learning_rate": 0.0008860656656207979, + "loss": 0.5084, + "step": 4175 + }, + { + "epoch": 0.23331564097550073, + "grad_norm": 0.42989441752433777, + "learning_rate": 0.0008860376512774541, + "loss": 0.5038, + "step": 4176 + }, + { + "epoch": 0.23337151157918262, + "grad_norm": 0.5076581835746765, + "learning_rate": 0.0008860096369341103, + "loss": 0.5537, + "step": 4177 + }, + { + "epoch": 0.2334273821828645, + "grad_norm": 0.5735991597175598, + "learning_rate": 0.0008859816225907666, + "loss": 0.4575, + "step": 4178 + }, + { + "epoch": 0.23348325278654636, + "grad_norm": 0.6869438886642456, + "learning_rate": 0.0008859536082474227, + "loss": 0.4661, + "step": 4179 + }, + { + "epoch": 0.23353912339022823, + "grad_norm": 5.422333240509033, + "learning_rate": 0.0008859255939040789, + "loss": 0.4962, + "step": 4180 + }, + { + "epoch": 0.2335949939939101, + "grad_norm": 0.489166259765625, + "learning_rate": 0.0008858975795607351, + "loss": 0.4366, + "step": 4181 + }, + { + "epoch": 0.233650864597592, + "grad_norm": 0.5072371959686279, + "learning_rate": 0.0008858695652173914, + "loss": 0.3488, + "step": 4182 + }, + { + "epoch": 0.23370673520127386, + "grad_norm": 0.529441237449646, + "learning_rate": 0.0008858415508740475, + "loss": 0.6002, + "step": 4183 + }, + { + "epoch": 0.23376260580495573, + "grad_norm": 0.6955692172050476, + "learning_rate": 0.0008858135365307037, + "loss": 0.5219, + "step": 4184 + }, + { + "epoch": 0.2338184764086376, + "grad_norm": 0.7149230241775513, + "learning_rate": 0.00088578552218736, + "loss": 0.5274, + "step": 4185 + }, + { + "epoch": 0.23387434701231946, + "grad_norm": 0.8157262802124023, + "learning_rate": 0.0008857575078440162, + "loss": 0.4758, + "step": 4186 + }, + { + "epoch": 0.23393021761600133, + "grad_norm": 0.7421581745147705, + "learning_rate": 0.0008857294935006723, + "loss": 0.5784, + "step": 4187 + }, + { + "epoch": 0.23398608821968322, + "grad_norm": 0.8523985147476196, + "learning_rate": 0.0008857014791573285, + "loss": 0.5245, + "step": 4188 + }, + { + "epoch": 0.2340419588233651, + "grad_norm": 0.6057237386703491, + "learning_rate": 0.0008856734648139848, + "loss": 0.395, + "step": 4189 + }, + { + "epoch": 0.23409782942704696, + "grad_norm": 0.710608184337616, + "learning_rate": 0.0008856454504706409, + "loss": 0.5282, + "step": 4190 + }, + { + "epoch": 0.23415370003072883, + "grad_norm": 0.5958489775657654, + "learning_rate": 0.0008856174361272971, + "loss": 0.5662, + "step": 4191 + }, + { + "epoch": 0.2342095706344107, + "grad_norm": 3.5915932655334473, + "learning_rate": 0.0008855894217839535, + "loss": 0.5172, + "step": 4192 + }, + { + "epoch": 0.23426544123809256, + "grad_norm": 1.037196397781372, + "learning_rate": 0.0008855614074406097, + "loss": 0.5627, + "step": 4193 + }, + { + "epoch": 0.23432131184177446, + "grad_norm": 0.8437638282775879, + "learning_rate": 0.0008855333930972658, + "loss": 0.4615, + "step": 4194 + }, + { + "epoch": 0.23437718244545633, + "grad_norm": 0.5504167079925537, + "learning_rate": 0.000885505378753922, + "loss": 0.4797, + "step": 4195 + }, + { + "epoch": 0.2344330530491382, + "grad_norm": 2.8451380729675293, + "learning_rate": 0.0008854773644105783, + "loss": 0.4173, + "step": 4196 + }, + { + "epoch": 0.23448892365282006, + "grad_norm": 1.0700137615203857, + "learning_rate": 0.0008854493500672345, + "loss": 0.5434, + "step": 4197 + }, + { + "epoch": 0.23454479425650193, + "grad_norm": 0.904691219329834, + "learning_rate": 0.0008854213357238906, + "loss": 0.3855, + "step": 4198 + }, + { + "epoch": 0.23460066486018383, + "grad_norm": 0.49944645166397095, + "learning_rate": 0.0008853933213805468, + "loss": 0.4228, + "step": 4199 + }, + { + "epoch": 0.2346565354638657, + "grad_norm": 2.5445780754089355, + "learning_rate": 0.0008853653070372031, + "loss": 0.543, + "step": 4200 + }, + { + "epoch": 0.23471240606754756, + "grad_norm": 0.5192331075668335, + "learning_rate": 0.0008853372926938593, + "loss": 0.4554, + "step": 4201 + }, + { + "epoch": 0.23476827667122943, + "grad_norm": 0.9988318681716919, + "learning_rate": 0.0008853092783505154, + "loss": 0.4159, + "step": 4202 + }, + { + "epoch": 0.2348241472749113, + "grad_norm": 0.9794596433639526, + "learning_rate": 0.0008852812640071717, + "loss": 0.4655, + "step": 4203 + }, + { + "epoch": 0.23488001787859317, + "grad_norm": 0.6887660622596741, + "learning_rate": 0.0008852532496638279, + "loss": 0.5283, + "step": 4204 + }, + { + "epoch": 0.23493588848227506, + "grad_norm": 0.9118051528930664, + "learning_rate": 0.0008852252353204841, + "loss": 0.6218, + "step": 4205 + }, + { + "epoch": 0.23499175908595693, + "grad_norm": 1.6195811033248901, + "learning_rate": 0.0008851972209771402, + "loss": 0.487, + "step": 4206 + }, + { + "epoch": 0.2350476296896388, + "grad_norm": 0.8179507255554199, + "learning_rate": 0.0008851692066337965, + "loss": 0.3385, + "step": 4207 + }, + { + "epoch": 0.23510350029332067, + "grad_norm": 3.3914875984191895, + "learning_rate": 0.0008851411922904528, + "loss": 0.4054, + "step": 4208 + }, + { + "epoch": 0.23515937089700253, + "grad_norm": 2.501088857650757, + "learning_rate": 0.0008851131779471089, + "loss": 0.5266, + "step": 4209 + }, + { + "epoch": 0.2352152415006844, + "grad_norm": 0.5762655735015869, + "learning_rate": 0.0008850851636037652, + "loss": 0.4898, + "step": 4210 + }, + { + "epoch": 0.2352711121043663, + "grad_norm": 0.5274091362953186, + "learning_rate": 0.0008850571492604214, + "loss": 0.4342, + "step": 4211 + }, + { + "epoch": 0.23532698270804817, + "grad_norm": 0.4818646013736725, + "learning_rate": 0.0008850291349170776, + "loss": 0.3862, + "step": 4212 + }, + { + "epoch": 0.23538285331173003, + "grad_norm": 0.6624579429626465, + "learning_rate": 0.0008850011205737337, + "loss": 0.4406, + "step": 4213 + }, + { + "epoch": 0.2354387239154119, + "grad_norm": 0.5341525673866272, + "learning_rate": 0.00088497310623039, + "loss": 0.4819, + "step": 4214 + }, + { + "epoch": 0.23549459451909377, + "grad_norm": 0.5675890445709229, + "learning_rate": 0.0008849450918870462, + "loss": 0.4617, + "step": 4215 + }, + { + "epoch": 0.23555046512277567, + "grad_norm": 0.47306960821151733, + "learning_rate": 0.0008849170775437024, + "loss": 0.3462, + "step": 4216 + }, + { + "epoch": 0.23560633572645753, + "grad_norm": 0.5306919813156128, + "learning_rate": 0.0008848890632003586, + "loss": 0.4651, + "step": 4217 + }, + { + "epoch": 0.2356622063301394, + "grad_norm": 0.4023691415786743, + "learning_rate": 0.0008848610488570148, + "loss": 0.3822, + "step": 4218 + }, + { + "epoch": 0.23571807693382127, + "grad_norm": 0.5351731181144714, + "learning_rate": 0.000884833034513671, + "loss": 0.5193, + "step": 4219 + }, + { + "epoch": 0.23577394753750314, + "grad_norm": 0.7928450107574463, + "learning_rate": 0.0008848050201703272, + "loss": 0.3996, + "step": 4220 + }, + { + "epoch": 0.235829818141185, + "grad_norm": 0.804872989654541, + "learning_rate": 0.0008847770058269834, + "loss": 0.5708, + "step": 4221 + }, + { + "epoch": 0.2358856887448669, + "grad_norm": 0.5764588713645935, + "learning_rate": 0.0008847489914836396, + "loss": 0.4993, + "step": 4222 + }, + { + "epoch": 0.23594155934854877, + "grad_norm": 0.7982374429702759, + "learning_rate": 0.0008847209771402958, + "loss": 0.5, + "step": 4223 + }, + { + "epoch": 0.23599742995223064, + "grad_norm": 0.44566041231155396, + "learning_rate": 0.0008846929627969522, + "loss": 0.5077, + "step": 4224 + }, + { + "epoch": 0.2360533005559125, + "grad_norm": 0.4217692017555237, + "learning_rate": 0.0008846649484536083, + "loss": 0.5462, + "step": 4225 + }, + { + "epoch": 0.23610917115959437, + "grad_norm": 0.38973963260650635, + "learning_rate": 0.0008846369341102645, + "loss": 0.4421, + "step": 4226 + }, + { + "epoch": 0.23616504176327624, + "grad_norm": 0.5745378732681274, + "learning_rate": 0.0008846089197669207, + "loss": 0.4506, + "step": 4227 + }, + { + "epoch": 0.23622091236695814, + "grad_norm": 0.5290927290916443, + "learning_rate": 0.000884580905423577, + "loss": 0.4118, + "step": 4228 + }, + { + "epoch": 0.23627678297064, + "grad_norm": 0.559008002281189, + "learning_rate": 0.0008845528910802331, + "loss": 0.5591, + "step": 4229 + }, + { + "epoch": 0.23633265357432187, + "grad_norm": 0.8064879179000854, + "learning_rate": 0.0008845248767368893, + "loss": 0.4806, + "step": 4230 + }, + { + "epoch": 0.23638852417800374, + "grad_norm": 0.45895543694496155, + "learning_rate": 0.0008844968623935456, + "loss": 0.4316, + "step": 4231 + }, + { + "epoch": 0.2364443947816856, + "grad_norm": 0.7882242798805237, + "learning_rate": 0.0008844688480502017, + "loss": 0.4681, + "step": 4232 + }, + { + "epoch": 0.23650026538536748, + "grad_norm": 0.4650938808917999, + "learning_rate": 0.0008844408337068579, + "loss": 0.457, + "step": 4233 + }, + { + "epoch": 0.23655613598904937, + "grad_norm": 1.0505732297897339, + "learning_rate": 0.0008844128193635141, + "loss": 0.5422, + "step": 4234 + }, + { + "epoch": 0.23661200659273124, + "grad_norm": 3.038475751876831, + "learning_rate": 0.0008843848050201704, + "loss": 0.406, + "step": 4235 + }, + { + "epoch": 0.2366678771964131, + "grad_norm": 2.6374447345733643, + "learning_rate": 0.0008843567906768265, + "loss": 0.3744, + "step": 4236 + }, + { + "epoch": 0.23672374780009497, + "grad_norm": 0.7039818167686462, + "learning_rate": 0.0008843287763334827, + "loss": 0.4668, + "step": 4237 + }, + { + "epoch": 0.23677961840377684, + "grad_norm": 0.6271088123321533, + "learning_rate": 0.0008843007619901389, + "loss": 0.4161, + "step": 4238 + }, + { + "epoch": 0.23683548900745874, + "grad_norm": 1.6361582279205322, + "learning_rate": 0.0008842727476467952, + "loss": 0.4255, + "step": 4239 + }, + { + "epoch": 0.2368913596111406, + "grad_norm": 0.4566899538040161, + "learning_rate": 0.0008842447333034513, + "loss": 0.3852, + "step": 4240 + }, + { + "epoch": 0.23694723021482247, + "grad_norm": 0.41672447323799133, + "learning_rate": 0.0008842167189601075, + "loss": 0.3869, + "step": 4241 + }, + { + "epoch": 0.23700310081850434, + "grad_norm": 0.5750576853752136, + "learning_rate": 0.0008841887046167639, + "loss": 0.4062, + "step": 4242 + }, + { + "epoch": 0.2370589714221862, + "grad_norm": 0.4703211784362793, + "learning_rate": 0.0008841606902734201, + "loss": 0.4356, + "step": 4243 + }, + { + "epoch": 0.23711484202586808, + "grad_norm": 0.6200658679008484, + "learning_rate": 0.0008841326759300762, + "loss": 0.5998, + "step": 4244 + }, + { + "epoch": 0.23717071262954997, + "grad_norm": 0.6404126882553101, + "learning_rate": 0.0008841046615867324, + "loss": 0.4943, + "step": 4245 + }, + { + "epoch": 0.23722658323323184, + "grad_norm": 0.9376481771469116, + "learning_rate": 0.0008840766472433887, + "loss": 0.377, + "step": 4246 + }, + { + "epoch": 0.2372824538369137, + "grad_norm": 0.45378533005714417, + "learning_rate": 0.0008840486329000449, + "loss": 0.5193, + "step": 4247 + }, + { + "epoch": 0.23733832444059558, + "grad_norm": 0.6312829852104187, + "learning_rate": 0.000884020618556701, + "loss": 0.6419, + "step": 4248 + }, + { + "epoch": 0.23739419504427745, + "grad_norm": 0.4994530975818634, + "learning_rate": 0.0008839926042133573, + "loss": 0.444, + "step": 4249 + }, + { + "epoch": 0.2374500656479593, + "grad_norm": 0.8314129114151001, + "learning_rate": 0.0008839645898700135, + "loss": 0.4925, + "step": 4250 + }, + { + "epoch": 0.2375059362516412, + "grad_norm": 0.9790481925010681, + "learning_rate": 0.0008839365755266697, + "loss": 0.4668, + "step": 4251 + }, + { + "epoch": 0.23756180685532308, + "grad_norm": 0.48881545662879944, + "learning_rate": 0.0008839085611833258, + "loss": 0.5233, + "step": 4252 + }, + { + "epoch": 0.23761767745900494, + "grad_norm": 0.9651234745979309, + "learning_rate": 0.0008838805468399821, + "loss": 0.4088, + "step": 4253 + }, + { + "epoch": 0.2376735480626868, + "grad_norm": 0.6589600443840027, + "learning_rate": 0.0008838525324966383, + "loss": 0.5178, + "step": 4254 + }, + { + "epoch": 0.23772941866636868, + "grad_norm": 0.4721037745475769, + "learning_rate": 0.0008838245181532944, + "loss": 0.4211, + "step": 4255 + }, + { + "epoch": 0.23778528927005058, + "grad_norm": 0.8703773021697998, + "learning_rate": 0.0008837965038099507, + "loss": 0.5119, + "step": 4256 + }, + { + "epoch": 0.23784115987373244, + "grad_norm": 1.2180160284042358, + "learning_rate": 0.000883768489466607, + "loss": 0.4086, + "step": 4257 + }, + { + "epoch": 0.2378970304774143, + "grad_norm": 0.821360170841217, + "learning_rate": 0.0008837404751232632, + "loss": 0.6186, + "step": 4258 + }, + { + "epoch": 0.23795290108109618, + "grad_norm": 0.4745534360408783, + "learning_rate": 0.0008837124607799193, + "loss": 0.427, + "step": 4259 + }, + { + "epoch": 0.23800877168477805, + "grad_norm": 1.4504953622817993, + "learning_rate": 0.0008836844464365756, + "loss": 0.4507, + "step": 4260 + }, + { + "epoch": 0.23806464228845992, + "grad_norm": 0.46805354952812195, + "learning_rate": 0.0008836564320932318, + "loss": 0.4707, + "step": 4261 + }, + { + "epoch": 0.2381205128921418, + "grad_norm": 0.5915274620056152, + "learning_rate": 0.000883628417749888, + "loss": 0.4923, + "step": 4262 + }, + { + "epoch": 0.23817638349582368, + "grad_norm": 0.7700567245483398, + "learning_rate": 0.0008836004034065442, + "loss": 0.373, + "step": 4263 + }, + { + "epoch": 0.23823225409950555, + "grad_norm": 0.5113480091094971, + "learning_rate": 0.0008835723890632004, + "loss": 0.4818, + "step": 4264 + }, + { + "epoch": 0.23828812470318742, + "grad_norm": 0.848321259021759, + "learning_rate": 0.0008835443747198566, + "loss": 0.5481, + "step": 4265 + }, + { + "epoch": 0.23834399530686928, + "grad_norm": 0.5361381769180298, + "learning_rate": 0.0008835163603765128, + "loss": 0.4349, + "step": 4266 + }, + { + "epoch": 0.23839986591055115, + "grad_norm": 0.5383433103561401, + "learning_rate": 0.000883488346033169, + "loss": 0.5581, + "step": 4267 + }, + { + "epoch": 0.23845573651423305, + "grad_norm": 0.7325769662857056, + "learning_rate": 0.0008834603316898252, + "loss": 0.453, + "step": 4268 + }, + { + "epoch": 0.23851160711791491, + "grad_norm": 0.4301282465457916, + "learning_rate": 0.0008834323173464814, + "loss": 0.4288, + "step": 4269 + }, + { + "epoch": 0.23856747772159678, + "grad_norm": 1.4636316299438477, + "learning_rate": 0.0008834043030031376, + "loss": 0.5438, + "step": 4270 + }, + { + "epoch": 0.23862334832527865, + "grad_norm": 1.3826282024383545, + "learning_rate": 0.0008833762886597938, + "loss": 0.5757, + "step": 4271 + }, + { + "epoch": 0.23867921892896052, + "grad_norm": 0.5977545380592346, + "learning_rate": 0.00088334827431645, + "loss": 0.3701, + "step": 4272 + }, + { + "epoch": 0.2387350895326424, + "grad_norm": 0.5707186460494995, + "learning_rate": 0.0008833202599731062, + "loss": 0.5467, + "step": 4273 + }, + { + "epoch": 0.23879096013632428, + "grad_norm": 0.976052463054657, + "learning_rate": 0.0008832922456297624, + "loss": 0.4633, + "step": 4274 + }, + { + "epoch": 0.23884683074000615, + "grad_norm": 0.5536332130432129, + "learning_rate": 0.0008832642312864187, + "loss": 0.4397, + "step": 4275 + }, + { + "epoch": 0.23890270134368802, + "grad_norm": 0.43175211548805237, + "learning_rate": 0.0008832362169430749, + "loss": 0.4421, + "step": 4276 + }, + { + "epoch": 0.23895857194736989, + "grad_norm": 0.4942137897014618, + "learning_rate": 0.0008832082025997311, + "loss": 0.4045, + "step": 4277 + }, + { + "epoch": 0.23901444255105175, + "grad_norm": 0.527834415435791, + "learning_rate": 0.0008831801882563873, + "loss": 0.4458, + "step": 4278 + }, + { + "epoch": 0.23907031315473365, + "grad_norm": 0.5546362996101379, + "learning_rate": 0.0008831521739130435, + "loss": 0.5283, + "step": 4279 + }, + { + "epoch": 0.23912618375841552, + "grad_norm": 0.5700463652610779, + "learning_rate": 0.0008831241595696997, + "loss": 0.4108, + "step": 4280 + }, + { + "epoch": 0.23918205436209738, + "grad_norm": 1.534970760345459, + "learning_rate": 0.000883096145226356, + "loss": 0.473, + "step": 4281 + }, + { + "epoch": 0.23923792496577925, + "grad_norm": 0.47014686465263367, + "learning_rate": 0.0008830681308830121, + "loss": 0.4535, + "step": 4282 + }, + { + "epoch": 0.23929379556946112, + "grad_norm": 0.4483439326286316, + "learning_rate": 0.0008830401165396683, + "loss": 0.4075, + "step": 4283 + }, + { + "epoch": 0.239349666173143, + "grad_norm": 1.3196183443069458, + "learning_rate": 0.0008830121021963245, + "loss": 0.5578, + "step": 4284 + }, + { + "epoch": 0.23940553677682488, + "grad_norm": 0.505124568939209, + "learning_rate": 0.0008829840878529808, + "loss": 0.4428, + "step": 4285 + }, + { + "epoch": 0.23946140738050675, + "grad_norm": 0.571721076965332, + "learning_rate": 0.0008829560735096369, + "loss": 0.602, + "step": 4286 + }, + { + "epoch": 0.23951727798418862, + "grad_norm": 0.5268236994743347, + "learning_rate": 0.0008829280591662931, + "loss": 0.5252, + "step": 4287 + }, + { + "epoch": 0.2395731485878705, + "grad_norm": 0.4758719503879547, + "learning_rate": 0.0008829000448229494, + "loss": 0.3768, + "step": 4288 + }, + { + "epoch": 0.23962901919155236, + "grad_norm": 0.5382834076881409, + "learning_rate": 0.0008828720304796056, + "loss": 0.5169, + "step": 4289 + }, + { + "epoch": 0.23968488979523422, + "grad_norm": 0.9572944045066833, + "learning_rate": 0.0008828440161362617, + "loss": 0.5048, + "step": 4290 + }, + { + "epoch": 0.23974076039891612, + "grad_norm": 0.6876468062400818, + "learning_rate": 0.0008828160017929179, + "loss": 0.4753, + "step": 4291 + }, + { + "epoch": 0.239796631002598, + "grad_norm": 0.9336578845977783, + "learning_rate": 0.0008827879874495743, + "loss": 0.5578, + "step": 4292 + }, + { + "epoch": 0.23985250160627986, + "grad_norm": 0.5807604193687439, + "learning_rate": 0.0008827599731062305, + "loss": 0.5069, + "step": 4293 + }, + { + "epoch": 0.23990837220996172, + "grad_norm": 0.8924793004989624, + "learning_rate": 0.0008827319587628866, + "loss": 0.4465, + "step": 4294 + }, + { + "epoch": 0.2399642428136436, + "grad_norm": 0.8446182012557983, + "learning_rate": 0.0008827039444195429, + "loss": 0.6157, + "step": 4295 + }, + { + "epoch": 0.2400201134173255, + "grad_norm": 0.6588237285614014, + "learning_rate": 0.0008826759300761991, + "loss": 0.4165, + "step": 4296 + }, + { + "epoch": 0.24007598402100735, + "grad_norm": 0.907112181186676, + "learning_rate": 0.0008826479157328552, + "loss": 0.5889, + "step": 4297 + }, + { + "epoch": 0.24013185462468922, + "grad_norm": 0.6333624124526978, + "learning_rate": 0.0008826199013895114, + "loss": 0.44, + "step": 4298 + }, + { + "epoch": 0.2401877252283711, + "grad_norm": 0.9743716716766357, + "learning_rate": 0.0008825918870461677, + "loss": 0.4366, + "step": 4299 + }, + { + "epoch": 0.24024359583205296, + "grad_norm": 0.6167629957199097, + "learning_rate": 0.0008825638727028239, + "loss": 0.5228, + "step": 4300 + }, + { + "epoch": 0.24029946643573483, + "grad_norm": 0.6480828523635864, + "learning_rate": 0.00088253585835948, + "loss": 0.4906, + "step": 4301 + }, + { + "epoch": 0.24035533703941672, + "grad_norm": 0.6434074640274048, + "learning_rate": 0.0008825078440161362, + "loss": 0.6244, + "step": 4302 + }, + { + "epoch": 0.2404112076430986, + "grad_norm": 1.404990553855896, + "learning_rate": 0.0008824798296727925, + "loss": 0.5044, + "step": 4303 + }, + { + "epoch": 0.24046707824678046, + "grad_norm": 0.5399903059005737, + "learning_rate": 0.0008824518153294487, + "loss": 0.5264, + "step": 4304 + }, + { + "epoch": 0.24052294885046233, + "grad_norm": 1.2506591081619263, + "learning_rate": 0.0008824238009861048, + "loss": 0.4755, + "step": 4305 + }, + { + "epoch": 0.2405788194541442, + "grad_norm": 1.6921353340148926, + "learning_rate": 0.0008823957866427611, + "loss": 0.5239, + "step": 4306 + }, + { + "epoch": 0.24063469005782606, + "grad_norm": 0.5752133727073669, + "learning_rate": 0.0008823677722994173, + "loss": 0.4652, + "step": 4307 + }, + { + "epoch": 0.24069056066150796, + "grad_norm": 0.6807466745376587, + "learning_rate": 0.0008823397579560736, + "loss": 0.4002, + "step": 4308 + }, + { + "epoch": 0.24074643126518983, + "grad_norm": 0.7182154059410095, + "learning_rate": 0.0008823117436127296, + "loss": 0.535, + "step": 4309 + }, + { + "epoch": 0.2408023018688717, + "grad_norm": 0.5059728026390076, + "learning_rate": 0.000882283729269386, + "loss": 0.3614, + "step": 4310 + }, + { + "epoch": 0.24085817247255356, + "grad_norm": 0.6142624020576477, + "learning_rate": 0.0008822557149260422, + "loss": 0.4789, + "step": 4311 + }, + { + "epoch": 0.24091404307623543, + "grad_norm": 0.5324705839157104, + "learning_rate": 0.0008822277005826984, + "loss": 0.3918, + "step": 4312 + }, + { + "epoch": 0.24096991367991732, + "grad_norm": 0.5393995046615601, + "learning_rate": 0.0008821996862393546, + "loss": 0.3825, + "step": 4313 + }, + { + "epoch": 0.2410257842835992, + "grad_norm": 0.44544553756713867, + "learning_rate": 0.0008821716718960108, + "loss": 0.4053, + "step": 4314 + }, + { + "epoch": 0.24108165488728106, + "grad_norm": 0.8924400210380554, + "learning_rate": 0.000882143657552667, + "loss": 0.5782, + "step": 4315 + }, + { + "epoch": 0.24113752549096293, + "grad_norm": 0.7637109160423279, + "learning_rate": 0.0008821156432093231, + "loss": 0.6026, + "step": 4316 + }, + { + "epoch": 0.2411933960946448, + "grad_norm": 2.6056230068206787, + "learning_rate": 0.0008820876288659794, + "loss": 0.4113, + "step": 4317 + }, + { + "epoch": 0.24124926669832666, + "grad_norm": 0.5834433436393738, + "learning_rate": 0.0008820596145226356, + "loss": 0.3904, + "step": 4318 + }, + { + "epoch": 0.24130513730200856, + "grad_norm": 6.947653293609619, + "learning_rate": 0.0008820316001792918, + "loss": 0.5408, + "step": 4319 + }, + { + "epoch": 0.24136100790569043, + "grad_norm": 0.5851271748542786, + "learning_rate": 0.000882003585835948, + "loss": 0.5138, + "step": 4320 + }, + { + "epoch": 0.2414168785093723, + "grad_norm": 0.4363996386528015, + "learning_rate": 0.0008819755714926042, + "loss": 0.5599, + "step": 4321 + }, + { + "epoch": 0.24147274911305416, + "grad_norm": 0.6079261898994446, + "learning_rate": 0.0008819475571492604, + "loss": 0.5225, + "step": 4322 + }, + { + "epoch": 0.24152861971673603, + "grad_norm": 0.6317042708396912, + "learning_rate": 0.0008819195428059166, + "loss": 0.5464, + "step": 4323 + }, + { + "epoch": 0.2415844903204179, + "grad_norm": 0.5486106872558594, + "learning_rate": 0.0008818915284625728, + "loss": 0.5552, + "step": 4324 + }, + { + "epoch": 0.2416403609240998, + "grad_norm": 0.4987889230251312, + "learning_rate": 0.000881863514119229, + "loss": 0.4251, + "step": 4325 + }, + { + "epoch": 0.24169623152778166, + "grad_norm": 0.576528787612915, + "learning_rate": 0.0008818354997758853, + "loss": 0.4194, + "step": 4326 + }, + { + "epoch": 0.24175210213146353, + "grad_norm": 0.5335289239883423, + "learning_rate": 0.0008818074854325416, + "loss": 0.478, + "step": 4327 + }, + { + "epoch": 0.2418079727351454, + "grad_norm": 0.4711935520172119, + "learning_rate": 0.0008817794710891977, + "loss": 0.4135, + "step": 4328 + }, + { + "epoch": 0.24186384333882727, + "grad_norm": 10.794709205627441, + "learning_rate": 0.0008817514567458539, + "loss": 0.4095, + "step": 4329 + }, + { + "epoch": 0.24191971394250916, + "grad_norm": 0.678676426410675, + "learning_rate": 0.0008817234424025101, + "loss": 0.3332, + "step": 4330 + }, + { + "epoch": 0.24197558454619103, + "grad_norm": 0.7200943231582642, + "learning_rate": 0.0008816954280591664, + "loss": 0.4955, + "step": 4331 + }, + { + "epoch": 0.2420314551498729, + "grad_norm": 0.5162609219551086, + "learning_rate": 0.0008816674137158225, + "loss": 0.5778, + "step": 4332 + }, + { + "epoch": 0.24208732575355477, + "grad_norm": 0.6007496118545532, + "learning_rate": 0.0008816393993724787, + "loss": 0.4812, + "step": 4333 + }, + { + "epoch": 0.24214319635723663, + "grad_norm": 0.9287883639335632, + "learning_rate": 0.000881611385029135, + "loss": 0.6423, + "step": 4334 + }, + { + "epoch": 0.2421990669609185, + "grad_norm": 2.7612037658691406, + "learning_rate": 0.0008815833706857912, + "loss": 0.5849, + "step": 4335 + }, + { + "epoch": 0.2422549375646004, + "grad_norm": 2.072117805480957, + "learning_rate": 0.0008815553563424473, + "loss": 0.4409, + "step": 4336 + }, + { + "epoch": 0.24231080816828227, + "grad_norm": 0.6003206372261047, + "learning_rate": 0.0008815273419991035, + "loss": 0.3887, + "step": 4337 + }, + { + "epoch": 0.24236667877196413, + "grad_norm": 0.7381556034088135, + "learning_rate": 0.0008814993276557598, + "loss": 0.5583, + "step": 4338 + }, + { + "epoch": 0.242422549375646, + "grad_norm": 0.5890124440193176, + "learning_rate": 0.0008814713133124159, + "loss": 0.4035, + "step": 4339 + }, + { + "epoch": 0.24247841997932787, + "grad_norm": 0.48684296011924744, + "learning_rate": 0.0008814432989690721, + "loss": 0.4236, + "step": 4340 + }, + { + "epoch": 0.24253429058300974, + "grad_norm": 0.9939450025558472, + "learning_rate": 0.0008814152846257283, + "loss": 0.5196, + "step": 4341 + }, + { + "epoch": 0.24259016118669163, + "grad_norm": 0.793200671672821, + "learning_rate": 0.0008813872702823847, + "loss": 0.6271, + "step": 4342 + }, + { + "epoch": 0.2426460317903735, + "grad_norm": 0.7231776118278503, + "learning_rate": 0.0008813592559390408, + "loss": 0.507, + "step": 4343 + }, + { + "epoch": 0.24270190239405537, + "grad_norm": 0.6038108468055725, + "learning_rate": 0.000881331241595697, + "loss": 0.4627, + "step": 4344 + }, + { + "epoch": 0.24275777299773724, + "grad_norm": 0.8145732283592224, + "learning_rate": 0.0008813032272523533, + "loss": 0.5424, + "step": 4345 + }, + { + "epoch": 0.2428136436014191, + "grad_norm": 0.6509616374969482, + "learning_rate": 0.0008812752129090095, + "loss": 0.5066, + "step": 4346 + }, + { + "epoch": 0.24286951420510097, + "grad_norm": 0.900558352470398, + "learning_rate": 0.0008812471985656656, + "loss": 0.3978, + "step": 4347 + }, + { + "epoch": 0.24292538480878287, + "grad_norm": 0.8102753162384033, + "learning_rate": 0.0008812191842223218, + "loss": 0.4681, + "step": 4348 + }, + { + "epoch": 0.24298125541246474, + "grad_norm": 2.3449559211730957, + "learning_rate": 0.0008811911698789781, + "loss": 0.4766, + "step": 4349 + }, + { + "epoch": 0.2430371260161466, + "grad_norm": 0.7419986724853516, + "learning_rate": 0.0008811631555356343, + "loss": 0.551, + "step": 4350 + }, + { + "epoch": 0.24309299661982847, + "grad_norm": 0.3914238512516022, + "learning_rate": 0.0008811351411922904, + "loss": 0.3914, + "step": 4351 + }, + { + "epoch": 0.24314886722351034, + "grad_norm": 0.7410634756088257, + "learning_rate": 0.0008811071268489467, + "loss": 0.5304, + "step": 4352 + }, + { + "epoch": 0.24320473782719224, + "grad_norm": 3.1059303283691406, + "learning_rate": 0.0008810791125056029, + "loss": 0.4303, + "step": 4353 + }, + { + "epoch": 0.2432606084308741, + "grad_norm": 1.0598517656326294, + "learning_rate": 0.0008810510981622591, + "loss": 0.5653, + "step": 4354 + }, + { + "epoch": 0.24331647903455597, + "grad_norm": 0.5405341982841492, + "learning_rate": 0.0008810230838189152, + "loss": 0.567, + "step": 4355 + }, + { + "epoch": 0.24337234963823784, + "grad_norm": 1.0491735935211182, + "learning_rate": 0.0008809950694755715, + "loss": 0.5375, + "step": 4356 + }, + { + "epoch": 0.2434282202419197, + "grad_norm": 0.6657913327217102, + "learning_rate": 0.0008809670551322277, + "loss": 0.5777, + "step": 4357 + }, + { + "epoch": 0.24348409084560158, + "grad_norm": 0.5535717606544495, + "learning_rate": 0.0008809390407888838, + "loss": 0.4424, + "step": 4358 + }, + { + "epoch": 0.24353996144928347, + "grad_norm": 0.5317490696907043, + "learning_rate": 0.0008809110264455402, + "loss": 0.4083, + "step": 4359 + }, + { + "epoch": 0.24359583205296534, + "grad_norm": 0.5491297245025635, + "learning_rate": 0.0008808830121021964, + "loss": 0.5303, + "step": 4360 + }, + { + "epoch": 0.2436517026566472, + "grad_norm": 1.7465602159500122, + "learning_rate": 0.0008808549977588526, + "loss": 0.4875, + "step": 4361 + }, + { + "epoch": 0.24370757326032907, + "grad_norm": 0.4889052212238312, + "learning_rate": 0.0008808269834155087, + "loss": 0.4831, + "step": 4362 + }, + { + "epoch": 0.24376344386401094, + "grad_norm": 0.5024933218955994, + "learning_rate": 0.000880798969072165, + "loss": 0.3691, + "step": 4363 + }, + { + "epoch": 0.2438193144676928, + "grad_norm": 0.83844393491745, + "learning_rate": 0.0008807709547288212, + "loss": 0.4319, + "step": 4364 + }, + { + "epoch": 0.2438751850713747, + "grad_norm": 0.47784411907196045, + "learning_rate": 0.0008807429403854774, + "loss": 0.4202, + "step": 4365 + }, + { + "epoch": 0.24393105567505657, + "grad_norm": 0.675568163394928, + "learning_rate": 0.0008807149260421336, + "loss": 0.4694, + "step": 4366 + }, + { + "epoch": 0.24398692627873844, + "grad_norm": 1.1086933612823486, + "learning_rate": 0.0008806869116987898, + "loss": 0.459, + "step": 4367 + }, + { + "epoch": 0.2440427968824203, + "grad_norm": 0.4460262656211853, + "learning_rate": 0.000880658897355446, + "loss": 0.4231, + "step": 4368 + }, + { + "epoch": 0.24409866748610218, + "grad_norm": 1.2327508926391602, + "learning_rate": 0.0008806308830121022, + "loss": 0.6164, + "step": 4369 + }, + { + "epoch": 0.24415453808978407, + "grad_norm": 0.9984464049339294, + "learning_rate": 0.0008806028686687584, + "loss": 0.5194, + "step": 4370 + }, + { + "epoch": 0.24421040869346594, + "grad_norm": 0.9557931423187256, + "learning_rate": 0.0008805748543254146, + "loss": 0.4281, + "step": 4371 + }, + { + "epoch": 0.2442662792971478, + "grad_norm": 0.4948663115501404, + "learning_rate": 0.0008805468399820708, + "loss": 0.4754, + "step": 4372 + }, + { + "epoch": 0.24432214990082968, + "grad_norm": 0.5255029201507568, + "learning_rate": 0.0008805188256387271, + "loss": 0.4652, + "step": 4373 + }, + { + "epoch": 0.24437802050451155, + "grad_norm": 0.544701337814331, + "learning_rate": 0.0008804908112953832, + "loss": 0.4008, + "step": 4374 + }, + { + "epoch": 0.2444338911081934, + "grad_norm": 0.5394237041473389, + "learning_rate": 0.0008804627969520395, + "loss": 0.3521, + "step": 4375 + }, + { + "epoch": 0.2444897617118753, + "grad_norm": 0.5174205303192139, + "learning_rate": 0.0008804347826086957, + "loss": 0.4135, + "step": 4376 + }, + { + "epoch": 0.24454563231555718, + "grad_norm": 1.8061649799346924, + "learning_rate": 0.000880406768265352, + "loss": 0.5338, + "step": 4377 + }, + { + "epoch": 0.24460150291923904, + "grad_norm": 0.38308894634246826, + "learning_rate": 0.0008803787539220081, + "loss": 0.3753, + "step": 4378 + }, + { + "epoch": 0.2446573735229209, + "grad_norm": 0.39739081263542175, + "learning_rate": 0.0008803507395786643, + "loss": 0.4588, + "step": 4379 + }, + { + "epoch": 0.24471324412660278, + "grad_norm": 0.47891682386398315, + "learning_rate": 0.0008803227252353205, + "loss": 0.4744, + "step": 4380 + }, + { + "epoch": 0.24476911473028465, + "grad_norm": 0.6469098329544067, + "learning_rate": 0.0008802947108919767, + "loss": 0.5073, + "step": 4381 + }, + { + "epoch": 0.24482498533396654, + "grad_norm": 0.6632043123245239, + "learning_rate": 0.0008802666965486329, + "loss": 0.352, + "step": 4382 + }, + { + "epoch": 0.2448808559376484, + "grad_norm": 0.8871246576309204, + "learning_rate": 0.0008802386822052891, + "loss": 0.4559, + "step": 4383 + }, + { + "epoch": 0.24493672654133028, + "grad_norm": 0.8177871108055115, + "learning_rate": 0.0008802106678619454, + "loss": 0.4266, + "step": 4384 + }, + { + "epoch": 0.24499259714501215, + "grad_norm": 0.6537104249000549, + "learning_rate": 0.0008801826535186015, + "loss": 0.4509, + "step": 4385 + }, + { + "epoch": 0.24504846774869402, + "grad_norm": 0.7549874186515808, + "learning_rate": 0.0008801546391752577, + "loss": 0.4733, + "step": 4386 + }, + { + "epoch": 0.2451043383523759, + "grad_norm": 1.2187671661376953, + "learning_rate": 0.0008801266248319139, + "loss": 0.4903, + "step": 4387 + }, + { + "epoch": 0.24516020895605778, + "grad_norm": 0.6959953904151917, + "learning_rate": 0.0008800986104885702, + "loss": 0.5222, + "step": 4388 + }, + { + "epoch": 0.24521607955973965, + "grad_norm": 0.6470445394515991, + "learning_rate": 0.0008800705961452263, + "loss": 0.4538, + "step": 4389 + }, + { + "epoch": 0.24527195016342151, + "grad_norm": 0.5597547292709351, + "learning_rate": 0.0008800425818018825, + "loss": 0.4365, + "step": 4390 + }, + { + "epoch": 0.24532782076710338, + "grad_norm": 7.383012294769287, + "learning_rate": 0.0008800145674585389, + "loss": 0.4478, + "step": 4391 + }, + { + "epoch": 0.24538369137078525, + "grad_norm": 0.7303881049156189, + "learning_rate": 0.0008799865531151951, + "loss": 0.4562, + "step": 4392 + }, + { + "epoch": 0.24543956197446715, + "grad_norm": 0.8856653571128845, + "learning_rate": 0.0008799585387718512, + "loss": 0.4498, + "step": 4393 + }, + { + "epoch": 0.24549543257814901, + "grad_norm": 0.5023199915885925, + "learning_rate": 0.0008799305244285074, + "loss": 0.4081, + "step": 4394 + }, + { + "epoch": 0.24555130318183088, + "grad_norm": 0.7843542098999023, + "learning_rate": 0.0008799025100851637, + "loss": 0.472, + "step": 4395 + }, + { + "epoch": 0.24560717378551275, + "grad_norm": 0.7630894184112549, + "learning_rate": 0.0008798744957418199, + "loss": 0.6061, + "step": 4396 + }, + { + "epoch": 0.24566304438919462, + "grad_norm": 0.7754666209220886, + "learning_rate": 0.000879846481398476, + "loss": 0.4514, + "step": 4397 + }, + { + "epoch": 0.24571891499287649, + "grad_norm": 13.255594253540039, + "learning_rate": 0.0008798184670551323, + "loss": 0.5371, + "step": 4398 + }, + { + "epoch": 0.24577478559655838, + "grad_norm": 0.8719791173934937, + "learning_rate": 0.0008797904527117885, + "loss": 0.5316, + "step": 4399 + }, + { + "epoch": 0.24583065620024025, + "grad_norm": 0.929517388343811, + "learning_rate": 0.0008797624383684447, + "loss": 0.5881, + "step": 4400 + }, + { + "epoch": 0.24588652680392212, + "grad_norm": 0.522274374961853, + "learning_rate": 0.0008797344240251008, + "loss": 0.4284, + "step": 4401 + }, + { + "epoch": 0.24594239740760399, + "grad_norm": 0.748342752456665, + "learning_rate": 0.0008797064096817571, + "loss": 0.4728, + "step": 4402 + }, + { + "epoch": 0.24599826801128585, + "grad_norm": 0.6738157272338867, + "learning_rate": 0.0008796783953384133, + "loss": 0.606, + "step": 4403 + }, + { + "epoch": 0.24605413861496772, + "grad_norm": 0.9681056141853333, + "learning_rate": 0.0008796503809950694, + "loss": 0.4504, + "step": 4404 + }, + { + "epoch": 0.24611000921864962, + "grad_norm": 0.9361034035682678, + "learning_rate": 0.0008796223666517257, + "loss": 0.641, + "step": 4405 + }, + { + "epoch": 0.24616587982233148, + "grad_norm": 0.6141452789306641, + "learning_rate": 0.0008795943523083819, + "loss": 0.4615, + "step": 4406 + }, + { + "epoch": 0.24622175042601335, + "grad_norm": 0.5437930226325989, + "learning_rate": 0.0008795663379650381, + "loss": 0.4002, + "step": 4407 + }, + { + "epoch": 0.24627762102969522, + "grad_norm": 0.6894605159759521, + "learning_rate": 0.0008795383236216942, + "loss": 0.4824, + "step": 4408 + }, + { + "epoch": 0.2463334916333771, + "grad_norm": 1.5491564273834229, + "learning_rate": 0.0008795103092783506, + "loss": 0.5504, + "step": 4409 + }, + { + "epoch": 0.24638936223705898, + "grad_norm": 0.9197080731391907, + "learning_rate": 0.0008794822949350068, + "loss": 0.4354, + "step": 4410 + }, + { + "epoch": 0.24644523284074085, + "grad_norm": 1.1580934524536133, + "learning_rate": 0.000879454280591663, + "loss": 0.505, + "step": 4411 + }, + { + "epoch": 0.24650110344442272, + "grad_norm": 1.779961347579956, + "learning_rate": 0.0008794262662483191, + "loss": 0.476, + "step": 4412 + }, + { + "epoch": 0.2465569740481046, + "grad_norm": 0.6172492504119873, + "learning_rate": 0.0008793982519049754, + "loss": 0.5103, + "step": 4413 + }, + { + "epoch": 0.24661284465178646, + "grad_norm": 0.6478009819984436, + "learning_rate": 0.0008793702375616316, + "loss": 0.4768, + "step": 4414 + }, + { + "epoch": 0.24666871525546832, + "grad_norm": 0.8619832396507263, + "learning_rate": 0.0008793422232182878, + "loss": 0.3989, + "step": 4415 + }, + { + "epoch": 0.24672458585915022, + "grad_norm": 0.62435382604599, + "learning_rate": 0.000879314208874944, + "loss": 0.5074, + "step": 4416 + }, + { + "epoch": 0.2467804564628321, + "grad_norm": 2.0136325359344482, + "learning_rate": 0.0008792861945316002, + "loss": 0.5698, + "step": 4417 + }, + { + "epoch": 0.24683632706651396, + "grad_norm": 0.9953793883323669, + "learning_rate": 0.0008792581801882564, + "loss": 0.6107, + "step": 4418 + }, + { + "epoch": 0.24689219767019582, + "grad_norm": 0.8158494234085083, + "learning_rate": 0.0008792301658449126, + "loss": 0.6068, + "step": 4419 + }, + { + "epoch": 0.2469480682738777, + "grad_norm": 0.5526538491249084, + "learning_rate": 0.0008792021515015688, + "loss": 0.397, + "step": 4420 + }, + { + "epoch": 0.24700393887755956, + "grad_norm": 0.4353998899459839, + "learning_rate": 0.000879174137158225, + "loss": 0.4552, + "step": 4421 + }, + { + "epoch": 0.24705980948124145, + "grad_norm": 1.2353062629699707, + "learning_rate": 0.0008791461228148812, + "loss": 0.6104, + "step": 4422 + }, + { + "epoch": 0.24711568008492332, + "grad_norm": 0.5494514107704163, + "learning_rate": 0.0008791181084715374, + "loss": 0.5146, + "step": 4423 + }, + { + "epoch": 0.2471715506886052, + "grad_norm": 0.48015597462654114, + "learning_rate": 0.0008790900941281936, + "loss": 0.4385, + "step": 4424 + }, + { + "epoch": 0.24722742129228706, + "grad_norm": 0.5655499696731567, + "learning_rate": 0.0008790620797848498, + "loss": 0.4416, + "step": 4425 + }, + { + "epoch": 0.24728329189596893, + "grad_norm": 0.9782427549362183, + "learning_rate": 0.0008790340654415061, + "loss": 0.5935, + "step": 4426 + }, + { + "epoch": 0.24733916249965082, + "grad_norm": 0.6500933170318604, + "learning_rate": 0.0008790060510981623, + "loss": 0.5051, + "step": 4427 + }, + { + "epoch": 0.2473950331033327, + "grad_norm": 0.8659023642539978, + "learning_rate": 0.0008789780367548185, + "loss": 0.4954, + "step": 4428 + }, + { + "epoch": 0.24745090370701456, + "grad_norm": 0.5707963109016418, + "learning_rate": 0.0008789500224114747, + "loss": 0.4818, + "step": 4429 + }, + { + "epoch": 0.24750677431069643, + "grad_norm": 0.8789964318275452, + "learning_rate": 0.000878922008068131, + "loss": 0.5144, + "step": 4430 + }, + { + "epoch": 0.2475626449143783, + "grad_norm": 10.552491188049316, + "learning_rate": 0.0008788939937247871, + "loss": 0.5124, + "step": 4431 + }, + { + "epoch": 0.24761851551806016, + "grad_norm": 0.7904112935066223, + "learning_rate": 0.0008788659793814433, + "loss": 0.5941, + "step": 4432 + }, + { + "epoch": 0.24767438612174206, + "grad_norm": 0.6745386719703674, + "learning_rate": 0.0008788379650380995, + "loss": 0.6249, + "step": 4433 + }, + { + "epoch": 0.24773025672542393, + "grad_norm": 12.759353637695312, + "learning_rate": 0.0008788099506947558, + "loss": 0.4478, + "step": 4434 + }, + { + "epoch": 0.2477861273291058, + "grad_norm": 0.9143531322479248, + "learning_rate": 0.0008787819363514119, + "loss": 0.4156, + "step": 4435 + }, + { + "epoch": 0.24784199793278766, + "grad_norm": 0.8839870691299438, + "learning_rate": 0.0008787539220080681, + "loss": 0.518, + "step": 4436 + }, + { + "epoch": 0.24789786853646953, + "grad_norm": 0.5257184505462646, + "learning_rate": 0.0008787259076647244, + "loss": 0.5826, + "step": 4437 + }, + { + "epoch": 0.2479537391401514, + "grad_norm": 0.5187451839447021, + "learning_rate": 0.0008786978933213806, + "loss": 0.4556, + "step": 4438 + }, + { + "epoch": 0.2480096097438333, + "grad_norm": 0.4774661064147949, + "learning_rate": 0.0008786698789780367, + "loss": 0.3976, + "step": 4439 + }, + { + "epoch": 0.24806548034751516, + "grad_norm": 0.6748642921447754, + "learning_rate": 0.0008786418646346929, + "loss": 0.543, + "step": 4440 + }, + { + "epoch": 0.24812135095119703, + "grad_norm": 0.7633100152015686, + "learning_rate": 0.0008786138502913493, + "loss": 0.3895, + "step": 4441 + }, + { + "epoch": 0.2481772215548789, + "grad_norm": 0.5409467220306396, + "learning_rate": 0.0008785858359480055, + "loss": 0.5216, + "step": 4442 + }, + { + "epoch": 0.24823309215856076, + "grad_norm": 0.6531580686569214, + "learning_rate": 0.0008785578216046616, + "loss": 0.3912, + "step": 4443 + }, + { + "epoch": 0.24828896276224266, + "grad_norm": 0.7026460766792297, + "learning_rate": 0.0008785298072613179, + "loss": 0.5423, + "step": 4444 + }, + { + "epoch": 0.24834483336592453, + "grad_norm": 0.9667498469352722, + "learning_rate": 0.0008785017929179741, + "loss": 0.4168, + "step": 4445 + }, + { + "epoch": 0.2484007039696064, + "grad_norm": 0.6347891092300415, + "learning_rate": 0.0008784737785746302, + "loss": 0.4586, + "step": 4446 + }, + { + "epoch": 0.24845657457328826, + "grad_norm": 4.224841594696045, + "learning_rate": 0.0008784457642312864, + "loss": 0.5823, + "step": 4447 + }, + { + "epoch": 0.24851244517697013, + "grad_norm": 0.48881351947784424, + "learning_rate": 0.0008784177498879427, + "loss": 0.4765, + "step": 4448 + }, + { + "epoch": 0.248568315780652, + "grad_norm": 0.5234783887863159, + "learning_rate": 0.0008783897355445989, + "loss": 0.4847, + "step": 4449 + }, + { + "epoch": 0.2486241863843339, + "grad_norm": 0.7555892467498779, + "learning_rate": 0.000878361721201255, + "loss": 0.5022, + "step": 4450 + }, + { + "epoch": 0.24868005698801576, + "grad_norm": 0.6011260747909546, + "learning_rate": 0.0008783337068579112, + "loss": 0.746, + "step": 4451 + }, + { + "epoch": 0.24873592759169763, + "grad_norm": 0.8352910876274109, + "learning_rate": 0.0008783056925145675, + "loss": 0.3498, + "step": 4452 + }, + { + "epoch": 0.2487917981953795, + "grad_norm": 0.9082228541374207, + "learning_rate": 0.0008782776781712237, + "loss": 0.4819, + "step": 4453 + }, + { + "epoch": 0.24884766879906137, + "grad_norm": 0.5091032385826111, + "learning_rate": 0.0008782496638278798, + "loss": 0.4599, + "step": 4454 + }, + { + "epoch": 0.24890353940274323, + "grad_norm": 2.742262363433838, + "learning_rate": 0.0008782216494845361, + "loss": 0.4424, + "step": 4455 + }, + { + "epoch": 0.24895941000642513, + "grad_norm": 0.739380419254303, + "learning_rate": 0.0008781936351411923, + "loss": 0.5537, + "step": 4456 + }, + { + "epoch": 0.249015280610107, + "grad_norm": 0.5859549641609192, + "learning_rate": 0.0008781656207978485, + "loss": 0.4977, + "step": 4457 + }, + { + "epoch": 0.24907115121378887, + "grad_norm": 5.564648151397705, + "learning_rate": 0.0008781376064545046, + "loss": 0.4298, + "step": 4458 + }, + { + "epoch": 0.24912702181747073, + "grad_norm": 0.3788720667362213, + "learning_rate": 0.000878109592111161, + "loss": 0.4501, + "step": 4459 + }, + { + "epoch": 0.2491828924211526, + "grad_norm": 4.4596123695373535, + "learning_rate": 0.0008780815777678172, + "loss": 0.363, + "step": 4460 + }, + { + "epoch": 0.24923876302483447, + "grad_norm": 0.44355159997940063, + "learning_rate": 0.0008780535634244734, + "loss": 0.4286, + "step": 4461 + }, + { + "epoch": 0.24929463362851637, + "grad_norm": 0.5725240707397461, + "learning_rate": 0.0008780255490811296, + "loss": 0.5031, + "step": 4462 + }, + { + "epoch": 0.24935050423219823, + "grad_norm": 0.5631911754608154, + "learning_rate": 0.0008779975347377858, + "loss": 0.4337, + "step": 4463 + }, + { + "epoch": 0.2494063748358801, + "grad_norm": 0.6471314430236816, + "learning_rate": 0.000877969520394442, + "loss": 0.5142, + "step": 4464 + }, + { + "epoch": 0.24946224543956197, + "grad_norm": 0.5749391913414001, + "learning_rate": 0.0008779415060510981, + "loss": 0.541, + "step": 4465 + }, + { + "epoch": 0.24951811604324384, + "grad_norm": 0.5896140336990356, + "learning_rate": 0.0008779134917077544, + "loss": 0.4219, + "step": 4466 + }, + { + "epoch": 0.24957398664692573, + "grad_norm": 0.5436257719993591, + "learning_rate": 0.0008778854773644106, + "loss": 0.5867, + "step": 4467 + }, + { + "epoch": 0.2496298572506076, + "grad_norm": 1.5994360446929932, + "learning_rate": 0.0008778574630210668, + "loss": 0.5204, + "step": 4468 + }, + { + "epoch": 0.24968572785428947, + "grad_norm": 0.5834673643112183, + "learning_rate": 0.000877829448677723, + "loss": 0.562, + "step": 4469 + }, + { + "epoch": 0.24974159845797134, + "grad_norm": 0.4802553653717041, + "learning_rate": 0.0008778014343343792, + "loss": 0.4713, + "step": 4470 + }, + { + "epoch": 0.2497974690616532, + "grad_norm": 0.5403347611427307, + "learning_rate": 0.0008777734199910354, + "loss": 0.5132, + "step": 4471 + }, + { + "epoch": 0.24985333966533507, + "grad_norm": 0.4536905288696289, + "learning_rate": 0.0008777454056476916, + "loss": 0.4281, + "step": 4472 + }, + { + "epoch": 0.24990921026901697, + "grad_norm": 0.5973700881004333, + "learning_rate": 0.0008777173913043478, + "loss": 0.3653, + "step": 4473 + }, + { + "epoch": 0.24996508087269884, + "grad_norm": 0.5336066484451294, + "learning_rate": 0.000877689376961004, + "loss": 0.4163, + "step": 4474 + }, + { + "epoch": 0.2500209514763807, + "grad_norm": 0.5147836804389954, + "learning_rate": 0.0008776613626176602, + "loss": 0.4261, + "step": 4475 + }, + { + "epoch": 0.2500768220800626, + "grad_norm": 1.6578412055969238, + "learning_rate": 0.0008776333482743166, + "loss": 0.3785, + "step": 4476 + }, + { + "epoch": 0.25013269268374444, + "grad_norm": 0.6089770197868347, + "learning_rate": 0.0008776053339309727, + "loss": 0.4475, + "step": 4477 + }, + { + "epoch": 0.25018856328742634, + "grad_norm": 0.5108608603477478, + "learning_rate": 0.0008775773195876289, + "loss": 0.5309, + "step": 4478 + }, + { + "epoch": 0.2502444338911082, + "grad_norm": 0.5055390000343323, + "learning_rate": 0.0008775493052442851, + "loss": 0.52, + "step": 4479 + }, + { + "epoch": 0.25030030449479007, + "grad_norm": 0.8839969635009766, + "learning_rate": 0.0008775212909009414, + "loss": 0.4931, + "step": 4480 + }, + { + "epoch": 0.2503561750984719, + "grad_norm": 0.7981281876564026, + "learning_rate": 0.0008774932765575975, + "loss": 0.4344, + "step": 4481 + }, + { + "epoch": 0.2504120457021538, + "grad_norm": 0.7508647441864014, + "learning_rate": 0.0008774652622142537, + "loss": 0.5358, + "step": 4482 + }, + { + "epoch": 0.2504679163058357, + "grad_norm": 0.5672193169593811, + "learning_rate": 0.0008774372478709099, + "loss": 0.4817, + "step": 4483 + }, + { + "epoch": 0.25052378690951754, + "grad_norm": 0.5385472178459167, + "learning_rate": 0.0008774092335275662, + "loss": 0.446, + "step": 4484 + }, + { + "epoch": 0.25057965751319944, + "grad_norm": 0.5659594535827637, + "learning_rate": 0.0008773812191842223, + "loss": 0.4755, + "step": 4485 + }, + { + "epoch": 0.2506355281168813, + "grad_norm": 0.45119574666023254, + "learning_rate": 0.0008773532048408785, + "loss": 0.4698, + "step": 4486 + }, + { + "epoch": 0.2506913987205632, + "grad_norm": 0.5960628986358643, + "learning_rate": 0.0008773251904975348, + "loss": 0.4627, + "step": 4487 + }, + { + "epoch": 0.25074726932424507, + "grad_norm": 0.45046505331993103, + "learning_rate": 0.0008772971761541909, + "loss": 0.4513, + "step": 4488 + }, + { + "epoch": 0.2508031399279269, + "grad_norm": 1.1369861364364624, + "learning_rate": 0.0008772691618108471, + "loss": 0.4515, + "step": 4489 + }, + { + "epoch": 0.2508590105316088, + "grad_norm": 1.485497236251831, + "learning_rate": 0.0008772411474675033, + "loss": 0.5203, + "step": 4490 + }, + { + "epoch": 0.25091488113529065, + "grad_norm": 0.7478820085525513, + "learning_rate": 0.0008772131331241597, + "loss": 0.5027, + "step": 4491 + }, + { + "epoch": 0.25097075173897254, + "grad_norm": 0.5810976624488831, + "learning_rate": 0.0008771851187808157, + "loss": 0.3932, + "step": 4492 + }, + { + "epoch": 0.25102662234265444, + "grad_norm": 0.44721418619155884, + "learning_rate": 0.000877157104437472, + "loss": 0.4689, + "step": 4493 + }, + { + "epoch": 0.2510824929463363, + "grad_norm": 0.5779846906661987, + "learning_rate": 0.0008771290900941283, + "loss": 0.5017, + "step": 4494 + }, + { + "epoch": 0.2511383635500182, + "grad_norm": 0.8801562786102295, + "learning_rate": 0.0008771010757507845, + "loss": 0.5571, + "step": 4495 + }, + { + "epoch": 0.2511942341537, + "grad_norm": 0.44875892996788025, + "learning_rate": 0.0008770730614074406, + "loss": 0.4804, + "step": 4496 + }, + { + "epoch": 0.2512501047573819, + "grad_norm": 0.7611216902732849, + "learning_rate": 0.0008770450470640968, + "loss": 0.4898, + "step": 4497 + }, + { + "epoch": 0.25130597536106375, + "grad_norm": 0.6831035017967224, + "learning_rate": 0.0008770170327207531, + "loss": 0.5158, + "step": 4498 + }, + { + "epoch": 0.25136184596474564, + "grad_norm": 0.6584053039550781, + "learning_rate": 0.0008769890183774093, + "loss": 0.6353, + "step": 4499 + }, + { + "epoch": 0.25141771656842754, + "grad_norm": 0.8408846855163574, + "learning_rate": 0.0008769610040340654, + "loss": 0.6497, + "step": 4500 + }, + { + "epoch": 0.25141771656842754, + "eval_cer": 0.09585610945631894, + "eval_loss": 0.3616579473018646, + "eval_runtime": 56.0752, + "eval_samples_per_second": 80.927, + "eval_steps_per_second": 5.065, + "eval_wer": 0.37406776347029413, + "step": 4500 + }, + { + "epoch": 0.2514735871721094, + "grad_norm": 0.5994569063186646, + "learning_rate": 0.0008769329896907217, + "loss": 0.4942, + "step": 4501 + }, + { + "epoch": 0.2515294577757913, + "grad_norm": 0.505544126033783, + "learning_rate": 0.0008769049753473779, + "loss": 0.4938, + "step": 4502 + }, + { + "epoch": 0.2515853283794731, + "grad_norm": 1.2937158346176147, + "learning_rate": 0.0008768769610040341, + "loss": 0.4918, + "step": 4503 + }, + { + "epoch": 0.251641198983155, + "grad_norm": 0.8728862404823303, + "learning_rate": 0.0008768489466606902, + "loss": 0.4758, + "step": 4504 + }, + { + "epoch": 0.2516970695868369, + "grad_norm": 0.6806188821792603, + "learning_rate": 0.0008768209323173465, + "loss": 0.5381, + "step": 4505 + }, + { + "epoch": 0.25175294019051875, + "grad_norm": 1.5402776002883911, + "learning_rate": 0.0008767929179740027, + "loss": 0.549, + "step": 4506 + }, + { + "epoch": 0.25180881079420064, + "grad_norm": 0.417765736579895, + "learning_rate": 0.0008767649036306588, + "loss": 0.3705, + "step": 4507 + }, + { + "epoch": 0.2518646813978825, + "grad_norm": 0.7775244116783142, + "learning_rate": 0.0008767368892873151, + "loss": 0.4861, + "step": 4508 + }, + { + "epoch": 0.2519205520015644, + "grad_norm": 0.7343034148216248, + "learning_rate": 0.0008767088749439714, + "loss": 0.4971, + "step": 4509 + }, + { + "epoch": 0.2519764226052463, + "grad_norm": 1.1027699708938599, + "learning_rate": 0.0008766808606006276, + "loss": 0.496, + "step": 4510 + }, + { + "epoch": 0.2520322932089281, + "grad_norm": 0.49558597803115845, + "learning_rate": 0.0008766528462572837, + "loss": 0.4903, + "step": 4511 + }, + { + "epoch": 0.25208816381261, + "grad_norm": 1.2529432773590088, + "learning_rate": 0.00087662483191394, + "loss": 0.4359, + "step": 4512 + }, + { + "epoch": 0.25214403441629185, + "grad_norm": 0.6917349100112915, + "learning_rate": 0.0008765968175705962, + "loss": 0.4824, + "step": 4513 + }, + { + "epoch": 0.25219990501997375, + "grad_norm": 0.7514471411705017, + "learning_rate": 0.0008765688032272524, + "loss": 0.601, + "step": 4514 + }, + { + "epoch": 0.2522557756236556, + "grad_norm": 0.5384849905967712, + "learning_rate": 0.0008765407888839085, + "loss": 0.4774, + "step": 4515 + }, + { + "epoch": 0.2523116462273375, + "grad_norm": 0.7101408839225769, + "learning_rate": 0.0008765127745405648, + "loss": 0.5229, + "step": 4516 + }, + { + "epoch": 0.2523675168310194, + "grad_norm": 0.45907604694366455, + "learning_rate": 0.000876484760197221, + "loss": 0.406, + "step": 4517 + }, + { + "epoch": 0.2524233874347012, + "grad_norm": 0.43495306372642517, + "learning_rate": 0.0008764567458538772, + "loss": 0.4825, + "step": 4518 + }, + { + "epoch": 0.2524792580383831, + "grad_norm": 0.5003460049629211, + "learning_rate": 0.0008764287315105334, + "loss": 0.4878, + "step": 4519 + }, + { + "epoch": 0.25253512864206495, + "grad_norm": 0.5842649936676025, + "learning_rate": 0.0008764007171671896, + "loss": 0.3893, + "step": 4520 + }, + { + "epoch": 0.25259099924574685, + "grad_norm": 1.872408151626587, + "learning_rate": 0.0008763727028238458, + "loss": 0.5805, + "step": 4521 + }, + { + "epoch": 0.25264686984942875, + "grad_norm": 0.4570554494857788, + "learning_rate": 0.000876344688480502, + "loss": 0.4873, + "step": 4522 + }, + { + "epoch": 0.2527027404531106, + "grad_norm": 1.6559475660324097, + "learning_rate": 0.0008763166741371582, + "loss": 0.476, + "step": 4523 + }, + { + "epoch": 0.2527586110567925, + "grad_norm": 0.5000524520874023, + "learning_rate": 0.0008762886597938144, + "loss": 0.5069, + "step": 4524 + }, + { + "epoch": 0.2528144816604743, + "grad_norm": 1.6677573919296265, + "learning_rate": 0.0008762606454504706, + "loss": 0.6029, + "step": 4525 + }, + { + "epoch": 0.2528703522641562, + "grad_norm": 3.6717915534973145, + "learning_rate": 0.000876232631107127, + "loss": 0.39, + "step": 4526 + }, + { + "epoch": 0.2529262228678381, + "grad_norm": 0.5246142745018005, + "learning_rate": 0.0008762046167637831, + "loss": 0.406, + "step": 4527 + }, + { + "epoch": 0.25298209347151995, + "grad_norm": 0.7614737153053284, + "learning_rate": 0.0008761766024204393, + "loss": 0.5282, + "step": 4528 + }, + { + "epoch": 0.25303796407520185, + "grad_norm": 0.5161161422729492, + "learning_rate": 0.0008761485880770955, + "loss": 0.4439, + "step": 4529 + }, + { + "epoch": 0.2530938346788837, + "grad_norm": 0.8924152255058289, + "learning_rate": 0.0008761205737337517, + "loss": 0.4959, + "step": 4530 + }, + { + "epoch": 0.2531497052825656, + "grad_norm": 0.85765540599823, + "learning_rate": 0.0008760925593904079, + "loss": 0.6436, + "step": 4531 + }, + { + "epoch": 0.2532055758862474, + "grad_norm": 0.5701043605804443, + "learning_rate": 0.0008760645450470641, + "loss": 0.5266, + "step": 4532 + }, + { + "epoch": 0.2532614464899293, + "grad_norm": 0.7890478372573853, + "learning_rate": 0.0008760365307037204, + "loss": 0.5525, + "step": 4533 + }, + { + "epoch": 0.2533173170936112, + "grad_norm": 0.46904614567756653, + "learning_rate": 0.0008760085163603765, + "loss": 0.4356, + "step": 4534 + }, + { + "epoch": 0.25337318769729306, + "grad_norm": 0.5806142687797546, + "learning_rate": 0.0008759805020170327, + "loss": 0.5936, + "step": 4535 + }, + { + "epoch": 0.25342905830097495, + "grad_norm": 0.6894263029098511, + "learning_rate": 0.0008759524876736889, + "loss": 0.5002, + "step": 4536 + }, + { + "epoch": 0.2534849289046568, + "grad_norm": 0.574905276298523, + "learning_rate": 0.0008759244733303452, + "loss": 0.4702, + "step": 4537 + }, + { + "epoch": 0.2535407995083387, + "grad_norm": 0.6272011399269104, + "learning_rate": 0.0008758964589870013, + "loss": 0.5566, + "step": 4538 + }, + { + "epoch": 0.2535966701120206, + "grad_norm": 0.702812671661377, + "learning_rate": 0.0008758684446436575, + "loss": 0.5068, + "step": 4539 + }, + { + "epoch": 0.2536525407157024, + "grad_norm": 11.420536041259766, + "learning_rate": 0.0008758404303003138, + "loss": 0.4162, + "step": 4540 + }, + { + "epoch": 0.2537084113193843, + "grad_norm": 1.8507262468338013, + "learning_rate": 0.00087581241595697, + "loss": 0.5042, + "step": 4541 + }, + { + "epoch": 0.25376428192306616, + "grad_norm": 0.4640161991119385, + "learning_rate": 0.0008757844016136261, + "loss": 0.3937, + "step": 4542 + }, + { + "epoch": 0.25382015252674806, + "grad_norm": 0.6007736325263977, + "learning_rate": 0.0008757563872702824, + "loss": 0.5374, + "step": 4543 + }, + { + "epoch": 0.25387602313042995, + "grad_norm": 0.5580400228500366, + "learning_rate": 0.0008757283729269387, + "loss": 0.5221, + "step": 4544 + }, + { + "epoch": 0.2539318937341118, + "grad_norm": 3.3067331314086914, + "learning_rate": 0.0008757003585835949, + "loss": 0.4331, + "step": 4545 + }, + { + "epoch": 0.2539877643377937, + "grad_norm": 1.3475291728973389, + "learning_rate": 0.000875672344240251, + "loss": 0.6087, + "step": 4546 + }, + { + "epoch": 0.2540436349414755, + "grad_norm": 0.5574381351470947, + "learning_rate": 0.0008756443298969073, + "loss": 0.3528, + "step": 4547 + }, + { + "epoch": 0.2540995055451574, + "grad_norm": 0.7941306829452515, + "learning_rate": 0.0008756163155535635, + "loss": 0.5782, + "step": 4548 + }, + { + "epoch": 0.25415537614883926, + "grad_norm": 1.2263448238372803, + "learning_rate": 0.0008755883012102196, + "loss": 0.5807, + "step": 4549 + }, + { + "epoch": 0.25421124675252116, + "grad_norm": 0.651995837688446, + "learning_rate": 0.0008755602868668758, + "loss": 0.641, + "step": 4550 + }, + { + "epoch": 0.25426711735620305, + "grad_norm": 2.3254761695861816, + "learning_rate": 0.0008755322725235321, + "loss": 0.4837, + "step": 4551 + }, + { + "epoch": 0.2543229879598849, + "grad_norm": 2.0723471641540527, + "learning_rate": 0.0008755042581801883, + "loss": 0.4458, + "step": 4552 + }, + { + "epoch": 0.2543788585635668, + "grad_norm": 0.9961845874786377, + "learning_rate": 0.0008754762438368444, + "loss": 0.6488, + "step": 4553 + }, + { + "epoch": 0.25443472916724863, + "grad_norm": 0.5777174830436707, + "learning_rate": 0.0008754482294935006, + "loss": 0.5664, + "step": 4554 + }, + { + "epoch": 0.2544905997709305, + "grad_norm": 0.942571759223938, + "learning_rate": 0.0008754202151501569, + "loss": 0.6211, + "step": 4555 + }, + { + "epoch": 0.2545464703746124, + "grad_norm": 0.7065373063087463, + "learning_rate": 0.0008753922008068131, + "loss": 0.7075, + "step": 4556 + }, + { + "epoch": 0.25460234097829426, + "grad_norm": 1.6006433963775635, + "learning_rate": 0.0008753641864634692, + "loss": 0.4591, + "step": 4557 + }, + { + "epoch": 0.25465821158197616, + "grad_norm": 0.8434430956840515, + "learning_rate": 0.0008753361721201255, + "loss": 0.536, + "step": 4558 + }, + { + "epoch": 0.254714082185658, + "grad_norm": 0.6806038618087769, + "learning_rate": 0.0008753081577767818, + "loss": 0.623, + "step": 4559 + }, + { + "epoch": 0.2547699527893399, + "grad_norm": 0.559968113899231, + "learning_rate": 0.000875280143433438, + "loss": 0.5411, + "step": 4560 + }, + { + "epoch": 0.2548258233930218, + "grad_norm": 0.8287050724029541, + "learning_rate": 0.0008752521290900941, + "loss": 0.4368, + "step": 4561 + }, + { + "epoch": 0.25488169399670363, + "grad_norm": 2.7951900959014893, + "learning_rate": 0.0008752241147467504, + "loss": 0.4953, + "step": 4562 + }, + { + "epoch": 0.2549375646003855, + "grad_norm": 0.6453946828842163, + "learning_rate": 0.0008751961004034066, + "loss": 0.4527, + "step": 4563 + }, + { + "epoch": 0.25499343520406736, + "grad_norm": 1.2333531379699707, + "learning_rate": 0.0008751680860600628, + "loss": 0.4786, + "step": 4564 + }, + { + "epoch": 0.25504930580774926, + "grad_norm": 0.5305621027946472, + "learning_rate": 0.000875140071716719, + "loss": 0.4023, + "step": 4565 + }, + { + "epoch": 0.2551051764114311, + "grad_norm": 1.563538670539856, + "learning_rate": 0.0008751120573733752, + "loss": 0.4351, + "step": 4566 + }, + { + "epoch": 0.255161047015113, + "grad_norm": 1.3014531135559082, + "learning_rate": 0.0008750840430300314, + "loss": 0.5256, + "step": 4567 + }, + { + "epoch": 0.2552169176187949, + "grad_norm": 0.6398177742958069, + "learning_rate": 0.0008750560286866876, + "loss": 0.3386, + "step": 4568 + }, + { + "epoch": 0.25527278822247673, + "grad_norm": 0.6654679775238037, + "learning_rate": 0.0008750280143433438, + "loss": 0.3688, + "step": 4569 + }, + { + "epoch": 0.2553286588261586, + "grad_norm": 0.5269203782081604, + "learning_rate": 0.000875, + "loss": 0.4623, + "step": 4570 + }, + { + "epoch": 0.25538452942984047, + "grad_norm": 0.4870298504829407, + "learning_rate": 0.0008749719856566562, + "loss": 0.6299, + "step": 4571 + }, + { + "epoch": 0.25544040003352236, + "grad_norm": 0.6083543300628662, + "learning_rate": 0.0008749439713133124, + "loss": 0.3995, + "step": 4572 + }, + { + "epoch": 0.25549627063720426, + "grad_norm": 6.351544380187988, + "learning_rate": 0.0008749159569699686, + "loss": 0.4018, + "step": 4573 + }, + { + "epoch": 0.2555521412408861, + "grad_norm": 0.8722611665725708, + "learning_rate": 0.0008748879426266248, + "loss": 0.484, + "step": 4574 + }, + { + "epoch": 0.255608011844568, + "grad_norm": 1.2385854721069336, + "learning_rate": 0.000874859928283281, + "loss": 0.5025, + "step": 4575 + }, + { + "epoch": 0.25566388244824984, + "grad_norm": 1.189807415008545, + "learning_rate": 0.0008748319139399373, + "loss": 0.4815, + "step": 4576 + }, + { + "epoch": 0.25571975305193173, + "grad_norm": 0.49584412574768066, + "learning_rate": 0.0008748038995965935, + "loss": 0.6122, + "step": 4577 + }, + { + "epoch": 0.2557756236556136, + "grad_norm": 0.6272901296615601, + "learning_rate": 0.0008747758852532497, + "loss": 0.3845, + "step": 4578 + }, + { + "epoch": 0.25583149425929547, + "grad_norm": 0.663088858127594, + "learning_rate": 0.000874747870909906, + "loss": 0.4682, + "step": 4579 + }, + { + "epoch": 0.25588736486297736, + "grad_norm": 8.649540901184082, + "learning_rate": 0.0008747198565665621, + "loss": 0.3964, + "step": 4580 + }, + { + "epoch": 0.2559432354666592, + "grad_norm": 0.9562984108924866, + "learning_rate": 0.0008746918422232183, + "loss": 0.4401, + "step": 4581 + }, + { + "epoch": 0.2559991060703411, + "grad_norm": 0.7663208246231079, + "learning_rate": 0.0008746638278798745, + "loss": 0.4579, + "step": 4582 + }, + { + "epoch": 0.25605497667402294, + "grad_norm": 0.5860781669616699, + "learning_rate": 0.0008746358135365308, + "loss": 0.4507, + "step": 4583 + }, + { + "epoch": 0.25611084727770483, + "grad_norm": 0.37219107151031494, + "learning_rate": 0.0008746077991931869, + "loss": 0.3811, + "step": 4584 + }, + { + "epoch": 0.25616671788138673, + "grad_norm": 0.9458240866661072, + "learning_rate": 0.0008745797848498431, + "loss": 0.4876, + "step": 4585 + }, + { + "epoch": 0.25622258848506857, + "grad_norm": 0.8911159634590149, + "learning_rate": 0.0008745517705064994, + "loss": 0.5867, + "step": 4586 + }, + { + "epoch": 0.25627845908875047, + "grad_norm": 0.5728241801261902, + "learning_rate": 0.0008745237561631556, + "loss": 0.4153, + "step": 4587 + }, + { + "epoch": 0.2563343296924323, + "grad_norm": 0.653093695640564, + "learning_rate": 0.0008744957418198117, + "loss": 0.4127, + "step": 4588 + }, + { + "epoch": 0.2563902002961142, + "grad_norm": 0.6609985828399658, + "learning_rate": 0.0008744677274764679, + "loss": 0.517, + "step": 4589 + }, + { + "epoch": 0.2564460708997961, + "grad_norm": 1.4050332307815552, + "learning_rate": 0.0008744397131331242, + "loss": 0.4228, + "step": 4590 + }, + { + "epoch": 0.25650194150347794, + "grad_norm": 0.619671106338501, + "learning_rate": 0.0008744116987897804, + "loss": 0.4517, + "step": 4591 + }, + { + "epoch": 0.25655781210715983, + "grad_norm": 0.7691707015037537, + "learning_rate": 0.0008743836844464365, + "loss": 0.4771, + "step": 4592 + }, + { + "epoch": 0.2566136827108417, + "grad_norm": 0.5567960143089294, + "learning_rate": 0.0008743556701030928, + "loss": 0.4949, + "step": 4593 + }, + { + "epoch": 0.25666955331452357, + "grad_norm": 0.6117062568664551, + "learning_rate": 0.0008743276557597491, + "loss": 0.3366, + "step": 4594 + }, + { + "epoch": 0.2567254239182054, + "grad_norm": 0.5037212371826172, + "learning_rate": 0.0008742996414164052, + "loss": 0.6315, + "step": 4595 + }, + { + "epoch": 0.2567812945218873, + "grad_norm": 0.8742517828941345, + "learning_rate": 0.0008742716270730614, + "loss": 0.5866, + "step": 4596 + }, + { + "epoch": 0.2568371651255692, + "grad_norm": 0.8511859178543091, + "learning_rate": 0.0008742436127297177, + "loss": 0.489, + "step": 4597 + }, + { + "epoch": 0.25689303572925104, + "grad_norm": 0.8782104253768921, + "learning_rate": 0.0008742155983863739, + "loss": 0.4945, + "step": 4598 + }, + { + "epoch": 0.25694890633293294, + "grad_norm": 0.7723953723907471, + "learning_rate": 0.00087418758404303, + "loss": 0.4752, + "step": 4599 + }, + { + "epoch": 0.2570047769366148, + "grad_norm": 0.4924076199531555, + "learning_rate": 0.0008741595696996862, + "loss": 0.4619, + "step": 4600 + }, + { + "epoch": 0.25706064754029667, + "grad_norm": 0.5952222347259521, + "learning_rate": 0.0008741315553563425, + "loss": 0.4664, + "step": 4601 + }, + { + "epoch": 0.25711651814397857, + "grad_norm": 2.254061460494995, + "learning_rate": 0.0008741035410129987, + "loss": 0.5193, + "step": 4602 + }, + { + "epoch": 0.2571723887476604, + "grad_norm": 0.9220318794250488, + "learning_rate": 0.0008740755266696548, + "loss": 0.5136, + "step": 4603 + }, + { + "epoch": 0.2572282593513423, + "grad_norm": 0.7268227338790894, + "learning_rate": 0.0008740475123263111, + "loss": 0.7325, + "step": 4604 + }, + { + "epoch": 0.25728412995502414, + "grad_norm": 4.9351325035095215, + "learning_rate": 0.0008740194979829673, + "loss": 0.4176, + "step": 4605 + }, + { + "epoch": 0.25734000055870604, + "grad_norm": 0.7900199890136719, + "learning_rate": 0.0008739914836396235, + "loss": 0.5157, + "step": 4606 + }, + { + "epoch": 0.25739587116238793, + "grad_norm": 0.47038888931274414, + "learning_rate": 0.0008739634692962796, + "loss": 0.4497, + "step": 4607 + }, + { + "epoch": 0.2574517417660698, + "grad_norm": 0.6443123817443848, + "learning_rate": 0.000873935454952936, + "loss": 0.5013, + "step": 4608 + }, + { + "epoch": 0.25750761236975167, + "grad_norm": 0.5772133469581604, + "learning_rate": 0.0008739074406095922, + "loss": 0.4939, + "step": 4609 + }, + { + "epoch": 0.2575634829734335, + "grad_norm": 1.0206618309020996, + "learning_rate": 0.0008738794262662484, + "loss": 0.4136, + "step": 4610 + }, + { + "epoch": 0.2576193535771154, + "grad_norm": 0.6612287759780884, + "learning_rate": 0.0008738514119229046, + "loss": 0.5213, + "step": 4611 + }, + { + "epoch": 0.25767522418079725, + "grad_norm": 0.6255256533622742, + "learning_rate": 0.0008738233975795608, + "loss": 0.4462, + "step": 4612 + }, + { + "epoch": 0.25773109478447914, + "grad_norm": 11.765560150146484, + "learning_rate": 0.000873795383236217, + "loss": 0.5325, + "step": 4613 + }, + { + "epoch": 0.25778696538816104, + "grad_norm": 0.44668611884117126, + "learning_rate": 0.0008737673688928731, + "loss": 0.3307, + "step": 4614 + }, + { + "epoch": 0.2578428359918429, + "grad_norm": 0.9124888777732849, + "learning_rate": 0.0008737393545495294, + "loss": 0.4851, + "step": 4615 + }, + { + "epoch": 0.2578987065955248, + "grad_norm": 0.6073693633079529, + "learning_rate": 0.0008737113402061856, + "loss": 0.4749, + "step": 4616 + }, + { + "epoch": 0.2579545771992066, + "grad_norm": 0.6579868197441101, + "learning_rate": 0.0008736833258628418, + "loss": 0.7047, + "step": 4617 + }, + { + "epoch": 0.2580104478028885, + "grad_norm": 2.5662033557891846, + "learning_rate": 0.000873655311519498, + "loss": 0.4678, + "step": 4618 + }, + { + "epoch": 0.2580663184065704, + "grad_norm": 0.523700475692749, + "learning_rate": 0.0008736272971761542, + "loss": 0.4373, + "step": 4619 + }, + { + "epoch": 0.25812218901025225, + "grad_norm": 0.6096883416175842, + "learning_rate": 0.0008735992828328104, + "loss": 0.4071, + "step": 4620 + }, + { + "epoch": 0.25817805961393414, + "grad_norm": 0.4396882951259613, + "learning_rate": 0.0008735712684894666, + "loss": 0.4174, + "step": 4621 + }, + { + "epoch": 0.258233930217616, + "grad_norm": 0.6588603258132935, + "learning_rate": 0.0008735432541461228, + "loss": 0.4936, + "step": 4622 + }, + { + "epoch": 0.2582898008212979, + "grad_norm": 0.5835273265838623, + "learning_rate": 0.000873515239802779, + "loss": 0.5852, + "step": 4623 + }, + { + "epoch": 0.2583456714249798, + "grad_norm": 1.4423692226409912, + "learning_rate": 0.0008734872254594352, + "loss": 0.4791, + "step": 4624 + }, + { + "epoch": 0.2584015420286616, + "grad_norm": 0.7043792009353638, + "learning_rate": 0.0008734592111160916, + "loss": 0.5147, + "step": 4625 + }, + { + "epoch": 0.2584574126323435, + "grad_norm": 0.9753476977348328, + "learning_rate": 0.0008734311967727477, + "loss": 0.483, + "step": 4626 + }, + { + "epoch": 0.25851328323602535, + "grad_norm": 3.1001148223876953, + "learning_rate": 0.0008734031824294039, + "loss": 0.5416, + "step": 4627 + }, + { + "epoch": 0.25856915383970724, + "grad_norm": 0.6619459390640259, + "learning_rate": 0.0008733751680860601, + "loss": 0.5611, + "step": 4628 + }, + { + "epoch": 0.2586250244433891, + "grad_norm": 0.4754568040370941, + "learning_rate": 0.0008733471537427164, + "loss": 0.4449, + "step": 4629 + }, + { + "epoch": 0.258680895047071, + "grad_norm": 0.5536420941352844, + "learning_rate": 0.0008733191393993725, + "loss": 0.4811, + "step": 4630 + }, + { + "epoch": 0.2587367656507529, + "grad_norm": 0.840639591217041, + "learning_rate": 0.0008732911250560287, + "loss": 0.4182, + "step": 4631 + }, + { + "epoch": 0.2587926362544347, + "grad_norm": 6.785843372344971, + "learning_rate": 0.0008732631107126849, + "loss": 0.4023, + "step": 4632 + }, + { + "epoch": 0.2588485068581166, + "grad_norm": 0.743794322013855, + "learning_rate": 0.0008732350963693412, + "loss": 0.5161, + "step": 4633 + }, + { + "epoch": 0.25890437746179845, + "grad_norm": 3.6816630363464355, + "learning_rate": 0.0008732070820259973, + "loss": 0.5111, + "step": 4634 + }, + { + "epoch": 0.25896024806548035, + "grad_norm": 0.6175789833068848, + "learning_rate": 0.0008731790676826535, + "loss": 0.5491, + "step": 4635 + }, + { + "epoch": 0.25901611866916224, + "grad_norm": 0.7265775799751282, + "learning_rate": 0.0008731510533393098, + "loss": 0.5877, + "step": 4636 + }, + { + "epoch": 0.2590719892728441, + "grad_norm": 0.4267091453075409, + "learning_rate": 0.0008731230389959659, + "loss": 0.4575, + "step": 4637 + }, + { + "epoch": 0.259127859876526, + "grad_norm": 0.737727701663971, + "learning_rate": 0.0008730950246526221, + "loss": 0.5375, + "step": 4638 + }, + { + "epoch": 0.2591837304802078, + "grad_norm": 0.47817155718803406, + "learning_rate": 0.0008730670103092783, + "loss": 0.436, + "step": 4639 + }, + { + "epoch": 0.2592396010838897, + "grad_norm": 0.6717612743377686, + "learning_rate": 0.0008730389959659346, + "loss": 0.4136, + "step": 4640 + }, + { + "epoch": 0.2592954716875716, + "grad_norm": 0.5808565020561218, + "learning_rate": 0.0008730109816225907, + "loss": 0.4053, + "step": 4641 + }, + { + "epoch": 0.25935134229125345, + "grad_norm": 0.5759374499320984, + "learning_rate": 0.0008729829672792469, + "loss": 0.3375, + "step": 4642 + }, + { + "epoch": 0.25940721289493535, + "grad_norm": 0.888604998588562, + "learning_rate": 0.0008729549529359033, + "loss": 0.4881, + "step": 4643 + }, + { + "epoch": 0.2594630834986172, + "grad_norm": 0.5695391893386841, + "learning_rate": 0.0008729269385925595, + "loss": 0.4158, + "step": 4644 + }, + { + "epoch": 0.2595189541022991, + "grad_norm": 0.45352569222450256, + "learning_rate": 0.0008728989242492156, + "loss": 0.4799, + "step": 4645 + }, + { + "epoch": 0.2595748247059809, + "grad_norm": 0.4908115863800049, + "learning_rate": 0.0008728709099058718, + "loss": 0.4349, + "step": 4646 + }, + { + "epoch": 0.2596306953096628, + "grad_norm": 1.162288784980774, + "learning_rate": 0.0008728428955625281, + "loss": 0.5233, + "step": 4647 + }, + { + "epoch": 0.2596865659133447, + "grad_norm": 0.6949254274368286, + "learning_rate": 0.0008728148812191843, + "loss": 0.4356, + "step": 4648 + }, + { + "epoch": 0.25974243651702655, + "grad_norm": 0.8284937739372253, + "learning_rate": 0.0008727868668758404, + "loss": 0.5251, + "step": 4649 + }, + { + "epoch": 0.25979830712070845, + "grad_norm": 0.720591127872467, + "learning_rate": 0.0008727588525324967, + "loss": 0.5179, + "step": 4650 + }, + { + "epoch": 0.2598541777243903, + "grad_norm": 0.5841206908226013, + "learning_rate": 0.0008727308381891529, + "loss": 0.6201, + "step": 4651 + }, + { + "epoch": 0.2599100483280722, + "grad_norm": 0.46547409892082214, + "learning_rate": 0.0008727028238458091, + "loss": 0.355, + "step": 4652 + }, + { + "epoch": 0.2599659189317541, + "grad_norm": 0.6249770522117615, + "learning_rate": 0.0008726748095024652, + "loss": 0.4397, + "step": 4653 + }, + { + "epoch": 0.2600217895354359, + "grad_norm": 0.7630364298820496, + "learning_rate": 0.0008726467951591215, + "loss": 0.4419, + "step": 4654 + }, + { + "epoch": 0.2600776601391178, + "grad_norm": 0.7235999703407288, + "learning_rate": 0.0008726187808157777, + "loss": 0.4727, + "step": 4655 + }, + { + "epoch": 0.26013353074279966, + "grad_norm": 0.5611547231674194, + "learning_rate": 0.0008725907664724338, + "loss": 0.6366, + "step": 4656 + }, + { + "epoch": 0.26018940134648155, + "grad_norm": 0.7370139956474304, + "learning_rate": 0.0008725627521290901, + "loss": 0.4217, + "step": 4657 + }, + { + "epoch": 0.26024527195016345, + "grad_norm": 1.095665454864502, + "learning_rate": 0.0008725347377857463, + "loss": 0.601, + "step": 4658 + }, + { + "epoch": 0.2603011425538453, + "grad_norm": 0.6648024916648865, + "learning_rate": 0.0008725067234424026, + "loss": 0.4594, + "step": 4659 + }, + { + "epoch": 0.2603570131575272, + "grad_norm": 0.40195682644844055, + "learning_rate": 0.0008724787090990587, + "loss": 0.3517, + "step": 4660 + }, + { + "epoch": 0.260412883761209, + "grad_norm": 0.5704547166824341, + "learning_rate": 0.000872450694755715, + "loss": 0.473, + "step": 4661 + }, + { + "epoch": 0.2604687543648909, + "grad_norm": 0.7281366586685181, + "learning_rate": 0.0008724226804123712, + "loss": 0.4425, + "step": 4662 + }, + { + "epoch": 0.26052462496857276, + "grad_norm": 0.4010917544364929, + "learning_rate": 0.0008723946660690274, + "loss": 0.3487, + "step": 4663 + }, + { + "epoch": 0.26058049557225466, + "grad_norm": 0.46709099411964417, + "learning_rate": 0.0008723666517256835, + "loss": 0.4263, + "step": 4664 + }, + { + "epoch": 0.26063636617593655, + "grad_norm": 0.516634464263916, + "learning_rate": 0.0008723386373823398, + "loss": 0.4315, + "step": 4665 + }, + { + "epoch": 0.2606922367796184, + "grad_norm": 1.548061490058899, + "learning_rate": 0.000872310623038996, + "loss": 0.5076, + "step": 4666 + }, + { + "epoch": 0.2607481073833003, + "grad_norm": 0.7696959972381592, + "learning_rate": 0.0008722826086956522, + "loss": 0.497, + "step": 4667 + }, + { + "epoch": 0.2608039779869821, + "grad_norm": 0.5852643251419067, + "learning_rate": 0.0008722545943523084, + "loss": 0.3945, + "step": 4668 + }, + { + "epoch": 0.260859848590664, + "grad_norm": 0.3686451315879822, + "learning_rate": 0.0008722265800089646, + "loss": 0.3693, + "step": 4669 + }, + { + "epoch": 0.2609157191943459, + "grad_norm": 0.47836166620254517, + "learning_rate": 0.0008721985656656208, + "loss": 0.5478, + "step": 4670 + }, + { + "epoch": 0.26097158979802776, + "grad_norm": 0.5803844928741455, + "learning_rate": 0.000872170551322277, + "loss": 0.5764, + "step": 4671 + }, + { + "epoch": 0.26102746040170965, + "grad_norm": 0.40867510437965393, + "learning_rate": 0.0008721425369789332, + "loss": 0.3883, + "step": 4672 + }, + { + "epoch": 0.2610833310053915, + "grad_norm": 0.587527871131897, + "learning_rate": 0.0008721145226355894, + "loss": 0.4918, + "step": 4673 + }, + { + "epoch": 0.2611392016090734, + "grad_norm": 1.291203260421753, + "learning_rate": 0.0008720865082922456, + "loss": 0.5513, + "step": 4674 + }, + { + "epoch": 0.2611950722127553, + "grad_norm": 2.53665828704834, + "learning_rate": 0.000872058493948902, + "loss": 0.4946, + "step": 4675 + }, + { + "epoch": 0.2612509428164371, + "grad_norm": 0.4247376322746277, + "learning_rate": 0.000872030479605558, + "loss": 0.3736, + "step": 4676 + }, + { + "epoch": 0.261306813420119, + "grad_norm": 0.4429069459438324, + "learning_rate": 0.0008720024652622143, + "loss": 0.4071, + "step": 4677 + }, + { + "epoch": 0.26136268402380086, + "grad_norm": 0.6786647439002991, + "learning_rate": 0.0008719744509188705, + "loss": 0.3796, + "step": 4678 + }, + { + "epoch": 0.26141855462748276, + "grad_norm": 0.49399513006210327, + "learning_rate": 0.0008719464365755267, + "loss": 0.4615, + "step": 4679 + }, + { + "epoch": 0.2614744252311646, + "grad_norm": 0.4844259023666382, + "learning_rate": 0.0008719184222321829, + "loss": 0.4228, + "step": 4680 + }, + { + "epoch": 0.2615302958348465, + "grad_norm": 1.6712499856948853, + "learning_rate": 0.0008718904078888391, + "loss": 0.4317, + "step": 4681 + }, + { + "epoch": 0.2615861664385284, + "grad_norm": 0.834905743598938, + "learning_rate": 0.0008718623935454954, + "loss": 0.5148, + "step": 4682 + }, + { + "epoch": 0.26164203704221023, + "grad_norm": 0.4423810839653015, + "learning_rate": 0.0008718343792021515, + "loss": 0.4504, + "step": 4683 + }, + { + "epoch": 0.2616979076458921, + "grad_norm": 0.980494499206543, + "learning_rate": 0.0008718063648588077, + "loss": 0.53, + "step": 4684 + }, + { + "epoch": 0.26175377824957397, + "grad_norm": 0.894912838935852, + "learning_rate": 0.0008717783505154639, + "loss": 0.4808, + "step": 4685 + }, + { + "epoch": 0.26180964885325586, + "grad_norm": 1.6339601278305054, + "learning_rate": 0.0008717503361721202, + "loss": 0.5657, + "step": 4686 + }, + { + "epoch": 0.26186551945693776, + "grad_norm": 0.4985036551952362, + "learning_rate": 0.0008717223218287763, + "loss": 0.4724, + "step": 4687 + }, + { + "epoch": 0.2619213900606196, + "grad_norm": 0.6166756749153137, + "learning_rate": 0.0008716943074854325, + "loss": 0.4711, + "step": 4688 + }, + { + "epoch": 0.2619772606643015, + "grad_norm": 0.5356043577194214, + "learning_rate": 0.0008716662931420888, + "loss": 0.5799, + "step": 4689 + }, + { + "epoch": 0.26203313126798333, + "grad_norm": 0.653160572052002, + "learning_rate": 0.000871638278798745, + "loss": 0.4766, + "step": 4690 + }, + { + "epoch": 0.26208900187166523, + "grad_norm": 0.7461988925933838, + "learning_rate": 0.0008716102644554011, + "loss": 0.5261, + "step": 4691 + }, + { + "epoch": 0.2621448724753471, + "grad_norm": 0.6307992339134216, + "learning_rate": 0.0008715822501120573, + "loss": 0.4311, + "step": 4692 + }, + { + "epoch": 0.26220074307902896, + "grad_norm": 0.5220402479171753, + "learning_rate": 0.0008715542357687137, + "loss": 0.4191, + "step": 4693 + }, + { + "epoch": 0.26225661368271086, + "grad_norm": 0.7969569563865662, + "learning_rate": 0.0008715262214253699, + "loss": 0.5239, + "step": 4694 + }, + { + "epoch": 0.2623124842863927, + "grad_norm": 0.5475456118583679, + "learning_rate": 0.000871498207082026, + "loss": 0.5796, + "step": 4695 + }, + { + "epoch": 0.2623683548900746, + "grad_norm": 0.5020490288734436, + "learning_rate": 0.0008714701927386822, + "loss": 0.4352, + "step": 4696 + }, + { + "epoch": 0.26242422549375644, + "grad_norm": 0.5164951682090759, + "learning_rate": 0.0008714421783953385, + "loss": 0.5449, + "step": 4697 + }, + { + "epoch": 0.26248009609743833, + "grad_norm": 0.9321594834327698, + "learning_rate": 0.0008714141640519946, + "loss": 0.5676, + "step": 4698 + }, + { + "epoch": 0.2625359667011202, + "grad_norm": 0.545346736907959, + "learning_rate": 0.0008713861497086508, + "loss": 0.4306, + "step": 4699 + }, + { + "epoch": 0.26259183730480207, + "grad_norm": 0.5505691766738892, + "learning_rate": 0.0008713581353653071, + "loss": 0.3418, + "step": 4700 + }, + { + "epoch": 0.26264770790848396, + "grad_norm": 0.6262491941452026, + "learning_rate": 0.0008713301210219633, + "loss": 0.6202, + "step": 4701 + }, + { + "epoch": 0.2627035785121658, + "grad_norm": 0.8456186652183533, + "learning_rate": 0.0008713021066786194, + "loss": 0.6018, + "step": 4702 + }, + { + "epoch": 0.2627594491158477, + "grad_norm": 0.7811360359191895, + "learning_rate": 0.0008712740923352756, + "loss": 0.4865, + "step": 4703 + }, + { + "epoch": 0.2628153197195296, + "grad_norm": 0.4319712817668915, + "learning_rate": 0.0008712460779919319, + "loss": 0.4192, + "step": 4704 + }, + { + "epoch": 0.26287119032321143, + "grad_norm": 0.46278849244117737, + "learning_rate": 0.0008712180636485881, + "loss": 0.3255, + "step": 4705 + }, + { + "epoch": 0.26292706092689333, + "grad_norm": 0.7857874035835266, + "learning_rate": 0.0008711900493052442, + "loss": 0.5587, + "step": 4706 + }, + { + "epoch": 0.26298293153057517, + "grad_norm": 0.5678580403327942, + "learning_rate": 0.0008711620349619005, + "loss": 0.5266, + "step": 4707 + }, + { + "epoch": 0.26303880213425707, + "grad_norm": 0.8211227655410767, + "learning_rate": 0.0008711340206185567, + "loss": 0.4261, + "step": 4708 + }, + { + "epoch": 0.2630946727379389, + "grad_norm": 0.5850849747657776, + "learning_rate": 0.000871106006275213, + "loss": 0.3998, + "step": 4709 + }, + { + "epoch": 0.2631505433416208, + "grad_norm": 0.6358022093772888, + "learning_rate": 0.000871077991931869, + "loss": 0.4572, + "step": 4710 + }, + { + "epoch": 0.2632064139453027, + "grad_norm": 0.5420745015144348, + "learning_rate": 0.0008710499775885254, + "loss": 0.3783, + "step": 4711 + }, + { + "epoch": 0.26326228454898454, + "grad_norm": 0.7901658415794373, + "learning_rate": 0.0008710219632451816, + "loss": 0.5877, + "step": 4712 + }, + { + "epoch": 0.26331815515266643, + "grad_norm": 0.7474997639656067, + "learning_rate": 0.0008709939489018378, + "loss": 0.4967, + "step": 4713 + }, + { + "epoch": 0.2633740257563483, + "grad_norm": 0.7283620238304138, + "learning_rate": 0.000870965934558494, + "loss": 0.5291, + "step": 4714 + }, + { + "epoch": 0.26342989636003017, + "grad_norm": 0.5086837410926819, + "learning_rate": 0.0008709379202151502, + "loss": 0.5, + "step": 4715 + }, + { + "epoch": 0.26348576696371206, + "grad_norm": 0.5028250217437744, + "learning_rate": 0.0008709099058718064, + "loss": 0.4599, + "step": 4716 + }, + { + "epoch": 0.2635416375673939, + "grad_norm": 0.8519473671913147, + "learning_rate": 0.0008708818915284626, + "loss": 0.4714, + "step": 4717 + }, + { + "epoch": 0.2635975081710758, + "grad_norm": 1.743090033531189, + "learning_rate": 0.0008708538771851188, + "loss": 0.5213, + "step": 4718 + }, + { + "epoch": 0.26365337877475764, + "grad_norm": 0.46617743372917175, + "learning_rate": 0.000870825862841775, + "loss": 0.3884, + "step": 4719 + }, + { + "epoch": 0.26370924937843954, + "grad_norm": 0.6012043356895447, + "learning_rate": 0.0008707978484984312, + "loss": 0.5252, + "step": 4720 + }, + { + "epoch": 0.26376511998212143, + "grad_norm": 0.5680378079414368, + "learning_rate": 0.0008707698341550874, + "loss": 0.5392, + "step": 4721 + }, + { + "epoch": 0.26382099058580327, + "grad_norm": 0.7541109323501587, + "learning_rate": 0.0008707418198117436, + "loss": 0.417, + "step": 4722 + }, + { + "epoch": 0.26387686118948517, + "grad_norm": 0.534678041934967, + "learning_rate": 0.0008707138054683998, + "loss": 0.4683, + "step": 4723 + }, + { + "epoch": 0.263932731793167, + "grad_norm": 0.8287755846977234, + "learning_rate": 0.000870685791125056, + "loss": 0.5028, + "step": 4724 + }, + { + "epoch": 0.2639886023968489, + "grad_norm": 0.6174692511558533, + "learning_rate": 0.0008706577767817122, + "loss": 0.6912, + "step": 4725 + }, + { + "epoch": 0.26404447300053074, + "grad_norm": 0.5131943821907043, + "learning_rate": 0.0008706297624383685, + "loss": 0.3839, + "step": 4726 + }, + { + "epoch": 0.26410034360421264, + "grad_norm": 0.5556840300559998, + "learning_rate": 0.0008706017480950247, + "loss": 0.4905, + "step": 4727 + }, + { + "epoch": 0.26415621420789454, + "grad_norm": 0.4358578324317932, + "learning_rate": 0.000870573733751681, + "loss": 0.3701, + "step": 4728 + }, + { + "epoch": 0.2642120848115764, + "grad_norm": 0.6839627623558044, + "learning_rate": 0.0008705457194083371, + "loss": 0.4363, + "step": 4729 + }, + { + "epoch": 0.26426795541525827, + "grad_norm": 0.4954952299594879, + "learning_rate": 0.0008705177050649933, + "loss": 0.5605, + "step": 4730 + }, + { + "epoch": 0.2643238260189401, + "grad_norm": 0.6469412446022034, + "learning_rate": 0.0008704896907216495, + "loss": 0.5025, + "step": 4731 + }, + { + "epoch": 0.264379696622622, + "grad_norm": 0.5000861883163452, + "learning_rate": 0.0008704616763783058, + "loss": 0.3918, + "step": 4732 + }, + { + "epoch": 0.2644355672263039, + "grad_norm": 0.6531603336334229, + "learning_rate": 0.0008704336620349619, + "loss": 0.6545, + "step": 4733 + }, + { + "epoch": 0.26449143782998574, + "grad_norm": 0.6222965121269226, + "learning_rate": 0.0008704056476916181, + "loss": 0.6071, + "step": 4734 + }, + { + "epoch": 0.26454730843366764, + "grad_norm": 0.5838983654975891, + "learning_rate": 0.0008703776333482743, + "loss": 0.4933, + "step": 4735 + }, + { + "epoch": 0.2646031790373495, + "grad_norm": 0.6175536513328552, + "learning_rate": 0.0008703496190049306, + "loss": 0.4888, + "step": 4736 + }, + { + "epoch": 0.2646590496410314, + "grad_norm": 0.5576623678207397, + "learning_rate": 0.0008703216046615867, + "loss": 0.4768, + "step": 4737 + }, + { + "epoch": 0.26471492024471327, + "grad_norm": 0.6325699090957642, + "learning_rate": 0.0008702935903182429, + "loss": 0.4529, + "step": 4738 + }, + { + "epoch": 0.2647707908483951, + "grad_norm": 0.49728691577911377, + "learning_rate": 0.0008702655759748992, + "loss": 0.4454, + "step": 4739 + }, + { + "epoch": 0.264826661452077, + "grad_norm": 0.8242083787918091, + "learning_rate": 0.0008702375616315553, + "loss": 0.4319, + "step": 4740 + }, + { + "epoch": 0.26488253205575885, + "grad_norm": 1.0361502170562744, + "learning_rate": 0.0008702095472882115, + "loss": 0.5137, + "step": 4741 + }, + { + "epoch": 0.26493840265944074, + "grad_norm": 0.40767091512680054, + "learning_rate": 0.0008701815329448677, + "loss": 0.3421, + "step": 4742 + }, + { + "epoch": 0.2649942732631226, + "grad_norm": 0.4936685264110565, + "learning_rate": 0.0008701535186015241, + "loss": 0.3834, + "step": 4743 + }, + { + "epoch": 0.2650501438668045, + "grad_norm": 0.7340288162231445, + "learning_rate": 0.0008701255042581802, + "loss": 0.5722, + "step": 4744 + }, + { + "epoch": 0.2651060144704864, + "grad_norm": 0.9337022304534912, + "learning_rate": 0.0008700974899148364, + "loss": 0.5417, + "step": 4745 + }, + { + "epoch": 0.2651618850741682, + "grad_norm": 1.5300657749176025, + "learning_rate": 0.0008700694755714927, + "loss": 0.5071, + "step": 4746 + }, + { + "epoch": 0.2652177556778501, + "grad_norm": 0.40065398812294006, + "learning_rate": 0.0008700414612281489, + "loss": 0.3959, + "step": 4747 + }, + { + "epoch": 0.26527362628153195, + "grad_norm": 0.48796263337135315, + "learning_rate": 0.000870013446884805, + "loss": 0.5303, + "step": 4748 + }, + { + "epoch": 0.26532949688521384, + "grad_norm": 0.5345280766487122, + "learning_rate": 0.0008699854325414612, + "loss": 0.5788, + "step": 4749 + }, + { + "epoch": 0.26538536748889574, + "grad_norm": 0.43540409207344055, + "learning_rate": 0.0008699574181981175, + "loss": 0.4567, + "step": 4750 + }, + { + "epoch": 0.2654412380925776, + "grad_norm": 1.4380534887313843, + "learning_rate": 0.0008699294038547737, + "loss": 0.4994, + "step": 4751 + }, + { + "epoch": 0.2654971086962595, + "grad_norm": 0.8916324973106384, + "learning_rate": 0.0008699013895114298, + "loss": 0.4071, + "step": 4752 + }, + { + "epoch": 0.2655529792999413, + "grad_norm": 0.49485719203948975, + "learning_rate": 0.0008698733751680861, + "loss": 0.5195, + "step": 4753 + }, + { + "epoch": 0.2656088499036232, + "grad_norm": 0.6744000315666199, + "learning_rate": 0.0008698453608247423, + "loss": 0.4753, + "step": 4754 + }, + { + "epoch": 0.2656647205073051, + "grad_norm": 0.4359937310218811, + "learning_rate": 0.0008698173464813985, + "loss": 0.4271, + "step": 4755 + }, + { + "epoch": 0.26572059111098695, + "grad_norm": 1.641929268836975, + "learning_rate": 0.0008697893321380546, + "loss": 0.5602, + "step": 4756 + }, + { + "epoch": 0.26577646171466884, + "grad_norm": 1.9947052001953125, + "learning_rate": 0.0008697613177947109, + "loss": 0.4433, + "step": 4757 + }, + { + "epoch": 0.2658323323183507, + "grad_norm": 0.640821635723114, + "learning_rate": 0.0008697333034513671, + "loss": 0.5127, + "step": 4758 + }, + { + "epoch": 0.2658882029220326, + "grad_norm": 0.502997100353241, + "learning_rate": 0.0008697052891080234, + "loss": 0.435, + "step": 4759 + }, + { + "epoch": 0.2659440735257144, + "grad_norm": 0.7411304116249084, + "learning_rate": 0.0008696772747646796, + "loss": 0.4741, + "step": 4760 + }, + { + "epoch": 0.2659999441293963, + "grad_norm": 0.6266261339187622, + "learning_rate": 0.0008696492604213358, + "loss": 0.609, + "step": 4761 + }, + { + "epoch": 0.2660558147330782, + "grad_norm": 0.47190189361572266, + "learning_rate": 0.000869621246077992, + "loss": 0.5032, + "step": 4762 + }, + { + "epoch": 0.26611168533676005, + "grad_norm": 0.4543841779232025, + "learning_rate": 0.0008695932317346481, + "loss": 0.5345, + "step": 4763 + }, + { + "epoch": 0.26616755594044195, + "grad_norm": 0.5485124588012695, + "learning_rate": 0.0008695652173913044, + "loss": 0.4441, + "step": 4764 + }, + { + "epoch": 0.2662234265441238, + "grad_norm": 0.4349532723426819, + "learning_rate": 0.0008695372030479606, + "loss": 0.388, + "step": 4765 + }, + { + "epoch": 0.2662792971478057, + "grad_norm": 4.072417736053467, + "learning_rate": 0.0008695091887046168, + "loss": 0.3986, + "step": 4766 + }, + { + "epoch": 0.2663351677514876, + "grad_norm": 0.47908347845077515, + "learning_rate": 0.0008694811743612729, + "loss": 0.5053, + "step": 4767 + }, + { + "epoch": 0.2663910383551694, + "grad_norm": 0.44371849298477173, + "learning_rate": 0.0008694531600179292, + "loss": 0.4771, + "step": 4768 + }, + { + "epoch": 0.2664469089588513, + "grad_norm": 0.9746691584587097, + "learning_rate": 0.0008694251456745854, + "loss": 0.4719, + "step": 4769 + }, + { + "epoch": 0.26650277956253315, + "grad_norm": 0.5877187848091125, + "learning_rate": 0.0008693971313312416, + "loss": 0.4851, + "step": 4770 + }, + { + "epoch": 0.26655865016621505, + "grad_norm": 3.924994468688965, + "learning_rate": 0.0008693691169878978, + "loss": 0.6163, + "step": 4771 + }, + { + "epoch": 0.26661452076989695, + "grad_norm": 0.5479152798652649, + "learning_rate": 0.000869341102644554, + "loss": 0.4605, + "step": 4772 + }, + { + "epoch": 0.2666703913735788, + "grad_norm": 0.5398334860801697, + "learning_rate": 0.0008693130883012102, + "loss": 0.4422, + "step": 4773 + }, + { + "epoch": 0.2667262619772607, + "grad_norm": 1.4306913614273071, + "learning_rate": 0.0008692850739578664, + "loss": 0.5196, + "step": 4774 + }, + { + "epoch": 0.2667821325809425, + "grad_norm": 0.8768619298934937, + "learning_rate": 0.0008692570596145226, + "loss": 0.486, + "step": 4775 + }, + { + "epoch": 0.2668380031846244, + "grad_norm": 0.4381319582462311, + "learning_rate": 0.0008692290452711788, + "loss": 0.4409, + "step": 4776 + }, + { + "epoch": 0.26689387378830626, + "grad_norm": 1.0785605907440186, + "learning_rate": 0.0008692010309278351, + "loss": 0.5048, + "step": 4777 + }, + { + "epoch": 0.26694974439198815, + "grad_norm": 0.517749547958374, + "learning_rate": 0.0008691730165844914, + "loss": 0.5259, + "step": 4778 + }, + { + "epoch": 0.26700561499567005, + "grad_norm": 0.44754505157470703, + "learning_rate": 0.0008691450022411475, + "loss": 0.4231, + "step": 4779 + }, + { + "epoch": 0.2670614855993519, + "grad_norm": 0.4076138436794281, + "learning_rate": 0.0008691169878978037, + "loss": 0.4063, + "step": 4780 + }, + { + "epoch": 0.2671173562030338, + "grad_norm": 0.9256775975227356, + "learning_rate": 0.0008690889735544599, + "loss": 0.6184, + "step": 4781 + }, + { + "epoch": 0.2671732268067156, + "grad_norm": 0.5820056796073914, + "learning_rate": 0.0008690609592111162, + "loss": 0.4604, + "step": 4782 + }, + { + "epoch": 0.2672290974103975, + "grad_norm": 0.5159952640533447, + "learning_rate": 0.0008690329448677723, + "loss": 0.515, + "step": 4783 + }, + { + "epoch": 0.2672849680140794, + "grad_norm": 0.7165277600288391, + "learning_rate": 0.0008690049305244285, + "loss": 0.6397, + "step": 4784 + }, + { + "epoch": 0.26734083861776126, + "grad_norm": 1.5877052545547485, + "learning_rate": 0.0008689769161810848, + "loss": 0.7085, + "step": 4785 + }, + { + "epoch": 0.26739670922144315, + "grad_norm": 0.4714047610759735, + "learning_rate": 0.0008689489018377409, + "loss": 0.4731, + "step": 4786 + }, + { + "epoch": 0.267452579825125, + "grad_norm": 0.5159148573875427, + "learning_rate": 0.0008689208874943971, + "loss": 0.5139, + "step": 4787 + }, + { + "epoch": 0.2675084504288069, + "grad_norm": 0.5486021041870117, + "learning_rate": 0.0008688928731510533, + "loss": 0.468, + "step": 4788 + }, + { + "epoch": 0.2675643210324888, + "grad_norm": 0.5520483255386353, + "learning_rate": 0.0008688648588077096, + "loss": 0.4514, + "step": 4789 + }, + { + "epoch": 0.2676201916361706, + "grad_norm": 0.5883941650390625, + "learning_rate": 0.0008688368444643657, + "loss": 0.603, + "step": 4790 + }, + { + "epoch": 0.2676760622398525, + "grad_norm": 0.5158499479293823, + "learning_rate": 0.0008688088301210219, + "loss": 0.4083, + "step": 4791 + }, + { + "epoch": 0.26773193284353436, + "grad_norm": 0.5791929364204407, + "learning_rate": 0.0008687808157776783, + "loss": 0.5275, + "step": 4792 + }, + { + "epoch": 0.26778780344721625, + "grad_norm": 0.5774570107460022, + "learning_rate": 0.0008687528014343345, + "loss": 0.3586, + "step": 4793 + }, + { + "epoch": 0.2678436740508981, + "grad_norm": 0.5573208332061768, + "learning_rate": 0.0008687247870909906, + "loss": 0.5484, + "step": 4794 + }, + { + "epoch": 0.26789954465458, + "grad_norm": 0.7580753564834595, + "learning_rate": 0.0008686967727476468, + "loss": 0.5425, + "step": 4795 + }, + { + "epoch": 0.2679554152582619, + "grad_norm": 0.8818480372428894, + "learning_rate": 0.0008686687584043031, + "loss": 0.4394, + "step": 4796 + }, + { + "epoch": 0.2680112858619437, + "grad_norm": 0.5485820174217224, + "learning_rate": 0.0008686407440609593, + "loss": 0.4479, + "step": 4797 + }, + { + "epoch": 0.2680671564656256, + "grad_norm": 0.6696102619171143, + "learning_rate": 0.0008686127297176154, + "loss": 0.5242, + "step": 4798 + }, + { + "epoch": 0.26812302706930746, + "grad_norm": 1.8664488792419434, + "learning_rate": 0.0008685847153742717, + "loss": 0.3865, + "step": 4799 + }, + { + "epoch": 0.26817889767298936, + "grad_norm": 0.686521589756012, + "learning_rate": 0.0008685567010309279, + "loss": 0.4368, + "step": 4800 + }, + { + "epoch": 0.26823476827667125, + "grad_norm": 0.6686986088752747, + "learning_rate": 0.0008685286866875841, + "loss": 0.5137, + "step": 4801 + }, + { + "epoch": 0.2682906388803531, + "grad_norm": 0.6202827095985413, + "learning_rate": 0.0008685006723442402, + "loss": 0.4968, + "step": 4802 + }, + { + "epoch": 0.268346509484035, + "grad_norm": 0.5236333608627319, + "learning_rate": 0.0008684726580008965, + "loss": 0.5507, + "step": 4803 + }, + { + "epoch": 0.26840238008771683, + "grad_norm": 0.5110465884208679, + "learning_rate": 0.0008684446436575527, + "loss": 0.5025, + "step": 4804 + }, + { + "epoch": 0.2684582506913987, + "grad_norm": 0.5366842746734619, + "learning_rate": 0.0008684166293142088, + "loss": 0.5474, + "step": 4805 + }, + { + "epoch": 0.2685141212950806, + "grad_norm": 0.7544980049133301, + "learning_rate": 0.000868388614970865, + "loss": 0.4214, + "step": 4806 + }, + { + "epoch": 0.26856999189876246, + "grad_norm": 0.5390056371688843, + "learning_rate": 0.0008683606006275213, + "loss": 0.5794, + "step": 4807 + }, + { + "epoch": 0.26862586250244436, + "grad_norm": 0.4397297501564026, + "learning_rate": 0.0008683325862841775, + "loss": 0.4238, + "step": 4808 + }, + { + "epoch": 0.2686817331061262, + "grad_norm": 0.5640954971313477, + "learning_rate": 0.0008683045719408336, + "loss": 0.5401, + "step": 4809 + }, + { + "epoch": 0.2687376037098081, + "grad_norm": 0.7891275882720947, + "learning_rate": 0.00086827655759749, + "loss": 0.4687, + "step": 4810 + }, + { + "epoch": 0.26879347431348993, + "grad_norm": 0.46548452973365784, + "learning_rate": 0.0008682485432541462, + "loss": 0.3873, + "step": 4811 + }, + { + "epoch": 0.26884934491717183, + "grad_norm": 1.03413987159729, + "learning_rate": 0.0008682205289108024, + "loss": 0.6233, + "step": 4812 + }, + { + "epoch": 0.2689052155208537, + "grad_norm": 1.9921543598175049, + "learning_rate": 0.0008681925145674585, + "loss": 0.4822, + "step": 4813 + }, + { + "epoch": 0.26896108612453556, + "grad_norm": 0.7475831508636475, + "learning_rate": 0.0008681645002241148, + "loss": 0.3947, + "step": 4814 + }, + { + "epoch": 0.26901695672821746, + "grad_norm": 4.4075703620910645, + "learning_rate": 0.000868136485880771, + "loss": 0.4694, + "step": 4815 + }, + { + "epoch": 0.2690728273318993, + "grad_norm": 0.5445650219917297, + "learning_rate": 0.0008681084715374272, + "loss": 0.637, + "step": 4816 + }, + { + "epoch": 0.2691286979355812, + "grad_norm": 0.4802640676498413, + "learning_rate": 0.0008680804571940834, + "loss": 0.4734, + "step": 4817 + }, + { + "epoch": 0.2691845685392631, + "grad_norm": 0.9599683284759521, + "learning_rate": 0.0008680524428507396, + "loss": 0.5269, + "step": 4818 + }, + { + "epoch": 0.26924043914294493, + "grad_norm": 17.124006271362305, + "learning_rate": 0.0008680244285073958, + "loss": 0.3249, + "step": 4819 + }, + { + "epoch": 0.2692963097466268, + "grad_norm": 0.446634978055954, + "learning_rate": 0.000867996414164052, + "loss": 0.4057, + "step": 4820 + }, + { + "epoch": 0.26935218035030867, + "grad_norm": 0.8109078407287598, + "learning_rate": 0.0008679683998207082, + "loss": 0.5015, + "step": 4821 + }, + { + "epoch": 0.26940805095399056, + "grad_norm": 0.5955027341842651, + "learning_rate": 0.0008679403854773644, + "loss": 0.5342, + "step": 4822 + }, + { + "epoch": 0.2694639215576724, + "grad_norm": 0.7648717164993286, + "learning_rate": 0.0008679123711340206, + "loss": 0.5597, + "step": 4823 + }, + { + "epoch": 0.2695197921613543, + "grad_norm": 10.452116012573242, + "learning_rate": 0.000867884356790677, + "loss": 0.444, + "step": 4824 + }, + { + "epoch": 0.2695756627650362, + "grad_norm": 0.4436444044113159, + "learning_rate": 0.000867856342447333, + "loss": 0.3307, + "step": 4825 + }, + { + "epoch": 0.26963153336871803, + "grad_norm": 0.4791375994682312, + "learning_rate": 0.0008678283281039892, + "loss": 0.4403, + "step": 4826 + }, + { + "epoch": 0.26968740397239993, + "grad_norm": 1.1390658617019653, + "learning_rate": 0.0008678003137606455, + "loss": 0.5585, + "step": 4827 + }, + { + "epoch": 0.26974327457608177, + "grad_norm": 7.98125696182251, + "learning_rate": 0.0008677722994173017, + "loss": 0.4933, + "step": 4828 + }, + { + "epoch": 0.26979914517976367, + "grad_norm": 0.544477105140686, + "learning_rate": 0.0008677442850739579, + "loss": 0.5083, + "step": 4829 + }, + { + "epoch": 0.26985501578344556, + "grad_norm": 0.8008679151535034, + "learning_rate": 0.0008677162707306141, + "loss": 0.6129, + "step": 4830 + }, + { + "epoch": 0.2699108863871274, + "grad_norm": 0.6135130524635315, + "learning_rate": 0.0008676882563872704, + "loss": 0.4042, + "step": 4831 + }, + { + "epoch": 0.2699667569908093, + "grad_norm": 0.8530373573303223, + "learning_rate": 0.0008676602420439265, + "loss": 0.4802, + "step": 4832 + }, + { + "epoch": 0.27002262759449114, + "grad_norm": 1.9335838556289673, + "learning_rate": 0.0008676322277005827, + "loss": 0.436, + "step": 4833 + }, + { + "epoch": 0.27007849819817303, + "grad_norm": 0.4706903100013733, + "learning_rate": 0.0008676042133572389, + "loss": 0.4767, + "step": 4834 + }, + { + "epoch": 0.27013436880185493, + "grad_norm": 2.144615411758423, + "learning_rate": 0.0008675761990138952, + "loss": 0.5007, + "step": 4835 + }, + { + "epoch": 0.27019023940553677, + "grad_norm": 0.555341899394989, + "learning_rate": 0.0008675481846705513, + "loss": 0.5139, + "step": 4836 + }, + { + "epoch": 0.27024611000921867, + "grad_norm": 0.5322878360748291, + "learning_rate": 0.0008675201703272075, + "loss": 0.5098, + "step": 4837 + }, + { + "epoch": 0.2703019806129005, + "grad_norm": 0.43717771768569946, + "learning_rate": 0.0008674921559838638, + "loss": 0.4463, + "step": 4838 + }, + { + "epoch": 0.2703578512165824, + "grad_norm": 0.8694549202919006, + "learning_rate": 0.00086746414164052, + "loss": 0.459, + "step": 4839 + }, + { + "epoch": 0.27041372182026424, + "grad_norm": 0.8854798078536987, + "learning_rate": 0.0008674361272971761, + "loss": 0.4286, + "step": 4840 + }, + { + "epoch": 0.27046959242394614, + "grad_norm": 0.47070422768592834, + "learning_rate": 0.0008674081129538323, + "loss": 0.4518, + "step": 4841 + }, + { + "epoch": 0.27052546302762803, + "grad_norm": 0.4735255837440491, + "learning_rate": 0.0008673800986104887, + "loss": 0.4014, + "step": 4842 + }, + { + "epoch": 0.2705813336313099, + "grad_norm": 0.4060108959674835, + "learning_rate": 0.0008673520842671449, + "loss": 0.459, + "step": 4843 + }, + { + "epoch": 0.27063720423499177, + "grad_norm": 0.38718920946121216, + "learning_rate": 0.000867324069923801, + "loss": 0.4746, + "step": 4844 + }, + { + "epoch": 0.2706930748386736, + "grad_norm": 0.38666102290153503, + "learning_rate": 0.0008672960555804572, + "loss": 0.4596, + "step": 4845 + }, + { + "epoch": 0.2707489454423555, + "grad_norm": 0.9720858931541443, + "learning_rate": 0.0008672680412371135, + "loss": 0.4634, + "step": 4846 + }, + { + "epoch": 0.2708048160460374, + "grad_norm": 2.1870150566101074, + "learning_rate": 0.0008672400268937696, + "loss": 0.4656, + "step": 4847 + }, + { + "epoch": 0.27086068664971924, + "grad_norm": 0.6250362396240234, + "learning_rate": 0.0008672120125504258, + "loss": 0.4276, + "step": 4848 + }, + { + "epoch": 0.27091655725340114, + "grad_norm": 0.9147055149078369, + "learning_rate": 0.0008671839982070821, + "loss": 0.5659, + "step": 4849 + }, + { + "epoch": 0.270972427857083, + "grad_norm": 0.5132003426551819, + "learning_rate": 0.0008671559838637383, + "loss": 0.4995, + "step": 4850 + }, + { + "epoch": 0.27102829846076487, + "grad_norm": 0.3818069100379944, + "learning_rate": 0.0008671279695203944, + "loss": 0.3908, + "step": 4851 + }, + { + "epoch": 0.27108416906444677, + "grad_norm": 0.860121488571167, + "learning_rate": 0.0008670999551770506, + "loss": 0.5307, + "step": 4852 + }, + { + "epoch": 0.2711400396681286, + "grad_norm": 0.5585142970085144, + "learning_rate": 0.0008670719408337069, + "loss": 0.4522, + "step": 4853 + }, + { + "epoch": 0.2711959102718105, + "grad_norm": 0.5466330647468567, + "learning_rate": 0.0008670439264903631, + "loss": 0.5835, + "step": 4854 + }, + { + "epoch": 0.27125178087549234, + "grad_norm": 1.443958044052124, + "learning_rate": 0.0008670159121470192, + "loss": 0.505, + "step": 4855 + }, + { + "epoch": 0.27130765147917424, + "grad_norm": 0.6244962811470032, + "learning_rate": 0.0008669878978036755, + "loss": 0.4772, + "step": 4856 + }, + { + "epoch": 0.2713635220828561, + "grad_norm": 0.7367831468582153, + "learning_rate": 0.0008669598834603317, + "loss": 0.4822, + "step": 4857 + }, + { + "epoch": 0.271419392686538, + "grad_norm": 0.6725254654884338, + "learning_rate": 0.0008669318691169879, + "loss": 0.558, + "step": 4858 + }, + { + "epoch": 0.27147526329021987, + "grad_norm": 0.6520898938179016, + "learning_rate": 0.000866903854773644, + "loss": 0.5436, + "step": 4859 + }, + { + "epoch": 0.2715311338939017, + "grad_norm": 1.9654810428619385, + "learning_rate": 0.0008668758404303004, + "loss": 0.4294, + "step": 4860 + }, + { + "epoch": 0.2715870044975836, + "grad_norm": 1.1603329181671143, + "learning_rate": 0.0008668478260869566, + "loss": 0.4242, + "step": 4861 + }, + { + "epoch": 0.27164287510126545, + "grad_norm": 1.1054619550704956, + "learning_rate": 0.0008668198117436128, + "loss": 0.508, + "step": 4862 + }, + { + "epoch": 0.27169874570494734, + "grad_norm": 0.8218368291854858, + "learning_rate": 0.000866791797400269, + "loss": 0.5111, + "step": 4863 + }, + { + "epoch": 0.27175461630862924, + "grad_norm": 2.0514349937438965, + "learning_rate": 0.0008667637830569252, + "loss": 0.5647, + "step": 4864 + }, + { + "epoch": 0.2718104869123111, + "grad_norm": 1.0037235021591187, + "learning_rate": 0.0008667357687135814, + "loss": 0.6323, + "step": 4865 + }, + { + "epoch": 0.271866357515993, + "grad_norm": 1.2113057374954224, + "learning_rate": 0.0008667077543702376, + "loss": 0.487, + "step": 4866 + }, + { + "epoch": 0.2719222281196748, + "grad_norm": 1.0357584953308105, + "learning_rate": 0.0008666797400268938, + "loss": 0.4615, + "step": 4867 + }, + { + "epoch": 0.2719780987233567, + "grad_norm": 0.5572765469551086, + "learning_rate": 0.00086665172568355, + "loss": 0.4184, + "step": 4868 + }, + { + "epoch": 0.2720339693270386, + "grad_norm": 0.5530126094818115, + "learning_rate": 0.0008666237113402062, + "loss": 0.58, + "step": 4869 + }, + { + "epoch": 0.27208983993072045, + "grad_norm": 1.0826482772827148, + "learning_rate": 0.0008665956969968624, + "loss": 0.5262, + "step": 4870 + }, + { + "epoch": 0.27214571053440234, + "grad_norm": 0.6669971346855164, + "learning_rate": 0.0008665676826535186, + "loss": 0.4422, + "step": 4871 + }, + { + "epoch": 0.2722015811380842, + "grad_norm": 0.9388304948806763, + "learning_rate": 0.0008665396683101748, + "loss": 0.6206, + "step": 4872 + }, + { + "epoch": 0.2722574517417661, + "grad_norm": 1.1693689823150635, + "learning_rate": 0.000866511653966831, + "loss": 0.5447, + "step": 4873 + }, + { + "epoch": 0.2723133223454479, + "grad_norm": 3.5709316730499268, + "learning_rate": 0.0008664836396234872, + "loss": 0.4703, + "step": 4874 + }, + { + "epoch": 0.2723691929491298, + "grad_norm": 0.7356288433074951, + "learning_rate": 0.0008664556252801434, + "loss": 0.4856, + "step": 4875 + }, + { + "epoch": 0.2724250635528117, + "grad_norm": 1.1344703435897827, + "learning_rate": 0.0008664276109367996, + "loss": 0.4295, + "step": 4876 + }, + { + "epoch": 0.27248093415649355, + "grad_norm": 0.6246911883354187, + "learning_rate": 0.0008663995965934559, + "loss": 0.4164, + "step": 4877 + }, + { + "epoch": 0.27253680476017544, + "grad_norm": 4.9226508140563965, + "learning_rate": 0.0008663715822501121, + "loss": 0.5261, + "step": 4878 + }, + { + "epoch": 0.2725926753638573, + "grad_norm": 1.0077414512634277, + "learning_rate": 0.0008663435679067683, + "loss": 0.427, + "step": 4879 + }, + { + "epoch": 0.2726485459675392, + "grad_norm": 2.759584665298462, + "learning_rate": 0.0008663155535634245, + "loss": 0.6973, + "step": 4880 + }, + { + "epoch": 0.2727044165712211, + "grad_norm": 0.6498048901557922, + "learning_rate": 0.0008662875392200808, + "loss": 0.5516, + "step": 4881 + }, + { + "epoch": 0.2727602871749029, + "grad_norm": 0.6650070548057556, + "learning_rate": 0.0008662595248767369, + "loss": 0.4938, + "step": 4882 + }, + { + "epoch": 0.2728161577785848, + "grad_norm": 0.7618356347084045, + "learning_rate": 0.0008662315105333931, + "loss": 0.4256, + "step": 4883 + }, + { + "epoch": 0.27287202838226665, + "grad_norm": 0.8110017776489258, + "learning_rate": 0.0008662034961900493, + "loss": 0.5527, + "step": 4884 + }, + { + "epoch": 0.27292789898594855, + "grad_norm": 0.7745519876480103, + "learning_rate": 0.0008661754818467056, + "loss": 0.506, + "step": 4885 + }, + { + "epoch": 0.27298376958963044, + "grad_norm": 1.484757423400879, + "learning_rate": 0.0008661474675033617, + "loss": 0.4992, + "step": 4886 + }, + { + "epoch": 0.2730396401933123, + "grad_norm": 0.8740671277046204, + "learning_rate": 0.0008661194531600179, + "loss": 0.58, + "step": 4887 + }, + { + "epoch": 0.2730955107969942, + "grad_norm": 0.614793062210083, + "learning_rate": 0.0008660914388166742, + "loss": 0.4037, + "step": 4888 + }, + { + "epoch": 0.273151381400676, + "grad_norm": 0.5358225703239441, + "learning_rate": 0.0008660634244733303, + "loss": 0.4457, + "step": 4889 + }, + { + "epoch": 0.2732072520043579, + "grad_norm": 1.360268235206604, + "learning_rate": 0.0008660354101299865, + "loss": 0.5452, + "step": 4890 + }, + { + "epoch": 0.27326312260803975, + "grad_norm": 0.866869330406189, + "learning_rate": 0.0008660073957866427, + "loss": 0.4362, + "step": 4891 + }, + { + "epoch": 0.27331899321172165, + "grad_norm": 4.54645299911499, + "learning_rate": 0.000865979381443299, + "loss": 0.4724, + "step": 4892 + }, + { + "epoch": 0.27337486381540355, + "grad_norm": 1.1594613790512085, + "learning_rate": 0.0008659513670999551, + "loss": 0.5525, + "step": 4893 + }, + { + "epoch": 0.2734307344190854, + "grad_norm": 1.2743513584136963, + "learning_rate": 0.0008659233527566114, + "loss": 0.5013, + "step": 4894 + }, + { + "epoch": 0.2734866050227673, + "grad_norm": 0.856540322303772, + "learning_rate": 0.0008658953384132677, + "loss": 0.5898, + "step": 4895 + }, + { + "epoch": 0.2735424756264491, + "grad_norm": 0.5565491914749146, + "learning_rate": 0.0008658673240699239, + "loss": 0.4558, + "step": 4896 + }, + { + "epoch": 0.273598346230131, + "grad_norm": 0.6366143822669983, + "learning_rate": 0.00086583930972658, + "loss": 0.4633, + "step": 4897 + }, + { + "epoch": 0.2736542168338129, + "grad_norm": 0.6410838961601257, + "learning_rate": 0.0008658112953832362, + "loss": 0.5046, + "step": 4898 + }, + { + "epoch": 0.27371008743749475, + "grad_norm": 0.6369917988777161, + "learning_rate": 0.0008657832810398925, + "loss": 0.4698, + "step": 4899 + }, + { + "epoch": 0.27376595804117665, + "grad_norm": 2.7282145023345947, + "learning_rate": 0.0008657552666965487, + "loss": 0.4813, + "step": 4900 + }, + { + "epoch": 0.2738218286448585, + "grad_norm": 0.5758755207061768, + "learning_rate": 0.0008657272523532048, + "loss": 0.4337, + "step": 4901 + }, + { + "epoch": 0.2738776992485404, + "grad_norm": 0.5336123704910278, + "learning_rate": 0.0008656992380098611, + "loss": 0.4678, + "step": 4902 + }, + { + "epoch": 0.2739335698522223, + "grad_norm": 0.6329205632209778, + "learning_rate": 0.0008656712236665173, + "loss": 0.6043, + "step": 4903 + }, + { + "epoch": 0.2739894404559041, + "grad_norm": 0.43444398045539856, + "learning_rate": 0.0008656432093231735, + "loss": 0.4212, + "step": 4904 + }, + { + "epoch": 0.274045311059586, + "grad_norm": 0.49284207820892334, + "learning_rate": 0.0008656151949798296, + "loss": 0.4818, + "step": 4905 + }, + { + "epoch": 0.27410118166326786, + "grad_norm": 1.2231004238128662, + "learning_rate": 0.0008655871806364859, + "loss": 0.5329, + "step": 4906 + }, + { + "epoch": 0.27415705226694975, + "grad_norm": 0.5449196696281433, + "learning_rate": 0.0008655591662931421, + "loss": 0.4864, + "step": 4907 + }, + { + "epoch": 0.2742129228706316, + "grad_norm": 0.4043261408805847, + "learning_rate": 0.0008655311519497983, + "loss": 0.3674, + "step": 4908 + }, + { + "epoch": 0.2742687934743135, + "grad_norm": 0.8185630440711975, + "learning_rate": 0.0008655031376064544, + "loss": 0.4331, + "step": 4909 + }, + { + "epoch": 0.2743246640779954, + "grad_norm": 0.868583083152771, + "learning_rate": 0.0008654751232631108, + "loss": 0.5433, + "step": 4910 + }, + { + "epoch": 0.2743805346816772, + "grad_norm": 0.566251814365387, + "learning_rate": 0.000865447108919767, + "loss": 0.5956, + "step": 4911 + }, + { + "epoch": 0.2744364052853591, + "grad_norm": 0.5048642158508301, + "learning_rate": 0.0008654190945764231, + "loss": 0.4303, + "step": 4912 + }, + { + "epoch": 0.27449227588904096, + "grad_norm": 0.4740193486213684, + "learning_rate": 0.0008653910802330794, + "loss": 0.4494, + "step": 4913 + }, + { + "epoch": 0.27454814649272286, + "grad_norm": 0.4400840103626251, + "learning_rate": 0.0008653630658897356, + "loss": 0.373, + "step": 4914 + }, + { + "epoch": 0.27460401709640475, + "grad_norm": 3.038297414779663, + "learning_rate": 0.0008653350515463918, + "loss": 0.4977, + "step": 4915 + }, + { + "epoch": 0.2746598877000866, + "grad_norm": 2.4509341716766357, + "learning_rate": 0.0008653070372030479, + "loss": 0.5578, + "step": 4916 + }, + { + "epoch": 0.2747157583037685, + "grad_norm": 0.8091301918029785, + "learning_rate": 0.0008652790228597042, + "loss": 0.4499, + "step": 4917 + }, + { + "epoch": 0.2747716289074503, + "grad_norm": 0.7286555767059326, + "learning_rate": 0.0008652510085163604, + "loss": 0.4712, + "step": 4918 + }, + { + "epoch": 0.2748274995111322, + "grad_norm": 2.028243064880371, + "learning_rate": 0.0008652229941730166, + "loss": 0.6195, + "step": 4919 + }, + { + "epoch": 0.27488337011481406, + "grad_norm": 0.5542038083076477, + "learning_rate": 0.0008651949798296728, + "loss": 0.5168, + "step": 4920 + }, + { + "epoch": 0.27493924071849596, + "grad_norm": 0.4808811843395233, + "learning_rate": 0.000865166965486329, + "loss": 0.431, + "step": 4921 + }, + { + "epoch": 0.27499511132217785, + "grad_norm": 0.9802090525627136, + "learning_rate": 0.0008651389511429852, + "loss": 0.5387, + "step": 4922 + }, + { + "epoch": 0.2750509819258597, + "grad_norm": 0.8516334295272827, + "learning_rate": 0.0008651109367996414, + "loss": 0.4981, + "step": 4923 + }, + { + "epoch": 0.2751068525295416, + "grad_norm": 0.6784085631370544, + "learning_rate": 0.0008650829224562976, + "loss": 0.5429, + "step": 4924 + }, + { + "epoch": 0.27516272313322343, + "grad_norm": 0.5094239115715027, + "learning_rate": 0.0008650549081129538, + "loss": 0.4265, + "step": 4925 + }, + { + "epoch": 0.2752185937369053, + "grad_norm": 0.5779902935028076, + "learning_rate": 0.00086502689376961, + "loss": 0.4103, + "step": 4926 + }, + { + "epoch": 0.2752744643405872, + "grad_norm": 0.6518490314483643, + "learning_rate": 0.0008649988794262664, + "loss": 0.5856, + "step": 4927 + }, + { + "epoch": 0.27533033494426906, + "grad_norm": 0.636206865310669, + "learning_rate": 0.0008649708650829225, + "loss": 0.3761, + "step": 4928 + }, + { + "epoch": 0.27538620554795096, + "grad_norm": 0.7733147144317627, + "learning_rate": 0.0008649428507395787, + "loss": 0.4001, + "step": 4929 + }, + { + "epoch": 0.2754420761516328, + "grad_norm": 0.9372087717056274, + "learning_rate": 0.0008649148363962349, + "loss": 0.6212, + "step": 4930 + }, + { + "epoch": 0.2754979467553147, + "grad_norm": 0.7932202816009521, + "learning_rate": 0.0008648868220528912, + "loss": 0.4709, + "step": 4931 + }, + { + "epoch": 0.2755538173589966, + "grad_norm": 0.3518871068954468, + "learning_rate": 0.0008648588077095473, + "loss": 0.4183, + "step": 4932 + }, + { + "epoch": 0.27560968796267843, + "grad_norm": 0.7865981459617615, + "learning_rate": 0.0008648307933662035, + "loss": 0.5914, + "step": 4933 + }, + { + "epoch": 0.2756655585663603, + "grad_norm": 0.7340072989463806, + "learning_rate": 0.0008648027790228598, + "loss": 0.5614, + "step": 4934 + }, + { + "epoch": 0.27572142917004216, + "grad_norm": 0.7754415273666382, + "learning_rate": 0.0008647747646795159, + "loss": 0.558, + "step": 4935 + }, + { + "epoch": 0.27577729977372406, + "grad_norm": 0.5848038196563721, + "learning_rate": 0.0008647467503361721, + "loss": 0.4461, + "step": 4936 + }, + { + "epoch": 0.2758331703774059, + "grad_norm": 0.9898869395256042, + "learning_rate": 0.0008647187359928283, + "loss": 0.4585, + "step": 4937 + }, + { + "epoch": 0.2758890409810878, + "grad_norm": 0.41112497448921204, + "learning_rate": 0.0008646907216494846, + "loss": 0.4061, + "step": 4938 + }, + { + "epoch": 0.2759449115847697, + "grad_norm": 0.8684262037277222, + "learning_rate": 0.0008646627073061407, + "loss": 0.6063, + "step": 4939 + }, + { + "epoch": 0.27600078218845153, + "grad_norm": 0.7563137412071228, + "learning_rate": 0.0008646346929627969, + "loss": 0.4085, + "step": 4940 + }, + { + "epoch": 0.27605665279213343, + "grad_norm": 0.7687857747077942, + "learning_rate": 0.0008646066786194532, + "loss": 0.5382, + "step": 4941 + }, + { + "epoch": 0.27611252339581527, + "grad_norm": 0.7805190086364746, + "learning_rate": 0.0008645786642761094, + "loss": 0.5483, + "step": 4942 + }, + { + "epoch": 0.27616839399949716, + "grad_norm": 0.4105584919452667, + "learning_rate": 0.0008645506499327655, + "loss": 0.3524, + "step": 4943 + }, + { + "epoch": 0.27622426460317906, + "grad_norm": 0.5850437879562378, + "learning_rate": 0.0008645226355894218, + "loss": 0.5401, + "step": 4944 + }, + { + "epoch": 0.2762801352068609, + "grad_norm": 3.956526517868042, + "learning_rate": 0.0008644946212460781, + "loss": 0.4395, + "step": 4945 + }, + { + "epoch": 0.2763360058105428, + "grad_norm": 0.7465658783912659, + "learning_rate": 0.0008644666069027343, + "loss": 0.4127, + "step": 4946 + }, + { + "epoch": 0.27639187641422464, + "grad_norm": 0.7865839600563049, + "learning_rate": 0.0008644385925593904, + "loss": 0.641, + "step": 4947 + }, + { + "epoch": 0.27644774701790653, + "grad_norm": 0.8092109560966492, + "learning_rate": 0.0008644105782160466, + "loss": 0.4055, + "step": 4948 + }, + { + "epoch": 0.2765036176215884, + "grad_norm": 0.5448030829429626, + "learning_rate": 0.0008643825638727029, + "loss": 0.4899, + "step": 4949 + }, + { + "epoch": 0.27655948822527027, + "grad_norm": 1.575411319732666, + "learning_rate": 0.0008643545495293591, + "loss": 0.6914, + "step": 4950 + }, + { + "epoch": 0.27661535882895216, + "grad_norm": 0.5164878368377686, + "learning_rate": 0.0008643265351860152, + "loss": 0.4693, + "step": 4951 + }, + { + "epoch": 0.276671229432634, + "grad_norm": 0.8145294785499573, + "learning_rate": 0.0008642985208426715, + "loss": 0.4565, + "step": 4952 + }, + { + "epoch": 0.2767271000363159, + "grad_norm": 0.7728568315505981, + "learning_rate": 0.0008642705064993277, + "loss": 0.5577, + "step": 4953 + }, + { + "epoch": 0.27678297063999774, + "grad_norm": 0.6196942925453186, + "learning_rate": 0.0008642424921559838, + "loss": 0.5383, + "step": 4954 + }, + { + "epoch": 0.27683884124367963, + "grad_norm": 0.5229423642158508, + "learning_rate": 0.00086421447781264, + "loss": 0.482, + "step": 4955 + }, + { + "epoch": 0.27689471184736153, + "grad_norm": 0.6103101372718811, + "learning_rate": 0.0008641864634692963, + "loss": 0.3466, + "step": 4956 + }, + { + "epoch": 0.27695058245104337, + "grad_norm": 1.3213756084442139, + "learning_rate": 0.0008641584491259525, + "loss": 0.4678, + "step": 4957 + }, + { + "epoch": 0.27700645305472527, + "grad_norm": 0.6521528959274292, + "learning_rate": 0.0008641304347826086, + "loss": 0.4371, + "step": 4958 + }, + { + "epoch": 0.2770623236584071, + "grad_norm": 0.50040203332901, + "learning_rate": 0.000864102420439265, + "loss": 0.4191, + "step": 4959 + }, + { + "epoch": 0.277118194262089, + "grad_norm": 0.8673403263092041, + "learning_rate": 0.0008640744060959212, + "loss": 0.5033, + "step": 4960 + }, + { + "epoch": 0.2771740648657709, + "grad_norm": 1.74127197265625, + "learning_rate": 0.0008640463917525774, + "loss": 0.5059, + "step": 4961 + }, + { + "epoch": 0.27722993546945274, + "grad_norm": 0.5775505900382996, + "learning_rate": 0.0008640183774092335, + "loss": 0.4137, + "step": 4962 + }, + { + "epoch": 0.27728580607313463, + "grad_norm": 0.6535037159919739, + "learning_rate": 0.0008639903630658898, + "loss": 0.4102, + "step": 4963 + }, + { + "epoch": 0.2773416766768165, + "grad_norm": 0.7942030429840088, + "learning_rate": 0.000863962348722546, + "loss": 0.4421, + "step": 4964 + }, + { + "epoch": 0.27739754728049837, + "grad_norm": 0.4343326985836029, + "learning_rate": 0.0008639343343792022, + "loss": 0.4016, + "step": 4965 + }, + { + "epoch": 0.27745341788418026, + "grad_norm": 1.3634666204452515, + "learning_rate": 0.0008639063200358584, + "loss": 0.4698, + "step": 4966 + }, + { + "epoch": 0.2775092884878621, + "grad_norm": 0.6712188124656677, + "learning_rate": 0.0008638783056925146, + "loss": 0.4293, + "step": 4967 + }, + { + "epoch": 0.277565159091544, + "grad_norm": 0.7760248780250549, + "learning_rate": 0.0008638502913491708, + "loss": 0.4654, + "step": 4968 + }, + { + "epoch": 0.27762102969522584, + "grad_norm": 0.5173790454864502, + "learning_rate": 0.000863822277005827, + "loss": 0.3734, + "step": 4969 + }, + { + "epoch": 0.27767690029890774, + "grad_norm": 0.6529727578163147, + "learning_rate": 0.0008637942626624832, + "loss": 0.3783, + "step": 4970 + }, + { + "epoch": 0.2777327709025896, + "grad_norm": 0.4705888032913208, + "learning_rate": 0.0008637662483191394, + "loss": 0.4833, + "step": 4971 + }, + { + "epoch": 0.27778864150627147, + "grad_norm": 0.6162322163581848, + "learning_rate": 0.0008637382339757956, + "loss": 0.5197, + "step": 4972 + }, + { + "epoch": 0.27784451210995337, + "grad_norm": 1.384852647781372, + "learning_rate": 0.0008637102196324519, + "loss": 0.4763, + "step": 4973 + }, + { + "epoch": 0.2779003827136352, + "grad_norm": 0.6582178473472595, + "learning_rate": 0.000863682205289108, + "loss": 0.4957, + "step": 4974 + }, + { + "epoch": 0.2779562533173171, + "grad_norm": 3.2424230575561523, + "learning_rate": 0.0008636541909457642, + "loss": 0.6533, + "step": 4975 + }, + { + "epoch": 0.27801212392099894, + "grad_norm": 1.4161373376846313, + "learning_rate": 0.0008636261766024204, + "loss": 0.4367, + "step": 4976 + }, + { + "epoch": 0.27806799452468084, + "grad_norm": 0.9001994132995605, + "learning_rate": 0.0008635981622590767, + "loss": 0.7071, + "step": 4977 + }, + { + "epoch": 0.27812386512836273, + "grad_norm": 0.6062504649162292, + "learning_rate": 0.0008635701479157329, + "loss": 0.51, + "step": 4978 + }, + { + "epoch": 0.2781797357320446, + "grad_norm": 0.41843074560165405, + "learning_rate": 0.0008635421335723891, + "loss": 0.3754, + "step": 4979 + }, + { + "epoch": 0.27823560633572647, + "grad_norm": 0.6130450367927551, + "learning_rate": 0.0008635141192290454, + "loss": 0.6064, + "step": 4980 + }, + { + "epoch": 0.2782914769394083, + "grad_norm": 0.410011887550354, + "learning_rate": 0.0008634861048857015, + "loss": 0.4289, + "step": 4981 + }, + { + "epoch": 0.2783473475430902, + "grad_norm": 0.8402948975563049, + "learning_rate": 0.0008634580905423577, + "loss": 0.4489, + "step": 4982 + }, + { + "epoch": 0.2784032181467721, + "grad_norm": 0.7748754024505615, + "learning_rate": 0.0008634300761990139, + "loss": 0.5585, + "step": 4983 + }, + { + "epoch": 0.27845908875045394, + "grad_norm": 0.503722071647644, + "learning_rate": 0.0008634020618556702, + "loss": 0.4231, + "step": 4984 + }, + { + "epoch": 0.27851495935413584, + "grad_norm": 0.38928523659706116, + "learning_rate": 0.0008633740475123263, + "loss": 0.4769, + "step": 4985 + }, + { + "epoch": 0.2785708299578177, + "grad_norm": 0.8118487000465393, + "learning_rate": 0.0008633460331689825, + "loss": 0.4371, + "step": 4986 + }, + { + "epoch": 0.2786267005614996, + "grad_norm": 1.0253537893295288, + "learning_rate": 0.0008633180188256387, + "loss": 0.5538, + "step": 4987 + }, + { + "epoch": 0.2786825711651814, + "grad_norm": 1.5670850276947021, + "learning_rate": 0.000863290004482295, + "loss": 0.5218, + "step": 4988 + }, + { + "epoch": 0.2787384417688633, + "grad_norm": 0.491548091173172, + "learning_rate": 0.0008632619901389511, + "loss": 0.4968, + "step": 4989 + }, + { + "epoch": 0.2787943123725452, + "grad_norm": 1.1772825717926025, + "learning_rate": 0.0008632339757956073, + "loss": 0.5043, + "step": 4990 + }, + { + "epoch": 0.27885018297622705, + "grad_norm": 0.5697044134140015, + "learning_rate": 0.0008632059614522636, + "loss": 0.5824, + "step": 4991 + }, + { + "epoch": 0.27890605357990894, + "grad_norm": 0.6100141406059265, + "learning_rate": 0.0008631779471089198, + "loss": 0.5144, + "step": 4992 + }, + { + "epoch": 0.2789619241835908, + "grad_norm": 0.49396592378616333, + "learning_rate": 0.0008631499327655759, + "loss": 0.4865, + "step": 4993 + }, + { + "epoch": 0.2790177947872727, + "grad_norm": 0.44310474395751953, + "learning_rate": 0.0008631219184222322, + "loss": 0.3795, + "step": 4994 + }, + { + "epoch": 0.2790736653909546, + "grad_norm": 0.5251303315162659, + "learning_rate": 0.0008630939040788885, + "loss": 0.4529, + "step": 4995 + }, + { + "epoch": 0.2791295359946364, + "grad_norm": 0.7194657325744629, + "learning_rate": 0.0008630658897355446, + "loss": 0.4909, + "step": 4996 + }, + { + "epoch": 0.2791854065983183, + "grad_norm": 0.9757006764411926, + "learning_rate": 0.0008630378753922008, + "loss": 0.467, + "step": 4997 + }, + { + "epoch": 0.27924127720200015, + "grad_norm": 0.9638657569885254, + "learning_rate": 0.0008630098610488571, + "loss": 0.763, + "step": 4998 + }, + { + "epoch": 0.27929714780568204, + "grad_norm": 0.8412012457847595, + "learning_rate": 0.0008629818467055133, + "loss": 0.4174, + "step": 4999 + }, + { + "epoch": 0.27935301840936394, + "grad_norm": 0.7205965518951416, + "learning_rate": 0.0008629538323621694, + "loss": 0.3887, + "step": 5000 + }, + { + "epoch": 0.27935301840936394, + "eval_cer": 0.09850195847380883, + "eval_loss": 0.3619626462459564, + "eval_runtime": 55.9571, + "eval_samples_per_second": 81.098, + "eval_steps_per_second": 5.075, + "eval_wer": 0.3924750705287562, + "step": 5000 + }, + { + "epoch": 0.2794088890130458, + "grad_norm": 0.6415395140647888, + "learning_rate": 0.0008629258180188256, + "loss": 0.4845, + "step": 5001 + }, + { + "epoch": 0.2794647596167277, + "grad_norm": 0.5367297530174255, + "learning_rate": 0.0008628978036754819, + "loss": 0.5365, + "step": 5002 + }, + { + "epoch": 0.2795206302204095, + "grad_norm": 0.648182213306427, + "learning_rate": 0.0008628697893321381, + "loss": 0.4719, + "step": 5003 + }, + { + "epoch": 0.2795765008240914, + "grad_norm": 0.5080143809318542, + "learning_rate": 0.0008628417749887942, + "loss": 0.3673, + "step": 5004 + }, + { + "epoch": 0.27963237142777325, + "grad_norm": 0.7081291079521179, + "learning_rate": 0.0008628137606454505, + "loss": 0.4054, + "step": 5005 + }, + { + "epoch": 0.27968824203145515, + "grad_norm": 0.4722939133644104, + "learning_rate": 0.0008627857463021067, + "loss": 0.5069, + "step": 5006 + }, + { + "epoch": 0.27974411263513704, + "grad_norm": 0.5381889343261719, + "learning_rate": 0.0008627577319587629, + "loss": 0.5001, + "step": 5007 + }, + { + "epoch": 0.2797999832388189, + "grad_norm": 0.4618498980998993, + "learning_rate": 0.000862729717615419, + "loss": 0.5016, + "step": 5008 + }, + { + "epoch": 0.2798558538425008, + "grad_norm": 0.4697301387786865, + "learning_rate": 0.0008627017032720753, + "loss": 0.371, + "step": 5009 + }, + { + "epoch": 0.2799117244461826, + "grad_norm": Infinity, + "learning_rate": 0.0008627017032720753, + "loss": 0.4719, + "step": 5010 + }, + { + "epoch": 0.2799675950498645, + "grad_norm": 1.2877476215362549, + "learning_rate": 0.0008626736889287316, + "loss": 0.5975, + "step": 5011 + }, + { + "epoch": 0.2800234656535464, + "grad_norm": 0.6845802068710327, + "learning_rate": 0.0008626456745853878, + "loss": 0.4796, + "step": 5012 + }, + { + "epoch": 0.28007933625722825, + "grad_norm": 0.4130478799343109, + "learning_rate": 0.000862617660242044, + "loss": 0.409, + "step": 5013 + }, + { + "epoch": 0.28013520686091015, + "grad_norm": 0.44432687759399414, + "learning_rate": 0.0008625896458987002, + "loss": 0.4377, + "step": 5014 + }, + { + "epoch": 0.280191077464592, + "grad_norm": 0.6324438452720642, + "learning_rate": 0.0008625616315553564, + "loss": 0.5373, + "step": 5015 + }, + { + "epoch": 0.2802469480682739, + "grad_norm": 0.3965279757976532, + "learning_rate": 0.0008625336172120126, + "loss": 0.4893, + "step": 5016 + }, + { + "epoch": 0.2803028186719558, + "grad_norm": 0.8192760944366455, + "learning_rate": 0.0008625056028686688, + "loss": 0.4956, + "step": 5017 + }, + { + "epoch": 0.2803586892756376, + "grad_norm": Infinity, + "learning_rate": 0.0008625056028686688, + "loss": 0.4927, + "step": 5018 + }, + { + "epoch": 0.2804145598793195, + "grad_norm": 0.5879487991333008, + "learning_rate": 0.000862477588525325, + "loss": 0.4449, + "step": 5019 + }, + { + "epoch": 0.28047043048300135, + "grad_norm": 0.5210895538330078, + "learning_rate": 0.0008624495741819812, + "loss": 0.3944, + "step": 5020 + }, + { + "epoch": 0.28052630108668325, + "grad_norm": 0.7636909484863281, + "learning_rate": 0.0008624215598386373, + "loss": 0.5073, + "step": 5021 + }, + { + "epoch": 0.2805821716903651, + "grad_norm": 3.93509578704834, + "learning_rate": 0.0008623935454952936, + "loss": 0.4153, + "step": 5022 + }, + { + "epoch": 0.280638042294047, + "grad_norm": 1.0877807140350342, + "learning_rate": 0.0008623655311519498, + "loss": 0.6254, + "step": 5023 + }, + { + "epoch": 0.2806939128977289, + "grad_norm": 0.7232125401496887, + "learning_rate": 0.000862337516808606, + "loss": 0.4831, + "step": 5024 + }, + { + "epoch": 0.2807497835014107, + "grad_norm": 0.6046668291091919, + "learning_rate": 0.0008623095024652622, + "loss": 0.3945, + "step": 5025 + }, + { + "epoch": 0.2808056541050926, + "grad_norm": 0.5011358857154846, + "learning_rate": 0.0008622814881219184, + "loss": 0.4281, + "step": 5026 + }, + { + "epoch": 0.28086152470877446, + "grad_norm": 0.47909364104270935, + "learning_rate": 0.0008622534737785746, + "loss": 0.3878, + "step": 5027 + }, + { + "epoch": 0.28091739531245635, + "grad_norm": 0.6834143400192261, + "learning_rate": 0.0008622254594352308, + "loss": 0.432, + "step": 5028 + }, + { + "epoch": 0.28097326591613825, + "grad_norm": 1.0216916799545288, + "learning_rate": 0.000862197445091887, + "loss": 0.457, + "step": 5029 + }, + { + "epoch": 0.2810291365198201, + "grad_norm": 0.6524214744567871, + "learning_rate": 0.0008621694307485433, + "loss": 0.5326, + "step": 5030 + }, + { + "epoch": 0.281085007123502, + "grad_norm": 0.45287981629371643, + "learning_rate": 0.0008621414164051995, + "loss": 0.4521, + "step": 5031 + }, + { + "epoch": 0.2811408777271838, + "grad_norm": 0.5708306431770325, + "learning_rate": 0.0008621134020618558, + "loss": 0.4977, + "step": 5032 + }, + { + "epoch": 0.2811967483308657, + "grad_norm": 0.5449576377868652, + "learning_rate": 0.0008620853877185119, + "loss": 0.4538, + "step": 5033 + }, + { + "epoch": 0.28125261893454756, + "grad_norm": 0.6853867769241333, + "learning_rate": 0.0008620573733751681, + "loss": 0.5203, + "step": 5034 + }, + { + "epoch": 0.28130848953822946, + "grad_norm": 0.6843395829200745, + "learning_rate": 0.0008620293590318243, + "loss": 0.4606, + "step": 5035 + }, + { + "epoch": 0.28136436014191135, + "grad_norm": 0.42617490887641907, + "learning_rate": 0.0008620013446884806, + "loss": 0.5498, + "step": 5036 + }, + { + "epoch": 0.2814202307455932, + "grad_norm": 0.5373966693878174, + "learning_rate": 0.0008619733303451367, + "loss": 0.4287, + "step": 5037 + }, + { + "epoch": 0.2814761013492751, + "grad_norm": Infinity, + "learning_rate": 0.0008619733303451367, + "loss": 0.4062, + "step": 5038 + }, + { + "epoch": 0.2815319719529569, + "grad_norm": 0.6020870208740234, + "learning_rate": 0.0008619453160017929, + "loss": 0.5459, + "step": 5039 + }, + { + "epoch": 0.2815878425566388, + "grad_norm": 0.4885038435459137, + "learning_rate": 0.0008619173016584492, + "loss": 0.4593, + "step": 5040 + }, + { + "epoch": 0.2816437131603207, + "grad_norm": 0.6149919629096985, + "learning_rate": 0.0008618892873151053, + "loss": 0.5143, + "step": 5041 + }, + { + "epoch": 0.28169958376400256, + "grad_norm": Infinity, + "learning_rate": 0.0008618892873151053, + "loss": 0.5589, + "step": 5042 + }, + { + "epoch": 0.28175545436768445, + "grad_norm": 4.473811149597168, + "learning_rate": 0.0008618612729717615, + "loss": 0.4806, + "step": 5043 + }, + { + "epoch": 0.2818113249713663, + "grad_norm": 0.701213002204895, + "learning_rate": 0.0008618332586284177, + "loss": 0.5777, + "step": 5044 + }, + { + "epoch": 0.2818671955750482, + "grad_norm": 1.131921648979187, + "learning_rate": 0.000861805244285074, + "loss": 0.4463, + "step": 5045 + }, + { + "epoch": 0.2819230661787301, + "grad_norm": 0.5974535942077637, + "learning_rate": 0.0008617772299417301, + "loss": 0.5634, + "step": 5046 + }, + { + "epoch": 0.2819789367824119, + "grad_norm": 1.778108835220337, + "learning_rate": 0.0008617492155983863, + "loss": 0.5067, + "step": 5047 + }, + { + "epoch": 0.2820348073860938, + "grad_norm": 0.4950951337814331, + "learning_rate": 0.0008617212012550427, + "loss": 0.3553, + "step": 5048 + }, + { + "epoch": 0.28209067798977566, + "grad_norm": 0.6047303080558777, + "learning_rate": 0.0008616931869116989, + "loss": 0.4867, + "step": 5049 + }, + { + "epoch": 0.28214654859345756, + "grad_norm": 0.8404955863952637, + "learning_rate": 0.000861665172568355, + "loss": 0.4499, + "step": 5050 + }, + { + "epoch": 0.2822024191971394, + "grad_norm": 0.41792699694633484, + "learning_rate": 0.0008616371582250112, + "loss": 0.471, + "step": 5051 + }, + { + "epoch": 0.2822582898008213, + "grad_norm": 1.0072171688079834, + "learning_rate": 0.0008616091438816675, + "loss": 0.4645, + "step": 5052 + }, + { + "epoch": 0.2823141604045032, + "grad_norm": 0.592352569103241, + "learning_rate": 0.0008615811295383237, + "loss": 0.5229, + "step": 5053 + }, + { + "epoch": 0.28237003100818503, + "grad_norm": 0.5170997381210327, + "learning_rate": 0.0008615531151949798, + "loss": 0.4451, + "step": 5054 + }, + { + "epoch": 0.2824259016118669, + "grad_norm": 0.4069174528121948, + "learning_rate": 0.0008615251008516361, + "loss": 0.4692, + "step": 5055 + }, + { + "epoch": 0.28248177221554877, + "grad_norm": 0.7693159580230713, + "learning_rate": 0.0008614970865082923, + "loss": 0.4658, + "step": 5056 + }, + { + "epoch": 0.28253764281923066, + "grad_norm": 0.6280845403671265, + "learning_rate": 0.0008614690721649485, + "loss": 0.4488, + "step": 5057 + }, + { + "epoch": 0.28259351342291256, + "grad_norm": 0.5164259672164917, + "learning_rate": 0.0008614410578216046, + "loss": 0.4724, + "step": 5058 + }, + { + "epoch": 0.2826493840265944, + "grad_norm": 0.5467740297317505, + "learning_rate": 0.0008614130434782609, + "loss": 0.457, + "step": 5059 + }, + { + "epoch": 0.2827052546302763, + "grad_norm": 0.5902903079986572, + "learning_rate": 0.0008613850291349171, + "loss": 0.4019, + "step": 5060 + }, + { + "epoch": 0.28276112523395813, + "grad_norm": 1.206576943397522, + "learning_rate": 0.0008613570147915733, + "loss": 0.625, + "step": 5061 + }, + { + "epoch": 0.28281699583764003, + "grad_norm": 0.773082435131073, + "learning_rate": 0.0008613290004482294, + "loss": 0.8476, + "step": 5062 + }, + { + "epoch": 0.2828728664413219, + "grad_norm": 0.7244957089424133, + "learning_rate": 0.0008613009861048857, + "loss": 0.5027, + "step": 5063 + }, + { + "epoch": 0.28292873704500376, + "grad_norm": 0.5481377243995667, + "learning_rate": 0.000861272971761542, + "loss": 0.5469, + "step": 5064 + }, + { + "epoch": 0.28298460764868566, + "grad_norm": 0.9994580149650574, + "learning_rate": 0.000861244957418198, + "loss": 0.4983, + "step": 5065 + }, + { + "epoch": 0.2830404782523675, + "grad_norm": 0.8076426386833191, + "learning_rate": 0.0008612169430748544, + "loss": 0.4614, + "step": 5066 + }, + { + "epoch": 0.2830963488560494, + "grad_norm": 1.9266659021377563, + "learning_rate": 0.0008611889287315106, + "loss": 0.521, + "step": 5067 + }, + { + "epoch": 0.28315221945973124, + "grad_norm": 0.45378053188323975, + "learning_rate": 0.0008611609143881668, + "loss": 0.4203, + "step": 5068 + }, + { + "epoch": 0.28320809006341313, + "grad_norm": 0.5800686478614807, + "learning_rate": 0.0008611329000448229, + "loss": 0.433, + "step": 5069 + }, + { + "epoch": 0.283263960667095, + "grad_norm": 1.8816627264022827, + "learning_rate": 0.0008611048857014792, + "loss": 0.5158, + "step": 5070 + }, + { + "epoch": 0.28331983127077687, + "grad_norm": 0.7481634020805359, + "learning_rate": 0.0008610768713581354, + "loss": 0.5356, + "step": 5071 + }, + { + "epoch": 0.28337570187445876, + "grad_norm": 7.411871910095215, + "learning_rate": 0.0008610488570147916, + "loss": 0.3938, + "step": 5072 + }, + { + "epoch": 0.2834315724781406, + "grad_norm": 0.47670990228652954, + "learning_rate": 0.0008610208426714478, + "loss": 0.4974, + "step": 5073 + }, + { + "epoch": 0.2834874430818225, + "grad_norm": 0.7607618570327759, + "learning_rate": 0.000860992828328104, + "loss": 0.3941, + "step": 5074 + }, + { + "epoch": 0.2835433136855044, + "grad_norm": 0.6216470003128052, + "learning_rate": 0.0008609648139847602, + "loss": 0.4674, + "step": 5075 + }, + { + "epoch": 0.28359918428918623, + "grad_norm": 0.8534097075462341, + "learning_rate": 0.0008609367996414164, + "loss": 0.4959, + "step": 5076 + }, + { + "epoch": 0.28365505489286813, + "grad_norm": 0.5903803706169128, + "learning_rate": 0.0008609087852980726, + "loss": 0.5708, + "step": 5077 + }, + { + "epoch": 0.28371092549654997, + "grad_norm": 0.5069766640663147, + "learning_rate": 0.0008608807709547288, + "loss": 0.4849, + "step": 5078 + }, + { + "epoch": 0.28376679610023187, + "grad_norm": 0.4698599576950073, + "learning_rate": 0.000860852756611385, + "loss": 0.346, + "step": 5079 + }, + { + "epoch": 0.28382266670391376, + "grad_norm": 0.7980721592903137, + "learning_rate": 0.0008608247422680414, + "loss": 0.4495, + "step": 5080 + }, + { + "epoch": 0.2838785373075956, + "grad_norm": 0.6971267461776733, + "learning_rate": 0.0008607967279246975, + "loss": 0.5058, + "step": 5081 + }, + { + "epoch": 0.2839344079112775, + "grad_norm": 0.9363576173782349, + "learning_rate": 0.0008607687135813537, + "loss": 0.52, + "step": 5082 + }, + { + "epoch": 0.28399027851495934, + "grad_norm": 0.4060189425945282, + "learning_rate": 0.0008607406992380099, + "loss": 0.4477, + "step": 5083 + }, + { + "epoch": 0.28404614911864123, + "grad_norm": 0.5335767269134521, + "learning_rate": 0.0008607126848946661, + "loss": 0.6607, + "step": 5084 + }, + { + "epoch": 0.2841020197223231, + "grad_norm": 1.3191856145858765, + "learning_rate": 0.0008606846705513223, + "loss": 0.7115, + "step": 5085 + }, + { + "epoch": 0.28415789032600497, + "grad_norm": 0.5459403991699219, + "learning_rate": 0.0008606566562079785, + "loss": 0.339, + "step": 5086 + }, + { + "epoch": 0.28421376092968686, + "grad_norm": 0.5574341416358948, + "learning_rate": 0.0008606286418646348, + "loss": 0.349, + "step": 5087 + }, + { + "epoch": 0.2842696315333687, + "grad_norm": 1.4570033550262451, + "learning_rate": 0.0008606006275212909, + "loss": 0.4576, + "step": 5088 + }, + { + "epoch": 0.2843255021370506, + "grad_norm": 2.2694790363311768, + "learning_rate": 0.0008605726131779471, + "loss": 0.3537, + "step": 5089 + }, + { + "epoch": 0.28438137274073244, + "grad_norm": 0.6529256105422974, + "learning_rate": 0.0008605445988346033, + "loss": 0.4832, + "step": 5090 + }, + { + "epoch": 0.28443724334441434, + "grad_norm": 0.7416061162948608, + "learning_rate": 0.0008605165844912596, + "loss": 0.4614, + "step": 5091 + }, + { + "epoch": 0.28449311394809623, + "grad_norm": 0.48411810398101807, + "learning_rate": 0.0008604885701479157, + "loss": 0.3751, + "step": 5092 + }, + { + "epoch": 0.2845489845517781, + "grad_norm": 1.0309022665023804, + "learning_rate": 0.0008604605558045719, + "loss": 0.4967, + "step": 5093 + }, + { + "epoch": 0.28460485515545997, + "grad_norm": 0.41780856251716614, + "learning_rate": 0.0008604325414612281, + "loss": 0.3314, + "step": 5094 + }, + { + "epoch": 0.2846607257591418, + "grad_norm": 0.4191058278083801, + "learning_rate": 0.0008604045271178844, + "loss": 0.4116, + "step": 5095 + }, + { + "epoch": 0.2847165963628237, + "grad_norm": 0.4674198627471924, + "learning_rate": 0.0008603765127745405, + "loss": 0.4408, + "step": 5096 + }, + { + "epoch": 0.2847724669665056, + "grad_norm": 2.6929423809051514, + "learning_rate": 0.0008603484984311967, + "loss": 0.5618, + "step": 5097 + }, + { + "epoch": 0.28482833757018744, + "grad_norm": 0.5997753739356995, + "learning_rate": 0.0008603204840878531, + "loss": 0.5762, + "step": 5098 + }, + { + "epoch": 0.28488420817386934, + "grad_norm": 0.4394136965274811, + "learning_rate": 0.0008602924697445093, + "loss": 0.5105, + "step": 5099 + }, + { + "epoch": 0.2849400787775512, + "grad_norm": 2.442070245742798, + "learning_rate": 0.0008602644554011654, + "loss": 0.5031, + "step": 5100 + }, + { + "epoch": 0.28499594938123307, + "grad_norm": 0.4321898818016052, + "learning_rate": 0.0008602364410578216, + "loss": 0.4235, + "step": 5101 + }, + { + "epoch": 0.2850518199849149, + "grad_norm": 0.5619220733642578, + "learning_rate": 0.0008602084267144779, + "loss": 0.375, + "step": 5102 + }, + { + "epoch": 0.2851076905885968, + "grad_norm": 0.5887995362281799, + "learning_rate": 0.0008601804123711341, + "loss": 0.4497, + "step": 5103 + }, + { + "epoch": 0.2851635611922787, + "grad_norm": 1.116566777229309, + "learning_rate": 0.0008601523980277902, + "loss": 0.5919, + "step": 5104 + }, + { + "epoch": 0.28521943179596054, + "grad_norm": 0.43820908665657043, + "learning_rate": 0.0008601243836844465, + "loss": 0.3856, + "step": 5105 + }, + { + "epoch": 0.28527530239964244, + "grad_norm": 0.4309489130973816, + "learning_rate": 0.0008600963693411027, + "loss": 0.3996, + "step": 5106 + }, + { + "epoch": 0.2853311730033243, + "grad_norm": 2.3346338272094727, + "learning_rate": 0.0008600683549977588, + "loss": 0.4506, + "step": 5107 + }, + { + "epoch": 0.2853870436070062, + "grad_norm": 0.5066880583763123, + "learning_rate": 0.000860040340654415, + "loss": 0.3952, + "step": 5108 + }, + { + "epoch": 0.28544291421068807, + "grad_norm": 1.6543737649917603, + "learning_rate": 0.0008600123263110713, + "loss": 0.4284, + "step": 5109 + }, + { + "epoch": 0.2854987848143699, + "grad_norm": 0.5023285150527954, + "learning_rate": 0.0008599843119677275, + "loss": 0.4319, + "step": 5110 + }, + { + "epoch": 0.2855546554180518, + "grad_norm": 0.6296706199645996, + "learning_rate": 0.0008599562976243836, + "loss": 0.6092, + "step": 5111 + }, + { + "epoch": 0.28561052602173365, + "grad_norm": 0.959672749042511, + "learning_rate": 0.0008599282832810399, + "loss": 0.4329, + "step": 5112 + }, + { + "epoch": 0.28566639662541554, + "grad_norm": 0.753082811832428, + "learning_rate": 0.0008599002689376961, + "loss": 0.425, + "step": 5113 + }, + { + "epoch": 0.28572226722909744, + "grad_norm": 0.5708947777748108, + "learning_rate": 0.0008598722545943524, + "loss": 0.5918, + "step": 5114 + }, + { + "epoch": 0.2857781378327793, + "grad_norm": 0.435630202293396, + "learning_rate": 0.0008598442402510084, + "loss": 0.4892, + "step": 5115 + }, + { + "epoch": 0.2858340084364612, + "grad_norm": 0.7615248560905457, + "learning_rate": 0.0008598162259076648, + "loss": 0.4987, + "step": 5116 + }, + { + "epoch": 0.285889879040143, + "grad_norm": 0.5200396180152893, + "learning_rate": 0.000859788211564321, + "loss": 0.4896, + "step": 5117 + }, + { + "epoch": 0.2859457496438249, + "grad_norm": 2.1619670391082764, + "learning_rate": 0.0008597601972209772, + "loss": 0.5848, + "step": 5118 + }, + { + "epoch": 0.28600162024750675, + "grad_norm": 0.4210617244243622, + "learning_rate": 0.0008597321828776334, + "loss": 0.5376, + "step": 5119 + }, + { + "epoch": 0.28605749085118864, + "grad_norm": 0.43604776263237, + "learning_rate": 0.0008597041685342896, + "loss": 0.4496, + "step": 5120 + }, + { + "epoch": 0.28611336145487054, + "grad_norm": 0.7282569408416748, + "learning_rate": 0.0008596761541909458, + "loss": 0.451, + "step": 5121 + }, + { + "epoch": 0.2861692320585524, + "grad_norm": 1.4295270442962646, + "learning_rate": 0.000859648139847602, + "loss": 0.4705, + "step": 5122 + }, + { + "epoch": 0.2862251026622343, + "grad_norm": 0.5452918410301208, + "learning_rate": 0.0008596201255042582, + "loss": 0.5072, + "step": 5123 + }, + { + "epoch": 0.2862809732659161, + "grad_norm": 0.6320980191230774, + "learning_rate": 0.0008595921111609144, + "loss": 0.4901, + "step": 5124 + }, + { + "epoch": 0.286336843869598, + "grad_norm": 0.9969284534454346, + "learning_rate": 0.0008595640968175706, + "loss": 0.409, + "step": 5125 + }, + { + "epoch": 0.2863927144732799, + "grad_norm": 0.40059348940849304, + "learning_rate": 0.0008595360824742269, + "loss": 0.5231, + "step": 5126 + }, + { + "epoch": 0.28644858507696175, + "grad_norm": 0.5144661068916321, + "learning_rate": 0.000859508068130883, + "loss": 0.4879, + "step": 5127 + }, + { + "epoch": 0.28650445568064364, + "grad_norm": 0.7929937243461609, + "learning_rate": 0.0008594800537875392, + "loss": 0.446, + "step": 5128 + }, + { + "epoch": 0.2865603262843255, + "grad_norm": 0.5264194011688232, + "learning_rate": 0.0008594520394441954, + "loss": 0.4425, + "step": 5129 + }, + { + "epoch": 0.2866161968880074, + "grad_norm": 0.5411107540130615, + "learning_rate": 0.0008594240251008516, + "loss": 0.5423, + "step": 5130 + }, + { + "epoch": 0.2866720674916893, + "grad_norm": 2.099186658859253, + "learning_rate": 0.0008593960107575078, + "loss": 0.5239, + "step": 5131 + }, + { + "epoch": 0.2867279380953711, + "grad_norm": 0.49832189083099365, + "learning_rate": 0.0008593679964141641, + "loss": 0.4176, + "step": 5132 + }, + { + "epoch": 0.286783808699053, + "grad_norm": 0.4782625138759613, + "learning_rate": 0.0008593399820708203, + "loss": 0.5057, + "step": 5133 + }, + { + "epoch": 0.28683967930273485, + "grad_norm": 0.5767653584480286, + "learning_rate": 0.0008593119677274765, + "loss": 0.5549, + "step": 5134 + }, + { + "epoch": 0.28689554990641675, + "grad_norm": 0.5049887895584106, + "learning_rate": 0.0008592839533841327, + "loss": 0.4352, + "step": 5135 + }, + { + "epoch": 0.2869514205100986, + "grad_norm": 0.7541242837905884, + "learning_rate": 0.0008592559390407889, + "loss": 0.4587, + "step": 5136 + }, + { + "epoch": 0.2870072911137805, + "grad_norm": 0.7345057725906372, + "learning_rate": 0.0008592279246974452, + "loss": 0.4499, + "step": 5137 + }, + { + "epoch": 0.2870631617174624, + "grad_norm": 0.8523752093315125, + "learning_rate": 0.0008591999103541013, + "loss": 0.4681, + "step": 5138 + }, + { + "epoch": 0.2871190323211442, + "grad_norm": 0.40026456117630005, + "learning_rate": 0.0008591718960107575, + "loss": 0.4431, + "step": 5139 + }, + { + "epoch": 0.2871749029248261, + "grad_norm": 0.45606839656829834, + "learning_rate": 0.0008591438816674137, + "loss": 0.4258, + "step": 5140 + }, + { + "epoch": 0.28723077352850795, + "grad_norm": 0.6231154203414917, + "learning_rate": 0.00085911586732407, + "loss": 0.4523, + "step": 5141 + }, + { + "epoch": 0.28728664413218985, + "grad_norm": 0.7361146807670593, + "learning_rate": 0.0008590878529807261, + "loss": 0.4525, + "step": 5142 + }, + { + "epoch": 0.28734251473587175, + "grad_norm": 0.5148532390594482, + "learning_rate": 0.0008590598386373823, + "loss": 0.5446, + "step": 5143 + }, + { + "epoch": 0.2873983853395536, + "grad_norm": 0.475721150636673, + "learning_rate": 0.0008590318242940386, + "loss": 0.4387, + "step": 5144 + }, + { + "epoch": 0.2874542559432355, + "grad_norm": 0.8652244806289673, + "learning_rate": 0.0008590038099506948, + "loss": 0.3961, + "step": 5145 + }, + { + "epoch": 0.2875101265469173, + "grad_norm": 0.45319199562072754, + "learning_rate": 0.0008589757956073509, + "loss": 0.4405, + "step": 5146 + }, + { + "epoch": 0.2875659971505992, + "grad_norm": 2.067124128341675, + "learning_rate": 0.0008589477812640071, + "loss": 0.465, + "step": 5147 + }, + { + "epoch": 0.28762186775428106, + "grad_norm": 0.7132276296615601, + "learning_rate": 0.0008589197669206635, + "loss": 0.5475, + "step": 5148 + }, + { + "epoch": 0.28767773835796295, + "grad_norm": 1.0015418529510498, + "learning_rate": 0.0008588917525773196, + "loss": 0.4066, + "step": 5149 + }, + { + "epoch": 0.28773360896164485, + "grad_norm": 0.47315195202827454, + "learning_rate": 0.0008588637382339758, + "loss": 0.3996, + "step": 5150 + }, + { + "epoch": 0.2877894795653267, + "grad_norm": 0.43202054500579834, + "learning_rate": 0.0008588357238906321, + "loss": 0.3122, + "step": 5151 + }, + { + "epoch": 0.2878453501690086, + "grad_norm": 0.9747262597084045, + "learning_rate": 0.0008588077095472883, + "loss": 0.4654, + "step": 5152 + }, + { + "epoch": 0.2879012207726904, + "grad_norm": 0.48219799995422363, + "learning_rate": 0.0008587796952039444, + "loss": 0.4108, + "step": 5153 + }, + { + "epoch": 0.2879570913763723, + "grad_norm": 0.42748692631721497, + "learning_rate": 0.0008587516808606006, + "loss": 0.4295, + "step": 5154 + }, + { + "epoch": 0.2880129619800542, + "grad_norm": 0.9712622165679932, + "learning_rate": 0.0008587236665172569, + "loss": 0.62, + "step": 5155 + }, + { + "epoch": 0.28806883258373606, + "grad_norm": 0.7765264511108398, + "learning_rate": 0.0008586956521739131, + "loss": 0.5532, + "step": 5156 + }, + { + "epoch": 0.28812470318741795, + "grad_norm": 0.493882417678833, + "learning_rate": 0.0008586676378305692, + "loss": 0.5354, + "step": 5157 + }, + { + "epoch": 0.2881805737910998, + "grad_norm": 2.2065958976745605, + "learning_rate": 0.0008586396234872255, + "loss": 0.5035, + "step": 5158 + }, + { + "epoch": 0.2882364443947817, + "grad_norm": 0.532984733581543, + "learning_rate": 0.0008586116091438817, + "loss": 0.6294, + "step": 5159 + }, + { + "epoch": 0.2882923149984636, + "grad_norm": 0.9140126705169678, + "learning_rate": 0.0008585835948005379, + "loss": 0.669, + "step": 5160 + }, + { + "epoch": 0.2883481856021454, + "grad_norm": 0.37814292311668396, + "learning_rate": 0.000858555580457194, + "loss": 0.3092, + "step": 5161 + }, + { + "epoch": 0.2884040562058273, + "grad_norm": 0.6134832501411438, + "learning_rate": 0.0008585275661138503, + "loss": 0.4055, + "step": 5162 + }, + { + "epoch": 0.28845992680950916, + "grad_norm": 0.4940750300884247, + "learning_rate": 0.0008584995517705065, + "loss": 0.4424, + "step": 5163 + }, + { + "epoch": 0.28851579741319106, + "grad_norm": 0.5522485375404358, + "learning_rate": 0.0008584715374271628, + "loss": 0.4408, + "step": 5164 + }, + { + "epoch": 0.2885716680168729, + "grad_norm": 0.47439441084861755, + "learning_rate": 0.0008584435230838188, + "loss": 0.3394, + "step": 5165 + }, + { + "epoch": 0.2886275386205548, + "grad_norm": 0.7996551990509033, + "learning_rate": 0.0008584155087404752, + "loss": 0.5588, + "step": 5166 + }, + { + "epoch": 0.2886834092242367, + "grad_norm": 0.4808255434036255, + "learning_rate": 0.0008583874943971314, + "loss": 0.4476, + "step": 5167 + }, + { + "epoch": 0.2887392798279185, + "grad_norm": 1.0434166193008423, + "learning_rate": 0.0008583594800537876, + "loss": 0.5045, + "step": 5168 + }, + { + "epoch": 0.2887951504316004, + "grad_norm": 0.5984846949577332, + "learning_rate": 0.0008583314657104438, + "loss": 0.3756, + "step": 5169 + }, + { + "epoch": 0.28885102103528226, + "grad_norm": 0.7431084513664246, + "learning_rate": 0.0008583034513671, + "loss": 0.4671, + "step": 5170 + }, + { + "epoch": 0.28890689163896416, + "grad_norm": 0.6003817915916443, + "learning_rate": 0.0008582754370237562, + "loss": 0.4033, + "step": 5171 + }, + { + "epoch": 0.28896276224264605, + "grad_norm": 0.8317606449127197, + "learning_rate": 0.0008582474226804123, + "loss": 0.561, + "step": 5172 + }, + { + "epoch": 0.2890186328463279, + "grad_norm": 0.4678092896938324, + "learning_rate": 0.0008582194083370686, + "loss": 0.3987, + "step": 5173 + }, + { + "epoch": 0.2890745034500098, + "grad_norm": 0.6951267123222351, + "learning_rate": 0.0008581913939937248, + "loss": 0.485, + "step": 5174 + }, + { + "epoch": 0.28913037405369163, + "grad_norm": 0.4330856502056122, + "learning_rate": 0.000858163379650381, + "loss": 0.3957, + "step": 5175 + }, + { + "epoch": 0.2891862446573735, + "grad_norm": 0.3409773111343384, + "learning_rate": 0.0008581353653070372, + "loss": 0.4047, + "step": 5176 + }, + { + "epoch": 0.2892421152610554, + "grad_norm": 0.5107835531234741, + "learning_rate": 0.0008581073509636934, + "loss": 0.4821, + "step": 5177 + }, + { + "epoch": 0.28929798586473726, + "grad_norm": 0.4871034026145935, + "learning_rate": 0.0008580793366203496, + "loss": 0.386, + "step": 5178 + }, + { + "epoch": 0.28935385646841916, + "grad_norm": 0.536902666091919, + "learning_rate": 0.0008580513222770058, + "loss": 0.4415, + "step": 5179 + }, + { + "epoch": 0.289409727072101, + "grad_norm": 0.37378138303756714, + "learning_rate": 0.000858023307933662, + "loss": 0.375, + "step": 5180 + }, + { + "epoch": 0.2894655976757829, + "grad_norm": 0.5859538316726685, + "learning_rate": 0.0008579952935903182, + "loss": 0.4089, + "step": 5181 + }, + { + "epoch": 0.28952146827946473, + "grad_norm": 0.6631330847740173, + "learning_rate": 0.0008579672792469745, + "loss": 0.4947, + "step": 5182 + }, + { + "epoch": 0.28957733888314663, + "grad_norm": 0.6698817014694214, + "learning_rate": 0.0008579392649036308, + "loss": 0.4648, + "step": 5183 + }, + { + "epoch": 0.2896332094868285, + "grad_norm": 0.5039183497428894, + "learning_rate": 0.0008579112505602869, + "loss": 0.4054, + "step": 5184 + }, + { + "epoch": 0.28968908009051036, + "grad_norm": 2.972846269607544, + "learning_rate": 0.0008578832362169431, + "loss": 0.3874, + "step": 5185 + }, + { + "epoch": 0.28974495069419226, + "grad_norm": 0.5175074934959412, + "learning_rate": 0.0008578552218735993, + "loss": 0.4929, + "step": 5186 + }, + { + "epoch": 0.2898008212978741, + "grad_norm": 0.952195405960083, + "learning_rate": 0.0008578272075302556, + "loss": 0.5321, + "step": 5187 + }, + { + "epoch": 0.289856691901556, + "grad_norm": 0.9069812297821045, + "learning_rate": 0.0008577991931869117, + "loss": 0.4799, + "step": 5188 + }, + { + "epoch": 0.2899125625052379, + "grad_norm": 0.4598220884799957, + "learning_rate": 0.0008577711788435679, + "loss": 0.4247, + "step": 5189 + }, + { + "epoch": 0.28996843310891973, + "grad_norm": 0.6881436705589294, + "learning_rate": 0.0008577431645002242, + "loss": 0.6624, + "step": 5190 + }, + { + "epoch": 0.2900243037126016, + "grad_norm": 0.4898996651172638, + "learning_rate": 0.0008577151501568803, + "loss": 0.5013, + "step": 5191 + }, + { + "epoch": 0.29008017431628347, + "grad_norm": 0.5728178024291992, + "learning_rate": 0.0008576871358135365, + "loss": 0.5067, + "step": 5192 + }, + { + "epoch": 0.29013604491996536, + "grad_norm": 0.41239744424819946, + "learning_rate": 0.0008576591214701927, + "loss": 0.3894, + "step": 5193 + }, + { + "epoch": 0.29019191552364726, + "grad_norm": 0.6647194623947144, + "learning_rate": 0.000857631107126849, + "loss": 0.3836, + "step": 5194 + }, + { + "epoch": 0.2902477861273291, + "grad_norm": 0.7594090104103088, + "learning_rate": 0.0008576030927835051, + "loss": 0.5128, + "step": 5195 + }, + { + "epoch": 0.290303656731011, + "grad_norm": 0.5641520023345947, + "learning_rate": 0.0008575750784401613, + "loss": 0.4695, + "step": 5196 + }, + { + "epoch": 0.29035952733469284, + "grad_norm": 0.648765504360199, + "learning_rate": 0.0008575470640968177, + "loss": 0.4711, + "step": 5197 + }, + { + "epoch": 0.29041539793837473, + "grad_norm": 0.896328866481781, + "learning_rate": 0.0008575190497534739, + "loss": 0.3839, + "step": 5198 + }, + { + "epoch": 0.29047126854205657, + "grad_norm": 0.7045734524726868, + "learning_rate": 0.00085749103541013, + "loss": 0.4511, + "step": 5199 + }, + { + "epoch": 0.29052713914573847, + "grad_norm": 0.5737088918685913, + "learning_rate": 0.0008574630210667862, + "loss": 0.4384, + "step": 5200 + }, + { + "epoch": 0.29058300974942036, + "grad_norm": 0.5456222891807556, + "learning_rate": 0.0008574350067234425, + "loss": 0.4877, + "step": 5201 + }, + { + "epoch": 0.2906388803531022, + "grad_norm": 0.8107205033302307, + "learning_rate": 0.0008574069923800987, + "loss": 0.5629, + "step": 5202 + }, + { + "epoch": 0.2906947509567841, + "grad_norm": 1.324686050415039, + "learning_rate": 0.0008573789780367548, + "loss": 0.4623, + "step": 5203 + }, + { + "epoch": 0.29075062156046594, + "grad_norm": 0.473239541053772, + "learning_rate": 0.000857350963693411, + "loss": 0.424, + "step": 5204 + }, + { + "epoch": 0.29080649216414783, + "grad_norm": 0.7552129626274109, + "learning_rate": 0.0008573229493500673, + "loss": 0.4015, + "step": 5205 + }, + { + "epoch": 0.29086236276782973, + "grad_norm": 0.8216189742088318, + "learning_rate": 0.0008572949350067235, + "loss": 0.6043, + "step": 5206 + }, + { + "epoch": 0.29091823337151157, + "grad_norm": 1.2283083200454712, + "learning_rate": 0.0008572669206633796, + "loss": 0.4782, + "step": 5207 + }, + { + "epoch": 0.29097410397519347, + "grad_norm": 0.9336365461349487, + "learning_rate": 0.0008572389063200359, + "loss": 0.5208, + "step": 5208 + }, + { + "epoch": 0.2910299745788753, + "grad_norm": 0.7877665758132935, + "learning_rate": 0.0008572108919766921, + "loss": 0.7808, + "step": 5209 + }, + { + "epoch": 0.2910858451825572, + "grad_norm": 0.5314787030220032, + "learning_rate": 0.0008571828776333483, + "loss": 0.4039, + "step": 5210 + }, + { + "epoch": 0.2911417157862391, + "grad_norm": 0.6445666551589966, + "learning_rate": 0.0008571548632900044, + "loss": 0.3924, + "step": 5211 + }, + { + "epoch": 0.29119758638992094, + "grad_norm": 0.4650024473667145, + "learning_rate": 0.0008571268489466607, + "loss": 0.4185, + "step": 5212 + }, + { + "epoch": 0.29125345699360283, + "grad_norm": 0.5741336941719055, + "learning_rate": 0.0008570988346033169, + "loss": 0.5158, + "step": 5213 + }, + { + "epoch": 0.2913093275972847, + "grad_norm": 0.5563361644744873, + "learning_rate": 0.000857070820259973, + "loss": 0.3706, + "step": 5214 + }, + { + "epoch": 0.29136519820096657, + "grad_norm": 0.6563366651535034, + "learning_rate": 0.0008570428059166294, + "loss": 0.497, + "step": 5215 + }, + { + "epoch": 0.2914210688046484, + "grad_norm": 0.7086651921272278, + "learning_rate": 0.0008570147915732856, + "loss": 0.4316, + "step": 5216 + }, + { + "epoch": 0.2914769394083303, + "grad_norm": 0.5208603143692017, + "learning_rate": 0.0008569867772299418, + "loss": 0.4511, + "step": 5217 + }, + { + "epoch": 0.2915328100120122, + "grad_norm": 0.4251033067703247, + "learning_rate": 0.0008569587628865979, + "loss": 0.4415, + "step": 5218 + }, + { + "epoch": 0.29158868061569404, + "grad_norm": 0.6054595112800598, + "learning_rate": 0.0008569307485432542, + "loss": 0.6555, + "step": 5219 + }, + { + "epoch": 0.29164455121937594, + "grad_norm": 0.8300573825836182, + "learning_rate": 0.0008569027341999104, + "loss": 0.4355, + "step": 5220 + }, + { + "epoch": 0.2917004218230578, + "grad_norm": 0.6783788204193115, + "learning_rate": 0.0008568747198565666, + "loss": 0.5079, + "step": 5221 + }, + { + "epoch": 0.29175629242673967, + "grad_norm": 0.6132943630218506, + "learning_rate": 0.0008568467055132228, + "loss": 0.3737, + "step": 5222 + }, + { + "epoch": 0.29181216303042157, + "grad_norm": 1.6471081972122192, + "learning_rate": 0.000856818691169879, + "loss": 0.5553, + "step": 5223 + }, + { + "epoch": 0.2918680336341034, + "grad_norm": 0.47969570755958557, + "learning_rate": 0.0008567906768265352, + "loss": 0.4473, + "step": 5224 + }, + { + "epoch": 0.2919239042377853, + "grad_norm": 0.5476240515708923, + "learning_rate": 0.0008567626624831914, + "loss": 0.446, + "step": 5225 + }, + { + "epoch": 0.29197977484146714, + "grad_norm": 0.49233880639076233, + "learning_rate": 0.0008567346481398476, + "loss": 0.4776, + "step": 5226 + }, + { + "epoch": 0.29203564544514904, + "grad_norm": 1.6070255041122437, + "learning_rate": 0.0008567066337965038, + "loss": 0.4313, + "step": 5227 + }, + { + "epoch": 0.29209151604883093, + "grad_norm": 0.4554293155670166, + "learning_rate": 0.00085667861945316, + "loss": 0.5372, + "step": 5228 + }, + { + "epoch": 0.2921473866525128, + "grad_norm": 1.4059510231018066, + "learning_rate": 0.0008566506051098163, + "loss": 0.5045, + "step": 5229 + }, + { + "epoch": 0.29220325725619467, + "grad_norm": 0.5737929940223694, + "learning_rate": 0.0008566225907664724, + "loss": 0.4033, + "step": 5230 + }, + { + "epoch": 0.2922591278598765, + "grad_norm": 0.6567088961601257, + "learning_rate": 0.0008565945764231286, + "loss": 0.5669, + "step": 5231 + }, + { + "epoch": 0.2923149984635584, + "grad_norm": 1.3200700283050537, + "learning_rate": 0.0008565665620797849, + "loss": 0.4043, + "step": 5232 + }, + { + "epoch": 0.29237086906724025, + "grad_norm": 0.8189688324928284, + "learning_rate": 0.0008565385477364411, + "loss": 0.4817, + "step": 5233 + }, + { + "epoch": 0.29242673967092214, + "grad_norm": 1.194512128829956, + "learning_rate": 0.0008565105333930973, + "loss": 0.4303, + "step": 5234 + }, + { + "epoch": 0.29248261027460404, + "grad_norm": 0.5779362320899963, + "learning_rate": 0.0008564825190497535, + "loss": 0.4689, + "step": 5235 + }, + { + "epoch": 0.2925384808782859, + "grad_norm": 0.4842422306537628, + "learning_rate": 0.0008564545047064098, + "loss": 0.4777, + "step": 5236 + }, + { + "epoch": 0.2925943514819678, + "grad_norm": 0.7960717082023621, + "learning_rate": 0.0008564264903630659, + "loss": 0.6312, + "step": 5237 + }, + { + "epoch": 0.2926502220856496, + "grad_norm": 0.5011351704597473, + "learning_rate": 0.0008563984760197221, + "loss": 0.4941, + "step": 5238 + }, + { + "epoch": 0.2927060926893315, + "grad_norm": 0.6191487908363342, + "learning_rate": 0.0008563704616763783, + "loss": 0.4457, + "step": 5239 + }, + { + "epoch": 0.2927619632930134, + "grad_norm": 0.5990877151489258, + "learning_rate": 0.0008563424473330346, + "loss": 0.7194, + "step": 5240 + }, + { + "epoch": 0.29281783389669525, + "grad_norm": 0.48869988322257996, + "learning_rate": 0.0008563144329896907, + "loss": 0.5662, + "step": 5241 + }, + { + "epoch": 0.29287370450037714, + "grad_norm": 0.5712915062904358, + "learning_rate": 0.0008562864186463469, + "loss": 0.4584, + "step": 5242 + }, + { + "epoch": 0.292929575104059, + "grad_norm": 1.9412747621536255, + "learning_rate": 0.0008562584043030031, + "loss": 0.4815, + "step": 5243 + }, + { + "epoch": 0.2929854457077409, + "grad_norm": 0.6045790910720825, + "learning_rate": 0.0008562303899596594, + "loss": 0.4333, + "step": 5244 + }, + { + "epoch": 0.2930413163114228, + "grad_norm": 0.4629065990447998, + "learning_rate": 0.0008562023756163155, + "loss": 0.3948, + "step": 5245 + }, + { + "epoch": 0.2930971869151046, + "grad_norm": 0.48573926091194153, + "learning_rate": 0.0008561743612729717, + "loss": 0.4323, + "step": 5246 + }, + { + "epoch": 0.2931530575187865, + "grad_norm": 0.8892102837562561, + "learning_rate": 0.000856146346929628, + "loss": 0.4101, + "step": 5247 + }, + { + "epoch": 0.29320892812246835, + "grad_norm": 0.6316726207733154, + "learning_rate": 0.0008561183325862843, + "loss": 0.5593, + "step": 5248 + }, + { + "epoch": 0.29326479872615024, + "grad_norm": 2.861544609069824, + "learning_rate": 0.0008560903182429404, + "loss": 0.5111, + "step": 5249 + }, + { + "epoch": 0.2933206693298321, + "grad_norm": 4.144753456115723, + "learning_rate": 0.0008560623038995966, + "loss": 0.4035, + "step": 5250 + }, + { + "epoch": 0.293376539933514, + "grad_norm": 0.5897698998451233, + "learning_rate": 0.0008560342895562529, + "loss": 0.4751, + "step": 5251 + }, + { + "epoch": 0.2934324105371959, + "grad_norm": 0.4294137954711914, + "learning_rate": 0.0008560062752129091, + "loss": 0.3896, + "step": 5252 + }, + { + "epoch": 0.2934882811408777, + "grad_norm": 0.8198501467704773, + "learning_rate": 0.0008559782608695652, + "loss": 0.4865, + "step": 5253 + }, + { + "epoch": 0.2935441517445596, + "grad_norm": 0.4218595027923584, + "learning_rate": 0.0008559502465262215, + "loss": 0.46, + "step": 5254 + }, + { + "epoch": 0.29360002234824145, + "grad_norm": 0.4902544617652893, + "learning_rate": 0.0008559222321828777, + "loss": 0.4735, + "step": 5255 + }, + { + "epoch": 0.29365589295192335, + "grad_norm": 0.6466330289840698, + "learning_rate": 0.0008558942178395338, + "loss": 0.5437, + "step": 5256 + }, + { + "epoch": 0.29371176355560524, + "grad_norm": 0.7908904552459717, + "learning_rate": 0.00085586620349619, + "loss": 0.5747, + "step": 5257 + }, + { + "epoch": 0.2937676341592871, + "grad_norm": 0.38620877265930176, + "learning_rate": 0.0008558381891528463, + "loss": 0.4581, + "step": 5258 + }, + { + "epoch": 0.293823504762969, + "grad_norm": 0.42276978492736816, + "learning_rate": 0.0008558101748095025, + "loss": 0.5323, + "step": 5259 + }, + { + "epoch": 0.2938793753666508, + "grad_norm": 0.4575797915458679, + "learning_rate": 0.0008557821604661586, + "loss": 0.4789, + "step": 5260 + }, + { + "epoch": 0.2939352459703327, + "grad_norm": 0.4116096794605255, + "learning_rate": 0.0008557541461228149, + "loss": 0.425, + "step": 5261 + }, + { + "epoch": 0.29399111657401455, + "grad_norm": 0.7457348704338074, + "learning_rate": 0.0008557261317794711, + "loss": 0.7411, + "step": 5262 + }, + { + "epoch": 0.29404698717769645, + "grad_norm": 1.0529967546463013, + "learning_rate": 0.0008556981174361273, + "loss": 0.4293, + "step": 5263 + }, + { + "epoch": 0.29410285778137835, + "grad_norm": 0.5143582820892334, + "learning_rate": 0.0008556701030927834, + "loss": 0.4186, + "step": 5264 + }, + { + "epoch": 0.2941587283850602, + "grad_norm": 0.4400825500488281, + "learning_rate": 0.0008556420887494398, + "loss": 0.3579, + "step": 5265 + }, + { + "epoch": 0.2942145989887421, + "grad_norm": 1.3242347240447998, + "learning_rate": 0.000855614074406096, + "loss": 0.3634, + "step": 5266 + }, + { + "epoch": 0.2942704695924239, + "grad_norm": 0.5448470115661621, + "learning_rate": 0.0008555860600627522, + "loss": 0.6011, + "step": 5267 + }, + { + "epoch": 0.2943263401961058, + "grad_norm": 0.37621331214904785, + "learning_rate": 0.0008555580457194084, + "loss": 0.4315, + "step": 5268 + }, + { + "epoch": 0.2943822107997877, + "grad_norm": 0.4608527719974518, + "learning_rate": 0.0008555300313760646, + "loss": 0.4598, + "step": 5269 + }, + { + "epoch": 0.29443808140346955, + "grad_norm": 0.4513334035873413, + "learning_rate": 0.0008555020170327208, + "loss": 0.4894, + "step": 5270 + }, + { + "epoch": 0.29449395200715145, + "grad_norm": 0.6755939722061157, + "learning_rate": 0.000855474002689377, + "loss": 0.5793, + "step": 5271 + }, + { + "epoch": 0.2945498226108333, + "grad_norm": 0.7949337363243103, + "learning_rate": 0.0008554459883460332, + "loss": 0.4537, + "step": 5272 + }, + { + "epoch": 0.2946056932145152, + "grad_norm": 0.6416296362876892, + "learning_rate": 0.0008554179740026894, + "loss": 0.5314, + "step": 5273 + }, + { + "epoch": 0.2946615638181971, + "grad_norm": 0.592043936252594, + "learning_rate": 0.0008553899596593456, + "loss": 0.411, + "step": 5274 + }, + { + "epoch": 0.2947174344218789, + "grad_norm": 0.352333128452301, + "learning_rate": 0.0008553619453160018, + "loss": 0.3864, + "step": 5275 + }, + { + "epoch": 0.2947733050255608, + "grad_norm": 0.5143254995346069, + "learning_rate": 0.000855333930972658, + "loss": 0.5129, + "step": 5276 + }, + { + "epoch": 0.29482917562924266, + "grad_norm": 0.3922172784805298, + "learning_rate": 0.0008553059166293142, + "loss": 0.3782, + "step": 5277 + }, + { + "epoch": 0.29488504623292455, + "grad_norm": 0.6009125709533691, + "learning_rate": 0.0008552779022859704, + "loss": 0.4635, + "step": 5278 + }, + { + "epoch": 0.2949409168366064, + "grad_norm": 0.5753782391548157, + "learning_rate": 0.0008552498879426266, + "loss": 0.4781, + "step": 5279 + }, + { + "epoch": 0.2949967874402883, + "grad_norm": 0.46513912081718445, + "learning_rate": 0.0008552218735992828, + "loss": 0.3487, + "step": 5280 + }, + { + "epoch": 0.2950526580439702, + "grad_norm": 0.8257052302360535, + "learning_rate": 0.000855193859255939, + "loss": 0.539, + "step": 5281 + }, + { + "epoch": 0.295108528647652, + "grad_norm": 0.9083658456802368, + "learning_rate": 0.0008551658449125953, + "loss": 0.5333, + "step": 5282 + }, + { + "epoch": 0.2951643992513339, + "grad_norm": 0.6228927969932556, + "learning_rate": 0.0008551378305692515, + "loss": 0.531, + "step": 5283 + }, + { + "epoch": 0.29522026985501576, + "grad_norm": 0.4731937646865845, + "learning_rate": 0.0008551098162259077, + "loss": 0.3869, + "step": 5284 + }, + { + "epoch": 0.29527614045869766, + "grad_norm": 0.6993194222450256, + "learning_rate": 0.0008550818018825639, + "loss": 0.4388, + "step": 5285 + }, + { + "epoch": 0.29533201106237955, + "grad_norm": 2.623518466949463, + "learning_rate": 0.0008550537875392202, + "loss": 0.4562, + "step": 5286 + }, + { + "epoch": 0.2953878816660614, + "grad_norm": 0.5517109632492065, + "learning_rate": 0.0008550257731958763, + "loss": 0.4991, + "step": 5287 + }, + { + "epoch": 0.2954437522697433, + "grad_norm": 0.6360973119735718, + "learning_rate": 0.0008549977588525325, + "loss": 0.6202, + "step": 5288 + }, + { + "epoch": 0.2954996228734251, + "grad_norm": 4.042226314544678, + "learning_rate": 0.0008549697445091887, + "loss": 0.4523, + "step": 5289 + }, + { + "epoch": 0.295555493477107, + "grad_norm": 0.5723655819892883, + "learning_rate": 0.000854941730165845, + "loss": 0.4962, + "step": 5290 + }, + { + "epoch": 0.2956113640807889, + "grad_norm": 0.5483366847038269, + "learning_rate": 0.0008549137158225011, + "loss": 0.4743, + "step": 5291 + }, + { + "epoch": 0.29566723468447076, + "grad_norm": 0.4283278286457062, + "learning_rate": 0.0008548857014791573, + "loss": 0.4718, + "step": 5292 + }, + { + "epoch": 0.29572310528815265, + "grad_norm": 0.4512265622615814, + "learning_rate": 0.0008548576871358136, + "loss": 0.4679, + "step": 5293 + }, + { + "epoch": 0.2957789758918345, + "grad_norm": 0.47476309537887573, + "learning_rate": 0.0008548296727924698, + "loss": 0.4386, + "step": 5294 + }, + { + "epoch": 0.2958348464955164, + "grad_norm": 1.2830440998077393, + "learning_rate": 0.0008548016584491259, + "loss": 0.501, + "step": 5295 + }, + { + "epoch": 0.29589071709919823, + "grad_norm": 0.7941354513168335, + "learning_rate": 0.0008547736441057821, + "loss": 0.5473, + "step": 5296 + }, + { + "epoch": 0.2959465877028801, + "grad_norm": 0.46051594614982605, + "learning_rate": 0.0008547456297624384, + "loss": 0.6034, + "step": 5297 + }, + { + "epoch": 0.296002458306562, + "grad_norm": 0.675963819026947, + "learning_rate": 0.0008547176154190945, + "loss": 0.5331, + "step": 5298 + }, + { + "epoch": 0.29605832891024386, + "grad_norm": 0.5764648914337158, + "learning_rate": 0.0008546896010757508, + "loss": 0.5987, + "step": 5299 + }, + { + "epoch": 0.29611419951392576, + "grad_norm": 0.6275779008865356, + "learning_rate": 0.0008546615867324071, + "loss": 0.4663, + "step": 5300 + }, + { + "epoch": 0.2961700701176076, + "grad_norm": 0.6299389004707336, + "learning_rate": 0.0008546335723890633, + "loss": 0.4664, + "step": 5301 + }, + { + "epoch": 0.2962259407212895, + "grad_norm": 0.543125569820404, + "learning_rate": 0.0008546055580457194, + "loss": 0.41, + "step": 5302 + }, + { + "epoch": 0.2962818113249714, + "grad_norm": 0.6403347849845886, + "learning_rate": 0.0008545775437023756, + "loss": 0.635, + "step": 5303 + }, + { + "epoch": 0.29633768192865323, + "grad_norm": 0.48012274503707886, + "learning_rate": 0.0008545495293590319, + "loss": 0.4087, + "step": 5304 + }, + { + "epoch": 0.2963935525323351, + "grad_norm": 1.021001935005188, + "learning_rate": 0.0008545215150156881, + "loss": 0.5591, + "step": 5305 + }, + { + "epoch": 0.29644942313601697, + "grad_norm": 0.5080171227455139, + "learning_rate": 0.0008544935006723442, + "loss": 0.4201, + "step": 5306 + }, + { + "epoch": 0.29650529373969886, + "grad_norm": 0.5647750496864319, + "learning_rate": 0.0008544654863290004, + "loss": 0.449, + "step": 5307 + }, + { + "epoch": 0.29656116434338076, + "grad_norm": 0.609820544719696, + "learning_rate": 0.0008544374719856567, + "loss": 0.5067, + "step": 5308 + }, + { + "epoch": 0.2966170349470626, + "grad_norm": 0.4286586344242096, + "learning_rate": 0.0008544094576423129, + "loss": 0.4383, + "step": 5309 + }, + { + "epoch": 0.2966729055507445, + "grad_norm": 1.4970097541809082, + "learning_rate": 0.000854381443298969, + "loss": 0.382, + "step": 5310 + }, + { + "epoch": 0.29672877615442633, + "grad_norm": 1.001288652420044, + "learning_rate": 0.0008543534289556253, + "loss": 0.465, + "step": 5311 + }, + { + "epoch": 0.29678464675810823, + "grad_norm": 2.0194990634918213, + "learning_rate": 0.0008543254146122815, + "loss": 0.5236, + "step": 5312 + }, + { + "epoch": 0.29684051736179007, + "grad_norm": 0.41774800419807434, + "learning_rate": 0.0008542974002689377, + "loss": 0.3668, + "step": 5313 + }, + { + "epoch": 0.29689638796547196, + "grad_norm": 0.5028137564659119, + "learning_rate": 0.0008542693859255938, + "loss": 0.4632, + "step": 5314 + }, + { + "epoch": 0.29695225856915386, + "grad_norm": 0.6111031770706177, + "learning_rate": 0.0008542413715822502, + "loss": 0.4813, + "step": 5315 + }, + { + "epoch": 0.2970081291728357, + "grad_norm": 0.8771689534187317, + "learning_rate": 0.0008542133572389064, + "loss": 0.559, + "step": 5316 + }, + { + "epoch": 0.2970639997765176, + "grad_norm": 0.7115880846977234, + "learning_rate": 0.0008541853428955626, + "loss": 0.3775, + "step": 5317 + }, + { + "epoch": 0.29711987038019944, + "grad_norm": 0.6602780222892761, + "learning_rate": 0.0008541573285522188, + "loss": 0.4904, + "step": 5318 + }, + { + "epoch": 0.29717574098388133, + "grad_norm": 0.44066429138183594, + "learning_rate": 0.000854129314208875, + "loss": 0.4803, + "step": 5319 + }, + { + "epoch": 0.2972316115875632, + "grad_norm": 0.5072330236434937, + "learning_rate": 0.0008541012998655312, + "loss": 0.4258, + "step": 5320 + }, + { + "epoch": 0.29728748219124507, + "grad_norm": 0.5969917178153992, + "learning_rate": 0.0008540732855221873, + "loss": 0.4035, + "step": 5321 + }, + { + "epoch": 0.29734335279492696, + "grad_norm": 0.6411219835281372, + "learning_rate": 0.0008540452711788436, + "loss": 0.6095, + "step": 5322 + }, + { + "epoch": 0.2973992233986088, + "grad_norm": 0.6842495203018188, + "learning_rate": 0.0008540172568354998, + "loss": 0.5208, + "step": 5323 + }, + { + "epoch": 0.2974550940022907, + "grad_norm": 0.4121076464653015, + "learning_rate": 0.000853989242492156, + "loss": 0.4303, + "step": 5324 + }, + { + "epoch": 0.2975109646059726, + "grad_norm": 0.6089886426925659, + "learning_rate": 0.0008539612281488122, + "loss": 0.4863, + "step": 5325 + }, + { + "epoch": 0.29756683520965443, + "grad_norm": 0.4883969724178314, + "learning_rate": 0.0008539332138054684, + "loss": 0.4568, + "step": 5326 + }, + { + "epoch": 0.29762270581333633, + "grad_norm": 0.47139298915863037, + "learning_rate": 0.0008539051994621246, + "loss": 0.4208, + "step": 5327 + }, + { + "epoch": 0.29767857641701817, + "grad_norm": 0.8847037553787231, + "learning_rate": 0.0008538771851187808, + "loss": 0.4645, + "step": 5328 + }, + { + "epoch": 0.29773444702070007, + "grad_norm": 0.7110500335693359, + "learning_rate": 0.000853849170775437, + "loss": 0.5193, + "step": 5329 + }, + { + "epoch": 0.2977903176243819, + "grad_norm": 0.45008566975593567, + "learning_rate": 0.0008538211564320932, + "loss": 0.479, + "step": 5330 + }, + { + "epoch": 0.2978461882280638, + "grad_norm": 0.44434022903442383, + "learning_rate": 0.0008537931420887494, + "loss": 0.4375, + "step": 5331 + }, + { + "epoch": 0.2979020588317457, + "grad_norm": 0.5495310425758362, + "learning_rate": 0.0008537651277454058, + "loss": 0.427, + "step": 5332 + }, + { + "epoch": 0.29795792943542754, + "grad_norm": 0.4081675112247467, + "learning_rate": 0.0008537371134020619, + "loss": 0.4201, + "step": 5333 + }, + { + "epoch": 0.29801380003910943, + "grad_norm": 1.0175385475158691, + "learning_rate": 0.0008537090990587181, + "loss": 0.455, + "step": 5334 + }, + { + "epoch": 0.2980696706427913, + "grad_norm": 0.3903183937072754, + "learning_rate": 0.0008536810847153743, + "loss": 0.3493, + "step": 5335 + }, + { + "epoch": 0.29812554124647317, + "grad_norm": 0.41805437207221985, + "learning_rate": 0.0008536530703720306, + "loss": 0.4193, + "step": 5336 + }, + { + "epoch": 0.29818141185015506, + "grad_norm": 0.7151769399642944, + "learning_rate": 0.0008536250560286867, + "loss": 0.4117, + "step": 5337 + }, + { + "epoch": 0.2982372824538369, + "grad_norm": 0.5958705544471741, + "learning_rate": 0.0008535970416853429, + "loss": 0.4475, + "step": 5338 + }, + { + "epoch": 0.2982931530575188, + "grad_norm": 0.6471501588821411, + "learning_rate": 0.0008535690273419992, + "loss": 0.5837, + "step": 5339 + }, + { + "epoch": 0.29834902366120064, + "grad_norm": 2.162607192993164, + "learning_rate": 0.0008535410129986553, + "loss": 0.5855, + "step": 5340 + }, + { + "epoch": 0.29840489426488254, + "grad_norm": 4.6282267570495605, + "learning_rate": 0.0008535129986553115, + "loss": 0.5262, + "step": 5341 + }, + { + "epoch": 0.29846076486856443, + "grad_norm": 2.4276576042175293, + "learning_rate": 0.0008534849843119677, + "loss": 0.4732, + "step": 5342 + }, + { + "epoch": 0.2985166354722463, + "grad_norm": 0.5532031059265137, + "learning_rate": 0.000853456969968624, + "loss": 0.4449, + "step": 5343 + }, + { + "epoch": 0.29857250607592817, + "grad_norm": 1.0967040061950684, + "learning_rate": 0.0008534289556252801, + "loss": 0.4636, + "step": 5344 + }, + { + "epoch": 0.29862837667961, + "grad_norm": 0.5266315340995789, + "learning_rate": 0.0008534009412819363, + "loss": 0.5205, + "step": 5345 + }, + { + "epoch": 0.2986842472832919, + "grad_norm": 0.5642468929290771, + "learning_rate": 0.0008533729269385925, + "loss": 0.4718, + "step": 5346 + }, + { + "epoch": 0.29874011788697374, + "grad_norm": 0.5531579256057739, + "learning_rate": 0.0008533449125952488, + "loss": 0.3899, + "step": 5347 + }, + { + "epoch": 0.29879598849065564, + "grad_norm": 0.4796006381511688, + "learning_rate": 0.0008533168982519049, + "loss": 0.4071, + "step": 5348 + }, + { + "epoch": 0.29885185909433754, + "grad_norm": 0.537839412689209, + "learning_rate": 0.0008532888839085612, + "loss": 0.4808, + "step": 5349 + }, + { + "epoch": 0.2989077296980194, + "grad_norm": 0.6825734972953796, + "learning_rate": 0.0008532608695652175, + "loss": 0.4816, + "step": 5350 + }, + { + "epoch": 0.29896360030170127, + "grad_norm": 0.5019391179084778, + "learning_rate": 0.0008532328552218737, + "loss": 0.3778, + "step": 5351 + }, + { + "epoch": 0.2990194709053831, + "grad_norm": 2.371469497680664, + "learning_rate": 0.0008532048408785298, + "loss": 0.5076, + "step": 5352 + }, + { + "epoch": 0.299075341509065, + "grad_norm": 0.4299662709236145, + "learning_rate": 0.000853176826535186, + "loss": 0.4318, + "step": 5353 + }, + { + "epoch": 0.2991312121127469, + "grad_norm": 0.5568432807922363, + "learning_rate": 0.0008531488121918423, + "loss": 0.4702, + "step": 5354 + }, + { + "epoch": 0.29918708271642874, + "grad_norm": 1.036288857460022, + "learning_rate": 0.0008531207978484985, + "loss": 0.4716, + "step": 5355 + }, + { + "epoch": 0.29924295332011064, + "grad_norm": 0.4771662950515747, + "learning_rate": 0.0008530927835051546, + "loss": 0.4623, + "step": 5356 + }, + { + "epoch": 0.2992988239237925, + "grad_norm": 0.6418251991271973, + "learning_rate": 0.0008530647691618109, + "loss": 0.5151, + "step": 5357 + }, + { + "epoch": 0.2993546945274744, + "grad_norm": 0.9430022239685059, + "learning_rate": 0.0008530367548184671, + "loss": 0.5003, + "step": 5358 + }, + { + "epoch": 0.2994105651311562, + "grad_norm": 0.6362647414207458, + "learning_rate": 0.0008530087404751233, + "loss": 0.4096, + "step": 5359 + }, + { + "epoch": 0.2994664357348381, + "grad_norm": 11.975069046020508, + "learning_rate": 0.0008529807261317794, + "loss": 0.7641, + "step": 5360 + }, + { + "epoch": 0.29952230633852, + "grad_norm": 0.7321158051490784, + "learning_rate": 0.0008529527117884357, + "loss": 0.4832, + "step": 5361 + }, + { + "epoch": 0.29957817694220185, + "grad_norm": 0.7941243052482605, + "learning_rate": 0.0008529246974450919, + "loss": 0.5645, + "step": 5362 + }, + { + "epoch": 0.29963404754588374, + "grad_norm": 0.7994101643562317, + "learning_rate": 0.000852896683101748, + "loss": 0.5157, + "step": 5363 + }, + { + "epoch": 0.2996899181495656, + "grad_norm": 2.75980281829834, + "learning_rate": 0.0008528686687584043, + "loss": 0.459, + "step": 5364 + }, + { + "epoch": 0.2997457887532475, + "grad_norm": 0.6013509631156921, + "learning_rate": 0.0008528406544150606, + "loss": 0.4442, + "step": 5365 + }, + { + "epoch": 0.2998016593569294, + "grad_norm": 2.0474040508270264, + "learning_rate": 0.0008528126400717168, + "loss": 0.5358, + "step": 5366 + }, + { + "epoch": 0.2998575299606112, + "grad_norm": 0.9430153965950012, + "learning_rate": 0.0008527846257283729, + "loss": 0.5048, + "step": 5367 + }, + { + "epoch": 0.2999134005642931, + "grad_norm": 0.7737842798233032, + "learning_rate": 0.0008527566113850292, + "loss": 0.4495, + "step": 5368 + }, + { + "epoch": 0.29996927116797495, + "grad_norm": 0.9555544853210449, + "learning_rate": 0.0008527285970416854, + "loss": 0.5435, + "step": 5369 + }, + { + "epoch": 0.30002514177165684, + "grad_norm": 0.6766431927680969, + "learning_rate": 0.0008527005826983416, + "loss": 0.4686, + "step": 5370 + }, + { + "epoch": 0.30008101237533874, + "grad_norm": 0.5441358089447021, + "learning_rate": 0.0008526725683549978, + "loss": 0.4927, + "step": 5371 + }, + { + "epoch": 0.3001368829790206, + "grad_norm": 1.0371239185333252, + "learning_rate": 0.000852644554011654, + "loss": 0.5505, + "step": 5372 + }, + { + "epoch": 0.3001927535827025, + "grad_norm": 1.732988953590393, + "learning_rate": 0.0008526165396683102, + "loss": 0.4315, + "step": 5373 + }, + { + "epoch": 0.3002486241863843, + "grad_norm": 0.9105690717697144, + "learning_rate": 0.0008525885253249664, + "loss": 0.4502, + "step": 5374 + }, + { + "epoch": 0.3003044947900662, + "grad_norm": 0.5795272588729858, + "learning_rate": 0.0008525605109816226, + "loss": 0.4778, + "step": 5375 + }, + { + "epoch": 0.30036036539374805, + "grad_norm": 0.49400749802589417, + "learning_rate": 0.0008525324966382788, + "loss": 0.4775, + "step": 5376 + }, + { + "epoch": 0.30041623599742995, + "grad_norm": 0.9107844233512878, + "learning_rate": 0.000852504482294935, + "loss": 0.4887, + "step": 5377 + }, + { + "epoch": 0.30047210660111184, + "grad_norm": 0.7486609816551208, + "learning_rate": 0.0008524764679515913, + "loss": 0.4752, + "step": 5378 + }, + { + "epoch": 0.3005279772047937, + "grad_norm": 0.9030241370201111, + "learning_rate": 0.0008524484536082474, + "loss": 0.4081, + "step": 5379 + }, + { + "epoch": 0.3005838478084756, + "grad_norm": 0.8056597709655762, + "learning_rate": 0.0008524204392649036, + "loss": 0.4478, + "step": 5380 + }, + { + "epoch": 0.3006397184121574, + "grad_norm": 0.4028540253639221, + "learning_rate": 0.0008523924249215598, + "loss": 0.4528, + "step": 5381 + }, + { + "epoch": 0.3006955890158393, + "grad_norm": 0.6983227133750916, + "learning_rate": 0.000852364410578216, + "loss": 0.4716, + "step": 5382 + }, + { + "epoch": 0.3007514596195212, + "grad_norm": 3.3386282920837402, + "learning_rate": 0.0008523363962348723, + "loss": 0.4405, + "step": 5383 + }, + { + "epoch": 0.30080733022320305, + "grad_norm": 0.7818995118141174, + "learning_rate": 0.0008523083818915285, + "loss": 0.4605, + "step": 5384 + }, + { + "epoch": 0.30086320082688495, + "grad_norm": 0.6799901723861694, + "learning_rate": 0.0008522803675481847, + "loss": 0.442, + "step": 5385 + }, + { + "epoch": 0.3009190714305668, + "grad_norm": 0.6667523384094238, + "learning_rate": 0.0008522523532048409, + "loss": 0.4334, + "step": 5386 + }, + { + "epoch": 0.3009749420342487, + "grad_norm": 0.5523688197135925, + "learning_rate": 0.0008522243388614971, + "loss": 0.4587, + "step": 5387 + }, + { + "epoch": 0.3010308126379306, + "grad_norm": 0.44369572401046753, + "learning_rate": 0.0008521963245181533, + "loss": 0.3426, + "step": 5388 + }, + { + "epoch": 0.3010866832416124, + "grad_norm": 0.4815022647380829, + "learning_rate": 0.0008521683101748096, + "loss": 0.4098, + "step": 5389 + }, + { + "epoch": 0.3011425538452943, + "grad_norm": 5.4134674072265625, + "learning_rate": 0.0008521402958314657, + "loss": 1.5266, + "step": 5390 + }, + { + "epoch": 0.30119842444897615, + "grad_norm": 1.5139812231063843, + "learning_rate": 0.0008521122814881219, + "loss": 0.5326, + "step": 5391 + }, + { + "epoch": 0.30125429505265805, + "grad_norm": 0.5053992867469788, + "learning_rate": 0.0008520842671447781, + "loss": 0.4527, + "step": 5392 + }, + { + "epoch": 0.3013101656563399, + "grad_norm": 1.549210786819458, + "learning_rate": 0.0008520562528014344, + "loss": 0.5756, + "step": 5393 + }, + { + "epoch": 0.3013660362600218, + "grad_norm": 0.6469733715057373, + "learning_rate": 0.0008520282384580905, + "loss": 0.4149, + "step": 5394 + }, + { + "epoch": 0.3014219068637037, + "grad_norm": 0.5179146528244019, + "learning_rate": 0.0008520002241147467, + "loss": 0.3953, + "step": 5395 + }, + { + "epoch": 0.3014777774673855, + "grad_norm": 0.5425530672073364, + "learning_rate": 0.000851972209771403, + "loss": 0.3676, + "step": 5396 + }, + { + "epoch": 0.3015336480710674, + "grad_norm": 0.5068067908287048, + "learning_rate": 0.0008519441954280592, + "loss": 0.4762, + "step": 5397 + }, + { + "epoch": 0.30158951867474926, + "grad_norm": 0.7817181348800659, + "learning_rate": 0.0008519161810847153, + "loss": 0.5663, + "step": 5398 + }, + { + "epoch": 0.30164538927843115, + "grad_norm": 0.7991998195648193, + "learning_rate": 0.0008518881667413716, + "loss": 0.5206, + "step": 5399 + }, + { + "epoch": 0.30170125988211305, + "grad_norm": 0.4967855215072632, + "learning_rate": 0.0008518601523980279, + "loss": 0.4153, + "step": 5400 + }, + { + "epoch": 0.3017571304857949, + "grad_norm": 0.6873252987861633, + "learning_rate": 0.0008518321380546841, + "loss": 0.4834, + "step": 5401 + }, + { + "epoch": 0.3018130010894768, + "grad_norm": 1.8710860013961792, + "learning_rate": 0.0008518041237113402, + "loss": 0.4617, + "step": 5402 + }, + { + "epoch": 0.3018688716931586, + "grad_norm": 0.6885605454444885, + "learning_rate": 0.0008517761093679965, + "loss": 0.4578, + "step": 5403 + }, + { + "epoch": 0.3019247422968405, + "grad_norm": 0.5564698576927185, + "learning_rate": 0.0008517480950246527, + "loss": 0.5454, + "step": 5404 + }, + { + "epoch": 0.3019806129005224, + "grad_norm": 1.0019111633300781, + "learning_rate": 0.0008517200806813088, + "loss": 0.4871, + "step": 5405 + }, + { + "epoch": 0.30203648350420426, + "grad_norm": 1.906326413154602, + "learning_rate": 0.000851692066337965, + "loss": 0.4209, + "step": 5406 + }, + { + "epoch": 0.30209235410788615, + "grad_norm": 0.7756844162940979, + "learning_rate": 0.0008516640519946213, + "loss": 0.4875, + "step": 5407 + }, + { + "epoch": 0.302148224711568, + "grad_norm": 0.48338866233825684, + "learning_rate": 0.0008516360376512775, + "loss": 0.6518, + "step": 5408 + }, + { + "epoch": 0.3022040953152499, + "grad_norm": 0.47145217657089233, + "learning_rate": 0.0008516080233079336, + "loss": 0.3646, + "step": 5409 + }, + { + "epoch": 0.30225996591893173, + "grad_norm": 5.8820295333862305, + "learning_rate": 0.0008515800089645899, + "loss": 0.3776, + "step": 5410 + }, + { + "epoch": 0.3023158365226136, + "grad_norm": 1.3987972736358643, + "learning_rate": 0.0008515519946212461, + "loss": 0.8578, + "step": 5411 + }, + { + "epoch": 0.3023717071262955, + "grad_norm": 0.576457142829895, + "learning_rate": 0.0008515239802779023, + "loss": 0.4197, + "step": 5412 + }, + { + "epoch": 0.30242757772997736, + "grad_norm": 0.4935535788536072, + "learning_rate": 0.0008514959659345584, + "loss": 0.4098, + "step": 5413 + }, + { + "epoch": 0.30248344833365925, + "grad_norm": 0.8208840489387512, + "learning_rate": 0.0008514679515912147, + "loss": 0.4809, + "step": 5414 + }, + { + "epoch": 0.3025393189373411, + "grad_norm": 0.46751460433006287, + "learning_rate": 0.000851439937247871, + "loss": 0.4088, + "step": 5415 + }, + { + "epoch": 0.302595189541023, + "grad_norm": 0.47302207350730896, + "learning_rate": 0.0008514119229045272, + "loss": 0.4592, + "step": 5416 + }, + { + "epoch": 0.3026510601447049, + "grad_norm": 0.7057961225509644, + "learning_rate": 0.0008513839085611833, + "loss": 0.488, + "step": 5417 + }, + { + "epoch": 0.3027069307483867, + "grad_norm": 0.6732742190361023, + "learning_rate": 0.0008513558942178396, + "loss": 0.4538, + "step": 5418 + }, + { + "epoch": 0.3027628013520686, + "grad_norm": 0.7602300047874451, + "learning_rate": 0.0008513278798744958, + "loss": 0.4126, + "step": 5419 + }, + { + "epoch": 0.30281867195575046, + "grad_norm": 0.5770102739334106, + "learning_rate": 0.000851299865531152, + "loss": 0.4117, + "step": 5420 + }, + { + "epoch": 0.30287454255943236, + "grad_norm": 1.7580934762954712, + "learning_rate": 0.0008512718511878082, + "loss": 0.5397, + "step": 5421 + }, + { + "epoch": 0.30293041316311425, + "grad_norm": 0.9029746651649475, + "learning_rate": 0.0008512438368444644, + "loss": 0.5272, + "step": 5422 + }, + { + "epoch": 0.3029862837667961, + "grad_norm": 0.6063520908355713, + "learning_rate": 0.0008512158225011206, + "loss": 0.5075, + "step": 5423 + }, + { + "epoch": 0.303042154370478, + "grad_norm": 0.5737748742103577, + "learning_rate": 0.0008511878081577767, + "loss": 0.4483, + "step": 5424 + }, + { + "epoch": 0.30309802497415983, + "grad_norm": 0.6065922379493713, + "learning_rate": 0.000851159793814433, + "loss": 0.4444, + "step": 5425 + }, + { + "epoch": 0.3031538955778417, + "grad_norm": 0.6958575248718262, + "learning_rate": 0.0008511317794710892, + "loss": 0.565, + "step": 5426 + }, + { + "epoch": 0.30320976618152357, + "grad_norm": 0.7924972176551819, + "learning_rate": 0.0008511037651277454, + "loss": 0.5073, + "step": 5427 + }, + { + "epoch": 0.30326563678520546, + "grad_norm": 0.9135690927505493, + "learning_rate": 0.0008510757507844016, + "loss": 0.5038, + "step": 5428 + }, + { + "epoch": 0.30332150738888736, + "grad_norm": 0.5357514023780823, + "learning_rate": 0.0008510477364410578, + "loss": 0.4795, + "step": 5429 + }, + { + "epoch": 0.3033773779925692, + "grad_norm": 0.9919570088386536, + "learning_rate": 0.000851019722097714, + "loss": 0.4501, + "step": 5430 + }, + { + "epoch": 0.3034332485962511, + "grad_norm": 0.6758516430854797, + "learning_rate": 0.0008509917077543702, + "loss": 0.6006, + "step": 5431 + }, + { + "epoch": 0.30348911919993293, + "grad_norm": 0.6771132946014404, + "learning_rate": 0.0008509636934110265, + "loss": 0.5347, + "step": 5432 + }, + { + "epoch": 0.30354498980361483, + "grad_norm": 0.8029685020446777, + "learning_rate": 0.0008509356790676827, + "loss": 0.5531, + "step": 5433 + }, + { + "epoch": 0.3036008604072967, + "grad_norm": 0.5046650171279907, + "learning_rate": 0.0008509076647243389, + "loss": 0.3668, + "step": 5434 + }, + { + "epoch": 0.30365673101097856, + "grad_norm": 0.5147439241409302, + "learning_rate": 0.0008508796503809952, + "loss": 0.4089, + "step": 5435 + }, + { + "epoch": 0.30371260161466046, + "grad_norm": 0.9645848870277405, + "learning_rate": 0.0008508516360376513, + "loss": 0.5932, + "step": 5436 + }, + { + "epoch": 0.3037684722183423, + "grad_norm": 0.6256189346313477, + "learning_rate": 0.0008508236216943075, + "loss": 0.5132, + "step": 5437 + }, + { + "epoch": 0.3038243428220242, + "grad_norm": 0.4374461770057678, + "learning_rate": 0.0008507956073509637, + "loss": 0.4545, + "step": 5438 + }, + { + "epoch": 0.3038802134257061, + "grad_norm": 0.5982236266136169, + "learning_rate": 0.00085076759300762, + "loss": 0.4329, + "step": 5439 + }, + { + "epoch": 0.30393608402938793, + "grad_norm": 1.588793158531189, + "learning_rate": 0.0008507395786642761, + "loss": 0.4609, + "step": 5440 + }, + { + "epoch": 0.3039919546330698, + "grad_norm": 1.0060601234436035, + "learning_rate": 0.0008507115643209323, + "loss": 0.5312, + "step": 5441 + }, + { + "epoch": 0.30404782523675167, + "grad_norm": 0.4920540153980255, + "learning_rate": 0.0008506835499775886, + "loss": 0.4138, + "step": 5442 + }, + { + "epoch": 0.30410369584043356, + "grad_norm": 0.6157340407371521, + "learning_rate": 0.0008506555356342448, + "loss": 0.4002, + "step": 5443 + }, + { + "epoch": 0.3041595664441154, + "grad_norm": 0.47835591435432434, + "learning_rate": 0.0008506275212909009, + "loss": 0.4544, + "step": 5444 + }, + { + "epoch": 0.3042154370477973, + "grad_norm": 0.6389055252075195, + "learning_rate": 0.0008505995069475571, + "loss": 0.5373, + "step": 5445 + }, + { + "epoch": 0.3042713076514792, + "grad_norm": 0.6588213443756104, + "learning_rate": 0.0008505714926042134, + "loss": 0.4497, + "step": 5446 + }, + { + "epoch": 0.30432717825516103, + "grad_norm": 0.7100991010665894, + "learning_rate": 0.0008505434782608695, + "loss": 0.519, + "step": 5447 + }, + { + "epoch": 0.30438304885884293, + "grad_norm": 0.45214101672172546, + "learning_rate": 0.0008505154639175257, + "loss": 0.5317, + "step": 5448 + }, + { + "epoch": 0.30443891946252477, + "grad_norm": 0.6625474095344543, + "learning_rate": 0.000850487449574182, + "loss": 0.4556, + "step": 5449 + }, + { + "epoch": 0.30449479006620667, + "grad_norm": 0.9110856056213379, + "learning_rate": 0.0008504594352308383, + "loss": 0.4368, + "step": 5450 + }, + { + "epoch": 0.30455066066988856, + "grad_norm": 0.5524949431419373, + "learning_rate": 0.0008504314208874944, + "loss": 0.5009, + "step": 5451 + }, + { + "epoch": 0.3046065312735704, + "grad_norm": 1.2332398891448975, + "learning_rate": 0.0008504034065441506, + "loss": 0.4833, + "step": 5452 + }, + { + "epoch": 0.3046624018772523, + "grad_norm": 0.9255529046058655, + "learning_rate": 0.0008503753922008069, + "loss": 0.4427, + "step": 5453 + }, + { + "epoch": 0.30471827248093414, + "grad_norm": 0.7364022135734558, + "learning_rate": 0.0008503473778574631, + "loss": 0.6067, + "step": 5454 + }, + { + "epoch": 0.30477414308461603, + "grad_norm": 0.8737490773200989, + "learning_rate": 0.0008503193635141192, + "loss": 0.476, + "step": 5455 + }, + { + "epoch": 0.30483001368829793, + "grad_norm": 0.49863874912261963, + "learning_rate": 0.0008502913491707754, + "loss": 0.4973, + "step": 5456 + }, + { + "epoch": 0.30488588429197977, + "grad_norm": 0.41027596592903137, + "learning_rate": 0.0008502633348274317, + "loss": 0.4004, + "step": 5457 + }, + { + "epoch": 0.30494175489566167, + "grad_norm": 0.5330960154533386, + "learning_rate": 0.0008502353204840879, + "loss": 0.5517, + "step": 5458 + }, + { + "epoch": 0.3049976254993435, + "grad_norm": 0.6761045455932617, + "learning_rate": 0.000850207306140744, + "loss": 0.5851, + "step": 5459 + }, + { + "epoch": 0.3050534961030254, + "grad_norm": 0.5528743863105774, + "learning_rate": 0.0008501792917974003, + "loss": 0.3882, + "step": 5460 + }, + { + "epoch": 0.30510936670670724, + "grad_norm": 0.8310927748680115, + "learning_rate": 0.0008501512774540565, + "loss": 0.5001, + "step": 5461 + }, + { + "epoch": 0.30516523731038914, + "grad_norm": 0.5166611671447754, + "learning_rate": 0.0008501232631107127, + "loss": 0.5078, + "step": 5462 + }, + { + "epoch": 0.30522110791407103, + "grad_norm": 3.688122272491455, + "learning_rate": 0.0008500952487673688, + "loss": 0.3721, + "step": 5463 + }, + { + "epoch": 0.3052769785177529, + "grad_norm": 0.5664483904838562, + "learning_rate": 0.0008500672344240251, + "loss": 0.4445, + "step": 5464 + }, + { + "epoch": 0.30533284912143477, + "grad_norm": 0.7255063652992249, + "learning_rate": 0.0008500392200806814, + "loss": 0.4261, + "step": 5465 + }, + { + "epoch": 0.3053887197251166, + "grad_norm": 0.9931159019470215, + "learning_rate": 0.0008500112057373376, + "loss": 0.6432, + "step": 5466 + }, + { + "epoch": 0.3054445903287985, + "grad_norm": 0.9298226237297058, + "learning_rate": 0.0008499831913939938, + "loss": 0.4622, + "step": 5467 + }, + { + "epoch": 0.3055004609324804, + "grad_norm": 0.5176401734352112, + "learning_rate": 0.00084995517705065, + "loss": 0.4449, + "step": 5468 + }, + { + "epoch": 0.30555633153616224, + "grad_norm": 1.0308665037155151, + "learning_rate": 0.0008499271627073062, + "loss": 0.6042, + "step": 5469 + }, + { + "epoch": 0.30561220213984414, + "grad_norm": 0.40117424726486206, + "learning_rate": 0.0008498991483639623, + "loss": 0.4074, + "step": 5470 + }, + { + "epoch": 0.305668072743526, + "grad_norm": 1.0603394508361816, + "learning_rate": 0.0008498711340206186, + "loss": 0.4482, + "step": 5471 + }, + { + "epoch": 0.30572394334720787, + "grad_norm": 0.5529935359954834, + "learning_rate": 0.0008498431196772748, + "loss": 0.4988, + "step": 5472 + }, + { + "epoch": 0.3057798139508897, + "grad_norm": 0.8265949487686157, + "learning_rate": 0.000849815105333931, + "loss": 0.4792, + "step": 5473 + }, + { + "epoch": 0.3058356845545716, + "grad_norm": 0.4085021913051605, + "learning_rate": 0.0008497870909905872, + "loss": 0.4662, + "step": 5474 + }, + { + "epoch": 0.3058915551582535, + "grad_norm": 0.5806437134742737, + "learning_rate": 0.0008497590766472434, + "loss": 0.364, + "step": 5475 + }, + { + "epoch": 0.30594742576193534, + "grad_norm": 0.5203709006309509, + "learning_rate": 0.0008497310623038996, + "loss": 0.4505, + "step": 5476 + }, + { + "epoch": 0.30600329636561724, + "grad_norm": 0.5257204174995422, + "learning_rate": 0.0008497030479605558, + "loss": 0.4827, + "step": 5477 + }, + { + "epoch": 0.3060591669692991, + "grad_norm": 3.472770929336548, + "learning_rate": 0.000849675033617212, + "loss": 0.4091, + "step": 5478 + }, + { + "epoch": 0.306115037572981, + "grad_norm": 0.6950019598007202, + "learning_rate": 0.0008496470192738682, + "loss": 0.3949, + "step": 5479 + }, + { + "epoch": 0.30617090817666287, + "grad_norm": 0.6670989394187927, + "learning_rate": 0.0008496190049305244, + "loss": 0.4229, + "step": 5480 + }, + { + "epoch": 0.3062267787803447, + "grad_norm": 0.5419312715530396, + "learning_rate": 0.0008495909905871808, + "loss": 0.4994, + "step": 5481 + }, + { + "epoch": 0.3062826493840266, + "grad_norm": 0.5931705832481384, + "learning_rate": 0.0008495629762438369, + "loss": 0.4802, + "step": 5482 + }, + { + "epoch": 0.30633851998770845, + "grad_norm": 0.468280553817749, + "learning_rate": 0.0008495349619004931, + "loss": 0.4473, + "step": 5483 + }, + { + "epoch": 0.30639439059139034, + "grad_norm": 0.4428548514842987, + "learning_rate": 0.0008495069475571493, + "loss": 0.4471, + "step": 5484 + }, + { + "epoch": 0.30645026119507224, + "grad_norm": 2.3595852851867676, + "learning_rate": 0.0008494789332138056, + "loss": 0.5967, + "step": 5485 + }, + { + "epoch": 0.3065061317987541, + "grad_norm": 0.6981626152992249, + "learning_rate": 0.0008494509188704617, + "loss": 0.521, + "step": 5486 + }, + { + "epoch": 0.306562002402436, + "grad_norm": 0.6752743721008301, + "learning_rate": 0.0008494229045271179, + "loss": 0.562, + "step": 5487 + }, + { + "epoch": 0.3066178730061178, + "grad_norm": 0.3810230493545532, + "learning_rate": 0.0008493948901837741, + "loss": 0.4077, + "step": 5488 + }, + { + "epoch": 0.3066737436097997, + "grad_norm": 2.602975606918335, + "learning_rate": 0.0008493668758404303, + "loss": 0.4663, + "step": 5489 + }, + { + "epoch": 0.30672961421348155, + "grad_norm": 1.2136085033416748, + "learning_rate": 0.0008493388614970865, + "loss": 0.4838, + "step": 5490 + }, + { + "epoch": 0.30678548481716345, + "grad_norm": 1.2993669509887695, + "learning_rate": 0.0008493108471537427, + "loss": 0.584, + "step": 5491 + }, + { + "epoch": 0.30684135542084534, + "grad_norm": 0.5152722597122192, + "learning_rate": 0.000849282832810399, + "loss": 0.5206, + "step": 5492 + }, + { + "epoch": 0.3068972260245272, + "grad_norm": 0.3931487500667572, + "learning_rate": 0.0008492548184670551, + "loss": 0.4421, + "step": 5493 + }, + { + "epoch": 0.3069530966282091, + "grad_norm": 1.7890452146530151, + "learning_rate": 0.0008492268041237113, + "loss": 0.3876, + "step": 5494 + }, + { + "epoch": 0.3070089672318909, + "grad_norm": 1.6168525218963623, + "learning_rate": 0.0008491987897803675, + "loss": 0.4868, + "step": 5495 + }, + { + "epoch": 0.3070648378355728, + "grad_norm": 0.5749663710594177, + "learning_rate": 0.0008491707754370238, + "loss": 0.4714, + "step": 5496 + }, + { + "epoch": 0.3071207084392547, + "grad_norm": 0.4699666500091553, + "learning_rate": 0.0008491427610936799, + "loss": 0.3894, + "step": 5497 + }, + { + "epoch": 0.30717657904293655, + "grad_norm": 0.8526723384857178, + "learning_rate": 0.0008491147467503361, + "loss": 0.4533, + "step": 5498 + }, + { + "epoch": 0.30723244964661844, + "grad_norm": 0.554200291633606, + "learning_rate": 0.0008490867324069925, + "loss": 0.4845, + "step": 5499 + }, + { + "epoch": 0.3072883202503003, + "grad_norm": 0.4599955081939697, + "learning_rate": 0.0008490587180636487, + "loss": 0.4604, + "step": 5500 + }, + { + "epoch": 0.3072883202503003, + "eval_cer": 0.09467775195574613, + "eval_loss": 0.35704582929611206, + "eval_runtime": 57.1145, + "eval_samples_per_second": 79.454, + "eval_steps_per_second": 4.972, + "eval_wer": 0.3726711544370269, + "step": 5500 + }, + { + "epoch": 0.3073441908539822, + "grad_norm": 0.4593367576599121, + "learning_rate": 0.0008490307037203048, + "loss": 0.4484, + "step": 5501 + }, + { + "epoch": 0.3074000614576641, + "grad_norm": 0.5224381685256958, + "learning_rate": 0.000849002689376961, + "loss": 0.4518, + "step": 5502 + }, + { + "epoch": 0.3074559320613459, + "grad_norm": 0.7090650796890259, + "learning_rate": 0.0008489746750336173, + "loss": 0.458, + "step": 5503 + }, + { + "epoch": 0.3075118026650278, + "grad_norm": 0.4909626245498657, + "learning_rate": 0.0008489466606902735, + "loss": 0.4829, + "step": 5504 + }, + { + "epoch": 0.30756767326870965, + "grad_norm": 0.7202634811401367, + "learning_rate": 0.0008489186463469296, + "loss": 0.4105, + "step": 5505 + }, + { + "epoch": 0.30762354387239155, + "grad_norm": 0.6702147722244263, + "learning_rate": 0.0008488906320035859, + "loss": 0.4491, + "step": 5506 + }, + { + "epoch": 0.3076794144760734, + "grad_norm": 0.708186686038971, + "learning_rate": 0.0008488626176602421, + "loss": 0.3231, + "step": 5507 + }, + { + "epoch": 0.3077352850797553, + "grad_norm": 0.6041837334632874, + "learning_rate": 0.0008488346033168983, + "loss": 0.5615, + "step": 5508 + }, + { + "epoch": 0.3077911556834372, + "grad_norm": 1.2007112503051758, + "learning_rate": 0.0008488065889735544, + "loss": 0.62, + "step": 5509 + }, + { + "epoch": 0.307847026287119, + "grad_norm": 0.4864131212234497, + "learning_rate": 0.0008487785746302107, + "loss": 0.4876, + "step": 5510 + }, + { + "epoch": 0.3079028968908009, + "grad_norm": 0.5683916211128235, + "learning_rate": 0.0008487505602868669, + "loss": 0.3889, + "step": 5511 + }, + { + "epoch": 0.30795876749448275, + "grad_norm": 0.38261479139328003, + "learning_rate": 0.000848722545943523, + "loss": 0.4666, + "step": 5512 + }, + { + "epoch": 0.30801463809816465, + "grad_norm": 6.176339149475098, + "learning_rate": 0.0008486945316001793, + "loss": 0.4113, + "step": 5513 + }, + { + "epoch": 0.30807050870184655, + "grad_norm": 0.6758056879043579, + "learning_rate": 0.0008486665172568355, + "loss": 0.4531, + "step": 5514 + }, + { + "epoch": 0.3081263793055284, + "grad_norm": 0.9775447845458984, + "learning_rate": 0.0008486385029134918, + "loss": 0.3975, + "step": 5515 + }, + { + "epoch": 0.3081822499092103, + "grad_norm": 0.4279237389564514, + "learning_rate": 0.0008486104885701478, + "loss": 0.4329, + "step": 5516 + }, + { + "epoch": 0.3082381205128921, + "grad_norm": 0.6433815360069275, + "learning_rate": 0.0008485824742268042, + "loss": 0.4753, + "step": 5517 + }, + { + "epoch": 0.308293991116574, + "grad_norm": 0.6737523674964905, + "learning_rate": 0.0008485544598834604, + "loss": 0.4609, + "step": 5518 + }, + { + "epoch": 0.3083498617202559, + "grad_norm": 0.7861856818199158, + "learning_rate": 0.0008485264455401166, + "loss": 0.4307, + "step": 5519 + }, + { + "epoch": 0.30840573232393775, + "grad_norm": 0.4873378872871399, + "learning_rate": 0.0008484984311967727, + "loss": 0.4621, + "step": 5520 + }, + { + "epoch": 0.30846160292761965, + "grad_norm": 0.5509065985679626, + "learning_rate": 0.000848470416853429, + "loss": 0.617, + "step": 5521 + }, + { + "epoch": 0.3085174735313015, + "grad_norm": 0.45215722918510437, + "learning_rate": 0.0008484424025100852, + "loss": 0.4936, + "step": 5522 + }, + { + "epoch": 0.3085733441349834, + "grad_norm": 0.8102368712425232, + "learning_rate": 0.0008484143881667414, + "loss": 0.4194, + "step": 5523 + }, + { + "epoch": 0.3086292147386652, + "grad_norm": 0.9361057877540588, + "learning_rate": 0.0008483863738233976, + "loss": 0.4078, + "step": 5524 + }, + { + "epoch": 0.3086850853423471, + "grad_norm": 0.5034268498420715, + "learning_rate": 0.0008483583594800538, + "loss": 0.5377, + "step": 5525 + }, + { + "epoch": 0.308740955946029, + "grad_norm": 0.7160896062850952, + "learning_rate": 0.00084833034513671, + "loss": 0.4393, + "step": 5526 + }, + { + "epoch": 0.30879682654971086, + "grad_norm": 0.40303611755371094, + "learning_rate": 0.0008483023307933662, + "loss": 0.3473, + "step": 5527 + }, + { + "epoch": 0.30885269715339275, + "grad_norm": 0.5000059008598328, + "learning_rate": 0.0008482743164500224, + "loss": 0.4952, + "step": 5528 + }, + { + "epoch": 0.3089085677570746, + "grad_norm": 2.6567676067352295, + "learning_rate": 0.0008482463021066786, + "loss": 0.7158, + "step": 5529 + }, + { + "epoch": 0.3089644383607565, + "grad_norm": 0.7796276807785034, + "learning_rate": 0.0008482182877633348, + "loss": 0.4585, + "step": 5530 + }, + { + "epoch": 0.3090203089644384, + "grad_norm": 0.9154967069625854, + "learning_rate": 0.000848190273419991, + "loss": 0.5877, + "step": 5531 + }, + { + "epoch": 0.3090761795681202, + "grad_norm": 1.3950281143188477, + "learning_rate": 0.0008481622590766472, + "loss": 0.6357, + "step": 5532 + }, + { + "epoch": 0.3091320501718021, + "grad_norm": 0.5266571640968323, + "learning_rate": 0.0008481342447333035, + "loss": 0.3789, + "step": 5533 + }, + { + "epoch": 0.30918792077548396, + "grad_norm": 0.6321330070495605, + "learning_rate": 0.0008481062303899597, + "loss": 0.5741, + "step": 5534 + }, + { + "epoch": 0.30924379137916586, + "grad_norm": 0.9014947414398193, + "learning_rate": 0.0008480782160466159, + "loss": 0.5489, + "step": 5535 + }, + { + "epoch": 0.30929966198284775, + "grad_norm": 0.7584993243217468, + "learning_rate": 0.0008480502017032721, + "loss": 0.67, + "step": 5536 + }, + { + "epoch": 0.3093555325865296, + "grad_norm": 3.413449764251709, + "learning_rate": 0.0008480221873599283, + "loss": 0.4918, + "step": 5537 + }, + { + "epoch": 0.3094114031902115, + "grad_norm": 0.7581272125244141, + "learning_rate": 0.0008479941730165846, + "loss": 0.5472, + "step": 5538 + }, + { + "epoch": 0.3094672737938933, + "grad_norm": 0.4519740045070648, + "learning_rate": 0.0008479661586732407, + "loss": 0.4723, + "step": 5539 + }, + { + "epoch": 0.3095231443975752, + "grad_norm": 0.8811973333358765, + "learning_rate": 0.0008479381443298969, + "loss": 0.5827, + "step": 5540 + }, + { + "epoch": 0.30957901500125706, + "grad_norm": 0.8245788216590881, + "learning_rate": 0.0008479101299865531, + "loss": 0.5444, + "step": 5541 + }, + { + "epoch": 0.30963488560493896, + "grad_norm": 0.5593982934951782, + "learning_rate": 0.0008478821156432094, + "loss": 0.4747, + "step": 5542 + }, + { + "epoch": 0.30969075620862085, + "grad_norm": 3.951293706893921, + "learning_rate": 0.0008478541012998655, + "loss": 0.4909, + "step": 5543 + }, + { + "epoch": 0.3097466268123027, + "grad_norm": 0.401913583278656, + "learning_rate": 0.0008478260869565217, + "loss": 0.4267, + "step": 5544 + }, + { + "epoch": 0.3098024974159846, + "grad_norm": 0.9992120862007141, + "learning_rate": 0.000847798072613178, + "loss": 0.4245, + "step": 5545 + }, + { + "epoch": 0.30985836801966643, + "grad_norm": 0.7159444689750671, + "learning_rate": 0.0008477700582698342, + "loss": 0.5986, + "step": 5546 + }, + { + "epoch": 0.3099142386233483, + "grad_norm": 0.47679394483566284, + "learning_rate": 0.0008477420439264903, + "loss": 0.4011, + "step": 5547 + }, + { + "epoch": 0.3099701092270302, + "grad_norm": 0.5748596787452698, + "learning_rate": 0.0008477140295831465, + "loss": 0.4448, + "step": 5548 + }, + { + "epoch": 0.31002597983071206, + "grad_norm": 0.5180022716522217, + "learning_rate": 0.0008476860152398029, + "loss": 0.4157, + "step": 5549 + }, + { + "epoch": 0.31008185043439396, + "grad_norm": 0.6510815620422363, + "learning_rate": 0.0008476580008964591, + "loss": 0.5676, + "step": 5550 + }, + { + "epoch": 0.3101377210380758, + "grad_norm": 1.0993666648864746, + "learning_rate": 0.0008476299865531152, + "loss": 0.4451, + "step": 5551 + }, + { + "epoch": 0.3101935916417577, + "grad_norm": 0.5017836093902588, + "learning_rate": 0.0008476019722097715, + "loss": 0.3944, + "step": 5552 + }, + { + "epoch": 0.3102494622454396, + "grad_norm": 0.4954838156700134, + "learning_rate": 0.0008475739578664277, + "loss": 0.3837, + "step": 5553 + }, + { + "epoch": 0.31030533284912143, + "grad_norm": 0.564395010471344, + "learning_rate": 0.0008475459435230838, + "loss": 0.5488, + "step": 5554 + }, + { + "epoch": 0.3103612034528033, + "grad_norm": 1.5262882709503174, + "learning_rate": 0.00084751792917974, + "loss": 0.897, + "step": 5555 + }, + { + "epoch": 0.31041707405648516, + "grad_norm": 0.5105319023132324, + "learning_rate": 0.0008474899148363963, + "loss": 0.4546, + "step": 5556 + }, + { + "epoch": 0.31047294466016706, + "grad_norm": 1.54734468460083, + "learning_rate": 0.0008474619004930525, + "loss": 0.5017, + "step": 5557 + }, + { + "epoch": 0.3105288152638489, + "grad_norm": 0.7369034290313721, + "learning_rate": 0.0008474338861497086, + "loss": 0.4563, + "step": 5558 + }, + { + "epoch": 0.3105846858675308, + "grad_norm": 0.629767119884491, + "learning_rate": 0.0008474058718063648, + "loss": 0.5833, + "step": 5559 + }, + { + "epoch": 0.3106405564712127, + "grad_norm": 1.7269259691238403, + "learning_rate": 0.0008473778574630211, + "loss": 0.4346, + "step": 5560 + }, + { + "epoch": 0.31069642707489453, + "grad_norm": 0.574377179145813, + "learning_rate": 0.0008473498431196773, + "loss": 0.4846, + "step": 5561 + }, + { + "epoch": 0.31075229767857643, + "grad_norm": 0.46182698011398315, + "learning_rate": 0.0008473218287763334, + "loss": 0.5356, + "step": 5562 + }, + { + "epoch": 0.31080816828225827, + "grad_norm": 0.8448501229286194, + "learning_rate": 0.0008472938144329897, + "loss": 0.4727, + "step": 5563 + }, + { + "epoch": 0.31086403888594016, + "grad_norm": 0.5345706939697266, + "learning_rate": 0.0008472658000896459, + "loss": 0.4187, + "step": 5564 + }, + { + "epoch": 0.31091990948962206, + "grad_norm": 0.4491683542728424, + "learning_rate": 0.0008472377857463021, + "loss": 0.3696, + "step": 5565 + }, + { + "epoch": 0.3109757800933039, + "grad_norm": 0.5052008628845215, + "learning_rate": 0.0008472097714029582, + "loss": 0.5725, + "step": 5566 + }, + { + "epoch": 0.3110316506969858, + "grad_norm": 0.5478227138519287, + "learning_rate": 0.0008471817570596146, + "loss": 0.4798, + "step": 5567 + }, + { + "epoch": 0.31108752130066764, + "grad_norm": 4.434738636016846, + "learning_rate": 0.0008471537427162708, + "loss": 0.3221, + "step": 5568 + }, + { + "epoch": 0.31114339190434953, + "grad_norm": 0.4904155731201172, + "learning_rate": 0.000847125728372927, + "loss": 0.5013, + "step": 5569 + }, + { + "epoch": 0.3111992625080314, + "grad_norm": 0.432204931974411, + "learning_rate": 0.0008470977140295832, + "loss": 0.3315, + "step": 5570 + }, + { + "epoch": 0.31125513311171327, + "grad_norm": 0.7502817511558533, + "learning_rate": 0.0008470696996862394, + "loss": 0.5718, + "step": 5571 + }, + { + "epoch": 0.31131100371539516, + "grad_norm": 0.5065408945083618, + "learning_rate": 0.0008470416853428956, + "loss": 0.4844, + "step": 5572 + }, + { + "epoch": 0.311366874319077, + "grad_norm": 0.5898789763450623, + "learning_rate": 0.0008470136709995517, + "loss": 0.4886, + "step": 5573 + }, + { + "epoch": 0.3114227449227589, + "grad_norm": 0.39644524455070496, + "learning_rate": 0.000846985656656208, + "loss": 0.485, + "step": 5574 + }, + { + "epoch": 0.31147861552644074, + "grad_norm": 0.809878945350647, + "learning_rate": 0.0008469576423128642, + "loss": 0.5316, + "step": 5575 + }, + { + "epoch": 0.31153448613012263, + "grad_norm": 0.9915091395378113, + "learning_rate": 0.0008469296279695204, + "loss": 0.5653, + "step": 5576 + }, + { + "epoch": 0.31159035673380453, + "grad_norm": 0.49817702174186707, + "learning_rate": 0.0008469016136261766, + "loss": 0.3943, + "step": 5577 + }, + { + "epoch": 0.31164622733748637, + "grad_norm": 0.5065735578536987, + "learning_rate": 0.0008468735992828328, + "loss": 0.4353, + "step": 5578 + }, + { + "epoch": 0.31170209794116827, + "grad_norm": 0.49890756607055664, + "learning_rate": 0.000846845584939489, + "loss": 0.3963, + "step": 5579 + }, + { + "epoch": 0.3117579685448501, + "grad_norm": 0.6314283609390259, + "learning_rate": 0.0008468175705961452, + "loss": 0.5923, + "step": 5580 + }, + { + "epoch": 0.311813839148532, + "grad_norm": 1.574580192565918, + "learning_rate": 0.0008467895562528014, + "loss": 0.4683, + "step": 5581 + }, + { + "epoch": 0.3118697097522139, + "grad_norm": 0.6348354816436768, + "learning_rate": 0.0008467615419094576, + "loss": 0.4306, + "step": 5582 + }, + { + "epoch": 0.31192558035589574, + "grad_norm": 0.530066967010498, + "learning_rate": 0.0008467335275661139, + "loss": 0.4068, + "step": 5583 + }, + { + "epoch": 0.31198145095957763, + "grad_norm": 0.8644289374351501, + "learning_rate": 0.0008467055132227702, + "loss": 0.5738, + "step": 5584 + }, + { + "epoch": 0.3120373215632595, + "grad_norm": 0.42082685232162476, + "learning_rate": 0.0008466774988794263, + "loss": 0.3811, + "step": 5585 + }, + { + "epoch": 0.31209319216694137, + "grad_norm": 0.5201012492179871, + "learning_rate": 0.0008466494845360825, + "loss": 0.5952, + "step": 5586 + }, + { + "epoch": 0.3121490627706232, + "grad_norm": 0.6166862845420837, + "learning_rate": 0.0008466214701927387, + "loss": 0.5705, + "step": 5587 + }, + { + "epoch": 0.3122049333743051, + "grad_norm": 0.5003292560577393, + "learning_rate": 0.000846593455849395, + "loss": 0.3813, + "step": 5588 + }, + { + "epoch": 0.312260803977987, + "grad_norm": 1.5041347742080688, + "learning_rate": 0.0008465654415060511, + "loss": 0.5422, + "step": 5589 + }, + { + "epoch": 0.31231667458166884, + "grad_norm": 0.39266303181648254, + "learning_rate": 0.0008465374271627073, + "loss": 0.4331, + "step": 5590 + }, + { + "epoch": 0.31237254518535074, + "grad_norm": 0.5897176265716553, + "learning_rate": 0.0008465094128193636, + "loss": 0.5624, + "step": 5591 + }, + { + "epoch": 0.3124284157890326, + "grad_norm": 0.566651463508606, + "learning_rate": 0.0008464813984760198, + "loss": 0.4335, + "step": 5592 + }, + { + "epoch": 0.31248428639271447, + "grad_norm": 0.6047028303146362, + "learning_rate": 0.0008464533841326759, + "loss": 0.5434, + "step": 5593 + }, + { + "epoch": 0.31254015699639637, + "grad_norm": 4.277653694152832, + "learning_rate": 0.0008464253697893321, + "loss": 0.4892, + "step": 5594 + }, + { + "epoch": 0.3125960276000782, + "grad_norm": 0.44817426800727844, + "learning_rate": 0.0008463973554459884, + "loss": 0.4766, + "step": 5595 + }, + { + "epoch": 0.3126518982037601, + "grad_norm": 0.40493664145469666, + "learning_rate": 0.0008463693411026445, + "loss": 0.3596, + "step": 5596 + }, + { + "epoch": 0.31270776880744194, + "grad_norm": 0.6015021800994873, + "learning_rate": 0.0008463413267593007, + "loss": 0.5401, + "step": 5597 + }, + { + "epoch": 0.31276363941112384, + "grad_norm": 0.7387809753417969, + "learning_rate": 0.0008463133124159569, + "loss": 0.344, + "step": 5598 + }, + { + "epoch": 0.31281951001480574, + "grad_norm": 0.4570474326610565, + "learning_rate": 0.0008462852980726133, + "loss": 0.3777, + "step": 5599 + }, + { + "epoch": 0.3128753806184876, + "grad_norm": 0.6257132887840271, + "learning_rate": 0.0008462572837292694, + "loss": 0.4471, + "step": 5600 + }, + { + "epoch": 0.31293125122216947, + "grad_norm": 0.5994383096694946, + "learning_rate": 0.0008462292693859256, + "loss": 0.4931, + "step": 5601 + }, + { + "epoch": 0.3129871218258513, + "grad_norm": 0.5457228422164917, + "learning_rate": 0.0008462012550425819, + "loss": 0.5819, + "step": 5602 + }, + { + "epoch": 0.3130429924295332, + "grad_norm": 0.5879257321357727, + "learning_rate": 0.0008461732406992381, + "loss": 0.4356, + "step": 5603 + }, + { + "epoch": 0.31309886303321505, + "grad_norm": 0.8198217749595642, + "learning_rate": 0.0008461452263558942, + "loss": 0.5664, + "step": 5604 + }, + { + "epoch": 0.31315473363689694, + "grad_norm": 0.685960054397583, + "learning_rate": 0.0008461172120125504, + "loss": 0.5183, + "step": 5605 + }, + { + "epoch": 0.31321060424057884, + "grad_norm": 0.5605204701423645, + "learning_rate": 0.0008460891976692067, + "loss": 0.5364, + "step": 5606 + }, + { + "epoch": 0.3132664748442607, + "grad_norm": 0.45028555393218994, + "learning_rate": 0.0008460611833258629, + "loss": 0.4825, + "step": 5607 + }, + { + "epoch": 0.3133223454479426, + "grad_norm": 0.5055218935012817, + "learning_rate": 0.000846033168982519, + "loss": 0.4638, + "step": 5608 + }, + { + "epoch": 0.3133782160516244, + "grad_norm": 0.5085271596908569, + "learning_rate": 0.0008460051546391753, + "loss": 0.4709, + "step": 5609 + }, + { + "epoch": 0.3134340866553063, + "grad_norm": 0.5966741442680359, + "learning_rate": 0.0008459771402958315, + "loss": 0.4571, + "step": 5610 + }, + { + "epoch": 0.3134899572589882, + "grad_norm": 0.670189380645752, + "learning_rate": 0.0008459491259524877, + "loss": 0.5908, + "step": 5611 + }, + { + "epoch": 0.31354582786267005, + "grad_norm": 0.4916702210903168, + "learning_rate": 0.0008459211116091438, + "loss": 0.4442, + "step": 5612 + }, + { + "epoch": 0.31360169846635194, + "grad_norm": 1.1918748617172241, + "learning_rate": 0.0008458930972658001, + "loss": 0.3982, + "step": 5613 + }, + { + "epoch": 0.3136575690700338, + "grad_norm": 0.6493722796440125, + "learning_rate": 0.0008458650829224563, + "loss": 0.4024, + "step": 5614 + }, + { + "epoch": 0.3137134396737157, + "grad_norm": 1.4247260093688965, + "learning_rate": 0.0008458370685791125, + "loss": 0.4934, + "step": 5615 + }, + { + "epoch": 0.3137693102773976, + "grad_norm": 0.5089534521102905, + "learning_rate": 0.0008458090542357688, + "loss": 0.4256, + "step": 5616 + }, + { + "epoch": 0.3138251808810794, + "grad_norm": 1.4272161722183228, + "learning_rate": 0.000845781039892425, + "loss": 0.389, + "step": 5617 + }, + { + "epoch": 0.3138810514847613, + "grad_norm": 0.4404493570327759, + "learning_rate": 0.0008457530255490812, + "loss": 0.4821, + "step": 5618 + }, + { + "epoch": 0.31393692208844315, + "grad_norm": 0.5874308943748474, + "learning_rate": 0.0008457250112057373, + "loss": 0.5103, + "step": 5619 + }, + { + "epoch": 0.31399279269212504, + "grad_norm": 0.5645403861999512, + "learning_rate": 0.0008456969968623936, + "loss": 0.4282, + "step": 5620 + }, + { + "epoch": 0.3140486632958069, + "grad_norm": 2.057006359100342, + "learning_rate": 0.0008456689825190498, + "loss": 0.3736, + "step": 5621 + }, + { + "epoch": 0.3141045338994888, + "grad_norm": 0.647833526134491, + "learning_rate": 0.000845640968175706, + "loss": 0.4486, + "step": 5622 + }, + { + "epoch": 0.3141604045031707, + "grad_norm": 4.43013334274292, + "learning_rate": 0.0008456129538323622, + "loss": 0.382, + "step": 5623 + }, + { + "epoch": 0.3142162751068525, + "grad_norm": 0.8136746883392334, + "learning_rate": 0.0008455849394890184, + "loss": 0.4565, + "step": 5624 + }, + { + "epoch": 0.3142721457105344, + "grad_norm": 0.7927437424659729, + "learning_rate": 0.0008455569251456746, + "loss": 0.5259, + "step": 5625 + }, + { + "epoch": 0.31432801631421625, + "grad_norm": 0.8155714273452759, + "learning_rate": 0.0008455289108023308, + "loss": 0.437, + "step": 5626 + }, + { + "epoch": 0.31438388691789815, + "grad_norm": 0.6431964039802551, + "learning_rate": 0.000845500896458987, + "loss": 0.5554, + "step": 5627 + }, + { + "epoch": 0.31443975752158004, + "grad_norm": 0.5642159581184387, + "learning_rate": 0.0008454728821156432, + "loss": 0.5107, + "step": 5628 + }, + { + "epoch": 0.3144956281252619, + "grad_norm": 0.4544612765312195, + "learning_rate": 0.0008454448677722994, + "loss": 0.4235, + "step": 5629 + }, + { + "epoch": 0.3145514987289438, + "grad_norm": 0.6413934230804443, + "learning_rate": 0.0008454168534289556, + "loss": 0.4584, + "step": 5630 + }, + { + "epoch": 0.3146073693326256, + "grad_norm": 0.6362944841384888, + "learning_rate": 0.0008453888390856118, + "loss": 0.481, + "step": 5631 + }, + { + "epoch": 0.3146632399363075, + "grad_norm": 0.5639865398406982, + "learning_rate": 0.000845360824742268, + "loss": 0.4231, + "step": 5632 + }, + { + "epoch": 0.3147191105399894, + "grad_norm": 0.4940715432167053, + "learning_rate": 0.0008453328103989243, + "loss": 0.5081, + "step": 5633 + }, + { + "epoch": 0.31477498114367125, + "grad_norm": 0.559850811958313, + "learning_rate": 0.0008453047960555806, + "loss": 0.3895, + "step": 5634 + }, + { + "epoch": 0.31483085174735315, + "grad_norm": 0.4550793468952179, + "learning_rate": 0.0008452767817122367, + "loss": 0.404, + "step": 5635 + }, + { + "epoch": 0.314886722351035, + "grad_norm": 0.5403375029563904, + "learning_rate": 0.0008452487673688929, + "loss": 0.5609, + "step": 5636 + }, + { + "epoch": 0.3149425929547169, + "grad_norm": 0.4877609312534332, + "learning_rate": 0.0008452207530255491, + "loss": 0.474, + "step": 5637 + }, + { + "epoch": 0.3149984635583987, + "grad_norm": 0.6696338057518005, + "learning_rate": 0.0008451927386822053, + "loss": 0.474, + "step": 5638 + }, + { + "epoch": 0.3150543341620806, + "grad_norm": 0.7584934234619141, + "learning_rate": 0.0008451647243388615, + "loss": 0.6215, + "step": 5639 + }, + { + "epoch": 0.3151102047657625, + "grad_norm": 0.46672090888023376, + "learning_rate": 0.0008451367099955177, + "loss": 0.4357, + "step": 5640 + }, + { + "epoch": 0.31516607536944435, + "grad_norm": Infinity, + "learning_rate": 0.0008451367099955177, + "loss": 0.4689, + "step": 5641 + }, + { + "epoch": 0.31522194597312625, + "grad_norm": 0.4713086187839508, + "learning_rate": 0.000845108695652174, + "loss": 0.5149, + "step": 5642 + }, + { + "epoch": 0.3152778165768081, + "grad_norm": 0.6608066558837891, + "learning_rate": 0.0008450806813088301, + "loss": 0.5634, + "step": 5643 + }, + { + "epoch": 0.31533368718049, + "grad_norm": 0.5421141386032104, + "learning_rate": 0.0008450526669654863, + "loss": 0.3614, + "step": 5644 + }, + { + "epoch": 0.3153895577841719, + "grad_norm": 0.8061105608940125, + "learning_rate": 0.0008450246526221425, + "loss": 0.4741, + "step": 5645 + }, + { + "epoch": 0.3154454283878537, + "grad_norm": 0.586520254611969, + "learning_rate": 0.0008449966382787988, + "loss": 0.4566, + "step": 5646 + }, + { + "epoch": 0.3155012989915356, + "grad_norm": 0.45880261063575745, + "learning_rate": 0.0008449686239354549, + "loss": 0.5178, + "step": 5647 + }, + { + "epoch": 0.31555716959521746, + "grad_norm": 0.5670356154441833, + "learning_rate": 0.0008449406095921111, + "loss": 0.5662, + "step": 5648 + }, + { + "epoch": 0.31561304019889935, + "grad_norm": 0.8781639933586121, + "learning_rate": 0.0008449125952487674, + "loss": 0.5785, + "step": 5649 + }, + { + "epoch": 0.31566891080258125, + "grad_norm": 0.6879367828369141, + "learning_rate": 0.0008448845809054237, + "loss": 0.4959, + "step": 5650 + }, + { + "epoch": 0.3157247814062631, + "grad_norm": 2.1492221355438232, + "learning_rate": 0.0008448565665620798, + "loss": 0.5532, + "step": 5651 + }, + { + "epoch": 0.315780652009945, + "grad_norm": 0.6593572497367859, + "learning_rate": 0.000844828552218736, + "loss": 0.7348, + "step": 5652 + }, + { + "epoch": 0.3158365226136268, + "grad_norm": 0.5745440721511841, + "learning_rate": 0.0008448005378753923, + "loss": 0.4871, + "step": 5653 + }, + { + "epoch": 0.3158923932173087, + "grad_norm": 0.5804295539855957, + "learning_rate": 0.0008447725235320485, + "loss": 0.4163, + "step": 5654 + }, + { + "epoch": 0.31594826382099056, + "grad_norm": 1.0495027303695679, + "learning_rate": 0.0008447445091887046, + "loss": 0.3947, + "step": 5655 + }, + { + "epoch": 0.31600413442467246, + "grad_norm": 0.6198778748512268, + "learning_rate": 0.0008447164948453609, + "loss": 0.5275, + "step": 5656 + }, + { + "epoch": 0.31606000502835435, + "grad_norm": 0.4754329025745392, + "learning_rate": 0.0008446884805020171, + "loss": 0.5197, + "step": 5657 + }, + { + "epoch": 0.3161158756320362, + "grad_norm": 0.4649758040904999, + "learning_rate": 0.0008446604661586733, + "loss": 0.4538, + "step": 5658 + }, + { + "epoch": 0.3161717462357181, + "grad_norm": 0.4759577810764313, + "learning_rate": 0.0008446324518153294, + "loss": 0.3892, + "step": 5659 + }, + { + "epoch": 0.3162276168393999, + "grad_norm": 0.4955660104751587, + "learning_rate": 0.0008446044374719857, + "loss": 0.4737, + "step": 5660 + }, + { + "epoch": 0.3162834874430818, + "grad_norm": 0.6217213869094849, + "learning_rate": 0.0008445764231286419, + "loss": 0.4647, + "step": 5661 + }, + { + "epoch": 0.3163393580467637, + "grad_norm": 1.00480318069458, + "learning_rate": 0.000844548408785298, + "loss": 0.5548, + "step": 5662 + }, + { + "epoch": 0.31639522865044556, + "grad_norm": 0.5961635112762451, + "learning_rate": 0.0008445203944419542, + "loss": 0.6061, + "step": 5663 + }, + { + "epoch": 0.31645109925412745, + "grad_norm": 0.5389479398727417, + "learning_rate": 0.0008444923800986105, + "loss": 0.4038, + "step": 5664 + }, + { + "epoch": 0.3165069698578093, + "grad_norm": 0.6081658005714417, + "learning_rate": 0.0008444643657552667, + "loss": 0.5936, + "step": 5665 + }, + { + "epoch": 0.3165628404614912, + "grad_norm": 0.52402663230896, + "learning_rate": 0.0008444363514119228, + "loss": 0.5042, + "step": 5666 + }, + { + "epoch": 0.3166187110651731, + "grad_norm": 1.1891117095947266, + "learning_rate": 0.0008444083370685792, + "loss": 0.5497, + "step": 5667 + }, + { + "epoch": 0.3166745816688549, + "grad_norm": 0.5008256435394287, + "learning_rate": 0.0008443803227252354, + "loss": 0.3671, + "step": 5668 + }, + { + "epoch": 0.3167304522725368, + "grad_norm": 0.8811865448951721, + "learning_rate": 0.0008443523083818916, + "loss": 0.4157, + "step": 5669 + }, + { + "epoch": 0.31678632287621866, + "grad_norm": 0.5213171243667603, + "learning_rate": 0.0008443242940385477, + "loss": 0.5041, + "step": 5670 + }, + { + "epoch": 0.31684219347990056, + "grad_norm": 0.5423316955566406, + "learning_rate": 0.000844296279695204, + "loss": 0.5328, + "step": 5671 + }, + { + "epoch": 0.3168980640835824, + "grad_norm": 0.4642821252346039, + "learning_rate": 0.0008442682653518602, + "loss": 0.4403, + "step": 5672 + }, + { + "epoch": 0.3169539346872643, + "grad_norm": 0.6070437431335449, + "learning_rate": 0.0008442402510085164, + "loss": 0.4994, + "step": 5673 + }, + { + "epoch": 0.3170098052909462, + "grad_norm": 0.7321299910545349, + "learning_rate": 0.0008442122366651726, + "loss": 0.4939, + "step": 5674 + }, + { + "epoch": 0.31706567589462803, + "grad_norm": 0.6181579828262329, + "learning_rate": 0.0008441842223218288, + "loss": 0.5218, + "step": 5675 + }, + { + "epoch": 0.3171215464983099, + "grad_norm": 0.6955077052116394, + "learning_rate": 0.000844156207978485, + "loss": 0.5636, + "step": 5676 + }, + { + "epoch": 0.31717741710199177, + "grad_norm": 0.39743903279304504, + "learning_rate": 0.0008441281936351412, + "loss": 0.3383, + "step": 5677 + }, + { + "epoch": 0.31723328770567366, + "grad_norm": 0.8710095882415771, + "learning_rate": 0.0008441001792917974, + "loss": 0.5801, + "step": 5678 + }, + { + "epoch": 0.31728915830935556, + "grad_norm": 0.6728836894035339, + "learning_rate": 0.0008440721649484536, + "loss": 0.3623, + "step": 5679 + }, + { + "epoch": 0.3173450289130374, + "grad_norm": 0.5670413970947266, + "learning_rate": 0.0008440441506051098, + "loss": 0.4922, + "step": 5680 + }, + { + "epoch": 0.3174008995167193, + "grad_norm": 0.40467533469200134, + "learning_rate": 0.000844016136261766, + "loss": 0.3878, + "step": 5681 + }, + { + "epoch": 0.31745677012040113, + "grad_norm": 0.9013327360153198, + "learning_rate": 0.0008439881219184222, + "loss": 0.5612, + "step": 5682 + }, + { + "epoch": 0.31751264072408303, + "grad_norm": 0.8515595197677612, + "learning_rate": 0.0008439601075750784, + "loss": 0.57, + "step": 5683 + }, + { + "epoch": 0.3175685113277649, + "grad_norm": 0.8291237354278564, + "learning_rate": 0.0008439320932317347, + "loss": 0.4553, + "step": 5684 + }, + { + "epoch": 0.31762438193144676, + "grad_norm": 0.6156446933746338, + "learning_rate": 0.0008439040788883909, + "loss": 0.487, + "step": 5685 + }, + { + "epoch": 0.31768025253512866, + "grad_norm": 0.4463694989681244, + "learning_rate": 0.0008438760645450471, + "loss": 0.4841, + "step": 5686 + }, + { + "epoch": 0.3177361231388105, + "grad_norm": 0.8430772423744202, + "learning_rate": 0.0008438480502017033, + "loss": 0.4866, + "step": 5687 + }, + { + "epoch": 0.3177919937424924, + "grad_norm": 0.5457513332366943, + "learning_rate": 0.0008438200358583596, + "loss": 0.4373, + "step": 5688 + }, + { + "epoch": 0.31784786434617424, + "grad_norm": 0.4296136200428009, + "learning_rate": 0.0008437920215150157, + "loss": 0.4353, + "step": 5689 + }, + { + "epoch": 0.31790373494985613, + "grad_norm": 0.6625611782073975, + "learning_rate": 0.0008437640071716719, + "loss": 0.4326, + "step": 5690 + }, + { + "epoch": 0.317959605553538, + "grad_norm": 0.5338588356971741, + "learning_rate": 0.0008437359928283281, + "loss": 0.4373, + "step": 5691 + }, + { + "epoch": 0.31801547615721987, + "grad_norm": 0.6292380094528198, + "learning_rate": 0.0008437079784849844, + "loss": 0.4771, + "step": 5692 + }, + { + "epoch": 0.31807134676090176, + "grad_norm": 0.5471693873405457, + "learning_rate": 0.0008436799641416405, + "loss": 0.3936, + "step": 5693 + }, + { + "epoch": 0.3181272173645836, + "grad_norm": 0.5223928689956665, + "learning_rate": 0.0008436519497982967, + "loss": 0.4135, + "step": 5694 + }, + { + "epoch": 0.3181830879682655, + "grad_norm": 1.0042071342468262, + "learning_rate": 0.000843623935454953, + "loss": 0.5143, + "step": 5695 + }, + { + "epoch": 0.3182389585719474, + "grad_norm": 1.0258073806762695, + "learning_rate": 0.0008435959211116092, + "loss": 0.3969, + "step": 5696 + }, + { + "epoch": 0.31829482917562923, + "grad_norm": 0.5824647545814514, + "learning_rate": 0.0008435679067682653, + "loss": 0.4642, + "step": 5697 + }, + { + "epoch": 0.31835069977931113, + "grad_norm": 0.5146098136901855, + "learning_rate": 0.0008435398924249215, + "loss": 0.5135, + "step": 5698 + }, + { + "epoch": 0.31840657038299297, + "grad_norm": 0.44073325395584106, + "learning_rate": 0.0008435118780815778, + "loss": 0.4064, + "step": 5699 + }, + { + "epoch": 0.31846244098667487, + "grad_norm": 0.5395626425743103, + "learning_rate": 0.000843483863738234, + "loss": 0.378, + "step": 5700 + }, + { + "epoch": 0.3185183115903567, + "grad_norm": 0.6167239546775818, + "learning_rate": 0.0008434558493948902, + "loss": 0.5089, + "step": 5701 + }, + { + "epoch": 0.3185741821940386, + "grad_norm": 0.4653970003128052, + "learning_rate": 0.0008434278350515464, + "loss": 0.3984, + "step": 5702 + }, + { + "epoch": 0.3186300527977205, + "grad_norm": 0.4422249495983124, + "learning_rate": 0.0008433998207082027, + "loss": 0.4347, + "step": 5703 + }, + { + "epoch": 0.31868592340140234, + "grad_norm": 0.4248867332935333, + "learning_rate": 0.0008433718063648588, + "loss": 0.48, + "step": 5704 + }, + { + "epoch": 0.31874179400508423, + "grad_norm": 1.554665207862854, + "learning_rate": 0.000843343792021515, + "loss": 0.4612, + "step": 5705 + }, + { + "epoch": 0.3187976646087661, + "grad_norm": 0.5066379904747009, + "learning_rate": 0.0008433157776781713, + "loss": 0.4713, + "step": 5706 + }, + { + "epoch": 0.31885353521244797, + "grad_norm": 1.0493048429489136, + "learning_rate": 0.0008432877633348275, + "loss": 0.5646, + "step": 5707 + }, + { + "epoch": 0.31890940581612987, + "grad_norm": 0.5598887801170349, + "learning_rate": 0.0008432597489914836, + "loss": 0.4778, + "step": 5708 + }, + { + "epoch": 0.3189652764198117, + "grad_norm": 0.5651242136955261, + "learning_rate": 0.0008432317346481398, + "loss": 0.4103, + "step": 5709 + }, + { + "epoch": 0.3190211470234936, + "grad_norm": 0.45974886417388916, + "learning_rate": 0.0008432037203047961, + "loss": 0.3906, + "step": 5710 + }, + { + "epoch": 0.31907701762717544, + "grad_norm": 1.9129929542541504, + "learning_rate": 0.0008431757059614523, + "loss": 0.5668, + "step": 5711 + }, + { + "epoch": 0.31913288823085734, + "grad_norm": 1.0095785856246948, + "learning_rate": 0.0008431476916181084, + "loss": 0.457, + "step": 5712 + }, + { + "epoch": 0.31918875883453923, + "grad_norm": 0.7854469418525696, + "learning_rate": 0.0008431196772747647, + "loss": 0.4301, + "step": 5713 + }, + { + "epoch": 0.3192446294382211, + "grad_norm": 0.6350091099739075, + "learning_rate": 0.0008430916629314209, + "loss": 0.4138, + "step": 5714 + }, + { + "epoch": 0.31930050004190297, + "grad_norm": 0.8685070872306824, + "learning_rate": 0.0008430636485880771, + "loss": 0.5991, + "step": 5715 + }, + { + "epoch": 0.3193563706455848, + "grad_norm": 1.211094856262207, + "learning_rate": 0.0008430356342447332, + "loss": 0.4314, + "step": 5716 + }, + { + "epoch": 0.3194122412492667, + "grad_norm": 1.197924256324768, + "learning_rate": 0.0008430076199013896, + "loss": 0.5236, + "step": 5717 + }, + { + "epoch": 0.31946811185294854, + "grad_norm": 0.46552807092666626, + "learning_rate": 0.0008429796055580458, + "loss": 0.308, + "step": 5718 + }, + { + "epoch": 0.31952398245663044, + "grad_norm": 0.6090052127838135, + "learning_rate": 0.000842951591214702, + "loss": 0.4581, + "step": 5719 + }, + { + "epoch": 0.31957985306031234, + "grad_norm": 0.45006507635116577, + "learning_rate": 0.0008429235768713582, + "loss": 0.5241, + "step": 5720 + }, + { + "epoch": 0.3196357236639942, + "grad_norm": 0.7563645243644714, + "learning_rate": 0.0008428955625280144, + "loss": 0.4227, + "step": 5721 + }, + { + "epoch": 0.31969159426767607, + "grad_norm": 0.455435574054718, + "learning_rate": 0.0008428675481846706, + "loss": 0.5446, + "step": 5722 + }, + { + "epoch": 0.3197474648713579, + "grad_norm": 0.7369527220726013, + "learning_rate": 0.0008428395338413267, + "loss": 0.6786, + "step": 5723 + }, + { + "epoch": 0.3198033354750398, + "grad_norm": 4.164901256561279, + "learning_rate": 0.000842811519497983, + "loss": 0.5396, + "step": 5724 + }, + { + "epoch": 0.3198592060787217, + "grad_norm": 0.7074564099311829, + "learning_rate": 0.0008427835051546392, + "loss": 0.5192, + "step": 5725 + }, + { + "epoch": 0.31991507668240354, + "grad_norm": 0.5243586301803589, + "learning_rate": 0.0008427554908112954, + "loss": 0.4607, + "step": 5726 + }, + { + "epoch": 0.31997094728608544, + "grad_norm": 0.6509024500846863, + "learning_rate": 0.0008427274764679516, + "loss": 0.449, + "step": 5727 + }, + { + "epoch": 0.3200268178897673, + "grad_norm": 0.5137689113616943, + "learning_rate": 0.0008426994621246078, + "loss": 0.4454, + "step": 5728 + }, + { + "epoch": 0.3200826884934492, + "grad_norm": 0.8672430515289307, + "learning_rate": 0.000842671447781264, + "loss": 0.578, + "step": 5729 + }, + { + "epoch": 0.32013855909713107, + "grad_norm": 0.5496703386306763, + "learning_rate": 0.0008426434334379202, + "loss": 0.3951, + "step": 5730 + }, + { + "epoch": 0.3201944297008129, + "grad_norm": 1.9245225191116333, + "learning_rate": 0.0008426154190945764, + "loss": 0.8048, + "step": 5731 + }, + { + "epoch": 0.3202503003044948, + "grad_norm": 3.0367231369018555, + "learning_rate": 0.0008425874047512326, + "loss": 0.5514, + "step": 5732 + }, + { + "epoch": 0.32030617090817665, + "grad_norm": 0.7176940441131592, + "learning_rate": 0.0008425593904078888, + "loss": 0.5, + "step": 5733 + }, + { + "epoch": 0.32036204151185854, + "grad_norm": 0.6960157752037048, + "learning_rate": 0.0008425313760645452, + "loss": 0.4339, + "step": 5734 + }, + { + "epoch": 0.3204179121155404, + "grad_norm": 0.6517798900604248, + "learning_rate": 0.0008425033617212013, + "loss": 0.5072, + "step": 5735 + }, + { + "epoch": 0.3204737827192223, + "grad_norm": 0.7413956522941589, + "learning_rate": 0.0008424753473778575, + "loss": 0.4949, + "step": 5736 + }, + { + "epoch": 0.3205296533229042, + "grad_norm": 0.4817879796028137, + "learning_rate": 0.0008424473330345137, + "loss": 0.5286, + "step": 5737 + }, + { + "epoch": 0.320585523926586, + "grad_norm": 0.4470662772655487, + "learning_rate": 0.00084241931869117, + "loss": 0.4514, + "step": 5738 + }, + { + "epoch": 0.3206413945302679, + "grad_norm": 0.6301866173744202, + "learning_rate": 0.0008423913043478261, + "loss": 0.5539, + "step": 5739 + }, + { + "epoch": 0.32069726513394975, + "grad_norm": 0.5976293683052063, + "learning_rate": 0.0008423632900044823, + "loss": 0.5072, + "step": 5740 + }, + { + "epoch": 0.32075313573763164, + "grad_norm": 0.7600700259208679, + "learning_rate": 0.0008423352756611385, + "loss": 0.4394, + "step": 5741 + }, + { + "epoch": 0.32080900634131354, + "grad_norm": 0.9195538759231567, + "learning_rate": 0.0008423072613177948, + "loss": 0.6619, + "step": 5742 + }, + { + "epoch": 0.3208648769449954, + "grad_norm": 0.5721901655197144, + "learning_rate": 0.0008422792469744509, + "loss": 0.4076, + "step": 5743 + }, + { + "epoch": 0.3209207475486773, + "grad_norm": 1.3041919469833374, + "learning_rate": 0.0008422512326311071, + "loss": 0.4436, + "step": 5744 + }, + { + "epoch": 0.3209766181523591, + "grad_norm": 0.5308111906051636, + "learning_rate": 0.0008422232182877634, + "loss": 0.492, + "step": 5745 + }, + { + "epoch": 0.321032488756041, + "grad_norm": 0.5494526624679565, + "learning_rate": 0.0008421952039444195, + "loss": 0.4435, + "step": 5746 + }, + { + "epoch": 0.3210883593597229, + "grad_norm": 0.5290229916572571, + "learning_rate": 0.0008421671896010757, + "loss": 0.5105, + "step": 5747 + }, + { + "epoch": 0.32114422996340475, + "grad_norm": 0.4549524486064911, + "learning_rate": 0.0008421391752577319, + "loss": 0.4853, + "step": 5748 + }, + { + "epoch": 0.32120010056708664, + "grad_norm": 0.7797739505767822, + "learning_rate": 0.0008421111609143882, + "loss": 0.3533, + "step": 5749 + }, + { + "epoch": 0.3212559711707685, + "grad_norm": 6.581976413726807, + "learning_rate": 0.0008420831465710443, + "loss": 0.4361, + "step": 5750 + }, + { + "epoch": 0.3213118417744504, + "grad_norm": 0.634002149105072, + "learning_rate": 0.0008420551322277006, + "loss": 0.5798, + "step": 5751 + }, + { + "epoch": 0.3213677123781322, + "grad_norm": 1.3383805751800537, + "learning_rate": 0.0008420271178843569, + "loss": 0.4518, + "step": 5752 + }, + { + "epoch": 0.3214235829818141, + "grad_norm": 0.5985991954803467, + "learning_rate": 0.0008419991035410131, + "loss": 0.4177, + "step": 5753 + }, + { + "epoch": 0.321479453585496, + "grad_norm": 0.5462119579315186, + "learning_rate": 0.0008419710891976692, + "loss": 0.4907, + "step": 5754 + }, + { + "epoch": 0.32153532418917785, + "grad_norm": 0.4514373242855072, + "learning_rate": 0.0008419430748543254, + "loss": 0.4925, + "step": 5755 + }, + { + "epoch": 0.32159119479285975, + "grad_norm": 0.5686076283454895, + "learning_rate": 0.0008419150605109817, + "loss": 0.4622, + "step": 5756 + }, + { + "epoch": 0.3216470653965416, + "grad_norm": 0.9872475266456604, + "learning_rate": 0.0008418870461676379, + "loss": 0.4351, + "step": 5757 + }, + { + "epoch": 0.3217029360002235, + "grad_norm": 0.5440011024475098, + "learning_rate": 0.000841859031824294, + "loss": 0.4533, + "step": 5758 + }, + { + "epoch": 0.3217588066039054, + "grad_norm": 1.0949106216430664, + "learning_rate": 0.0008418310174809503, + "loss": 0.4395, + "step": 5759 + }, + { + "epoch": 0.3218146772075872, + "grad_norm": 0.5777234435081482, + "learning_rate": 0.0008418030031376065, + "loss": 0.4612, + "step": 5760 + }, + { + "epoch": 0.3218705478112691, + "grad_norm": 0.5535335540771484, + "learning_rate": 0.0008417749887942627, + "loss": 0.5199, + "step": 5761 + }, + { + "epoch": 0.32192641841495095, + "grad_norm": 0.6251397728919983, + "learning_rate": 0.0008417469744509188, + "loss": 0.4816, + "step": 5762 + }, + { + "epoch": 0.32198228901863285, + "grad_norm": 0.5595818758010864, + "learning_rate": 0.0008417189601075751, + "loss": 0.4346, + "step": 5763 + }, + { + "epoch": 0.32203815962231475, + "grad_norm": 0.4895046353340149, + "learning_rate": 0.0008416909457642313, + "loss": 0.5196, + "step": 5764 + }, + { + "epoch": 0.3220940302259966, + "grad_norm": 0.5337347984313965, + "learning_rate": 0.0008416629314208874, + "loss": 0.4458, + "step": 5765 + }, + { + "epoch": 0.3221499008296785, + "grad_norm": 0.5535709261894226, + "learning_rate": 0.0008416349170775437, + "loss": 0.5972, + "step": 5766 + }, + { + "epoch": 0.3222057714333603, + "grad_norm": 1.2344169616699219, + "learning_rate": 0.0008416069027342, + "loss": 0.4285, + "step": 5767 + }, + { + "epoch": 0.3222616420370422, + "grad_norm": 0.6218457221984863, + "learning_rate": 0.0008415788883908562, + "loss": 0.5452, + "step": 5768 + }, + { + "epoch": 0.32231751264072406, + "grad_norm": 0.7028220891952515, + "learning_rate": 0.0008415508740475123, + "loss": 0.5537, + "step": 5769 + }, + { + "epoch": 0.32237338324440595, + "grad_norm": 0.4943113923072815, + "learning_rate": 0.0008415228597041686, + "loss": 0.4439, + "step": 5770 + }, + { + "epoch": 0.32242925384808785, + "grad_norm": 1.0100640058517456, + "learning_rate": 0.0008414948453608248, + "loss": 0.5104, + "step": 5771 + }, + { + "epoch": 0.3224851244517697, + "grad_norm": 0.6047587990760803, + "learning_rate": 0.000841466831017481, + "loss": 0.4522, + "step": 5772 + }, + { + "epoch": 0.3225409950554516, + "grad_norm": 0.6005681753158569, + "learning_rate": 0.0008414388166741371, + "loss": 0.444, + "step": 5773 + }, + { + "epoch": 0.3225968656591334, + "grad_norm": 0.5053476095199585, + "learning_rate": 0.0008414108023307934, + "loss": 0.4564, + "step": 5774 + }, + { + "epoch": 0.3226527362628153, + "grad_norm": 0.5141583681106567, + "learning_rate": 0.0008413827879874496, + "loss": 0.5861, + "step": 5775 + }, + { + "epoch": 0.3227086068664972, + "grad_norm": 0.45340272784233093, + "learning_rate": 0.0008413547736441058, + "loss": 0.4405, + "step": 5776 + }, + { + "epoch": 0.32276447747017906, + "grad_norm": 1.0324243307113647, + "learning_rate": 0.000841326759300762, + "loss": 0.4327, + "step": 5777 + }, + { + "epoch": 0.32282034807386095, + "grad_norm": 0.4770086705684662, + "learning_rate": 0.0008412987449574182, + "loss": 0.5238, + "step": 5778 + }, + { + "epoch": 0.3228762186775428, + "grad_norm": 0.759972095489502, + "learning_rate": 0.0008412707306140744, + "loss": 0.6065, + "step": 5779 + }, + { + "epoch": 0.3229320892812247, + "grad_norm": 0.7512027025222778, + "learning_rate": 0.0008412427162707306, + "loss": 0.455, + "step": 5780 + }, + { + "epoch": 0.3229879598849066, + "grad_norm": 0.46663129329681396, + "learning_rate": 0.0008412147019273868, + "loss": 0.5621, + "step": 5781 + }, + { + "epoch": 0.3230438304885884, + "grad_norm": 0.5139055252075195, + "learning_rate": 0.000841186687584043, + "loss": 0.4483, + "step": 5782 + }, + { + "epoch": 0.3230997010922703, + "grad_norm": 0.7393901348114014, + "learning_rate": 0.0008411586732406992, + "loss": 0.4765, + "step": 5783 + }, + { + "epoch": 0.32315557169595216, + "grad_norm": 0.5088405013084412, + "learning_rate": 0.0008411306588973556, + "loss": 0.4795, + "step": 5784 + }, + { + "epoch": 0.32321144229963406, + "grad_norm": 0.6588709354400635, + "learning_rate": 0.0008411026445540117, + "loss": 0.4002, + "step": 5785 + }, + { + "epoch": 0.3232673129033159, + "grad_norm": 0.531298816204071, + "learning_rate": 0.0008410746302106679, + "loss": 0.4083, + "step": 5786 + }, + { + "epoch": 0.3233231835069978, + "grad_norm": 0.6255412101745605, + "learning_rate": 0.0008410466158673241, + "loss": 0.5018, + "step": 5787 + }, + { + "epoch": 0.3233790541106797, + "grad_norm": 1.0007073879241943, + "learning_rate": 0.0008410186015239803, + "loss": 0.4631, + "step": 5788 + }, + { + "epoch": 0.3234349247143615, + "grad_norm": 0.8083999156951904, + "learning_rate": 0.0008409905871806365, + "loss": 0.4673, + "step": 5789 + }, + { + "epoch": 0.3234907953180434, + "grad_norm": 0.6917881965637207, + "learning_rate": 0.0008409625728372927, + "loss": 0.4282, + "step": 5790 + }, + { + "epoch": 0.32354666592172526, + "grad_norm": 1.7364935874938965, + "learning_rate": 0.000840934558493949, + "loss": 0.4769, + "step": 5791 + }, + { + "epoch": 0.32360253652540716, + "grad_norm": 0.7396431565284729, + "learning_rate": 0.0008409065441506051, + "loss": 0.5379, + "step": 5792 + }, + { + "epoch": 0.32365840712908905, + "grad_norm": 0.7546364665031433, + "learning_rate": 0.0008408785298072613, + "loss": 0.4511, + "step": 5793 + }, + { + "epoch": 0.3237142777327709, + "grad_norm": 0.5039647221565247, + "learning_rate": 0.0008408505154639175, + "loss": 0.3748, + "step": 5794 + }, + { + "epoch": 0.3237701483364528, + "grad_norm": 1.0018582344055176, + "learning_rate": 0.0008408225011205738, + "loss": 0.3474, + "step": 5795 + }, + { + "epoch": 0.32382601894013463, + "grad_norm": 0.9577719569206238, + "learning_rate": 0.0008407944867772299, + "loss": 0.5899, + "step": 5796 + }, + { + "epoch": 0.3238818895438165, + "grad_norm": 0.596130907535553, + "learning_rate": 0.0008407664724338861, + "loss": 0.4384, + "step": 5797 + }, + { + "epoch": 0.3239377601474984, + "grad_norm": 1.1732115745544434, + "learning_rate": 0.0008407384580905424, + "loss": 0.4682, + "step": 5798 + }, + { + "epoch": 0.32399363075118026, + "grad_norm": 2.429823398590088, + "learning_rate": 0.0008407104437471986, + "loss": 0.3756, + "step": 5799 + }, + { + "epoch": 0.32404950135486216, + "grad_norm": 1.6957430839538574, + "learning_rate": 0.0008406824294038547, + "loss": 0.433, + "step": 5800 + }, + { + "epoch": 0.324105371958544, + "grad_norm": 0.5568557977676392, + "learning_rate": 0.000840654415060511, + "loss": 0.5079, + "step": 5801 + }, + { + "epoch": 0.3241612425622259, + "grad_norm": 0.5447269082069397, + "learning_rate": 0.0008406264007171673, + "loss": 0.533, + "step": 5802 + }, + { + "epoch": 0.32421711316590773, + "grad_norm": 0.6060471534729004, + "learning_rate": 0.0008405983863738235, + "loss": 0.5073, + "step": 5803 + }, + { + "epoch": 0.32427298376958963, + "grad_norm": 0.9440334439277649, + "learning_rate": 0.0008405703720304796, + "loss": 0.5651, + "step": 5804 + }, + { + "epoch": 0.3243288543732715, + "grad_norm": 0.5530333518981934, + "learning_rate": 0.0008405423576871359, + "loss": 0.4977, + "step": 5805 + }, + { + "epoch": 0.32438472497695336, + "grad_norm": 1.8049042224884033, + "learning_rate": 0.0008405143433437921, + "loss": 0.4804, + "step": 5806 + }, + { + "epoch": 0.32444059558063526, + "grad_norm": 0.4725913405418396, + "learning_rate": 0.0008404863290004483, + "loss": 0.4666, + "step": 5807 + }, + { + "epoch": 0.3244964661843171, + "grad_norm": 0.5317930579185486, + "learning_rate": 0.0008404583146571044, + "loss": 0.6157, + "step": 5808 + }, + { + "epoch": 0.324552336787999, + "grad_norm": 2.429047107696533, + "learning_rate": 0.0008404303003137607, + "loss": 0.3577, + "step": 5809 + }, + { + "epoch": 0.3246082073916809, + "grad_norm": 0.5937598347663879, + "learning_rate": 0.0008404022859704169, + "loss": 0.5113, + "step": 5810 + }, + { + "epoch": 0.32466407799536273, + "grad_norm": 0.5039104223251343, + "learning_rate": 0.000840374271627073, + "loss": 0.425, + "step": 5811 + }, + { + "epoch": 0.3247199485990446, + "grad_norm": 0.6126775741577148, + "learning_rate": 0.0008403462572837292, + "loss": 0.4755, + "step": 5812 + }, + { + "epoch": 0.32477581920272647, + "grad_norm": 0.518125593662262, + "learning_rate": 0.0008403182429403855, + "loss": 0.5124, + "step": 5813 + }, + { + "epoch": 0.32483168980640836, + "grad_norm": 0.5383478999137878, + "learning_rate": 0.0008402902285970417, + "loss": 0.5409, + "step": 5814 + }, + { + "epoch": 0.3248875604100902, + "grad_norm": 0.5686440467834473, + "learning_rate": 0.0008402622142536978, + "loss": 0.5309, + "step": 5815 + }, + { + "epoch": 0.3249434310137721, + "grad_norm": 0.6829490661621094, + "learning_rate": 0.0008402341999103541, + "loss": 0.4529, + "step": 5816 + }, + { + "epoch": 0.324999301617454, + "grad_norm": 0.4380023181438446, + "learning_rate": 0.0008402061855670104, + "loss": 0.4989, + "step": 5817 + }, + { + "epoch": 0.32505517222113584, + "grad_norm": 0.5205996036529541, + "learning_rate": 0.0008401781712236666, + "loss": 0.5142, + "step": 5818 + }, + { + "epoch": 0.32511104282481773, + "grad_norm": 0.6291648745536804, + "learning_rate": 0.0008401501568803227, + "loss": 0.5183, + "step": 5819 + }, + { + "epoch": 0.32516691342849957, + "grad_norm": 0.57550048828125, + "learning_rate": 0.000840122142536979, + "loss": 0.5508, + "step": 5820 + }, + { + "epoch": 0.32522278403218147, + "grad_norm": 0.5553627014160156, + "learning_rate": 0.0008400941281936352, + "loss": 0.3826, + "step": 5821 + }, + { + "epoch": 0.32527865463586336, + "grad_norm": 0.9448410868644714, + "learning_rate": 0.0008400661138502914, + "loss": 0.3938, + "step": 5822 + }, + { + "epoch": 0.3253345252395452, + "grad_norm": 11.009489059448242, + "learning_rate": 0.0008400380995069476, + "loss": 0.4293, + "step": 5823 + }, + { + "epoch": 0.3253903958432271, + "grad_norm": 0.5300685167312622, + "learning_rate": 0.0008400100851636038, + "loss": 0.4204, + "step": 5824 + }, + { + "epoch": 0.32544626644690894, + "grad_norm": 0.4270522892475128, + "learning_rate": 0.00083998207082026, + "loss": 0.5442, + "step": 5825 + }, + { + "epoch": 0.32550213705059083, + "grad_norm": 0.4659240245819092, + "learning_rate": 0.0008399540564769162, + "loss": 0.4636, + "step": 5826 + }, + { + "epoch": 0.32555800765427273, + "grad_norm": 0.5245846509933472, + "learning_rate": 0.0008399260421335724, + "loss": 0.5033, + "step": 5827 + }, + { + "epoch": 0.32561387825795457, + "grad_norm": 0.7690344452857971, + "learning_rate": 0.0008398980277902286, + "loss": 0.4525, + "step": 5828 + }, + { + "epoch": 0.32566974886163647, + "grad_norm": 0.5502141714096069, + "learning_rate": 0.0008398700134468848, + "loss": 0.4809, + "step": 5829 + }, + { + "epoch": 0.3257256194653183, + "grad_norm": 0.8152344822883606, + "learning_rate": 0.000839841999103541, + "loss": 0.5407, + "step": 5830 + }, + { + "epoch": 0.3257814900690002, + "grad_norm": 0.6886998414993286, + "learning_rate": 0.0008398139847601972, + "loss": 0.5508, + "step": 5831 + }, + { + "epoch": 0.32583736067268204, + "grad_norm": 0.6517998576164246, + "learning_rate": 0.0008397859704168534, + "loss": 0.5082, + "step": 5832 + }, + { + "epoch": 0.32589323127636394, + "grad_norm": 0.48014524579048157, + "learning_rate": 0.0008397579560735096, + "loss": 0.4812, + "step": 5833 + }, + { + "epoch": 0.32594910188004583, + "grad_norm": 3.0967001914978027, + "learning_rate": 0.0008397299417301659, + "loss": 0.4284, + "step": 5834 + }, + { + "epoch": 0.3260049724837277, + "grad_norm": 0.5601823925971985, + "learning_rate": 0.0008397019273868221, + "loss": 0.4884, + "step": 5835 + }, + { + "epoch": 0.32606084308740957, + "grad_norm": 0.5677313208580017, + "learning_rate": 0.0008396739130434783, + "loss": 0.4124, + "step": 5836 + }, + { + "epoch": 0.3261167136910914, + "grad_norm": 0.5907091498374939, + "learning_rate": 0.0008396458987001346, + "loss": 0.3868, + "step": 5837 + }, + { + "epoch": 0.3261725842947733, + "grad_norm": 0.7597725987434387, + "learning_rate": 0.0008396178843567907, + "loss": 0.518, + "step": 5838 + }, + { + "epoch": 0.3262284548984552, + "grad_norm": 0.6268984079360962, + "learning_rate": 0.0008395898700134469, + "loss": 0.4676, + "step": 5839 + }, + { + "epoch": 0.32628432550213704, + "grad_norm": 0.5040469765663147, + "learning_rate": 0.0008395618556701031, + "loss": 0.6082, + "step": 5840 + }, + { + "epoch": 0.32634019610581894, + "grad_norm": 0.6248039603233337, + "learning_rate": 0.0008395338413267594, + "loss": 0.4824, + "step": 5841 + }, + { + "epoch": 0.3263960667095008, + "grad_norm": 0.4690035581588745, + "learning_rate": 0.0008395058269834155, + "loss": 0.4883, + "step": 5842 + }, + { + "epoch": 0.32645193731318267, + "grad_norm": 0.9344251751899719, + "learning_rate": 0.0008394778126400717, + "loss": 0.5849, + "step": 5843 + }, + { + "epoch": 0.32650780791686457, + "grad_norm": 0.5747566819190979, + "learning_rate": 0.0008394497982967279, + "loss": 0.6071, + "step": 5844 + }, + { + "epoch": 0.3265636785205464, + "grad_norm": 0.5151383876800537, + "learning_rate": 0.0008394217839533842, + "loss": 0.3645, + "step": 5845 + }, + { + "epoch": 0.3266195491242283, + "grad_norm": 0.421424001455307, + "learning_rate": 0.0008393937696100403, + "loss": 0.3499, + "step": 5846 + }, + { + "epoch": 0.32667541972791014, + "grad_norm": 1.0172964334487915, + "learning_rate": 0.0008393657552666965, + "loss": 0.5416, + "step": 5847 + }, + { + "epoch": 0.32673129033159204, + "grad_norm": 0.49959075450897217, + "learning_rate": 0.0008393377409233528, + "loss": 0.4521, + "step": 5848 + }, + { + "epoch": 0.3267871609352739, + "grad_norm": 2.364264965057373, + "learning_rate": 0.000839309726580009, + "loss": 0.5334, + "step": 5849 + }, + { + "epoch": 0.3268430315389558, + "grad_norm": 0.4716517925262451, + "learning_rate": 0.0008392817122366651, + "loss": 0.496, + "step": 5850 + }, + { + "epoch": 0.32689890214263767, + "grad_norm": 0.7658815979957581, + "learning_rate": 0.0008392536978933213, + "loss": 0.4265, + "step": 5851 + }, + { + "epoch": 0.3269547727463195, + "grad_norm": 0.6428964138031006, + "learning_rate": 0.0008392256835499777, + "loss": 0.3631, + "step": 5852 + }, + { + "epoch": 0.3270106433500014, + "grad_norm": 0.5278635621070862, + "learning_rate": 0.0008391976692066338, + "loss": 0.4845, + "step": 5853 + }, + { + "epoch": 0.32706651395368325, + "grad_norm": 0.5636388659477234, + "learning_rate": 0.00083916965486329, + "loss": 0.4047, + "step": 5854 + }, + { + "epoch": 0.32712238455736514, + "grad_norm": 0.8224744200706482, + "learning_rate": 0.0008391416405199463, + "loss": 0.411, + "step": 5855 + }, + { + "epoch": 0.32717825516104704, + "grad_norm": 0.7387667894363403, + "learning_rate": 0.0008391136261766025, + "loss": 0.4951, + "step": 5856 + }, + { + "epoch": 0.3272341257647289, + "grad_norm": 0.6481671333312988, + "learning_rate": 0.0008390856118332586, + "loss": 0.3782, + "step": 5857 + }, + { + "epoch": 0.3272899963684108, + "grad_norm": 0.4536457657814026, + "learning_rate": 0.0008390575974899148, + "loss": 0.3496, + "step": 5858 + }, + { + "epoch": 0.3273458669720926, + "grad_norm": 0.9402965903282166, + "learning_rate": 0.0008390295831465711, + "loss": 0.5404, + "step": 5859 + }, + { + "epoch": 0.3274017375757745, + "grad_norm": 0.6868078708648682, + "learning_rate": 0.0008390015688032273, + "loss": 0.4971, + "step": 5860 + }, + { + "epoch": 0.3274576081794564, + "grad_norm": 0.8112798929214478, + "learning_rate": 0.0008389735544598834, + "loss": 0.6096, + "step": 5861 + }, + { + "epoch": 0.32751347878313825, + "grad_norm": 0.5263088941574097, + "learning_rate": 0.0008389455401165397, + "loss": 0.3897, + "step": 5862 + }, + { + "epoch": 0.32756934938682014, + "grad_norm": 0.5304678678512573, + "learning_rate": 0.0008389175257731959, + "loss": 0.4443, + "step": 5863 + }, + { + "epoch": 0.327625219990502, + "grad_norm": 0.5453434586524963, + "learning_rate": 0.0008388895114298521, + "loss": 0.3684, + "step": 5864 + }, + { + "epoch": 0.3276810905941839, + "grad_norm": 0.815732479095459, + "learning_rate": 0.0008388614970865082, + "loss": 0.4095, + "step": 5865 + }, + { + "epoch": 0.3277369611978657, + "grad_norm": 0.3897377848625183, + "learning_rate": 0.0008388334827431645, + "loss": 0.3945, + "step": 5866 + }, + { + "epoch": 0.3277928318015476, + "grad_norm": 0.6417025923728943, + "learning_rate": 0.0008388054683998208, + "loss": 0.4955, + "step": 5867 + }, + { + "epoch": 0.3278487024052295, + "grad_norm": 0.4803791642189026, + "learning_rate": 0.000838777454056477, + "loss": 0.5128, + "step": 5868 + }, + { + "epoch": 0.32790457300891135, + "grad_norm": 0.689471423625946, + "learning_rate": 0.0008387494397131332, + "loss": 0.5119, + "step": 5869 + }, + { + "epoch": 0.32796044361259324, + "grad_norm": 0.4585915803909302, + "learning_rate": 0.0008387214253697894, + "loss": 0.5187, + "step": 5870 + }, + { + "epoch": 0.3280163142162751, + "grad_norm": 0.8238504528999329, + "learning_rate": 0.0008386934110264456, + "loss": 0.4803, + "step": 5871 + }, + { + "epoch": 0.328072184819957, + "grad_norm": 1.3114200830459595, + "learning_rate": 0.0008386653966831017, + "loss": 0.4459, + "step": 5872 + }, + { + "epoch": 0.3281280554236389, + "grad_norm": 0.44244739413261414, + "learning_rate": 0.000838637382339758, + "loss": 0.3846, + "step": 5873 + }, + { + "epoch": 0.3281839260273207, + "grad_norm": 0.5128824710845947, + "learning_rate": 0.0008386093679964142, + "loss": 0.5801, + "step": 5874 + }, + { + "epoch": 0.3282397966310026, + "grad_norm": 0.4500565528869629, + "learning_rate": 0.0008385813536530704, + "loss": 0.3449, + "step": 5875 + }, + { + "epoch": 0.32829566723468445, + "grad_norm": 0.39117172360420227, + "learning_rate": 0.0008385533393097265, + "loss": 0.3886, + "step": 5876 + }, + { + "epoch": 0.32835153783836635, + "grad_norm": 0.3558556139469147, + "learning_rate": 0.0008385253249663828, + "loss": 0.4139, + "step": 5877 + }, + { + "epoch": 0.32840740844204824, + "grad_norm": 0.5822017192840576, + "learning_rate": 0.000838497310623039, + "loss": 0.3802, + "step": 5878 + }, + { + "epoch": 0.3284632790457301, + "grad_norm": 0.41836312413215637, + "learning_rate": 0.0008384692962796952, + "loss": 0.4334, + "step": 5879 + }, + { + "epoch": 0.328519149649412, + "grad_norm": 0.38732874393463135, + "learning_rate": 0.0008384412819363514, + "loss": 0.4449, + "step": 5880 + }, + { + "epoch": 0.3285750202530938, + "grad_norm": 0.4562824070453644, + "learning_rate": 0.0008384132675930076, + "loss": 0.4645, + "step": 5881 + }, + { + "epoch": 0.3286308908567757, + "grad_norm": 0.5329294204711914, + "learning_rate": 0.0008383852532496638, + "loss": 0.5096, + "step": 5882 + }, + { + "epoch": 0.32868676146045755, + "grad_norm": 0.48447686433792114, + "learning_rate": 0.00083835723890632, + "loss": 0.5696, + "step": 5883 + }, + { + "epoch": 0.32874263206413945, + "grad_norm": 0.5979647040367126, + "learning_rate": 0.0008383292245629762, + "loss": 0.5066, + "step": 5884 + }, + { + "epoch": 0.32879850266782135, + "grad_norm": 0.5574808120727539, + "learning_rate": 0.0008383012102196325, + "loss": 0.4819, + "step": 5885 + }, + { + "epoch": 0.3288543732715032, + "grad_norm": 2.000960111618042, + "learning_rate": 0.0008382731958762887, + "loss": 0.4507, + "step": 5886 + }, + { + "epoch": 0.3289102438751851, + "grad_norm": 1.0557894706726074, + "learning_rate": 0.000838245181532945, + "loss": 0.482, + "step": 5887 + }, + { + "epoch": 0.3289661144788669, + "grad_norm": 0.5398057103157043, + "learning_rate": 0.0008382171671896011, + "loss": 0.4478, + "step": 5888 + }, + { + "epoch": 0.3290219850825488, + "grad_norm": 2.3232414722442627, + "learning_rate": 0.0008381891528462573, + "loss": 0.4322, + "step": 5889 + }, + { + "epoch": 0.3290778556862307, + "grad_norm": 1.204603672027588, + "learning_rate": 0.0008381611385029135, + "loss": 0.847, + "step": 5890 + }, + { + "epoch": 0.32913372628991255, + "grad_norm": 3.0454654693603516, + "learning_rate": 0.0008381331241595698, + "loss": 0.4349, + "step": 5891 + }, + { + "epoch": 0.32918959689359445, + "grad_norm": 0.5506415367126465, + "learning_rate": 0.0008381051098162259, + "loss": 0.4613, + "step": 5892 + }, + { + "epoch": 0.3292454674972763, + "grad_norm": 0.532383382320404, + "learning_rate": 0.0008380770954728821, + "loss": 0.4338, + "step": 5893 + }, + { + "epoch": 0.3293013381009582, + "grad_norm": 0.5107948780059814, + "learning_rate": 0.0008380490811295384, + "loss": 0.4929, + "step": 5894 + }, + { + "epoch": 0.3293572087046401, + "grad_norm": 0.5045707821846008, + "learning_rate": 0.0008380210667861945, + "loss": 0.4968, + "step": 5895 + }, + { + "epoch": 0.3294130793083219, + "grad_norm": 0.49284127354621887, + "learning_rate": 0.0008379930524428507, + "loss": 0.4737, + "step": 5896 + }, + { + "epoch": 0.3294689499120038, + "grad_norm": 0.5354411602020264, + "learning_rate": 0.0008379650380995069, + "loss": 0.4586, + "step": 5897 + }, + { + "epoch": 0.32952482051568566, + "grad_norm": 0.5591123700141907, + "learning_rate": 0.0008379370237561632, + "loss": 0.4545, + "step": 5898 + }, + { + "epoch": 0.32958069111936755, + "grad_norm": 0.4780065715312958, + "learning_rate": 0.0008379090094128193, + "loss": 0.4279, + "step": 5899 + }, + { + "epoch": 0.3296365617230494, + "grad_norm": 0.8920430541038513, + "learning_rate": 0.0008378809950694755, + "loss": 0.5389, + "step": 5900 + }, + { + "epoch": 0.3296924323267313, + "grad_norm": 0.9430884122848511, + "learning_rate": 0.0008378529807261319, + "loss": 0.3746, + "step": 5901 + }, + { + "epoch": 0.3297483029304132, + "grad_norm": 0.44697996973991394, + "learning_rate": 0.0008378249663827881, + "loss": 0.4062, + "step": 5902 + }, + { + "epoch": 0.329804173534095, + "grad_norm": 0.7806053161621094, + "learning_rate": 0.0008377969520394442, + "loss": 0.5763, + "step": 5903 + }, + { + "epoch": 0.3298600441377769, + "grad_norm": 0.6445209383964539, + "learning_rate": 0.0008377689376961004, + "loss": 0.5867, + "step": 5904 + }, + { + "epoch": 0.32991591474145876, + "grad_norm": 0.718333899974823, + "learning_rate": 0.0008377409233527567, + "loss": 0.6174, + "step": 5905 + }, + { + "epoch": 0.32997178534514066, + "grad_norm": 0.4744546711444855, + "learning_rate": 0.0008377129090094129, + "loss": 0.4051, + "step": 5906 + }, + { + "epoch": 0.33002765594882255, + "grad_norm": 0.7473469972610474, + "learning_rate": 0.000837684894666069, + "loss": 0.4494, + "step": 5907 + }, + { + "epoch": 0.3300835265525044, + "grad_norm": 0.4381925165653229, + "learning_rate": 0.0008376568803227253, + "loss": 0.5233, + "step": 5908 + }, + { + "epoch": 0.3301393971561863, + "grad_norm": 1.063980221748352, + "learning_rate": 0.0008376288659793815, + "loss": 0.4466, + "step": 5909 + }, + { + "epoch": 0.3301952677598681, + "grad_norm": 0.8965221643447876, + "learning_rate": 0.0008376008516360377, + "loss": 0.803, + "step": 5910 + }, + { + "epoch": 0.33025113836355, + "grad_norm": 0.455841064453125, + "learning_rate": 0.0008375728372926938, + "loss": 0.5074, + "step": 5911 + }, + { + "epoch": 0.33030700896723186, + "grad_norm": 1.0197128057479858, + "learning_rate": 0.0008375448229493501, + "loss": 0.4497, + "step": 5912 + }, + { + "epoch": 0.33036287957091376, + "grad_norm": 0.5721043348312378, + "learning_rate": 0.0008375168086060063, + "loss": 0.7176, + "step": 5913 + }, + { + "epoch": 0.33041875017459565, + "grad_norm": 0.4164842665195465, + "learning_rate": 0.0008374887942626624, + "loss": 0.4796, + "step": 5914 + }, + { + "epoch": 0.3304746207782775, + "grad_norm": 0.37165722250938416, + "learning_rate": 0.0008374607799193186, + "loss": 0.421, + "step": 5915 + }, + { + "epoch": 0.3305304913819594, + "grad_norm": 0.7646094560623169, + "learning_rate": 0.0008374327655759749, + "loss": 0.4026, + "step": 5916 + }, + { + "epoch": 0.33058636198564123, + "grad_norm": 0.5235036015510559, + "learning_rate": 0.0008374047512326311, + "loss": 0.3691, + "step": 5917 + }, + { + "epoch": 0.3306422325893231, + "grad_norm": 0.5434574484825134, + "learning_rate": 0.0008373767368892872, + "loss": 0.5494, + "step": 5918 + }, + { + "epoch": 0.330698103193005, + "grad_norm": 0.4330061078071594, + "learning_rate": 0.0008373487225459436, + "loss": 0.4961, + "step": 5919 + }, + { + "epoch": 0.33075397379668686, + "grad_norm": 0.4893794655799866, + "learning_rate": 0.0008373207082025998, + "loss": 0.41, + "step": 5920 + }, + { + "epoch": 0.33080984440036876, + "grad_norm": 0.5035836696624756, + "learning_rate": 0.000837292693859256, + "loss": 0.5408, + "step": 5921 + }, + { + "epoch": 0.3308657150040506, + "grad_norm": 3.476487874984741, + "learning_rate": 0.0008372646795159121, + "loss": 0.5533, + "step": 5922 + }, + { + "epoch": 0.3309215856077325, + "grad_norm": 0.5900025963783264, + "learning_rate": 0.0008372366651725684, + "loss": 0.4402, + "step": 5923 + }, + { + "epoch": 0.3309774562114144, + "grad_norm": 0.5224894285202026, + "learning_rate": 0.0008372086508292246, + "loss": 0.4957, + "step": 5924 + }, + { + "epoch": 0.33103332681509623, + "grad_norm": 0.5066608190536499, + "learning_rate": 0.0008371806364858808, + "loss": 0.6015, + "step": 5925 + }, + { + "epoch": 0.3310891974187781, + "grad_norm": 0.8200652599334717, + "learning_rate": 0.000837152622142537, + "loss": 0.5577, + "step": 5926 + }, + { + "epoch": 0.33114506802245997, + "grad_norm": 0.7393449544906616, + "learning_rate": 0.0008371246077991932, + "loss": 0.7152, + "step": 5927 + }, + { + "epoch": 0.33120093862614186, + "grad_norm": 0.7821397185325623, + "learning_rate": 0.0008370965934558494, + "loss": 0.4953, + "step": 5928 + }, + { + "epoch": 0.3312568092298237, + "grad_norm": 0.4219760000705719, + "learning_rate": 0.0008370685791125056, + "loss": 0.3928, + "step": 5929 + }, + { + "epoch": 0.3313126798335056, + "grad_norm": 2.0250930786132812, + "learning_rate": 0.0008370405647691618, + "loss": 0.3671, + "step": 5930 + }, + { + "epoch": 0.3313685504371875, + "grad_norm": 0.675646185874939, + "learning_rate": 0.000837012550425818, + "loss": 0.6318, + "step": 5931 + }, + { + "epoch": 0.33142442104086933, + "grad_norm": 0.39513352513313293, + "learning_rate": 0.0008369845360824742, + "loss": 0.5865, + "step": 5932 + }, + { + "epoch": 0.33148029164455123, + "grad_norm": 0.5595794916152954, + "learning_rate": 0.0008369565217391306, + "loss": 0.446, + "step": 5933 + }, + { + "epoch": 0.33153616224823307, + "grad_norm": 0.4058817923069, + "learning_rate": 0.0008369285073957866, + "loss": 0.498, + "step": 5934 + }, + { + "epoch": 0.33159203285191496, + "grad_norm": 0.5330256223678589, + "learning_rate": 0.0008369004930524429, + "loss": 0.4379, + "step": 5935 + }, + { + "epoch": 0.33164790345559686, + "grad_norm": 0.5514656901359558, + "learning_rate": 0.0008368724787090991, + "loss": 0.5145, + "step": 5936 + }, + { + "epoch": 0.3317037740592787, + "grad_norm": 0.48577630519866943, + "learning_rate": 0.0008368444643657553, + "loss": 0.3977, + "step": 5937 + }, + { + "epoch": 0.3317596446629606, + "grad_norm": 0.5819946527481079, + "learning_rate": 0.0008368164500224115, + "loss": 0.4117, + "step": 5938 + }, + { + "epoch": 0.33181551526664244, + "grad_norm": 0.5892490744590759, + "learning_rate": 0.0008367884356790677, + "loss": 0.5014, + "step": 5939 + }, + { + "epoch": 0.33187138587032433, + "grad_norm": 0.9845032095909119, + "learning_rate": 0.000836760421335724, + "loss": 0.5683, + "step": 5940 + }, + { + "epoch": 0.3319272564740062, + "grad_norm": 2.3519675731658936, + "learning_rate": 0.0008367324069923801, + "loss": 0.4674, + "step": 5941 + }, + { + "epoch": 0.33198312707768807, + "grad_norm": 0.529681384563446, + "learning_rate": 0.0008367043926490363, + "loss": 0.3924, + "step": 5942 + }, + { + "epoch": 0.33203899768136996, + "grad_norm": 1.8950473070144653, + "learning_rate": 0.0008366763783056925, + "loss": 0.7954, + "step": 5943 + }, + { + "epoch": 0.3320948682850518, + "grad_norm": 0.5739957094192505, + "learning_rate": 0.0008366483639623488, + "loss": 0.5358, + "step": 5944 + }, + { + "epoch": 0.3321507388887337, + "grad_norm": 0.42215290665626526, + "learning_rate": 0.0008366203496190049, + "loss": 0.4893, + "step": 5945 + }, + { + "epoch": 0.33220660949241554, + "grad_norm": 0.4647344648838043, + "learning_rate": 0.0008365923352756611, + "loss": 0.4358, + "step": 5946 + }, + { + "epoch": 0.33226248009609743, + "grad_norm": 0.6621007919311523, + "learning_rate": 0.0008365643209323174, + "loss": 0.5794, + "step": 5947 + }, + { + "epoch": 0.33231835069977933, + "grad_norm": 0.5649006366729736, + "learning_rate": 0.0008365363065889736, + "loss": 0.4553, + "step": 5948 + }, + { + "epoch": 0.33237422130346117, + "grad_norm": 0.6902429461479187, + "learning_rate": 0.0008365082922456297, + "loss": 0.4509, + "step": 5949 + }, + { + "epoch": 0.33243009190714307, + "grad_norm": 2.195847511291504, + "learning_rate": 0.0008364802779022859, + "loss": 0.5192, + "step": 5950 + }, + { + "epoch": 0.3324859625108249, + "grad_norm": 0.5779674053192139, + "learning_rate": 0.0008364522635589423, + "loss": 0.4733, + "step": 5951 + }, + { + "epoch": 0.3325418331145068, + "grad_norm": 0.6292220950126648, + "learning_rate": 0.0008364242492155985, + "loss": 0.4563, + "step": 5952 + }, + { + "epoch": 0.3325977037181887, + "grad_norm": 0.539089560508728, + "learning_rate": 0.0008363962348722546, + "loss": 0.4768, + "step": 5953 + }, + { + "epoch": 0.33265357432187054, + "grad_norm": 0.5532649755477905, + "learning_rate": 0.0008363682205289108, + "loss": 0.4889, + "step": 5954 + }, + { + "epoch": 0.33270944492555243, + "grad_norm": 0.4986051321029663, + "learning_rate": 0.0008363402061855671, + "loss": 0.4544, + "step": 5955 + }, + { + "epoch": 0.3327653155292343, + "grad_norm": 2.293778896331787, + "learning_rate": 0.0008363121918422233, + "loss": 0.5822, + "step": 5956 + }, + { + "epoch": 0.33282118613291617, + "grad_norm": 0.6248571872711182, + "learning_rate": 0.0008362841774988794, + "loss": 0.5846, + "step": 5957 + }, + { + "epoch": 0.33287705673659806, + "grad_norm": 1.4036259651184082, + "learning_rate": 0.0008362561631555357, + "loss": 0.4824, + "step": 5958 + }, + { + "epoch": 0.3329329273402799, + "grad_norm": 2.1768393516540527, + "learning_rate": 0.0008362281488121919, + "loss": 0.5125, + "step": 5959 + }, + { + "epoch": 0.3329887979439618, + "grad_norm": 0.6846258044242859, + "learning_rate": 0.000836200134468848, + "loss": 0.4871, + "step": 5960 + }, + { + "epoch": 0.33304466854764364, + "grad_norm": 0.49456077814102173, + "learning_rate": 0.0008361721201255042, + "loss": 0.3644, + "step": 5961 + }, + { + "epoch": 0.33310053915132554, + "grad_norm": 0.6060474514961243, + "learning_rate": 0.0008361441057821605, + "loss": 0.5961, + "step": 5962 + }, + { + "epoch": 0.3331564097550074, + "grad_norm": 0.560760498046875, + "learning_rate": 0.0008361160914388167, + "loss": 0.4832, + "step": 5963 + }, + { + "epoch": 0.3332122803586893, + "grad_norm": 0.687765896320343, + "learning_rate": 0.0008360880770954728, + "loss": 0.4164, + "step": 5964 + }, + { + "epoch": 0.33326815096237117, + "grad_norm": 1.2702951431274414, + "learning_rate": 0.0008360600627521291, + "loss": 0.5365, + "step": 5965 + }, + { + "epoch": 0.333324021566053, + "grad_norm": 3.8869433403015137, + "learning_rate": 0.0008360320484087853, + "loss": 0.4314, + "step": 5966 + }, + { + "epoch": 0.3333798921697349, + "grad_norm": 0.6954916715621948, + "learning_rate": 0.0008360040340654415, + "loss": 0.4348, + "step": 5967 + }, + { + "epoch": 0.33343576277341674, + "grad_norm": 0.6135568022727966, + "learning_rate": 0.0008359760197220976, + "loss": 0.7068, + "step": 5968 + }, + { + "epoch": 0.33349163337709864, + "grad_norm": 0.621398389339447, + "learning_rate": 0.000835948005378754, + "loss": 0.5152, + "step": 5969 + }, + { + "epoch": 0.33354750398078054, + "grad_norm": 0.5300223231315613, + "learning_rate": 0.0008359199910354102, + "loss": 0.53, + "step": 5970 + }, + { + "epoch": 0.3336033745844624, + "grad_norm": 0.6034727692604065, + "learning_rate": 0.0008358919766920664, + "loss": 0.6015, + "step": 5971 + }, + { + "epoch": 0.33365924518814427, + "grad_norm": 0.7202126383781433, + "learning_rate": 0.0008358639623487226, + "loss": 0.4762, + "step": 5972 + }, + { + "epoch": 0.3337151157918261, + "grad_norm": 0.6417959928512573, + "learning_rate": 0.0008358359480053788, + "loss": 0.4387, + "step": 5973 + }, + { + "epoch": 0.333770986395508, + "grad_norm": 0.6173086762428284, + "learning_rate": 0.000835807933662035, + "loss": 0.4748, + "step": 5974 + }, + { + "epoch": 0.3338268569991899, + "grad_norm": 0.5175118446350098, + "learning_rate": 0.0008357799193186912, + "loss": 0.5345, + "step": 5975 + }, + { + "epoch": 0.33388272760287174, + "grad_norm": 0.5983191132545471, + "learning_rate": 0.0008357519049753474, + "loss": 0.4523, + "step": 5976 + }, + { + "epoch": 0.33393859820655364, + "grad_norm": 0.5592941045761108, + "learning_rate": 0.0008357238906320036, + "loss": 0.4754, + "step": 5977 + }, + { + "epoch": 0.3339944688102355, + "grad_norm": 0.6687396168708801, + "learning_rate": 0.0008356958762886598, + "loss": 0.4809, + "step": 5978 + }, + { + "epoch": 0.3340503394139174, + "grad_norm": 0.496591717004776, + "learning_rate": 0.000835667861945316, + "loss": 0.5429, + "step": 5979 + }, + { + "epoch": 0.3341062100175992, + "grad_norm": 0.6368250250816345, + "learning_rate": 0.0008356398476019722, + "loss": 0.396, + "step": 5980 + }, + { + "epoch": 0.3341620806212811, + "grad_norm": 0.987069845199585, + "learning_rate": 0.0008356118332586284, + "loss": 0.423, + "step": 5981 + }, + { + "epoch": 0.334217951224963, + "grad_norm": 0.47866007685661316, + "learning_rate": 0.0008355838189152846, + "loss": 0.3965, + "step": 5982 + }, + { + "epoch": 0.33427382182864485, + "grad_norm": 0.5743852853775024, + "learning_rate": 0.0008355558045719408, + "loss": 0.595, + "step": 5983 + }, + { + "epoch": 0.33432969243232674, + "grad_norm": 0.55213463306427, + "learning_rate": 0.000835527790228597, + "loss": 0.6482, + "step": 5984 + }, + { + "epoch": 0.3343855630360086, + "grad_norm": 0.364165335893631, + "learning_rate": 0.0008354997758852533, + "loss": 0.5129, + "step": 5985 + }, + { + "epoch": 0.3344414336396905, + "grad_norm": 0.47106194496154785, + "learning_rate": 0.0008354717615419096, + "loss": 0.4386, + "step": 5986 + }, + { + "epoch": 0.3344973042433724, + "grad_norm": 1.1401678323745728, + "learning_rate": 0.0008354437471985657, + "loss": 0.4711, + "step": 5987 + }, + { + "epoch": 0.3345531748470542, + "grad_norm": 0.5687565207481384, + "learning_rate": 0.0008354157328552219, + "loss": 0.4481, + "step": 5988 + }, + { + "epoch": 0.3346090454507361, + "grad_norm": 0.47635167837142944, + "learning_rate": 0.0008353877185118781, + "loss": 0.6819, + "step": 5989 + }, + { + "epoch": 0.33466491605441795, + "grad_norm": 0.7622010111808777, + "learning_rate": 0.0008353597041685344, + "loss": 0.5027, + "step": 5990 + }, + { + "epoch": 0.33472078665809984, + "grad_norm": 0.41560545563697815, + "learning_rate": 0.0008353316898251905, + "loss": 0.5055, + "step": 5991 + }, + { + "epoch": 0.33477665726178174, + "grad_norm": 0.4639936089515686, + "learning_rate": 0.0008353036754818467, + "loss": 0.4298, + "step": 5992 + }, + { + "epoch": 0.3348325278654636, + "grad_norm": 0.4849075973033905, + "learning_rate": 0.0008352756611385029, + "loss": 0.4098, + "step": 5993 + }, + { + "epoch": 0.3348883984691455, + "grad_norm": 0.5485953092575073, + "learning_rate": 0.0008352476467951592, + "loss": 0.4544, + "step": 5994 + }, + { + "epoch": 0.3349442690728273, + "grad_norm": 0.37615296244621277, + "learning_rate": 0.0008352196324518153, + "loss": 0.428, + "step": 5995 + }, + { + "epoch": 0.3350001396765092, + "grad_norm": 0.45558837056159973, + "learning_rate": 0.0008351916181084715, + "loss": 0.5944, + "step": 5996 + }, + { + "epoch": 0.33505601028019105, + "grad_norm": 1.5046547651290894, + "learning_rate": 0.0008351636037651278, + "loss": 0.4525, + "step": 5997 + }, + { + "epoch": 0.33511188088387295, + "grad_norm": 0.667465090751648, + "learning_rate": 0.000835135589421784, + "loss": 0.5645, + "step": 5998 + }, + { + "epoch": 0.33516775148755484, + "grad_norm": 0.5656726956367493, + "learning_rate": 0.0008351075750784401, + "loss": 0.4231, + "step": 5999 + }, + { + "epoch": 0.3352236220912367, + "grad_norm": 0.4004129469394684, + "learning_rate": 0.0008350795607350963, + "loss": 0.4349, + "step": 6000 + }, + { + "epoch": 0.3352236220912367, + "eval_cer": 0.0940176535410734, + "eval_loss": 0.3528870642185211, + "eval_runtime": 56.3869, + "eval_samples_per_second": 80.48, + "eval_steps_per_second": 5.037, + "eval_wer": 0.3686489204212173, + "step": 6000 + }, + { + "epoch": 0.3352794926949186, + "grad_norm": 0.9836883544921875, + "learning_rate": 0.0008350515463917527, + "loss": 0.5687, + "step": 6001 + }, + { + "epoch": 0.3353353632986004, + "grad_norm": 0.6683579683303833, + "learning_rate": 0.0008350235320484088, + "loss": 0.4399, + "step": 6002 + }, + { + "epoch": 0.3353912339022823, + "grad_norm": 0.5578369498252869, + "learning_rate": 0.000834995517705065, + "loss": 0.3147, + "step": 6003 + }, + { + "epoch": 0.3354471045059642, + "grad_norm": 0.7072634100914001, + "learning_rate": 0.0008349675033617213, + "loss": 0.4751, + "step": 6004 + }, + { + "epoch": 0.33550297510964605, + "grad_norm": 0.7269778251647949, + "learning_rate": 0.0008349394890183775, + "loss": 0.6567, + "step": 6005 + }, + { + "epoch": 0.33555884571332795, + "grad_norm": 0.5661282539367676, + "learning_rate": 0.0008349114746750336, + "loss": 0.582, + "step": 6006 + }, + { + "epoch": 0.3356147163170098, + "grad_norm": 0.4674564599990845, + "learning_rate": 0.0008348834603316898, + "loss": 0.4051, + "step": 6007 + }, + { + "epoch": 0.3356705869206917, + "grad_norm": 0.8183049559593201, + "learning_rate": 0.0008348554459883461, + "loss": 0.6191, + "step": 6008 + }, + { + "epoch": 0.3357264575243736, + "grad_norm": 0.6088268160820007, + "learning_rate": 0.0008348274316450023, + "loss": 0.4152, + "step": 6009 + }, + { + "epoch": 0.3357823281280554, + "grad_norm": 0.6021192669868469, + "learning_rate": 0.0008347994173016584, + "loss": 0.4058, + "step": 6010 + }, + { + "epoch": 0.3358381987317373, + "grad_norm": 0.5310773253440857, + "learning_rate": 0.0008347714029583147, + "loss": 0.526, + "step": 6011 + }, + { + "epoch": 0.33589406933541915, + "grad_norm": 0.5602021813392639, + "learning_rate": 0.0008347433886149709, + "loss": 0.5913, + "step": 6012 + }, + { + "epoch": 0.33594993993910105, + "grad_norm": 18.694793701171875, + "learning_rate": 0.0008347153742716271, + "loss": 0.6024, + "step": 6013 + }, + { + "epoch": 0.3360058105427829, + "grad_norm": 0.5870672464370728, + "learning_rate": 0.0008346873599282832, + "loss": 0.4978, + "step": 6014 + }, + { + "epoch": 0.3360616811464648, + "grad_norm": 0.9685060977935791, + "learning_rate": 0.0008346593455849395, + "loss": 0.5438, + "step": 6015 + }, + { + "epoch": 0.3361175517501467, + "grad_norm": 1.3502471446990967, + "learning_rate": 0.0008346313312415957, + "loss": 0.4473, + "step": 6016 + }, + { + "epoch": 0.3361734223538285, + "grad_norm": 1.0061477422714233, + "learning_rate": 0.000834603316898252, + "loss": 0.6137, + "step": 6017 + }, + { + "epoch": 0.3362292929575104, + "grad_norm": 0.5825634598731995, + "learning_rate": 0.0008345753025549082, + "loss": 0.4896, + "step": 6018 + }, + { + "epoch": 0.33628516356119226, + "grad_norm": 0.506363570690155, + "learning_rate": 0.0008345472882115644, + "loss": 0.3933, + "step": 6019 + }, + { + "epoch": 0.33634103416487415, + "grad_norm": 0.6312246918678284, + "learning_rate": 0.0008345192738682206, + "loss": 0.4248, + "step": 6020 + }, + { + "epoch": 0.33639690476855605, + "grad_norm": 0.9314213991165161, + "learning_rate": 0.0008344912595248767, + "loss": 0.3815, + "step": 6021 + }, + { + "epoch": 0.3364527753722379, + "grad_norm": 0.5302849411964417, + "learning_rate": 0.000834463245181533, + "loss": 0.4724, + "step": 6022 + }, + { + "epoch": 0.3365086459759198, + "grad_norm": 0.8375819325447083, + "learning_rate": 0.0008344352308381892, + "loss": 0.6362, + "step": 6023 + }, + { + "epoch": 0.3365645165796016, + "grad_norm": 0.629793643951416, + "learning_rate": 0.0008344072164948454, + "loss": 0.5431, + "step": 6024 + }, + { + "epoch": 0.3366203871832835, + "grad_norm": 0.45484989881515503, + "learning_rate": 0.0008343792021515015, + "loss": 0.4104, + "step": 6025 + }, + { + "epoch": 0.33667625778696536, + "grad_norm": 0.47180238366127014, + "learning_rate": 0.0008343511878081578, + "loss": 0.5963, + "step": 6026 + }, + { + "epoch": 0.33673212839064726, + "grad_norm": 0.4123101830482483, + "learning_rate": 0.000834323173464814, + "loss": 0.394, + "step": 6027 + }, + { + "epoch": 0.33678799899432915, + "grad_norm": 0.8123453855514526, + "learning_rate": 0.0008342951591214702, + "loss": 0.5174, + "step": 6028 + }, + { + "epoch": 0.336843869598011, + "grad_norm": 0.7732170820236206, + "learning_rate": 0.0008342671447781264, + "loss": 0.3939, + "step": 6029 + }, + { + "epoch": 0.3368997402016929, + "grad_norm": 2.8999900817871094, + "learning_rate": 0.0008342391304347826, + "loss": 0.5531, + "step": 6030 + }, + { + "epoch": 0.33695561080537473, + "grad_norm": 0.6057357788085938, + "learning_rate": 0.0008342111160914388, + "loss": 0.5019, + "step": 6031 + }, + { + "epoch": 0.3370114814090566, + "grad_norm": 2.0265963077545166, + "learning_rate": 0.000834183101748095, + "loss": 0.3871, + "step": 6032 + }, + { + "epoch": 0.3370673520127385, + "grad_norm": 1.0017009973526, + "learning_rate": 0.0008341550874047512, + "loss": 0.4269, + "step": 6033 + }, + { + "epoch": 0.33712322261642036, + "grad_norm": 0.47352704405784607, + "learning_rate": 0.0008341270730614074, + "loss": 0.453, + "step": 6034 + }, + { + "epoch": 0.33717909322010226, + "grad_norm": 0.4794802665710449, + "learning_rate": 0.0008340990587180637, + "loss": 0.5604, + "step": 6035 + }, + { + "epoch": 0.3372349638237841, + "grad_norm": 0.5524861216545105, + "learning_rate": 0.00083407104437472, + "loss": 0.5335, + "step": 6036 + }, + { + "epoch": 0.337290834427466, + "grad_norm": 0.4704053997993469, + "learning_rate": 0.0008340430300313761, + "loss": 0.5479, + "step": 6037 + }, + { + "epoch": 0.3373467050311479, + "grad_norm": 0.7685915231704712, + "learning_rate": 0.0008340150156880323, + "loss": 0.4541, + "step": 6038 + }, + { + "epoch": 0.3374025756348297, + "grad_norm": 0.5565403699874878, + "learning_rate": 0.0008339870013446885, + "loss": 0.4984, + "step": 6039 + }, + { + "epoch": 0.3374584462385116, + "grad_norm": 0.46013322472572327, + "learning_rate": 0.0008339589870013448, + "loss": 0.3971, + "step": 6040 + }, + { + "epoch": 0.33751431684219346, + "grad_norm": 0.8790013194084167, + "learning_rate": 0.0008339309726580009, + "loss": 0.528, + "step": 6041 + }, + { + "epoch": 0.33757018744587536, + "grad_norm": 1.8310908079147339, + "learning_rate": 0.0008339029583146571, + "loss": 0.4089, + "step": 6042 + }, + { + "epoch": 0.3376260580495572, + "grad_norm": 0.518828272819519, + "learning_rate": 0.0008338749439713134, + "loss": 0.4942, + "step": 6043 + }, + { + "epoch": 0.3376819286532391, + "grad_norm": 0.6181161403656006, + "learning_rate": 0.0008338469296279695, + "loss": 0.3828, + "step": 6044 + }, + { + "epoch": 0.337737799256921, + "grad_norm": 1.1456477642059326, + "learning_rate": 0.0008338189152846257, + "loss": 0.4935, + "step": 6045 + }, + { + "epoch": 0.33779366986060283, + "grad_norm": 0.4142249822616577, + "learning_rate": 0.0008337909009412819, + "loss": 0.3417, + "step": 6046 + }, + { + "epoch": 0.3378495404642847, + "grad_norm": 0.49381735920906067, + "learning_rate": 0.0008337628865979382, + "loss": 0.4115, + "step": 6047 + }, + { + "epoch": 0.33790541106796657, + "grad_norm": 0.46165505051612854, + "learning_rate": 0.0008337348722545943, + "loss": 0.4701, + "step": 6048 + }, + { + "epoch": 0.33796128167164846, + "grad_norm": 0.8138267397880554, + "learning_rate": 0.0008337068579112505, + "loss": 0.4982, + "step": 6049 + }, + { + "epoch": 0.33801715227533036, + "grad_norm": 0.6115504503250122, + "learning_rate": 0.0008336788435679068, + "loss": 0.4649, + "step": 6050 + }, + { + "epoch": 0.3380730228790122, + "grad_norm": 0.5007449388504028, + "learning_rate": 0.0008336508292245631, + "loss": 0.4682, + "step": 6051 + }, + { + "epoch": 0.3381288934826941, + "grad_norm": 0.5062131285667419, + "learning_rate": 0.0008336228148812192, + "loss": 0.5207, + "step": 6052 + }, + { + "epoch": 0.33818476408637593, + "grad_norm": 0.40274789929389954, + "learning_rate": 0.0008335948005378754, + "loss": 0.3811, + "step": 6053 + }, + { + "epoch": 0.33824063469005783, + "grad_norm": 1.2009791135787964, + "learning_rate": 0.0008335667861945317, + "loss": 0.488, + "step": 6054 + }, + { + "epoch": 0.3382965052937397, + "grad_norm": 2.8915350437164307, + "learning_rate": 0.0008335387718511879, + "loss": 0.5103, + "step": 6055 + }, + { + "epoch": 0.33835237589742156, + "grad_norm": 0.6894481778144836, + "learning_rate": 0.000833510757507844, + "loss": 0.5168, + "step": 6056 + }, + { + "epoch": 0.33840824650110346, + "grad_norm": 0.616943895816803, + "learning_rate": 0.0008334827431645002, + "loss": 0.4926, + "step": 6057 + }, + { + "epoch": 0.3384641171047853, + "grad_norm": 0.5179904699325562, + "learning_rate": 0.0008334547288211565, + "loss": 0.4843, + "step": 6058 + }, + { + "epoch": 0.3385199877084672, + "grad_norm": 0.6686420440673828, + "learning_rate": 0.0008334267144778127, + "loss": 0.4632, + "step": 6059 + }, + { + "epoch": 0.33857585831214904, + "grad_norm": 0.5201950669288635, + "learning_rate": 0.0008333987001344688, + "loss": 0.5113, + "step": 6060 + }, + { + "epoch": 0.33863172891583093, + "grad_norm": 0.9428126215934753, + "learning_rate": 0.0008333706857911251, + "loss": 0.421, + "step": 6061 + }, + { + "epoch": 0.3386875995195128, + "grad_norm": 0.6356209516525269, + "learning_rate": 0.0008333426714477813, + "loss": 0.5263, + "step": 6062 + }, + { + "epoch": 0.33874347012319467, + "grad_norm": 0.6516725420951843, + "learning_rate": 0.0008333146571044374, + "loss": 0.5166, + "step": 6063 + }, + { + "epoch": 0.33879934072687656, + "grad_norm": 0.8693323731422424, + "learning_rate": 0.0008332866427610936, + "loss": 0.4514, + "step": 6064 + }, + { + "epoch": 0.3388552113305584, + "grad_norm": 0.5338800549507141, + "learning_rate": 0.0008332586284177499, + "loss": 0.4877, + "step": 6065 + }, + { + "epoch": 0.3389110819342403, + "grad_norm": 0.4513671398162842, + "learning_rate": 0.0008332306140744061, + "loss": 0.4669, + "step": 6066 + }, + { + "epoch": 0.3389669525379222, + "grad_norm": 0.4930427074432373, + "learning_rate": 0.0008332025997310622, + "loss": 0.502, + "step": 6067 + }, + { + "epoch": 0.33902282314160403, + "grad_norm": 0.6420872807502747, + "learning_rate": 0.0008331745853877186, + "loss": 0.6655, + "step": 6068 + }, + { + "epoch": 0.33907869374528593, + "grad_norm": 0.7885256409645081, + "learning_rate": 0.0008331465710443748, + "loss": 0.3776, + "step": 6069 + }, + { + "epoch": 0.33913456434896777, + "grad_norm": 0.8877418041229248, + "learning_rate": 0.000833118556701031, + "loss": 0.6483, + "step": 6070 + }, + { + "epoch": 0.33919043495264967, + "grad_norm": 0.4932022988796234, + "learning_rate": 0.0008330905423576871, + "loss": 0.4838, + "step": 6071 + }, + { + "epoch": 0.33924630555633156, + "grad_norm": 0.47334370017051697, + "learning_rate": 0.0008330625280143434, + "loss": 0.4352, + "step": 6072 + }, + { + "epoch": 0.3393021761600134, + "grad_norm": 0.6141536831855774, + "learning_rate": 0.0008330345136709996, + "loss": 0.552, + "step": 6073 + }, + { + "epoch": 0.3393580467636953, + "grad_norm": 0.592214822769165, + "learning_rate": 0.0008330064993276558, + "loss": 0.58, + "step": 6074 + }, + { + "epoch": 0.33941391736737714, + "grad_norm": 0.47659116983413696, + "learning_rate": 0.000832978484984312, + "loss": 0.4906, + "step": 6075 + }, + { + "epoch": 0.33946978797105903, + "grad_norm": 0.49063989520072937, + "learning_rate": 0.0008329504706409682, + "loss": 0.4602, + "step": 6076 + }, + { + "epoch": 0.3395256585747409, + "grad_norm": 1.2895300388336182, + "learning_rate": 0.0008329224562976244, + "loss": 0.563, + "step": 6077 + }, + { + "epoch": 0.33958152917842277, + "grad_norm": 4.8244524002075195, + "learning_rate": 0.0008328944419542806, + "loss": 0.4214, + "step": 6078 + }, + { + "epoch": 0.33963739978210467, + "grad_norm": 0.6701483726501465, + "learning_rate": 0.0008328664276109368, + "loss": 0.3978, + "step": 6079 + }, + { + "epoch": 0.3396932703857865, + "grad_norm": 0.37649405002593994, + "learning_rate": 0.000832838413267593, + "loss": 0.2942, + "step": 6080 + }, + { + "epoch": 0.3397491409894684, + "grad_norm": 0.580350399017334, + "learning_rate": 0.0008328103989242492, + "loss": 0.3911, + "step": 6081 + }, + { + "epoch": 0.33980501159315024, + "grad_norm": 0.46767187118530273, + "learning_rate": 0.0008327823845809055, + "loss": 0.4521, + "step": 6082 + }, + { + "epoch": 0.33986088219683214, + "grad_norm": 0.6988583207130432, + "learning_rate": 0.0008327543702375616, + "loss": 0.5902, + "step": 6083 + }, + { + "epoch": 0.33991675280051403, + "grad_norm": 0.6017526984214783, + "learning_rate": 0.0008327263558942178, + "loss": 0.465, + "step": 6084 + }, + { + "epoch": 0.3399726234041959, + "grad_norm": 0.9230008721351624, + "learning_rate": 0.000832698341550874, + "loss": 0.4367, + "step": 6085 + }, + { + "epoch": 0.34002849400787777, + "grad_norm": 10.056321144104004, + "learning_rate": 0.0008326703272075303, + "loss": 0.5203, + "step": 6086 + }, + { + "epoch": 0.3400843646115596, + "grad_norm": 0.5818965435028076, + "learning_rate": 0.0008326423128641865, + "loss": 0.6886, + "step": 6087 + }, + { + "epoch": 0.3401402352152415, + "grad_norm": 0.45233145356178284, + "learning_rate": 0.0008326142985208427, + "loss": 0.4705, + "step": 6088 + }, + { + "epoch": 0.3401961058189234, + "grad_norm": 0.379586786031723, + "learning_rate": 0.000832586284177499, + "loss": 0.3646, + "step": 6089 + }, + { + "epoch": 0.34025197642260524, + "grad_norm": 0.8485622406005859, + "learning_rate": 0.0008325582698341551, + "loss": 0.4759, + "step": 6090 + }, + { + "epoch": 0.34030784702628714, + "grad_norm": 0.5860520005226135, + "learning_rate": 0.0008325302554908113, + "loss": 0.6618, + "step": 6091 + }, + { + "epoch": 0.340363717629969, + "grad_norm": 0.44540074467658997, + "learning_rate": 0.0008325022411474675, + "loss": 0.3805, + "step": 6092 + }, + { + "epoch": 0.34041958823365087, + "grad_norm": 0.61279296875, + "learning_rate": 0.0008324742268041238, + "loss": 0.5929, + "step": 6093 + }, + { + "epoch": 0.3404754588373327, + "grad_norm": 0.630282998085022, + "learning_rate": 0.0008324462124607799, + "loss": 0.7484, + "step": 6094 + }, + { + "epoch": 0.3405313294410146, + "grad_norm": 0.4983368515968323, + "learning_rate": 0.0008324181981174361, + "loss": 0.5092, + "step": 6095 + }, + { + "epoch": 0.3405872000446965, + "grad_norm": 0.5497864484786987, + "learning_rate": 0.0008323901837740923, + "loss": 0.5789, + "step": 6096 + }, + { + "epoch": 0.34064307064837834, + "grad_norm": 0.485186368227005, + "learning_rate": 0.0008323621694307486, + "loss": 0.3493, + "step": 6097 + }, + { + "epoch": 0.34069894125206024, + "grad_norm": 0.6519353985786438, + "learning_rate": 0.0008323341550874047, + "loss": 0.5303, + "step": 6098 + }, + { + "epoch": 0.3407548118557421, + "grad_norm": 0.4950815737247467, + "learning_rate": 0.0008323061407440609, + "loss": 0.4482, + "step": 6099 + }, + { + "epoch": 0.340810682459424, + "grad_norm": 0.4178243577480316, + "learning_rate": 0.0008322781264007172, + "loss": 0.473, + "step": 6100 + }, + { + "epoch": 0.34086655306310587, + "grad_norm": 0.6563735604286194, + "learning_rate": 0.0008322501120573735, + "loss": 0.602, + "step": 6101 + }, + { + "epoch": 0.3409224236667877, + "grad_norm": 0.6491042375564575, + "learning_rate": 0.0008322220977140296, + "loss": 0.4598, + "step": 6102 + }, + { + "epoch": 0.3409782942704696, + "grad_norm": 0.6762677431106567, + "learning_rate": 0.0008321940833706858, + "loss": 0.3906, + "step": 6103 + }, + { + "epoch": 0.34103416487415145, + "grad_norm": 0.5846664905548096, + "learning_rate": 0.0008321660690273421, + "loss": 0.447, + "step": 6104 + }, + { + "epoch": 0.34109003547783334, + "grad_norm": 0.8738669157028198, + "learning_rate": 0.0008321380546839982, + "loss": 0.4686, + "step": 6105 + }, + { + "epoch": 0.34114590608151524, + "grad_norm": 0.7721189260482788, + "learning_rate": 0.0008321100403406544, + "loss": 0.4267, + "step": 6106 + }, + { + "epoch": 0.3412017766851971, + "grad_norm": 6.3531060218811035, + "learning_rate": 0.0008320820259973107, + "loss": 0.5145, + "step": 6107 + }, + { + "epoch": 0.341257647288879, + "grad_norm": 0.49005770683288574, + "learning_rate": 0.0008320540116539669, + "loss": 0.5152, + "step": 6108 + }, + { + "epoch": 0.3413135178925608, + "grad_norm": 0.502199113368988, + "learning_rate": 0.000832025997310623, + "loss": 0.6121, + "step": 6109 + }, + { + "epoch": 0.3413693884962427, + "grad_norm": 1.0549256801605225, + "learning_rate": 0.0008319979829672792, + "loss": 0.6946, + "step": 6110 + }, + { + "epoch": 0.34142525909992455, + "grad_norm": 0.5613778233528137, + "learning_rate": 0.0008319699686239355, + "loss": 0.4798, + "step": 6111 + }, + { + "epoch": 0.34148112970360645, + "grad_norm": 0.4630190134048462, + "learning_rate": 0.0008319419542805917, + "loss": 0.5107, + "step": 6112 + }, + { + "epoch": 0.34153700030728834, + "grad_norm": 0.46027010679244995, + "learning_rate": 0.0008319139399372478, + "loss": 0.4253, + "step": 6113 + }, + { + "epoch": 0.3415928709109702, + "grad_norm": 0.7958740592002869, + "learning_rate": 0.0008318859255939041, + "loss": 0.7857, + "step": 6114 + }, + { + "epoch": 0.3416487415146521, + "grad_norm": 0.5841127038002014, + "learning_rate": 0.0008318579112505603, + "loss": 0.5578, + "step": 6115 + }, + { + "epoch": 0.3417046121183339, + "grad_norm": 0.43776145577430725, + "learning_rate": 0.0008318298969072165, + "loss": 0.4727, + "step": 6116 + }, + { + "epoch": 0.3417604827220158, + "grad_norm": 0.5868346691131592, + "learning_rate": 0.0008318018825638726, + "loss": 0.3955, + "step": 6117 + }, + { + "epoch": 0.3418163533256977, + "grad_norm": 0.5411125421524048, + "learning_rate": 0.000831773868220529, + "loss": 0.377, + "step": 6118 + }, + { + "epoch": 0.34187222392937955, + "grad_norm": 2.756865978240967, + "learning_rate": 0.0008317458538771852, + "loss": 0.477, + "step": 6119 + }, + { + "epoch": 0.34192809453306144, + "grad_norm": 0.5920282006263733, + "learning_rate": 0.0008317178395338414, + "loss": 0.4533, + "step": 6120 + }, + { + "epoch": 0.3419839651367433, + "grad_norm": 0.6929852366447449, + "learning_rate": 0.0008316898251904976, + "loss": 0.4887, + "step": 6121 + }, + { + "epoch": 0.3420398357404252, + "grad_norm": 0.5776175856590271, + "learning_rate": 0.0008316618108471538, + "loss": 0.4496, + "step": 6122 + }, + { + "epoch": 0.3420957063441071, + "grad_norm": 0.4436497390270233, + "learning_rate": 0.00083163379650381, + "loss": 0.4177, + "step": 6123 + }, + { + "epoch": 0.3421515769477889, + "grad_norm": 1.0830237865447998, + "learning_rate": 0.0008316057821604662, + "loss": 0.5745, + "step": 6124 + }, + { + "epoch": 0.3422074475514708, + "grad_norm": 1.7280224561691284, + "learning_rate": 0.0008315777678171224, + "loss": 0.4234, + "step": 6125 + }, + { + "epoch": 0.34226331815515265, + "grad_norm": 0.508019208908081, + "learning_rate": 0.0008315497534737786, + "loss": 0.4329, + "step": 6126 + }, + { + "epoch": 0.34231918875883455, + "grad_norm": 0.4005994200706482, + "learning_rate": 0.0008315217391304348, + "loss": 0.4788, + "step": 6127 + }, + { + "epoch": 0.3423750593625164, + "grad_norm": 0.5472623109817505, + "learning_rate": 0.0008314937247870909, + "loss": 0.4685, + "step": 6128 + }, + { + "epoch": 0.3424309299661983, + "grad_norm": 0.6321690678596497, + "learning_rate": 0.0008314657104437472, + "loss": 0.4572, + "step": 6129 + }, + { + "epoch": 0.3424868005698802, + "grad_norm": 0.6403376460075378, + "learning_rate": 0.0008314376961004034, + "loss": 0.6541, + "step": 6130 + }, + { + "epoch": 0.342542671173562, + "grad_norm": 0.5664817690849304, + "learning_rate": 0.0008314096817570596, + "loss": 0.4018, + "step": 6131 + }, + { + "epoch": 0.3425985417772439, + "grad_norm": 0.48583513498306274, + "learning_rate": 0.0008313816674137158, + "loss": 0.4116, + "step": 6132 + }, + { + "epoch": 0.34265441238092575, + "grad_norm": 0.5045223236083984, + "learning_rate": 0.000831353653070372, + "loss": 0.4563, + "step": 6133 + }, + { + "epoch": 0.34271028298460765, + "grad_norm": 0.5106861591339111, + "learning_rate": 0.0008313256387270282, + "loss": 0.4427, + "step": 6134 + }, + { + "epoch": 0.34276615358828955, + "grad_norm": 0.6021305918693542, + "learning_rate": 0.0008312976243836845, + "loss": 0.5114, + "step": 6135 + }, + { + "epoch": 0.3428220241919714, + "grad_norm": 0.48696351051330566, + "learning_rate": 0.0008312696100403407, + "loss": 0.4239, + "step": 6136 + }, + { + "epoch": 0.3428778947956533, + "grad_norm": 0.4993094503879547, + "learning_rate": 0.0008312415956969969, + "loss": 0.5348, + "step": 6137 + }, + { + "epoch": 0.3429337653993351, + "grad_norm": 0.5975547432899475, + "learning_rate": 0.0008312135813536531, + "loss": 0.4963, + "step": 6138 + }, + { + "epoch": 0.342989636003017, + "grad_norm": 1.2739489078521729, + "learning_rate": 0.0008311855670103094, + "loss": 0.6757, + "step": 6139 + }, + { + "epoch": 0.34304550660669886, + "grad_norm": 0.6974195837974548, + "learning_rate": 0.0008311575526669655, + "loss": 0.4993, + "step": 6140 + }, + { + "epoch": 0.34310137721038075, + "grad_norm": 0.8443767428398132, + "learning_rate": 0.0008311295383236217, + "loss": 0.4549, + "step": 6141 + }, + { + "epoch": 0.34315724781406265, + "grad_norm": 1.4535877704620361, + "learning_rate": 0.0008311015239802779, + "loss": 0.4911, + "step": 6142 + }, + { + "epoch": 0.3432131184177445, + "grad_norm": 0.49021628499031067, + "learning_rate": 0.0008310735096369342, + "loss": 0.4786, + "step": 6143 + }, + { + "epoch": 0.3432689890214264, + "grad_norm": 0.5146598815917969, + "learning_rate": 0.0008310454952935903, + "loss": 0.4457, + "step": 6144 + }, + { + "epoch": 0.3433248596251082, + "grad_norm": 0.7747002243995667, + "learning_rate": 0.0008310174809502465, + "loss": 0.568, + "step": 6145 + }, + { + "epoch": 0.3433807302287901, + "grad_norm": 0.6313403844833374, + "learning_rate": 0.0008309894666069028, + "loss": 0.4249, + "step": 6146 + }, + { + "epoch": 0.343436600832472, + "grad_norm": 4.2319159507751465, + "learning_rate": 0.000830961452263559, + "loss": 0.4964, + "step": 6147 + }, + { + "epoch": 0.34349247143615386, + "grad_norm": 0.39181947708129883, + "learning_rate": 0.0008309334379202151, + "loss": 0.4253, + "step": 6148 + }, + { + "epoch": 0.34354834203983575, + "grad_norm": 3.83128023147583, + "learning_rate": 0.0008309054235768713, + "loss": 0.3991, + "step": 6149 + }, + { + "epoch": 0.3436042126435176, + "grad_norm": 0.41732731461524963, + "learning_rate": 0.0008308774092335276, + "loss": 0.4743, + "step": 6150 + }, + { + "epoch": 0.3436600832471995, + "grad_norm": 0.6124885678291321, + "learning_rate": 0.0008308493948901837, + "loss": 0.5614, + "step": 6151 + }, + { + "epoch": 0.3437159538508814, + "grad_norm": 0.4508824646472931, + "learning_rate": 0.00083082138054684, + "loss": 0.3651, + "step": 6152 + }, + { + "epoch": 0.3437718244545632, + "grad_norm": 0.5231598615646362, + "learning_rate": 0.0008307933662034963, + "loss": 0.4571, + "step": 6153 + }, + { + "epoch": 0.3438276950582451, + "grad_norm": 0.6544235348701477, + "learning_rate": 0.0008307653518601525, + "loss": 0.5686, + "step": 6154 + }, + { + "epoch": 0.34388356566192696, + "grad_norm": 0.604475736618042, + "learning_rate": 0.0008307373375168086, + "loss": 0.4513, + "step": 6155 + }, + { + "epoch": 0.34393943626560886, + "grad_norm": 0.5342380404472351, + "learning_rate": 0.0008307093231734648, + "loss": 0.4157, + "step": 6156 + }, + { + "epoch": 0.3439953068692907, + "grad_norm": 0.5321977734565735, + "learning_rate": 0.0008306813088301211, + "loss": 0.4511, + "step": 6157 + }, + { + "epoch": 0.3440511774729726, + "grad_norm": 0.8506699800491333, + "learning_rate": 0.0008306532944867773, + "loss": 0.7504, + "step": 6158 + }, + { + "epoch": 0.3441070480766545, + "grad_norm": 0.43939918279647827, + "learning_rate": 0.0008306252801434334, + "loss": 0.4345, + "step": 6159 + }, + { + "epoch": 0.3441629186803363, + "grad_norm": 0.5629659295082092, + "learning_rate": 0.0008305972658000897, + "loss": 0.4765, + "step": 6160 + }, + { + "epoch": 0.3442187892840182, + "grad_norm": 0.5209752321243286, + "learning_rate": 0.0008305692514567459, + "loss": 0.4618, + "step": 6161 + }, + { + "epoch": 0.34427465988770006, + "grad_norm": 0.5833732485771179, + "learning_rate": 0.0008305412371134021, + "loss": 0.4452, + "step": 6162 + }, + { + "epoch": 0.34433053049138196, + "grad_norm": 0.5703694820404053, + "learning_rate": 0.0008305132227700582, + "loss": 0.5085, + "step": 6163 + }, + { + "epoch": 0.34438640109506385, + "grad_norm": 0.53708815574646, + "learning_rate": 0.0008304852084267145, + "loss": 0.4311, + "step": 6164 + }, + { + "epoch": 0.3444422716987457, + "grad_norm": 0.4205625653266907, + "learning_rate": 0.0008304571940833707, + "loss": 0.4375, + "step": 6165 + }, + { + "epoch": 0.3444981423024276, + "grad_norm": 0.5607991814613342, + "learning_rate": 0.0008304291797400269, + "loss": 0.5895, + "step": 6166 + }, + { + "epoch": 0.34455401290610943, + "grad_norm": 0.4679194390773773, + "learning_rate": 0.000830401165396683, + "loss": 0.416, + "step": 6167 + }, + { + "epoch": 0.3446098835097913, + "grad_norm": 0.6496146321296692, + "learning_rate": 0.0008303731510533394, + "loss": 0.5193, + "step": 6168 + }, + { + "epoch": 0.3446657541134732, + "grad_norm": 0.6962479948997498, + "learning_rate": 0.0008303451367099956, + "loss": 0.5596, + "step": 6169 + }, + { + "epoch": 0.34472162471715506, + "grad_norm": 0.5263936519622803, + "learning_rate": 0.0008303171223666517, + "loss": 0.4642, + "step": 6170 + }, + { + "epoch": 0.34477749532083696, + "grad_norm": 0.5194091796875, + "learning_rate": 0.000830289108023308, + "loss": 0.4805, + "step": 6171 + }, + { + "epoch": 0.3448333659245188, + "grad_norm": 0.6688709855079651, + "learning_rate": 0.0008302610936799642, + "loss": 0.5446, + "step": 6172 + }, + { + "epoch": 0.3448892365282007, + "grad_norm": 0.5382821559906006, + "learning_rate": 0.0008302330793366204, + "loss": 0.5285, + "step": 6173 + }, + { + "epoch": 0.34494510713188253, + "grad_norm": 0.5943978428840637, + "learning_rate": 0.0008302050649932765, + "loss": 0.4775, + "step": 6174 + }, + { + "epoch": 0.34500097773556443, + "grad_norm": 0.6922537088394165, + "learning_rate": 0.0008301770506499328, + "loss": 0.5487, + "step": 6175 + }, + { + "epoch": 0.3450568483392463, + "grad_norm": 0.625964879989624, + "learning_rate": 0.000830149036306589, + "loss": 0.398, + "step": 6176 + }, + { + "epoch": 0.34511271894292816, + "grad_norm": 0.4470444917678833, + "learning_rate": 0.0008301210219632452, + "loss": 0.4513, + "step": 6177 + }, + { + "epoch": 0.34516858954661006, + "grad_norm": 0.8022071123123169, + "learning_rate": 0.0008300930076199014, + "loss": 0.4529, + "step": 6178 + }, + { + "epoch": 0.3452244601502919, + "grad_norm": 0.6939103007316589, + "learning_rate": 0.0008300649932765576, + "loss": 0.5768, + "step": 6179 + }, + { + "epoch": 0.3452803307539738, + "grad_norm": 0.6340508460998535, + "learning_rate": 0.0008300369789332138, + "loss": 0.4015, + "step": 6180 + }, + { + "epoch": 0.3453362013576557, + "grad_norm": 0.5141483545303345, + "learning_rate": 0.00083000896458987, + "loss": 0.4412, + "step": 6181 + }, + { + "epoch": 0.34539207196133753, + "grad_norm": 0.9686450362205505, + "learning_rate": 0.0008299809502465262, + "loss": 0.5748, + "step": 6182 + }, + { + "epoch": 0.34544794256501943, + "grad_norm": 0.5904701352119446, + "learning_rate": 0.0008299529359031824, + "loss": 0.5438, + "step": 6183 + }, + { + "epoch": 0.34550381316870127, + "grad_norm": 0.39603784680366516, + "learning_rate": 0.0008299249215598386, + "loss": 0.3262, + "step": 6184 + }, + { + "epoch": 0.34555968377238316, + "grad_norm": 0.5666400790214539, + "learning_rate": 0.000829896907216495, + "loss": 0.4216, + "step": 6185 + }, + { + "epoch": 0.34561555437606506, + "grad_norm": 0.4973011016845703, + "learning_rate": 0.0008298688928731511, + "loss": 0.4473, + "step": 6186 + }, + { + "epoch": 0.3456714249797469, + "grad_norm": 0.4737824499607086, + "learning_rate": 0.0008298408785298073, + "loss": 0.4738, + "step": 6187 + }, + { + "epoch": 0.3457272955834288, + "grad_norm": 0.5035756230354309, + "learning_rate": 0.0008298128641864635, + "loss": 0.4727, + "step": 6188 + }, + { + "epoch": 0.34578316618711064, + "grad_norm": 0.5603522658348083, + "learning_rate": 0.0008297848498431198, + "loss": 0.3746, + "step": 6189 + }, + { + "epoch": 0.34583903679079253, + "grad_norm": 0.5274117588996887, + "learning_rate": 0.0008297568354997759, + "loss": 0.536, + "step": 6190 + }, + { + "epoch": 0.34589490739447437, + "grad_norm": 0.478831022977829, + "learning_rate": 0.0008297288211564321, + "loss": 0.4823, + "step": 6191 + }, + { + "epoch": 0.34595077799815627, + "grad_norm": 0.666968822479248, + "learning_rate": 0.0008297008068130884, + "loss": 0.4817, + "step": 6192 + }, + { + "epoch": 0.34600664860183816, + "grad_norm": 0.5971948504447937, + "learning_rate": 0.0008296727924697445, + "loss": 0.5223, + "step": 6193 + }, + { + "epoch": 0.34606251920552, + "grad_norm": 0.4955073297023773, + "learning_rate": 0.0008296447781264007, + "loss": 0.5017, + "step": 6194 + }, + { + "epoch": 0.3461183898092019, + "grad_norm": 0.38683512806892395, + "learning_rate": 0.0008296167637830569, + "loss": 0.3762, + "step": 6195 + }, + { + "epoch": 0.34617426041288374, + "grad_norm": 0.6253572106361389, + "learning_rate": 0.0008295887494397132, + "loss": 0.6018, + "step": 6196 + }, + { + "epoch": 0.34623013101656563, + "grad_norm": 0.6597263813018799, + "learning_rate": 0.0008295607350963693, + "loss": 0.5591, + "step": 6197 + }, + { + "epoch": 0.34628600162024753, + "grad_norm": 0.6589916944503784, + "learning_rate": 0.0008295327207530255, + "loss": 0.3913, + "step": 6198 + }, + { + "epoch": 0.34634187222392937, + "grad_norm": 0.36921948194503784, + "learning_rate": 0.0008295047064096818, + "loss": 0.4178, + "step": 6199 + }, + { + "epoch": 0.34639774282761127, + "grad_norm": 0.5219224691390991, + "learning_rate": 0.000829476692066338, + "loss": 0.4101, + "step": 6200 + }, + { + "epoch": 0.3464536134312931, + "grad_norm": 0.7392685413360596, + "learning_rate": 0.0008294486777229941, + "loss": 0.3876, + "step": 6201 + }, + { + "epoch": 0.346509484034975, + "grad_norm": 0.4934576153755188, + "learning_rate": 0.0008294206633796503, + "loss": 0.4538, + "step": 6202 + }, + { + "epoch": 0.3465653546386569, + "grad_norm": 0.40815356373786926, + "learning_rate": 0.0008293926490363067, + "loss": 0.4387, + "step": 6203 + }, + { + "epoch": 0.34662122524233874, + "grad_norm": 0.6324625611305237, + "learning_rate": 0.0008293646346929629, + "loss": 0.5473, + "step": 6204 + }, + { + "epoch": 0.34667709584602063, + "grad_norm": 4.384434223175049, + "learning_rate": 0.000829336620349619, + "loss": 0.4969, + "step": 6205 + }, + { + "epoch": 0.3467329664497025, + "grad_norm": 0.6843063235282898, + "learning_rate": 0.0008293086060062752, + "loss": 0.3354, + "step": 6206 + }, + { + "epoch": 0.34678883705338437, + "grad_norm": 0.5570499897003174, + "learning_rate": 0.0008292805916629315, + "loss": 0.3927, + "step": 6207 + }, + { + "epoch": 0.3468447076570662, + "grad_norm": 0.504330575466156, + "learning_rate": 0.0008292525773195877, + "loss": 0.4842, + "step": 6208 + }, + { + "epoch": 0.3469005782607481, + "grad_norm": 0.33471328020095825, + "learning_rate": 0.0008292245629762438, + "loss": 0.3581, + "step": 6209 + }, + { + "epoch": 0.34695644886443, + "grad_norm": 0.5047796964645386, + "learning_rate": 0.0008291965486329001, + "loss": 0.395, + "step": 6210 + }, + { + "epoch": 0.34701231946811184, + "grad_norm": 7.391432762145996, + "learning_rate": 0.0008291685342895563, + "loss": 0.4119, + "step": 6211 + }, + { + "epoch": 0.34706819007179374, + "grad_norm": 0.5307571291923523, + "learning_rate": 0.0008291405199462124, + "loss": 0.4399, + "step": 6212 + }, + { + "epoch": 0.3471240606754756, + "grad_norm": 0.5118415951728821, + "learning_rate": 0.0008291125056028686, + "loss": 0.4757, + "step": 6213 + }, + { + "epoch": 0.34717993127915747, + "grad_norm": 1.2128592729568481, + "learning_rate": 0.0008290844912595249, + "loss": 0.4736, + "step": 6214 + }, + { + "epoch": 0.34723580188283937, + "grad_norm": 0.4628819525241852, + "learning_rate": 0.0008290564769161811, + "loss": 0.3908, + "step": 6215 + }, + { + "epoch": 0.3472916724865212, + "grad_norm": 0.3868899941444397, + "learning_rate": 0.0008290284625728372, + "loss": 0.3718, + "step": 6216 + }, + { + "epoch": 0.3473475430902031, + "grad_norm": 0.6230060458183289, + "learning_rate": 0.0008290004482294935, + "loss": 0.4496, + "step": 6217 + }, + { + "epoch": 0.34740341369388494, + "grad_norm": 0.397953599691391, + "learning_rate": 0.0008289724338861498, + "loss": 0.3291, + "step": 6218 + }, + { + "epoch": 0.34745928429756684, + "grad_norm": 0.8943667411804199, + "learning_rate": 0.000828944419542806, + "loss": 0.4669, + "step": 6219 + }, + { + "epoch": 0.34751515490124874, + "grad_norm": 3.1154303550720215, + "learning_rate": 0.000828916405199462, + "loss": 0.405, + "step": 6220 + }, + { + "epoch": 0.3475710255049306, + "grad_norm": 0.7706595063209534, + "learning_rate": 0.0008288883908561184, + "loss": 0.3971, + "step": 6221 + }, + { + "epoch": 0.34762689610861247, + "grad_norm": 0.626971960067749, + "learning_rate": 0.0008288603765127746, + "loss": 0.5246, + "step": 6222 + }, + { + "epoch": 0.3476827667122943, + "grad_norm": 0.7301790118217468, + "learning_rate": 0.0008288323621694308, + "loss": 0.4941, + "step": 6223 + }, + { + "epoch": 0.3477386373159762, + "grad_norm": 0.5693417191505432, + "learning_rate": 0.000828804347826087, + "loss": 0.5444, + "step": 6224 + }, + { + "epoch": 0.34779450791965805, + "grad_norm": 0.6701192259788513, + "learning_rate": 0.0008287763334827432, + "loss": 0.6505, + "step": 6225 + }, + { + "epoch": 0.34785037852333994, + "grad_norm": 4.143788814544678, + "learning_rate": 0.0008287483191393994, + "loss": 0.3954, + "step": 6226 + }, + { + "epoch": 0.34790624912702184, + "grad_norm": 0.45081260800361633, + "learning_rate": 0.0008287203047960556, + "loss": 0.3912, + "step": 6227 + }, + { + "epoch": 0.3479621197307037, + "grad_norm": 0.6436135172843933, + "learning_rate": 0.0008286922904527118, + "loss": 0.5505, + "step": 6228 + }, + { + "epoch": 0.3480179903343856, + "grad_norm": 0.5074776411056519, + "learning_rate": 0.000828664276109368, + "loss": 0.4908, + "step": 6229 + }, + { + "epoch": 0.3480738609380674, + "grad_norm": 2.792149543762207, + "learning_rate": 0.0008286362617660242, + "loss": 0.4693, + "step": 6230 + }, + { + "epoch": 0.3481297315417493, + "grad_norm": 0.49164900183677673, + "learning_rate": 0.0008286082474226805, + "loss": 0.4625, + "step": 6231 + }, + { + "epoch": 0.3481856021454312, + "grad_norm": 0.742106556892395, + "learning_rate": 0.0008285802330793366, + "loss": 0.4498, + "step": 6232 + }, + { + "epoch": 0.34824147274911305, + "grad_norm": 0.5886502861976624, + "learning_rate": 0.0008285522187359928, + "loss": 0.3438, + "step": 6233 + }, + { + "epoch": 0.34829734335279494, + "grad_norm": 0.5255749225616455, + "learning_rate": 0.000828524204392649, + "loss": 0.5296, + "step": 6234 + }, + { + "epoch": 0.3483532139564768, + "grad_norm": 0.5893737077713013, + "learning_rate": 0.0008284961900493052, + "loss": 0.4503, + "step": 6235 + }, + { + "epoch": 0.3484090845601587, + "grad_norm": 0.5382351875305176, + "learning_rate": 0.0008284681757059615, + "loss": 0.4859, + "step": 6236 + }, + { + "epoch": 0.3484649551638406, + "grad_norm": 1.0489513874053955, + "learning_rate": 0.0008284401613626177, + "loss": 0.4329, + "step": 6237 + }, + { + "epoch": 0.3485208257675224, + "grad_norm": 0.38606777787208557, + "learning_rate": 0.0008284121470192739, + "loss": 0.4675, + "step": 6238 + }, + { + "epoch": 0.3485766963712043, + "grad_norm": 0.4044322371482849, + "learning_rate": 0.0008283841326759301, + "loss": 0.4663, + "step": 6239 + }, + { + "epoch": 0.34863256697488615, + "grad_norm": 0.4742310345172882, + "learning_rate": 0.0008283561183325863, + "loss": 0.6109, + "step": 6240 + }, + { + "epoch": 0.34868843757856804, + "grad_norm": 0.5648439526557922, + "learning_rate": 0.0008283281039892425, + "loss": 0.4871, + "step": 6241 + }, + { + "epoch": 0.3487443081822499, + "grad_norm": 0.49997326731681824, + "learning_rate": 0.0008283000896458988, + "loss": 0.4176, + "step": 6242 + }, + { + "epoch": 0.3488001787859318, + "grad_norm": 0.7731658220291138, + "learning_rate": 0.0008282720753025549, + "loss": 0.5491, + "step": 6243 + }, + { + "epoch": 0.3488560493896137, + "grad_norm": 0.5663182139396667, + "learning_rate": 0.0008282440609592111, + "loss": 0.3587, + "step": 6244 + }, + { + "epoch": 0.3489119199932955, + "grad_norm": 0.576610267162323, + "learning_rate": 0.0008282160466158673, + "loss": 0.5893, + "step": 6245 + }, + { + "epoch": 0.3489677905969774, + "grad_norm": 0.6167611479759216, + "learning_rate": 0.0008281880322725236, + "loss": 0.3461, + "step": 6246 + }, + { + "epoch": 0.34902366120065925, + "grad_norm": 0.4240652620792389, + "learning_rate": 0.0008281600179291797, + "loss": 0.4349, + "step": 6247 + }, + { + "epoch": 0.34907953180434115, + "grad_norm": 1.1053932905197144, + "learning_rate": 0.0008281320035858359, + "loss": 0.5343, + "step": 6248 + }, + { + "epoch": 0.34913540240802304, + "grad_norm": 1.1944725513458252, + "learning_rate": 0.0008281039892424922, + "loss": 0.526, + "step": 6249 + }, + { + "epoch": 0.3491912730117049, + "grad_norm": 0.5837888121604919, + "learning_rate": 0.0008280759748991484, + "loss": 0.5715, + "step": 6250 + }, + { + "epoch": 0.3492471436153868, + "grad_norm": 0.7926021218299866, + "learning_rate": 0.0008280479605558045, + "loss": 0.3962, + "step": 6251 + }, + { + "epoch": 0.3493030142190686, + "grad_norm": 0.46949300169944763, + "learning_rate": 0.0008280199462124607, + "loss": 0.4121, + "step": 6252 + }, + { + "epoch": 0.3493588848227505, + "grad_norm": 0.6085501909255981, + "learning_rate": 0.0008279919318691171, + "loss": 0.3992, + "step": 6253 + }, + { + "epoch": 0.34941475542643236, + "grad_norm": 0.4024346172809601, + "learning_rate": 0.0008279639175257732, + "loss": 0.3643, + "step": 6254 + }, + { + "epoch": 0.34947062603011425, + "grad_norm": 0.6190791130065918, + "learning_rate": 0.0008279359031824294, + "loss": 0.5094, + "step": 6255 + }, + { + "epoch": 0.34952649663379615, + "grad_norm": 4.586511611938477, + "learning_rate": 0.0008279078888390857, + "loss": 0.4528, + "step": 6256 + }, + { + "epoch": 0.349582367237478, + "grad_norm": 0.5229771137237549, + "learning_rate": 0.0008278798744957419, + "loss": 0.5473, + "step": 6257 + }, + { + "epoch": 0.3496382378411599, + "grad_norm": 0.3977530300617218, + "learning_rate": 0.000827851860152398, + "loss": 0.4333, + "step": 6258 + }, + { + "epoch": 0.3496941084448417, + "grad_norm": 0.6942669153213501, + "learning_rate": 0.0008278238458090542, + "loss": 0.4856, + "step": 6259 + }, + { + "epoch": 0.3497499790485236, + "grad_norm": 0.5638141632080078, + "learning_rate": 0.0008277958314657105, + "loss": 0.5965, + "step": 6260 + }, + { + "epoch": 0.3498058496522055, + "grad_norm": 0.5180565714836121, + "learning_rate": 0.0008277678171223667, + "loss": 0.4009, + "step": 6261 + }, + { + "epoch": 0.34986172025588735, + "grad_norm": 0.7079669237136841, + "learning_rate": 0.0008277398027790228, + "loss": 0.5925, + "step": 6262 + }, + { + "epoch": 0.34991759085956925, + "grad_norm": 0.5055522322654724, + "learning_rate": 0.0008277117884356791, + "loss": 0.5569, + "step": 6263 + }, + { + "epoch": 0.3499734614632511, + "grad_norm": 0.4697633683681488, + "learning_rate": 0.0008276837740923353, + "loss": 0.4357, + "step": 6264 + }, + { + "epoch": 0.350029332066933, + "grad_norm": 0.4925629198551178, + "learning_rate": 0.0008276557597489915, + "loss": 0.535, + "step": 6265 + }, + { + "epoch": 0.3500852026706149, + "grad_norm": 0.42854583263397217, + "learning_rate": 0.0008276277454056476, + "loss": 0.3858, + "step": 6266 + }, + { + "epoch": 0.3501410732742967, + "grad_norm": 0.5348297357559204, + "learning_rate": 0.0008275997310623039, + "loss": 0.3943, + "step": 6267 + }, + { + "epoch": 0.3501969438779786, + "grad_norm": 0.43173831701278687, + "learning_rate": 0.0008275717167189601, + "loss": 0.4665, + "step": 6268 + }, + { + "epoch": 0.35025281448166046, + "grad_norm": 0.5315508246421814, + "learning_rate": 0.0008275437023756164, + "loss": 0.3876, + "step": 6269 + }, + { + "epoch": 0.35030868508534235, + "grad_norm": 0.8165509104728699, + "learning_rate": 0.0008275156880322725, + "loss": 0.4371, + "step": 6270 + }, + { + "epoch": 0.3503645556890242, + "grad_norm": 1.4363017082214355, + "learning_rate": 0.0008274876736889288, + "loss": 0.4738, + "step": 6271 + }, + { + "epoch": 0.3504204262927061, + "grad_norm": 0.5078338384628296, + "learning_rate": 0.000827459659345585, + "loss": 0.4417, + "step": 6272 + }, + { + "epoch": 0.350476296896388, + "grad_norm": 0.978949785232544, + "learning_rate": 0.0008274316450022412, + "loss": 0.5534, + "step": 6273 + }, + { + "epoch": 0.3505321675000698, + "grad_norm": 0.7218210697174072, + "learning_rate": 0.0008274036306588974, + "loss": 0.5256, + "step": 6274 + }, + { + "epoch": 0.3505880381037517, + "grad_norm": 0.8835763931274414, + "learning_rate": 0.0008273756163155536, + "loss": 0.7023, + "step": 6275 + }, + { + "epoch": 0.35064390870743356, + "grad_norm": 0.5215734243392944, + "learning_rate": 0.0008273476019722098, + "loss": 0.4697, + "step": 6276 + }, + { + "epoch": 0.35069977931111546, + "grad_norm": 0.5817258358001709, + "learning_rate": 0.0008273195876288659, + "loss": 0.3562, + "step": 6277 + }, + { + "epoch": 0.35075564991479735, + "grad_norm": 0.8038312792778015, + "learning_rate": 0.0008272915732855222, + "loss": 0.3981, + "step": 6278 + }, + { + "epoch": 0.3508115205184792, + "grad_norm": 0.5963142514228821, + "learning_rate": 0.0008272635589421784, + "loss": 0.3828, + "step": 6279 + }, + { + "epoch": 0.3508673911221611, + "grad_norm": 0.4890156388282776, + "learning_rate": 0.0008272355445988346, + "loss": 0.3602, + "step": 6280 + }, + { + "epoch": 0.3509232617258429, + "grad_norm": 0.4957914352416992, + "learning_rate": 0.0008272075302554908, + "loss": 0.3725, + "step": 6281 + }, + { + "epoch": 0.3509791323295248, + "grad_norm": 0.40280386805534363, + "learning_rate": 0.000827179515912147, + "loss": 0.3965, + "step": 6282 + }, + { + "epoch": 0.3510350029332067, + "grad_norm": 0.49780380725860596, + "learning_rate": 0.0008271515015688032, + "loss": 0.5203, + "step": 6283 + }, + { + "epoch": 0.35109087353688856, + "grad_norm": 0.46002820134162903, + "learning_rate": 0.0008271234872254594, + "loss": 0.5076, + "step": 6284 + }, + { + "epoch": 0.35114674414057045, + "grad_norm": 0.4131922423839569, + "learning_rate": 0.0008270954728821156, + "loss": 0.3807, + "step": 6285 + }, + { + "epoch": 0.3512026147442523, + "grad_norm": 0.514276385307312, + "learning_rate": 0.0008270674585387719, + "loss": 0.4174, + "step": 6286 + }, + { + "epoch": 0.3512584853479342, + "grad_norm": 0.5849116444587708, + "learning_rate": 0.0008270394441954281, + "loss": 0.4039, + "step": 6287 + }, + { + "epoch": 0.35131435595161603, + "grad_norm": 0.3968336880207062, + "learning_rate": 0.0008270114298520844, + "loss": 0.4495, + "step": 6288 + }, + { + "epoch": 0.3513702265552979, + "grad_norm": 0.8204522132873535, + "learning_rate": 0.0008269834155087405, + "loss": 0.6892, + "step": 6289 + }, + { + "epoch": 0.3514260971589798, + "grad_norm": 0.79197758436203, + "learning_rate": 0.0008269554011653967, + "loss": 0.9206, + "step": 6290 + }, + { + "epoch": 0.35148196776266166, + "grad_norm": 0.5600926280021667, + "learning_rate": 0.0008269273868220529, + "loss": 0.4636, + "step": 6291 + }, + { + "epoch": 0.35153783836634356, + "grad_norm": 0.8454629182815552, + "learning_rate": 0.0008268993724787092, + "loss": 0.4103, + "step": 6292 + }, + { + "epoch": 0.3515937089700254, + "grad_norm": 0.5113289952278137, + "learning_rate": 0.0008268713581353653, + "loss": 0.4657, + "step": 6293 + }, + { + "epoch": 0.3516495795737073, + "grad_norm": 2.243358612060547, + "learning_rate": 0.0008268433437920215, + "loss": 0.493, + "step": 6294 + }, + { + "epoch": 0.3517054501773892, + "grad_norm": 0.6552174091339111, + "learning_rate": 0.0008268153294486778, + "loss": 0.5168, + "step": 6295 + }, + { + "epoch": 0.35176132078107103, + "grad_norm": 0.7504728436470032, + "learning_rate": 0.000826787315105334, + "loss": 0.5225, + "step": 6296 + }, + { + "epoch": 0.3518171913847529, + "grad_norm": 1.0198179483413696, + "learning_rate": 0.0008267593007619901, + "loss": 0.4886, + "step": 6297 + }, + { + "epoch": 0.35187306198843477, + "grad_norm": 0.5285964012145996, + "learning_rate": 0.0008267312864186463, + "loss": 0.5199, + "step": 6298 + }, + { + "epoch": 0.35192893259211666, + "grad_norm": 0.489683598279953, + "learning_rate": 0.0008267032720753026, + "loss": 0.5345, + "step": 6299 + }, + { + "epoch": 0.35198480319579856, + "grad_norm": 0.4981469511985779, + "learning_rate": 0.0008266752577319587, + "loss": 0.4397, + "step": 6300 + }, + { + "epoch": 0.3520406737994804, + "grad_norm": 1.0357290506362915, + "learning_rate": 0.0008266472433886149, + "loss": 0.4552, + "step": 6301 + }, + { + "epoch": 0.3520965444031623, + "grad_norm": 0.9967161417007446, + "learning_rate": 0.0008266192290452713, + "loss": 0.4337, + "step": 6302 + }, + { + "epoch": 0.35215241500684413, + "grad_norm": 0.45584893226623535, + "learning_rate": 0.0008265912147019275, + "loss": 0.4878, + "step": 6303 + }, + { + "epoch": 0.35220828561052603, + "grad_norm": 0.4536430537700653, + "learning_rate": 0.0008265632003585836, + "loss": 0.4896, + "step": 6304 + }, + { + "epoch": 0.35226415621420787, + "grad_norm": 0.6491416692733765, + "learning_rate": 0.0008265351860152398, + "loss": 0.3961, + "step": 6305 + }, + { + "epoch": 0.35232002681788976, + "grad_norm": 0.7829586863517761, + "learning_rate": 0.0008265071716718961, + "loss": 0.3991, + "step": 6306 + }, + { + "epoch": 0.35237589742157166, + "grad_norm": 0.5000866651535034, + "learning_rate": 0.0008264791573285523, + "loss": 0.4252, + "step": 6307 + }, + { + "epoch": 0.3524317680252535, + "grad_norm": 0.7431145906448364, + "learning_rate": 0.0008264511429852084, + "loss": 0.5126, + "step": 6308 + }, + { + "epoch": 0.3524876386289354, + "grad_norm": 0.7291324138641357, + "learning_rate": 0.0008264231286418646, + "loss": 0.4405, + "step": 6309 + }, + { + "epoch": 0.35254350923261724, + "grad_norm": 0.8134508728981018, + "learning_rate": 0.0008263951142985209, + "loss": 0.6333, + "step": 6310 + }, + { + "epoch": 0.35259937983629913, + "grad_norm": 0.5022749900817871, + "learning_rate": 0.0008263670999551771, + "loss": 0.4157, + "step": 6311 + }, + { + "epoch": 0.352655250439981, + "grad_norm": 0.6167768239974976, + "learning_rate": 0.0008263390856118332, + "loss": 0.4419, + "step": 6312 + }, + { + "epoch": 0.35271112104366287, + "grad_norm": 3.062406063079834, + "learning_rate": 0.0008263110712684895, + "loss": 0.6253, + "step": 6313 + }, + { + "epoch": 0.35276699164734476, + "grad_norm": 0.4065723717212677, + "learning_rate": 0.0008262830569251457, + "loss": 0.405, + "step": 6314 + }, + { + "epoch": 0.3528228622510266, + "grad_norm": 0.6413792371749878, + "learning_rate": 0.0008262550425818019, + "loss": 0.4255, + "step": 6315 + }, + { + "epoch": 0.3528787328547085, + "grad_norm": 0.7272178530693054, + "learning_rate": 0.000826227028238458, + "loss": 0.4722, + "step": 6316 + }, + { + "epoch": 0.3529346034583904, + "grad_norm": 0.48313286900520325, + "learning_rate": 0.0008261990138951143, + "loss": 0.4216, + "step": 6317 + }, + { + "epoch": 0.35299047406207223, + "grad_norm": 0.46013686060905457, + "learning_rate": 0.0008261709995517705, + "loss": 0.4755, + "step": 6318 + }, + { + "epoch": 0.35304634466575413, + "grad_norm": 1.9778319597244263, + "learning_rate": 0.0008261429852084266, + "loss": 0.4727, + "step": 6319 + }, + { + "epoch": 0.35310221526943597, + "grad_norm": 0.7171513438224792, + "learning_rate": 0.000826114970865083, + "loss": 0.5624, + "step": 6320 + }, + { + "epoch": 0.35315808587311787, + "grad_norm": 0.5011780261993408, + "learning_rate": 0.0008260869565217392, + "loss": 0.3734, + "step": 6321 + }, + { + "epoch": 0.3532139564767997, + "grad_norm": 0.5717525482177734, + "learning_rate": 0.0008260589421783954, + "loss": 0.4783, + "step": 6322 + }, + { + "epoch": 0.3532698270804816, + "grad_norm": 0.7278454899787903, + "learning_rate": 0.0008260309278350515, + "loss": 0.4896, + "step": 6323 + }, + { + "epoch": 0.3533256976841635, + "grad_norm": 0.5919605493545532, + "learning_rate": 0.0008260029134917078, + "loss": 0.3913, + "step": 6324 + }, + { + "epoch": 0.35338156828784534, + "grad_norm": 0.5572449564933777, + "learning_rate": 0.000825974899148364, + "loss": 0.6, + "step": 6325 + }, + { + "epoch": 0.35343743889152723, + "grad_norm": 0.9174687266349792, + "learning_rate": 0.0008259468848050202, + "loss": 0.3614, + "step": 6326 + }, + { + "epoch": 0.3534933094952091, + "grad_norm": 0.6517714858055115, + "learning_rate": 0.0008259188704616764, + "loss": 0.5588, + "step": 6327 + }, + { + "epoch": 0.35354918009889097, + "grad_norm": 1.0218665599822998, + "learning_rate": 0.0008258908561183326, + "loss": 0.4483, + "step": 6328 + }, + { + "epoch": 0.35360505070257287, + "grad_norm": 0.550144612789154, + "learning_rate": 0.0008258628417749888, + "loss": 0.4559, + "step": 6329 + }, + { + "epoch": 0.3536609213062547, + "grad_norm": 0.4544374942779541, + "learning_rate": 0.000825834827431645, + "loss": 0.6815, + "step": 6330 + }, + { + "epoch": 0.3537167919099366, + "grad_norm": 0.5173408389091492, + "learning_rate": 0.0008258068130883012, + "loss": 0.4734, + "step": 6331 + }, + { + "epoch": 0.35377266251361844, + "grad_norm": 10.76001262664795, + "learning_rate": 0.0008257787987449574, + "loss": 0.5567, + "step": 6332 + }, + { + "epoch": 0.35382853311730034, + "grad_norm": 4.885684490203857, + "learning_rate": 0.0008257507844016136, + "loss": 0.4141, + "step": 6333 + }, + { + "epoch": 0.35388440372098223, + "grad_norm": 1.7115710973739624, + "learning_rate": 0.00082572277005827, + "loss": 0.5896, + "step": 6334 + }, + { + "epoch": 0.3539402743246641, + "grad_norm": 1.4591562747955322, + "learning_rate": 0.000825694755714926, + "loss": 0.4361, + "step": 6335 + }, + { + "epoch": 0.35399614492834597, + "grad_norm": 1.3025938272476196, + "learning_rate": 0.0008256667413715823, + "loss": 0.5949, + "step": 6336 + }, + { + "epoch": 0.3540520155320278, + "grad_norm": 0.4330524206161499, + "learning_rate": 0.0008256387270282385, + "loss": 0.4247, + "step": 6337 + }, + { + "epoch": 0.3541078861357097, + "grad_norm": 2.158782482147217, + "learning_rate": 0.0008256107126848948, + "loss": 0.3984, + "step": 6338 + }, + { + "epoch": 0.35416375673939154, + "grad_norm": 0.6874232292175293, + "learning_rate": 0.0008255826983415509, + "loss": 0.5585, + "step": 6339 + }, + { + "epoch": 0.35421962734307344, + "grad_norm": 0.6519833207130432, + "learning_rate": 0.0008255546839982071, + "loss": 0.5522, + "step": 6340 + }, + { + "epoch": 0.35427549794675534, + "grad_norm": 0.4480941891670227, + "learning_rate": 0.0008255266696548634, + "loss": 0.4734, + "step": 6341 + }, + { + "epoch": 0.3543313685504372, + "grad_norm": 0.8993548154830933, + "learning_rate": 0.0008254986553115195, + "loss": 0.5131, + "step": 6342 + }, + { + "epoch": 0.35438723915411907, + "grad_norm": 0.6886534094810486, + "learning_rate": 0.0008254706409681757, + "loss": 0.4235, + "step": 6343 + }, + { + "epoch": 0.3544431097578009, + "grad_norm": 0.43922916054725647, + "learning_rate": 0.0008254426266248319, + "loss": 0.4617, + "step": 6344 + }, + { + "epoch": 0.3544989803614828, + "grad_norm": 0.9847414493560791, + "learning_rate": 0.0008254146122814882, + "loss": 0.5132, + "step": 6345 + }, + { + "epoch": 0.3545548509651647, + "grad_norm": 0.4762301445007324, + "learning_rate": 0.0008253865979381443, + "loss": 0.4106, + "step": 6346 + }, + { + "epoch": 0.35461072156884654, + "grad_norm": 0.6528300046920776, + "learning_rate": 0.0008253585835948005, + "loss": 0.4888, + "step": 6347 + }, + { + "epoch": 0.35466659217252844, + "grad_norm": 1.622235655784607, + "learning_rate": 0.0008253305692514567, + "loss": 0.6096, + "step": 6348 + }, + { + "epoch": 0.3547224627762103, + "grad_norm": 0.48887038230895996, + "learning_rate": 0.000825302554908113, + "loss": 0.4037, + "step": 6349 + }, + { + "epoch": 0.3547783333798922, + "grad_norm": 1.2915005683898926, + "learning_rate": 0.0008252745405647691, + "loss": 0.3466, + "step": 6350 + }, + { + "epoch": 0.35483420398357407, + "grad_norm": 1.482955813407898, + "learning_rate": 0.0008252465262214253, + "loss": 0.3733, + "step": 6351 + }, + { + "epoch": 0.3548900745872559, + "grad_norm": 0.7276626229286194, + "learning_rate": 0.0008252185118780817, + "loss": 0.5777, + "step": 6352 + }, + { + "epoch": 0.3549459451909378, + "grad_norm": 5.269224643707275, + "learning_rate": 0.0008251904975347379, + "loss": 0.4208, + "step": 6353 + }, + { + "epoch": 0.35500181579461965, + "grad_norm": 0.6284651160240173, + "learning_rate": 0.000825162483191394, + "loss": 0.4678, + "step": 6354 + }, + { + "epoch": 0.35505768639830154, + "grad_norm": 0.3737070858478546, + "learning_rate": 0.0008251344688480502, + "loss": 0.412, + "step": 6355 + }, + { + "epoch": 0.3551135570019834, + "grad_norm": 0.5439226031303406, + "learning_rate": 0.0008251064545047065, + "loss": 0.485, + "step": 6356 + }, + { + "epoch": 0.3551694276056653, + "grad_norm": 0.919047474861145, + "learning_rate": 0.0008250784401613627, + "loss": 0.4957, + "step": 6357 + }, + { + "epoch": 0.3552252982093472, + "grad_norm": 0.5146211981773376, + "learning_rate": 0.0008250504258180188, + "loss": 0.546, + "step": 6358 + }, + { + "epoch": 0.355281168813029, + "grad_norm": 0.5898237228393555, + "learning_rate": 0.0008250224114746751, + "loss": 0.534, + "step": 6359 + }, + { + "epoch": 0.3553370394167109, + "grad_norm": 0.6761288642883301, + "learning_rate": 0.0008249943971313313, + "loss": 0.387, + "step": 6360 + }, + { + "epoch": 0.35539291002039275, + "grad_norm": 1.0436325073242188, + "learning_rate": 0.0008249663827879874, + "loss": 0.7254, + "step": 6361 + }, + { + "epoch": 0.35544878062407465, + "grad_norm": 0.6188142895698547, + "learning_rate": 0.0008249383684446436, + "loss": 0.5498, + "step": 6362 + }, + { + "epoch": 0.35550465122775654, + "grad_norm": 0.49158623814582825, + "learning_rate": 0.0008249103541012999, + "loss": 0.5081, + "step": 6363 + }, + { + "epoch": 0.3555605218314384, + "grad_norm": 0.5876120924949646, + "learning_rate": 0.0008248823397579561, + "loss": 0.5545, + "step": 6364 + }, + { + "epoch": 0.3556163924351203, + "grad_norm": 0.47148317098617554, + "learning_rate": 0.0008248543254146122, + "loss": 0.5354, + "step": 6365 + }, + { + "epoch": 0.3556722630388021, + "grad_norm": 0.6199638247489929, + "learning_rate": 0.0008248263110712685, + "loss": 0.4797, + "step": 6366 + }, + { + "epoch": 0.355728133642484, + "grad_norm": 0.7910218238830566, + "learning_rate": 0.0008247982967279247, + "loss": 0.395, + "step": 6367 + }, + { + "epoch": 0.35578400424616585, + "grad_norm": 0.8025599122047424, + "learning_rate": 0.000824770282384581, + "loss": 0.471, + "step": 6368 + }, + { + "epoch": 0.35583987484984775, + "grad_norm": 2.9539408683776855, + "learning_rate": 0.000824742268041237, + "loss": 0.4184, + "step": 6369 + }, + { + "epoch": 0.35589574545352964, + "grad_norm": 2.9960551261901855, + "learning_rate": 0.0008247142536978934, + "loss": 0.6209, + "step": 6370 + }, + { + "epoch": 0.3559516160572115, + "grad_norm": 0.6449220180511475, + "learning_rate": 0.0008246862393545496, + "loss": 0.4809, + "step": 6371 + }, + { + "epoch": 0.3560074866608934, + "grad_norm": 0.7139711976051331, + "learning_rate": 0.0008246582250112058, + "loss": 0.4174, + "step": 6372 + }, + { + "epoch": 0.3560633572645752, + "grad_norm": 0.6111873388290405, + "learning_rate": 0.000824630210667862, + "loss": 0.4806, + "step": 6373 + }, + { + "epoch": 0.3561192278682571, + "grad_norm": 0.8181046843528748, + "learning_rate": 0.0008246021963245182, + "loss": 0.5784, + "step": 6374 + }, + { + "epoch": 0.356175098471939, + "grad_norm": 1.242393136024475, + "learning_rate": 0.0008245741819811744, + "loss": 0.3863, + "step": 6375 + }, + { + "epoch": 0.35623096907562085, + "grad_norm": 0.4295813739299774, + "learning_rate": 0.0008245461676378306, + "loss": 0.4469, + "step": 6376 + }, + { + "epoch": 0.35628683967930275, + "grad_norm": 0.5454009771347046, + "learning_rate": 0.0008245181532944868, + "loss": 0.4, + "step": 6377 + }, + { + "epoch": 0.3563427102829846, + "grad_norm": 1.2228155136108398, + "learning_rate": 0.000824490138951143, + "loss": 0.4596, + "step": 6378 + }, + { + "epoch": 0.3563985808866665, + "grad_norm": 2.869554281234741, + "learning_rate": 0.0008244621246077992, + "loss": 0.4965, + "step": 6379 + }, + { + "epoch": 0.3564544514903484, + "grad_norm": 0.5591950416564941, + "learning_rate": 0.0008244341102644555, + "loss": 0.5331, + "step": 6380 + }, + { + "epoch": 0.3565103220940302, + "grad_norm": 0.8910638689994812, + "learning_rate": 0.0008244060959211116, + "loss": 0.5943, + "step": 6381 + }, + { + "epoch": 0.3565661926977121, + "grad_norm": 0.4550562798976898, + "learning_rate": 0.0008243780815777678, + "loss": 0.4073, + "step": 6382 + }, + { + "epoch": 0.35662206330139395, + "grad_norm": 1.1828960180282593, + "learning_rate": 0.000824350067234424, + "loss": 0.3817, + "step": 6383 + }, + { + "epoch": 0.35667793390507585, + "grad_norm": 0.5765656232833862, + "learning_rate": 0.0008243220528910802, + "loss": 0.4524, + "step": 6384 + }, + { + "epoch": 0.3567338045087577, + "grad_norm": 0.3477967381477356, + "learning_rate": 0.0008242940385477364, + "loss": 0.3707, + "step": 6385 + }, + { + "epoch": 0.3567896751124396, + "grad_norm": 0.8962097764015198, + "learning_rate": 0.0008242660242043927, + "loss": 0.4701, + "step": 6386 + }, + { + "epoch": 0.3568455457161215, + "grad_norm": 0.6188319325447083, + "learning_rate": 0.0008242380098610489, + "loss": 0.5652, + "step": 6387 + }, + { + "epoch": 0.3569014163198033, + "grad_norm": 1.373807668685913, + "learning_rate": 0.0008242099955177051, + "loss": 0.5345, + "step": 6388 + }, + { + "epoch": 0.3569572869234852, + "grad_norm": 0.4321742355823517, + "learning_rate": 0.0008241819811743613, + "loss": 0.4311, + "step": 6389 + }, + { + "epoch": 0.35701315752716706, + "grad_norm": 0.5121470093727112, + "learning_rate": 0.0008241539668310175, + "loss": 0.4886, + "step": 6390 + }, + { + "epoch": 0.35706902813084895, + "grad_norm": 0.3865099251270294, + "learning_rate": 0.0008241259524876738, + "loss": 0.3465, + "step": 6391 + }, + { + "epoch": 0.35712489873453085, + "grad_norm": 0.5115795731544495, + "learning_rate": 0.0008240979381443299, + "loss": 0.5519, + "step": 6392 + }, + { + "epoch": 0.3571807693382127, + "grad_norm": 0.45735809206962585, + "learning_rate": 0.0008240699238009861, + "loss": 0.3922, + "step": 6393 + }, + { + "epoch": 0.3572366399418946, + "grad_norm": 0.5153868198394775, + "learning_rate": 0.0008240419094576423, + "loss": 0.567, + "step": 6394 + }, + { + "epoch": 0.3572925105455764, + "grad_norm": 0.46305209398269653, + "learning_rate": 0.0008240138951142986, + "loss": 0.4364, + "step": 6395 + }, + { + "epoch": 0.3573483811492583, + "grad_norm": 0.784221351146698, + "learning_rate": 0.0008239858807709547, + "loss": 0.4285, + "step": 6396 + }, + { + "epoch": 0.3574042517529402, + "grad_norm": 0.4202946722507477, + "learning_rate": 0.0008239578664276109, + "loss": 0.4652, + "step": 6397 + }, + { + "epoch": 0.35746012235662206, + "grad_norm": 0.5859045386314392, + "learning_rate": 0.0008239298520842672, + "loss": 0.5491, + "step": 6398 + }, + { + "epoch": 0.35751599296030395, + "grad_norm": 1.1369577646255493, + "learning_rate": 0.0008239018377409234, + "loss": 0.4032, + "step": 6399 + }, + { + "epoch": 0.3575718635639858, + "grad_norm": 0.5728000998497009, + "learning_rate": 0.0008238738233975795, + "loss": 0.4757, + "step": 6400 + }, + { + "epoch": 0.3576277341676677, + "grad_norm": 0.7874393463134766, + "learning_rate": 0.0008238458090542357, + "loss": 0.3882, + "step": 6401 + }, + { + "epoch": 0.35768360477134953, + "grad_norm": 0.7946639060974121, + "learning_rate": 0.0008238177947108921, + "loss": 0.5687, + "step": 6402 + }, + { + "epoch": 0.3577394753750314, + "grad_norm": 0.5648518204689026, + "learning_rate": 0.0008237897803675482, + "loss": 0.5143, + "step": 6403 + }, + { + "epoch": 0.3577953459787133, + "grad_norm": 0.3675774037837982, + "learning_rate": 0.0008237617660242044, + "loss": 0.4401, + "step": 6404 + }, + { + "epoch": 0.35785121658239516, + "grad_norm": 0.5797474980354309, + "learning_rate": 0.0008237337516808607, + "loss": 0.4185, + "step": 6405 + }, + { + "epoch": 0.35790708718607706, + "grad_norm": 0.48755931854248047, + "learning_rate": 0.0008237057373375169, + "loss": 0.4589, + "step": 6406 + }, + { + "epoch": 0.3579629577897589, + "grad_norm": 0.7678301930427551, + "learning_rate": 0.000823677722994173, + "loss": 0.4524, + "step": 6407 + }, + { + "epoch": 0.3580188283934408, + "grad_norm": 0.7032248973846436, + "learning_rate": 0.0008236497086508292, + "loss": 0.4129, + "step": 6408 + }, + { + "epoch": 0.3580746989971227, + "grad_norm": 1.816131353378296, + "learning_rate": 0.0008236216943074855, + "loss": 0.5626, + "step": 6409 + }, + { + "epoch": 0.3581305696008045, + "grad_norm": 0.48849526047706604, + "learning_rate": 0.0008235936799641417, + "loss": 0.4179, + "step": 6410 + }, + { + "epoch": 0.3581864402044864, + "grad_norm": 0.6107448935508728, + "learning_rate": 0.0008235656656207978, + "loss": 0.4808, + "step": 6411 + }, + { + "epoch": 0.35824231080816826, + "grad_norm": 0.5292683839797974, + "learning_rate": 0.0008235376512774541, + "loss": 0.4902, + "step": 6412 + }, + { + "epoch": 0.35829818141185016, + "grad_norm": 1.0175635814666748, + "learning_rate": 0.0008235096369341103, + "loss": 0.4637, + "step": 6413 + }, + { + "epoch": 0.35835405201553205, + "grad_norm": 0.4654546082019806, + "learning_rate": 0.0008234816225907665, + "loss": 0.4084, + "step": 6414 + }, + { + "epoch": 0.3584099226192139, + "grad_norm": 0.7552298903465271, + "learning_rate": 0.0008234536082474226, + "loss": 0.4172, + "step": 6415 + }, + { + "epoch": 0.3584657932228958, + "grad_norm": 0.39880427718162537, + "learning_rate": 0.0008234255939040789, + "loss": 0.3852, + "step": 6416 + }, + { + "epoch": 0.35852166382657763, + "grad_norm": 0.4776149392127991, + "learning_rate": 0.0008233975795607351, + "loss": 0.5051, + "step": 6417 + }, + { + "epoch": 0.3585775344302595, + "grad_norm": 2.926621198654175, + "learning_rate": 0.0008233695652173913, + "loss": 0.5902, + "step": 6418 + }, + { + "epoch": 0.35863340503394137, + "grad_norm": 0.5482921004295349, + "learning_rate": 0.0008233415508740474, + "loss": 0.6186, + "step": 6419 + }, + { + "epoch": 0.35868927563762326, + "grad_norm": 1.1431884765625, + "learning_rate": 0.0008233135365307038, + "loss": 0.5593, + "step": 6420 + }, + { + "epoch": 0.35874514624130516, + "grad_norm": 1.5101624727249146, + "learning_rate": 0.00082328552218736, + "loss": 0.4897, + "step": 6421 + }, + { + "epoch": 0.358801016844987, + "grad_norm": 0.7703801393508911, + "learning_rate": 0.0008232575078440162, + "loss": 0.3715, + "step": 6422 + }, + { + "epoch": 0.3588568874486689, + "grad_norm": 0.37360915541648865, + "learning_rate": 0.0008232294935006724, + "loss": 0.4556, + "step": 6423 + }, + { + "epoch": 0.35891275805235073, + "grad_norm": 1.140684962272644, + "learning_rate": 0.0008232014791573286, + "loss": 0.4292, + "step": 6424 + }, + { + "epoch": 0.35896862865603263, + "grad_norm": 0.6387362480163574, + "learning_rate": 0.0008231734648139848, + "loss": 0.5689, + "step": 6425 + }, + { + "epoch": 0.3590244992597145, + "grad_norm": 0.5890669822692871, + "learning_rate": 0.0008231454504706409, + "loss": 0.4958, + "step": 6426 + }, + { + "epoch": 0.35908036986339636, + "grad_norm": 0.500892162322998, + "learning_rate": 0.0008231174361272972, + "loss": 0.4688, + "step": 6427 + }, + { + "epoch": 0.35913624046707826, + "grad_norm": 0.7243421673774719, + "learning_rate": 0.0008230894217839534, + "loss": 0.6332, + "step": 6428 + }, + { + "epoch": 0.3591921110707601, + "grad_norm": 1.5617477893829346, + "learning_rate": 0.0008230614074406096, + "loss": 0.5218, + "step": 6429 + }, + { + "epoch": 0.359247981674442, + "grad_norm": 0.7575781345367432, + "learning_rate": 0.0008230333930972658, + "loss": 0.3864, + "step": 6430 + }, + { + "epoch": 0.3593038522781239, + "grad_norm": 0.5030255317687988, + "learning_rate": 0.000823005378753922, + "loss": 0.6009, + "step": 6431 + }, + { + "epoch": 0.35935972288180573, + "grad_norm": 9.549452781677246, + "learning_rate": 0.0008229773644105782, + "loss": 0.4335, + "step": 6432 + }, + { + "epoch": 0.35941559348548763, + "grad_norm": 0.39250752329826355, + "learning_rate": 0.0008229493500672344, + "loss": 0.4324, + "step": 6433 + }, + { + "epoch": 0.35947146408916947, + "grad_norm": 0.4698900282382965, + "learning_rate": 0.0008229213357238906, + "loss": 0.5136, + "step": 6434 + }, + { + "epoch": 0.35952733469285136, + "grad_norm": 8.296608924865723, + "learning_rate": 0.0008228933213805468, + "loss": 0.4335, + "step": 6435 + }, + { + "epoch": 0.3595832052965332, + "grad_norm": 0.6039557456970215, + "learning_rate": 0.000822865307037203, + "loss": 0.4965, + "step": 6436 + }, + { + "epoch": 0.3596390759002151, + "grad_norm": 0.6718931794166565, + "learning_rate": 0.0008228372926938594, + "loss": 0.5039, + "step": 6437 + }, + { + "epoch": 0.359694946503897, + "grad_norm": 0.6942671537399292, + "learning_rate": 0.0008228092783505155, + "loss": 0.3727, + "step": 6438 + }, + { + "epoch": 0.35975081710757884, + "grad_norm": 0.8868228197097778, + "learning_rate": 0.0008227812640071717, + "loss": 0.4662, + "step": 6439 + }, + { + "epoch": 0.35980668771126073, + "grad_norm": 0.6035279631614685, + "learning_rate": 0.0008227532496638279, + "loss": 0.4797, + "step": 6440 + }, + { + "epoch": 0.35986255831494257, + "grad_norm": 0.9580274224281311, + "learning_rate": 0.0008227252353204842, + "loss": 0.8019, + "step": 6441 + }, + { + "epoch": 0.35991842891862447, + "grad_norm": 0.5210030674934387, + "learning_rate": 0.0008226972209771403, + "loss": 0.472, + "step": 6442 + }, + { + "epoch": 0.35997429952230636, + "grad_norm": 0.6664880514144897, + "learning_rate": 0.0008226692066337965, + "loss": 0.4942, + "step": 6443 + }, + { + "epoch": 0.3600301701259882, + "grad_norm": 0.7182350158691406, + "learning_rate": 0.0008226411922904528, + "loss": 0.4546, + "step": 6444 + }, + { + "epoch": 0.3600860407296701, + "grad_norm": 0.5951759219169617, + "learning_rate": 0.0008226131779471089, + "loss": 0.5238, + "step": 6445 + }, + { + "epoch": 0.36014191133335194, + "grad_norm": 0.56834477186203, + "learning_rate": 0.0008225851636037651, + "loss": 0.5294, + "step": 6446 + }, + { + "epoch": 0.36019778193703383, + "grad_norm": 0.6476634740829468, + "learning_rate": 0.0008225571492604213, + "loss": 0.4876, + "step": 6447 + }, + { + "epoch": 0.36025365254071573, + "grad_norm": 0.7154616117477417, + "learning_rate": 0.0008225291349170776, + "loss": 0.4334, + "step": 6448 + }, + { + "epoch": 0.36030952314439757, + "grad_norm": 0.5109213590621948, + "learning_rate": 0.0008225011205737337, + "loss": 0.4768, + "step": 6449 + }, + { + "epoch": 0.36036539374807947, + "grad_norm": 0.5803002715110779, + "learning_rate": 0.0008224731062303899, + "loss": 0.5219, + "step": 6450 + }, + { + "epoch": 0.3604212643517613, + "grad_norm": 0.7561616897583008, + "learning_rate": 0.0008224450918870461, + "loss": 0.5268, + "step": 6451 + }, + { + "epoch": 0.3604771349554432, + "grad_norm": 0.6444684267044067, + "learning_rate": 0.0008224170775437025, + "loss": 0.5737, + "step": 6452 + }, + { + "epoch": 0.36053300555912504, + "grad_norm": 0.7180501818656921, + "learning_rate": 0.0008223890632003586, + "loss": 0.4877, + "step": 6453 + }, + { + "epoch": 0.36058887616280694, + "grad_norm": 0.6681928634643555, + "learning_rate": 0.0008223610488570148, + "loss": 0.4917, + "step": 6454 + }, + { + "epoch": 0.36064474676648883, + "grad_norm": 0.7270925641059875, + "learning_rate": 0.0008223330345136711, + "loss": 0.3791, + "step": 6455 + }, + { + "epoch": 0.3607006173701707, + "grad_norm": 0.6743019223213196, + "learning_rate": 0.0008223050201703273, + "loss": 0.573, + "step": 6456 + }, + { + "epoch": 0.36075648797385257, + "grad_norm": 1.4530447721481323, + "learning_rate": 0.0008222770058269834, + "loss": 0.5042, + "step": 6457 + }, + { + "epoch": 0.3608123585775344, + "grad_norm": 0.5506135821342468, + "learning_rate": 0.0008222489914836396, + "loss": 0.3861, + "step": 6458 + }, + { + "epoch": 0.3608682291812163, + "grad_norm": 0.42397403717041016, + "learning_rate": 0.0008222209771402959, + "loss": 0.3587, + "step": 6459 + }, + { + "epoch": 0.3609240997848982, + "grad_norm": 0.5143830180168152, + "learning_rate": 0.0008221929627969521, + "loss": 0.5231, + "step": 6460 + }, + { + "epoch": 0.36097997038858004, + "grad_norm": 0.4448000192642212, + "learning_rate": 0.0008221649484536082, + "loss": 0.4334, + "step": 6461 + }, + { + "epoch": 0.36103584099226194, + "grad_norm": 0.8825182914733887, + "learning_rate": 0.0008221369341102645, + "loss": 0.4265, + "step": 6462 + }, + { + "epoch": 0.3610917115959438, + "grad_norm": 0.470602422952652, + "learning_rate": 0.0008221089197669207, + "loss": 0.4129, + "step": 6463 + }, + { + "epoch": 0.36114758219962567, + "grad_norm": 1.321081519126892, + "learning_rate": 0.0008220809054235769, + "loss": 0.4785, + "step": 6464 + }, + { + "epoch": 0.3612034528033075, + "grad_norm": 0.5496178865432739, + "learning_rate": 0.000822052891080233, + "loss": 0.5595, + "step": 6465 + }, + { + "epoch": 0.3612593234069894, + "grad_norm": 0.6210770606994629, + "learning_rate": 0.0008220248767368893, + "loss": 0.4344, + "step": 6466 + }, + { + "epoch": 0.3613151940106713, + "grad_norm": 0.5699465870857239, + "learning_rate": 0.0008219968623935455, + "loss": 0.4794, + "step": 6467 + }, + { + "epoch": 0.36137106461435314, + "grad_norm": 0.632429838180542, + "learning_rate": 0.0008219688480502016, + "loss": 0.3716, + "step": 6468 + }, + { + "epoch": 0.36142693521803504, + "grad_norm": 0.4308183491230011, + "learning_rate": 0.000821940833706858, + "loss": 0.4884, + "step": 6469 + }, + { + "epoch": 0.3614828058217169, + "grad_norm": 1.747057318687439, + "learning_rate": 0.0008219128193635142, + "loss": 0.5626, + "step": 6470 + }, + { + "epoch": 0.3615386764253988, + "grad_norm": 0.4984160363674164, + "learning_rate": 0.0008218848050201704, + "loss": 0.4579, + "step": 6471 + }, + { + "epoch": 0.36159454702908067, + "grad_norm": 0.6626816391944885, + "learning_rate": 0.0008218567906768265, + "loss": 0.484, + "step": 6472 + }, + { + "epoch": 0.3616504176327625, + "grad_norm": 6.185814380645752, + "learning_rate": 0.0008218287763334828, + "loss": 0.5273, + "step": 6473 + }, + { + "epoch": 0.3617062882364444, + "grad_norm": 0.5676392912864685, + "learning_rate": 0.000821800761990139, + "loss": 0.5612, + "step": 6474 + }, + { + "epoch": 0.36176215884012625, + "grad_norm": 0.763146698474884, + "learning_rate": 0.0008217727476467952, + "loss": 0.5833, + "step": 6475 + }, + { + "epoch": 0.36181802944380814, + "grad_norm": 0.6734564304351807, + "learning_rate": 0.0008217447333034514, + "loss": 0.4907, + "step": 6476 + }, + { + "epoch": 0.36187390004749004, + "grad_norm": 0.46116355061531067, + "learning_rate": 0.0008217167189601076, + "loss": 0.4491, + "step": 6477 + }, + { + "epoch": 0.3619297706511719, + "grad_norm": 0.690844714641571, + "learning_rate": 0.0008216887046167638, + "loss": 0.6198, + "step": 6478 + }, + { + "epoch": 0.3619856412548538, + "grad_norm": 3.6262733936309814, + "learning_rate": 0.00082166069027342, + "loss": 0.4946, + "step": 6479 + }, + { + "epoch": 0.3620415118585356, + "grad_norm": 0.8427473306655884, + "learning_rate": 0.0008216326759300762, + "loss": 0.4515, + "step": 6480 + }, + { + "epoch": 0.3620973824622175, + "grad_norm": 0.6426787376403809, + "learning_rate": 0.0008216046615867324, + "loss": 0.6192, + "step": 6481 + }, + { + "epoch": 0.36215325306589935, + "grad_norm": 0.6486719250679016, + "learning_rate": 0.0008215766472433886, + "loss": 0.4185, + "step": 6482 + }, + { + "epoch": 0.36220912366958125, + "grad_norm": 0.548964262008667, + "learning_rate": 0.0008215486329000449, + "loss": 0.5166, + "step": 6483 + }, + { + "epoch": 0.36226499427326314, + "grad_norm": 0.5687814950942993, + "learning_rate": 0.000821520618556701, + "loss": 0.4499, + "step": 6484 + }, + { + "epoch": 0.362320864876945, + "grad_norm": 0.5195846557617188, + "learning_rate": 0.0008214926042133572, + "loss": 0.5155, + "step": 6485 + }, + { + "epoch": 0.3623767354806269, + "grad_norm": 0.5589302778244019, + "learning_rate": 0.0008214645898700135, + "loss": 0.4209, + "step": 6486 + }, + { + "epoch": 0.3624326060843087, + "grad_norm": 0.5253198146820068, + "learning_rate": 0.0008214365755266698, + "loss": 0.4644, + "step": 6487 + }, + { + "epoch": 0.3624884766879906, + "grad_norm": 0.5239326357841492, + "learning_rate": 0.0008214085611833259, + "loss": 0.5406, + "step": 6488 + }, + { + "epoch": 0.3625443472916725, + "grad_norm": 0.7761749029159546, + "learning_rate": 0.0008213805468399821, + "loss": 0.3877, + "step": 6489 + }, + { + "epoch": 0.36260021789535435, + "grad_norm": 0.8062101602554321, + "learning_rate": 0.0008213525324966383, + "loss": 0.5099, + "step": 6490 + }, + { + "epoch": 0.36265608849903624, + "grad_norm": 0.4624994397163391, + "learning_rate": 0.0008213245181532945, + "loss": 0.3489, + "step": 6491 + }, + { + "epoch": 0.3627119591027181, + "grad_norm": 1.0873057842254639, + "learning_rate": 0.0008212965038099507, + "loss": 0.6638, + "step": 6492 + }, + { + "epoch": 0.3627678297064, + "grad_norm": 0.5620323419570923, + "learning_rate": 0.0008212684894666069, + "loss": 0.5053, + "step": 6493 + }, + { + "epoch": 0.3628237003100819, + "grad_norm": 0.5227258801460266, + "learning_rate": 0.0008212404751232632, + "loss": 0.5737, + "step": 6494 + }, + { + "epoch": 0.3628795709137637, + "grad_norm": 0.7506962418556213, + "learning_rate": 0.0008212124607799193, + "loss": 0.4614, + "step": 6495 + }, + { + "epoch": 0.3629354415174456, + "grad_norm": 0.5590639710426331, + "learning_rate": 0.0008211844464365755, + "loss": 0.5448, + "step": 6496 + }, + { + "epoch": 0.36299131212112745, + "grad_norm": 0.679283857345581, + "learning_rate": 0.0008211564320932317, + "loss": 0.4975, + "step": 6497 + }, + { + "epoch": 0.36304718272480935, + "grad_norm": 0.5421339869499207, + "learning_rate": 0.000821128417749888, + "loss": 0.5176, + "step": 6498 + }, + { + "epoch": 0.3631030533284912, + "grad_norm": 0.419164776802063, + "learning_rate": 0.0008211004034065441, + "loss": 0.4841, + "step": 6499 + }, + { + "epoch": 0.3631589239321731, + "grad_norm": 0.4994907081127167, + "learning_rate": 0.0008210723890632003, + "loss": 0.5403, + "step": 6500 + }, + { + "epoch": 0.3631589239321731, + "eval_cer": 0.09405584105266604, + "eval_loss": 0.3518145680427551, + "eval_runtime": 56.3123, + "eval_samples_per_second": 80.586, + "eval_steps_per_second": 5.043, + "eval_wer": 0.372196307365716, + "step": 6500 + }, + { + "epoch": 0.363214794535855, + "grad_norm": 0.9762527942657471, + "learning_rate": 0.0008210443747198566, + "loss": 0.5723, + "step": 6501 + }, + { + "epoch": 0.3632706651395368, + "grad_norm": 0.47543710470199585, + "learning_rate": 0.0008210163603765129, + "loss": 0.6311, + "step": 6502 + }, + { + "epoch": 0.3633265357432187, + "grad_norm": 1.2459733486175537, + "learning_rate": 0.000820988346033169, + "loss": 0.676, + "step": 6503 + }, + { + "epoch": 0.36338240634690055, + "grad_norm": 0.6723775863647461, + "learning_rate": 0.0008209603316898252, + "loss": 0.5417, + "step": 6504 + }, + { + "epoch": 0.36343827695058245, + "grad_norm": 0.4232587516307831, + "learning_rate": 0.0008209323173464815, + "loss": 0.6506, + "step": 6505 + }, + { + "epoch": 0.36349414755426435, + "grad_norm": 0.6074180006980896, + "learning_rate": 0.0008209043030031377, + "loss": 0.5016, + "step": 6506 + }, + { + "epoch": 0.3635500181579462, + "grad_norm": 1.2157869338989258, + "learning_rate": 0.0008208762886597938, + "loss": 0.3523, + "step": 6507 + }, + { + "epoch": 0.3636058887616281, + "grad_norm": 0.6063497066497803, + "learning_rate": 0.0008208482743164501, + "loss": 0.5989, + "step": 6508 + }, + { + "epoch": 0.3636617593653099, + "grad_norm": 0.46300405263900757, + "learning_rate": 0.0008208202599731063, + "loss": 0.4465, + "step": 6509 + }, + { + "epoch": 0.3637176299689918, + "grad_norm": 0.532004177570343, + "learning_rate": 0.0008207922456297624, + "loss": 0.4724, + "step": 6510 + }, + { + "epoch": 0.3637735005726737, + "grad_norm": 0.5922155380249023, + "learning_rate": 0.0008207642312864186, + "loss": 0.4421, + "step": 6511 + }, + { + "epoch": 0.36382937117635555, + "grad_norm": 0.6165679097175598, + "learning_rate": 0.0008207362169430749, + "loss": 0.5122, + "step": 6512 + }, + { + "epoch": 0.36388524178003745, + "grad_norm": 0.43577855825424194, + "learning_rate": 0.0008207082025997311, + "loss": 0.4393, + "step": 6513 + }, + { + "epoch": 0.3639411123837193, + "grad_norm": 0.5168145895004272, + "learning_rate": 0.0008206801882563872, + "loss": 0.4242, + "step": 6514 + }, + { + "epoch": 0.3639969829874012, + "grad_norm": 0.9994063377380371, + "learning_rate": 0.0008206521739130435, + "loss": 0.4863, + "step": 6515 + }, + { + "epoch": 0.364052853591083, + "grad_norm": 1.6142085790634155, + "learning_rate": 0.0008206241595696997, + "loss": 0.3705, + "step": 6516 + }, + { + "epoch": 0.3641087241947649, + "grad_norm": 0.43708184361457825, + "learning_rate": 0.0008205961452263559, + "loss": 0.4594, + "step": 6517 + }, + { + "epoch": 0.3641645947984468, + "grad_norm": 3.4635186195373535, + "learning_rate": 0.000820568130883012, + "loss": 0.3999, + "step": 6518 + }, + { + "epoch": 0.36422046540212866, + "grad_norm": 0.5377785563468933, + "learning_rate": 0.0008205401165396684, + "loss": 0.4265, + "step": 6519 + }, + { + "epoch": 0.36427633600581055, + "grad_norm": 0.8294495940208435, + "learning_rate": 0.0008205121021963246, + "loss": 0.695, + "step": 6520 + }, + { + "epoch": 0.3643322066094924, + "grad_norm": 2.2361953258514404, + "learning_rate": 0.0008204840878529808, + "loss": 0.4143, + "step": 6521 + }, + { + "epoch": 0.3643880772131743, + "grad_norm": 0.5819830894470215, + "learning_rate": 0.0008204560735096369, + "loss": 0.4428, + "step": 6522 + }, + { + "epoch": 0.3644439478168562, + "grad_norm": 1.057499647140503, + "learning_rate": 0.0008204280591662932, + "loss": 0.5895, + "step": 6523 + }, + { + "epoch": 0.364499818420538, + "grad_norm": 0.45362040400505066, + "learning_rate": 0.0008204000448229494, + "loss": 0.4078, + "step": 6524 + }, + { + "epoch": 0.3645556890242199, + "grad_norm": 0.6551773548126221, + "learning_rate": 0.0008203720304796056, + "loss": 0.5275, + "step": 6525 + }, + { + "epoch": 0.36461155962790176, + "grad_norm": 0.6580588221549988, + "learning_rate": 0.0008203440161362618, + "loss": 0.5336, + "step": 6526 + }, + { + "epoch": 0.36466743023158366, + "grad_norm": 1.166867971420288, + "learning_rate": 0.000820316001792918, + "loss": 0.4146, + "step": 6527 + }, + { + "epoch": 0.36472330083526555, + "grad_norm": 0.4090692698955536, + "learning_rate": 0.0008202879874495742, + "loss": 0.4398, + "step": 6528 + }, + { + "epoch": 0.3647791714389474, + "grad_norm": 0.5556562542915344, + "learning_rate": 0.0008202599731062304, + "loss": 0.3274, + "step": 6529 + }, + { + "epoch": 0.3648350420426293, + "grad_norm": 0.8365010619163513, + "learning_rate": 0.0008202319587628866, + "loss": 0.6801, + "step": 6530 + }, + { + "epoch": 0.3648909126463111, + "grad_norm": 0.6008497476577759, + "learning_rate": 0.0008202039444195428, + "loss": 0.4668, + "step": 6531 + }, + { + "epoch": 0.364946783249993, + "grad_norm": 1.2929158210754395, + "learning_rate": 0.000820175930076199, + "loss": 0.4186, + "step": 6532 + }, + { + "epoch": 0.36500265385367486, + "grad_norm": 0.8284275531768799, + "learning_rate": 0.0008201479157328552, + "loss": 0.4583, + "step": 6533 + }, + { + "epoch": 0.36505852445735676, + "grad_norm": 1.2362267971038818, + "learning_rate": 0.0008201199013895114, + "loss": 0.4533, + "step": 6534 + }, + { + "epoch": 0.36511439506103865, + "grad_norm": 0.8599357604980469, + "learning_rate": 0.0008200918870461676, + "loss": 0.4127, + "step": 6535 + }, + { + "epoch": 0.3651702656647205, + "grad_norm": 0.46917831897735596, + "learning_rate": 0.0008200638727028239, + "loss": 0.3575, + "step": 6536 + }, + { + "epoch": 0.3652261362684024, + "grad_norm": 0.517162561416626, + "learning_rate": 0.0008200358583594801, + "loss": 0.3793, + "step": 6537 + }, + { + "epoch": 0.36528200687208423, + "grad_norm": 0.5193377137184143, + "learning_rate": 0.0008200078440161363, + "loss": 0.4159, + "step": 6538 + }, + { + "epoch": 0.3653378774757661, + "grad_norm": 4.4327216148376465, + "learning_rate": 0.0008199798296727925, + "loss": 0.5094, + "step": 6539 + }, + { + "epoch": 0.365393748079448, + "grad_norm": 0.4832030236721039, + "learning_rate": 0.0008199518153294488, + "loss": 0.4635, + "step": 6540 + }, + { + "epoch": 0.36544961868312986, + "grad_norm": 0.4460844397544861, + "learning_rate": 0.0008199238009861049, + "loss": 0.4119, + "step": 6541 + }, + { + "epoch": 0.36550548928681176, + "grad_norm": 0.49096861481666565, + "learning_rate": 0.0008198957866427611, + "loss": 0.476, + "step": 6542 + }, + { + "epoch": 0.3655613598904936, + "grad_norm": 0.45243725180625916, + "learning_rate": 0.0008198677722994173, + "loss": 0.489, + "step": 6543 + }, + { + "epoch": 0.3656172304941755, + "grad_norm": 0.9679161906242371, + "learning_rate": 0.0008198397579560736, + "loss": 0.4171, + "step": 6544 + }, + { + "epoch": 0.3656731010978574, + "grad_norm": 0.4893883466720581, + "learning_rate": 0.0008198117436127297, + "loss": 0.3717, + "step": 6545 + }, + { + "epoch": 0.36572897170153923, + "grad_norm": 0.5249677300453186, + "learning_rate": 0.0008197837292693859, + "loss": 0.5492, + "step": 6546 + }, + { + "epoch": 0.3657848423052211, + "grad_norm": 0.6052812337875366, + "learning_rate": 0.0008197557149260422, + "loss": 0.4904, + "step": 6547 + }, + { + "epoch": 0.36584071290890297, + "grad_norm": 0.8571063280105591, + "learning_rate": 0.0008197277005826984, + "loss": 0.4965, + "step": 6548 + }, + { + "epoch": 0.36589658351258486, + "grad_norm": 0.4626041352748871, + "learning_rate": 0.0008196996862393545, + "loss": 0.3999, + "step": 6549 + }, + { + "epoch": 0.3659524541162667, + "grad_norm": 0.7431975603103638, + "learning_rate": 0.0008196716718960107, + "loss": 0.4582, + "step": 6550 + }, + { + "epoch": 0.3660083247199486, + "grad_norm": 0.41719651222229004, + "learning_rate": 0.000819643657552667, + "loss": 0.3655, + "step": 6551 + }, + { + "epoch": 0.3660641953236305, + "grad_norm": 1.0047013759613037, + "learning_rate": 0.0008196156432093231, + "loss": 0.4382, + "step": 6552 + }, + { + "epoch": 0.36612006592731233, + "grad_norm": 0.7039257287979126, + "learning_rate": 0.0008195876288659793, + "loss": 0.5797, + "step": 6553 + }, + { + "epoch": 0.36617593653099423, + "grad_norm": 0.5657756328582764, + "learning_rate": 0.0008195596145226357, + "loss": 0.4773, + "step": 6554 + }, + { + "epoch": 0.36623180713467607, + "grad_norm": 0.5879949331283569, + "learning_rate": 0.0008195316001792919, + "loss": 0.5601, + "step": 6555 + }, + { + "epoch": 0.36628767773835796, + "grad_norm": 1.4781841039657593, + "learning_rate": 0.000819503585835948, + "loss": 0.6907, + "step": 6556 + }, + { + "epoch": 0.36634354834203986, + "grad_norm": 0.6795819997787476, + "learning_rate": 0.0008194755714926042, + "loss": 0.5281, + "step": 6557 + }, + { + "epoch": 0.3663994189457217, + "grad_norm": 0.6521258354187012, + "learning_rate": 0.0008194475571492605, + "loss": 0.4421, + "step": 6558 + }, + { + "epoch": 0.3664552895494036, + "grad_norm": 0.6119281053543091, + "learning_rate": 0.0008194195428059167, + "loss": 0.465, + "step": 6559 + }, + { + "epoch": 0.36651116015308544, + "grad_norm": 0.9667111039161682, + "learning_rate": 0.0008193915284625728, + "loss": 0.5282, + "step": 6560 + }, + { + "epoch": 0.36656703075676733, + "grad_norm": 0.6326295137405396, + "learning_rate": 0.000819363514119229, + "loss": 0.4302, + "step": 6561 + }, + { + "epoch": 0.3666229013604492, + "grad_norm": 1.6582541465759277, + "learning_rate": 0.0008193354997758853, + "loss": 0.8966, + "step": 6562 + }, + { + "epoch": 0.36667877196413107, + "grad_norm": 0.7928370237350464, + "learning_rate": 0.0008193074854325415, + "loss": 0.4549, + "step": 6563 + }, + { + "epoch": 0.36673464256781296, + "grad_norm": 0.6433700919151306, + "learning_rate": 0.0008192794710891976, + "loss": 0.4035, + "step": 6564 + }, + { + "epoch": 0.3667905131714948, + "grad_norm": 0.6846978068351746, + "learning_rate": 0.0008192514567458539, + "loss": 0.4197, + "step": 6565 + }, + { + "epoch": 0.3668463837751767, + "grad_norm": 0.8632194995880127, + "learning_rate": 0.0008192234424025101, + "loss": 0.5638, + "step": 6566 + }, + { + "epoch": 0.36690225437885854, + "grad_norm": 0.8496072292327881, + "learning_rate": 0.0008191954280591663, + "loss": 0.4352, + "step": 6567 + }, + { + "epoch": 0.36695812498254043, + "grad_norm": 0.5769191980361938, + "learning_rate": 0.0008191674137158224, + "loss": 0.4167, + "step": 6568 + }, + { + "epoch": 0.36701399558622233, + "grad_norm": 0.5202886462211609, + "learning_rate": 0.0008191393993724788, + "loss": 0.4632, + "step": 6569 + }, + { + "epoch": 0.36706986618990417, + "grad_norm": 0.5157213807106018, + "learning_rate": 0.000819111385029135, + "loss": 0.4914, + "step": 6570 + }, + { + "epoch": 0.36712573679358607, + "grad_norm": 0.5006386637687683, + "learning_rate": 0.0008190833706857912, + "loss": 0.4587, + "step": 6571 + }, + { + "epoch": 0.3671816073972679, + "grad_norm": 0.9026569128036499, + "learning_rate": 0.0008190553563424474, + "loss": 0.6271, + "step": 6572 + }, + { + "epoch": 0.3672374780009498, + "grad_norm": 1.0592459440231323, + "learning_rate": 0.0008190273419991036, + "loss": 0.4823, + "step": 6573 + }, + { + "epoch": 0.3672933486046317, + "grad_norm": 1.1585725545883179, + "learning_rate": 0.0008189993276557598, + "loss": 0.4397, + "step": 6574 + }, + { + "epoch": 0.36734921920831354, + "grad_norm": 0.5333094000816345, + "learning_rate": 0.0008189713133124159, + "loss": 0.4515, + "step": 6575 + }, + { + "epoch": 0.36740508981199543, + "grad_norm": 0.62296062707901, + "learning_rate": 0.0008189432989690722, + "loss": 0.5199, + "step": 6576 + }, + { + "epoch": 0.3674609604156773, + "grad_norm": 0.5213127732276917, + "learning_rate": 0.0008189152846257284, + "loss": 0.413, + "step": 6577 + }, + { + "epoch": 0.36751683101935917, + "grad_norm": 0.4567759335041046, + "learning_rate": 0.0008188872702823846, + "loss": 0.422, + "step": 6578 + }, + { + "epoch": 0.367572701623041, + "grad_norm": 0.7984210848808289, + "learning_rate": 0.0008188592559390408, + "loss": 0.3932, + "step": 6579 + }, + { + "epoch": 0.3676285722267229, + "grad_norm": 0.6783561706542969, + "learning_rate": 0.000818831241595697, + "loss": 0.417, + "step": 6580 + }, + { + "epoch": 0.3676844428304048, + "grad_norm": 1.4970120191574097, + "learning_rate": 0.0008188032272523532, + "loss": 0.407, + "step": 6581 + }, + { + "epoch": 0.36774031343408664, + "grad_norm": 0.6435659527778625, + "learning_rate": 0.0008187752129090094, + "loss": 0.5526, + "step": 6582 + }, + { + "epoch": 0.36779618403776854, + "grad_norm": 0.5356265306472778, + "learning_rate": 0.0008187471985656656, + "loss": 0.4156, + "step": 6583 + }, + { + "epoch": 0.3678520546414504, + "grad_norm": 0.48186591267585754, + "learning_rate": 0.0008187191842223218, + "loss": 0.4188, + "step": 6584 + }, + { + "epoch": 0.3679079252451323, + "grad_norm": 0.5500417351722717, + "learning_rate": 0.000818691169878978, + "loss": 0.4512, + "step": 6585 + }, + { + "epoch": 0.36796379584881417, + "grad_norm": 0.4727037847042084, + "learning_rate": 0.0008186631555356344, + "loss": 0.6774, + "step": 6586 + }, + { + "epoch": 0.368019666452496, + "grad_norm": 0.6300316452980042, + "learning_rate": 0.0008186351411922905, + "loss": 0.6072, + "step": 6587 + }, + { + "epoch": 0.3680755370561779, + "grad_norm": 0.5410597920417786, + "learning_rate": 0.0008186071268489467, + "loss": 0.4599, + "step": 6588 + }, + { + "epoch": 0.36813140765985974, + "grad_norm": 0.48540642857551575, + "learning_rate": 0.0008185791125056029, + "loss": 0.4791, + "step": 6589 + }, + { + "epoch": 0.36818727826354164, + "grad_norm": 0.507685661315918, + "learning_rate": 0.0008185510981622592, + "loss": 0.4827, + "step": 6590 + }, + { + "epoch": 0.36824314886722354, + "grad_norm": 0.563389241695404, + "learning_rate": 0.0008185230838189153, + "loss": 0.4942, + "step": 6591 + }, + { + "epoch": 0.3682990194709054, + "grad_norm": 0.6008557677268982, + "learning_rate": 0.0008184950694755715, + "loss": 0.5134, + "step": 6592 + }, + { + "epoch": 0.36835489007458727, + "grad_norm": 0.7914635539054871, + "learning_rate": 0.0008184670551322278, + "loss": 0.5167, + "step": 6593 + }, + { + "epoch": 0.3684107606782691, + "grad_norm": 0.5750457644462585, + "learning_rate": 0.0008184390407888839, + "loss": 0.4263, + "step": 6594 + }, + { + "epoch": 0.368466631281951, + "grad_norm": 0.8050326108932495, + "learning_rate": 0.0008184110264455401, + "loss": 0.4402, + "step": 6595 + }, + { + "epoch": 0.36852250188563285, + "grad_norm": 0.6835136413574219, + "learning_rate": 0.0008183830121021963, + "loss": 0.3698, + "step": 6596 + }, + { + "epoch": 0.36857837248931474, + "grad_norm": 0.41056814789772034, + "learning_rate": 0.0008183549977588526, + "loss": 0.4864, + "step": 6597 + }, + { + "epoch": 0.36863424309299664, + "grad_norm": 0.46239927411079407, + "learning_rate": 0.0008183269834155087, + "loss": 0.4712, + "step": 6598 + }, + { + "epoch": 0.3686901136966785, + "grad_norm": 1.6798182725906372, + "learning_rate": 0.0008182989690721649, + "loss": 0.5755, + "step": 6599 + }, + { + "epoch": 0.3687459843003604, + "grad_norm": 0.44019970297813416, + "learning_rate": 0.0008182709547288211, + "loss": 0.3838, + "step": 6600 + }, + { + "epoch": 0.3688018549040422, + "grad_norm": 0.6524181962013245, + "learning_rate": 0.0008182429403854774, + "loss": 0.4661, + "step": 6601 + }, + { + "epoch": 0.3688577255077241, + "grad_norm": 0.47720056772232056, + "learning_rate": 0.0008182149260421335, + "loss": 0.4397, + "step": 6602 + }, + { + "epoch": 0.368913596111406, + "grad_norm": 0.8136889934539795, + "learning_rate": 0.0008181869116987897, + "loss": 0.4254, + "step": 6603 + }, + { + "epoch": 0.36896946671508785, + "grad_norm": 0.5047023892402649, + "learning_rate": 0.0008181588973554461, + "loss": 0.5215, + "step": 6604 + }, + { + "epoch": 0.36902533731876974, + "grad_norm": 0.4541560709476471, + "learning_rate": 0.0008181308830121023, + "loss": 0.5751, + "step": 6605 + }, + { + "epoch": 0.3690812079224516, + "grad_norm": 0.5690813660621643, + "learning_rate": 0.0008181028686687584, + "loss": 0.4466, + "step": 6606 + }, + { + "epoch": 0.3691370785261335, + "grad_norm": 0.5140795111656189, + "learning_rate": 0.0008180748543254146, + "loss": 0.4683, + "step": 6607 + }, + { + "epoch": 0.3691929491298154, + "grad_norm": 10.096177101135254, + "learning_rate": 0.0008180468399820709, + "loss": 0.3398, + "step": 6608 + }, + { + "epoch": 0.3692488197334972, + "grad_norm": 0.5055945515632629, + "learning_rate": 0.0008180188256387271, + "loss": 0.4279, + "step": 6609 + }, + { + "epoch": 0.3693046903371791, + "grad_norm": 0.49428024888038635, + "learning_rate": 0.0008179908112953832, + "loss": 0.4612, + "step": 6610 + }, + { + "epoch": 0.36936056094086095, + "grad_norm": 0.5755905508995056, + "learning_rate": 0.0008179627969520395, + "loss": 0.5348, + "step": 6611 + }, + { + "epoch": 0.36941643154454284, + "grad_norm": 0.4374230206012726, + "learning_rate": 0.0008179347826086957, + "loss": 0.3987, + "step": 6612 + }, + { + "epoch": 0.3694723021482247, + "grad_norm": 3.0422019958496094, + "learning_rate": 0.0008179067682653519, + "loss": 0.4948, + "step": 6613 + }, + { + "epoch": 0.3695281727519066, + "grad_norm": 0.5649192333221436, + "learning_rate": 0.000817878753922008, + "loss": 0.5873, + "step": 6614 + }, + { + "epoch": 0.3695840433555885, + "grad_norm": 0.5922582149505615, + "learning_rate": 0.0008178507395786643, + "loss": 0.409, + "step": 6615 + }, + { + "epoch": 0.3696399139592703, + "grad_norm": 0.5909778475761414, + "learning_rate": 0.0008178227252353205, + "loss": 0.5138, + "step": 6616 + }, + { + "epoch": 0.3696957845629522, + "grad_norm": 0.6824950575828552, + "learning_rate": 0.0008177947108919766, + "loss": 0.496, + "step": 6617 + }, + { + "epoch": 0.36975165516663405, + "grad_norm": 0.48061642050743103, + "learning_rate": 0.0008177666965486329, + "loss": 0.3741, + "step": 6618 + }, + { + "epoch": 0.36980752577031595, + "grad_norm": 0.42387300729751587, + "learning_rate": 0.0008177386822052891, + "loss": 0.4279, + "step": 6619 + }, + { + "epoch": 0.36986339637399784, + "grad_norm": 0.7366540431976318, + "learning_rate": 0.0008177106678619454, + "loss": 0.5335, + "step": 6620 + }, + { + "epoch": 0.3699192669776797, + "grad_norm": 0.5195976495742798, + "learning_rate": 0.0008176826535186015, + "loss": 0.4801, + "step": 6621 + }, + { + "epoch": 0.3699751375813616, + "grad_norm": 0.8107237219810486, + "learning_rate": 0.0008176546391752578, + "loss": 0.3746, + "step": 6622 + }, + { + "epoch": 0.3700310081850434, + "grad_norm": 0.48926037549972534, + "learning_rate": 0.000817626624831914, + "loss": 0.487, + "step": 6623 + }, + { + "epoch": 0.3700868787887253, + "grad_norm": 0.5074799656867981, + "learning_rate": 0.0008175986104885702, + "loss": 0.368, + "step": 6624 + }, + { + "epoch": 0.3701427493924072, + "grad_norm": 1.2977454662322998, + "learning_rate": 0.0008175705961452264, + "loss": 0.5352, + "step": 6625 + }, + { + "epoch": 0.37019861999608905, + "grad_norm": 2.0323469638824463, + "learning_rate": 0.0008175425818018826, + "loss": 0.4, + "step": 6626 + }, + { + "epoch": 0.37025449059977095, + "grad_norm": 0.675969123840332, + "learning_rate": 0.0008175145674585388, + "loss": 0.4302, + "step": 6627 + }, + { + "epoch": 0.3703103612034528, + "grad_norm": 0.44191858172416687, + "learning_rate": 0.000817486553115195, + "loss": 0.4639, + "step": 6628 + }, + { + "epoch": 0.3703662318071347, + "grad_norm": 0.4987929165363312, + "learning_rate": 0.0008174585387718512, + "loss": 0.5339, + "step": 6629 + }, + { + "epoch": 0.3704221024108165, + "grad_norm": 0.5412311553955078, + "learning_rate": 0.0008174305244285074, + "loss": 0.6024, + "step": 6630 + }, + { + "epoch": 0.3704779730144984, + "grad_norm": 0.39180827140808105, + "learning_rate": 0.0008174025100851636, + "loss": 0.3844, + "step": 6631 + }, + { + "epoch": 0.3705338436181803, + "grad_norm": 3.434215545654297, + "learning_rate": 0.0008173744957418198, + "loss": 0.4358, + "step": 6632 + }, + { + "epoch": 0.37058971422186215, + "grad_norm": 0.45864835381507874, + "learning_rate": 0.000817346481398476, + "loss": 0.6341, + "step": 6633 + }, + { + "epoch": 0.37064558482554405, + "grad_norm": 0.5561710000038147, + "learning_rate": 0.0008173184670551322, + "loss": 0.4551, + "step": 6634 + }, + { + "epoch": 0.3707014554292259, + "grad_norm": 1.4298810958862305, + "learning_rate": 0.0008172904527117884, + "loss": 0.4072, + "step": 6635 + }, + { + "epoch": 0.3707573260329078, + "grad_norm": 0.45250171422958374, + "learning_rate": 0.0008172624383684448, + "loss": 0.4438, + "step": 6636 + }, + { + "epoch": 0.3708131966365897, + "grad_norm": 0.4745189845561981, + "learning_rate": 0.0008172344240251009, + "loss": 0.3824, + "step": 6637 + }, + { + "epoch": 0.3708690672402715, + "grad_norm": 0.4185824692249298, + "learning_rate": 0.0008172064096817571, + "loss": 0.5359, + "step": 6638 + }, + { + "epoch": 0.3709249378439534, + "grad_norm": 0.7233831882476807, + "learning_rate": 0.0008171783953384133, + "loss": 0.4611, + "step": 6639 + }, + { + "epoch": 0.37098080844763526, + "grad_norm": 2.905626058578491, + "learning_rate": 0.0008171503809950695, + "loss": 0.3587, + "step": 6640 + }, + { + "epoch": 0.37103667905131715, + "grad_norm": 0.4926472008228302, + "learning_rate": 0.0008171223666517257, + "loss": 0.5061, + "step": 6641 + }, + { + "epoch": 0.37109254965499905, + "grad_norm": 2.0837631225585938, + "learning_rate": 0.0008170943523083819, + "loss": 0.515, + "step": 6642 + }, + { + "epoch": 0.3711484202586809, + "grad_norm": 0.48780909180641174, + "learning_rate": 0.0008170663379650382, + "loss": 0.4652, + "step": 6643 + }, + { + "epoch": 0.3712042908623628, + "grad_norm": 0.608462929725647, + "learning_rate": 0.0008170383236216943, + "loss": 0.5353, + "step": 6644 + }, + { + "epoch": 0.3712601614660446, + "grad_norm": 0.4542344808578491, + "learning_rate": 0.0008170103092783505, + "loss": 0.3793, + "step": 6645 + }, + { + "epoch": 0.3713160320697265, + "grad_norm": 0.4778662323951721, + "learning_rate": 0.0008169822949350067, + "loss": 0.5372, + "step": 6646 + }, + { + "epoch": 0.37137190267340836, + "grad_norm": 0.6397621631622314, + "learning_rate": 0.000816954280591663, + "loss": 0.4398, + "step": 6647 + }, + { + "epoch": 0.37142777327709026, + "grad_norm": 0.5515587329864502, + "learning_rate": 0.0008169262662483191, + "loss": 0.449, + "step": 6648 + }, + { + "epoch": 0.37148364388077215, + "grad_norm": 0.5915719866752625, + "learning_rate": 0.0008168982519049753, + "loss": 0.6203, + "step": 6649 + }, + { + "epoch": 0.371539514484454, + "grad_norm": 0.727460503578186, + "learning_rate": 0.0008168702375616316, + "loss": 0.5193, + "step": 6650 + }, + { + "epoch": 0.3715953850881359, + "grad_norm": 0.5437660813331604, + "learning_rate": 0.0008168422232182878, + "loss": 0.3609, + "step": 6651 + }, + { + "epoch": 0.37165125569181773, + "grad_norm": 6.792923927307129, + "learning_rate": 0.0008168142088749439, + "loss": 0.5522, + "step": 6652 + }, + { + "epoch": 0.3717071262954996, + "grad_norm": 0.4036847949028015, + "learning_rate": 0.0008167861945316001, + "loss": 0.4252, + "step": 6653 + }, + { + "epoch": 0.3717629968991815, + "grad_norm": 0.5406880378723145, + "learning_rate": 0.0008167581801882565, + "loss": 0.432, + "step": 6654 + }, + { + "epoch": 0.37181886750286336, + "grad_norm": 0.7312514781951904, + "learning_rate": 0.0008167301658449127, + "loss": 0.4706, + "step": 6655 + }, + { + "epoch": 0.37187473810654526, + "grad_norm": 0.44737377762794495, + "learning_rate": 0.0008167021515015688, + "loss": 0.4051, + "step": 6656 + }, + { + "epoch": 0.3719306087102271, + "grad_norm": 0.48754268884658813, + "learning_rate": 0.0008166741371582251, + "loss": 0.49, + "step": 6657 + }, + { + "epoch": 0.371986479313909, + "grad_norm": 0.6235092878341675, + "learning_rate": 0.0008166461228148813, + "loss": 0.4161, + "step": 6658 + }, + { + "epoch": 0.3720423499175909, + "grad_norm": 0.724306046962738, + "learning_rate": 0.0008166181084715374, + "loss": 0.406, + "step": 6659 + }, + { + "epoch": 0.3720982205212727, + "grad_norm": 0.4820166528224945, + "learning_rate": 0.0008165900941281936, + "loss": 0.5254, + "step": 6660 + }, + { + "epoch": 0.3721540911249546, + "grad_norm": 1.3503793478012085, + "learning_rate": 0.0008165620797848499, + "loss": 0.5265, + "step": 6661 + }, + { + "epoch": 0.37220996172863646, + "grad_norm": 2.0472350120544434, + "learning_rate": 0.0008165340654415061, + "loss": 0.4827, + "step": 6662 + }, + { + "epoch": 0.37226583233231836, + "grad_norm": 0.746029794216156, + "learning_rate": 0.0008165060510981622, + "loss": 0.3829, + "step": 6663 + }, + { + "epoch": 0.3723217029360002, + "grad_norm": 0.632749617099762, + "learning_rate": 0.0008164780367548184, + "loss": 0.4482, + "step": 6664 + }, + { + "epoch": 0.3723775735396821, + "grad_norm": 0.9057317972183228, + "learning_rate": 0.0008164500224114747, + "loss": 0.5059, + "step": 6665 + }, + { + "epoch": 0.372433444143364, + "grad_norm": 0.5537781119346619, + "learning_rate": 0.0008164220080681309, + "loss": 0.5434, + "step": 6666 + }, + { + "epoch": 0.37248931474704583, + "grad_norm": 5.451834678649902, + "learning_rate": 0.000816393993724787, + "loss": 0.4606, + "step": 6667 + }, + { + "epoch": 0.3725451853507277, + "grad_norm": 0.510977566242218, + "learning_rate": 0.0008163659793814433, + "loss": 0.4021, + "step": 6668 + }, + { + "epoch": 0.37260105595440957, + "grad_norm": 0.6098328828811646, + "learning_rate": 0.0008163379650380995, + "loss": 0.4292, + "step": 6669 + }, + { + "epoch": 0.37265692655809146, + "grad_norm": 0.6375926733016968, + "learning_rate": 0.0008163099506947558, + "loss": 0.4919, + "step": 6670 + }, + { + "epoch": 0.37271279716177336, + "grad_norm": 0.7927713990211487, + "learning_rate": 0.0008162819363514119, + "loss": 0.4876, + "step": 6671 + }, + { + "epoch": 0.3727686677654552, + "grad_norm": 0.8612987995147705, + "learning_rate": 0.0008162539220080682, + "loss": 0.4755, + "step": 6672 + }, + { + "epoch": 0.3728245383691371, + "grad_norm": 0.5559167861938477, + "learning_rate": 0.0008162259076647244, + "loss": 0.5517, + "step": 6673 + }, + { + "epoch": 0.37288040897281893, + "grad_norm": 0.6895058751106262, + "learning_rate": 0.0008161978933213806, + "loss": 0.5407, + "step": 6674 + }, + { + "epoch": 0.37293627957650083, + "grad_norm": 0.555664598941803, + "learning_rate": 0.0008161698789780368, + "loss": 0.4201, + "step": 6675 + }, + { + "epoch": 0.3729921501801827, + "grad_norm": 0.4510287344455719, + "learning_rate": 0.000816141864634693, + "loss": 0.3617, + "step": 6676 + }, + { + "epoch": 0.37304802078386456, + "grad_norm": 0.6655372977256775, + "learning_rate": 0.0008161138502913492, + "loss": 0.5996, + "step": 6677 + }, + { + "epoch": 0.37310389138754646, + "grad_norm": 1.11930251121521, + "learning_rate": 0.0008160858359480054, + "loss": 0.5525, + "step": 6678 + }, + { + "epoch": 0.3731597619912283, + "grad_norm": 0.6488763689994812, + "learning_rate": 0.0008160578216046616, + "loss": 0.5024, + "step": 6679 + }, + { + "epoch": 0.3732156325949102, + "grad_norm": 0.5727670788764954, + "learning_rate": 0.0008160298072613178, + "loss": 0.4159, + "step": 6680 + }, + { + "epoch": 0.37327150319859204, + "grad_norm": 0.6117333769798279, + "learning_rate": 0.000816001792917974, + "loss": 0.39, + "step": 6681 + }, + { + "epoch": 0.37332737380227393, + "grad_norm": 16.299915313720703, + "learning_rate": 0.0008159737785746302, + "loss": 0.6005, + "step": 6682 + }, + { + "epoch": 0.3733832444059558, + "grad_norm": 0.6162392497062683, + "learning_rate": 0.0008159457642312864, + "loss": 0.4992, + "step": 6683 + }, + { + "epoch": 0.37343911500963767, + "grad_norm": 0.9741047024726868, + "learning_rate": 0.0008159177498879426, + "loss": 0.557, + "step": 6684 + }, + { + "epoch": 0.37349498561331956, + "grad_norm": 1.5119662284851074, + "learning_rate": 0.0008158897355445988, + "loss": 0.5048, + "step": 6685 + }, + { + "epoch": 0.3735508562170014, + "grad_norm": 4.538615703582764, + "learning_rate": 0.000815861721201255, + "loss": 0.4824, + "step": 6686 + }, + { + "epoch": 0.3736067268206833, + "grad_norm": 0.5125417709350586, + "learning_rate": 0.0008158337068579113, + "loss": 0.4137, + "step": 6687 + }, + { + "epoch": 0.3736625974243652, + "grad_norm": 2.4957680702209473, + "learning_rate": 0.0008158056925145675, + "loss": 0.4444, + "step": 6688 + }, + { + "epoch": 0.37371846802804704, + "grad_norm": 1.8559585809707642, + "learning_rate": 0.0008157776781712238, + "loss": 0.5441, + "step": 6689 + }, + { + "epoch": 0.37377433863172893, + "grad_norm": 0.6209786534309387, + "learning_rate": 0.0008157496638278799, + "loss": 0.4646, + "step": 6690 + }, + { + "epoch": 0.37383020923541077, + "grad_norm": 1.2355750799179077, + "learning_rate": 0.0008157216494845361, + "loss": 0.3806, + "step": 6691 + }, + { + "epoch": 0.37388607983909267, + "grad_norm": 0.5585333108901978, + "learning_rate": 0.0008156936351411923, + "loss": 0.3975, + "step": 6692 + }, + { + "epoch": 0.3739419504427745, + "grad_norm": 0.8536207675933838, + "learning_rate": 0.0008156656207978486, + "loss": 0.5556, + "step": 6693 + }, + { + "epoch": 0.3739978210464564, + "grad_norm": 0.5239065885543823, + "learning_rate": 0.0008156376064545047, + "loss": 0.4438, + "step": 6694 + }, + { + "epoch": 0.3740536916501383, + "grad_norm": 0.46445462107658386, + "learning_rate": 0.0008156095921111609, + "loss": 0.488, + "step": 6695 + }, + { + "epoch": 0.37410956225382014, + "grad_norm": 0.5497421026229858, + "learning_rate": 0.0008155815777678172, + "loss": 0.5056, + "step": 6696 + }, + { + "epoch": 0.37416543285750203, + "grad_norm": 0.7097726464271545, + "learning_rate": 0.0008155535634244734, + "loss": 0.4238, + "step": 6697 + }, + { + "epoch": 0.3742213034611839, + "grad_norm": 0.5772419571876526, + "learning_rate": 0.0008155255490811295, + "loss": 0.514, + "step": 6698 + }, + { + "epoch": 0.37427717406486577, + "grad_norm": 3.218719244003296, + "learning_rate": 0.0008154975347377857, + "loss": 0.417, + "step": 6699 + }, + { + "epoch": 0.37433304466854767, + "grad_norm": 0.823305070400238, + "learning_rate": 0.000815469520394442, + "loss": 0.4012, + "step": 6700 + }, + { + "epoch": 0.3743889152722295, + "grad_norm": 0.6696406602859497, + "learning_rate": 0.0008154415060510981, + "loss": 0.5918, + "step": 6701 + }, + { + "epoch": 0.3744447858759114, + "grad_norm": 0.8430393934249878, + "learning_rate": 0.0008154134917077543, + "loss": 0.5702, + "step": 6702 + }, + { + "epoch": 0.37450065647959324, + "grad_norm": 0.4987841546535492, + "learning_rate": 0.0008153854773644105, + "loss": 0.3524, + "step": 6703 + }, + { + "epoch": 0.37455652708327514, + "grad_norm": 1.407572627067566, + "learning_rate": 0.0008153574630210669, + "loss": 0.4536, + "step": 6704 + }, + { + "epoch": 0.37461239768695703, + "grad_norm": 0.5236262679100037, + "learning_rate": 0.000815329448677723, + "loss": 0.3644, + "step": 6705 + }, + { + "epoch": 0.3746682682906389, + "grad_norm": 0.5744314193725586, + "learning_rate": 0.0008153014343343792, + "loss": 0.4455, + "step": 6706 + }, + { + "epoch": 0.37472413889432077, + "grad_norm": 0.5219522714614868, + "learning_rate": 0.0008152734199910355, + "loss": 0.4605, + "step": 6707 + }, + { + "epoch": 0.3747800094980026, + "grad_norm": 0.66902756690979, + "learning_rate": 0.0008152454056476917, + "loss": 0.4523, + "step": 6708 + }, + { + "epoch": 0.3748358801016845, + "grad_norm": 0.7072827816009521, + "learning_rate": 0.0008152173913043478, + "loss": 0.3897, + "step": 6709 + }, + { + "epoch": 0.37489175070536634, + "grad_norm": 0.518940806388855, + "learning_rate": 0.000815189376961004, + "loss": 0.3757, + "step": 6710 + }, + { + "epoch": 0.37494762130904824, + "grad_norm": 2.2978696823120117, + "learning_rate": 0.0008151613626176603, + "loss": 0.4416, + "step": 6711 + }, + { + "epoch": 0.37500349191273014, + "grad_norm": 0.9783353209495544, + "learning_rate": 0.0008151333482743165, + "loss": 0.5002, + "step": 6712 + }, + { + "epoch": 0.375059362516412, + "grad_norm": 7.894841194152832, + "learning_rate": 0.0008151053339309726, + "loss": 0.4635, + "step": 6713 + }, + { + "epoch": 0.37511523312009387, + "grad_norm": 0.7463032603263855, + "learning_rate": 0.0008150773195876289, + "loss": 0.4212, + "step": 6714 + }, + { + "epoch": 0.3751711037237757, + "grad_norm": 0.7906734347343445, + "learning_rate": 0.0008150493052442851, + "loss": 0.4834, + "step": 6715 + }, + { + "epoch": 0.3752269743274576, + "grad_norm": 0.566066324710846, + "learning_rate": 0.0008150212909009413, + "loss": 0.4937, + "step": 6716 + }, + { + "epoch": 0.3752828449311395, + "grad_norm": 0.5511184930801392, + "learning_rate": 0.0008149932765575974, + "loss": 0.4881, + "step": 6717 + }, + { + "epoch": 0.37533871553482134, + "grad_norm": 0.580925703048706, + "learning_rate": 0.0008149652622142537, + "loss": 0.682, + "step": 6718 + }, + { + "epoch": 0.37539458613850324, + "grad_norm": 0.6480460166931152, + "learning_rate": 0.00081493724787091, + "loss": 0.466, + "step": 6719 + }, + { + "epoch": 0.3754504567421851, + "grad_norm": 0.7140601873397827, + "learning_rate": 0.0008149092335275662, + "loss": 0.5178, + "step": 6720 + }, + { + "epoch": 0.375506327345867, + "grad_norm": 0.6865890622138977, + "learning_rate": 0.0008148812191842224, + "loss": 0.5569, + "step": 6721 + }, + { + "epoch": 0.37556219794954887, + "grad_norm": 0.8740230798721313, + "learning_rate": 0.0008148532048408786, + "loss": 0.4712, + "step": 6722 + }, + { + "epoch": 0.3756180685532307, + "grad_norm": 1.2320525646209717, + "learning_rate": 0.0008148251904975348, + "loss": 0.4869, + "step": 6723 + }, + { + "epoch": 0.3756739391569126, + "grad_norm": 0.41782450675964355, + "learning_rate": 0.0008147971761541909, + "loss": 0.4369, + "step": 6724 + }, + { + "epoch": 0.37572980976059445, + "grad_norm": 0.43218615651130676, + "learning_rate": 0.0008147691618108472, + "loss": 0.5028, + "step": 6725 + }, + { + "epoch": 0.37578568036427634, + "grad_norm": 0.5429787635803223, + "learning_rate": 0.0008147411474675034, + "loss": 0.4333, + "step": 6726 + }, + { + "epoch": 0.3758415509679582, + "grad_norm": 0.8591233491897583, + "learning_rate": 0.0008147131331241596, + "loss": 0.4437, + "step": 6727 + }, + { + "epoch": 0.3758974215716401, + "grad_norm": 0.9593701362609863, + "learning_rate": 0.0008146851187808158, + "loss": 0.3781, + "step": 6728 + }, + { + "epoch": 0.375953292175322, + "grad_norm": 0.6142887473106384, + "learning_rate": 0.000814657104437472, + "loss": 0.4345, + "step": 6729 + }, + { + "epoch": 0.3760091627790038, + "grad_norm": 0.5329455137252808, + "learning_rate": 0.0008146290900941282, + "loss": 0.4364, + "step": 6730 + }, + { + "epoch": 0.3760650333826857, + "grad_norm": 0.8405459523200989, + "learning_rate": 0.0008146010757507844, + "loss": 0.5235, + "step": 6731 + }, + { + "epoch": 0.37612090398636755, + "grad_norm": 1.4193490743637085, + "learning_rate": 0.0008145730614074406, + "loss": 0.3878, + "step": 6732 + }, + { + "epoch": 0.37617677459004945, + "grad_norm": 0.7616932392120361, + "learning_rate": 0.0008145450470640968, + "loss": 0.4468, + "step": 6733 + }, + { + "epoch": 0.37623264519373134, + "grad_norm": 0.800274670124054, + "learning_rate": 0.000814517032720753, + "loss": 0.4197, + "step": 6734 + }, + { + "epoch": 0.3762885157974132, + "grad_norm": 0.6258297562599182, + "learning_rate": 0.0008144890183774093, + "loss": 0.4053, + "step": 6735 + }, + { + "epoch": 0.3763443864010951, + "grad_norm": 0.7884092330932617, + "learning_rate": 0.0008144610040340654, + "loss": 0.5203, + "step": 6736 + }, + { + "epoch": 0.3764002570047769, + "grad_norm": 0.4380631744861603, + "learning_rate": 0.0008144329896907217, + "loss": 0.4837, + "step": 6737 + }, + { + "epoch": 0.3764561276084588, + "grad_norm": 0.6382566094398499, + "learning_rate": 0.0008144049753473779, + "loss": 0.5215, + "step": 6738 + }, + { + "epoch": 0.3765119982121407, + "grad_norm": 0.6008732318878174, + "learning_rate": 0.0008143769610040342, + "loss": 0.4087, + "step": 6739 + }, + { + "epoch": 0.37656786881582255, + "grad_norm": 1.2510894536972046, + "learning_rate": 0.0008143489466606903, + "loss": 0.5027, + "step": 6740 + }, + { + "epoch": 0.37662373941950444, + "grad_norm": 2.2030272483825684, + "learning_rate": 0.0008143209323173465, + "loss": 0.4582, + "step": 6741 + }, + { + "epoch": 0.3766796100231863, + "grad_norm": 1.2699209451675415, + "learning_rate": 0.0008142929179740027, + "loss": 0.496, + "step": 6742 + }, + { + "epoch": 0.3767354806268682, + "grad_norm": 1.513108253479004, + "learning_rate": 0.0008142649036306589, + "loss": 0.462, + "step": 6743 + }, + { + "epoch": 0.37679135123055, + "grad_norm": 1.55155348777771, + "learning_rate": 0.0008142368892873151, + "loss": 0.5738, + "step": 6744 + }, + { + "epoch": 0.3768472218342319, + "grad_norm": 1.5046916007995605, + "learning_rate": 0.0008142088749439713, + "loss": 0.4258, + "step": 6745 + }, + { + "epoch": 0.3769030924379138, + "grad_norm": 0.7348073720932007, + "learning_rate": 0.0008141808606006276, + "loss": 0.4592, + "step": 6746 + }, + { + "epoch": 0.37695896304159565, + "grad_norm": 1.2662646770477295, + "learning_rate": 0.0008141528462572837, + "loss": 0.6252, + "step": 6747 + }, + { + "epoch": 0.37701483364527755, + "grad_norm": 0.5587474703788757, + "learning_rate": 0.0008141248319139399, + "loss": 0.467, + "step": 6748 + }, + { + "epoch": 0.3770707042489594, + "grad_norm": 0.9602931141853333, + "learning_rate": 0.0008140968175705961, + "loss": 0.4776, + "step": 6749 + }, + { + "epoch": 0.3771265748526413, + "grad_norm": 0.6499449014663696, + "learning_rate": 0.0008140688032272524, + "loss": 0.4897, + "step": 6750 + }, + { + "epoch": 0.3771824454563232, + "grad_norm": 0.5379551649093628, + "learning_rate": 0.0008140407888839085, + "loss": 0.5432, + "step": 6751 + }, + { + "epoch": 0.377238316060005, + "grad_norm": 0.6730096936225891, + "learning_rate": 0.0008140127745405647, + "loss": 0.5219, + "step": 6752 + }, + { + "epoch": 0.3772941866636869, + "grad_norm": 0.8617908358573914, + "learning_rate": 0.0008139847601972211, + "loss": 0.461, + "step": 6753 + }, + { + "epoch": 0.37735005726736875, + "grad_norm": 0.6244693994522095, + "learning_rate": 0.0008139567458538773, + "loss": 0.4269, + "step": 6754 + }, + { + "epoch": 0.37740592787105065, + "grad_norm": 0.5209137201309204, + "learning_rate": 0.0008139287315105334, + "loss": 0.4854, + "step": 6755 + }, + { + "epoch": 0.37746179847473255, + "grad_norm": 0.6868574619293213, + "learning_rate": 0.0008139007171671896, + "loss": 0.4432, + "step": 6756 + }, + { + "epoch": 0.3775176690784144, + "grad_norm": 0.7285168170928955, + "learning_rate": 0.0008138727028238459, + "loss": 0.5604, + "step": 6757 + }, + { + "epoch": 0.3775735396820963, + "grad_norm": 0.45518067479133606, + "learning_rate": 0.0008138446884805021, + "loss": 0.4339, + "step": 6758 + }, + { + "epoch": 0.3776294102857781, + "grad_norm": 0.5928915739059448, + "learning_rate": 0.0008138166741371582, + "loss": 0.4177, + "step": 6759 + }, + { + "epoch": 0.37768528088946, + "grad_norm": 0.91423499584198, + "learning_rate": 0.0008137886597938145, + "loss": 0.3719, + "step": 6760 + }, + { + "epoch": 0.37774115149314186, + "grad_norm": 1.2936148643493652, + "learning_rate": 0.0008137606454504707, + "loss": 0.6197, + "step": 6761 + }, + { + "epoch": 0.37779702209682375, + "grad_norm": 0.5434927344322205, + "learning_rate": 0.0008137326311071269, + "loss": 0.498, + "step": 6762 + }, + { + "epoch": 0.37785289270050565, + "grad_norm": 0.5723403692245483, + "learning_rate": 0.000813704616763783, + "loss": 0.5338, + "step": 6763 + }, + { + "epoch": 0.3779087633041875, + "grad_norm": 0.5045139193534851, + "learning_rate": 0.0008136766024204393, + "loss": 0.515, + "step": 6764 + }, + { + "epoch": 0.3779646339078694, + "grad_norm": 1.981924295425415, + "learning_rate": 0.0008136485880770955, + "loss": 0.4414, + "step": 6765 + }, + { + "epoch": 0.3780205045115512, + "grad_norm": 0.6721435189247131, + "learning_rate": 0.0008136205737337516, + "loss": 0.4934, + "step": 6766 + }, + { + "epoch": 0.3780763751152331, + "grad_norm": 0.5695531368255615, + "learning_rate": 0.0008135925593904079, + "loss": 0.4863, + "step": 6767 + }, + { + "epoch": 0.378132245718915, + "grad_norm": 5.861961841583252, + "learning_rate": 0.0008135645450470641, + "loss": 0.4477, + "step": 6768 + }, + { + "epoch": 0.37818811632259686, + "grad_norm": 1.3190937042236328, + "learning_rate": 0.0008135365307037203, + "loss": 0.5166, + "step": 6769 + }, + { + "epoch": 0.37824398692627875, + "grad_norm": 0.7344817519187927, + "learning_rate": 0.0008135085163603764, + "loss": 0.4442, + "step": 6770 + }, + { + "epoch": 0.3782998575299606, + "grad_norm": 0.4467478394508362, + "learning_rate": 0.0008134805020170328, + "loss": 0.4168, + "step": 6771 + }, + { + "epoch": 0.3783557281336425, + "grad_norm": 3.088888168334961, + "learning_rate": 0.000813452487673689, + "loss": 0.4633, + "step": 6772 + }, + { + "epoch": 0.3784115987373244, + "grad_norm": 0.5535491108894348, + "learning_rate": 0.0008134244733303452, + "loss": 0.4177, + "step": 6773 + }, + { + "epoch": 0.3784674693410062, + "grad_norm": 0.5030247569084167, + "learning_rate": 0.0008133964589870013, + "loss": 0.4992, + "step": 6774 + }, + { + "epoch": 0.3785233399446881, + "grad_norm": 0.526638925075531, + "learning_rate": 0.0008133684446436576, + "loss": 0.5429, + "step": 6775 + }, + { + "epoch": 0.37857921054836996, + "grad_norm": 0.6782572865486145, + "learning_rate": 0.0008133404303003138, + "loss": 0.4253, + "step": 6776 + }, + { + "epoch": 0.37863508115205186, + "grad_norm": 0.6577927470207214, + "learning_rate": 0.00081331241595697, + "loss": 0.4131, + "step": 6777 + }, + { + "epoch": 0.3786909517557337, + "grad_norm": 0.7973887920379639, + "learning_rate": 0.0008132844016136262, + "loss": 0.5586, + "step": 6778 + }, + { + "epoch": 0.3787468223594156, + "grad_norm": 0.5729161500930786, + "learning_rate": 0.0008132563872702824, + "loss": 0.4832, + "step": 6779 + }, + { + "epoch": 0.3788026929630975, + "grad_norm": 0.6688548922538757, + "learning_rate": 0.0008132283729269386, + "loss": 0.4425, + "step": 6780 + }, + { + "epoch": 0.3788585635667793, + "grad_norm": 0.6319251656532288, + "learning_rate": 0.0008132003585835948, + "loss": 0.4233, + "step": 6781 + }, + { + "epoch": 0.3789144341704612, + "grad_norm": 1.6267236471176147, + "learning_rate": 0.000813172344240251, + "loss": 0.6121, + "step": 6782 + }, + { + "epoch": 0.37897030477414306, + "grad_norm": 0.8968880772590637, + "learning_rate": 0.0008131443298969072, + "loss": 0.5198, + "step": 6783 + }, + { + "epoch": 0.37902617537782496, + "grad_norm": 0.5867975950241089, + "learning_rate": 0.0008131163155535634, + "loss": 0.545, + "step": 6784 + }, + { + "epoch": 0.37908204598150685, + "grad_norm": 0.5366533398628235, + "learning_rate": 0.0008130883012102196, + "loss": 0.5198, + "step": 6785 + }, + { + "epoch": 0.3791379165851887, + "grad_norm": 1.6668174266815186, + "learning_rate": 0.0008130602868668758, + "loss": 0.362, + "step": 6786 + }, + { + "epoch": 0.3791937871888706, + "grad_norm": 0.9529000520706177, + "learning_rate": 0.000813032272523532, + "loss": 0.5178, + "step": 6787 + }, + { + "epoch": 0.37924965779255243, + "grad_norm": 0.4647759199142456, + "learning_rate": 0.0008130042581801883, + "loss": 0.4358, + "step": 6788 + }, + { + "epoch": 0.3793055283962343, + "grad_norm": 0.5090110898017883, + "learning_rate": 0.0008129762438368445, + "loss": 0.4441, + "step": 6789 + }, + { + "epoch": 0.3793613989999162, + "grad_norm": 0.5877071619033813, + "learning_rate": 0.0008129482294935007, + "loss": 0.5013, + "step": 6790 + }, + { + "epoch": 0.37941726960359806, + "grad_norm": 0.47896671295166016, + "learning_rate": 0.0008129202151501569, + "loss": 0.5281, + "step": 6791 + }, + { + "epoch": 0.37947314020727996, + "grad_norm": 0.501998245716095, + "learning_rate": 0.0008128922008068132, + "loss": 0.4628, + "step": 6792 + }, + { + "epoch": 0.3795290108109618, + "grad_norm": 0.854951024055481, + "learning_rate": 0.0008128641864634693, + "loss": 0.4016, + "step": 6793 + }, + { + "epoch": 0.3795848814146437, + "grad_norm": 0.5600664615631104, + "learning_rate": 0.0008128361721201255, + "loss": 0.699, + "step": 6794 + }, + { + "epoch": 0.37964075201832553, + "grad_norm": 7.98808479309082, + "learning_rate": 0.0008128081577767817, + "loss": 0.4233, + "step": 6795 + }, + { + "epoch": 0.37969662262200743, + "grad_norm": 0.44696804881095886, + "learning_rate": 0.000812780143433438, + "loss": 0.4659, + "step": 6796 + }, + { + "epoch": 0.3797524932256893, + "grad_norm": 1.3145151138305664, + "learning_rate": 0.0008127521290900941, + "loss": 0.5132, + "step": 6797 + }, + { + "epoch": 0.37980836382937117, + "grad_norm": 0.7199507355690002, + "learning_rate": 0.0008127241147467503, + "loss": 0.5379, + "step": 6798 + }, + { + "epoch": 0.37986423443305306, + "grad_norm": 0.7711859941482544, + "learning_rate": 0.0008126961004034066, + "loss": 0.451, + "step": 6799 + }, + { + "epoch": 0.3799201050367349, + "grad_norm": 0.41747036576271057, + "learning_rate": 0.0008126680860600628, + "loss": 0.512, + "step": 6800 + }, + { + "epoch": 0.3799759756404168, + "grad_norm": 1.533854365348816, + "learning_rate": 0.0008126400717167189, + "loss": 0.4558, + "step": 6801 + }, + { + "epoch": 0.3800318462440987, + "grad_norm": 0.5508436560630798, + "learning_rate": 0.0008126120573733751, + "loss": 0.7158, + "step": 6802 + }, + { + "epoch": 0.38008771684778053, + "grad_norm": 0.5729900598526001, + "learning_rate": 0.0008125840430300315, + "loss": 0.445, + "step": 6803 + }, + { + "epoch": 0.38014358745146243, + "grad_norm": 1.444897174835205, + "learning_rate": 0.0008125560286866877, + "loss": 0.3727, + "step": 6804 + }, + { + "epoch": 0.38019945805514427, + "grad_norm": 0.6523869037628174, + "learning_rate": 0.0008125280143433438, + "loss": 0.5981, + "step": 6805 + }, + { + "epoch": 0.38025532865882616, + "grad_norm": 0.552104651927948, + "learning_rate": 0.0008125000000000001, + "loss": 0.4371, + "step": 6806 + }, + { + "epoch": 0.380311199262508, + "grad_norm": 0.8897833824157715, + "learning_rate": 0.0008124719856566563, + "loss": 0.4719, + "step": 6807 + }, + { + "epoch": 0.3803670698661899, + "grad_norm": 0.8517739772796631, + "learning_rate": 0.0008124439713133124, + "loss": 0.4673, + "step": 6808 + }, + { + "epoch": 0.3804229404698718, + "grad_norm": 1.3626731634140015, + "learning_rate": 0.0008124159569699686, + "loss": 0.5373, + "step": 6809 + }, + { + "epoch": 0.38047881107355364, + "grad_norm": 0.5320659279823303, + "learning_rate": 0.0008123879426266249, + "loss": 0.3653, + "step": 6810 + }, + { + "epoch": 0.38053468167723553, + "grad_norm": 0.41657355427742004, + "learning_rate": 0.0008123599282832811, + "loss": 0.4532, + "step": 6811 + }, + { + "epoch": 0.38059055228091737, + "grad_norm": 0.696887731552124, + "learning_rate": 0.0008123319139399372, + "loss": 0.4959, + "step": 6812 + }, + { + "epoch": 0.38064642288459927, + "grad_norm": 7.508807182312012, + "learning_rate": 0.0008123038995965934, + "loss": 0.4148, + "step": 6813 + }, + { + "epoch": 0.38070229348828116, + "grad_norm": 0.5664615035057068, + "learning_rate": 0.0008122758852532497, + "loss": 0.5142, + "step": 6814 + }, + { + "epoch": 0.380758164091963, + "grad_norm": 0.4956309199333191, + "learning_rate": 0.0008122478709099059, + "loss": 0.3958, + "step": 6815 + }, + { + "epoch": 0.3808140346956449, + "grad_norm": 0.40488147735595703, + "learning_rate": 0.000812219856566562, + "loss": 0.3959, + "step": 6816 + }, + { + "epoch": 0.38086990529932674, + "grad_norm": 0.554977297782898, + "learning_rate": 0.0008121918422232183, + "loss": 0.4978, + "step": 6817 + }, + { + "epoch": 0.38092577590300863, + "grad_norm": 0.4365348517894745, + "learning_rate": 0.0008121638278798745, + "loss": 0.4336, + "step": 6818 + }, + { + "epoch": 0.38098164650669053, + "grad_norm": 6.319517135620117, + "learning_rate": 0.0008121358135365307, + "loss": 0.5024, + "step": 6819 + }, + { + "epoch": 0.38103751711037237, + "grad_norm": 1.8476691246032715, + "learning_rate": 0.0008121077991931868, + "loss": 0.4693, + "step": 6820 + }, + { + "epoch": 0.38109338771405427, + "grad_norm": 0.38763555884361267, + "learning_rate": 0.0008120797848498432, + "loss": 0.4217, + "step": 6821 + }, + { + "epoch": 0.3811492583177361, + "grad_norm": 0.5711219906806946, + "learning_rate": 0.0008120517705064994, + "loss": 0.4834, + "step": 6822 + }, + { + "epoch": 0.381205128921418, + "grad_norm": 0.4712193012237549, + "learning_rate": 0.0008120237561631556, + "loss": 0.4138, + "step": 6823 + }, + { + "epoch": 0.38126099952509984, + "grad_norm": 3.6711928844451904, + "learning_rate": 0.0008119957418198118, + "loss": 0.4466, + "step": 6824 + }, + { + "epoch": 0.38131687012878174, + "grad_norm": 4.365043640136719, + "learning_rate": 0.000811967727476468, + "loss": 0.6844, + "step": 6825 + }, + { + "epoch": 0.38137274073246363, + "grad_norm": 0.4277501106262207, + "learning_rate": 0.0008119397131331242, + "loss": 0.3705, + "step": 6826 + }, + { + "epoch": 0.3814286113361455, + "grad_norm": 0.47241446375846863, + "learning_rate": 0.0008119116987897804, + "loss": 0.4663, + "step": 6827 + }, + { + "epoch": 0.38148448193982737, + "grad_norm": 0.6501462459564209, + "learning_rate": 0.0008118836844464366, + "loss": 0.4834, + "step": 6828 + }, + { + "epoch": 0.3815403525435092, + "grad_norm": 0.4802459180355072, + "learning_rate": 0.0008118556701030928, + "loss": 0.4972, + "step": 6829 + }, + { + "epoch": 0.3815962231471911, + "grad_norm": 0.7526376247406006, + "learning_rate": 0.000811827655759749, + "loss": 0.7155, + "step": 6830 + }, + { + "epoch": 0.381652093750873, + "grad_norm": 0.4601394534111023, + "learning_rate": 0.0008117996414164052, + "loss": 0.4609, + "step": 6831 + }, + { + "epoch": 0.38170796435455484, + "grad_norm": 0.6650441288948059, + "learning_rate": 0.0008117716270730614, + "loss": 0.4386, + "step": 6832 + }, + { + "epoch": 0.38176383495823674, + "grad_norm": 0.4359271228313446, + "learning_rate": 0.0008117436127297176, + "loss": 0.4118, + "step": 6833 + }, + { + "epoch": 0.3818197055619186, + "grad_norm": 0.8490115404129028, + "learning_rate": 0.0008117155983863738, + "loss": 0.4725, + "step": 6834 + }, + { + "epoch": 0.38187557616560047, + "grad_norm": 0.6521044969558716, + "learning_rate": 0.00081168758404303, + "loss": 0.481, + "step": 6835 + }, + { + "epoch": 0.38193144676928237, + "grad_norm": 0.7323238849639893, + "learning_rate": 0.0008116595696996862, + "loss": 0.5588, + "step": 6836 + }, + { + "epoch": 0.3819873173729642, + "grad_norm": 0.4774148464202881, + "learning_rate": 0.0008116315553563425, + "loss": 0.4912, + "step": 6837 + }, + { + "epoch": 0.3820431879766461, + "grad_norm": 0.6009368300437927, + "learning_rate": 0.0008116035410129988, + "loss": 0.508, + "step": 6838 + }, + { + "epoch": 0.38209905858032794, + "grad_norm": 0.6363826990127563, + "learning_rate": 0.0008115755266696549, + "loss": 0.4602, + "step": 6839 + }, + { + "epoch": 0.38215492918400984, + "grad_norm": 0.8412860035896301, + "learning_rate": 0.0008115475123263111, + "loss": 0.435, + "step": 6840 + }, + { + "epoch": 0.3822107997876917, + "grad_norm": 0.6601228713989258, + "learning_rate": 0.0008115194979829673, + "loss": 0.4847, + "step": 6841 + }, + { + "epoch": 0.3822666703913736, + "grad_norm": 0.473509818315506, + "learning_rate": 0.0008114914836396236, + "loss": 0.5269, + "step": 6842 + }, + { + "epoch": 0.38232254099505547, + "grad_norm": 1.3382776975631714, + "learning_rate": 0.0008114634692962797, + "loss": 0.4272, + "step": 6843 + }, + { + "epoch": 0.3823784115987373, + "grad_norm": 0.4712091088294983, + "learning_rate": 0.0008114354549529359, + "loss": 0.4691, + "step": 6844 + }, + { + "epoch": 0.3824342822024192, + "grad_norm": 1.7667227983474731, + "learning_rate": 0.0008114074406095921, + "loss": 0.5184, + "step": 6845 + }, + { + "epoch": 0.38249015280610105, + "grad_norm": 0.44897449016571045, + "learning_rate": 0.0008113794262662484, + "loss": 0.409, + "step": 6846 + }, + { + "epoch": 0.38254602340978294, + "grad_norm": 0.697013258934021, + "learning_rate": 0.0008113514119229045, + "loss": 0.4865, + "step": 6847 + }, + { + "epoch": 0.38260189401346484, + "grad_norm": 0.5984876751899719, + "learning_rate": 0.0008113233975795607, + "loss": 0.4374, + "step": 6848 + }, + { + "epoch": 0.3826577646171467, + "grad_norm": 0.41162756085395813, + "learning_rate": 0.000811295383236217, + "loss": 0.4449, + "step": 6849 + }, + { + "epoch": 0.3827136352208286, + "grad_norm": 0.4246178865432739, + "learning_rate": 0.0008112673688928731, + "loss": 0.4585, + "step": 6850 + }, + { + "epoch": 0.3827695058245104, + "grad_norm": 0.385484904050827, + "learning_rate": 0.0008112393545495293, + "loss": 0.3983, + "step": 6851 + }, + { + "epoch": 0.3828253764281923, + "grad_norm": 0.597907304763794, + "learning_rate": 0.0008112113402061855, + "loss": 0.4035, + "step": 6852 + }, + { + "epoch": 0.3828812470318742, + "grad_norm": 0.6300098299980164, + "learning_rate": 0.0008111833258628419, + "loss": 0.4814, + "step": 6853 + }, + { + "epoch": 0.38293711763555605, + "grad_norm": 0.8925835490226746, + "learning_rate": 0.000811155311519498, + "loss": 0.4565, + "step": 6854 + }, + { + "epoch": 0.38299298823923794, + "grad_norm": 0.6614537835121155, + "learning_rate": 0.0008111272971761542, + "loss": 0.4607, + "step": 6855 + }, + { + "epoch": 0.3830488588429198, + "grad_norm": 0.9769797325134277, + "learning_rate": 0.0008110992828328105, + "loss": 0.447, + "step": 6856 + }, + { + "epoch": 0.3831047294466017, + "grad_norm": 0.4133673906326294, + "learning_rate": 0.0008110712684894667, + "loss": 0.3575, + "step": 6857 + }, + { + "epoch": 0.3831606000502835, + "grad_norm": 0.5448353290557861, + "learning_rate": 0.0008110432541461228, + "loss": 0.5296, + "step": 6858 + }, + { + "epoch": 0.3832164706539654, + "grad_norm": 1.6537526845932007, + "learning_rate": 0.000811015239802779, + "loss": 0.5359, + "step": 6859 + }, + { + "epoch": 0.3832723412576473, + "grad_norm": 1.0170962810516357, + "learning_rate": 0.0008109872254594353, + "loss": 0.5502, + "step": 6860 + }, + { + "epoch": 0.38332821186132915, + "grad_norm": 1.2182153463363647, + "learning_rate": 0.0008109592111160915, + "loss": 0.4304, + "step": 6861 + }, + { + "epoch": 0.38338408246501104, + "grad_norm": 0.49997401237487793, + "learning_rate": 0.0008109311967727476, + "loss": 0.536, + "step": 6862 + }, + { + "epoch": 0.3834399530686929, + "grad_norm": 0.5063858032226562, + "learning_rate": 0.0008109031824294039, + "loss": 0.5006, + "step": 6863 + }, + { + "epoch": 0.3834958236723748, + "grad_norm": 0.5820187330245972, + "learning_rate": 0.0008108751680860601, + "loss": 0.4872, + "step": 6864 + }, + { + "epoch": 0.3835516942760567, + "grad_norm": 2.4231956005096436, + "learning_rate": 0.0008108471537427163, + "loss": 0.481, + "step": 6865 + }, + { + "epoch": 0.3836075648797385, + "grad_norm": 0.7936228513717651, + "learning_rate": 0.0008108191393993724, + "loss": 0.4646, + "step": 6866 + }, + { + "epoch": 0.3836634354834204, + "grad_norm": 0.4091194272041321, + "learning_rate": 0.0008107911250560287, + "loss": 0.407, + "step": 6867 + }, + { + "epoch": 0.38371930608710225, + "grad_norm": 0.5841250419616699, + "learning_rate": 0.0008107631107126849, + "loss": 0.4762, + "step": 6868 + }, + { + "epoch": 0.38377517669078415, + "grad_norm": 1.4326436519622803, + "learning_rate": 0.0008107350963693411, + "loss": 0.4141, + "step": 6869 + }, + { + "epoch": 0.38383104729446604, + "grad_norm": 0.6079221963882446, + "learning_rate": 0.0008107070820259974, + "loss": 0.4641, + "step": 6870 + }, + { + "epoch": 0.3838869178981479, + "grad_norm": 0.9899337887763977, + "learning_rate": 0.0008106790676826536, + "loss": 0.6459, + "step": 6871 + }, + { + "epoch": 0.3839427885018298, + "grad_norm": 0.5905061960220337, + "learning_rate": 0.0008106510533393098, + "loss": 0.5401, + "step": 6872 + }, + { + "epoch": 0.3839986591055116, + "grad_norm": 0.4777269959449768, + "learning_rate": 0.0008106230389959659, + "loss": 0.4945, + "step": 6873 + }, + { + "epoch": 0.3840545297091935, + "grad_norm": 1.1134257316589355, + "learning_rate": 0.0008105950246526222, + "loss": 0.4565, + "step": 6874 + }, + { + "epoch": 0.38411040031287536, + "grad_norm": 0.5712931752204895, + "learning_rate": 0.0008105670103092784, + "loss": 0.5282, + "step": 6875 + }, + { + "epoch": 0.38416627091655725, + "grad_norm": 0.6395244598388672, + "learning_rate": 0.0008105389959659346, + "loss": 0.3837, + "step": 6876 + }, + { + "epoch": 0.38422214152023915, + "grad_norm": 0.6105409264564514, + "learning_rate": 0.0008105109816225907, + "loss": 0.4544, + "step": 6877 + }, + { + "epoch": 0.384278012123921, + "grad_norm": 2.024832010269165, + "learning_rate": 0.000810482967279247, + "loss": 0.4911, + "step": 6878 + }, + { + "epoch": 0.3843338827276029, + "grad_norm": 0.450237900018692, + "learning_rate": 0.0008104549529359032, + "loss": 0.5127, + "step": 6879 + }, + { + "epoch": 0.3843897533312847, + "grad_norm": 0.40020209550857544, + "learning_rate": 0.0008104269385925594, + "loss": 0.3379, + "step": 6880 + }, + { + "epoch": 0.3844456239349666, + "grad_norm": 0.8318966627120972, + "learning_rate": 0.0008103989242492156, + "loss": 0.4489, + "step": 6881 + }, + { + "epoch": 0.3845014945386485, + "grad_norm": 0.9016031622886658, + "learning_rate": 0.0008103709099058718, + "loss": 0.6324, + "step": 6882 + }, + { + "epoch": 0.38455736514233035, + "grad_norm": 0.36350488662719727, + "learning_rate": 0.000810342895562528, + "loss": 0.4449, + "step": 6883 + }, + { + "epoch": 0.38461323574601225, + "grad_norm": 17.038930892944336, + "learning_rate": 0.0008103148812191842, + "loss": 0.4699, + "step": 6884 + }, + { + "epoch": 0.3846691063496941, + "grad_norm": 0.5393266677856445, + "learning_rate": 0.0008102868668758404, + "loss": 0.382, + "step": 6885 + }, + { + "epoch": 0.384724976953376, + "grad_norm": 0.7417504191398621, + "learning_rate": 0.0008102588525324966, + "loss": 0.3556, + "step": 6886 + }, + { + "epoch": 0.3847808475570579, + "grad_norm": 0.6705668568611145, + "learning_rate": 0.0008102308381891529, + "loss": 0.4431, + "step": 6887 + }, + { + "epoch": 0.3848367181607397, + "grad_norm": 0.5073100924491882, + "learning_rate": 0.0008102028238458092, + "loss": 0.4306, + "step": 6888 + }, + { + "epoch": 0.3848925887644216, + "grad_norm": 0.506768524646759, + "learning_rate": 0.0008101748095024653, + "loss": 0.5282, + "step": 6889 + }, + { + "epoch": 0.38494845936810346, + "grad_norm": 1.3220324516296387, + "learning_rate": 0.0008101467951591215, + "loss": 0.487, + "step": 6890 + }, + { + "epoch": 0.38500432997178535, + "grad_norm": 0.9555814266204834, + "learning_rate": 0.0008101187808157777, + "loss": 0.3528, + "step": 6891 + }, + { + "epoch": 0.3850602005754672, + "grad_norm": 0.43792903423309326, + "learning_rate": 0.0008100907664724339, + "loss": 0.4265, + "step": 6892 + }, + { + "epoch": 0.3851160711791491, + "grad_norm": 0.5606759190559387, + "learning_rate": 0.0008100627521290901, + "loss": 0.5049, + "step": 6893 + }, + { + "epoch": 0.385171941782831, + "grad_norm": 0.6361874938011169, + "learning_rate": 0.0008100347377857463, + "loss": 0.5097, + "step": 6894 + }, + { + "epoch": 0.3852278123865128, + "grad_norm": 0.5118926167488098, + "learning_rate": 0.0008100067234424026, + "loss": 0.4278, + "step": 6895 + }, + { + "epoch": 0.3852836829901947, + "grad_norm": 0.8233524560928345, + "learning_rate": 0.0008099787090990587, + "loss": 0.6653, + "step": 6896 + }, + { + "epoch": 0.38533955359387656, + "grad_norm": 0.6488253474235535, + "learning_rate": 0.0008099506947557149, + "loss": 0.4201, + "step": 6897 + }, + { + "epoch": 0.38539542419755846, + "grad_norm": 0.7399587035179138, + "learning_rate": 0.0008099226804123711, + "loss": 0.5495, + "step": 6898 + }, + { + "epoch": 0.38545129480124035, + "grad_norm": 0.562983512878418, + "learning_rate": 0.0008098946660690274, + "loss": 0.3595, + "step": 6899 + }, + { + "epoch": 0.3855071654049222, + "grad_norm": 0.45240241289138794, + "learning_rate": 0.0008098666517256835, + "loss": 0.4557, + "step": 6900 + }, + { + "epoch": 0.3855630360086041, + "grad_norm": 0.6608217358589172, + "learning_rate": 0.0008098386373823397, + "loss": 0.5611, + "step": 6901 + }, + { + "epoch": 0.3856189066122859, + "grad_norm": 0.5169005990028381, + "learning_rate": 0.000809810623038996, + "loss": 0.4383, + "step": 6902 + }, + { + "epoch": 0.3856747772159678, + "grad_norm": 2.2305045127868652, + "learning_rate": 0.0008097826086956523, + "loss": 0.3616, + "step": 6903 + }, + { + "epoch": 0.3857306478196497, + "grad_norm": 0.5208249688148499, + "learning_rate": 0.0008097545943523083, + "loss": 0.5798, + "step": 6904 + }, + { + "epoch": 0.38578651842333156, + "grad_norm": 0.5344611406326294, + "learning_rate": 0.0008097265800089646, + "loss": 0.4104, + "step": 6905 + }, + { + "epoch": 0.38584238902701345, + "grad_norm": 0.555136501789093, + "learning_rate": 0.0008096985656656209, + "loss": 0.4216, + "step": 6906 + }, + { + "epoch": 0.3858982596306953, + "grad_norm": 0.6283798217773438, + "learning_rate": 0.0008096705513222771, + "loss": 0.5979, + "step": 6907 + }, + { + "epoch": 0.3859541302343772, + "grad_norm": 1.0212790966033936, + "learning_rate": 0.0008096425369789332, + "loss": 0.4535, + "step": 6908 + }, + { + "epoch": 0.38601000083805903, + "grad_norm": 0.5867560505867004, + "learning_rate": 0.0008096145226355895, + "loss": 0.5193, + "step": 6909 + }, + { + "epoch": 0.3860658714417409, + "grad_norm": 0.4453267455101013, + "learning_rate": 0.0008095865082922457, + "loss": 0.3905, + "step": 6910 + }, + { + "epoch": 0.3861217420454228, + "grad_norm": 0.5043413639068604, + "learning_rate": 0.0008095584939489019, + "loss": 0.496, + "step": 6911 + }, + { + "epoch": 0.38617761264910466, + "grad_norm": 0.49261629581451416, + "learning_rate": 0.000809530479605558, + "loss": 0.4349, + "step": 6912 + }, + { + "epoch": 0.38623348325278656, + "grad_norm": 0.6159219741821289, + "learning_rate": 0.0008095024652622143, + "loss": 0.5082, + "step": 6913 + }, + { + "epoch": 0.3862893538564684, + "grad_norm": 0.6478944420814514, + "learning_rate": 0.0008094744509188705, + "loss": 0.5887, + "step": 6914 + }, + { + "epoch": 0.3863452244601503, + "grad_norm": 0.5218268036842346, + "learning_rate": 0.0008094464365755266, + "loss": 0.5174, + "step": 6915 + }, + { + "epoch": 0.3864010950638322, + "grad_norm": 0.4158783555030823, + "learning_rate": 0.0008094184222321828, + "loss": 0.3884, + "step": 6916 + }, + { + "epoch": 0.38645696566751403, + "grad_norm": 0.8223123550415039, + "learning_rate": 0.0008093904078888391, + "loss": 0.5527, + "step": 6917 + }, + { + "epoch": 0.3865128362711959, + "grad_norm": 0.5168872475624084, + "learning_rate": 0.0008093623935454953, + "loss": 0.4561, + "step": 6918 + }, + { + "epoch": 0.38656870687487777, + "grad_norm": 0.5020731687545776, + "learning_rate": 0.0008093343792021514, + "loss": 0.4761, + "step": 6919 + }, + { + "epoch": 0.38662457747855966, + "grad_norm": 0.5581949353218079, + "learning_rate": 0.0008093063648588078, + "loss": 0.3863, + "step": 6920 + }, + { + "epoch": 0.3866804480822415, + "grad_norm": 0.5894461870193481, + "learning_rate": 0.000809278350515464, + "loss": 0.4232, + "step": 6921 + }, + { + "epoch": 0.3867363186859234, + "grad_norm": 0.7897534966468811, + "learning_rate": 0.0008092503361721202, + "loss": 0.4857, + "step": 6922 + }, + { + "epoch": 0.3867921892896053, + "grad_norm": 0.6698753833770752, + "learning_rate": 0.0008092223218287763, + "loss": 0.3726, + "step": 6923 + }, + { + "epoch": 0.38684805989328713, + "grad_norm": 0.5832765698432922, + "learning_rate": 0.0008091943074854326, + "loss": 0.5258, + "step": 6924 + }, + { + "epoch": 0.38690393049696903, + "grad_norm": 0.40830546617507935, + "learning_rate": 0.0008091662931420888, + "loss": 0.4135, + "step": 6925 + }, + { + "epoch": 0.38695980110065087, + "grad_norm": 1.6452839374542236, + "learning_rate": 0.000809138278798745, + "loss": 0.3918, + "step": 6926 + }, + { + "epoch": 0.38701567170433276, + "grad_norm": 0.930035412311554, + "learning_rate": 0.0008091102644554012, + "loss": 0.3327, + "step": 6927 + }, + { + "epoch": 0.38707154230801466, + "grad_norm": 0.5316284894943237, + "learning_rate": 0.0008090822501120574, + "loss": 0.3937, + "step": 6928 + }, + { + "epoch": 0.3871274129116965, + "grad_norm": 0.5715224742889404, + "learning_rate": 0.0008090542357687136, + "loss": 0.5109, + "step": 6929 + }, + { + "epoch": 0.3871832835153784, + "grad_norm": 0.7072285413742065, + "learning_rate": 0.0008090262214253698, + "loss": 0.4753, + "step": 6930 + }, + { + "epoch": 0.38723915411906024, + "grad_norm": 1.3766072988510132, + "learning_rate": 0.000808998207082026, + "loss": 0.3641, + "step": 6931 + }, + { + "epoch": 0.38729502472274213, + "grad_norm": 0.6340271830558777, + "learning_rate": 0.0008089701927386822, + "loss": 0.445, + "step": 6932 + }, + { + "epoch": 0.387350895326424, + "grad_norm": 0.7594043016433716, + "learning_rate": 0.0008089421783953384, + "loss": 0.6076, + "step": 6933 + }, + { + "epoch": 0.38740676593010587, + "grad_norm": 0.4606471657752991, + "learning_rate": 0.0008089141640519946, + "loss": 0.5636, + "step": 6934 + }, + { + "epoch": 0.38746263653378776, + "grad_norm": 0.7153794169425964, + "learning_rate": 0.0008088861497086508, + "loss": 0.526, + "step": 6935 + }, + { + "epoch": 0.3875185071374696, + "grad_norm": 3.4455480575561523, + "learning_rate": 0.000808858135365307, + "loss": 0.5709, + "step": 6936 + }, + { + "epoch": 0.3875743777411515, + "grad_norm": 0.9114776849746704, + "learning_rate": 0.0008088301210219632, + "loss": 0.3981, + "step": 6937 + }, + { + "epoch": 0.38763024834483334, + "grad_norm": 1.2270538806915283, + "learning_rate": 0.0008088021066786195, + "loss": 0.4393, + "step": 6938 + }, + { + "epoch": 0.38768611894851523, + "grad_norm": 2.589153528213501, + "learning_rate": 0.0008087740923352757, + "loss": 0.4681, + "step": 6939 + }, + { + "epoch": 0.38774198955219713, + "grad_norm": 0.45959147810935974, + "learning_rate": 0.0008087460779919319, + "loss": 0.4352, + "step": 6940 + }, + { + "epoch": 0.38779786015587897, + "grad_norm": 0.8245595097541809, + "learning_rate": 0.0008087180636485882, + "loss": 0.5477, + "step": 6941 + }, + { + "epoch": 0.38785373075956087, + "grad_norm": 3.621330738067627, + "learning_rate": 0.0008086900493052443, + "loss": 0.6936, + "step": 6942 + }, + { + "epoch": 0.3879096013632427, + "grad_norm": 0.6844802498817444, + "learning_rate": 0.0008086620349619005, + "loss": 0.427, + "step": 6943 + }, + { + "epoch": 0.3879654719669246, + "grad_norm": 0.5799052715301514, + "learning_rate": 0.0008086340206185567, + "loss": 0.373, + "step": 6944 + }, + { + "epoch": 0.3880213425706065, + "grad_norm": 0.6799577474594116, + "learning_rate": 0.000808606006275213, + "loss": 0.5079, + "step": 6945 + }, + { + "epoch": 0.38807721317428834, + "grad_norm": 1.3037645816802979, + "learning_rate": 0.0008085779919318691, + "loss": 0.3823, + "step": 6946 + }, + { + "epoch": 0.38813308377797023, + "grad_norm": 0.5306270718574524, + "learning_rate": 0.0008085499775885253, + "loss": 0.4247, + "step": 6947 + }, + { + "epoch": 0.3881889543816521, + "grad_norm": 1.8785653114318848, + "learning_rate": 0.0008085219632451816, + "loss": 0.4292, + "step": 6948 + }, + { + "epoch": 0.38824482498533397, + "grad_norm": 0.6126794219017029, + "learning_rate": 0.0008084939489018378, + "loss": 0.426, + "step": 6949 + }, + { + "epoch": 0.38830069558901587, + "grad_norm": 0.5835358500480652, + "learning_rate": 0.0008084659345584939, + "loss": 0.5673, + "step": 6950 + }, + { + "epoch": 0.3883565661926977, + "grad_norm": 0.4509269595146179, + "learning_rate": 0.0008084379202151501, + "loss": 0.4777, + "step": 6951 + }, + { + "epoch": 0.3884124367963796, + "grad_norm": 0.5442891716957092, + "learning_rate": 0.0008084099058718064, + "loss": 0.4153, + "step": 6952 + }, + { + "epoch": 0.38846830740006144, + "grad_norm": 0.56819087266922, + "learning_rate": 0.0008083818915284627, + "loss": 0.463, + "step": 6953 + }, + { + "epoch": 0.38852417800374334, + "grad_norm": 1.4963833093643188, + "learning_rate": 0.0008083538771851187, + "loss": 0.4505, + "step": 6954 + }, + { + "epoch": 0.3885800486074252, + "grad_norm": 0.6075685620307922, + "learning_rate": 0.000808325862841775, + "loss": 0.4535, + "step": 6955 + }, + { + "epoch": 0.3886359192111071, + "grad_norm": 0.6972708106040955, + "learning_rate": 0.0008082978484984313, + "loss": 0.44, + "step": 6956 + }, + { + "epoch": 0.38869178981478897, + "grad_norm": 0.48850536346435547, + "learning_rate": 0.0008082698341550874, + "loss": 0.5059, + "step": 6957 + }, + { + "epoch": 0.3887476604184708, + "grad_norm": 0.4809320271015167, + "learning_rate": 0.0008082418198117436, + "loss": 0.485, + "step": 6958 + }, + { + "epoch": 0.3888035310221527, + "grad_norm": 1.5307974815368652, + "learning_rate": 0.0008082138054683999, + "loss": 0.5712, + "step": 6959 + }, + { + "epoch": 0.38885940162583454, + "grad_norm": 0.5609871745109558, + "learning_rate": 0.0008081857911250561, + "loss": 0.423, + "step": 6960 + }, + { + "epoch": 0.38891527222951644, + "grad_norm": 0.4751880466938019, + "learning_rate": 0.0008081577767817122, + "loss": 0.4885, + "step": 6961 + }, + { + "epoch": 0.38897114283319834, + "grad_norm": 0.6001437902450562, + "learning_rate": 0.0008081297624383684, + "loss": 0.3226, + "step": 6962 + }, + { + "epoch": 0.3890270134368802, + "grad_norm": 0.7376618385314941, + "learning_rate": 0.0008081017480950247, + "loss": 0.5259, + "step": 6963 + }, + { + "epoch": 0.38908288404056207, + "grad_norm": 0.4292733371257782, + "learning_rate": 0.0008080737337516809, + "loss": 0.511, + "step": 6964 + }, + { + "epoch": 0.3891387546442439, + "grad_norm": 1.71282160282135, + "learning_rate": 0.000808045719408337, + "loss": 0.5407, + "step": 6965 + }, + { + "epoch": 0.3891946252479258, + "grad_norm": 0.4995880126953125, + "learning_rate": 0.0008080177050649933, + "loss": 0.5262, + "step": 6966 + }, + { + "epoch": 0.3892504958516077, + "grad_norm": 1.0387296676635742, + "learning_rate": 0.0008079896907216495, + "loss": 0.3778, + "step": 6967 + }, + { + "epoch": 0.38930636645528954, + "grad_norm": 0.6951598525047302, + "learning_rate": 0.0008079616763783057, + "loss": 0.5809, + "step": 6968 + }, + { + "epoch": 0.38936223705897144, + "grad_norm": 0.5530785918235779, + "learning_rate": 0.0008079336620349618, + "loss": 0.5036, + "step": 6969 + }, + { + "epoch": 0.3894181076626533, + "grad_norm": 1.4241291284561157, + "learning_rate": 0.0008079056476916182, + "loss": 0.4577, + "step": 6970 + }, + { + "epoch": 0.3894739782663352, + "grad_norm": 0.6513903141021729, + "learning_rate": 0.0008078776333482744, + "loss": 0.5147, + "step": 6971 + }, + { + "epoch": 0.389529848870017, + "grad_norm": 0.3735229969024658, + "learning_rate": 0.0008078496190049306, + "loss": 0.3926, + "step": 6972 + }, + { + "epoch": 0.3895857194736989, + "grad_norm": 0.5048012137413025, + "learning_rate": 0.0008078216046615868, + "loss": 0.3682, + "step": 6973 + }, + { + "epoch": 0.3896415900773808, + "grad_norm": 0.6283395886421204, + "learning_rate": 0.000807793590318243, + "loss": 0.4053, + "step": 6974 + }, + { + "epoch": 0.38969746068106265, + "grad_norm": 1.7325875759124756, + "learning_rate": 0.0008077655759748992, + "loss": 0.4298, + "step": 6975 + }, + { + "epoch": 0.38975333128474454, + "grad_norm": 0.47846144437789917, + "learning_rate": 0.0008077375616315553, + "loss": 0.4076, + "step": 6976 + }, + { + "epoch": 0.3898092018884264, + "grad_norm": 1.3808659315109253, + "learning_rate": 0.0008077095472882116, + "loss": 0.6723, + "step": 6977 + }, + { + "epoch": 0.3898650724921083, + "grad_norm": 0.4822274446487427, + "learning_rate": 0.0008076815329448678, + "loss": 0.4384, + "step": 6978 + }, + { + "epoch": 0.3899209430957902, + "grad_norm": 0.4955415725708008, + "learning_rate": 0.000807653518601524, + "loss": 0.4624, + "step": 6979 + }, + { + "epoch": 0.389976813699472, + "grad_norm": 0.6627122163772583, + "learning_rate": 0.0008076255042581802, + "loss": 0.5613, + "step": 6980 + }, + { + "epoch": 0.3900326843031539, + "grad_norm": 0.49047529697418213, + "learning_rate": 0.0008075974899148364, + "loss": 0.4527, + "step": 6981 + }, + { + "epoch": 0.39008855490683575, + "grad_norm": 0.6251693964004517, + "learning_rate": 0.0008075694755714926, + "loss": 0.4966, + "step": 6982 + }, + { + "epoch": 0.39014442551051765, + "grad_norm": 0.5330844521522522, + "learning_rate": 0.0008075414612281488, + "loss": 0.4172, + "step": 6983 + }, + { + "epoch": 0.39020029611419954, + "grad_norm": 0.8396794199943542, + "learning_rate": 0.000807513446884805, + "loss": 0.463, + "step": 6984 + }, + { + "epoch": 0.3902561667178814, + "grad_norm": 0.9876586198806763, + "learning_rate": 0.0008074854325414612, + "loss": 0.4443, + "step": 6985 + }, + { + "epoch": 0.3903120373215633, + "grad_norm": 0.8338679671287537, + "learning_rate": 0.0008074574181981174, + "loss": 0.3843, + "step": 6986 + }, + { + "epoch": 0.3903679079252451, + "grad_norm": 0.621367871761322, + "learning_rate": 0.0008074294038547736, + "loss": 0.4904, + "step": 6987 + }, + { + "epoch": 0.390423778528927, + "grad_norm": 0.8843216300010681, + "learning_rate": 0.0008074013895114299, + "loss": 0.5263, + "step": 6988 + }, + { + "epoch": 0.39047964913260885, + "grad_norm": 0.6882887482643127, + "learning_rate": 0.0008073733751680861, + "loss": 0.4309, + "step": 6989 + }, + { + "epoch": 0.39053551973629075, + "grad_norm": 0.6014006733894348, + "learning_rate": 0.0008073453608247423, + "loss": 0.3658, + "step": 6990 + }, + { + "epoch": 0.39059139033997264, + "grad_norm": 0.7708397507667542, + "learning_rate": 0.0008073173464813986, + "loss": 0.4495, + "step": 6991 + }, + { + "epoch": 0.3906472609436545, + "grad_norm": 1.1912121772766113, + "learning_rate": 0.0008072893321380547, + "loss": 0.4691, + "step": 6992 + }, + { + "epoch": 0.3907031315473364, + "grad_norm": 0.8313740491867065, + "learning_rate": 0.0008072613177947109, + "loss": 0.5386, + "step": 6993 + }, + { + "epoch": 0.3907590021510182, + "grad_norm": 0.36238792538642883, + "learning_rate": 0.0008072333034513671, + "loss": 0.4186, + "step": 6994 + }, + { + "epoch": 0.3908148727547001, + "grad_norm": 0.5884454846382141, + "learning_rate": 0.0008072052891080234, + "loss": 0.5166, + "step": 6995 + }, + { + "epoch": 0.390870743358382, + "grad_norm": 1.141343593597412, + "learning_rate": 0.0008071772747646795, + "loss": 0.4847, + "step": 6996 + }, + { + "epoch": 0.39092661396206385, + "grad_norm": 1.1846855878829956, + "learning_rate": 0.0008071492604213357, + "loss": 0.5089, + "step": 6997 + }, + { + "epoch": 0.39098248456574575, + "grad_norm": 1.6667908430099487, + "learning_rate": 0.000807121246077992, + "loss": 0.4094, + "step": 6998 + }, + { + "epoch": 0.3910383551694276, + "grad_norm": 1.3995994329452515, + "learning_rate": 0.0008070932317346481, + "loss": 0.4684, + "step": 6999 + }, + { + "epoch": 0.3910942257731095, + "grad_norm": 0.40428268909454346, + "learning_rate": 0.0008070652173913043, + "loss": 0.4455, + "step": 7000 + }, + { + "epoch": 0.3910942257731095, + "eval_cer": 0.0936685105779407, + "eval_loss": 0.35221439599990845, + "eval_runtime": 56.2428, + "eval_samples_per_second": 80.686, + "eval_steps_per_second": 5.05, + "eval_wer": 0.3694868858411776, + "step": 7000 + }, + { + "epoch": 0.3911500963767914, + "grad_norm": 0.40490999817848206, + "learning_rate": 0.0008070372030479605, + "loss": 0.4822, + "step": 7001 + }, + { + "epoch": 0.3912059669804732, + "grad_norm": 0.4826464354991913, + "learning_rate": 0.0008070091887046168, + "loss": 0.5557, + "step": 7002 + }, + { + "epoch": 0.3912618375841551, + "grad_norm": 4.189666748046875, + "learning_rate": 0.0008069811743612729, + "loss": 0.4709, + "step": 7003 + }, + { + "epoch": 0.39131770818783695, + "grad_norm": 0.6825993657112122, + "learning_rate": 0.0008069531600179291, + "loss": 0.5478, + "step": 7004 + }, + { + "epoch": 0.39137357879151885, + "grad_norm": 1.1585854291915894, + "learning_rate": 0.0008069251456745855, + "loss": 0.3788, + "step": 7005 + }, + { + "epoch": 0.3914294493952007, + "grad_norm": 0.7272922396659851, + "learning_rate": 0.0008068971313312417, + "loss": 0.5432, + "step": 7006 + }, + { + "epoch": 0.3914853199988826, + "grad_norm": 0.6562459468841553, + "learning_rate": 0.0008068691169878978, + "loss": 0.5608, + "step": 7007 + }, + { + "epoch": 0.3915411906025645, + "grad_norm": 0.6749001145362854, + "learning_rate": 0.000806841102644554, + "loss": 0.4902, + "step": 7008 + }, + { + "epoch": 0.3915970612062463, + "grad_norm": 0.9470919966697693, + "learning_rate": 0.0008068130883012103, + "loss": 0.5031, + "step": 7009 + }, + { + "epoch": 0.3916529318099282, + "grad_norm": 0.702444851398468, + "learning_rate": 0.0008067850739578665, + "loss": 0.3956, + "step": 7010 + }, + { + "epoch": 0.39170880241361006, + "grad_norm": 0.730505108833313, + "learning_rate": 0.0008067570596145226, + "loss": 0.6043, + "step": 7011 + }, + { + "epoch": 0.39176467301729195, + "grad_norm": 1.2646422386169434, + "learning_rate": 0.0008067290452711789, + "loss": 0.4208, + "step": 7012 + }, + { + "epoch": 0.39182054362097385, + "grad_norm": 0.7044093012809753, + "learning_rate": 0.0008067010309278351, + "loss": 0.5549, + "step": 7013 + }, + { + "epoch": 0.3918764142246557, + "grad_norm": 0.565847635269165, + "learning_rate": 0.0008066730165844913, + "loss": 0.377, + "step": 7014 + }, + { + "epoch": 0.3919322848283376, + "grad_norm": 0.6684389710426331, + "learning_rate": 0.0008066450022411474, + "loss": 0.4011, + "step": 7015 + }, + { + "epoch": 0.3919881554320194, + "grad_norm": 0.7841991186141968, + "learning_rate": 0.0008066169878978037, + "loss": 0.4989, + "step": 7016 + }, + { + "epoch": 0.3920440260357013, + "grad_norm": 0.5033202767372131, + "learning_rate": 0.0008065889735544599, + "loss": 0.3486, + "step": 7017 + }, + { + "epoch": 0.39209989663938316, + "grad_norm": 0.6548042297363281, + "learning_rate": 0.0008065609592111161, + "loss": 0.3974, + "step": 7018 + }, + { + "epoch": 0.39215576724306506, + "grad_norm": 0.6785100102424622, + "learning_rate": 0.0008065329448677722, + "loss": 0.4481, + "step": 7019 + }, + { + "epoch": 0.39221163784674695, + "grad_norm": 0.6446042656898499, + "learning_rate": 0.0008065049305244285, + "loss": 0.5111, + "step": 7020 + }, + { + "epoch": 0.3922675084504288, + "grad_norm": 0.8722866773605347, + "learning_rate": 0.0008064769161810848, + "loss": 0.4846, + "step": 7021 + }, + { + "epoch": 0.3923233790541107, + "grad_norm": 0.5611283779144287, + "learning_rate": 0.0008064489018377409, + "loss": 0.4631, + "step": 7022 + }, + { + "epoch": 0.39237924965779253, + "grad_norm": 1.0797009468078613, + "learning_rate": 0.0008064208874943972, + "loss": 0.5466, + "step": 7023 + }, + { + "epoch": 0.3924351202614744, + "grad_norm": 0.4275990128517151, + "learning_rate": 0.0008063928731510534, + "loss": 0.347, + "step": 7024 + }, + { + "epoch": 0.3924909908651563, + "grad_norm": 2.003610849380493, + "learning_rate": 0.0008063648588077096, + "loss": 0.4655, + "step": 7025 + }, + { + "epoch": 0.39254686146883816, + "grad_norm": 0.42713460326194763, + "learning_rate": 0.0008063368444643657, + "loss": 0.4046, + "step": 7026 + }, + { + "epoch": 0.39260273207252006, + "grad_norm": 0.6833263635635376, + "learning_rate": 0.000806308830121022, + "loss": 0.4882, + "step": 7027 + }, + { + "epoch": 0.3926586026762019, + "grad_norm": 1.607764720916748, + "learning_rate": 0.0008062808157776782, + "loss": 0.5737, + "step": 7028 + }, + { + "epoch": 0.3927144732798838, + "grad_norm": 1.6396644115447998, + "learning_rate": 0.0008062528014343344, + "loss": 0.4533, + "step": 7029 + }, + { + "epoch": 0.3927703438835657, + "grad_norm": 0.6118025779724121, + "learning_rate": 0.0008062247870909906, + "loss": 0.4875, + "step": 7030 + }, + { + "epoch": 0.3928262144872475, + "grad_norm": 1.1517512798309326, + "learning_rate": 0.0008061967727476468, + "loss": 0.5018, + "step": 7031 + }, + { + "epoch": 0.3928820850909294, + "grad_norm": 0.5140281915664673, + "learning_rate": 0.000806168758404303, + "loss": 0.5104, + "step": 7032 + }, + { + "epoch": 0.39293795569461126, + "grad_norm": 2.430715799331665, + "learning_rate": 0.0008061407440609592, + "loss": 0.4814, + "step": 7033 + }, + { + "epoch": 0.39299382629829316, + "grad_norm": 0.6719557046890259, + "learning_rate": 0.0008061127297176154, + "loss": 0.4744, + "step": 7034 + }, + { + "epoch": 0.393049696901975, + "grad_norm": 0.5886644721031189, + "learning_rate": 0.0008060847153742716, + "loss": 0.4723, + "step": 7035 + }, + { + "epoch": 0.3931055675056569, + "grad_norm": 0.6164878010749817, + "learning_rate": 0.0008060567010309278, + "loss": 0.5622, + "step": 7036 + }, + { + "epoch": 0.3931614381093388, + "grad_norm": 0.5846493244171143, + "learning_rate": 0.0008060286866875842, + "loss": 0.4328, + "step": 7037 + }, + { + "epoch": 0.39321730871302063, + "grad_norm": 0.5009116530418396, + "learning_rate": 0.0008060006723442403, + "loss": 0.3943, + "step": 7038 + }, + { + "epoch": 0.3932731793167025, + "grad_norm": 0.4654540419578552, + "learning_rate": 0.0008059726580008965, + "loss": 0.4479, + "step": 7039 + }, + { + "epoch": 0.39332904992038437, + "grad_norm": 0.4710710942745209, + "learning_rate": 0.0008059446436575527, + "loss": 0.3948, + "step": 7040 + }, + { + "epoch": 0.39338492052406626, + "grad_norm": 0.6461930871009827, + "learning_rate": 0.0008059166293142089, + "loss": 0.4598, + "step": 7041 + }, + { + "epoch": 0.39344079112774816, + "grad_norm": 2.4650542736053467, + "learning_rate": 0.0008058886149708651, + "loss": 0.4513, + "step": 7042 + }, + { + "epoch": 0.39349666173143, + "grad_norm": 0.6406357288360596, + "learning_rate": 0.0008058606006275213, + "loss": 0.4801, + "step": 7043 + }, + { + "epoch": 0.3935525323351119, + "grad_norm": 0.4064398407936096, + "learning_rate": 0.0008058325862841776, + "loss": 0.48, + "step": 7044 + }, + { + "epoch": 0.39360840293879373, + "grad_norm": 0.6963995695114136, + "learning_rate": 0.0008058045719408337, + "loss": 0.4162, + "step": 7045 + }, + { + "epoch": 0.39366427354247563, + "grad_norm": 0.4416905343532562, + "learning_rate": 0.0008057765575974899, + "loss": 0.4804, + "step": 7046 + }, + { + "epoch": 0.3937201441461575, + "grad_norm": 0.5299258828163147, + "learning_rate": 0.0008057485432541461, + "loss": 0.456, + "step": 7047 + }, + { + "epoch": 0.39377601474983936, + "grad_norm": 0.4521815776824951, + "learning_rate": 0.0008057205289108024, + "loss": 0.4796, + "step": 7048 + }, + { + "epoch": 0.39383188535352126, + "grad_norm": 0.5987514853477478, + "learning_rate": 0.0008056925145674585, + "loss": 0.4345, + "step": 7049 + }, + { + "epoch": 0.3938877559572031, + "grad_norm": 0.4292525351047516, + "learning_rate": 0.0008056645002241147, + "loss": 0.4225, + "step": 7050 + }, + { + "epoch": 0.393943626560885, + "grad_norm": 3.4692680835723877, + "learning_rate": 0.000805636485880771, + "loss": 0.4736, + "step": 7051 + }, + { + "epoch": 0.39399949716456684, + "grad_norm": 1.3877366781234741, + "learning_rate": 0.0008056084715374272, + "loss": 0.5921, + "step": 7052 + }, + { + "epoch": 0.39405536776824873, + "grad_norm": 0.6557919383049011, + "learning_rate": 0.0008055804571940833, + "loss": 0.4049, + "step": 7053 + }, + { + "epoch": 0.39411123837193063, + "grad_norm": 0.5572063326835632, + "learning_rate": 0.0008055524428507395, + "loss": 0.4575, + "step": 7054 + }, + { + "epoch": 0.39416710897561247, + "grad_norm": 0.5603480339050293, + "learning_rate": 0.0008055244285073959, + "loss": 0.4514, + "step": 7055 + }, + { + "epoch": 0.39422297957929436, + "grad_norm": 9.347752571105957, + "learning_rate": 0.0008054964141640521, + "loss": 0.5199, + "step": 7056 + }, + { + "epoch": 0.3942788501829762, + "grad_norm": 0.8515098690986633, + "learning_rate": 0.0008054683998207082, + "loss": 0.4523, + "step": 7057 + }, + { + "epoch": 0.3943347207866581, + "grad_norm": 0.7708616852760315, + "learning_rate": 0.0008054403854773644, + "loss": 0.4439, + "step": 7058 + }, + { + "epoch": 0.39439059139034, + "grad_norm": 0.5737174153327942, + "learning_rate": 0.0008054123711340207, + "loss": 0.5322, + "step": 7059 + }, + { + "epoch": 0.39444646199402184, + "grad_norm": 0.720364511013031, + "learning_rate": 0.0008053843567906769, + "loss": 0.4899, + "step": 7060 + }, + { + "epoch": 0.39450233259770373, + "grad_norm": 0.8781117796897888, + "learning_rate": 0.000805356342447333, + "loss": 0.6475, + "step": 7061 + }, + { + "epoch": 0.39455820320138557, + "grad_norm": 0.5328068733215332, + "learning_rate": 0.0008053283281039893, + "loss": 0.3559, + "step": 7062 + }, + { + "epoch": 0.39461407380506747, + "grad_norm": 0.7704755663871765, + "learning_rate": 0.0008053003137606455, + "loss": 0.5102, + "step": 7063 + }, + { + "epoch": 0.39466994440874936, + "grad_norm": 0.5940272808074951, + "learning_rate": 0.0008052722994173016, + "loss": 0.5272, + "step": 7064 + }, + { + "epoch": 0.3947258150124312, + "grad_norm": 0.752748429775238, + "learning_rate": 0.0008052442850739578, + "loss": 0.5709, + "step": 7065 + }, + { + "epoch": 0.3947816856161131, + "grad_norm": 1.2601189613342285, + "learning_rate": 0.0008052162707306141, + "loss": 0.5202, + "step": 7066 + }, + { + "epoch": 0.39483755621979494, + "grad_norm": 0.608449399471283, + "learning_rate": 0.0008051882563872703, + "loss": 0.4168, + "step": 7067 + }, + { + "epoch": 0.39489342682347683, + "grad_norm": 5.254899978637695, + "learning_rate": 0.0008051602420439264, + "loss": 0.4827, + "step": 7068 + }, + { + "epoch": 0.3949492974271587, + "grad_norm": 0.6568675637245178, + "learning_rate": 0.0008051322277005827, + "loss": 0.6326, + "step": 7069 + }, + { + "epoch": 0.39500516803084057, + "grad_norm": 0.7192527651786804, + "learning_rate": 0.000805104213357239, + "loss": 0.4629, + "step": 7070 + }, + { + "epoch": 0.39506103863452247, + "grad_norm": 1.1017980575561523, + "learning_rate": 0.0008050761990138952, + "loss": 0.4375, + "step": 7071 + }, + { + "epoch": 0.3951169092382043, + "grad_norm": 1.108139157295227, + "learning_rate": 0.0008050481846705513, + "loss": 0.3655, + "step": 7072 + }, + { + "epoch": 0.3951727798418862, + "grad_norm": 0.47852417826652527, + "learning_rate": 0.0008050201703272076, + "loss": 0.4948, + "step": 7073 + }, + { + "epoch": 0.39522865044556804, + "grad_norm": 1.3772486448287964, + "learning_rate": 0.0008049921559838638, + "loss": 0.436, + "step": 7074 + }, + { + "epoch": 0.39528452104924994, + "grad_norm": 0.5786846876144409, + "learning_rate": 0.00080496414164052, + "loss": 0.4755, + "step": 7075 + }, + { + "epoch": 0.39534039165293183, + "grad_norm": 0.9717200994491577, + "learning_rate": 0.0008049361272971762, + "loss": 0.5038, + "step": 7076 + }, + { + "epoch": 0.3953962622566137, + "grad_norm": 1.1353029012680054, + "learning_rate": 0.0008049081129538324, + "loss": 0.4068, + "step": 7077 + }, + { + "epoch": 0.39545213286029557, + "grad_norm": 0.6178039312362671, + "learning_rate": 0.0008048800986104886, + "loss": 0.5038, + "step": 7078 + }, + { + "epoch": 0.3955080034639774, + "grad_norm": 1.2133333683013916, + "learning_rate": 0.0008048520842671448, + "loss": 0.4895, + "step": 7079 + }, + { + "epoch": 0.3955638740676593, + "grad_norm": 1.1935129165649414, + "learning_rate": 0.000804824069923801, + "loss": 0.4478, + "step": 7080 + }, + { + "epoch": 0.3956197446713412, + "grad_norm": 0.5273128747940063, + "learning_rate": 0.0008047960555804572, + "loss": 0.4459, + "step": 7081 + }, + { + "epoch": 0.39567561527502304, + "grad_norm": 0.4863780736923218, + "learning_rate": 0.0008047680412371134, + "loss": 0.5526, + "step": 7082 + }, + { + "epoch": 0.39573148587870494, + "grad_norm": 0.9550788402557373, + "learning_rate": 0.0008047400268937696, + "loss": 0.6061, + "step": 7083 + }, + { + "epoch": 0.3957873564823868, + "grad_norm": 0.48134344816207886, + "learning_rate": 0.0008047120125504258, + "loss": 0.4779, + "step": 7084 + }, + { + "epoch": 0.39584322708606867, + "grad_norm": 0.44533395767211914, + "learning_rate": 0.000804683998207082, + "loss": 0.5535, + "step": 7085 + }, + { + "epoch": 0.3958990976897505, + "grad_norm": 0.4133225977420807, + "learning_rate": 0.0008046559838637382, + "loss": 0.4107, + "step": 7086 + }, + { + "epoch": 0.3959549682934324, + "grad_norm": 0.5812316536903381, + "learning_rate": 0.0008046279695203944, + "loss": 0.496, + "step": 7087 + }, + { + "epoch": 0.3960108388971143, + "grad_norm": 2.2453341484069824, + "learning_rate": 0.0008045999551770507, + "loss": 0.4447, + "step": 7088 + }, + { + "epoch": 0.39606670950079614, + "grad_norm": 0.6074969172477722, + "learning_rate": 0.0008045719408337069, + "loss": 0.4606, + "step": 7089 + }, + { + "epoch": 0.39612258010447804, + "grad_norm": 0.6073816418647766, + "learning_rate": 0.0008045439264903632, + "loss": 0.5389, + "step": 7090 + }, + { + "epoch": 0.3961784507081599, + "grad_norm": 0.48919567465782166, + "learning_rate": 0.0008045159121470193, + "loss": 0.5663, + "step": 7091 + }, + { + "epoch": 0.3962343213118418, + "grad_norm": 0.4172901511192322, + "learning_rate": 0.0008044878978036755, + "loss": 0.4051, + "step": 7092 + }, + { + "epoch": 0.39629019191552367, + "grad_norm": 0.4749389588832855, + "learning_rate": 0.0008044598834603317, + "loss": 0.4995, + "step": 7093 + }, + { + "epoch": 0.3963460625192055, + "grad_norm": 1.072256088256836, + "learning_rate": 0.000804431869116988, + "loss": 0.4383, + "step": 7094 + }, + { + "epoch": 0.3964019331228874, + "grad_norm": 0.5148457288742065, + "learning_rate": 0.0008044038547736441, + "loss": 0.6217, + "step": 7095 + }, + { + "epoch": 0.39645780372656925, + "grad_norm": 0.9998928308486938, + "learning_rate": 0.0008043758404303003, + "loss": 0.5327, + "step": 7096 + }, + { + "epoch": 0.39651367433025114, + "grad_norm": 0.8931484222412109, + "learning_rate": 0.0008043478260869565, + "loss": 0.6063, + "step": 7097 + }, + { + "epoch": 0.39656954493393304, + "grad_norm": 0.5696259140968323, + "learning_rate": 0.0008043198117436128, + "loss": 0.3531, + "step": 7098 + }, + { + "epoch": 0.3966254155376149, + "grad_norm": 3.484323740005493, + "learning_rate": 0.0008042917974002689, + "loss": 0.4473, + "step": 7099 + }, + { + "epoch": 0.3966812861412968, + "grad_norm": 0.40621596574783325, + "learning_rate": 0.0008042637830569251, + "loss": 0.4033, + "step": 7100 + }, + { + "epoch": 0.3967371567449786, + "grad_norm": 0.4742153286933899, + "learning_rate": 0.0008042357687135814, + "loss": 0.3899, + "step": 7101 + }, + { + "epoch": 0.3967930273486605, + "grad_norm": 0.5437451601028442, + "learning_rate": 0.0008042077543702376, + "loss": 0.4928, + "step": 7102 + }, + { + "epoch": 0.39684889795234235, + "grad_norm": 0.8460273146629333, + "learning_rate": 0.0008041797400268937, + "loss": 0.5471, + "step": 7103 + }, + { + "epoch": 0.39690476855602425, + "grad_norm": 0.4912261366844177, + "learning_rate": 0.00080415172568355, + "loss": 0.4241, + "step": 7104 + }, + { + "epoch": 0.39696063915970614, + "grad_norm": 0.7901257276535034, + "learning_rate": 0.0008041237113402063, + "loss": 0.4883, + "step": 7105 + }, + { + "epoch": 0.397016509763388, + "grad_norm": 0.49082717299461365, + "learning_rate": 0.0008040956969968624, + "loss": 0.5032, + "step": 7106 + }, + { + "epoch": 0.3970723803670699, + "grad_norm": 0.9747931957244873, + "learning_rate": 0.0008040676826535186, + "loss": 0.5874, + "step": 7107 + }, + { + "epoch": 0.3971282509707517, + "grad_norm": 0.6041909456253052, + "learning_rate": 0.0008040396683101749, + "loss": 0.3837, + "step": 7108 + }, + { + "epoch": 0.3971841215744336, + "grad_norm": 0.6338514089584351, + "learning_rate": 0.0008040116539668311, + "loss": 0.4641, + "step": 7109 + }, + { + "epoch": 0.3972399921781155, + "grad_norm": 0.7022388577461243, + "learning_rate": 0.0008039836396234872, + "loss": 0.4661, + "step": 7110 + }, + { + "epoch": 0.39729586278179735, + "grad_norm": 0.4761565029621124, + "learning_rate": 0.0008039556252801434, + "loss": 0.4505, + "step": 7111 + }, + { + "epoch": 0.39735173338547924, + "grad_norm": 0.46938544511795044, + "learning_rate": 0.0008039276109367997, + "loss": 0.3558, + "step": 7112 + }, + { + "epoch": 0.3974076039891611, + "grad_norm": 0.8038325905799866, + "learning_rate": 0.0008038995965934559, + "loss": 0.4893, + "step": 7113 + }, + { + "epoch": 0.397463474592843, + "grad_norm": 0.503667950630188, + "learning_rate": 0.000803871582250112, + "loss": 0.4275, + "step": 7114 + }, + { + "epoch": 0.3975193451965249, + "grad_norm": 0.5231167078018188, + "learning_rate": 0.0008038435679067683, + "loss": 0.6397, + "step": 7115 + }, + { + "epoch": 0.3975752158002067, + "grad_norm": 2.2609357833862305, + "learning_rate": 0.0008038155535634245, + "loss": 0.4178, + "step": 7116 + }, + { + "epoch": 0.3976310864038886, + "grad_norm": 0.9087424278259277, + "learning_rate": 0.0008037875392200807, + "loss": 0.5329, + "step": 7117 + }, + { + "epoch": 0.39768695700757045, + "grad_norm": 0.5329527258872986, + "learning_rate": 0.0008037595248767368, + "loss": 0.5272, + "step": 7118 + }, + { + "epoch": 0.39774282761125235, + "grad_norm": 3.035311698913574, + "learning_rate": 0.0008037315105333931, + "loss": 0.5582, + "step": 7119 + }, + { + "epoch": 0.3977986982149342, + "grad_norm": 1.1097420454025269, + "learning_rate": 0.0008037034961900493, + "loss": 0.4526, + "step": 7120 + }, + { + "epoch": 0.3978545688186161, + "grad_norm": 0.5787144899368286, + "learning_rate": 0.0008036754818467056, + "loss": 0.4072, + "step": 7121 + }, + { + "epoch": 0.397910439422298, + "grad_norm": 0.9428479075431824, + "learning_rate": 0.0008036474675033618, + "loss": 0.5241, + "step": 7122 + }, + { + "epoch": 0.3979663100259798, + "grad_norm": 0.795162558555603, + "learning_rate": 0.000803619453160018, + "loss": 0.6425, + "step": 7123 + }, + { + "epoch": 0.3980221806296617, + "grad_norm": 0.4295109212398529, + "learning_rate": 0.0008035914388166742, + "loss": 0.329, + "step": 7124 + }, + { + "epoch": 0.39807805123334356, + "grad_norm": 0.5008066296577454, + "learning_rate": 0.0008035634244733303, + "loss": 0.5757, + "step": 7125 + }, + { + "epoch": 0.39813392183702545, + "grad_norm": 0.4638589024543762, + "learning_rate": 0.0008035354101299866, + "loss": 0.4814, + "step": 7126 + }, + { + "epoch": 0.39818979244070735, + "grad_norm": 0.5590055584907532, + "learning_rate": 0.0008035073957866428, + "loss": 0.4801, + "step": 7127 + }, + { + "epoch": 0.3982456630443892, + "grad_norm": 0.5019645690917969, + "learning_rate": 0.000803479381443299, + "loss": 0.4232, + "step": 7128 + }, + { + "epoch": 0.3983015336480711, + "grad_norm": 0.5045915246009827, + "learning_rate": 0.0008034513670999551, + "loss": 0.4198, + "step": 7129 + }, + { + "epoch": 0.3983574042517529, + "grad_norm": 0.504115104675293, + "learning_rate": 0.0008034233527566114, + "loss": 0.4615, + "step": 7130 + }, + { + "epoch": 0.3984132748554348, + "grad_norm": 1.8904446363449097, + "learning_rate": 0.0008033953384132676, + "loss": 0.6113, + "step": 7131 + }, + { + "epoch": 0.39846914545911666, + "grad_norm": 0.45941078662872314, + "learning_rate": 0.0008033673240699238, + "loss": 0.3904, + "step": 7132 + }, + { + "epoch": 0.39852501606279855, + "grad_norm": 0.5465155243873596, + "learning_rate": 0.00080333930972658, + "loss": 0.4918, + "step": 7133 + }, + { + "epoch": 0.39858088666648045, + "grad_norm": 0.4900111258029938, + "learning_rate": 0.0008033112953832362, + "loss": 0.4083, + "step": 7134 + }, + { + "epoch": 0.3986367572701623, + "grad_norm": 3.1926937103271484, + "learning_rate": 0.0008032832810398924, + "loss": 0.3942, + "step": 7135 + }, + { + "epoch": 0.3986926278738442, + "grad_norm": 0.7979128360748291, + "learning_rate": 0.0008032552666965486, + "loss": 0.5082, + "step": 7136 + }, + { + "epoch": 0.398748498477526, + "grad_norm": 0.8961465358734131, + "learning_rate": 0.0008032272523532048, + "loss": 0.5042, + "step": 7137 + }, + { + "epoch": 0.3988043690812079, + "grad_norm": 0.8508933186531067, + "learning_rate": 0.000803199238009861, + "loss": 0.4395, + "step": 7138 + }, + { + "epoch": 0.3988602396848898, + "grad_norm": 0.625939667224884, + "learning_rate": 0.0008031712236665173, + "loss": 0.61, + "step": 7139 + }, + { + "epoch": 0.39891611028857166, + "grad_norm": 0.8425270915031433, + "learning_rate": 0.0008031432093231736, + "loss": 0.5029, + "step": 7140 + }, + { + "epoch": 0.39897198089225355, + "grad_norm": 0.7388454079627991, + "learning_rate": 0.0008031151949798297, + "loss": 0.5176, + "step": 7141 + }, + { + "epoch": 0.3990278514959354, + "grad_norm": 1.0234779119491577, + "learning_rate": 0.0008030871806364859, + "loss": 0.7837, + "step": 7142 + }, + { + "epoch": 0.3990837220996173, + "grad_norm": 0.45851391553878784, + "learning_rate": 0.0008030591662931421, + "loss": 0.442, + "step": 7143 + }, + { + "epoch": 0.3991395927032992, + "grad_norm": 0.6590308547019958, + "learning_rate": 0.0008030311519497984, + "loss": 0.4107, + "step": 7144 + }, + { + "epoch": 0.399195463306981, + "grad_norm": 0.4822482764720917, + "learning_rate": 0.0008030031376064545, + "loss": 0.4307, + "step": 7145 + }, + { + "epoch": 0.3992513339106629, + "grad_norm": 0.7174953818321228, + "learning_rate": 0.0008029751232631107, + "loss": 0.5195, + "step": 7146 + }, + { + "epoch": 0.39930720451434476, + "grad_norm": 0.41072356700897217, + "learning_rate": 0.000802947108919767, + "loss": 0.4536, + "step": 7147 + }, + { + "epoch": 0.39936307511802666, + "grad_norm": 0.543739914894104, + "learning_rate": 0.0008029190945764231, + "loss": 0.4423, + "step": 7148 + }, + { + "epoch": 0.3994189457217085, + "grad_norm": 0.8701680302619934, + "learning_rate": 0.0008028910802330793, + "loss": 0.5363, + "step": 7149 + }, + { + "epoch": 0.3994748163253904, + "grad_norm": 0.4381810426712036, + "learning_rate": 0.0008028630658897355, + "loss": 0.392, + "step": 7150 + }, + { + "epoch": 0.3995306869290723, + "grad_norm": 0.6614002585411072, + "learning_rate": 0.0008028350515463918, + "loss": 0.3372, + "step": 7151 + }, + { + "epoch": 0.3995865575327541, + "grad_norm": 2.292942762374878, + "learning_rate": 0.0008028070372030479, + "loss": 0.5232, + "step": 7152 + }, + { + "epoch": 0.399642428136436, + "grad_norm": 0.5718908309936523, + "learning_rate": 0.0008027790228597041, + "loss": 0.461, + "step": 7153 + }, + { + "epoch": 0.39969829874011786, + "grad_norm": 0.8802366852760315, + "learning_rate": 0.0008027510085163605, + "loss": 0.5204, + "step": 7154 + }, + { + "epoch": 0.39975416934379976, + "grad_norm": 0.5155148506164551, + "learning_rate": 0.0008027229941730167, + "loss": 0.3666, + "step": 7155 + }, + { + "epoch": 0.39981003994748165, + "grad_norm": 0.8661115169525146, + "learning_rate": 0.0008026949798296728, + "loss": 0.4942, + "step": 7156 + }, + { + "epoch": 0.3998659105511635, + "grad_norm": 1.946569561958313, + "learning_rate": 0.000802666965486329, + "loss": 0.5553, + "step": 7157 + }, + { + "epoch": 0.3999217811548454, + "grad_norm": 0.6115090250968933, + "learning_rate": 0.0008026389511429853, + "loss": 0.3595, + "step": 7158 + }, + { + "epoch": 0.39997765175852723, + "grad_norm": 0.5948266386985779, + "learning_rate": 0.0008026109367996415, + "loss": 0.4931, + "step": 7159 + }, + { + "epoch": 0.4000335223622091, + "grad_norm": 0.7198998332023621, + "learning_rate": 0.0008025829224562976, + "loss": 0.5336, + "step": 7160 + }, + { + "epoch": 0.400089392965891, + "grad_norm": 5.1747283935546875, + "learning_rate": 0.0008025549081129539, + "loss": 0.5234, + "step": 7161 + }, + { + "epoch": 0.40014526356957286, + "grad_norm": 0.46145981550216675, + "learning_rate": 0.0008025268937696101, + "loss": 0.4403, + "step": 7162 + }, + { + "epoch": 0.40020113417325476, + "grad_norm": 0.5754908919334412, + "learning_rate": 0.0008024988794262663, + "loss": 0.4218, + "step": 7163 + }, + { + "epoch": 0.4002570047769366, + "grad_norm": 0.4187431037425995, + "learning_rate": 0.0008024708650829224, + "loss": 0.4518, + "step": 7164 + }, + { + "epoch": 0.4003128753806185, + "grad_norm": 0.60135418176651, + "learning_rate": 0.0008024428507395787, + "loss": 0.426, + "step": 7165 + }, + { + "epoch": 0.40036874598430033, + "grad_norm": 0.4838568866252899, + "learning_rate": 0.0008024148363962349, + "loss": 0.4872, + "step": 7166 + }, + { + "epoch": 0.40042461658798223, + "grad_norm": 1.561225414276123, + "learning_rate": 0.0008023868220528911, + "loss": 0.4919, + "step": 7167 + }, + { + "epoch": 0.4004804871916641, + "grad_norm": 0.4696406126022339, + "learning_rate": 0.0008023588077095472, + "loss": 0.461, + "step": 7168 + }, + { + "epoch": 0.40053635779534597, + "grad_norm": 0.609910249710083, + "learning_rate": 0.0008023307933662035, + "loss": 0.4994, + "step": 7169 + }, + { + "epoch": 0.40059222839902786, + "grad_norm": 0.9595982432365417, + "learning_rate": 0.0008023027790228597, + "loss": 0.5127, + "step": 7170 + }, + { + "epoch": 0.4006480990027097, + "grad_norm": 0.7092859745025635, + "learning_rate": 0.0008022747646795158, + "loss": 0.4739, + "step": 7171 + }, + { + "epoch": 0.4007039696063916, + "grad_norm": 0.5040832757949829, + "learning_rate": 0.0008022467503361722, + "loss": 0.5066, + "step": 7172 + }, + { + "epoch": 0.4007598402100735, + "grad_norm": 6.338099002838135, + "learning_rate": 0.0008022187359928284, + "loss": 0.5057, + "step": 7173 + }, + { + "epoch": 0.40081571081375533, + "grad_norm": 0.5771711468696594, + "learning_rate": 0.0008021907216494846, + "loss": 0.3928, + "step": 7174 + }, + { + "epoch": 0.40087158141743723, + "grad_norm": 0.6248647570610046, + "learning_rate": 0.0008021627073061407, + "loss": 0.5198, + "step": 7175 + }, + { + "epoch": 0.40092745202111907, + "grad_norm": 0.5388386249542236, + "learning_rate": 0.000802134692962797, + "loss": 0.5408, + "step": 7176 + }, + { + "epoch": 0.40098332262480096, + "grad_norm": 0.5270634293556213, + "learning_rate": 0.0008021066786194532, + "loss": 0.5889, + "step": 7177 + }, + { + "epoch": 0.40103919322848286, + "grad_norm": 0.6599588394165039, + "learning_rate": 0.0008020786642761094, + "loss": 0.4243, + "step": 7178 + }, + { + "epoch": 0.4010950638321647, + "grad_norm": 0.6400723457336426, + "learning_rate": 0.0008020506499327656, + "loss": 0.4937, + "step": 7179 + }, + { + "epoch": 0.4011509344358466, + "grad_norm": 1.0210347175598145, + "learning_rate": 0.0008020226355894218, + "loss": 0.4692, + "step": 7180 + }, + { + "epoch": 0.40120680503952844, + "grad_norm": 0.5628160238265991, + "learning_rate": 0.000801994621246078, + "loss": 0.4327, + "step": 7181 + }, + { + "epoch": 0.40126267564321033, + "grad_norm": 0.7164530754089355, + "learning_rate": 0.0008019666069027342, + "loss": 0.6173, + "step": 7182 + }, + { + "epoch": 0.40131854624689217, + "grad_norm": 1.0394716262817383, + "learning_rate": 0.0008019385925593904, + "loss": 0.5181, + "step": 7183 + }, + { + "epoch": 0.40137441685057407, + "grad_norm": 0.7929601669311523, + "learning_rate": 0.0008019105782160466, + "loss": 0.4117, + "step": 7184 + }, + { + "epoch": 0.40143028745425596, + "grad_norm": 0.5934284329414368, + "learning_rate": 0.0008018825638727028, + "loss": 0.4143, + "step": 7185 + }, + { + "epoch": 0.4014861580579378, + "grad_norm": 0.5897178649902344, + "learning_rate": 0.0008018545495293591, + "loss": 0.4397, + "step": 7186 + }, + { + "epoch": 0.4015420286616197, + "grad_norm": 0.5764854550361633, + "learning_rate": 0.0008018265351860152, + "loss": 0.3833, + "step": 7187 + }, + { + "epoch": 0.40159789926530154, + "grad_norm": 0.49813756346702576, + "learning_rate": 0.0008017985208426715, + "loss": 0.3921, + "step": 7188 + }, + { + "epoch": 0.40165376986898343, + "grad_norm": 0.8071566820144653, + "learning_rate": 0.0008017705064993277, + "loss": 0.558, + "step": 7189 + }, + { + "epoch": 0.40170964047266533, + "grad_norm": 0.41440466046333313, + "learning_rate": 0.0008017424921559839, + "loss": 0.425, + "step": 7190 + }, + { + "epoch": 0.40176551107634717, + "grad_norm": 0.39291641116142273, + "learning_rate": 0.0008017144778126401, + "loss": 0.5016, + "step": 7191 + }, + { + "epoch": 0.40182138168002907, + "grad_norm": 0.6048544049263, + "learning_rate": 0.0008016864634692963, + "loss": 0.4338, + "step": 7192 + }, + { + "epoch": 0.4018772522837109, + "grad_norm": 0.5008641481399536, + "learning_rate": 0.0008016584491259526, + "loss": 0.48, + "step": 7193 + }, + { + "epoch": 0.4019331228873928, + "grad_norm": 0.8335583209991455, + "learning_rate": 0.0008016304347826087, + "loss": 0.5449, + "step": 7194 + }, + { + "epoch": 0.4019889934910747, + "grad_norm": 0.43796443939208984, + "learning_rate": 0.0008016024204392649, + "loss": 0.459, + "step": 7195 + }, + { + "epoch": 0.40204486409475654, + "grad_norm": 0.6704034209251404, + "learning_rate": 0.0008015744060959211, + "loss": 0.4011, + "step": 7196 + }, + { + "epoch": 0.40210073469843843, + "grad_norm": 0.5232049226760864, + "learning_rate": 0.0008015463917525774, + "loss": 0.5931, + "step": 7197 + }, + { + "epoch": 0.4021566053021203, + "grad_norm": 0.46309325098991394, + "learning_rate": 0.0008015183774092335, + "loss": 0.4327, + "step": 7198 + }, + { + "epoch": 0.40221247590580217, + "grad_norm": 0.8592037558555603, + "learning_rate": 0.0008014903630658897, + "loss": 0.5305, + "step": 7199 + }, + { + "epoch": 0.402268346509484, + "grad_norm": 0.4452190101146698, + "learning_rate": 0.0008014623487225459, + "loss": 0.3412, + "step": 7200 + }, + { + "epoch": 0.4023242171131659, + "grad_norm": 1.0413541793823242, + "learning_rate": 0.0008014343343792022, + "loss": 0.4093, + "step": 7201 + }, + { + "epoch": 0.4023800877168478, + "grad_norm": 0.6014522314071655, + "learning_rate": 0.0008014063200358583, + "loss": 0.4602, + "step": 7202 + }, + { + "epoch": 0.40243595832052964, + "grad_norm": 0.4288637042045593, + "learning_rate": 0.0008013783056925145, + "loss": 0.3986, + "step": 7203 + }, + { + "epoch": 0.40249182892421154, + "grad_norm": 0.4494834542274475, + "learning_rate": 0.0008013502913491709, + "loss": 0.4331, + "step": 7204 + }, + { + "epoch": 0.4025476995278934, + "grad_norm": 1.0962045192718506, + "learning_rate": 0.0008013222770058271, + "loss": 0.5946, + "step": 7205 + }, + { + "epoch": 0.4026035701315753, + "grad_norm": 0.518665611743927, + "learning_rate": 0.0008012942626624832, + "loss": 0.4933, + "step": 7206 + }, + { + "epoch": 0.40265944073525717, + "grad_norm": 0.5924087762832642, + "learning_rate": 0.0008012662483191394, + "loss": 0.6297, + "step": 7207 + }, + { + "epoch": 0.402715311338939, + "grad_norm": 0.5693612694740295, + "learning_rate": 0.0008012382339757957, + "loss": 0.3537, + "step": 7208 + }, + { + "epoch": 0.4027711819426209, + "grad_norm": 2.4433648586273193, + "learning_rate": 0.0008012102196324519, + "loss": 0.4872, + "step": 7209 + }, + { + "epoch": 0.40282705254630274, + "grad_norm": 0.9072322249412537, + "learning_rate": 0.000801182205289108, + "loss": 0.4209, + "step": 7210 + }, + { + "epoch": 0.40288292314998464, + "grad_norm": 0.5287766456604004, + "learning_rate": 0.0008011541909457643, + "loss": 0.5286, + "step": 7211 + }, + { + "epoch": 0.40293879375366654, + "grad_norm": 0.4649658203125, + "learning_rate": 0.0008011261766024205, + "loss": 0.4626, + "step": 7212 + }, + { + "epoch": 0.4029946643573484, + "grad_norm": 1.8517752885818481, + "learning_rate": 0.0008010981622590766, + "loss": 0.4585, + "step": 7213 + }, + { + "epoch": 0.40305053496103027, + "grad_norm": 0.5913764238357544, + "learning_rate": 0.0008010701479157328, + "loss": 0.4529, + "step": 7214 + }, + { + "epoch": 0.4031064055647121, + "grad_norm": 1.839235782623291, + "learning_rate": 0.0008010421335723891, + "loss": 0.409, + "step": 7215 + }, + { + "epoch": 0.403162276168394, + "grad_norm": 0.6820927262306213, + "learning_rate": 0.0008010141192290453, + "loss": 0.3878, + "step": 7216 + }, + { + "epoch": 0.40321814677207585, + "grad_norm": 0.5538651347160339, + "learning_rate": 0.0008009861048857014, + "loss": 0.5763, + "step": 7217 + }, + { + "epoch": 0.40327401737575774, + "grad_norm": 0.4081014394760132, + "learning_rate": 0.0008009580905423577, + "loss": 0.3971, + "step": 7218 + }, + { + "epoch": 0.40332988797943964, + "grad_norm": 0.41973599791526794, + "learning_rate": 0.0008009300761990139, + "loss": 0.3863, + "step": 7219 + }, + { + "epoch": 0.4033857585831215, + "grad_norm": 0.6846643686294556, + "learning_rate": 0.0008009020618556701, + "loss": 0.562, + "step": 7220 + }, + { + "epoch": 0.4034416291868034, + "grad_norm": 0.4224082827568054, + "learning_rate": 0.0008008740475123262, + "loss": 0.4253, + "step": 7221 + }, + { + "epoch": 0.4034974997904852, + "grad_norm": 0.5336311459541321, + "learning_rate": 0.0008008460331689826, + "loss": 0.5825, + "step": 7222 + }, + { + "epoch": 0.4035533703941671, + "grad_norm": 0.5295037031173706, + "learning_rate": 0.0008008180188256388, + "loss": 0.5405, + "step": 7223 + }, + { + "epoch": 0.403609240997849, + "grad_norm": 0.5889543890953064, + "learning_rate": 0.000800790004482295, + "loss": 0.525, + "step": 7224 + }, + { + "epoch": 0.40366511160153085, + "grad_norm": 1.7275726795196533, + "learning_rate": 0.0008007619901389512, + "loss": 0.4504, + "step": 7225 + }, + { + "epoch": 0.40372098220521274, + "grad_norm": 0.845880389213562, + "learning_rate": 0.0008007339757956074, + "loss": 0.4512, + "step": 7226 + }, + { + "epoch": 0.4037768528088946, + "grad_norm": 0.4183385372161865, + "learning_rate": 0.0008007059614522636, + "loss": 0.3954, + "step": 7227 + }, + { + "epoch": 0.4038327234125765, + "grad_norm": 2.356472969055176, + "learning_rate": 0.0008006779471089198, + "loss": 0.5059, + "step": 7228 + }, + { + "epoch": 0.4038885940162584, + "grad_norm": 0.5289495587348938, + "learning_rate": 0.000800649932765576, + "loss": 0.5875, + "step": 7229 + }, + { + "epoch": 0.4039444646199402, + "grad_norm": 0.39848724007606506, + "learning_rate": 0.0008006219184222322, + "loss": 0.3992, + "step": 7230 + }, + { + "epoch": 0.4040003352236221, + "grad_norm": 0.4798985719680786, + "learning_rate": 0.0008005939040788884, + "loss": 0.4687, + "step": 7231 + }, + { + "epoch": 0.40405620582730395, + "grad_norm": 3.535616397857666, + "learning_rate": 0.0008005658897355445, + "loss": 0.4768, + "step": 7232 + }, + { + "epoch": 0.40411207643098584, + "grad_norm": 0.5190005302429199, + "learning_rate": 0.0008005378753922008, + "loss": 0.4387, + "step": 7233 + }, + { + "epoch": 0.4041679470346677, + "grad_norm": 0.6099901795387268, + "learning_rate": 0.000800509861048857, + "loss": 0.5761, + "step": 7234 + }, + { + "epoch": 0.4042238176383496, + "grad_norm": 0.5553138256072998, + "learning_rate": 0.0008004818467055132, + "loss": 0.4392, + "step": 7235 + }, + { + "epoch": 0.4042796882420315, + "grad_norm": 0.5428395867347717, + "learning_rate": 0.0008004538323621694, + "loss": 0.503, + "step": 7236 + }, + { + "epoch": 0.4043355588457133, + "grad_norm": 1.3277359008789062, + "learning_rate": 0.0008004258180188256, + "loss": 0.7194, + "step": 7237 + }, + { + "epoch": 0.4043914294493952, + "grad_norm": 0.7265246510505676, + "learning_rate": 0.0008003978036754819, + "loss": 0.4642, + "step": 7238 + }, + { + "epoch": 0.40444730005307705, + "grad_norm": 0.7072876691818237, + "learning_rate": 0.0008003697893321381, + "loss": 0.5462, + "step": 7239 + }, + { + "epoch": 0.40450317065675895, + "grad_norm": 0.8661325573921204, + "learning_rate": 0.0008003417749887943, + "loss": 0.5281, + "step": 7240 + }, + { + "epoch": 0.40455904126044084, + "grad_norm": 0.530582845211029, + "learning_rate": 0.0008003137606454505, + "loss": 0.5193, + "step": 7241 + }, + { + "epoch": 0.4046149118641227, + "grad_norm": 1.3013896942138672, + "learning_rate": 0.0008002857463021067, + "loss": 0.6449, + "step": 7242 + }, + { + "epoch": 0.4046707824678046, + "grad_norm": 0.4303540885448456, + "learning_rate": 0.000800257731958763, + "loss": 0.3867, + "step": 7243 + }, + { + "epoch": 0.4047266530714864, + "grad_norm": 0.5901951789855957, + "learning_rate": 0.0008002297176154191, + "loss": 0.5159, + "step": 7244 + }, + { + "epoch": 0.4047825236751683, + "grad_norm": 0.6293992400169373, + "learning_rate": 0.0008002017032720753, + "loss": 0.4931, + "step": 7245 + }, + { + "epoch": 0.40483839427885016, + "grad_norm": 0.44343316555023193, + "learning_rate": 0.0008001736889287315, + "loss": 0.3849, + "step": 7246 + }, + { + "epoch": 0.40489426488253205, + "grad_norm": 0.5195592045783997, + "learning_rate": 0.0008001456745853878, + "loss": 0.514, + "step": 7247 + }, + { + "epoch": 0.40495013548621395, + "grad_norm": 0.6387735605239868, + "learning_rate": 0.0008001176602420439, + "loss": 0.3597, + "step": 7248 + }, + { + "epoch": 0.4050060060898958, + "grad_norm": 1.533046841621399, + "learning_rate": 0.0008000896458987001, + "loss": 0.4026, + "step": 7249 + }, + { + "epoch": 0.4050618766935777, + "grad_norm": 0.6363122463226318, + "learning_rate": 0.0008000616315553564, + "loss": 0.3592, + "step": 7250 + }, + { + "epoch": 0.4051177472972595, + "grad_norm": 0.8003740906715393, + "learning_rate": 0.0008000336172120126, + "loss": 0.4607, + "step": 7251 + }, + { + "epoch": 0.4051736179009414, + "grad_norm": 0.5563868284225464, + "learning_rate": 0.0008000056028686687, + "loss": 0.3954, + "step": 7252 + }, + { + "epoch": 0.4052294885046233, + "grad_norm": 1.2920414209365845, + "learning_rate": 0.0007999775885253249, + "loss": 0.4653, + "step": 7253 + }, + { + "epoch": 0.40528535910830515, + "grad_norm": 0.8037341833114624, + "learning_rate": 0.0007999495741819813, + "loss": 0.3927, + "step": 7254 + }, + { + "epoch": 0.40534122971198705, + "grad_norm": 1.1311622858047485, + "learning_rate": 0.0007999215598386373, + "loss": 0.475, + "step": 7255 + }, + { + "epoch": 0.4053971003156689, + "grad_norm": 6.275296688079834, + "learning_rate": 0.0007998935454952936, + "loss": 0.4978, + "step": 7256 + }, + { + "epoch": 0.4054529709193508, + "grad_norm": 0.7145070433616638, + "learning_rate": 0.0007998655311519499, + "loss": 0.5174, + "step": 7257 + }, + { + "epoch": 0.4055088415230327, + "grad_norm": 0.6679105758666992, + "learning_rate": 0.0007998375168086061, + "loss": 0.5412, + "step": 7258 + }, + { + "epoch": 0.4055647121267145, + "grad_norm": 0.4610825777053833, + "learning_rate": 0.0007998095024652622, + "loss": 0.46, + "step": 7259 + }, + { + "epoch": 0.4056205827303964, + "grad_norm": 0.8428614735603333, + "learning_rate": 0.0007997814881219184, + "loss": 0.6042, + "step": 7260 + }, + { + "epoch": 0.40567645333407826, + "grad_norm": 0.709949791431427, + "learning_rate": 0.0007997534737785747, + "loss": 0.5288, + "step": 7261 + }, + { + "epoch": 0.40573232393776015, + "grad_norm": 0.6587054133415222, + "learning_rate": 0.0007997254594352309, + "loss": 0.4203, + "step": 7262 + }, + { + "epoch": 0.405788194541442, + "grad_norm": 0.5689655542373657, + "learning_rate": 0.000799697445091887, + "loss": 0.5686, + "step": 7263 + }, + { + "epoch": 0.4058440651451239, + "grad_norm": 0.7366065979003906, + "learning_rate": 0.0007996694307485433, + "loss": 0.4727, + "step": 7264 + }, + { + "epoch": 0.4058999357488058, + "grad_norm": 0.9383274912834167, + "learning_rate": 0.0007996414164051995, + "loss": 0.6796, + "step": 7265 + }, + { + "epoch": 0.4059558063524876, + "grad_norm": 0.9883775115013123, + "learning_rate": 0.0007996134020618557, + "loss": 0.4377, + "step": 7266 + }, + { + "epoch": 0.4060116769561695, + "grad_norm": 0.6392148733139038, + "learning_rate": 0.0007995853877185118, + "loss": 0.4302, + "step": 7267 + }, + { + "epoch": 0.40606754755985136, + "grad_norm": 0.905890703201294, + "learning_rate": 0.0007995573733751681, + "loss": 0.4168, + "step": 7268 + }, + { + "epoch": 0.40612341816353326, + "grad_norm": 0.4576319754123688, + "learning_rate": 0.0007995293590318243, + "loss": 0.4154, + "step": 7269 + }, + { + "epoch": 0.40617928876721515, + "grad_norm": 0.7958630323410034, + "learning_rate": 0.0007995013446884805, + "loss": 0.4688, + "step": 7270 + }, + { + "epoch": 0.406235159370897, + "grad_norm": 0.5564049482345581, + "learning_rate": 0.0007994733303451366, + "loss": 0.3635, + "step": 7271 + }, + { + "epoch": 0.4062910299745789, + "grad_norm": 0.9726436734199524, + "learning_rate": 0.000799445316001793, + "loss": 0.5199, + "step": 7272 + }, + { + "epoch": 0.40634690057826073, + "grad_norm": 0.4519191086292267, + "learning_rate": 0.0007994173016584492, + "loss": 0.3902, + "step": 7273 + }, + { + "epoch": 0.4064027711819426, + "grad_norm": 0.5917185544967651, + "learning_rate": 0.0007993892873151053, + "loss": 0.4585, + "step": 7274 + }, + { + "epoch": 0.4064586417856245, + "grad_norm": 0.5546278357505798, + "learning_rate": 0.0007993612729717616, + "loss": 0.4674, + "step": 7275 + }, + { + "epoch": 0.40651451238930636, + "grad_norm": 0.5514824390411377, + "learning_rate": 0.0007993332586284178, + "loss": 0.4774, + "step": 7276 + }, + { + "epoch": 0.40657038299298826, + "grad_norm": 0.5692644715309143, + "learning_rate": 0.000799305244285074, + "loss": 0.433, + "step": 7277 + }, + { + "epoch": 0.4066262535966701, + "grad_norm": 0.6580110192298889, + "learning_rate": 0.0007992772299417301, + "loss": 0.5463, + "step": 7278 + }, + { + "epoch": 0.406682124200352, + "grad_norm": 0.9670417308807373, + "learning_rate": 0.0007992492155983864, + "loss": 0.4812, + "step": 7279 + }, + { + "epoch": 0.40673799480403383, + "grad_norm": 1.0694148540496826, + "learning_rate": 0.0007992212012550426, + "loss": 0.46, + "step": 7280 + }, + { + "epoch": 0.4067938654077157, + "grad_norm": 0.5336366891860962, + "learning_rate": 0.0007991931869116988, + "loss": 0.4146, + "step": 7281 + }, + { + "epoch": 0.4068497360113976, + "grad_norm": 0.46272164583206177, + "learning_rate": 0.000799165172568355, + "loss": 0.4929, + "step": 7282 + }, + { + "epoch": 0.40690560661507946, + "grad_norm": 0.7408347129821777, + "learning_rate": 0.0007991371582250112, + "loss": 0.5858, + "step": 7283 + }, + { + "epoch": 0.40696147721876136, + "grad_norm": 0.5362839698791504, + "learning_rate": 0.0007991091438816674, + "loss": 0.432, + "step": 7284 + }, + { + "epoch": 0.4070173478224432, + "grad_norm": 0.6249840259552002, + "learning_rate": 0.0007990811295383236, + "loss": 0.4501, + "step": 7285 + }, + { + "epoch": 0.4070732184261251, + "grad_norm": 0.4433291256427765, + "learning_rate": 0.0007990531151949798, + "loss": 0.482, + "step": 7286 + }, + { + "epoch": 0.407129089029807, + "grad_norm": 0.42593035101890564, + "learning_rate": 0.000799025100851636, + "loss": 0.3816, + "step": 7287 + }, + { + "epoch": 0.40718495963348883, + "grad_norm": 0.47582343220710754, + "learning_rate": 0.0007989970865082922, + "loss": 0.5322, + "step": 7288 + }, + { + "epoch": 0.4072408302371707, + "grad_norm": 0.4629213809967041, + "learning_rate": 0.0007989690721649486, + "loss": 0.505, + "step": 7289 + }, + { + "epoch": 0.40729670084085257, + "grad_norm": 0.6807497143745422, + "learning_rate": 0.0007989410578216047, + "loss": 0.4203, + "step": 7290 + }, + { + "epoch": 0.40735257144453446, + "grad_norm": 1.2047176361083984, + "learning_rate": 0.0007989130434782609, + "loss": 0.5709, + "step": 7291 + }, + { + "epoch": 0.40740844204821636, + "grad_norm": 0.5300371050834656, + "learning_rate": 0.0007988850291349171, + "loss": 0.3921, + "step": 7292 + }, + { + "epoch": 0.4074643126518982, + "grad_norm": 0.6668831706047058, + "learning_rate": 0.0007988570147915734, + "loss": 0.3777, + "step": 7293 + }, + { + "epoch": 0.4075201832555801, + "grad_norm": 0.8419818878173828, + "learning_rate": 0.0007988290004482295, + "loss": 0.5499, + "step": 7294 + }, + { + "epoch": 0.40757605385926193, + "grad_norm": 0.43385419249534607, + "learning_rate": 0.0007988009861048857, + "loss": 0.4389, + "step": 7295 + }, + { + "epoch": 0.40763192446294383, + "grad_norm": 0.93235182762146, + "learning_rate": 0.000798772971761542, + "loss": 0.6207, + "step": 7296 + }, + { + "epoch": 0.40768779506662567, + "grad_norm": 0.44813811779022217, + "learning_rate": 0.0007987449574181981, + "loss": 0.3941, + "step": 7297 + }, + { + "epoch": 0.40774366567030756, + "grad_norm": 6.153847694396973, + "learning_rate": 0.0007987169430748543, + "loss": 0.5157, + "step": 7298 + }, + { + "epoch": 0.40779953627398946, + "grad_norm": 0.3824983537197113, + "learning_rate": 0.0007986889287315105, + "loss": 0.4126, + "step": 7299 + }, + { + "epoch": 0.4078554068776713, + "grad_norm": 0.6154986023902893, + "learning_rate": 0.0007986609143881668, + "loss": 0.4255, + "step": 7300 + }, + { + "epoch": 0.4079112774813532, + "grad_norm": 0.8405534625053406, + "learning_rate": 0.0007986329000448229, + "loss": 0.5395, + "step": 7301 + }, + { + "epoch": 0.40796714808503504, + "grad_norm": 6.053187370300293, + "learning_rate": 0.0007986048857014791, + "loss": 0.6695, + "step": 7302 + }, + { + "epoch": 0.40802301868871693, + "grad_norm": 1.143372893333435, + "learning_rate": 0.0007985768713581354, + "loss": 0.4777, + "step": 7303 + }, + { + "epoch": 0.4080788892923988, + "grad_norm": 0.8155447840690613, + "learning_rate": 0.0007985488570147917, + "loss": 0.5187, + "step": 7304 + }, + { + "epoch": 0.40813475989608067, + "grad_norm": 0.5386424660682678, + "learning_rate": 0.0007985208426714477, + "loss": 0.4549, + "step": 7305 + }, + { + "epoch": 0.40819063049976256, + "grad_norm": 0.6848695278167725, + "learning_rate": 0.000798492828328104, + "loss": 0.4351, + "step": 7306 + }, + { + "epoch": 0.4082465011034444, + "grad_norm": 0.6646403670310974, + "learning_rate": 0.0007984648139847603, + "loss": 0.5163, + "step": 7307 + }, + { + "epoch": 0.4083023717071263, + "grad_norm": 1.5901843309402466, + "learning_rate": 0.0007984367996414165, + "loss": 0.5681, + "step": 7308 + }, + { + "epoch": 0.4083582423108082, + "grad_norm": 0.5681113600730896, + "learning_rate": 0.0007984087852980726, + "loss": 0.5055, + "step": 7309 + }, + { + "epoch": 0.40841411291449004, + "grad_norm": 0.5612027049064636, + "learning_rate": 0.0007983807709547288, + "loss": 0.4138, + "step": 7310 + }, + { + "epoch": 0.40846998351817193, + "grad_norm": 0.6554590463638306, + "learning_rate": 0.0007983527566113851, + "loss": 0.4413, + "step": 7311 + }, + { + "epoch": 0.40852585412185377, + "grad_norm": 0.46362826228141785, + "learning_rate": 0.0007983247422680413, + "loss": 0.4139, + "step": 7312 + }, + { + "epoch": 0.40858172472553567, + "grad_norm": 0.531375527381897, + "learning_rate": 0.0007982967279246974, + "loss": 0.4029, + "step": 7313 + }, + { + "epoch": 0.4086375953292175, + "grad_norm": 4.001208782196045, + "learning_rate": 0.0007982687135813537, + "loss": 0.4639, + "step": 7314 + }, + { + "epoch": 0.4086934659328994, + "grad_norm": 0.47802942991256714, + "learning_rate": 0.0007982406992380099, + "loss": 0.5925, + "step": 7315 + }, + { + "epoch": 0.4087493365365813, + "grad_norm": 1.0980303287506104, + "learning_rate": 0.000798212684894666, + "loss": 0.4969, + "step": 7316 + }, + { + "epoch": 0.40880520714026314, + "grad_norm": 0.4626041054725647, + "learning_rate": 0.0007981846705513222, + "loss": 0.4219, + "step": 7317 + }, + { + "epoch": 0.40886107774394503, + "grad_norm": 0.42365318536758423, + "learning_rate": 0.0007981566562079785, + "loss": 0.3906, + "step": 7318 + }, + { + "epoch": 0.4089169483476269, + "grad_norm": 0.5960249900817871, + "learning_rate": 0.0007981286418646347, + "loss": 0.4485, + "step": 7319 + }, + { + "epoch": 0.40897281895130877, + "grad_norm": 0.7884456515312195, + "learning_rate": 0.0007981006275212908, + "loss": 0.3731, + "step": 7320 + }, + { + "epoch": 0.40902868955499067, + "grad_norm": 0.5946395397186279, + "learning_rate": 0.0007980726131779472, + "loss": 0.5777, + "step": 7321 + }, + { + "epoch": 0.4090845601586725, + "grad_norm": 3.926441192626953, + "learning_rate": 0.0007980445988346034, + "loss": 0.4411, + "step": 7322 + }, + { + "epoch": 0.4091404307623544, + "grad_norm": 0.41013291478157043, + "learning_rate": 0.0007980165844912596, + "loss": 0.414, + "step": 7323 + }, + { + "epoch": 0.40919630136603624, + "grad_norm": 0.6229344010353088, + "learning_rate": 0.0007979885701479157, + "loss": 0.528, + "step": 7324 + }, + { + "epoch": 0.40925217196971814, + "grad_norm": 0.4366622865200043, + "learning_rate": 0.000797960555804572, + "loss": 0.5164, + "step": 7325 + }, + { + "epoch": 0.40930804257340003, + "grad_norm": 0.5889489650726318, + "learning_rate": 0.0007979325414612282, + "loss": 0.4713, + "step": 7326 + }, + { + "epoch": 0.4093639131770819, + "grad_norm": 0.36655446887016296, + "learning_rate": 0.0007979045271178844, + "loss": 0.3281, + "step": 7327 + }, + { + "epoch": 0.40941978378076377, + "grad_norm": 0.6175941824913025, + "learning_rate": 0.0007978765127745406, + "loss": 0.4339, + "step": 7328 + }, + { + "epoch": 0.4094756543844456, + "grad_norm": 0.8622004985809326, + "learning_rate": 0.0007978484984311968, + "loss": 0.4318, + "step": 7329 + }, + { + "epoch": 0.4095315249881275, + "grad_norm": 0.5395911335945129, + "learning_rate": 0.000797820484087853, + "loss": 0.4792, + "step": 7330 + }, + { + "epoch": 0.40958739559180934, + "grad_norm": 1.3475881814956665, + "learning_rate": 0.0007977924697445092, + "loss": 0.4691, + "step": 7331 + }, + { + "epoch": 0.40964326619549124, + "grad_norm": 0.650614321231842, + "learning_rate": 0.0007977644554011654, + "loss": 0.4367, + "step": 7332 + }, + { + "epoch": 0.40969913679917314, + "grad_norm": 1.0451488494873047, + "learning_rate": 0.0007977364410578216, + "loss": 0.7155, + "step": 7333 + }, + { + "epoch": 0.409755007402855, + "grad_norm": 0.5434666275978088, + "learning_rate": 0.0007977084267144778, + "loss": 0.4593, + "step": 7334 + }, + { + "epoch": 0.40981087800653687, + "grad_norm": 0.6027839183807373, + "learning_rate": 0.0007976804123711341, + "loss": 0.3934, + "step": 7335 + }, + { + "epoch": 0.4098667486102187, + "grad_norm": 0.40781882405281067, + "learning_rate": 0.0007976523980277902, + "loss": 0.3956, + "step": 7336 + }, + { + "epoch": 0.4099226192139006, + "grad_norm": 0.3998832404613495, + "learning_rate": 0.0007976243836844464, + "loss": 0.4444, + "step": 7337 + }, + { + "epoch": 0.4099784898175825, + "grad_norm": 0.634172260761261, + "learning_rate": 0.0007975963693411026, + "loss": 0.5018, + "step": 7338 + }, + { + "epoch": 0.41003436042126434, + "grad_norm": 0.9627041220664978, + "learning_rate": 0.0007975683549977589, + "loss": 0.4308, + "step": 7339 + }, + { + "epoch": 0.41009023102494624, + "grad_norm": 0.480617880821228, + "learning_rate": 0.0007975403406544151, + "loss": 0.4568, + "step": 7340 + }, + { + "epoch": 0.4101461016286281, + "grad_norm": 0.6346330642700195, + "learning_rate": 0.0007975123263110713, + "loss": 0.4432, + "step": 7341 + }, + { + "epoch": 0.41020197223231, + "grad_norm": 0.4500509202480316, + "learning_rate": 0.0007974843119677276, + "loss": 0.5884, + "step": 7342 + }, + { + "epoch": 0.41025784283599187, + "grad_norm": 4.6128339767456055, + "learning_rate": 0.0007974562976243837, + "loss": 0.4733, + "step": 7343 + }, + { + "epoch": 0.4103137134396737, + "grad_norm": 0.41268059611320496, + "learning_rate": 0.0007974282832810399, + "loss": 0.3678, + "step": 7344 + }, + { + "epoch": 0.4103695840433556, + "grad_norm": 0.7860644459724426, + "learning_rate": 0.0007974002689376961, + "loss": 0.4488, + "step": 7345 + }, + { + "epoch": 0.41042545464703745, + "grad_norm": 0.49355819821357727, + "learning_rate": 0.0007973722545943524, + "loss": 0.4094, + "step": 7346 + }, + { + "epoch": 0.41048132525071934, + "grad_norm": 0.39029210805892944, + "learning_rate": 0.0007973442402510085, + "loss": 0.402, + "step": 7347 + }, + { + "epoch": 0.4105371958544012, + "grad_norm": 0.5204662084579468, + "learning_rate": 0.0007973162259076647, + "loss": 0.5501, + "step": 7348 + }, + { + "epoch": 0.4105930664580831, + "grad_norm": 1.0078450441360474, + "learning_rate": 0.0007972882115643209, + "loss": 0.4337, + "step": 7349 + }, + { + "epoch": 0.410648937061765, + "grad_norm": 0.6721319556236267, + "learning_rate": 0.0007972601972209772, + "loss": 0.4624, + "step": 7350 + }, + { + "epoch": 0.4107048076654468, + "grad_norm": 0.48497632145881653, + "learning_rate": 0.0007972321828776333, + "loss": 0.4229, + "step": 7351 + }, + { + "epoch": 0.4107606782691287, + "grad_norm": 0.4659441113471985, + "learning_rate": 0.0007972041685342895, + "loss": 0.4623, + "step": 7352 + }, + { + "epoch": 0.41081654887281055, + "grad_norm": 0.6055890321731567, + "learning_rate": 0.0007971761541909458, + "loss": 0.4916, + "step": 7353 + }, + { + "epoch": 0.41087241947649245, + "grad_norm": 0.41717827320098877, + "learning_rate": 0.000797148139847602, + "loss": 0.4124, + "step": 7354 + }, + { + "epoch": 0.41092829008017434, + "grad_norm": 0.5366477966308594, + "learning_rate": 0.0007971201255042581, + "loss": 0.4271, + "step": 7355 + }, + { + "epoch": 0.4109841606838562, + "grad_norm": 0.761496901512146, + "learning_rate": 0.0007970921111609144, + "loss": 0.5111, + "step": 7356 + }, + { + "epoch": 0.4110400312875381, + "grad_norm": 0.5319584012031555, + "learning_rate": 0.0007970640968175707, + "loss": 0.451, + "step": 7357 + }, + { + "epoch": 0.4110959018912199, + "grad_norm": 0.48750853538513184, + "learning_rate": 0.0007970360824742269, + "loss": 0.4212, + "step": 7358 + }, + { + "epoch": 0.4111517724949018, + "grad_norm": 0.7592357397079468, + "learning_rate": 0.000797008068130883, + "loss": 0.6041, + "step": 7359 + }, + { + "epoch": 0.41120764309858365, + "grad_norm": 0.43445178866386414, + "learning_rate": 0.0007969800537875393, + "loss": 0.5054, + "step": 7360 + }, + { + "epoch": 0.41126351370226555, + "grad_norm": 0.6225228905677795, + "learning_rate": 0.0007969520394441955, + "loss": 0.4403, + "step": 7361 + }, + { + "epoch": 0.41131938430594744, + "grad_norm": 0.5137776136398315, + "learning_rate": 0.0007969240251008516, + "loss": 0.5683, + "step": 7362 + }, + { + "epoch": 0.4113752549096293, + "grad_norm": 0.40974608063697815, + "learning_rate": 0.0007968960107575078, + "loss": 0.4258, + "step": 7363 + }, + { + "epoch": 0.4114311255133112, + "grad_norm": 0.6849403381347656, + "learning_rate": 0.0007968679964141641, + "loss": 0.4748, + "step": 7364 + }, + { + "epoch": 0.411486996116993, + "grad_norm": 1.9418227672576904, + "learning_rate": 0.0007968399820708203, + "loss": 0.5061, + "step": 7365 + }, + { + "epoch": 0.4115428667206749, + "grad_norm": 0.41972726583480835, + "learning_rate": 0.0007968119677274764, + "loss": 0.4507, + "step": 7366 + }, + { + "epoch": 0.4115987373243568, + "grad_norm": 0.45774099230766296, + "learning_rate": 0.0007967839533841327, + "loss": 0.4267, + "step": 7367 + }, + { + "epoch": 0.41165460792803865, + "grad_norm": 0.4839518070220947, + "learning_rate": 0.0007967559390407889, + "loss": 0.4331, + "step": 7368 + }, + { + "epoch": 0.41171047853172055, + "grad_norm": 1.7802921533584595, + "learning_rate": 0.0007967279246974451, + "loss": 0.3693, + "step": 7369 + }, + { + "epoch": 0.4117663491354024, + "grad_norm": 1.0773507356643677, + "learning_rate": 0.0007966999103541012, + "loss": 0.7202, + "step": 7370 + }, + { + "epoch": 0.4118222197390843, + "grad_norm": 0.5298578143119812, + "learning_rate": 0.0007966718960107575, + "loss": 0.4444, + "step": 7371 + }, + { + "epoch": 0.4118780903427662, + "grad_norm": 0.6861273050308228, + "learning_rate": 0.0007966438816674138, + "loss": 0.6245, + "step": 7372 + }, + { + "epoch": 0.411933960946448, + "grad_norm": 1.2633931636810303, + "learning_rate": 0.00079661586732407, + "loss": 0.5432, + "step": 7373 + }, + { + "epoch": 0.4119898315501299, + "grad_norm": 0.7048432230949402, + "learning_rate": 0.0007965878529807262, + "loss": 0.4215, + "step": 7374 + }, + { + "epoch": 0.41204570215381175, + "grad_norm": 0.402104914188385, + "learning_rate": 0.0007965598386373824, + "loss": 0.3977, + "step": 7375 + }, + { + "epoch": 0.41210157275749365, + "grad_norm": 0.48883765935897827, + "learning_rate": 0.0007965318242940386, + "loss": 0.4803, + "step": 7376 + }, + { + "epoch": 0.4121574433611755, + "grad_norm": 1.0407682657241821, + "learning_rate": 0.0007965038099506948, + "loss": 0.4435, + "step": 7377 + }, + { + "epoch": 0.4122133139648574, + "grad_norm": 0.48395299911499023, + "learning_rate": 0.000796475795607351, + "loss": 0.4457, + "step": 7378 + }, + { + "epoch": 0.4122691845685393, + "grad_norm": 0.4650857448577881, + "learning_rate": 0.0007964477812640072, + "loss": 0.4498, + "step": 7379 + }, + { + "epoch": 0.4123250551722211, + "grad_norm": 0.5360819697380066, + "learning_rate": 0.0007964197669206634, + "loss": 0.4535, + "step": 7380 + }, + { + "epoch": 0.412380925775903, + "grad_norm": 0.5488067269325256, + "learning_rate": 0.0007963917525773195, + "loss": 0.4702, + "step": 7381 + }, + { + "epoch": 0.41243679637958486, + "grad_norm": 0.5401497483253479, + "learning_rate": 0.0007963637382339758, + "loss": 0.5025, + "step": 7382 + }, + { + "epoch": 0.41249266698326675, + "grad_norm": 0.9952354431152344, + "learning_rate": 0.000796335723890632, + "loss": 0.4465, + "step": 7383 + }, + { + "epoch": 0.41254853758694865, + "grad_norm": 0.4035298824310303, + "learning_rate": 0.0007963077095472882, + "loss": 0.4765, + "step": 7384 + }, + { + "epoch": 0.4126044081906305, + "grad_norm": 0.5940676927566528, + "learning_rate": 0.0007962796952039444, + "loss": 0.4639, + "step": 7385 + }, + { + "epoch": 0.4126602787943124, + "grad_norm": 0.6619634032249451, + "learning_rate": 0.0007962516808606006, + "loss": 0.4446, + "step": 7386 + }, + { + "epoch": 0.4127161493979942, + "grad_norm": 0.5255485773086548, + "learning_rate": 0.0007962236665172568, + "loss": 0.4938, + "step": 7387 + }, + { + "epoch": 0.4127720200016761, + "grad_norm": 0.5367745757102966, + "learning_rate": 0.000796195652173913, + "loss": 0.4965, + "step": 7388 + }, + { + "epoch": 0.412827890605358, + "grad_norm": 0.6245740056037903, + "learning_rate": 0.0007961676378305693, + "loss": 0.5726, + "step": 7389 + }, + { + "epoch": 0.41288376120903986, + "grad_norm": 0.4335363507270813, + "learning_rate": 0.0007961396234872255, + "loss": 0.3604, + "step": 7390 + }, + { + "epoch": 0.41293963181272175, + "grad_norm": 0.7676070928573608, + "learning_rate": 0.0007961116091438817, + "loss": 0.4631, + "step": 7391 + }, + { + "epoch": 0.4129955024164036, + "grad_norm": 0.4489234387874603, + "learning_rate": 0.000796083594800538, + "loss": 0.4344, + "step": 7392 + }, + { + "epoch": 0.4130513730200855, + "grad_norm": 0.8555400371551514, + "learning_rate": 0.0007960555804571941, + "loss": 0.4437, + "step": 7393 + }, + { + "epoch": 0.41310724362376733, + "grad_norm": 0.5603752732276917, + "learning_rate": 0.0007960275661138503, + "loss": 0.6339, + "step": 7394 + }, + { + "epoch": 0.4131631142274492, + "grad_norm": 0.6124736070632935, + "learning_rate": 0.0007959995517705065, + "loss": 0.3792, + "step": 7395 + }, + { + "epoch": 0.4132189848311311, + "grad_norm": 0.5264571309089661, + "learning_rate": 0.0007959715374271628, + "loss": 0.3537, + "step": 7396 + }, + { + "epoch": 0.41327485543481296, + "grad_norm": 0.700009286403656, + "learning_rate": 0.0007959435230838189, + "loss": 0.5748, + "step": 7397 + }, + { + "epoch": 0.41333072603849486, + "grad_norm": 0.4320071041584015, + "learning_rate": 0.0007959155087404751, + "loss": 0.3746, + "step": 7398 + }, + { + "epoch": 0.4133865966421767, + "grad_norm": 0.6614438891410828, + "learning_rate": 0.0007958874943971314, + "loss": 0.4931, + "step": 7399 + }, + { + "epoch": 0.4134424672458586, + "grad_norm": 0.3816594183444977, + "learning_rate": 0.0007958594800537876, + "loss": 0.4011, + "step": 7400 + }, + { + "epoch": 0.4134983378495405, + "grad_norm": 1.0405455827713013, + "learning_rate": 0.0007958314657104437, + "loss": 0.5786, + "step": 7401 + }, + { + "epoch": 0.4135542084532223, + "grad_norm": 0.5190457701683044, + "learning_rate": 0.0007958034513670999, + "loss": 0.5065, + "step": 7402 + }, + { + "epoch": 0.4136100790569042, + "grad_norm": 0.4316205382347107, + "learning_rate": 0.0007957754370237562, + "loss": 0.5724, + "step": 7403 + }, + { + "epoch": 0.41366594966058606, + "grad_norm": 0.5989258885383606, + "learning_rate": 0.0007957474226804123, + "loss": 0.471, + "step": 7404 + }, + { + "epoch": 0.41372182026426796, + "grad_norm": 1.3919334411621094, + "learning_rate": 0.0007957194083370685, + "loss": 0.5183, + "step": 7405 + }, + { + "epoch": 0.41377769086794985, + "grad_norm": 2.4555342197418213, + "learning_rate": 0.0007956913939937249, + "loss": 0.3785, + "step": 7406 + }, + { + "epoch": 0.4138335614716317, + "grad_norm": 0.5988123416900635, + "learning_rate": 0.0007956633796503811, + "loss": 0.4334, + "step": 7407 + }, + { + "epoch": 0.4138894320753136, + "grad_norm": 0.44483232498168945, + "learning_rate": 0.0007956353653070372, + "loss": 0.4819, + "step": 7408 + }, + { + "epoch": 0.41394530267899543, + "grad_norm": 0.6447468996047974, + "learning_rate": 0.0007956073509636934, + "loss": 0.5585, + "step": 7409 + }, + { + "epoch": 0.4140011732826773, + "grad_norm": 0.6913833022117615, + "learning_rate": 0.0007955793366203497, + "loss": 0.5371, + "step": 7410 + }, + { + "epoch": 0.41405704388635917, + "grad_norm": 0.6886458396911621, + "learning_rate": 0.0007955513222770059, + "loss": 0.464, + "step": 7411 + }, + { + "epoch": 0.41411291449004106, + "grad_norm": 0.6045411825180054, + "learning_rate": 0.000795523307933662, + "loss": 0.3901, + "step": 7412 + }, + { + "epoch": 0.41416878509372296, + "grad_norm": 0.751338541507721, + "learning_rate": 0.0007954952935903182, + "loss": 0.5198, + "step": 7413 + }, + { + "epoch": 0.4142246556974048, + "grad_norm": 0.5967824459075928, + "learning_rate": 0.0007954672792469745, + "loss": 0.4614, + "step": 7414 + }, + { + "epoch": 0.4142805263010867, + "grad_norm": 0.669353723526001, + "learning_rate": 0.0007954392649036307, + "loss": 0.4118, + "step": 7415 + }, + { + "epoch": 0.41433639690476853, + "grad_norm": 0.500142514705658, + "learning_rate": 0.0007954112505602868, + "loss": 0.6011, + "step": 7416 + }, + { + "epoch": 0.41439226750845043, + "grad_norm": 0.9639164805412292, + "learning_rate": 0.0007953832362169431, + "loss": 0.4341, + "step": 7417 + }, + { + "epoch": 0.4144481381121323, + "grad_norm": 0.9428437352180481, + "learning_rate": 0.0007953552218735993, + "loss": 0.5075, + "step": 7418 + }, + { + "epoch": 0.41450400871581417, + "grad_norm": 0.5915223360061646, + "learning_rate": 0.0007953272075302555, + "loss": 0.4795, + "step": 7419 + }, + { + "epoch": 0.41455987931949606, + "grad_norm": 0.7575157880783081, + "learning_rate": 0.0007952991931869116, + "loss": 0.3962, + "step": 7420 + }, + { + "epoch": 0.4146157499231779, + "grad_norm": 0.703601598739624, + "learning_rate": 0.000795271178843568, + "loss": 0.4448, + "step": 7421 + }, + { + "epoch": 0.4146716205268598, + "grad_norm": 0.5531327128410339, + "learning_rate": 0.0007952431645002242, + "loss": 0.3072, + "step": 7422 + }, + { + "epoch": 0.4147274911305417, + "grad_norm": 0.5098794102668762, + "learning_rate": 0.0007952151501568803, + "loss": 0.5749, + "step": 7423 + }, + { + "epoch": 0.41478336173422353, + "grad_norm": 0.5209490656852722, + "learning_rate": 0.0007951871358135366, + "loss": 0.4515, + "step": 7424 + }, + { + "epoch": 0.41483923233790543, + "grad_norm": 0.628422200679779, + "learning_rate": 0.0007951591214701928, + "loss": 0.4498, + "step": 7425 + }, + { + "epoch": 0.41489510294158727, + "grad_norm": 0.8758570551872253, + "learning_rate": 0.000795131107126849, + "loss": 0.4336, + "step": 7426 + }, + { + "epoch": 0.41495097354526916, + "grad_norm": 0.5288830995559692, + "learning_rate": 0.0007951030927835051, + "loss": 0.5209, + "step": 7427 + }, + { + "epoch": 0.415006844148951, + "grad_norm": 0.6451472640037537, + "learning_rate": 0.0007950750784401614, + "loss": 0.4947, + "step": 7428 + }, + { + "epoch": 0.4150627147526329, + "grad_norm": 0.5947410464286804, + "learning_rate": 0.0007950470640968176, + "loss": 0.4398, + "step": 7429 + }, + { + "epoch": 0.4151185853563148, + "grad_norm": 0.4635494649410248, + "learning_rate": 0.0007950190497534738, + "loss": 0.5287, + "step": 7430 + }, + { + "epoch": 0.41517445595999664, + "grad_norm": 0.41025254130363464, + "learning_rate": 0.00079499103541013, + "loss": 0.4702, + "step": 7431 + }, + { + "epoch": 0.41523032656367853, + "grad_norm": 0.4618853032588959, + "learning_rate": 0.0007949630210667862, + "loss": 0.446, + "step": 7432 + }, + { + "epoch": 0.41528619716736037, + "grad_norm": 0.5825926065444946, + "learning_rate": 0.0007949350067234424, + "loss": 0.604, + "step": 7433 + }, + { + "epoch": 0.41534206777104227, + "grad_norm": 0.6515415906906128, + "learning_rate": 0.0007949069923800986, + "loss": 0.5096, + "step": 7434 + }, + { + "epoch": 0.41539793837472416, + "grad_norm": 0.6207972764968872, + "learning_rate": 0.0007948789780367548, + "loss": 0.5552, + "step": 7435 + }, + { + "epoch": 0.415453808978406, + "grad_norm": 0.46532493829727173, + "learning_rate": 0.000794850963693411, + "loss": 0.3873, + "step": 7436 + }, + { + "epoch": 0.4155096795820879, + "grad_norm": 0.7189059853553772, + "learning_rate": 0.0007948229493500672, + "loss": 0.3971, + "step": 7437 + }, + { + "epoch": 0.41556555018576974, + "grad_norm": 1.5672879219055176, + "learning_rate": 0.0007947949350067236, + "loss": 0.5315, + "step": 7438 + }, + { + "epoch": 0.41562142078945163, + "grad_norm": 3.5705080032348633, + "learning_rate": 0.0007947669206633797, + "loss": 0.5249, + "step": 7439 + }, + { + "epoch": 0.41567729139313353, + "grad_norm": 0.45453619956970215, + "learning_rate": 0.0007947389063200359, + "loss": 0.4713, + "step": 7440 + }, + { + "epoch": 0.41573316199681537, + "grad_norm": 0.47521886229515076, + "learning_rate": 0.0007947108919766921, + "loss": 0.4706, + "step": 7441 + }, + { + "epoch": 0.41578903260049727, + "grad_norm": 0.5168255567550659, + "learning_rate": 0.0007946828776333484, + "loss": 0.4364, + "step": 7442 + }, + { + "epoch": 0.4158449032041791, + "grad_norm": 0.865293562412262, + "learning_rate": 0.0007946548632900045, + "loss": 0.4057, + "step": 7443 + }, + { + "epoch": 0.415900773807861, + "grad_norm": 1.5949372053146362, + "learning_rate": 0.0007946268489466607, + "loss": 0.5258, + "step": 7444 + }, + { + "epoch": 0.41595664441154284, + "grad_norm": 2.190420389175415, + "learning_rate": 0.000794598834603317, + "loss": 0.6487, + "step": 7445 + }, + { + "epoch": 0.41601251501522474, + "grad_norm": 0.6028539538383484, + "learning_rate": 0.0007945708202599731, + "loss": 0.4522, + "step": 7446 + }, + { + "epoch": 0.41606838561890663, + "grad_norm": 0.44147440791130066, + "learning_rate": 0.0007945428059166293, + "loss": 0.5089, + "step": 7447 + }, + { + "epoch": 0.4161242562225885, + "grad_norm": 0.6065190434455872, + "learning_rate": 0.0007945147915732855, + "loss": 0.4505, + "step": 7448 + }, + { + "epoch": 0.41618012682627037, + "grad_norm": 0.6282077431678772, + "learning_rate": 0.0007944867772299418, + "loss": 0.6746, + "step": 7449 + }, + { + "epoch": 0.4162359974299522, + "grad_norm": 0.49435845017433167, + "learning_rate": 0.0007944587628865979, + "loss": 0.5822, + "step": 7450 + }, + { + "epoch": 0.4162918680336341, + "grad_norm": 0.38725602626800537, + "learning_rate": 0.0007944307485432541, + "loss": 0.4234, + "step": 7451 + }, + { + "epoch": 0.416347738637316, + "grad_norm": 1.9893271923065186, + "learning_rate": 0.0007944027341999103, + "loss": 0.3722, + "step": 7452 + }, + { + "epoch": 0.41640360924099784, + "grad_norm": 0.5186235904693604, + "learning_rate": 0.0007943747198565666, + "loss": 0.4069, + "step": 7453 + }, + { + "epoch": 0.41645947984467974, + "grad_norm": 0.5328706502914429, + "learning_rate": 0.0007943467055132227, + "loss": 0.5014, + "step": 7454 + }, + { + "epoch": 0.4165153504483616, + "grad_norm": 0.4973846673965454, + "learning_rate": 0.000794318691169879, + "loss": 0.4858, + "step": 7455 + }, + { + "epoch": 0.41657122105204347, + "grad_norm": 0.7880746126174927, + "learning_rate": 0.0007942906768265353, + "loss": 0.6194, + "step": 7456 + }, + { + "epoch": 0.41662709165572537, + "grad_norm": 0.7088366746902466, + "learning_rate": 0.0007942626624831915, + "loss": 0.4167, + "step": 7457 + }, + { + "epoch": 0.4166829622594072, + "grad_norm": 0.4730663597583771, + "learning_rate": 0.0007942346481398476, + "loss": 0.4988, + "step": 7458 + }, + { + "epoch": 0.4167388328630891, + "grad_norm": 0.7896215319633484, + "learning_rate": 0.0007942066337965038, + "loss": 0.4973, + "step": 7459 + }, + { + "epoch": 0.41679470346677094, + "grad_norm": 1.0006859302520752, + "learning_rate": 0.0007941786194531601, + "loss": 0.5672, + "step": 7460 + }, + { + "epoch": 0.41685057407045284, + "grad_norm": 4.826554298400879, + "learning_rate": 0.0007941506051098163, + "loss": 0.474, + "step": 7461 + }, + { + "epoch": 0.4169064446741347, + "grad_norm": 0.6643521785736084, + "learning_rate": 0.0007941225907664724, + "loss": 0.5917, + "step": 7462 + }, + { + "epoch": 0.4169623152778166, + "grad_norm": 0.9099168181419373, + "learning_rate": 0.0007940945764231287, + "loss": 0.5225, + "step": 7463 + }, + { + "epoch": 0.41701818588149847, + "grad_norm": 0.8093367218971252, + "learning_rate": 0.0007940665620797849, + "loss": 0.5619, + "step": 7464 + }, + { + "epoch": 0.4170740564851803, + "grad_norm": 0.9986276626586914, + "learning_rate": 0.000794038547736441, + "loss": 0.4021, + "step": 7465 + }, + { + "epoch": 0.4171299270888622, + "grad_norm": 0.4369891881942749, + "learning_rate": 0.0007940105333930972, + "loss": 0.5361, + "step": 7466 + }, + { + "epoch": 0.41718579769254405, + "grad_norm": 4.135344505310059, + "learning_rate": 0.0007939825190497535, + "loss": 0.4903, + "step": 7467 + }, + { + "epoch": 0.41724166829622594, + "grad_norm": 2.78395676612854, + "learning_rate": 0.0007939545047064097, + "loss": 0.5231, + "step": 7468 + }, + { + "epoch": 0.41729753889990784, + "grad_norm": 0.847667932510376, + "learning_rate": 0.0007939264903630658, + "loss": 0.4947, + "step": 7469 + }, + { + "epoch": 0.4173534095035897, + "grad_norm": 3.4946587085723877, + "learning_rate": 0.0007938984760197221, + "loss": 0.4724, + "step": 7470 + }, + { + "epoch": 0.4174092801072716, + "grad_norm": 1.0033974647521973, + "learning_rate": 0.0007938704616763783, + "loss": 0.4245, + "step": 7471 + }, + { + "epoch": 0.4174651507109534, + "grad_norm": 2.4381778240203857, + "learning_rate": 0.0007938424473330346, + "loss": 0.508, + "step": 7472 + }, + { + "epoch": 0.4175210213146353, + "grad_norm": 0.6341838240623474, + "learning_rate": 0.0007938144329896907, + "loss": 0.3749, + "step": 7473 + }, + { + "epoch": 0.41757689191831715, + "grad_norm": 2.553955554962158, + "learning_rate": 0.000793786418646347, + "loss": 0.3987, + "step": 7474 + }, + { + "epoch": 0.41763276252199905, + "grad_norm": 0.5246524214744568, + "learning_rate": 0.0007937584043030032, + "loss": 0.3962, + "step": 7475 + }, + { + "epoch": 0.41768863312568094, + "grad_norm": 0.6862634420394897, + "learning_rate": 0.0007937303899596594, + "loss": 0.4007, + "step": 7476 + }, + { + "epoch": 0.4177445037293628, + "grad_norm": 0.5762678980827332, + "learning_rate": 0.0007937023756163156, + "loss": 0.4558, + "step": 7477 + }, + { + "epoch": 0.4178003743330447, + "grad_norm": 0.6950821876525879, + "learning_rate": 0.0007936743612729718, + "loss": 0.3941, + "step": 7478 + }, + { + "epoch": 0.4178562449367265, + "grad_norm": 0.48215848207473755, + "learning_rate": 0.000793646346929628, + "loss": 0.4775, + "step": 7479 + }, + { + "epoch": 0.4179121155404084, + "grad_norm": 0.6278411149978638, + "learning_rate": 0.0007936183325862842, + "loss": 0.4759, + "step": 7480 + }, + { + "epoch": 0.4179679861440903, + "grad_norm": 0.5240369439125061, + "learning_rate": 0.0007935903182429404, + "loss": 0.448, + "step": 7481 + }, + { + "epoch": 0.41802385674777215, + "grad_norm": 0.8468273878097534, + "learning_rate": 0.0007935623038995966, + "loss": 0.5785, + "step": 7482 + }, + { + "epoch": 0.41807972735145404, + "grad_norm": 0.7094804644584656, + "learning_rate": 0.0007935342895562528, + "loss": 0.505, + "step": 7483 + }, + { + "epoch": 0.4181355979551359, + "grad_norm": 0.7669112682342529, + "learning_rate": 0.0007935062752129091, + "loss": 0.5376, + "step": 7484 + }, + { + "epoch": 0.4181914685588178, + "grad_norm": 0.5932039022445679, + "learning_rate": 0.0007934782608695652, + "loss": 0.4318, + "step": 7485 + }, + { + "epoch": 0.4182473391624997, + "grad_norm": 0.5540536046028137, + "learning_rate": 0.0007934502465262214, + "loss": 0.3848, + "step": 7486 + }, + { + "epoch": 0.4183032097661815, + "grad_norm": 0.49922171235084534, + "learning_rate": 0.0007934222321828776, + "loss": 0.4329, + "step": 7487 + }, + { + "epoch": 0.4183590803698634, + "grad_norm": 1.4219698905944824, + "learning_rate": 0.0007933942178395338, + "loss": 0.4037, + "step": 7488 + }, + { + "epoch": 0.41841495097354525, + "grad_norm": 1.1548793315887451, + "learning_rate": 0.00079336620349619, + "loss": 0.4791, + "step": 7489 + }, + { + "epoch": 0.41847082157722715, + "grad_norm": 0.5606076717376709, + "learning_rate": 0.0007933381891528463, + "loss": 0.4979, + "step": 7490 + }, + { + "epoch": 0.418526692180909, + "grad_norm": 0.540451169013977, + "learning_rate": 0.0007933101748095025, + "loss": 0.5508, + "step": 7491 + }, + { + "epoch": 0.4185825627845909, + "grad_norm": 0.7175434827804565, + "learning_rate": 0.0007932821604661587, + "loss": 0.4929, + "step": 7492 + }, + { + "epoch": 0.4186384333882728, + "grad_norm": 0.6486732363700867, + "learning_rate": 0.0007932541461228149, + "loss": 0.5188, + "step": 7493 + }, + { + "epoch": 0.4186943039919546, + "grad_norm": 0.5147367119789124, + "learning_rate": 0.0007932261317794711, + "loss": 0.4134, + "step": 7494 + }, + { + "epoch": 0.4187501745956365, + "grad_norm": 0.46374136209487915, + "learning_rate": 0.0007931981174361274, + "loss": 0.4544, + "step": 7495 + }, + { + "epoch": 0.41880604519931836, + "grad_norm": 0.5849515199661255, + "learning_rate": 0.0007931701030927835, + "loss": 0.42, + "step": 7496 + }, + { + "epoch": 0.41886191580300025, + "grad_norm": 0.42844462394714355, + "learning_rate": 0.0007931420887494397, + "loss": 0.3976, + "step": 7497 + }, + { + "epoch": 0.41891778640668215, + "grad_norm": 0.8546964526176453, + "learning_rate": 0.0007931140744060959, + "loss": 0.483, + "step": 7498 + }, + { + "epoch": 0.418973657010364, + "grad_norm": 1.042801856994629, + "learning_rate": 0.0007930860600627522, + "loss": 0.4293, + "step": 7499 + }, + { + "epoch": 0.4190295276140459, + "grad_norm": 0.587601900100708, + "learning_rate": 0.0007930580457194083, + "loss": 0.4454, + "step": 7500 + }, + { + "epoch": 0.4190295276140459, + "eval_cer": 0.09406675177026393, + "eval_loss": 0.3527156412601471, + "eval_runtime": 56.3899, + "eval_samples_per_second": 80.475, + "eval_steps_per_second": 5.036, + "eval_wer": 0.372196307365716, + "step": 7500 + }, + { + "epoch": 0.4190853982177277, + "grad_norm": 1.0335760116577148, + "learning_rate": 0.0007930300313760645, + "loss": 0.4347, + "step": 7501 + }, + { + "epoch": 0.4191412688214096, + "grad_norm": 0.6053768992424011, + "learning_rate": 0.0007930020170327208, + "loss": 0.4867, + "step": 7502 + }, + { + "epoch": 0.4191971394250915, + "grad_norm": 1.6739767789840698, + "learning_rate": 0.000792974002689377, + "loss": 0.5608, + "step": 7503 + }, + { + "epoch": 0.41925301002877335, + "grad_norm": 3.9701623916625977, + "learning_rate": 0.0007929459883460331, + "loss": 0.4915, + "step": 7504 + }, + { + "epoch": 0.41930888063245525, + "grad_norm": 1.1323524713516235, + "learning_rate": 0.0007929179740026893, + "loss": 0.3749, + "step": 7505 + }, + { + "epoch": 0.4193647512361371, + "grad_norm": 0.8780444264411926, + "learning_rate": 0.0007928899596593457, + "loss": 0.5082, + "step": 7506 + }, + { + "epoch": 0.419420621839819, + "grad_norm": 1.6464660167694092, + "learning_rate": 0.0007928619453160019, + "loss": 0.4747, + "step": 7507 + }, + { + "epoch": 0.4194764924435008, + "grad_norm": 0.5425184965133667, + "learning_rate": 0.000792833930972658, + "loss": 0.4534, + "step": 7508 + }, + { + "epoch": 0.4195323630471827, + "grad_norm": 1.2141865491867065, + "learning_rate": 0.0007928059166293143, + "loss": 0.4244, + "step": 7509 + }, + { + "epoch": 0.4195882336508646, + "grad_norm": 0.8190539479255676, + "learning_rate": 0.0007927779022859705, + "loss": 0.3973, + "step": 7510 + }, + { + "epoch": 0.41964410425454646, + "grad_norm": 0.6253194212913513, + "learning_rate": 0.0007927498879426266, + "loss": 0.4806, + "step": 7511 + }, + { + "epoch": 0.41969997485822835, + "grad_norm": 0.6336153745651245, + "learning_rate": 0.0007927218735992828, + "loss": 0.4186, + "step": 7512 + }, + { + "epoch": 0.4197558454619102, + "grad_norm": 0.696178138256073, + "learning_rate": 0.0007926938592559391, + "loss": 0.5288, + "step": 7513 + }, + { + "epoch": 0.4198117160655921, + "grad_norm": 1.617223858833313, + "learning_rate": 0.0007926658449125953, + "loss": 0.5492, + "step": 7514 + }, + { + "epoch": 0.419867586669274, + "grad_norm": 0.793877124786377, + "learning_rate": 0.0007926378305692514, + "loss": 0.517, + "step": 7515 + }, + { + "epoch": 0.4199234572729558, + "grad_norm": 1.1311171054840088, + "learning_rate": 0.0007926098162259077, + "loss": 0.4928, + "step": 7516 + }, + { + "epoch": 0.4199793278766377, + "grad_norm": 0.8464235663414001, + "learning_rate": 0.0007925818018825639, + "loss": 0.3491, + "step": 7517 + }, + { + "epoch": 0.42003519848031956, + "grad_norm": 1.5752507448196411, + "learning_rate": 0.0007925537875392201, + "loss": 0.4624, + "step": 7518 + }, + { + "epoch": 0.42009106908400146, + "grad_norm": 0.6557574272155762, + "learning_rate": 0.0007925257731958762, + "loss": 0.5117, + "step": 7519 + }, + { + "epoch": 0.42014693968768335, + "grad_norm": 0.5631721019744873, + "learning_rate": 0.0007924977588525325, + "loss": 0.5135, + "step": 7520 + }, + { + "epoch": 0.4202028102913652, + "grad_norm": 0.4959169626235962, + "learning_rate": 0.0007924697445091887, + "loss": 0.4502, + "step": 7521 + }, + { + "epoch": 0.4202586808950471, + "grad_norm": 0.6068229675292969, + "learning_rate": 0.000792441730165845, + "loss": 0.5163, + "step": 7522 + }, + { + "epoch": 0.4203145514987289, + "grad_norm": 0.7141276001930237, + "learning_rate": 0.000792413715822501, + "loss": 0.5221, + "step": 7523 + }, + { + "epoch": 0.4203704221024108, + "grad_norm": 0.5903587341308594, + "learning_rate": 0.0007923857014791574, + "loss": 0.576, + "step": 7524 + }, + { + "epoch": 0.42042629270609266, + "grad_norm": 0.4662926197052002, + "learning_rate": 0.0007923576871358136, + "loss": 0.4696, + "step": 7525 + }, + { + "epoch": 0.42048216330977456, + "grad_norm": 2.2158610820770264, + "learning_rate": 0.0007923296727924698, + "loss": 0.5771, + "step": 7526 + }, + { + "epoch": 0.42053803391345645, + "grad_norm": 4.52646541595459, + "learning_rate": 0.000792301658449126, + "loss": 0.462, + "step": 7527 + }, + { + "epoch": 0.4205939045171383, + "grad_norm": 0.39845988154411316, + "learning_rate": 0.0007922736441057822, + "loss": 0.447, + "step": 7528 + }, + { + "epoch": 0.4206497751208202, + "grad_norm": 0.4908253252506256, + "learning_rate": 0.0007922456297624384, + "loss": 0.5361, + "step": 7529 + }, + { + "epoch": 0.42070564572450203, + "grad_norm": 0.3701731562614441, + "learning_rate": 0.0007922176154190945, + "loss": 0.3451, + "step": 7530 + }, + { + "epoch": 0.4207615163281839, + "grad_norm": 0.5698756575584412, + "learning_rate": 0.0007921896010757508, + "loss": 0.4086, + "step": 7531 + }, + { + "epoch": 0.4208173869318658, + "grad_norm": 0.5652030110359192, + "learning_rate": 0.000792161586732407, + "loss": 0.4463, + "step": 7532 + }, + { + "epoch": 0.42087325753554766, + "grad_norm": 0.754665732383728, + "learning_rate": 0.0007921335723890632, + "loss": 0.4884, + "step": 7533 + }, + { + "epoch": 0.42092912813922956, + "grad_norm": 0.912115216255188, + "learning_rate": 0.0007921055580457194, + "loss": 0.5429, + "step": 7534 + }, + { + "epoch": 0.4209849987429114, + "grad_norm": 0.7340993881225586, + "learning_rate": 0.0007920775437023756, + "loss": 0.4303, + "step": 7535 + }, + { + "epoch": 0.4210408693465933, + "grad_norm": 0.6401121616363525, + "learning_rate": 0.0007920495293590318, + "loss": 0.6857, + "step": 7536 + }, + { + "epoch": 0.4210967399502752, + "grad_norm": 0.6313377618789673, + "learning_rate": 0.000792021515015688, + "loss": 0.5323, + "step": 7537 + }, + { + "epoch": 0.42115261055395703, + "grad_norm": 0.8026255369186401, + "learning_rate": 0.0007919935006723442, + "loss": 0.4291, + "step": 7538 + }, + { + "epoch": 0.4212084811576389, + "grad_norm": 0.6879578828811646, + "learning_rate": 0.0007919654863290005, + "loss": 0.5267, + "step": 7539 + }, + { + "epoch": 0.42126435176132077, + "grad_norm": 0.6343170404434204, + "learning_rate": 0.0007919374719856567, + "loss": 0.555, + "step": 7540 + }, + { + "epoch": 0.42132022236500266, + "grad_norm": 0.4215458929538727, + "learning_rate": 0.000791909457642313, + "loss": 0.3949, + "step": 7541 + }, + { + "epoch": 0.4213760929686845, + "grad_norm": 0.4684436321258545, + "learning_rate": 0.0007918814432989691, + "loss": 0.4268, + "step": 7542 + }, + { + "epoch": 0.4214319635723664, + "grad_norm": 0.8907071352005005, + "learning_rate": 0.0007918534289556253, + "loss": 0.4336, + "step": 7543 + }, + { + "epoch": 0.4214878341760483, + "grad_norm": 0.6243287920951843, + "learning_rate": 0.0007918254146122815, + "loss": 0.5574, + "step": 7544 + }, + { + "epoch": 0.42154370477973013, + "grad_norm": 0.3679237961769104, + "learning_rate": 0.0007917974002689378, + "loss": 0.4133, + "step": 7545 + }, + { + "epoch": 0.42159957538341203, + "grad_norm": 3.102598190307617, + "learning_rate": 0.0007917693859255939, + "loss": 0.46, + "step": 7546 + }, + { + "epoch": 0.42165544598709387, + "grad_norm": 0.4135434925556183, + "learning_rate": 0.0007917413715822501, + "loss": 0.4417, + "step": 7547 + }, + { + "epoch": 0.42171131659077576, + "grad_norm": 0.47992655634880066, + "learning_rate": 0.0007917133572389064, + "loss": 0.4696, + "step": 7548 + }, + { + "epoch": 0.42176718719445766, + "grad_norm": 0.7417668700218201, + "learning_rate": 0.0007916853428955626, + "loss": 0.7018, + "step": 7549 + }, + { + "epoch": 0.4218230577981395, + "grad_norm": 0.632980465888977, + "learning_rate": 0.0007916573285522187, + "loss": 0.4558, + "step": 7550 + }, + { + "epoch": 0.4218789284018214, + "grad_norm": 0.5510860681533813, + "learning_rate": 0.0007916293142088749, + "loss": 0.5717, + "step": 7551 + }, + { + "epoch": 0.42193479900550324, + "grad_norm": 2.3384077548980713, + "learning_rate": 0.0007916012998655312, + "loss": 0.4942, + "step": 7552 + }, + { + "epoch": 0.42199066960918513, + "grad_norm": 0.6067194938659668, + "learning_rate": 0.0007915732855221873, + "loss": 0.3864, + "step": 7553 + }, + { + "epoch": 0.422046540212867, + "grad_norm": 0.6270174384117126, + "learning_rate": 0.0007915452711788435, + "loss": 0.4778, + "step": 7554 + }, + { + "epoch": 0.42210241081654887, + "grad_norm": 0.426343709230423, + "learning_rate": 0.0007915172568354999, + "loss": 0.4277, + "step": 7555 + }, + { + "epoch": 0.42215828142023076, + "grad_norm": 0.45259323716163635, + "learning_rate": 0.0007914892424921561, + "loss": 0.4145, + "step": 7556 + }, + { + "epoch": 0.4222141520239126, + "grad_norm": 0.6662765145301819, + "learning_rate": 0.0007914612281488122, + "loss": 0.4452, + "step": 7557 + }, + { + "epoch": 0.4222700226275945, + "grad_norm": 0.5806347131729126, + "learning_rate": 0.0007914332138054684, + "loss": 0.3499, + "step": 7558 + }, + { + "epoch": 0.42232589323127634, + "grad_norm": 0.5749458074569702, + "learning_rate": 0.0007914051994621247, + "loss": 0.4225, + "step": 7559 + }, + { + "epoch": 0.42238176383495823, + "grad_norm": 0.8719210028648376, + "learning_rate": 0.0007913771851187809, + "loss": 0.4926, + "step": 7560 + }, + { + "epoch": 0.42243763443864013, + "grad_norm": 0.6037045121192932, + "learning_rate": 0.000791349170775437, + "loss": 0.6193, + "step": 7561 + }, + { + "epoch": 0.42249350504232197, + "grad_norm": 0.6813039183616638, + "learning_rate": 0.0007913211564320932, + "loss": 0.4178, + "step": 7562 + }, + { + "epoch": 0.42254937564600387, + "grad_norm": 0.8878533840179443, + "learning_rate": 0.0007912931420887495, + "loss": 0.5061, + "step": 7563 + }, + { + "epoch": 0.4226052462496857, + "grad_norm": 1.0003706216812134, + "learning_rate": 0.0007912651277454057, + "loss": 0.4529, + "step": 7564 + }, + { + "epoch": 0.4226611168533676, + "grad_norm": 0.3636043071746826, + "learning_rate": 0.0007912371134020618, + "loss": 0.3881, + "step": 7565 + }, + { + "epoch": 0.4227169874570495, + "grad_norm": 0.8265222311019897, + "learning_rate": 0.0007912090990587181, + "loss": 0.4867, + "step": 7566 + }, + { + "epoch": 0.42277285806073134, + "grad_norm": 0.5408148765563965, + "learning_rate": 0.0007911810847153743, + "loss": 0.5636, + "step": 7567 + }, + { + "epoch": 0.42282872866441323, + "grad_norm": 1.1007134914398193, + "learning_rate": 0.0007911530703720305, + "loss": 0.4566, + "step": 7568 + }, + { + "epoch": 0.4228845992680951, + "grad_norm": 1.0611541271209717, + "learning_rate": 0.0007911250560286866, + "loss": 0.3834, + "step": 7569 + }, + { + "epoch": 0.42294046987177697, + "grad_norm": 0.5110235810279846, + "learning_rate": 0.0007910970416853429, + "loss": 0.4351, + "step": 7570 + }, + { + "epoch": 0.4229963404754588, + "grad_norm": 0.6275303363800049, + "learning_rate": 0.0007910690273419991, + "loss": 0.4246, + "step": 7571 + }, + { + "epoch": 0.4230522110791407, + "grad_norm": 0.5007472038269043, + "learning_rate": 0.0007910410129986552, + "loss": 0.4973, + "step": 7572 + }, + { + "epoch": 0.4231080816828226, + "grad_norm": 0.8270429372787476, + "learning_rate": 0.0007910129986553116, + "loss": 0.5229, + "step": 7573 + }, + { + "epoch": 0.42316395228650444, + "grad_norm": 0.5822251439094543, + "learning_rate": 0.0007909849843119678, + "loss": 0.3902, + "step": 7574 + }, + { + "epoch": 0.42321982289018634, + "grad_norm": 0.7484460473060608, + "learning_rate": 0.000790956969968624, + "loss": 0.5429, + "step": 7575 + }, + { + "epoch": 0.4232756934938682, + "grad_norm": 0.5159305334091187, + "learning_rate": 0.0007909289556252801, + "loss": 0.4539, + "step": 7576 + }, + { + "epoch": 0.4233315640975501, + "grad_norm": 0.703018844127655, + "learning_rate": 0.0007909009412819364, + "loss": 0.4366, + "step": 7577 + }, + { + "epoch": 0.42338743470123197, + "grad_norm": 0.5889967083930969, + "learning_rate": 0.0007908729269385926, + "loss": 0.4364, + "step": 7578 + }, + { + "epoch": 0.4234433053049138, + "grad_norm": 0.7942429780960083, + "learning_rate": 0.0007908449125952488, + "loss": 0.4376, + "step": 7579 + }, + { + "epoch": 0.4234991759085957, + "grad_norm": 0.42330479621887207, + "learning_rate": 0.000790816898251905, + "loss": 0.4777, + "step": 7580 + }, + { + "epoch": 0.42355504651227754, + "grad_norm": 0.6660901308059692, + "learning_rate": 0.0007907888839085612, + "loss": 0.5228, + "step": 7581 + }, + { + "epoch": 0.42361091711595944, + "grad_norm": 0.5438134670257568, + "learning_rate": 0.0007907608695652174, + "loss": 0.4182, + "step": 7582 + }, + { + "epoch": 0.42366678771964134, + "grad_norm": 0.7630256414413452, + "learning_rate": 0.0007907328552218736, + "loss": 0.3746, + "step": 7583 + }, + { + "epoch": 0.4237226583233232, + "grad_norm": 1.4882526397705078, + "learning_rate": 0.0007907048408785298, + "loss": 0.4717, + "step": 7584 + }, + { + "epoch": 0.42377852892700507, + "grad_norm": 1.1985409259796143, + "learning_rate": 0.000790676826535186, + "loss": 0.4869, + "step": 7585 + }, + { + "epoch": 0.4238343995306869, + "grad_norm": 1.0353785753250122, + "learning_rate": 0.0007906488121918422, + "loss": 0.4964, + "step": 7586 + }, + { + "epoch": 0.4238902701343688, + "grad_norm": 1.1396843194961548, + "learning_rate": 0.0007906207978484985, + "loss": 0.3684, + "step": 7587 + }, + { + "epoch": 0.42394614073805065, + "grad_norm": 1.3597553968429565, + "learning_rate": 0.0007905927835051546, + "loss": 0.495, + "step": 7588 + }, + { + "epoch": 0.42400201134173254, + "grad_norm": 0.5065939426422119, + "learning_rate": 0.0007905647691618109, + "loss": 0.4671, + "step": 7589 + }, + { + "epoch": 0.42405788194541444, + "grad_norm": 1.192055106163025, + "learning_rate": 0.0007905367548184671, + "loss": 0.4243, + "step": 7590 + }, + { + "epoch": 0.4241137525490963, + "grad_norm": 0.3538742661476135, + "learning_rate": 0.0007905087404751234, + "loss": 0.3728, + "step": 7591 + }, + { + "epoch": 0.4241696231527782, + "grad_norm": 1.3724205493927002, + "learning_rate": 0.0007904807261317795, + "loss": 0.5473, + "step": 7592 + }, + { + "epoch": 0.42422549375646, + "grad_norm": 0.9103184938430786, + "learning_rate": 0.0007904527117884357, + "loss": 0.441, + "step": 7593 + }, + { + "epoch": 0.4242813643601419, + "grad_norm": 0.49006471037864685, + "learning_rate": 0.0007904246974450919, + "loss": 0.4771, + "step": 7594 + }, + { + "epoch": 0.4243372349638238, + "grad_norm": 0.5364717841148376, + "learning_rate": 0.0007903966831017481, + "loss": 0.493, + "step": 7595 + }, + { + "epoch": 0.42439310556750565, + "grad_norm": 0.5158135294914246, + "learning_rate": 0.0007903686687584043, + "loss": 0.4547, + "step": 7596 + }, + { + "epoch": 0.42444897617118754, + "grad_norm": 5.089476585388184, + "learning_rate": 0.0007903406544150605, + "loss": 0.4849, + "step": 7597 + }, + { + "epoch": 0.4245048467748694, + "grad_norm": 0.5306967496871948, + "learning_rate": 0.0007903126400717168, + "loss": 0.4816, + "step": 7598 + }, + { + "epoch": 0.4245607173785513, + "grad_norm": 0.7608968615531921, + "learning_rate": 0.0007902846257283729, + "loss": 0.5285, + "step": 7599 + }, + { + "epoch": 0.4246165879822332, + "grad_norm": 0.7697194218635559, + "learning_rate": 0.0007902566113850291, + "loss": 0.5042, + "step": 7600 + }, + { + "epoch": 0.424672458585915, + "grad_norm": 1.0011788606643677, + "learning_rate": 0.0007902285970416853, + "loss": 0.4869, + "step": 7601 + }, + { + "epoch": 0.4247283291895969, + "grad_norm": 0.8343632817268372, + "learning_rate": 0.0007902005826983416, + "loss": 0.6817, + "step": 7602 + }, + { + "epoch": 0.42478419979327875, + "grad_norm": 1.519160270690918, + "learning_rate": 0.0007901725683549977, + "loss": 0.4481, + "step": 7603 + }, + { + "epoch": 0.42484007039696065, + "grad_norm": 0.8376628756523132, + "learning_rate": 0.0007901445540116539, + "loss": 0.4354, + "step": 7604 + }, + { + "epoch": 0.4248959410006425, + "grad_norm": 0.5370498299598694, + "learning_rate": 0.0007901165396683103, + "loss": 0.5022, + "step": 7605 + }, + { + "epoch": 0.4249518116043244, + "grad_norm": 0.7866578102111816, + "learning_rate": 0.0007900885253249665, + "loss": 0.4749, + "step": 7606 + }, + { + "epoch": 0.4250076822080063, + "grad_norm": 0.6034008264541626, + "learning_rate": 0.0007900605109816226, + "loss": 0.555, + "step": 7607 + }, + { + "epoch": 0.4250635528116881, + "grad_norm": 0.6044673323631287, + "learning_rate": 0.0007900324966382788, + "loss": 0.4701, + "step": 7608 + }, + { + "epoch": 0.42511942341537, + "grad_norm": 0.6636052131652832, + "learning_rate": 0.0007900044822949351, + "loss": 0.4744, + "step": 7609 + }, + { + "epoch": 0.42517529401905185, + "grad_norm": 0.7916946411132812, + "learning_rate": 0.0007899764679515913, + "loss": 0.4044, + "step": 7610 + }, + { + "epoch": 0.42523116462273375, + "grad_norm": 1.1864466667175293, + "learning_rate": 0.0007899484536082474, + "loss": 0.4187, + "step": 7611 + }, + { + "epoch": 0.42528703522641564, + "grad_norm": 0.6465734839439392, + "learning_rate": 0.0007899204392649037, + "loss": 0.4498, + "step": 7612 + }, + { + "epoch": 0.4253429058300975, + "grad_norm": 0.6115749478340149, + "learning_rate": 0.0007898924249215599, + "loss": 0.3535, + "step": 7613 + }, + { + "epoch": 0.4253987764337794, + "grad_norm": 0.5533115863800049, + "learning_rate": 0.000789864410578216, + "loss": 0.5444, + "step": 7614 + }, + { + "epoch": 0.4254546470374612, + "grad_norm": 0.5277127027511597, + "learning_rate": 0.0007898363962348722, + "loss": 0.4674, + "step": 7615 + }, + { + "epoch": 0.4255105176411431, + "grad_norm": 0.783004641532898, + "learning_rate": 0.0007898083818915285, + "loss": 0.5308, + "step": 7616 + }, + { + "epoch": 0.425566388244825, + "grad_norm": 0.6420401930809021, + "learning_rate": 0.0007897803675481847, + "loss": 0.4715, + "step": 7617 + }, + { + "epoch": 0.42562225884850685, + "grad_norm": 0.5794914960861206, + "learning_rate": 0.0007897523532048408, + "loss": 0.4513, + "step": 7618 + }, + { + "epoch": 0.42567812945218875, + "grad_norm": 0.7524482011795044, + "learning_rate": 0.0007897243388614971, + "loss": 0.4409, + "step": 7619 + }, + { + "epoch": 0.4257340000558706, + "grad_norm": 0.53031325340271, + "learning_rate": 0.0007896963245181533, + "loss": 0.3705, + "step": 7620 + }, + { + "epoch": 0.4257898706595525, + "grad_norm": 1.1012158393859863, + "learning_rate": 0.0007896683101748095, + "loss": 0.4263, + "step": 7621 + }, + { + "epoch": 0.4258457412632343, + "grad_norm": 0.5490109920501709, + "learning_rate": 0.0007896402958314656, + "loss": 0.5358, + "step": 7622 + }, + { + "epoch": 0.4259016118669162, + "grad_norm": 1.3085273504257202, + "learning_rate": 0.000789612281488122, + "loss": 0.426, + "step": 7623 + }, + { + "epoch": 0.4259574824705981, + "grad_norm": 0.4191201329231262, + "learning_rate": 0.0007895842671447782, + "loss": 0.4637, + "step": 7624 + }, + { + "epoch": 0.42601335307427995, + "grad_norm": 0.5764442086219788, + "learning_rate": 0.0007895562528014344, + "loss": 0.4613, + "step": 7625 + }, + { + "epoch": 0.42606922367796185, + "grad_norm": 1.7072713375091553, + "learning_rate": 0.0007895282384580905, + "loss": 0.4172, + "step": 7626 + }, + { + "epoch": 0.4261250942816437, + "grad_norm": 0.4484826326370239, + "learning_rate": 0.0007895002241147468, + "loss": 0.4566, + "step": 7627 + }, + { + "epoch": 0.4261809648853256, + "grad_norm": 0.45744505524635315, + "learning_rate": 0.000789472209771403, + "loss": 0.4618, + "step": 7628 + }, + { + "epoch": 0.4262368354890075, + "grad_norm": 1.9666938781738281, + "learning_rate": 0.0007894441954280592, + "loss": 0.6168, + "step": 7629 + }, + { + "epoch": 0.4262927060926893, + "grad_norm": 0.49108728766441345, + "learning_rate": 0.0007894161810847154, + "loss": 0.398, + "step": 7630 + }, + { + "epoch": 0.4263485766963712, + "grad_norm": 0.4404587745666504, + "learning_rate": 0.0007893881667413716, + "loss": 0.4986, + "step": 7631 + }, + { + "epoch": 0.42640444730005306, + "grad_norm": 0.7471987009048462, + "learning_rate": 0.0007893601523980278, + "loss": 0.4818, + "step": 7632 + }, + { + "epoch": 0.42646031790373495, + "grad_norm": 0.4581044912338257, + "learning_rate": 0.000789332138054684, + "loss": 0.4897, + "step": 7633 + }, + { + "epoch": 0.42651618850741685, + "grad_norm": 9.572473526000977, + "learning_rate": 0.0007893041237113402, + "loss": 0.5123, + "step": 7634 + }, + { + "epoch": 0.4265720591110987, + "grad_norm": 0.8764320015907288, + "learning_rate": 0.0007892761093679964, + "loss": 0.509, + "step": 7635 + }, + { + "epoch": 0.4266279297147806, + "grad_norm": 0.7493451237678528, + "learning_rate": 0.0007892480950246526, + "loss": 0.4778, + "step": 7636 + }, + { + "epoch": 0.4266838003184624, + "grad_norm": 0.37569090723991394, + "learning_rate": 0.0007892200806813088, + "loss": 0.3607, + "step": 7637 + }, + { + "epoch": 0.4267396709221443, + "grad_norm": 7.541540622711182, + "learning_rate": 0.000789192066337965, + "loss": 0.4011, + "step": 7638 + }, + { + "epoch": 0.42679554152582616, + "grad_norm": 0.4331929683685303, + "learning_rate": 0.0007891640519946213, + "loss": 0.4112, + "step": 7639 + }, + { + "epoch": 0.42685141212950806, + "grad_norm": 0.44391798973083496, + "learning_rate": 0.0007891360376512775, + "loss": 0.4042, + "step": 7640 + }, + { + "epoch": 0.42690728273318995, + "grad_norm": 1.527060627937317, + "learning_rate": 0.0007891080233079337, + "loss": 0.4513, + "step": 7641 + }, + { + "epoch": 0.4269631533368718, + "grad_norm": 1.088496446609497, + "learning_rate": 0.0007890800089645899, + "loss": 0.5065, + "step": 7642 + }, + { + "epoch": 0.4270190239405537, + "grad_norm": 0.8388476967811584, + "learning_rate": 0.0007890519946212461, + "loss": 0.5795, + "step": 7643 + }, + { + "epoch": 0.42707489454423553, + "grad_norm": 0.465645432472229, + "learning_rate": 0.0007890239802779024, + "loss": 0.6525, + "step": 7644 + }, + { + "epoch": 0.4271307651479174, + "grad_norm": 0.4444662630558014, + "learning_rate": 0.0007889959659345585, + "loss": 0.4533, + "step": 7645 + }, + { + "epoch": 0.4271866357515993, + "grad_norm": 0.6915349960327148, + "learning_rate": 0.0007889679515912147, + "loss": 0.4272, + "step": 7646 + }, + { + "epoch": 0.42724250635528116, + "grad_norm": 0.7253225445747375, + "learning_rate": 0.0007889399372478709, + "loss": 0.5158, + "step": 7647 + }, + { + "epoch": 0.42729837695896306, + "grad_norm": 0.7311456203460693, + "learning_rate": 0.0007889119229045272, + "loss": 0.4593, + "step": 7648 + }, + { + "epoch": 0.4273542475626449, + "grad_norm": 0.6924473643302917, + "learning_rate": 0.0007888839085611833, + "loss": 0.507, + "step": 7649 + }, + { + "epoch": 0.4274101181663268, + "grad_norm": 0.5202884674072266, + "learning_rate": 0.0007888558942178395, + "loss": 0.4438, + "step": 7650 + }, + { + "epoch": 0.4274659887700087, + "grad_norm": 3.5896129608154297, + "learning_rate": 0.0007888278798744958, + "loss": 0.5173, + "step": 7651 + }, + { + "epoch": 0.4275218593736905, + "grad_norm": 2.00431227684021, + "learning_rate": 0.000788799865531152, + "loss": 0.5752, + "step": 7652 + }, + { + "epoch": 0.4275777299773724, + "grad_norm": 0.8906704187393188, + "learning_rate": 0.0007887718511878081, + "loss": 0.4154, + "step": 7653 + }, + { + "epoch": 0.42763360058105426, + "grad_norm": 0.8937134742736816, + "learning_rate": 0.0007887438368444643, + "loss": 0.4754, + "step": 7654 + }, + { + "epoch": 0.42768947118473616, + "grad_norm": 0.4856739342212677, + "learning_rate": 0.0007887158225011207, + "loss": 0.4777, + "step": 7655 + }, + { + "epoch": 0.427745341788418, + "grad_norm": 0.5493022203445435, + "learning_rate": 0.0007886878081577767, + "loss": 0.3835, + "step": 7656 + }, + { + "epoch": 0.4278012123920999, + "grad_norm": 0.5662045478820801, + "learning_rate": 0.000788659793814433, + "loss": 0.6191, + "step": 7657 + }, + { + "epoch": 0.4278570829957818, + "grad_norm": 1.7822294235229492, + "learning_rate": 0.0007886317794710893, + "loss": 0.5573, + "step": 7658 + }, + { + "epoch": 0.42791295359946363, + "grad_norm": 0.47523534297943115, + "learning_rate": 0.0007886037651277455, + "loss": 0.5227, + "step": 7659 + }, + { + "epoch": 0.4279688242031455, + "grad_norm": 0.6089125275611877, + "learning_rate": 0.0007885757507844016, + "loss": 0.5062, + "step": 7660 + }, + { + "epoch": 0.42802469480682737, + "grad_norm": 0.830024242401123, + "learning_rate": 0.0007885477364410578, + "loss": 0.4877, + "step": 7661 + }, + { + "epoch": 0.42808056541050926, + "grad_norm": 4.463840007781982, + "learning_rate": 0.0007885197220977141, + "loss": 0.4922, + "step": 7662 + }, + { + "epoch": 0.42813643601419116, + "grad_norm": 0.6856356263160706, + "learning_rate": 0.0007884917077543703, + "loss": 0.4516, + "step": 7663 + }, + { + "epoch": 0.428192306617873, + "grad_norm": 0.432841956615448, + "learning_rate": 0.0007884636934110264, + "loss": 0.4153, + "step": 7664 + }, + { + "epoch": 0.4282481772215549, + "grad_norm": 1.445918083190918, + "learning_rate": 0.0007884356790676826, + "loss": 0.5302, + "step": 7665 + }, + { + "epoch": 0.42830404782523673, + "grad_norm": 0.7912014126777649, + "learning_rate": 0.0007884076647243389, + "loss": 0.5202, + "step": 7666 + }, + { + "epoch": 0.42835991842891863, + "grad_norm": 0.5630530118942261, + "learning_rate": 0.0007883796503809951, + "loss": 0.4552, + "step": 7667 + }, + { + "epoch": 0.4284157890326005, + "grad_norm": 0.5699687004089355, + "learning_rate": 0.0007883516360376512, + "loss": 0.5104, + "step": 7668 + }, + { + "epoch": 0.42847165963628236, + "grad_norm": 1.1191076040267944, + "learning_rate": 0.0007883236216943075, + "loss": 0.54, + "step": 7669 + }, + { + "epoch": 0.42852753023996426, + "grad_norm": 2.0853936672210693, + "learning_rate": 0.0007882956073509637, + "loss": 0.4978, + "step": 7670 + }, + { + "epoch": 0.4285834008436461, + "grad_norm": 0.5287569761276245, + "learning_rate": 0.0007882675930076199, + "loss": 0.5659, + "step": 7671 + }, + { + "epoch": 0.428639271447328, + "grad_norm": 0.7810014486312866, + "learning_rate": 0.000788239578664276, + "loss": 0.3966, + "step": 7672 + }, + { + "epoch": 0.42869514205100984, + "grad_norm": 0.8725608587265015, + "learning_rate": 0.0007882115643209324, + "loss": 0.4652, + "step": 7673 + }, + { + "epoch": 0.42875101265469173, + "grad_norm": 0.4048239290714264, + "learning_rate": 0.0007881835499775886, + "loss": 0.4065, + "step": 7674 + }, + { + "epoch": 0.42880688325837363, + "grad_norm": 0.5355663299560547, + "learning_rate": 0.0007881555356342448, + "loss": 0.492, + "step": 7675 + }, + { + "epoch": 0.42886275386205547, + "grad_norm": 1.2814215421676636, + "learning_rate": 0.000788127521290901, + "loss": 0.4932, + "step": 7676 + }, + { + "epoch": 0.42891862446573736, + "grad_norm": 1.206464171409607, + "learning_rate": 0.0007880995069475572, + "loss": 0.4075, + "step": 7677 + }, + { + "epoch": 0.4289744950694192, + "grad_norm": 1.3877902030944824, + "learning_rate": 0.0007880714926042134, + "loss": 0.7363, + "step": 7678 + }, + { + "epoch": 0.4290303656731011, + "grad_norm": 1.6719281673431396, + "learning_rate": 0.0007880434782608695, + "loss": 0.5104, + "step": 7679 + }, + { + "epoch": 0.429086236276783, + "grad_norm": 1.0803380012512207, + "learning_rate": 0.0007880154639175258, + "loss": 0.4486, + "step": 7680 + }, + { + "epoch": 0.42914210688046484, + "grad_norm": 0.6703947186470032, + "learning_rate": 0.000787987449574182, + "loss": 0.3854, + "step": 7681 + }, + { + "epoch": 0.42919797748414673, + "grad_norm": 0.5913898348808289, + "learning_rate": 0.0007879594352308382, + "loss": 0.411, + "step": 7682 + }, + { + "epoch": 0.42925384808782857, + "grad_norm": 0.5547372102737427, + "learning_rate": 0.0007879314208874944, + "loss": 0.3703, + "step": 7683 + }, + { + "epoch": 0.42930971869151047, + "grad_norm": 0.5901825428009033, + "learning_rate": 0.0007879034065441506, + "loss": 0.4942, + "step": 7684 + }, + { + "epoch": 0.4293655892951923, + "grad_norm": 0.4701577425003052, + "learning_rate": 0.0007878753922008068, + "loss": 0.5603, + "step": 7685 + }, + { + "epoch": 0.4294214598988742, + "grad_norm": 0.5960360169410706, + "learning_rate": 0.000787847377857463, + "loss": 0.5777, + "step": 7686 + }, + { + "epoch": 0.4294773305025561, + "grad_norm": 0.6679201722145081, + "learning_rate": 0.0007878193635141192, + "loss": 0.575, + "step": 7687 + }, + { + "epoch": 0.42953320110623794, + "grad_norm": 0.48743000626564026, + "learning_rate": 0.0007877913491707754, + "loss": 0.4185, + "step": 7688 + }, + { + "epoch": 0.42958907170991983, + "grad_norm": 1.337043046951294, + "learning_rate": 0.0007877633348274316, + "loss": 0.5934, + "step": 7689 + }, + { + "epoch": 0.4296449423136017, + "grad_norm": 0.5061859488487244, + "learning_rate": 0.000787735320484088, + "loss": 0.4546, + "step": 7690 + }, + { + "epoch": 0.42970081291728357, + "grad_norm": 0.43310481309890747, + "learning_rate": 0.0007877073061407441, + "loss": 0.3986, + "step": 7691 + }, + { + "epoch": 0.42975668352096547, + "grad_norm": 0.4701261520385742, + "learning_rate": 0.0007876792917974003, + "loss": 0.4155, + "step": 7692 + }, + { + "epoch": 0.4298125541246473, + "grad_norm": 1.034239649772644, + "learning_rate": 0.0007876512774540565, + "loss": 0.5086, + "step": 7693 + }, + { + "epoch": 0.4298684247283292, + "grad_norm": 0.4098905920982361, + "learning_rate": 0.0007876232631107128, + "loss": 0.4287, + "step": 7694 + }, + { + "epoch": 0.42992429533201104, + "grad_norm": 1.1247365474700928, + "learning_rate": 0.0007875952487673689, + "loss": 0.5753, + "step": 7695 + }, + { + "epoch": 0.42998016593569294, + "grad_norm": 11.183462142944336, + "learning_rate": 0.0007875672344240251, + "loss": 0.4026, + "step": 7696 + }, + { + "epoch": 0.43003603653937483, + "grad_norm": 0.6496208310127258, + "learning_rate": 0.0007875392200806814, + "loss": 0.425, + "step": 7697 + }, + { + "epoch": 0.4300919071430567, + "grad_norm": 0.7365894317626953, + "learning_rate": 0.0007875112057373376, + "loss": 0.3366, + "step": 7698 + }, + { + "epoch": 0.43014777774673857, + "grad_norm": 0.49366050958633423, + "learning_rate": 0.0007874831913939937, + "loss": 0.4209, + "step": 7699 + }, + { + "epoch": 0.4302036483504204, + "grad_norm": 0.4845854938030243, + "learning_rate": 0.0007874551770506499, + "loss": 0.4992, + "step": 7700 + }, + { + "epoch": 0.4302595189541023, + "grad_norm": 0.6987083554267883, + "learning_rate": 0.0007874271627073062, + "loss": 0.4681, + "step": 7701 + }, + { + "epoch": 0.43031538955778414, + "grad_norm": 0.8404329419136047, + "learning_rate": 0.0007873991483639623, + "loss": 0.5668, + "step": 7702 + }, + { + "epoch": 0.43037126016146604, + "grad_norm": 1.2692207098007202, + "learning_rate": 0.0007873711340206185, + "loss": 0.5814, + "step": 7703 + }, + { + "epoch": 0.43042713076514794, + "grad_norm": 2.916429042816162, + "learning_rate": 0.0007873431196772747, + "loss": 0.4817, + "step": 7704 + }, + { + "epoch": 0.4304830013688298, + "grad_norm": 1.1823042631149292, + "learning_rate": 0.000787315105333931, + "loss": 0.5201, + "step": 7705 + }, + { + "epoch": 0.43053887197251167, + "grad_norm": 0.7991429567337036, + "learning_rate": 0.0007872870909905871, + "loss": 0.446, + "step": 7706 + }, + { + "epoch": 0.4305947425761935, + "grad_norm": 0.615501344203949, + "learning_rate": 0.0007872590766472434, + "loss": 0.3977, + "step": 7707 + }, + { + "epoch": 0.4306506131798754, + "grad_norm": 1.4535013437271118, + "learning_rate": 0.0007872310623038997, + "loss": 0.3738, + "step": 7708 + }, + { + "epoch": 0.4307064837835573, + "grad_norm": 1.0330047607421875, + "learning_rate": 0.0007872030479605559, + "loss": 0.491, + "step": 7709 + }, + { + "epoch": 0.43076235438723914, + "grad_norm": 0.5989092588424683, + "learning_rate": 0.000787175033617212, + "loss": 0.5714, + "step": 7710 + }, + { + "epoch": 0.43081822499092104, + "grad_norm": 0.734026312828064, + "learning_rate": 0.0007871470192738682, + "loss": 0.432, + "step": 7711 + }, + { + "epoch": 0.4308740955946029, + "grad_norm": 1.2719428539276123, + "learning_rate": 0.0007871190049305245, + "loss": 0.3711, + "step": 7712 + }, + { + "epoch": 0.4309299661982848, + "grad_norm": 0.5172150731086731, + "learning_rate": 0.0007870909905871807, + "loss": 0.4799, + "step": 7713 + }, + { + "epoch": 0.43098583680196667, + "grad_norm": 0.6308301687240601, + "learning_rate": 0.0007870629762438368, + "loss": 0.5104, + "step": 7714 + }, + { + "epoch": 0.4310417074056485, + "grad_norm": 0.5157002210617065, + "learning_rate": 0.0007870349619004931, + "loss": 0.588, + "step": 7715 + }, + { + "epoch": 0.4310975780093304, + "grad_norm": 0.6802142262458801, + "learning_rate": 0.0007870069475571493, + "loss": 0.4314, + "step": 7716 + }, + { + "epoch": 0.43115344861301225, + "grad_norm": 1.396522879600525, + "learning_rate": 0.0007869789332138055, + "loss": 0.4309, + "step": 7717 + }, + { + "epoch": 0.43120931921669414, + "grad_norm": 0.5319505333900452, + "learning_rate": 0.0007869509188704616, + "loss": 0.4372, + "step": 7718 + }, + { + "epoch": 0.431265189820376, + "grad_norm": 0.4972088932991028, + "learning_rate": 0.0007869229045271179, + "loss": 0.3911, + "step": 7719 + }, + { + "epoch": 0.4313210604240579, + "grad_norm": 0.8087112903594971, + "learning_rate": 0.0007868948901837741, + "loss": 0.6056, + "step": 7720 + }, + { + "epoch": 0.4313769310277398, + "grad_norm": 0.6812052130699158, + "learning_rate": 0.0007868668758404302, + "loss": 0.4835, + "step": 7721 + }, + { + "epoch": 0.4314328016314216, + "grad_norm": 0.5767220854759216, + "learning_rate": 0.0007868388614970865, + "loss": 0.3621, + "step": 7722 + }, + { + "epoch": 0.4314886722351035, + "grad_norm": 0.7344948053359985, + "learning_rate": 0.0007868108471537428, + "loss": 0.5126, + "step": 7723 + }, + { + "epoch": 0.43154454283878535, + "grad_norm": 0.5572934746742249, + "learning_rate": 0.000786782832810399, + "loss": 0.6186, + "step": 7724 + }, + { + "epoch": 0.43160041344246725, + "grad_norm": 0.4777257740497589, + "learning_rate": 0.0007867548184670551, + "loss": 0.5489, + "step": 7725 + }, + { + "epoch": 0.43165628404614914, + "grad_norm": 1.089013934135437, + "learning_rate": 0.0007867268041237114, + "loss": 0.4011, + "step": 7726 + }, + { + "epoch": 0.431712154649831, + "grad_norm": 2.128246545791626, + "learning_rate": 0.0007866987897803676, + "loss": 0.7147, + "step": 7727 + }, + { + "epoch": 0.4317680252535129, + "grad_norm": 0.8349575996398926, + "learning_rate": 0.0007866707754370238, + "loss": 0.3682, + "step": 7728 + }, + { + "epoch": 0.4318238958571947, + "grad_norm": 0.43797191977500916, + "learning_rate": 0.00078664276109368, + "loss": 0.3895, + "step": 7729 + }, + { + "epoch": 0.4318797664608766, + "grad_norm": 0.4395776093006134, + "learning_rate": 0.0007866147467503362, + "loss": 0.4702, + "step": 7730 + }, + { + "epoch": 0.4319356370645585, + "grad_norm": 0.7568327188491821, + "learning_rate": 0.0007865867324069924, + "loss": 0.4702, + "step": 7731 + }, + { + "epoch": 0.43199150766824035, + "grad_norm": 1.1159371137619019, + "learning_rate": 0.0007865587180636486, + "loss": 0.5743, + "step": 7732 + }, + { + "epoch": 0.43204737827192224, + "grad_norm": 0.7205939888954163, + "learning_rate": 0.0007865307037203048, + "loss": 0.4191, + "step": 7733 + }, + { + "epoch": 0.4321032488756041, + "grad_norm": 0.46701329946517944, + "learning_rate": 0.000786502689376961, + "loss": 0.4047, + "step": 7734 + }, + { + "epoch": 0.432159119479286, + "grad_norm": 0.5116703510284424, + "learning_rate": 0.0007864746750336172, + "loss": 0.4743, + "step": 7735 + }, + { + "epoch": 0.4322149900829678, + "grad_norm": 0.534207284450531, + "learning_rate": 0.0007864466606902735, + "loss": 0.3695, + "step": 7736 + }, + { + "epoch": 0.4322708606866497, + "grad_norm": 0.7179378271102905, + "learning_rate": 0.0007864186463469296, + "loss": 0.4911, + "step": 7737 + }, + { + "epoch": 0.4323267312903316, + "grad_norm": 0.7805116176605225, + "learning_rate": 0.0007863906320035858, + "loss": 0.4966, + "step": 7738 + }, + { + "epoch": 0.43238260189401345, + "grad_norm": 0.5068917870521545, + "learning_rate": 0.000786362617660242, + "loss": 0.383, + "step": 7739 + }, + { + "epoch": 0.43243847249769535, + "grad_norm": 0.5705819129943848, + "learning_rate": 0.0007863346033168984, + "loss": 0.4301, + "step": 7740 + }, + { + "epoch": 0.4324943431013772, + "grad_norm": 1.4573924541473389, + "learning_rate": 0.0007863065889735545, + "loss": 0.3576, + "step": 7741 + }, + { + "epoch": 0.4325502137050591, + "grad_norm": 0.6321025490760803, + "learning_rate": 0.0007862785746302107, + "loss": 0.4549, + "step": 7742 + }, + { + "epoch": 0.432606084308741, + "grad_norm": 0.47242721915245056, + "learning_rate": 0.0007862505602868669, + "loss": 0.4256, + "step": 7743 + }, + { + "epoch": 0.4326619549124228, + "grad_norm": 0.4355708658695221, + "learning_rate": 0.0007862225459435231, + "loss": 0.4455, + "step": 7744 + }, + { + "epoch": 0.4327178255161047, + "grad_norm": 0.5267342925071716, + "learning_rate": 0.0007861945316001793, + "loss": 0.484, + "step": 7745 + }, + { + "epoch": 0.43277369611978656, + "grad_norm": 0.5952574610710144, + "learning_rate": 0.0007861665172568355, + "loss": 0.4705, + "step": 7746 + }, + { + "epoch": 0.43282956672346845, + "grad_norm": 0.7777061462402344, + "learning_rate": 0.0007861385029134918, + "loss": 0.4877, + "step": 7747 + }, + { + "epoch": 0.43288543732715035, + "grad_norm": 1.0699042081832886, + "learning_rate": 0.0007861104885701479, + "loss": 0.4721, + "step": 7748 + }, + { + "epoch": 0.4329413079308322, + "grad_norm": 0.42261549830436707, + "learning_rate": 0.0007860824742268041, + "loss": 0.357, + "step": 7749 + }, + { + "epoch": 0.4329971785345141, + "grad_norm": 0.5335983037948608, + "learning_rate": 0.0007860544598834603, + "loss": 0.6155, + "step": 7750 + }, + { + "epoch": 0.4330530491381959, + "grad_norm": Infinity, + "learning_rate": 0.0007860544598834603, + "loss": 0.7399, + "step": 7751 + }, + { + "epoch": 0.4331089197418778, + "grad_norm": 1.3638560771942139, + "learning_rate": 0.0007860264455401166, + "loss": 0.4214, + "step": 7752 + }, + { + "epoch": 0.43316479034555966, + "grad_norm": 4.952521324157715, + "learning_rate": 0.0007859984311967727, + "loss": 0.4097, + "step": 7753 + }, + { + "epoch": 0.43322066094924155, + "grad_norm": 0.43153703212738037, + "learning_rate": 0.0007859704168534289, + "loss": 0.4664, + "step": 7754 + }, + { + "epoch": 0.43327653155292345, + "grad_norm": 6.7690205574035645, + "learning_rate": 0.0007859424025100852, + "loss": 0.4342, + "step": 7755 + }, + { + "epoch": 0.4333324021566053, + "grad_norm": 0.5010371208190918, + "learning_rate": 0.0007859143881667414, + "loss": 0.4373, + "step": 7756 + }, + { + "epoch": 0.4333882727602872, + "grad_norm": 0.5382539629936218, + "learning_rate": 0.0007858863738233975, + "loss": 0.4589, + "step": 7757 + }, + { + "epoch": 0.433444143363969, + "grad_norm": 0.45392847061157227, + "learning_rate": 0.0007858583594800538, + "loss": 0.376, + "step": 7758 + }, + { + "epoch": 0.4335000139676509, + "grad_norm": 0.5627939701080322, + "learning_rate": 0.0007858303451367101, + "loss": 0.3626, + "step": 7759 + }, + { + "epoch": 0.4335558845713328, + "grad_norm": 0.424269437789917, + "learning_rate": 0.0007858023307933663, + "loss": 0.3726, + "step": 7760 + }, + { + "epoch": 0.43361175517501466, + "grad_norm": 0.8004353642463684, + "learning_rate": 0.0007857743164500224, + "loss": 0.3886, + "step": 7761 + }, + { + "epoch": 0.43366762577869655, + "grad_norm": 0.4775344133377075, + "learning_rate": 0.0007857463021066787, + "loss": 0.3889, + "step": 7762 + }, + { + "epoch": 0.4337234963823784, + "grad_norm": 4.630137920379639, + "learning_rate": 0.0007857182877633349, + "loss": 0.3509, + "step": 7763 + }, + { + "epoch": 0.4337793669860603, + "grad_norm": 0.5877236127853394, + "learning_rate": 0.000785690273419991, + "loss": 0.5325, + "step": 7764 + }, + { + "epoch": 0.4338352375897422, + "grad_norm": 0.5643927454948425, + "learning_rate": 0.0007856622590766472, + "loss": 0.5001, + "step": 7765 + }, + { + "epoch": 0.433891108193424, + "grad_norm": 0.6003684401512146, + "learning_rate": 0.0007856342447333035, + "loss": 0.4844, + "step": 7766 + }, + { + "epoch": 0.4339469787971059, + "grad_norm": 2.8461153507232666, + "learning_rate": 0.0007856062303899597, + "loss": 0.5031, + "step": 7767 + }, + { + "epoch": 0.43400284940078776, + "grad_norm": 0.38253292441368103, + "learning_rate": 0.0007855782160466158, + "loss": 0.2983, + "step": 7768 + }, + { + "epoch": 0.43405872000446966, + "grad_norm": 0.46417564153671265, + "learning_rate": 0.0007855502017032721, + "loss": 0.4061, + "step": 7769 + }, + { + "epoch": 0.4341145906081515, + "grad_norm": 0.7555800676345825, + "learning_rate": 0.0007855221873599283, + "loss": 0.5205, + "step": 7770 + }, + { + "epoch": 0.4341704612118334, + "grad_norm": 0.8537387251853943, + "learning_rate": 0.0007854941730165845, + "loss": 0.5745, + "step": 7771 + }, + { + "epoch": 0.4342263318155153, + "grad_norm": 0.5732616186141968, + "learning_rate": 0.0007854661586732406, + "loss": 0.5402, + "step": 7772 + }, + { + "epoch": 0.4342822024191971, + "grad_norm": 0.5685868263244629, + "learning_rate": 0.000785438144329897, + "loss": 0.4303, + "step": 7773 + }, + { + "epoch": 0.434338073022879, + "grad_norm": 0.6755401492118835, + "learning_rate": 0.0007854101299865532, + "loss": 0.6598, + "step": 7774 + }, + { + "epoch": 0.43439394362656086, + "grad_norm": 0.9088857769966125, + "learning_rate": 0.0007853821156432094, + "loss": 0.4296, + "step": 7775 + }, + { + "epoch": 0.43444981423024276, + "grad_norm": 0.4908600151538849, + "learning_rate": 0.0007853541012998655, + "loss": 0.5056, + "step": 7776 + }, + { + "epoch": 0.43450568483392465, + "grad_norm": 0.6894169449806213, + "learning_rate": 0.0007853260869565218, + "loss": 0.4282, + "step": 7777 + }, + { + "epoch": 0.4345615554376065, + "grad_norm": 0.7428377866744995, + "learning_rate": 0.000785298072613178, + "loss": 0.399, + "step": 7778 + }, + { + "epoch": 0.4346174260412884, + "grad_norm": 0.5645182728767395, + "learning_rate": 0.0007852700582698342, + "loss": 0.5394, + "step": 7779 + }, + { + "epoch": 0.43467329664497023, + "grad_norm": 0.4553984999656677, + "learning_rate": 0.0007852420439264904, + "loss": 0.4995, + "step": 7780 + }, + { + "epoch": 0.4347291672486521, + "grad_norm": 0.5431766510009766, + "learning_rate": 0.0007852140295831466, + "loss": 0.5464, + "step": 7781 + }, + { + "epoch": 0.434785037852334, + "grad_norm": 0.5561527013778687, + "learning_rate": 0.0007851860152398028, + "loss": 0.4686, + "step": 7782 + }, + { + "epoch": 0.43484090845601586, + "grad_norm": 0.6543338298797607, + "learning_rate": 0.000785158000896459, + "loss": 0.5688, + "step": 7783 + }, + { + "epoch": 0.43489677905969776, + "grad_norm": 0.5381743907928467, + "learning_rate": 0.0007851299865531152, + "loss": 0.5051, + "step": 7784 + }, + { + "epoch": 0.4349526496633796, + "grad_norm": 0.5600734949111938, + "learning_rate": 0.0007851019722097714, + "loss": 0.4563, + "step": 7785 + }, + { + "epoch": 0.4350085202670615, + "grad_norm": 0.7490230202674866, + "learning_rate": 0.0007850739578664276, + "loss": 0.5789, + "step": 7786 + }, + { + "epoch": 0.43506439087074333, + "grad_norm": 0.6440300345420837, + "learning_rate": 0.0007850459435230838, + "loss": 0.4159, + "step": 7787 + }, + { + "epoch": 0.43512026147442523, + "grad_norm": 0.4772951006889343, + "learning_rate": 0.00078501792917974, + "loss": 0.4606, + "step": 7788 + }, + { + "epoch": 0.4351761320781071, + "grad_norm": 3.0705270767211914, + "learning_rate": 0.0007849899148363962, + "loss": 0.381, + "step": 7789 + }, + { + "epoch": 0.43523200268178897, + "grad_norm": 1.178820252418518, + "learning_rate": 0.0007849619004930524, + "loss": 0.6356, + "step": 7790 + }, + { + "epoch": 0.43528787328547086, + "grad_norm": 0.7304395437240601, + "learning_rate": 0.0007849338861497087, + "loss": 0.4944, + "step": 7791 + }, + { + "epoch": 0.4353437438891527, + "grad_norm": 0.46784475445747375, + "learning_rate": 0.0007849058718063649, + "loss": 0.5924, + "step": 7792 + }, + { + "epoch": 0.4353996144928346, + "grad_norm": 0.879607617855072, + "learning_rate": 0.0007848778574630211, + "loss": 0.4541, + "step": 7793 + }, + { + "epoch": 0.4354554850965165, + "grad_norm": 0.5853348970413208, + "learning_rate": 0.0007848498431196774, + "loss": 0.3703, + "step": 7794 + }, + { + "epoch": 0.43551135570019833, + "grad_norm": 0.5265429019927979, + "learning_rate": 0.0007848218287763335, + "loss": 0.5195, + "step": 7795 + }, + { + "epoch": 0.43556722630388023, + "grad_norm": 0.6490007638931274, + "learning_rate": 0.0007847938144329897, + "loss": 0.4636, + "step": 7796 + }, + { + "epoch": 0.43562309690756207, + "grad_norm": 0.8049538731575012, + "learning_rate": 0.0007847658000896459, + "loss": 0.5647, + "step": 7797 + }, + { + "epoch": 0.43567896751124396, + "grad_norm": 0.9354789853096008, + "learning_rate": 0.0007847377857463022, + "loss": 0.5064, + "step": 7798 + }, + { + "epoch": 0.4357348381149258, + "grad_norm": 2.0743815898895264, + "learning_rate": 0.0007847097714029583, + "loss": 0.9468, + "step": 7799 + }, + { + "epoch": 0.4357907087186077, + "grad_norm": 0.7867363095283508, + "learning_rate": 0.0007846817570596145, + "loss": 0.4233, + "step": 7800 + }, + { + "epoch": 0.4358465793222896, + "grad_norm": 0.652283787727356, + "learning_rate": 0.0007846537427162708, + "loss": 0.5245, + "step": 7801 + }, + { + "epoch": 0.43590244992597144, + "grad_norm": 3.7336692810058594, + "learning_rate": 0.000784625728372927, + "loss": 0.3976, + "step": 7802 + }, + { + "epoch": 0.43595832052965333, + "grad_norm": 0.5767647624015808, + "learning_rate": 0.0007845977140295831, + "loss": 0.4825, + "step": 7803 + }, + { + "epoch": 0.43601419113333517, + "grad_norm": 0.8900271654129028, + "learning_rate": 0.0007845696996862393, + "loss": 0.5616, + "step": 7804 + }, + { + "epoch": 0.43607006173701707, + "grad_norm": 0.552810549736023, + "learning_rate": 0.0007845416853428956, + "loss": 0.4572, + "step": 7805 + }, + { + "epoch": 0.43612593234069896, + "grad_norm": 0.4933144450187683, + "learning_rate": 0.0007845136709995517, + "loss": 0.4893, + "step": 7806 + }, + { + "epoch": 0.4361818029443808, + "grad_norm": 0.5170777440071106, + "learning_rate": 0.000784485656656208, + "loss": 0.4797, + "step": 7807 + }, + { + "epoch": 0.4362376735480627, + "grad_norm": 0.44195854663848877, + "learning_rate": 0.0007844576423128642, + "loss": 0.4746, + "step": 7808 + }, + { + "epoch": 0.43629354415174454, + "grad_norm": 0.4521957337856293, + "learning_rate": 0.0007844296279695205, + "loss": 0.4218, + "step": 7809 + }, + { + "epoch": 0.43634941475542643, + "grad_norm": 0.6201589107513428, + "learning_rate": 0.0007844016136261766, + "loss": 0.4843, + "step": 7810 + }, + { + "epoch": 0.43640528535910833, + "grad_norm": 0.6758819818496704, + "learning_rate": 0.0007843735992828328, + "loss": 0.4199, + "step": 7811 + }, + { + "epoch": 0.43646115596279017, + "grad_norm": 1.1030936241149902, + "learning_rate": 0.0007843455849394891, + "loss": 0.4156, + "step": 7812 + }, + { + "epoch": 0.43651702656647207, + "grad_norm": 1.4472929239273071, + "learning_rate": 0.0007843175705961453, + "loss": 0.3425, + "step": 7813 + }, + { + "epoch": 0.4365728971701539, + "grad_norm": 0.5451103448867798, + "learning_rate": 0.0007842895562528014, + "loss": 0.5357, + "step": 7814 + }, + { + "epoch": 0.4366287677738358, + "grad_norm": 0.5627673864364624, + "learning_rate": 0.0007842615419094576, + "loss": 0.4293, + "step": 7815 + }, + { + "epoch": 0.43668463837751764, + "grad_norm": 0.42584657669067383, + "learning_rate": 0.0007842335275661139, + "loss": 0.4395, + "step": 7816 + }, + { + "epoch": 0.43674050898119954, + "grad_norm": 0.3842869997024536, + "learning_rate": 0.0007842055132227701, + "loss": 0.3606, + "step": 7817 + }, + { + "epoch": 0.43679637958488143, + "grad_norm": 0.4535905122756958, + "learning_rate": 0.0007841774988794262, + "loss": 0.4193, + "step": 7818 + }, + { + "epoch": 0.4368522501885633, + "grad_norm": 0.6157552599906921, + "learning_rate": 0.0007841494845360825, + "loss": 0.3203, + "step": 7819 + }, + { + "epoch": 0.43690812079224517, + "grad_norm": 0.5702608823776245, + "learning_rate": 0.0007841214701927387, + "loss": 0.4771, + "step": 7820 + }, + { + "epoch": 0.436963991395927, + "grad_norm": 0.6061253547668457, + "learning_rate": 0.0007840934558493949, + "loss": 0.453, + "step": 7821 + }, + { + "epoch": 0.4370198619996089, + "grad_norm": 0.40912923216819763, + "learning_rate": 0.000784065441506051, + "loss": 0.3274, + "step": 7822 + }, + { + "epoch": 0.4370757326032908, + "grad_norm": 0.8196632266044617, + "learning_rate": 0.0007840374271627073, + "loss": 0.5925, + "step": 7823 + }, + { + "epoch": 0.43713160320697264, + "grad_norm": 2.070019483566284, + "learning_rate": 0.0007840094128193636, + "loss": 0.5153, + "step": 7824 + }, + { + "epoch": 0.43718747381065454, + "grad_norm": 0.39177364110946655, + "learning_rate": 0.0007839813984760198, + "loss": 0.3662, + "step": 7825 + }, + { + "epoch": 0.4372433444143364, + "grad_norm": 0.6514804363250732, + "learning_rate": 0.000783953384132676, + "loss": 0.435, + "step": 7826 + }, + { + "epoch": 0.4372992150180183, + "grad_norm": 0.5712965726852417, + "learning_rate": 0.0007839253697893322, + "loss": 0.467, + "step": 7827 + }, + { + "epoch": 0.43735508562170017, + "grad_norm": 1.0324641466140747, + "learning_rate": 0.0007838973554459884, + "loss": 0.4312, + "step": 7828 + }, + { + "epoch": 0.437410956225382, + "grad_norm": 0.8263630270957947, + "learning_rate": 0.0007838693411026445, + "loss": 0.5702, + "step": 7829 + }, + { + "epoch": 0.4374668268290639, + "grad_norm": 1.057350754737854, + "learning_rate": 0.0007838413267593008, + "loss": 0.4182, + "step": 7830 + }, + { + "epoch": 0.43752269743274574, + "grad_norm": 0.4912106394767761, + "learning_rate": 0.000783813312415957, + "loss": 0.4436, + "step": 7831 + }, + { + "epoch": 0.43757856803642764, + "grad_norm": 0.6750321388244629, + "learning_rate": 0.0007837852980726132, + "loss": 0.4758, + "step": 7832 + }, + { + "epoch": 0.4376344386401095, + "grad_norm": 0.4723770022392273, + "learning_rate": 0.0007837572837292694, + "loss": 0.4953, + "step": 7833 + }, + { + "epoch": 0.4376903092437914, + "grad_norm": 0.5479949712753296, + "learning_rate": 0.0007837292693859256, + "loss": 0.5505, + "step": 7834 + }, + { + "epoch": 0.43774617984747327, + "grad_norm": 0.4666690230369568, + "learning_rate": 0.0007837012550425818, + "loss": 0.3119, + "step": 7835 + }, + { + "epoch": 0.4378020504511551, + "grad_norm": 2.2402453422546387, + "learning_rate": 0.000783673240699238, + "loss": 0.4953, + "step": 7836 + }, + { + "epoch": 0.437857921054837, + "grad_norm": 0.4994359612464905, + "learning_rate": 0.0007836452263558942, + "loss": 0.4303, + "step": 7837 + }, + { + "epoch": 0.43791379165851885, + "grad_norm": 0.772261917591095, + "learning_rate": 0.0007836172120125504, + "loss": 0.4577, + "step": 7838 + }, + { + "epoch": 0.43796966226220074, + "grad_norm": 0.5058759450912476, + "learning_rate": 0.0007835891976692066, + "loss": 0.4944, + "step": 7839 + }, + { + "epoch": 0.43802553286588264, + "grad_norm": 0.4668005704879761, + "learning_rate": 0.000783561183325863, + "loss": 0.3836, + "step": 7840 + }, + { + "epoch": 0.4380814034695645, + "grad_norm": 3.1909725666046143, + "learning_rate": 0.000783533168982519, + "loss": 0.5245, + "step": 7841 + }, + { + "epoch": 0.4381372740732464, + "grad_norm": 4.2246880531311035, + "learning_rate": 0.0007835051546391753, + "loss": 0.4331, + "step": 7842 + }, + { + "epoch": 0.4381931446769282, + "grad_norm": 0.6459624767303467, + "learning_rate": 0.0007834771402958315, + "loss": 0.4205, + "step": 7843 + }, + { + "epoch": 0.4382490152806101, + "grad_norm": 0.6148913502693176, + "learning_rate": 0.0007834491259524878, + "loss": 0.4305, + "step": 7844 + }, + { + "epoch": 0.438304885884292, + "grad_norm": 0.5510879755020142, + "learning_rate": 0.0007834211116091439, + "loss": 0.4139, + "step": 7845 + }, + { + "epoch": 0.43836075648797385, + "grad_norm": 1.6277180910110474, + "learning_rate": 0.0007833930972658001, + "loss": 0.5866, + "step": 7846 + }, + { + "epoch": 0.43841662709165574, + "grad_norm": 1.8073519468307495, + "learning_rate": 0.0007833650829224563, + "loss": 0.5329, + "step": 7847 + }, + { + "epoch": 0.4384724976953376, + "grad_norm": 0.8353016972541809, + "learning_rate": 0.0007833370685791126, + "loss": 0.6389, + "step": 7848 + }, + { + "epoch": 0.4385283682990195, + "grad_norm": 0.4199523329734802, + "learning_rate": 0.0007833090542357687, + "loss": 0.3355, + "step": 7849 + }, + { + "epoch": 0.4385842389027013, + "grad_norm": 0.4783954620361328, + "learning_rate": 0.0007832810398924249, + "loss": 0.4515, + "step": 7850 + }, + { + "epoch": 0.4386401095063832, + "grad_norm": 0.49449023604393005, + "learning_rate": 0.0007832530255490812, + "loss": 0.4434, + "step": 7851 + }, + { + "epoch": 0.4386959801100651, + "grad_norm": 0.4668128192424774, + "learning_rate": 0.0007832250112057373, + "loss": 0.5659, + "step": 7852 + }, + { + "epoch": 0.43875185071374695, + "grad_norm": 0.48179396986961365, + "learning_rate": 0.0007831969968623935, + "loss": 0.4586, + "step": 7853 + }, + { + "epoch": 0.43880772131742884, + "grad_norm": 4.710902214050293, + "learning_rate": 0.0007831689825190497, + "loss": 0.527, + "step": 7854 + }, + { + "epoch": 0.4388635919211107, + "grad_norm": 0.7292954325675964, + "learning_rate": 0.000783140968175706, + "loss": 0.4292, + "step": 7855 + }, + { + "epoch": 0.4389194625247926, + "grad_norm": 0.5307188630104065, + "learning_rate": 0.0007831129538323621, + "loss": 0.4492, + "step": 7856 + }, + { + "epoch": 0.4389753331284745, + "grad_norm": 0.7290502786636353, + "learning_rate": 0.0007830849394890183, + "loss": 0.4754, + "step": 7857 + }, + { + "epoch": 0.4390312037321563, + "grad_norm": 0.6243895292282104, + "learning_rate": 0.0007830569251456747, + "loss": 0.4534, + "step": 7858 + }, + { + "epoch": 0.4390870743358382, + "grad_norm": 0.8174663186073303, + "learning_rate": 0.0007830289108023309, + "loss": 0.4656, + "step": 7859 + }, + { + "epoch": 0.43914294493952005, + "grad_norm": 0.48605969548225403, + "learning_rate": 0.000783000896458987, + "loss": 0.4579, + "step": 7860 + }, + { + "epoch": 0.43919881554320195, + "grad_norm": 0.5538164973258972, + "learning_rate": 0.0007829728821156432, + "loss": 0.5264, + "step": 7861 + }, + { + "epoch": 0.43925468614688384, + "grad_norm": 0.4932382106781006, + "learning_rate": 0.0007829448677722995, + "loss": 0.4537, + "step": 7862 + }, + { + "epoch": 0.4393105567505657, + "grad_norm": 0.524013876914978, + "learning_rate": 0.0007829168534289557, + "loss": 0.5866, + "step": 7863 + }, + { + "epoch": 0.4393664273542476, + "grad_norm": 1.1099833250045776, + "learning_rate": 0.0007828888390856118, + "loss": 0.5285, + "step": 7864 + }, + { + "epoch": 0.4394222979579294, + "grad_norm": 0.8634665012359619, + "learning_rate": 0.0007828608247422681, + "loss": 0.4843, + "step": 7865 + }, + { + "epoch": 0.4394781685616113, + "grad_norm": 0.49480140209198, + "learning_rate": 0.0007828328103989243, + "loss": 0.523, + "step": 7866 + }, + { + "epoch": 0.43953403916529316, + "grad_norm": 0.4454292058944702, + "learning_rate": 0.0007828047960555805, + "loss": 0.4325, + "step": 7867 + }, + { + "epoch": 0.43958990976897505, + "grad_norm": 0.7884365916252136, + "learning_rate": 0.0007827767817122366, + "loss": 0.4596, + "step": 7868 + }, + { + "epoch": 0.43964578037265695, + "grad_norm": 0.48330187797546387, + "learning_rate": 0.0007827487673688929, + "loss": 0.4326, + "step": 7869 + }, + { + "epoch": 0.4397016509763388, + "grad_norm": 0.6814785003662109, + "learning_rate": 0.0007827207530255491, + "loss": 0.4113, + "step": 7870 + }, + { + "epoch": 0.4397575215800207, + "grad_norm": 0.8202301859855652, + "learning_rate": 0.0007826927386822052, + "loss": 0.5779, + "step": 7871 + }, + { + "epoch": 0.4398133921837025, + "grad_norm": 0.46822845935821533, + "learning_rate": 0.0007826647243388615, + "loss": 0.4623, + "step": 7872 + }, + { + "epoch": 0.4398692627873844, + "grad_norm": 0.4593704342842102, + "learning_rate": 0.0007826367099955177, + "loss": 0.5542, + "step": 7873 + }, + { + "epoch": 0.4399251333910663, + "grad_norm": 0.6360671520233154, + "learning_rate": 0.000782608695652174, + "loss": 0.4425, + "step": 7874 + }, + { + "epoch": 0.43998100399474815, + "grad_norm": 0.49936190247535706, + "learning_rate": 0.00078258068130883, + "loss": 0.409, + "step": 7875 + }, + { + "epoch": 0.44003687459843005, + "grad_norm": 0.7183742523193359, + "learning_rate": 0.0007825526669654864, + "loss": 0.4853, + "step": 7876 + }, + { + "epoch": 0.4400927452021119, + "grad_norm": 1.4659887552261353, + "learning_rate": 0.0007825246526221426, + "loss": 0.5722, + "step": 7877 + }, + { + "epoch": 0.4401486158057938, + "grad_norm": 1.0532869100570679, + "learning_rate": 0.0007824966382787988, + "loss": 0.4429, + "step": 7878 + }, + { + "epoch": 0.4402044864094757, + "grad_norm": 0.4652147889137268, + "learning_rate": 0.0007824686239354549, + "loss": 0.3864, + "step": 7879 + }, + { + "epoch": 0.4402603570131575, + "grad_norm": 1.1693874597549438, + "learning_rate": 0.0007824406095921112, + "loss": 0.4047, + "step": 7880 + }, + { + "epoch": 0.4403162276168394, + "grad_norm": 0.9290417432785034, + "learning_rate": 0.0007824125952487674, + "loss": 0.6547, + "step": 7881 + }, + { + "epoch": 0.44037209822052126, + "grad_norm": 0.5367454290390015, + "learning_rate": 0.0007823845809054236, + "loss": 0.5633, + "step": 7882 + }, + { + "epoch": 0.44042796882420315, + "grad_norm": 0.6384426951408386, + "learning_rate": 0.0007823565665620798, + "loss": 0.582, + "step": 7883 + }, + { + "epoch": 0.440483839427885, + "grad_norm": 2.2288830280303955, + "learning_rate": 0.000782328552218736, + "loss": 0.4622, + "step": 7884 + }, + { + "epoch": 0.4405397100315669, + "grad_norm": 0.5828258991241455, + "learning_rate": 0.0007823005378753922, + "loss": 0.5275, + "step": 7885 + }, + { + "epoch": 0.4405955806352488, + "grad_norm": 0.7946711778640747, + "learning_rate": 0.0007822725235320484, + "loss": 0.5744, + "step": 7886 + }, + { + "epoch": 0.4406514512389306, + "grad_norm": 0.4137624204158783, + "learning_rate": 0.0007822445091887046, + "loss": 0.4406, + "step": 7887 + }, + { + "epoch": 0.4407073218426125, + "grad_norm": 0.42304888367652893, + "learning_rate": 0.0007822164948453608, + "loss": 0.3785, + "step": 7888 + }, + { + "epoch": 0.44076319244629436, + "grad_norm": 0.6566001772880554, + "learning_rate": 0.000782188480502017, + "loss": 0.5809, + "step": 7889 + }, + { + "epoch": 0.44081906304997626, + "grad_norm": 0.5172570943832397, + "learning_rate": 0.0007821604661586734, + "loss": 0.4939, + "step": 7890 + }, + { + "epoch": 0.44087493365365815, + "grad_norm": 0.5839219689369202, + "learning_rate": 0.0007821324518153295, + "loss": 0.4417, + "step": 7891 + }, + { + "epoch": 0.44093080425734, + "grad_norm": 0.7553153038024902, + "learning_rate": 0.0007821044374719857, + "loss": 0.4455, + "step": 7892 + }, + { + "epoch": 0.4409866748610219, + "grad_norm": 0.5016595125198364, + "learning_rate": 0.0007820764231286419, + "loss": 0.4694, + "step": 7893 + }, + { + "epoch": 0.44104254546470373, + "grad_norm": 0.6183447241783142, + "learning_rate": 0.0007820484087852981, + "loss": 0.4621, + "step": 7894 + }, + { + "epoch": 0.4410984160683856, + "grad_norm": 0.4909334182739258, + "learning_rate": 0.0007820203944419543, + "loss": 0.419, + "step": 7895 + }, + { + "epoch": 0.4411542866720675, + "grad_norm": 0.4823377728462219, + "learning_rate": 0.0007819923800986105, + "loss": 0.4371, + "step": 7896 + }, + { + "epoch": 0.44121015727574936, + "grad_norm": 0.8961281776428223, + "learning_rate": 0.0007819643657552668, + "loss": 0.531, + "step": 7897 + }, + { + "epoch": 0.44126602787943126, + "grad_norm": 0.737258791923523, + "learning_rate": 0.0007819363514119229, + "loss": 0.4313, + "step": 7898 + }, + { + "epoch": 0.4413218984831131, + "grad_norm": 0.4910532236099243, + "learning_rate": 0.0007819083370685791, + "loss": 0.4362, + "step": 7899 + }, + { + "epoch": 0.441377769086795, + "grad_norm": 0.96375572681427, + "learning_rate": 0.0007818803227252353, + "loss": 0.4561, + "step": 7900 + }, + { + "epoch": 0.44143363969047683, + "grad_norm": 0.4375898838043213, + "learning_rate": 0.0007818523083818916, + "loss": 0.4224, + "step": 7901 + }, + { + "epoch": 0.4414895102941587, + "grad_norm": 0.47941991686820984, + "learning_rate": 0.0007818242940385477, + "loss": 0.6143, + "step": 7902 + }, + { + "epoch": 0.4415453808978406, + "grad_norm": 0.549765408039093, + "learning_rate": 0.0007817962796952039, + "loss": 0.4884, + "step": 7903 + }, + { + "epoch": 0.44160125150152246, + "grad_norm": 0.9462970495223999, + "learning_rate": 0.0007817682653518602, + "loss": 0.415, + "step": 7904 + }, + { + "epoch": 0.44165712210520436, + "grad_norm": 0.5564084649085999, + "learning_rate": 0.0007817402510085164, + "loss": 0.3954, + "step": 7905 + }, + { + "epoch": 0.4417129927088862, + "grad_norm": 0.4232589304447174, + "learning_rate": 0.0007817122366651725, + "loss": 0.4127, + "step": 7906 + }, + { + "epoch": 0.4417688633125681, + "grad_norm": 38.29853820800781, + "learning_rate": 0.0007816842223218287, + "loss": 0.4206, + "step": 7907 + }, + { + "epoch": 0.44182473391625, + "grad_norm": 0.4801867604255676, + "learning_rate": 0.0007816562079784851, + "loss": 0.4727, + "step": 7908 + }, + { + "epoch": 0.44188060451993183, + "grad_norm": 2.1491105556488037, + "learning_rate": 0.0007816281936351413, + "loss": 0.5783, + "step": 7909 + }, + { + "epoch": 0.4419364751236137, + "grad_norm": 0.454118549823761, + "learning_rate": 0.0007816001792917974, + "loss": 0.4858, + "step": 7910 + }, + { + "epoch": 0.44199234572729557, + "grad_norm": 0.7014228701591492, + "learning_rate": 0.0007815721649484537, + "loss": 0.4264, + "step": 7911 + }, + { + "epoch": 0.44204821633097746, + "grad_norm": 0.43942394852638245, + "learning_rate": 0.0007815441506051099, + "loss": 0.4505, + "step": 7912 + }, + { + "epoch": 0.4421040869346593, + "grad_norm": 7.314657211303711, + "learning_rate": 0.000781516136261766, + "loss": 0.5093, + "step": 7913 + }, + { + "epoch": 0.4421599575383412, + "grad_norm": 0.5486673712730408, + "learning_rate": 0.0007814881219184222, + "loss": 0.5069, + "step": 7914 + }, + { + "epoch": 0.4422158281420231, + "grad_norm": 0.5958411693572998, + "learning_rate": 0.0007814601075750785, + "loss": 0.4914, + "step": 7915 + }, + { + "epoch": 0.44227169874570493, + "grad_norm": 0.450834184885025, + "learning_rate": 0.0007814320932317347, + "loss": 0.4807, + "step": 7916 + }, + { + "epoch": 0.44232756934938683, + "grad_norm": 0.44258153438568115, + "learning_rate": 0.0007814040788883908, + "loss": 0.3904, + "step": 7917 + }, + { + "epoch": 0.44238343995306867, + "grad_norm": 0.5932272672653198, + "learning_rate": 0.000781376064545047, + "loss": 0.4439, + "step": 7918 + }, + { + "epoch": 0.44243931055675056, + "grad_norm": 0.704868733882904, + "learning_rate": 0.0007813480502017033, + "loss": 0.4216, + "step": 7919 + }, + { + "epoch": 0.44249518116043246, + "grad_norm": 1.1210800409317017, + "learning_rate": 0.0007813200358583595, + "loss": 0.4984, + "step": 7920 + }, + { + "epoch": 0.4425510517641143, + "grad_norm": 0.7728482484817505, + "learning_rate": 0.0007812920215150156, + "loss": 0.7253, + "step": 7921 + }, + { + "epoch": 0.4426069223677962, + "grad_norm": 0.5157327055931091, + "learning_rate": 0.0007812640071716719, + "loss": 0.5011, + "step": 7922 + }, + { + "epoch": 0.44266279297147804, + "grad_norm": 1.0701417922973633, + "learning_rate": 0.0007812359928283281, + "loss": 0.5642, + "step": 7923 + }, + { + "epoch": 0.44271866357515993, + "grad_norm": 0.535409152507782, + "learning_rate": 0.0007812079784849844, + "loss": 0.4137, + "step": 7924 + }, + { + "epoch": 0.4427745341788418, + "grad_norm": 0.4453178942203522, + "learning_rate": 0.0007811799641416404, + "loss": 0.4457, + "step": 7925 + }, + { + "epoch": 0.44283040478252367, + "grad_norm": 0.704255223274231, + "learning_rate": 0.0007811519497982968, + "loss": 0.5679, + "step": 7926 + }, + { + "epoch": 0.44288627538620556, + "grad_norm": 1.14179527759552, + "learning_rate": 0.000781123935454953, + "loss": 0.4547, + "step": 7927 + }, + { + "epoch": 0.4429421459898874, + "grad_norm": 0.6458470225334167, + "learning_rate": 0.0007810959211116092, + "loss": 0.44, + "step": 7928 + }, + { + "epoch": 0.4429980165935693, + "grad_norm": 0.6278243064880371, + "learning_rate": 0.0007810679067682654, + "loss": 0.4701, + "step": 7929 + }, + { + "epoch": 0.44305388719725114, + "grad_norm": 0.6086082458496094, + "learning_rate": 0.0007810398924249216, + "loss": 0.5485, + "step": 7930 + }, + { + "epoch": 0.44310975780093304, + "grad_norm": 0.7766179442405701, + "learning_rate": 0.0007810118780815778, + "loss": 0.4586, + "step": 7931 + }, + { + "epoch": 0.44316562840461493, + "grad_norm": 0.9152851104736328, + "learning_rate": 0.000780983863738234, + "loss": 0.4795, + "step": 7932 + }, + { + "epoch": 0.44322149900829677, + "grad_norm": 1.2003811597824097, + "learning_rate": 0.0007809558493948902, + "loss": 0.4101, + "step": 7933 + }, + { + "epoch": 0.44327736961197867, + "grad_norm": 0.7642979025840759, + "learning_rate": 0.0007809278350515464, + "loss": 0.4797, + "step": 7934 + }, + { + "epoch": 0.4433332402156605, + "grad_norm": 0.4137806296348572, + "learning_rate": 0.0007808998207082026, + "loss": 0.4643, + "step": 7935 + }, + { + "epoch": 0.4433891108193424, + "grad_norm": 0.7173876762390137, + "learning_rate": 0.0007808718063648588, + "loss": 0.3594, + "step": 7936 + }, + { + "epoch": 0.4434449814230243, + "grad_norm": 1.8529058694839478, + "learning_rate": 0.000780843792021515, + "loss": 0.4247, + "step": 7937 + }, + { + "epoch": 0.44350085202670614, + "grad_norm": 0.6029425263404846, + "learning_rate": 0.0007808157776781712, + "loss": 0.4348, + "step": 7938 + }, + { + "epoch": 0.44355672263038803, + "grad_norm": 1.2709530591964722, + "learning_rate": 0.0007807877633348274, + "loss": 0.55, + "step": 7939 + }, + { + "epoch": 0.4436125932340699, + "grad_norm": 2.284200668334961, + "learning_rate": 0.0007807597489914836, + "loss": 0.392, + "step": 7940 + }, + { + "epoch": 0.44366846383775177, + "grad_norm": 2.0302977561950684, + "learning_rate": 0.0007807317346481399, + "loss": 0.5167, + "step": 7941 + }, + { + "epoch": 0.44372433444143367, + "grad_norm": 0.6141613125801086, + "learning_rate": 0.0007807037203047961, + "loss": 0.4775, + "step": 7942 + }, + { + "epoch": 0.4437802050451155, + "grad_norm": 0.5750949382781982, + "learning_rate": 0.0007806757059614524, + "loss": 0.5535, + "step": 7943 + }, + { + "epoch": 0.4438360756487974, + "grad_norm": 0.5259182453155518, + "learning_rate": 0.0007806476916181085, + "loss": 0.4775, + "step": 7944 + }, + { + "epoch": 0.44389194625247924, + "grad_norm": 0.8210169672966003, + "learning_rate": 0.0007806196772747647, + "loss": 0.612, + "step": 7945 + }, + { + "epoch": 0.44394781685616114, + "grad_norm": 0.5459886193275452, + "learning_rate": 0.0007805916629314209, + "loss": 0.5466, + "step": 7946 + }, + { + "epoch": 0.444003687459843, + "grad_norm": 0.5144389867782593, + "learning_rate": 0.0007805636485880772, + "loss": 0.5203, + "step": 7947 + }, + { + "epoch": 0.4440595580635249, + "grad_norm": 0.8527531623840332, + "learning_rate": 0.0007805356342447333, + "loss": 0.5155, + "step": 7948 + }, + { + "epoch": 0.44411542866720677, + "grad_norm": 0.511607825756073, + "learning_rate": 0.0007805076199013895, + "loss": 0.5212, + "step": 7949 + }, + { + "epoch": 0.4441712992708886, + "grad_norm": 0.8030672073364258, + "learning_rate": 0.0007804796055580458, + "loss": 0.381, + "step": 7950 + }, + { + "epoch": 0.4442271698745705, + "grad_norm": 0.5442646145820618, + "learning_rate": 0.000780451591214702, + "loss": 0.5024, + "step": 7951 + }, + { + "epoch": 0.44428304047825234, + "grad_norm": 0.422220379114151, + "learning_rate": 0.0007804235768713581, + "loss": 0.4353, + "step": 7952 + }, + { + "epoch": 0.44433891108193424, + "grad_norm": 0.9808390736579895, + "learning_rate": 0.0007803955625280143, + "loss": 0.764, + "step": 7953 + }, + { + "epoch": 0.44439478168561614, + "grad_norm": 0.5067709684371948, + "learning_rate": 0.0007803675481846706, + "loss": 0.3587, + "step": 7954 + }, + { + "epoch": 0.444450652289298, + "grad_norm": 0.9889939427375793, + "learning_rate": 0.0007803395338413267, + "loss": 0.4801, + "step": 7955 + }, + { + "epoch": 0.44450652289297987, + "grad_norm": 0.6000940799713135, + "learning_rate": 0.0007803115194979829, + "loss": 0.4837, + "step": 7956 + }, + { + "epoch": 0.4445623934966617, + "grad_norm": 0.7291669249534607, + "learning_rate": 0.0007802835051546391, + "loss": 0.773, + "step": 7957 + }, + { + "epoch": 0.4446182641003436, + "grad_norm": 0.8974303603172302, + "learning_rate": 0.0007802554908112955, + "loss": 0.3582, + "step": 7958 + }, + { + "epoch": 0.4446741347040255, + "grad_norm": 0.6470269560813904, + "learning_rate": 0.0007802274764679516, + "loss": 0.3997, + "step": 7959 + }, + { + "epoch": 0.44473000530770734, + "grad_norm": 1.0419994592666626, + "learning_rate": 0.0007801994621246078, + "loss": 0.4155, + "step": 7960 + }, + { + "epoch": 0.44478587591138924, + "grad_norm": 5.476319789886475, + "learning_rate": 0.0007801714477812641, + "loss": 0.5206, + "step": 7961 + }, + { + "epoch": 0.4448417465150711, + "grad_norm": 0.5076470375061035, + "learning_rate": 0.0007801434334379203, + "loss": 0.4627, + "step": 7962 + }, + { + "epoch": 0.444897617118753, + "grad_norm": 0.6510022282600403, + "learning_rate": 0.0007801154190945764, + "loss": 0.5384, + "step": 7963 + }, + { + "epoch": 0.4449534877224348, + "grad_norm": 0.4840760827064514, + "learning_rate": 0.0007800874047512326, + "loss": 0.4427, + "step": 7964 + }, + { + "epoch": 0.4450093583261167, + "grad_norm": 0.4923135042190552, + "learning_rate": 0.0007800593904078889, + "loss": 0.4054, + "step": 7965 + }, + { + "epoch": 0.4450652289297986, + "grad_norm": 0.5873765349388123, + "learning_rate": 0.0007800313760645451, + "loss": 0.5179, + "step": 7966 + }, + { + "epoch": 0.44512109953348045, + "grad_norm": 0.6937300562858582, + "learning_rate": 0.0007800033617212012, + "loss": 0.5416, + "step": 7967 + }, + { + "epoch": 0.44517697013716234, + "grad_norm": 0.5610969662666321, + "learning_rate": 0.0007799753473778575, + "loss": 0.4551, + "step": 7968 + }, + { + "epoch": 0.4452328407408442, + "grad_norm": 0.49351391196250916, + "learning_rate": 0.0007799473330345137, + "loss": 0.4075, + "step": 7969 + }, + { + "epoch": 0.4452887113445261, + "grad_norm": 0.5411625504493713, + "learning_rate": 0.0007799193186911699, + "loss": 0.4665, + "step": 7970 + }, + { + "epoch": 0.445344581948208, + "grad_norm": 0.7768208384513855, + "learning_rate": 0.000779891304347826, + "loss": 0.4972, + "step": 7971 + }, + { + "epoch": 0.4454004525518898, + "grad_norm": 0.6550928950309753, + "learning_rate": 0.0007798632900044823, + "loss": 0.625, + "step": 7972 + }, + { + "epoch": 0.4454563231555717, + "grad_norm": 0.6898470520973206, + "learning_rate": 0.0007798352756611385, + "loss": 0.4476, + "step": 7973 + }, + { + "epoch": 0.44551219375925355, + "grad_norm": 0.5632213950157166, + "learning_rate": 0.0007798072613177948, + "loss": 0.5064, + "step": 7974 + }, + { + "epoch": 0.44556806436293545, + "grad_norm": 0.9017908573150635, + "learning_rate": 0.000779779246974451, + "loss": 0.5181, + "step": 7975 + }, + { + "epoch": 0.44562393496661734, + "grad_norm": 0.4662236273288727, + "learning_rate": 0.0007797512326311072, + "loss": 0.3797, + "step": 7976 + }, + { + "epoch": 0.4456798055702992, + "grad_norm": 0.7788488268852234, + "learning_rate": 0.0007797232182877634, + "loss": 0.4417, + "step": 7977 + }, + { + "epoch": 0.4457356761739811, + "grad_norm": 0.6283632516860962, + "learning_rate": 0.0007796952039444195, + "loss": 0.4063, + "step": 7978 + }, + { + "epoch": 0.4457915467776629, + "grad_norm": 0.5652953386306763, + "learning_rate": 0.0007796671896010758, + "loss": 0.442, + "step": 7979 + }, + { + "epoch": 0.4458474173813448, + "grad_norm": 0.5559142231941223, + "learning_rate": 0.000779639175257732, + "loss": 0.3996, + "step": 7980 + }, + { + "epoch": 0.44590328798502665, + "grad_norm": 0.8013315796852112, + "learning_rate": 0.0007796111609143882, + "loss": 0.4942, + "step": 7981 + }, + { + "epoch": 0.44595915858870855, + "grad_norm": 0.5744814872741699, + "learning_rate": 0.0007795831465710444, + "loss": 0.376, + "step": 7982 + }, + { + "epoch": 0.44601502919239044, + "grad_norm": 0.5005522966384888, + "learning_rate": 0.0007795551322277006, + "loss": 0.4793, + "step": 7983 + }, + { + "epoch": 0.4460708997960723, + "grad_norm": 0.9511809945106506, + "learning_rate": 0.0007795271178843568, + "loss": 0.5947, + "step": 7984 + }, + { + "epoch": 0.4461267703997542, + "grad_norm": 0.6295504570007324, + "learning_rate": 0.000779499103541013, + "loss": 0.4757, + "step": 7985 + }, + { + "epoch": 0.446182641003436, + "grad_norm": 0.6868298053741455, + "learning_rate": 0.0007794710891976692, + "loss": 0.4379, + "step": 7986 + }, + { + "epoch": 0.4462385116071179, + "grad_norm": 0.42851948738098145, + "learning_rate": 0.0007794430748543254, + "loss": 0.4265, + "step": 7987 + }, + { + "epoch": 0.4462943822107998, + "grad_norm": 0.5513201951980591, + "learning_rate": 0.0007794150605109816, + "loss": 0.4517, + "step": 7988 + }, + { + "epoch": 0.44635025281448165, + "grad_norm": 0.6842598915100098, + "learning_rate": 0.0007793870461676378, + "loss": 0.5519, + "step": 7989 + }, + { + "epoch": 0.44640612341816355, + "grad_norm": 0.8496996164321899, + "learning_rate": 0.000779359031824294, + "loss": 0.3929, + "step": 7990 + }, + { + "epoch": 0.4464619940218454, + "grad_norm": 0.592697024345398, + "learning_rate": 0.0007793310174809503, + "loss": 0.5474, + "step": 7991 + }, + { + "epoch": 0.4465178646255273, + "grad_norm": 0.6559323668479919, + "learning_rate": 0.0007793030031376065, + "loss": 0.4107, + "step": 7992 + }, + { + "epoch": 0.4465737352292092, + "grad_norm": 0.5140361785888672, + "learning_rate": 0.0007792749887942628, + "loss": 0.5654, + "step": 7993 + }, + { + "epoch": 0.446629605832891, + "grad_norm": 1.1213937997817993, + "learning_rate": 0.0007792469744509189, + "loss": 0.3963, + "step": 7994 + }, + { + "epoch": 0.4466854764365729, + "grad_norm": 0.5202721357345581, + "learning_rate": 0.0007792189601075751, + "loss": 0.3683, + "step": 7995 + }, + { + "epoch": 0.44674134704025475, + "grad_norm": 0.7426143288612366, + "learning_rate": 0.0007791909457642313, + "loss": 0.5075, + "step": 7996 + }, + { + "epoch": 0.44679721764393665, + "grad_norm": 1.3103926181793213, + "learning_rate": 0.0007791629314208875, + "loss": 0.4801, + "step": 7997 + }, + { + "epoch": 0.4468530882476185, + "grad_norm": 0.787112832069397, + "learning_rate": 0.0007791349170775437, + "loss": 0.585, + "step": 7998 + }, + { + "epoch": 0.4469089588513004, + "grad_norm": 0.420590877532959, + "learning_rate": 0.0007791069027341999, + "loss": 0.3569, + "step": 7999 + }, + { + "epoch": 0.4469648294549823, + "grad_norm": 0.4996187388896942, + "learning_rate": 0.0007790788883908562, + "loss": 0.3582, + "step": 8000 + }, + { + "epoch": 0.4469648294549823, + "eval_cer": 0.09321026043882906, + "eval_loss": 0.34790465235710144, + "eval_runtime": 56.5395, + "eval_samples_per_second": 80.262, + "eval_steps_per_second": 5.023, + "eval_wer": 0.369598614563839, + "step": 8000 + }, + { + "epoch": 0.4470207000586641, + "grad_norm": 1.4392036199569702, + "learning_rate": 0.0007790508740475123, + "loss": 0.5211, + "step": 8001 + }, + { + "epoch": 0.447076570662346, + "grad_norm": 4.92567777633667, + "learning_rate": 0.0007790228597041685, + "loss": 0.5348, + "step": 8002 + }, + { + "epoch": 0.44713244126602786, + "grad_norm": 0.5947796702384949, + "learning_rate": 0.0007789948453608247, + "loss": 0.4423, + "step": 8003 + }, + { + "epoch": 0.44718831186970975, + "grad_norm": 0.7216703295707703, + "learning_rate": 0.000778966831017481, + "loss": 0.5754, + "step": 8004 + }, + { + "epoch": 0.44724418247339165, + "grad_norm": 0.4521479904651642, + "learning_rate": 0.0007789388166741371, + "loss": 0.4983, + "step": 8005 + }, + { + "epoch": 0.4473000530770735, + "grad_norm": 0.5025793313980103, + "learning_rate": 0.0007789108023307933, + "loss": 0.4277, + "step": 8006 + }, + { + "epoch": 0.4473559236807554, + "grad_norm": 0.5304014086723328, + "learning_rate": 0.0007788827879874497, + "loss": 0.4114, + "step": 8007 + }, + { + "epoch": 0.4474117942844372, + "grad_norm": 6.52413272857666, + "learning_rate": 0.0007788547736441059, + "loss": 0.4259, + "step": 8008 + }, + { + "epoch": 0.4474676648881191, + "grad_norm": 0.5662716031074524, + "learning_rate": 0.000778826759300762, + "loss": 0.3887, + "step": 8009 + }, + { + "epoch": 0.447523535491801, + "grad_norm": 1.4946825504302979, + "learning_rate": 0.0007787987449574182, + "loss": 0.5935, + "step": 8010 + }, + { + "epoch": 0.44757940609548286, + "grad_norm": 0.407449334859848, + "learning_rate": 0.0007787707306140745, + "loss": 0.3944, + "step": 8011 + }, + { + "epoch": 0.44763527669916475, + "grad_norm": 0.8778036832809448, + "learning_rate": 0.0007787427162707307, + "loss": 0.5478, + "step": 8012 + }, + { + "epoch": 0.4476911473028466, + "grad_norm": 0.38912972807884216, + "learning_rate": 0.0007787147019273868, + "loss": 0.3442, + "step": 8013 + }, + { + "epoch": 0.4477470179065285, + "grad_norm": 0.41141921281814575, + "learning_rate": 0.0007786866875840431, + "loss": 0.3834, + "step": 8014 + }, + { + "epoch": 0.44780288851021033, + "grad_norm": 0.585678219795227, + "learning_rate": 0.0007786586732406993, + "loss": 0.4938, + "step": 8015 + }, + { + "epoch": 0.4478587591138922, + "grad_norm": 0.9156177639961243, + "learning_rate": 0.0007786306588973555, + "loss": 0.3523, + "step": 8016 + }, + { + "epoch": 0.4479146297175741, + "grad_norm": 0.40577417612075806, + "learning_rate": 0.0007786026445540116, + "loss": 0.3772, + "step": 8017 + }, + { + "epoch": 0.44797050032125596, + "grad_norm": 0.6540268063545227, + "learning_rate": 0.0007785746302106679, + "loss": 0.4546, + "step": 8018 + }, + { + "epoch": 0.44802637092493786, + "grad_norm": 0.4492729604244232, + "learning_rate": 0.0007785466158673241, + "loss": 0.4864, + "step": 8019 + }, + { + "epoch": 0.4480822415286197, + "grad_norm": 0.8272926807403564, + "learning_rate": 0.0007785186015239802, + "loss": 0.51, + "step": 8020 + }, + { + "epoch": 0.4481381121323016, + "grad_norm": 1.3054002523422241, + "learning_rate": 0.0007784905871806364, + "loss": 0.5596, + "step": 8021 + }, + { + "epoch": 0.4481939827359835, + "grad_norm": 0.4437592625617981, + "learning_rate": 0.0007784625728372927, + "loss": 0.357, + "step": 8022 + }, + { + "epoch": 0.4482498533396653, + "grad_norm": 0.6291231513023376, + "learning_rate": 0.0007784345584939489, + "loss": 0.4887, + "step": 8023 + }, + { + "epoch": 0.4483057239433472, + "grad_norm": 0.6051587462425232, + "learning_rate": 0.000778406544150605, + "loss": 0.4813, + "step": 8024 + }, + { + "epoch": 0.44836159454702906, + "grad_norm": 0.5539816617965698, + "learning_rate": 0.0007783785298072614, + "loss": 0.5134, + "step": 8025 + }, + { + "epoch": 0.44841746515071096, + "grad_norm": 0.42014849185943604, + "learning_rate": 0.0007783505154639176, + "loss": 0.4789, + "step": 8026 + }, + { + "epoch": 0.4484733357543928, + "grad_norm": 0.484589546918869, + "learning_rate": 0.0007783225011205738, + "loss": 0.4532, + "step": 8027 + }, + { + "epoch": 0.4485292063580747, + "grad_norm": 1.027127981185913, + "learning_rate": 0.0007782944867772299, + "loss": 0.4541, + "step": 8028 + }, + { + "epoch": 0.4485850769617566, + "grad_norm": 0.6886415481567383, + "learning_rate": 0.0007782664724338862, + "loss": 0.603, + "step": 8029 + }, + { + "epoch": 0.44864094756543843, + "grad_norm": 0.5215808153152466, + "learning_rate": 0.0007782384580905424, + "loss": 0.5064, + "step": 8030 + }, + { + "epoch": 0.4486968181691203, + "grad_norm": 0.44951131939888, + "learning_rate": 0.0007782104437471986, + "loss": 0.4917, + "step": 8031 + }, + { + "epoch": 0.44875268877280217, + "grad_norm": 1.2073651552200317, + "learning_rate": 0.0007781824294038548, + "loss": 0.4948, + "step": 8032 + }, + { + "epoch": 0.44880855937648406, + "grad_norm": 0.7643930315971375, + "learning_rate": 0.000778154415060511, + "loss": 0.4977, + "step": 8033 + }, + { + "epoch": 0.44886442998016596, + "grad_norm": 0.388230562210083, + "learning_rate": 0.0007781264007171672, + "loss": 0.4002, + "step": 8034 + }, + { + "epoch": 0.4489203005838478, + "grad_norm": 0.7097642421722412, + "learning_rate": 0.0007780983863738234, + "loss": 0.5409, + "step": 8035 + }, + { + "epoch": 0.4489761711875297, + "grad_norm": 0.41527292132377625, + "learning_rate": 0.0007780703720304796, + "loss": 0.4254, + "step": 8036 + }, + { + "epoch": 0.44903204179121153, + "grad_norm": 2.1986422538757324, + "learning_rate": 0.0007780423576871358, + "loss": 0.3891, + "step": 8037 + }, + { + "epoch": 0.44908791239489343, + "grad_norm": 2.2662041187286377, + "learning_rate": 0.000778014343343792, + "loss": 0.4343, + "step": 8038 + }, + { + "epoch": 0.4491437829985753, + "grad_norm": 0.6958847045898438, + "learning_rate": 0.0007779863290004483, + "loss": 0.5871, + "step": 8039 + }, + { + "epoch": 0.44919965360225717, + "grad_norm": 0.5303357243537903, + "learning_rate": 0.0007779583146571044, + "loss": 0.4118, + "step": 8040 + }, + { + "epoch": 0.44925552420593906, + "grad_norm": 0.8433823585510254, + "learning_rate": 0.0007779303003137606, + "loss": 0.5307, + "step": 8041 + }, + { + "epoch": 0.4493113948096209, + "grad_norm": 0.8284838795661926, + "learning_rate": 0.0007779022859704169, + "loss": 0.6238, + "step": 8042 + }, + { + "epoch": 0.4493672654133028, + "grad_norm": 0.5055535435676575, + "learning_rate": 0.0007778742716270731, + "loss": 0.5376, + "step": 8043 + }, + { + "epoch": 0.44942313601698464, + "grad_norm": 0.4274643361568451, + "learning_rate": 0.0007778462572837293, + "loss": 0.3902, + "step": 8044 + }, + { + "epoch": 0.44947900662066653, + "grad_norm": 1.0536147356033325, + "learning_rate": 0.0007778182429403855, + "loss": 0.4661, + "step": 8045 + }, + { + "epoch": 0.44953487722434843, + "grad_norm": 0.5550790429115295, + "learning_rate": 0.0007777902285970418, + "loss": 0.4004, + "step": 8046 + }, + { + "epoch": 0.44959074782803027, + "grad_norm": 0.5378637909889221, + "learning_rate": 0.0007777622142536979, + "loss": 0.5112, + "step": 8047 + }, + { + "epoch": 0.44964661843171216, + "grad_norm": 0.7249488830566406, + "learning_rate": 0.0007777341999103541, + "loss": 0.5284, + "step": 8048 + }, + { + "epoch": 0.449702489035394, + "grad_norm": 0.40668272972106934, + "learning_rate": 0.0007777061855670103, + "loss": 0.3767, + "step": 8049 + }, + { + "epoch": 0.4497583596390759, + "grad_norm": 1.02720308303833, + "learning_rate": 0.0007776781712236666, + "loss": 0.4799, + "step": 8050 + }, + { + "epoch": 0.4498142302427578, + "grad_norm": 0.585223913192749, + "learning_rate": 0.0007776501568803227, + "loss": 0.5047, + "step": 8051 + }, + { + "epoch": 0.44987010084643964, + "grad_norm": 0.6261221766471863, + "learning_rate": 0.0007776221425369789, + "loss": 0.4233, + "step": 8052 + }, + { + "epoch": 0.44992597145012153, + "grad_norm": 3.156111717224121, + "learning_rate": 0.0007775941281936352, + "loss": 0.4718, + "step": 8053 + }, + { + "epoch": 0.44998184205380337, + "grad_norm": 0.5490922331809998, + "learning_rate": 0.0007775661138502914, + "loss": 0.4493, + "step": 8054 + }, + { + "epoch": 0.45003771265748527, + "grad_norm": 0.48738738894462585, + "learning_rate": 0.0007775380995069475, + "loss": 0.5211, + "step": 8055 + }, + { + "epoch": 0.45009358326116716, + "grad_norm": 0.6228406429290771, + "learning_rate": 0.0007775100851636037, + "loss": 0.5451, + "step": 8056 + }, + { + "epoch": 0.450149453864849, + "grad_norm": 0.5049477815628052, + "learning_rate": 0.00077748207082026, + "loss": 0.5064, + "step": 8057 + }, + { + "epoch": 0.4502053244685309, + "grad_norm": 0.3756181001663208, + "learning_rate": 0.0007774540564769163, + "loss": 0.3971, + "step": 8058 + }, + { + "epoch": 0.45026119507221274, + "grad_norm": 0.7412424087524414, + "learning_rate": 0.0007774260421335724, + "loss": 0.4775, + "step": 8059 + }, + { + "epoch": 0.45031706567589463, + "grad_norm": 0.33585014939308167, + "learning_rate": 0.0007773980277902286, + "loss": 0.3481, + "step": 8060 + }, + { + "epoch": 0.4503729362795765, + "grad_norm": 1.1512696743011475, + "learning_rate": 0.0007773700134468849, + "loss": 0.4974, + "step": 8061 + }, + { + "epoch": 0.45042880688325837, + "grad_norm": 0.7091189622879028, + "learning_rate": 0.000777341999103541, + "loss": 0.5056, + "step": 8062 + }, + { + "epoch": 0.45048467748694027, + "grad_norm": 0.4950464367866516, + "learning_rate": 0.0007773139847601972, + "loss": 0.4894, + "step": 8063 + }, + { + "epoch": 0.4505405480906221, + "grad_norm": 0.5070292949676514, + "learning_rate": 0.0007772859704168535, + "loss": 0.4544, + "step": 8064 + }, + { + "epoch": 0.450596418694304, + "grad_norm": 0.34846213459968567, + "learning_rate": 0.0007772579560735097, + "loss": 0.3647, + "step": 8065 + }, + { + "epoch": 0.45065228929798584, + "grad_norm": 0.5921600461006165, + "learning_rate": 0.0007772299417301658, + "loss": 0.6068, + "step": 8066 + }, + { + "epoch": 0.45070815990166774, + "grad_norm": 1.3292644023895264, + "learning_rate": 0.000777201927386822, + "loss": 0.4492, + "step": 8067 + }, + { + "epoch": 0.45076403050534963, + "grad_norm": 0.5242418050765991, + "learning_rate": 0.0007771739130434783, + "loss": 0.4718, + "step": 8068 + }, + { + "epoch": 0.4508199011090315, + "grad_norm": 0.4818948805332184, + "learning_rate": 0.0007771458987001345, + "loss": 0.3762, + "step": 8069 + }, + { + "epoch": 0.45087577171271337, + "grad_norm": 0.8648528456687927, + "learning_rate": 0.0007771178843567906, + "loss": 0.5605, + "step": 8070 + }, + { + "epoch": 0.4509316423163952, + "grad_norm": 0.8025678396224976, + "learning_rate": 0.0007770898700134469, + "loss": 0.4368, + "step": 8071 + }, + { + "epoch": 0.4509875129200771, + "grad_norm": 1.2903923988342285, + "learning_rate": 0.0007770618556701031, + "loss": 0.5347, + "step": 8072 + }, + { + "epoch": 0.451043383523759, + "grad_norm": 0.8959177136421204, + "learning_rate": 0.0007770338413267593, + "loss": 0.5018, + "step": 8073 + }, + { + "epoch": 0.45109925412744084, + "grad_norm": 0.8317453265190125, + "learning_rate": 0.0007770058269834154, + "loss": 0.429, + "step": 8074 + }, + { + "epoch": 0.45115512473112274, + "grad_norm": 0.5357131958007812, + "learning_rate": 0.0007769778126400718, + "loss": 0.4205, + "step": 8075 + }, + { + "epoch": 0.4512109953348046, + "grad_norm": 0.6480845212936401, + "learning_rate": 0.000776949798296728, + "loss": 0.5501, + "step": 8076 + }, + { + "epoch": 0.45126686593848647, + "grad_norm": 1.056183099746704, + "learning_rate": 0.0007769217839533842, + "loss": 0.3814, + "step": 8077 + }, + { + "epoch": 0.4513227365421683, + "grad_norm": 0.6082419753074646, + "learning_rate": 0.0007768937696100404, + "loss": 0.4494, + "step": 8078 + }, + { + "epoch": 0.4513786071458502, + "grad_norm": 0.8366072177886963, + "learning_rate": 0.0007768657552666966, + "loss": 0.6891, + "step": 8079 + }, + { + "epoch": 0.4514344777495321, + "grad_norm": 0.7513146996498108, + "learning_rate": 0.0007768377409233528, + "loss": 0.6173, + "step": 8080 + }, + { + "epoch": 0.45149034835321394, + "grad_norm": 0.4889216721057892, + "learning_rate": 0.000776809726580009, + "loss": 0.465, + "step": 8081 + }, + { + "epoch": 0.45154621895689584, + "grad_norm": 0.44858917593955994, + "learning_rate": 0.0007767817122366652, + "loss": 0.4857, + "step": 8082 + }, + { + "epoch": 0.4516020895605777, + "grad_norm": 0.6358163356781006, + "learning_rate": 0.0007767536978933214, + "loss": 0.5265, + "step": 8083 + }, + { + "epoch": 0.4516579601642596, + "grad_norm": 0.3900914788246155, + "learning_rate": 0.0007767256835499776, + "loss": 0.4019, + "step": 8084 + }, + { + "epoch": 0.45171383076794147, + "grad_norm": 0.9428178071975708, + "learning_rate": 0.0007766976692066338, + "loss": 0.4914, + "step": 8085 + }, + { + "epoch": 0.4517697013716233, + "grad_norm": 1.0501978397369385, + "learning_rate": 0.00077666965486329, + "loss": 0.4756, + "step": 8086 + }, + { + "epoch": 0.4518255719753052, + "grad_norm": 0.4639809727668762, + "learning_rate": 0.0007766416405199462, + "loss": 0.3616, + "step": 8087 + }, + { + "epoch": 0.45188144257898705, + "grad_norm": 0.40257591009140015, + "learning_rate": 0.0007766136261766024, + "loss": 0.3958, + "step": 8088 + }, + { + "epoch": 0.45193731318266894, + "grad_norm": 0.5291482210159302, + "learning_rate": 0.0007765856118332586, + "loss": 0.4835, + "step": 8089 + }, + { + "epoch": 0.45199318378635084, + "grad_norm": 0.9035294651985168, + "learning_rate": 0.0007765575974899148, + "loss": 0.4611, + "step": 8090 + }, + { + "epoch": 0.4520490543900327, + "grad_norm": 0.8767934441566467, + "learning_rate": 0.000776529583146571, + "loss": 0.5711, + "step": 8091 + }, + { + "epoch": 0.4521049249937146, + "grad_norm": 0.547916054725647, + "learning_rate": 0.0007765015688032274, + "loss": 0.4426, + "step": 8092 + }, + { + "epoch": 0.4521607955973964, + "grad_norm": 0.5564181208610535, + "learning_rate": 0.0007764735544598835, + "loss": 0.4627, + "step": 8093 + }, + { + "epoch": 0.4522166662010783, + "grad_norm": 0.5539538264274597, + "learning_rate": 0.0007764455401165397, + "loss": 0.4651, + "step": 8094 + }, + { + "epoch": 0.45227253680476015, + "grad_norm": 0.5042906999588013, + "learning_rate": 0.0007764175257731959, + "loss": 0.5314, + "step": 8095 + }, + { + "epoch": 0.45232840740844205, + "grad_norm": 0.5499275326728821, + "learning_rate": 0.0007763895114298522, + "loss": 0.5324, + "step": 8096 + }, + { + "epoch": 0.45238427801212394, + "grad_norm": 0.7184106111526489, + "learning_rate": 0.0007763614970865083, + "loss": 0.4262, + "step": 8097 + }, + { + "epoch": 0.4524401486158058, + "grad_norm": 0.9336539506912231, + "learning_rate": 0.0007763334827431645, + "loss": 0.4064, + "step": 8098 + }, + { + "epoch": 0.4524960192194877, + "grad_norm": 0.45797494053840637, + "learning_rate": 0.0007763054683998207, + "loss": 0.3138, + "step": 8099 + }, + { + "epoch": 0.4525518898231695, + "grad_norm": 12.495793342590332, + "learning_rate": 0.000776277454056477, + "loss": 0.4922, + "step": 8100 + }, + { + "epoch": 0.4526077604268514, + "grad_norm": 0.6494421362876892, + "learning_rate": 0.0007762494397131331, + "loss": 0.5561, + "step": 8101 + }, + { + "epoch": 0.4526636310305333, + "grad_norm": 0.8632330298423767, + "learning_rate": 0.0007762214253697893, + "loss": 0.4894, + "step": 8102 + }, + { + "epoch": 0.45271950163421515, + "grad_norm": 0.7152793407440186, + "learning_rate": 0.0007761934110264456, + "loss": 0.469, + "step": 8103 + }, + { + "epoch": 0.45277537223789704, + "grad_norm": 0.5792052149772644, + "learning_rate": 0.0007761653966831017, + "loss": 0.4889, + "step": 8104 + }, + { + "epoch": 0.4528312428415789, + "grad_norm": 0.7978283166885376, + "learning_rate": 0.0007761373823397579, + "loss": 0.5781, + "step": 8105 + }, + { + "epoch": 0.4528871134452608, + "grad_norm": 3.558727502822876, + "learning_rate": 0.0007761093679964141, + "loss": 0.4816, + "step": 8106 + }, + { + "epoch": 0.4529429840489427, + "grad_norm": 0.6753368377685547, + "learning_rate": 0.0007760813536530704, + "loss": 0.4636, + "step": 8107 + }, + { + "epoch": 0.4529988546526245, + "grad_norm": 0.862029492855072, + "learning_rate": 0.0007760533393097265, + "loss": 0.394, + "step": 8108 + }, + { + "epoch": 0.4530547252563064, + "grad_norm": 0.8385518789291382, + "learning_rate": 0.0007760253249663828, + "loss": 0.4099, + "step": 8109 + }, + { + "epoch": 0.45311059585998825, + "grad_norm": 0.39902034401893616, + "learning_rate": 0.0007759973106230391, + "loss": 0.4275, + "step": 8110 + }, + { + "epoch": 0.45316646646367015, + "grad_norm": 0.5388319492340088, + "learning_rate": 0.0007759692962796953, + "loss": 0.5198, + "step": 8111 + }, + { + "epoch": 0.453222337067352, + "grad_norm": 0.6860141158103943, + "learning_rate": 0.0007759412819363514, + "loss": 0.455, + "step": 8112 + }, + { + "epoch": 0.4532782076710339, + "grad_norm": 0.7477019429206848, + "learning_rate": 0.0007759132675930076, + "loss": 0.507, + "step": 8113 + }, + { + "epoch": 0.4533340782747158, + "grad_norm": 0.8091833591461182, + "learning_rate": 0.0007758852532496639, + "loss": 0.4771, + "step": 8114 + }, + { + "epoch": 0.4533899488783976, + "grad_norm": 0.6076411604881287, + "learning_rate": 0.0007758572389063201, + "loss": 0.5718, + "step": 8115 + }, + { + "epoch": 0.4534458194820795, + "grad_norm": 0.49227389693260193, + "learning_rate": 0.0007758292245629762, + "loss": 0.4649, + "step": 8116 + }, + { + "epoch": 0.45350169008576136, + "grad_norm": 0.5947369337081909, + "learning_rate": 0.0007758012102196325, + "loss": 0.4496, + "step": 8117 + }, + { + "epoch": 0.45355756068944325, + "grad_norm": 0.4705881178379059, + "learning_rate": 0.0007757731958762887, + "loss": 0.435, + "step": 8118 + }, + { + "epoch": 0.45361343129312515, + "grad_norm": 2.0239665508270264, + "learning_rate": 0.0007757451815329449, + "loss": 0.4613, + "step": 8119 + }, + { + "epoch": 0.453669301896807, + "grad_norm": 0.5947259664535522, + "learning_rate": 0.000775717167189601, + "loss": 0.4107, + "step": 8120 + }, + { + "epoch": 0.4537251725004889, + "grad_norm": 0.9771949052810669, + "learning_rate": 0.0007756891528462573, + "loss": 0.4872, + "step": 8121 + }, + { + "epoch": 0.4537810431041707, + "grad_norm": 0.7677519917488098, + "learning_rate": 0.0007756611385029135, + "loss": 0.5822, + "step": 8122 + }, + { + "epoch": 0.4538369137078526, + "grad_norm": 0.51595139503479, + "learning_rate": 0.0007756331241595697, + "loss": 0.4005, + "step": 8123 + }, + { + "epoch": 0.45389278431153446, + "grad_norm": 0.5770518183708191, + "learning_rate": 0.000775605109816226, + "loss": 0.5382, + "step": 8124 + }, + { + "epoch": 0.45394865491521635, + "grad_norm": 0.48722589015960693, + "learning_rate": 0.0007755770954728822, + "loss": 0.3748, + "step": 8125 + }, + { + "epoch": 0.45400452551889825, + "grad_norm": 0.5913227200508118, + "learning_rate": 0.0007755490811295384, + "loss": 0.4756, + "step": 8126 + }, + { + "epoch": 0.4540603961225801, + "grad_norm": 0.5925344228744507, + "learning_rate": 0.0007755210667861945, + "loss": 0.4373, + "step": 8127 + }, + { + "epoch": 0.454116266726262, + "grad_norm": 0.5861858129501343, + "learning_rate": 0.0007754930524428508, + "loss": 0.5156, + "step": 8128 + }, + { + "epoch": 0.4541721373299438, + "grad_norm": 0.9243488311767578, + "learning_rate": 0.000775465038099507, + "loss": 0.5098, + "step": 8129 + }, + { + "epoch": 0.4542280079336257, + "grad_norm": 0.6399312019348145, + "learning_rate": 0.0007754370237561632, + "loss": 0.4899, + "step": 8130 + }, + { + "epoch": 0.4542838785373076, + "grad_norm": 1.263217806816101, + "learning_rate": 0.0007754090094128193, + "loss": 0.6301, + "step": 8131 + }, + { + "epoch": 0.45433974914098946, + "grad_norm": 0.3514134883880615, + "learning_rate": 0.0007753809950694756, + "loss": 0.2992, + "step": 8132 + }, + { + "epoch": 0.45439561974467135, + "grad_norm": 0.4696384072303772, + "learning_rate": 0.0007753529807261318, + "loss": 0.5063, + "step": 8133 + }, + { + "epoch": 0.4544514903483532, + "grad_norm": 2.875499963760376, + "learning_rate": 0.000775324966382788, + "loss": 0.4294, + "step": 8134 + }, + { + "epoch": 0.4545073609520351, + "grad_norm": 0.7450470924377441, + "learning_rate": 0.0007752969520394442, + "loss": 0.4789, + "step": 8135 + }, + { + "epoch": 0.454563231555717, + "grad_norm": 0.6961474418640137, + "learning_rate": 0.0007752689376961004, + "loss": 0.5326, + "step": 8136 + }, + { + "epoch": 0.4546191021593988, + "grad_norm": 0.42100790143013, + "learning_rate": 0.0007752409233527566, + "loss": 0.3837, + "step": 8137 + }, + { + "epoch": 0.4546749727630807, + "grad_norm": 0.5880357027053833, + "learning_rate": 0.0007752129090094128, + "loss": 0.4698, + "step": 8138 + }, + { + "epoch": 0.45473084336676256, + "grad_norm": 0.43983855843544006, + "learning_rate": 0.000775184894666069, + "loss": 0.4507, + "step": 8139 + }, + { + "epoch": 0.45478671397044446, + "grad_norm": 0.49879947304725647, + "learning_rate": 0.0007751568803227252, + "loss": 0.5163, + "step": 8140 + }, + { + "epoch": 0.4548425845741263, + "grad_norm": 0.9277803301811218, + "learning_rate": 0.0007751288659793814, + "loss": 0.378, + "step": 8141 + }, + { + "epoch": 0.4548984551778082, + "grad_norm": 1.0406291484832764, + "learning_rate": 0.0007751008516360378, + "loss": 0.4843, + "step": 8142 + }, + { + "epoch": 0.4549543257814901, + "grad_norm": 0.6525417566299438, + "learning_rate": 0.0007750728372926939, + "loss": 0.5341, + "step": 8143 + }, + { + "epoch": 0.45501019638517193, + "grad_norm": 0.6345599293708801, + "learning_rate": 0.0007750448229493501, + "loss": 0.4043, + "step": 8144 + }, + { + "epoch": 0.4550660669888538, + "grad_norm": 0.5419347286224365, + "learning_rate": 0.0007750168086060063, + "loss": 0.4369, + "step": 8145 + }, + { + "epoch": 0.45512193759253566, + "grad_norm": 0.4220944344997406, + "learning_rate": 0.0007749887942626625, + "loss": 0.453, + "step": 8146 + }, + { + "epoch": 0.45517780819621756, + "grad_norm": 0.6264843344688416, + "learning_rate": 0.0007749607799193187, + "loss": 0.4516, + "step": 8147 + }, + { + "epoch": 0.45523367879989945, + "grad_norm": 0.3116769790649414, + "learning_rate": 0.0007749327655759749, + "loss": 0.3667, + "step": 8148 + }, + { + "epoch": 0.4552895494035813, + "grad_norm": 0.35666951537132263, + "learning_rate": 0.0007749047512326312, + "loss": 0.3813, + "step": 8149 + }, + { + "epoch": 0.4553454200072632, + "grad_norm": 1.065974473953247, + "learning_rate": 0.0007748767368892873, + "loss": 0.4932, + "step": 8150 + }, + { + "epoch": 0.45540129061094503, + "grad_norm": 0.4811967611312866, + "learning_rate": 0.0007748487225459435, + "loss": 0.4673, + "step": 8151 + }, + { + "epoch": 0.4554571612146269, + "grad_norm": 0.5140681266784668, + "learning_rate": 0.0007748207082025997, + "loss": 0.5177, + "step": 8152 + }, + { + "epoch": 0.4555130318183088, + "grad_norm": 0.7629656791687012, + "learning_rate": 0.000774792693859256, + "loss": 0.4596, + "step": 8153 + }, + { + "epoch": 0.45556890242199066, + "grad_norm": 1.7971687316894531, + "learning_rate": 0.0007747646795159121, + "loss": 0.3576, + "step": 8154 + }, + { + "epoch": 0.45562477302567256, + "grad_norm": 0.761243999004364, + "learning_rate": 0.0007747366651725683, + "loss": 0.4487, + "step": 8155 + }, + { + "epoch": 0.4556806436293544, + "grad_norm": 0.5183126330375671, + "learning_rate": 0.0007747086508292246, + "loss": 0.4505, + "step": 8156 + }, + { + "epoch": 0.4557365142330363, + "grad_norm": 0.7913049459457397, + "learning_rate": 0.0007746806364858808, + "loss": 0.4235, + "step": 8157 + }, + { + "epoch": 0.45579238483671813, + "grad_norm": 0.6263706684112549, + "learning_rate": 0.000774652622142537, + "loss": 0.538, + "step": 8158 + }, + { + "epoch": 0.45584825544040003, + "grad_norm": 0.5356819033622742, + "learning_rate": 0.0007746246077991932, + "loss": 0.651, + "step": 8159 + }, + { + "epoch": 0.4559041260440819, + "grad_norm": 1.3614507913589478, + "learning_rate": 0.0007745965934558495, + "loss": 0.473, + "step": 8160 + }, + { + "epoch": 0.45595999664776377, + "grad_norm": 0.41016724705696106, + "learning_rate": 0.0007745685791125057, + "loss": 0.3947, + "step": 8161 + }, + { + "epoch": 0.45601586725144566, + "grad_norm": 1.9823416471481323, + "learning_rate": 0.0007745405647691618, + "loss": 0.5012, + "step": 8162 + }, + { + "epoch": 0.4560717378551275, + "grad_norm": 0.5029736757278442, + "learning_rate": 0.0007745125504258181, + "loss": 0.4342, + "step": 8163 + }, + { + "epoch": 0.4561276084588094, + "grad_norm": 0.5901368260383606, + "learning_rate": 0.0007744845360824743, + "loss": 0.4397, + "step": 8164 + }, + { + "epoch": 0.4561834790624913, + "grad_norm": 0.5624266266822815, + "learning_rate": 0.0007744565217391305, + "loss": 0.5438, + "step": 8165 + }, + { + "epoch": 0.45623934966617313, + "grad_norm": 0.43325504660606384, + "learning_rate": 0.0007744285073957866, + "loss": 0.3649, + "step": 8166 + }, + { + "epoch": 0.45629522026985503, + "grad_norm": 3.75466251373291, + "learning_rate": 0.0007744004930524429, + "loss": 0.7023, + "step": 8167 + }, + { + "epoch": 0.45635109087353687, + "grad_norm": 0.46257591247558594, + "learning_rate": 0.0007743724787090991, + "loss": 0.3833, + "step": 8168 + }, + { + "epoch": 0.45640696147721876, + "grad_norm": 0.5828871130943298, + "learning_rate": 0.0007743444643657552, + "loss": 0.5805, + "step": 8169 + }, + { + "epoch": 0.45646283208090066, + "grad_norm": 0.44198817014694214, + "learning_rate": 0.0007743164500224114, + "loss": 0.4299, + "step": 8170 + }, + { + "epoch": 0.4565187026845825, + "grad_norm": 0.556422770023346, + "learning_rate": 0.0007742884356790677, + "loss": 0.4773, + "step": 8171 + }, + { + "epoch": 0.4565745732882644, + "grad_norm": 0.5459975600242615, + "learning_rate": 0.0007742604213357239, + "loss": 0.4707, + "step": 8172 + }, + { + "epoch": 0.45663044389194624, + "grad_norm": 1.7838716506958008, + "learning_rate": 0.00077423240699238, + "loss": 0.5298, + "step": 8173 + }, + { + "epoch": 0.45668631449562813, + "grad_norm": 0.4103209674358368, + "learning_rate": 0.0007742043926490363, + "loss": 0.419, + "step": 8174 + }, + { + "epoch": 0.45674218509930997, + "grad_norm": 0.49292564392089844, + "learning_rate": 0.0007741763783056926, + "loss": 0.4572, + "step": 8175 + }, + { + "epoch": 0.45679805570299187, + "grad_norm": 0.4638883173465729, + "learning_rate": 0.0007741483639623488, + "loss": 0.4437, + "step": 8176 + }, + { + "epoch": 0.45685392630667376, + "grad_norm": 0.646161675453186, + "learning_rate": 0.0007741203496190049, + "loss": 0.3459, + "step": 8177 + }, + { + "epoch": 0.4569097969103556, + "grad_norm": 0.7570212483406067, + "learning_rate": 0.0007740923352756612, + "loss": 0.4479, + "step": 8178 + }, + { + "epoch": 0.4569656675140375, + "grad_norm": 0.5581491589546204, + "learning_rate": 0.0007740643209323174, + "loss": 0.4408, + "step": 8179 + }, + { + "epoch": 0.45702153811771934, + "grad_norm": 0.6426613926887512, + "learning_rate": 0.0007740363065889736, + "loss": 0.4255, + "step": 8180 + }, + { + "epoch": 0.45707740872140123, + "grad_norm": 0.6624823212623596, + "learning_rate": 0.0007740082922456298, + "loss": 0.6077, + "step": 8181 + }, + { + "epoch": 0.45713327932508313, + "grad_norm": 0.7615561485290527, + "learning_rate": 0.000773980277902286, + "loss": 0.3497, + "step": 8182 + }, + { + "epoch": 0.45718914992876497, + "grad_norm": 0.5040482878684998, + "learning_rate": 0.0007739522635589422, + "loss": 0.3872, + "step": 8183 + }, + { + "epoch": 0.45724502053244687, + "grad_norm": 0.39506110548973083, + "learning_rate": 0.0007739242492155984, + "loss": 0.4261, + "step": 8184 + }, + { + "epoch": 0.4573008911361287, + "grad_norm": 0.5818336009979248, + "learning_rate": 0.0007738962348722546, + "loss": 0.4913, + "step": 8185 + }, + { + "epoch": 0.4573567617398106, + "grad_norm": 0.6220343708992004, + "learning_rate": 0.0007738682205289108, + "loss": 0.4727, + "step": 8186 + }, + { + "epoch": 0.4574126323434925, + "grad_norm": 0.498084157705307, + "learning_rate": 0.000773840206185567, + "loss": 0.4791, + "step": 8187 + }, + { + "epoch": 0.45746850294717434, + "grad_norm": 0.48435091972351074, + "learning_rate": 0.0007738121918422233, + "loss": 0.4052, + "step": 8188 + }, + { + "epoch": 0.45752437355085623, + "grad_norm": 0.6058427691459656, + "learning_rate": 0.0007737841774988794, + "loss": 0.5, + "step": 8189 + }, + { + "epoch": 0.4575802441545381, + "grad_norm": 1.2231088876724243, + "learning_rate": 0.0007737561631555356, + "loss": 0.5975, + "step": 8190 + }, + { + "epoch": 0.45763611475821997, + "grad_norm": 0.4492492079734802, + "learning_rate": 0.0007737281488121918, + "loss": 0.4344, + "step": 8191 + }, + { + "epoch": 0.4576919853619018, + "grad_norm": 0.8306164741516113, + "learning_rate": 0.000773700134468848, + "loss": 0.683, + "step": 8192 + }, + { + "epoch": 0.4577478559655837, + "grad_norm": 0.40699878334999084, + "learning_rate": 0.0007736721201255043, + "loss": 0.4608, + "step": 8193 + }, + { + "epoch": 0.4578037265692656, + "grad_norm": 1.3946070671081543, + "learning_rate": 0.0007736441057821605, + "loss": 0.3777, + "step": 8194 + }, + { + "epoch": 0.45785959717294744, + "grad_norm": 0.5971101522445679, + "learning_rate": 0.0007736160914388168, + "loss": 0.5909, + "step": 8195 + }, + { + "epoch": 0.45791546777662934, + "grad_norm": 0.889625072479248, + "learning_rate": 0.0007735880770954729, + "loss": 0.5404, + "step": 8196 + }, + { + "epoch": 0.4579713383803112, + "grad_norm": 0.559731662273407, + "learning_rate": 0.0007735600627521291, + "loss": 0.4046, + "step": 8197 + }, + { + "epoch": 0.4580272089839931, + "grad_norm": 1.5335882902145386, + "learning_rate": 0.0007735320484087853, + "loss": 0.5472, + "step": 8198 + }, + { + "epoch": 0.45808307958767497, + "grad_norm": 0.6497212052345276, + "learning_rate": 0.0007735040340654416, + "loss": 0.6109, + "step": 8199 + }, + { + "epoch": 0.4581389501913568, + "grad_norm": 1.1712278127670288, + "learning_rate": 0.0007734760197220977, + "loss": 0.3824, + "step": 8200 + }, + { + "epoch": 0.4581948207950387, + "grad_norm": 0.6004265546798706, + "learning_rate": 0.0007734480053787539, + "loss": 0.6629, + "step": 8201 + }, + { + "epoch": 0.45825069139872054, + "grad_norm": 0.4282049536705017, + "learning_rate": 0.0007734199910354101, + "loss": 0.3501, + "step": 8202 + }, + { + "epoch": 0.45830656200240244, + "grad_norm": 0.4651907980442047, + "learning_rate": 0.0007733919766920664, + "loss": 0.4676, + "step": 8203 + }, + { + "epoch": 0.45836243260608434, + "grad_norm": 1.0782241821289062, + "learning_rate": 0.0007733639623487225, + "loss": 0.5481, + "step": 8204 + }, + { + "epoch": 0.4584183032097662, + "grad_norm": 0.7011036276817322, + "learning_rate": 0.0007733359480053787, + "loss": 0.466, + "step": 8205 + }, + { + "epoch": 0.45847417381344807, + "grad_norm": 0.594673752784729, + "learning_rate": 0.000773307933662035, + "loss": 0.6609, + "step": 8206 + }, + { + "epoch": 0.4585300444171299, + "grad_norm": 0.683305025100708, + "learning_rate": 0.0007732799193186912, + "loss": 0.6323, + "step": 8207 + }, + { + "epoch": 0.4585859150208118, + "grad_norm": 0.7980417609214783, + "learning_rate": 0.0007732519049753473, + "loss": 0.4477, + "step": 8208 + }, + { + "epoch": 0.45864178562449365, + "grad_norm": 0.6444413065910339, + "learning_rate": 0.0007732238906320036, + "loss": 0.4548, + "step": 8209 + }, + { + "epoch": 0.45869765622817554, + "grad_norm": 0.4794323742389679, + "learning_rate": 0.0007731958762886599, + "loss": 0.5526, + "step": 8210 + }, + { + "epoch": 0.45875352683185744, + "grad_norm": 3.138261079788208, + "learning_rate": 0.000773167861945316, + "loss": 0.4474, + "step": 8211 + }, + { + "epoch": 0.4588093974355393, + "grad_norm": 0.6292574405670166, + "learning_rate": 0.0007731398476019722, + "loss": 0.5224, + "step": 8212 + }, + { + "epoch": 0.4588652680392212, + "grad_norm": 0.7984927892684937, + "learning_rate": 0.0007731118332586285, + "loss": 0.633, + "step": 8213 + }, + { + "epoch": 0.458921138642903, + "grad_norm": 0.6685599088668823, + "learning_rate": 0.0007730838189152847, + "loss": 0.5468, + "step": 8214 + }, + { + "epoch": 0.4589770092465849, + "grad_norm": 0.8274825811386108, + "learning_rate": 0.0007730558045719408, + "loss": 0.4189, + "step": 8215 + }, + { + "epoch": 0.4590328798502668, + "grad_norm": 0.5826215147972107, + "learning_rate": 0.000773027790228597, + "loss": 0.531, + "step": 8216 + }, + { + "epoch": 0.45908875045394865, + "grad_norm": 0.6010775566101074, + "learning_rate": 0.0007729997758852533, + "loss": 0.5631, + "step": 8217 + }, + { + "epoch": 0.45914462105763054, + "grad_norm": 0.504357099533081, + "learning_rate": 0.0007729717615419095, + "loss": 0.3888, + "step": 8218 + }, + { + "epoch": 0.4592004916613124, + "grad_norm": 0.557871401309967, + "learning_rate": 0.0007729437471985656, + "loss": 0.3696, + "step": 8219 + }, + { + "epoch": 0.4592563622649943, + "grad_norm": 0.5744185447692871, + "learning_rate": 0.0007729157328552219, + "loss": 0.5066, + "step": 8220 + }, + { + "epoch": 0.4593122328686762, + "grad_norm": 0.5503948926925659, + "learning_rate": 0.0007728877185118781, + "loss": 0.4786, + "step": 8221 + }, + { + "epoch": 0.459368103472358, + "grad_norm": 3.3097870349884033, + "learning_rate": 0.0007728597041685343, + "loss": 0.4674, + "step": 8222 + }, + { + "epoch": 0.4594239740760399, + "grad_norm": 0.43693968653678894, + "learning_rate": 0.0007728316898251904, + "loss": 0.4053, + "step": 8223 + }, + { + "epoch": 0.45947984467972175, + "grad_norm": 0.5603408217430115, + "learning_rate": 0.0007728036754818467, + "loss": 0.3309, + "step": 8224 + }, + { + "epoch": 0.45953571528340365, + "grad_norm": 1.0255197286605835, + "learning_rate": 0.000772775661138503, + "loss": 0.6016, + "step": 8225 + }, + { + "epoch": 0.4595915858870855, + "grad_norm": 0.963461697101593, + "learning_rate": 0.0007727476467951592, + "loss": 0.4463, + "step": 8226 + }, + { + "epoch": 0.4596474564907674, + "grad_norm": 4.656682968139648, + "learning_rate": 0.0007727196324518154, + "loss": 0.3398, + "step": 8227 + }, + { + "epoch": 0.4597033270944493, + "grad_norm": 0.9072276949882507, + "learning_rate": 0.0007726916181084716, + "loss": 0.4556, + "step": 8228 + }, + { + "epoch": 0.4597591976981311, + "grad_norm": 0.6587374806404114, + "learning_rate": 0.0007726636037651278, + "loss": 0.6076, + "step": 8229 + }, + { + "epoch": 0.459815068301813, + "grad_norm": 0.7167946100234985, + "learning_rate": 0.000772635589421784, + "loss": 0.4476, + "step": 8230 + }, + { + "epoch": 0.45987093890549485, + "grad_norm": 0.5401821732521057, + "learning_rate": 0.0007726075750784402, + "loss": 0.4823, + "step": 8231 + }, + { + "epoch": 0.45992680950917675, + "grad_norm": 0.4604087173938751, + "learning_rate": 0.0007725795607350964, + "loss": 0.4106, + "step": 8232 + }, + { + "epoch": 0.45998268011285864, + "grad_norm": 0.7488629817962646, + "learning_rate": 0.0007725515463917526, + "loss": 0.4405, + "step": 8233 + }, + { + "epoch": 0.4600385507165405, + "grad_norm": 0.3567279577255249, + "learning_rate": 0.0007725235320484087, + "loss": 0.3919, + "step": 8234 + }, + { + "epoch": 0.4600944213202224, + "grad_norm": 0.7876489758491516, + "learning_rate": 0.000772495517705065, + "loss": 0.5217, + "step": 8235 + }, + { + "epoch": 0.4601502919239042, + "grad_norm": 0.5616865754127502, + "learning_rate": 0.0007724675033617212, + "loss": 0.4517, + "step": 8236 + }, + { + "epoch": 0.4602061625275861, + "grad_norm": 0.49177590012550354, + "learning_rate": 0.0007724394890183774, + "loss": 0.4973, + "step": 8237 + }, + { + "epoch": 0.46026203313126796, + "grad_norm": 0.648530125617981, + "learning_rate": 0.0007724114746750336, + "loss": 0.4208, + "step": 8238 + }, + { + "epoch": 0.46031790373494985, + "grad_norm": 6.211165904998779, + "learning_rate": 0.0007723834603316898, + "loss": 0.5795, + "step": 8239 + }, + { + "epoch": 0.46037377433863175, + "grad_norm": 0.6680028438568115, + "learning_rate": 0.000772355445988346, + "loss": 0.4823, + "step": 8240 + }, + { + "epoch": 0.4604296449423136, + "grad_norm": 0.4257245659828186, + "learning_rate": 0.0007723274316450022, + "loss": 0.3583, + "step": 8241 + }, + { + "epoch": 0.4604855155459955, + "grad_norm": 0.6422126889228821, + "learning_rate": 0.0007722994173016585, + "loss": 0.3695, + "step": 8242 + }, + { + "epoch": 0.4605413861496773, + "grad_norm": 0.4840780794620514, + "learning_rate": 0.0007722714029583147, + "loss": 0.4009, + "step": 8243 + }, + { + "epoch": 0.4605972567533592, + "grad_norm": 15.197183609008789, + "learning_rate": 0.0007722433886149709, + "loss": 0.4503, + "step": 8244 + }, + { + "epoch": 0.4606531273570411, + "grad_norm": 0.573782742023468, + "learning_rate": 0.0007722153742716272, + "loss": 0.3427, + "step": 8245 + }, + { + "epoch": 0.46070899796072295, + "grad_norm": 0.7925040125846863, + "learning_rate": 0.0007721873599282833, + "loss": 0.5104, + "step": 8246 + }, + { + "epoch": 0.46076486856440485, + "grad_norm": 0.7711607813835144, + "learning_rate": 0.0007721593455849395, + "loss": 0.5332, + "step": 8247 + }, + { + "epoch": 0.4608207391680867, + "grad_norm": 1.215278148651123, + "learning_rate": 0.0007721313312415957, + "loss": 0.6413, + "step": 8248 + }, + { + "epoch": 0.4608766097717686, + "grad_norm": 0.683078408241272, + "learning_rate": 0.000772103316898252, + "loss": 0.4561, + "step": 8249 + }, + { + "epoch": 0.4609324803754505, + "grad_norm": 0.5132353901863098, + "learning_rate": 0.0007720753025549081, + "loss": 0.4249, + "step": 8250 + }, + { + "epoch": 0.4609883509791323, + "grad_norm": 0.3757503926753998, + "learning_rate": 0.0007720472882115643, + "loss": 0.4, + "step": 8251 + }, + { + "epoch": 0.4610442215828142, + "grad_norm": 1.3180315494537354, + "learning_rate": 0.0007720192738682206, + "loss": 0.541, + "step": 8252 + }, + { + "epoch": 0.46110009218649606, + "grad_norm": 0.5565733313560486, + "learning_rate": 0.0007719912595248767, + "loss": 0.5032, + "step": 8253 + }, + { + "epoch": 0.46115596279017795, + "grad_norm": 0.44627735018730164, + "learning_rate": 0.0007719632451815329, + "loss": 0.3728, + "step": 8254 + }, + { + "epoch": 0.4612118333938598, + "grad_norm": 0.46313992142677307, + "learning_rate": 0.0007719352308381891, + "loss": 0.3992, + "step": 8255 + }, + { + "epoch": 0.4612677039975417, + "grad_norm": 0.5721021890640259, + "learning_rate": 0.0007719072164948454, + "loss": 0.4868, + "step": 8256 + }, + { + "epoch": 0.4613235746012236, + "grad_norm": 0.5018796920776367, + "learning_rate": 0.0007718792021515015, + "loss": 0.4936, + "step": 8257 + }, + { + "epoch": 0.4613794452049054, + "grad_norm": 0.5795652270317078, + "learning_rate": 0.0007718511878081577, + "loss": 0.4786, + "step": 8258 + }, + { + "epoch": 0.4614353158085873, + "grad_norm": 0.5268062353134155, + "learning_rate": 0.0007718231734648141, + "loss": 0.5278, + "step": 8259 + }, + { + "epoch": 0.46149118641226916, + "grad_norm": 0.5816810131072998, + "learning_rate": 0.0007717951591214703, + "loss": 0.5753, + "step": 8260 + }, + { + "epoch": 0.46154705701595106, + "grad_norm": 0.520905613899231, + "learning_rate": 0.0007717671447781264, + "loss": 0.5547, + "step": 8261 + }, + { + "epoch": 0.46160292761963295, + "grad_norm": 0.7159306406974792, + "learning_rate": 0.0007717391304347826, + "loss": 0.369, + "step": 8262 + }, + { + "epoch": 0.4616587982233148, + "grad_norm": 0.8719356656074524, + "learning_rate": 0.0007717111160914389, + "loss": 0.5079, + "step": 8263 + }, + { + "epoch": 0.4617146688269967, + "grad_norm": 0.38251110911369324, + "learning_rate": 0.0007716831017480951, + "loss": 0.4692, + "step": 8264 + }, + { + "epoch": 0.46177053943067853, + "grad_norm": 0.36945322155952454, + "learning_rate": 0.0007716550874047512, + "loss": 0.4179, + "step": 8265 + }, + { + "epoch": 0.4618264100343604, + "grad_norm": 0.5172230005264282, + "learning_rate": 0.0007716270730614075, + "loss": 0.5019, + "step": 8266 + }, + { + "epoch": 0.4618822806380423, + "grad_norm": 0.6776398420333862, + "learning_rate": 0.0007715990587180637, + "loss": 0.495, + "step": 8267 + }, + { + "epoch": 0.46193815124172416, + "grad_norm": 0.5525338053703308, + "learning_rate": 0.0007715710443747199, + "loss": 0.4878, + "step": 8268 + }, + { + "epoch": 0.46199402184540606, + "grad_norm": 0.4968549609184265, + "learning_rate": 0.000771543030031376, + "loss": 0.4373, + "step": 8269 + }, + { + "epoch": 0.4620498924490879, + "grad_norm": 0.5844044089317322, + "learning_rate": 0.0007715150156880323, + "loss": 0.4508, + "step": 8270 + }, + { + "epoch": 0.4621057630527698, + "grad_norm": 0.5884616374969482, + "learning_rate": 0.0007714870013446885, + "loss": 0.4105, + "step": 8271 + }, + { + "epoch": 0.46216163365645163, + "grad_norm": 0.7341116666793823, + "learning_rate": 0.0007714589870013447, + "loss": 0.562, + "step": 8272 + }, + { + "epoch": 0.4622175042601335, + "grad_norm": 0.6356759071350098, + "learning_rate": 0.0007714309726580008, + "loss": 0.317, + "step": 8273 + }, + { + "epoch": 0.4622733748638154, + "grad_norm": 0.780443012714386, + "learning_rate": 0.0007714029583146571, + "loss": 0.4808, + "step": 8274 + }, + { + "epoch": 0.46232924546749726, + "grad_norm": 0.6301320791244507, + "learning_rate": 0.0007713749439713134, + "loss": 0.4148, + "step": 8275 + }, + { + "epoch": 0.46238511607117916, + "grad_norm": 0.6813434958457947, + "learning_rate": 0.0007713469296279694, + "loss": 0.4724, + "step": 8276 + }, + { + "epoch": 0.462440986674861, + "grad_norm": 1.308084487915039, + "learning_rate": 0.0007713189152846258, + "loss": 0.3848, + "step": 8277 + }, + { + "epoch": 0.4624968572785429, + "grad_norm": 0.5943385362625122, + "learning_rate": 0.000771290900941282, + "loss": 0.4374, + "step": 8278 + }, + { + "epoch": 0.4625527278822248, + "grad_norm": 0.5375806093215942, + "learning_rate": 0.0007712628865979382, + "loss": 0.5109, + "step": 8279 + }, + { + "epoch": 0.46260859848590663, + "grad_norm": 0.9387848377227783, + "learning_rate": 0.0007712348722545943, + "loss": 0.536, + "step": 8280 + }, + { + "epoch": 0.4626644690895885, + "grad_norm": 0.5914612412452698, + "learning_rate": 0.0007712068579112506, + "loss": 0.5428, + "step": 8281 + }, + { + "epoch": 0.46272033969327037, + "grad_norm": 0.5309314727783203, + "learning_rate": 0.0007711788435679068, + "loss": 0.3925, + "step": 8282 + }, + { + "epoch": 0.46277621029695226, + "grad_norm": 0.375560998916626, + "learning_rate": 0.000771150829224563, + "loss": 0.3296, + "step": 8283 + }, + { + "epoch": 0.46283208090063416, + "grad_norm": 0.5384954214096069, + "learning_rate": 0.0007711228148812192, + "loss": 0.4685, + "step": 8284 + }, + { + "epoch": 0.462887951504316, + "grad_norm": 0.511833906173706, + "learning_rate": 0.0007710948005378754, + "loss": 0.3929, + "step": 8285 + }, + { + "epoch": 0.4629438221079979, + "grad_norm": 1.2386195659637451, + "learning_rate": 0.0007710667861945316, + "loss": 0.4818, + "step": 8286 + }, + { + "epoch": 0.46299969271167973, + "grad_norm": 0.520439088344574, + "learning_rate": 0.0007710387718511878, + "loss": 0.4544, + "step": 8287 + }, + { + "epoch": 0.46305556331536163, + "grad_norm": 1.190665602684021, + "learning_rate": 0.000771010757507844, + "loss": 0.5064, + "step": 8288 + }, + { + "epoch": 0.46311143391904347, + "grad_norm": 0.5156307816505432, + "learning_rate": 0.0007709827431645002, + "loss": 0.4732, + "step": 8289 + }, + { + "epoch": 0.46316730452272536, + "grad_norm": 0.5585314035415649, + "learning_rate": 0.0007709547288211564, + "loss": 0.4559, + "step": 8290 + }, + { + "epoch": 0.46322317512640726, + "grad_norm": 0.5319691300392151, + "learning_rate": 0.0007709267144778128, + "loss": 0.5532, + "step": 8291 + }, + { + "epoch": 0.4632790457300891, + "grad_norm": 0.45863866806030273, + "learning_rate": 0.0007708987001344689, + "loss": 0.4176, + "step": 8292 + }, + { + "epoch": 0.463334916333771, + "grad_norm": 2.8771169185638428, + "learning_rate": 0.0007708706857911251, + "loss": 0.439, + "step": 8293 + }, + { + "epoch": 0.46339078693745284, + "grad_norm": 0.9528853893280029, + "learning_rate": 0.0007708426714477813, + "loss": 0.5898, + "step": 8294 + }, + { + "epoch": 0.46344665754113473, + "grad_norm": 0.7021616101264954, + "learning_rate": 0.0007708146571044375, + "loss": 0.4033, + "step": 8295 + }, + { + "epoch": 0.46350252814481663, + "grad_norm": 0.4639327824115753, + "learning_rate": 0.0007707866427610937, + "loss": 0.4763, + "step": 8296 + }, + { + "epoch": 0.46355839874849847, + "grad_norm": 3.6278014183044434, + "learning_rate": 0.0007707586284177499, + "loss": 0.4249, + "step": 8297 + }, + { + "epoch": 0.46361426935218036, + "grad_norm": 2.6089680194854736, + "learning_rate": 0.0007707306140744062, + "loss": 0.5056, + "step": 8298 + }, + { + "epoch": 0.4636701399558622, + "grad_norm": 7.744719982147217, + "learning_rate": 0.0007707025997310623, + "loss": 0.4017, + "step": 8299 + }, + { + "epoch": 0.4637260105595441, + "grad_norm": 1.2931395769119263, + "learning_rate": 0.0007706745853877185, + "loss": 0.3906, + "step": 8300 + }, + { + "epoch": 0.463781881163226, + "grad_norm": 1.6457399129867554, + "learning_rate": 0.0007706465710443747, + "loss": 0.4873, + "step": 8301 + }, + { + "epoch": 0.46383775176690784, + "grad_norm": 4.362532138824463, + "learning_rate": 0.000770618556701031, + "loss": 0.4516, + "step": 8302 + }, + { + "epoch": 0.46389362237058973, + "grad_norm": 0.6651793718338013, + "learning_rate": 0.0007705905423576871, + "loss": 0.4508, + "step": 8303 + }, + { + "epoch": 0.46394949297427157, + "grad_norm": 1.408756971359253, + "learning_rate": 0.0007705625280143433, + "loss": 0.441, + "step": 8304 + }, + { + "epoch": 0.46400536357795347, + "grad_norm": 1.1080268621444702, + "learning_rate": 0.0007705345136709996, + "loss": 0.4591, + "step": 8305 + }, + { + "epoch": 0.4640612341816353, + "grad_norm": 1.0789482593536377, + "learning_rate": 0.0007705064993276558, + "loss": 0.4138, + "step": 8306 + }, + { + "epoch": 0.4641171047853172, + "grad_norm": 0.8339887261390686, + "learning_rate": 0.0007704784849843119, + "loss": 0.4489, + "step": 8307 + }, + { + "epoch": 0.4641729753889991, + "grad_norm": 0.7519840598106384, + "learning_rate": 0.0007704504706409681, + "loss": 0.4314, + "step": 8308 + }, + { + "epoch": 0.46422884599268094, + "grad_norm": 0.6297497153282166, + "learning_rate": 0.0007704224562976245, + "loss": 0.5071, + "step": 8309 + }, + { + "epoch": 0.46428471659636283, + "grad_norm": 0.6760733127593994, + "learning_rate": 0.0007703944419542807, + "loss": 0.3679, + "step": 8310 + }, + { + "epoch": 0.4643405872000447, + "grad_norm": 0.5687812566757202, + "learning_rate": 0.0007703664276109368, + "loss": 0.4613, + "step": 8311 + }, + { + "epoch": 0.46439645780372657, + "grad_norm": 0.7078191041946411, + "learning_rate": 0.000770338413267593, + "loss": 0.4497, + "step": 8312 + }, + { + "epoch": 0.46445232840740847, + "grad_norm": 0.6329132318496704, + "learning_rate": 0.0007703103989242493, + "loss": 0.4262, + "step": 8313 + }, + { + "epoch": 0.4645081990110903, + "grad_norm": 2.370523691177368, + "learning_rate": 0.0007702823845809055, + "loss": 0.4718, + "step": 8314 + }, + { + "epoch": 0.4645640696147722, + "grad_norm": 0.7218901515007019, + "learning_rate": 0.0007702543702375616, + "loss": 0.5341, + "step": 8315 + }, + { + "epoch": 0.46461994021845404, + "grad_norm": 0.4267449378967285, + "learning_rate": 0.0007702263558942179, + "loss": 0.4245, + "step": 8316 + }, + { + "epoch": 0.46467581082213594, + "grad_norm": 0.3710036873817444, + "learning_rate": 0.0007701983415508741, + "loss": 0.4055, + "step": 8317 + }, + { + "epoch": 0.46473168142581783, + "grad_norm": 0.7179917693138123, + "learning_rate": 0.0007701703272075302, + "loss": 0.6369, + "step": 8318 + }, + { + "epoch": 0.4647875520294997, + "grad_norm": 0.7782357335090637, + "learning_rate": 0.0007701423128641864, + "loss": 0.3738, + "step": 8319 + }, + { + "epoch": 0.46484342263318157, + "grad_norm": 3.1377995014190674, + "learning_rate": 0.0007701142985208427, + "loss": 0.3501, + "step": 8320 + }, + { + "epoch": 0.4648992932368634, + "grad_norm": 0.6463568806648254, + "learning_rate": 0.0007700862841774989, + "loss": 0.419, + "step": 8321 + }, + { + "epoch": 0.4649551638405453, + "grad_norm": 0.4959586262702942, + "learning_rate": 0.000770058269834155, + "loss": 0.448, + "step": 8322 + }, + { + "epoch": 0.46501103444422714, + "grad_norm": 0.5664860606193542, + "learning_rate": 0.0007700302554908113, + "loss": 0.4697, + "step": 8323 + }, + { + "epoch": 0.46506690504790904, + "grad_norm": 0.4424310326576233, + "learning_rate": 0.0007700022411474675, + "loss": 0.3949, + "step": 8324 + }, + { + "epoch": 0.46512277565159094, + "grad_norm": 0.4525221884250641, + "learning_rate": 0.0007699742268041238, + "loss": 0.5055, + "step": 8325 + }, + { + "epoch": 0.4651786462552728, + "grad_norm": 0.5327884554862976, + "learning_rate": 0.0007699462124607798, + "loss": 0.5208, + "step": 8326 + }, + { + "epoch": 0.46523451685895467, + "grad_norm": 0.5142462849617004, + "learning_rate": 0.0007699181981174362, + "loss": 0.3569, + "step": 8327 + }, + { + "epoch": 0.4652903874626365, + "grad_norm": 0.4812423884868622, + "learning_rate": 0.0007698901837740924, + "loss": 0.4659, + "step": 8328 + }, + { + "epoch": 0.4653462580663184, + "grad_norm": 0.5060938000679016, + "learning_rate": 0.0007698621694307486, + "loss": 0.5357, + "step": 8329 + }, + { + "epoch": 0.4654021286700003, + "grad_norm": 0.781578004360199, + "learning_rate": 0.0007698341550874048, + "loss": 0.4572, + "step": 8330 + }, + { + "epoch": 0.46545799927368214, + "grad_norm": 0.5741456151008606, + "learning_rate": 0.000769806140744061, + "loss": 0.529, + "step": 8331 + }, + { + "epoch": 0.46551386987736404, + "grad_norm": 0.5708630084991455, + "learning_rate": 0.0007697781264007172, + "loss": 0.4527, + "step": 8332 + }, + { + "epoch": 0.4655697404810459, + "grad_norm": 1.2990732192993164, + "learning_rate": 0.0007697501120573734, + "loss": 0.6171, + "step": 8333 + }, + { + "epoch": 0.4656256110847278, + "grad_norm": 0.5736995339393616, + "learning_rate": 0.0007697220977140296, + "loss": 0.4508, + "step": 8334 + }, + { + "epoch": 0.46568148168840967, + "grad_norm": 0.41251081228256226, + "learning_rate": 0.0007696940833706858, + "loss": 0.4595, + "step": 8335 + }, + { + "epoch": 0.4657373522920915, + "grad_norm": 0.7546110153198242, + "learning_rate": 0.000769666069027342, + "loss": 0.436, + "step": 8336 + }, + { + "epoch": 0.4657932228957734, + "grad_norm": 0.6450165510177612, + "learning_rate": 0.0007696380546839982, + "loss": 0.5105, + "step": 8337 + }, + { + "epoch": 0.46584909349945525, + "grad_norm": 1.3083370923995972, + "learning_rate": 0.0007696100403406544, + "loss": 0.4162, + "step": 8338 + }, + { + "epoch": 0.46590496410313714, + "grad_norm": 0.6559974551200867, + "learning_rate": 0.0007695820259973106, + "loss": 0.3995, + "step": 8339 + }, + { + "epoch": 0.465960834706819, + "grad_norm": 0.5145432949066162, + "learning_rate": 0.0007695540116539668, + "loss": 0.5276, + "step": 8340 + }, + { + "epoch": 0.4660167053105009, + "grad_norm": 0.6356304287910461, + "learning_rate": 0.000769525997310623, + "loss": 0.405, + "step": 8341 + }, + { + "epoch": 0.4660725759141828, + "grad_norm": 0.48437735438346863, + "learning_rate": 0.0007694979829672793, + "loss": 0.3621, + "step": 8342 + }, + { + "epoch": 0.4661284465178646, + "grad_norm": 0.8736655116081238, + "learning_rate": 0.0007694699686239355, + "loss": 0.4222, + "step": 8343 + }, + { + "epoch": 0.4661843171215465, + "grad_norm": 0.987747848033905, + "learning_rate": 0.0007694419542805917, + "loss": 0.3742, + "step": 8344 + }, + { + "epoch": 0.46624018772522835, + "grad_norm": 0.5263038277626038, + "learning_rate": 0.0007694139399372479, + "loss": 0.4148, + "step": 8345 + }, + { + "epoch": 0.46629605832891025, + "grad_norm": 0.6895851492881775, + "learning_rate": 0.0007693859255939041, + "loss": 0.4818, + "step": 8346 + }, + { + "epoch": 0.46635192893259214, + "grad_norm": 0.6131057739257812, + "learning_rate": 0.0007693579112505603, + "loss": 0.3973, + "step": 8347 + }, + { + "epoch": 0.466407799536274, + "grad_norm": 0.4243418276309967, + "learning_rate": 0.0007693298969072166, + "loss": 0.3933, + "step": 8348 + }, + { + "epoch": 0.4664636701399559, + "grad_norm": 0.8770838379859924, + "learning_rate": 0.0007693018825638727, + "loss": 0.5152, + "step": 8349 + }, + { + "epoch": 0.4665195407436377, + "grad_norm": 0.5045440793037415, + "learning_rate": 0.0007692738682205289, + "loss": 0.4761, + "step": 8350 + }, + { + "epoch": 0.4665754113473196, + "grad_norm": 0.5446005463600159, + "learning_rate": 0.0007692458538771851, + "loss": 0.3901, + "step": 8351 + }, + { + "epoch": 0.46663128195100145, + "grad_norm": 0.8204420208930969, + "learning_rate": 0.0007692178395338414, + "loss": 0.5274, + "step": 8352 + }, + { + "epoch": 0.46668715255468335, + "grad_norm": 4.641491889953613, + "learning_rate": 0.0007691898251904975, + "loss": 0.5582, + "step": 8353 + }, + { + "epoch": 0.46674302315836524, + "grad_norm": 0.8418676257133484, + "learning_rate": 0.0007691618108471537, + "loss": 0.4162, + "step": 8354 + }, + { + "epoch": 0.4667988937620471, + "grad_norm": 0.6711313128471375, + "learning_rate": 0.00076913379650381, + "loss": 0.4379, + "step": 8355 + }, + { + "epoch": 0.466854764365729, + "grad_norm": 0.46988874673843384, + "learning_rate": 0.0007691057821604662, + "loss": 0.3938, + "step": 8356 + }, + { + "epoch": 0.4669106349694108, + "grad_norm": 0.6195098757743835, + "learning_rate": 0.0007690777678171223, + "loss": 0.4283, + "step": 8357 + }, + { + "epoch": 0.4669665055730927, + "grad_norm": 0.47910916805267334, + "learning_rate": 0.0007690497534737785, + "loss": 0.4327, + "step": 8358 + }, + { + "epoch": 0.4670223761767746, + "grad_norm": 0.7250540256500244, + "learning_rate": 0.0007690217391304349, + "loss": 0.4007, + "step": 8359 + }, + { + "epoch": 0.46707824678045645, + "grad_norm": 0.6183105111122131, + "learning_rate": 0.000768993724787091, + "loss": 0.4175, + "step": 8360 + }, + { + "epoch": 0.46713411738413835, + "grad_norm": 0.6812803745269775, + "learning_rate": 0.0007689657104437472, + "loss": 0.3863, + "step": 8361 + }, + { + "epoch": 0.4671899879878202, + "grad_norm": 0.6463358998298645, + "learning_rate": 0.0007689376961004035, + "loss": 0.433, + "step": 8362 + }, + { + "epoch": 0.4672458585915021, + "grad_norm": 0.4519011974334717, + "learning_rate": 0.0007689096817570597, + "loss": 0.3849, + "step": 8363 + }, + { + "epoch": 0.467301729195184, + "grad_norm": 0.6182186603546143, + "learning_rate": 0.0007688816674137158, + "loss": 0.5727, + "step": 8364 + }, + { + "epoch": 0.4673575997988658, + "grad_norm": 0.44368797540664673, + "learning_rate": 0.000768853653070372, + "loss": 0.431, + "step": 8365 + }, + { + "epoch": 0.4674134704025477, + "grad_norm": 0.8842661380767822, + "learning_rate": 0.0007688256387270283, + "loss": 0.5319, + "step": 8366 + }, + { + "epoch": 0.46746934100622956, + "grad_norm": 0.4016326665878296, + "learning_rate": 0.0007687976243836845, + "loss": 0.4062, + "step": 8367 + }, + { + "epoch": 0.46752521160991145, + "grad_norm": 0.5155897736549377, + "learning_rate": 0.0007687696100403406, + "loss": 0.4434, + "step": 8368 + }, + { + "epoch": 0.4675810822135933, + "grad_norm": 0.4672565162181854, + "learning_rate": 0.0007687415956969969, + "loss": 0.3959, + "step": 8369 + }, + { + "epoch": 0.4676369528172752, + "grad_norm": 0.432827889919281, + "learning_rate": 0.0007687135813536531, + "loss": 0.3395, + "step": 8370 + }, + { + "epoch": 0.4676928234209571, + "grad_norm": 0.7310377955436707, + "learning_rate": 0.0007686855670103093, + "loss": 0.6166, + "step": 8371 + }, + { + "epoch": 0.4677486940246389, + "grad_norm": 0.8188595771789551, + "learning_rate": 0.0007686575526669654, + "loss": 0.4073, + "step": 8372 + }, + { + "epoch": 0.4678045646283208, + "grad_norm": 0.5443474054336548, + "learning_rate": 0.0007686295383236217, + "loss": 0.463, + "step": 8373 + }, + { + "epoch": 0.46786043523200266, + "grad_norm": 0.5615955591201782, + "learning_rate": 0.0007686015239802779, + "loss": 0.4735, + "step": 8374 + }, + { + "epoch": 0.46791630583568455, + "grad_norm": 0.4900813400745392, + "learning_rate": 0.0007685735096369342, + "loss": 0.4053, + "step": 8375 + }, + { + "epoch": 0.46797217643936645, + "grad_norm": 0.5002673864364624, + "learning_rate": 0.0007685454952935902, + "loss": 0.4416, + "step": 8376 + }, + { + "epoch": 0.4680280470430483, + "grad_norm": 0.6251047253608704, + "learning_rate": 0.0007685174809502466, + "loss": 0.6529, + "step": 8377 + }, + { + "epoch": 0.4680839176467302, + "grad_norm": 0.5734697580337524, + "learning_rate": 0.0007684894666069028, + "loss": 0.4221, + "step": 8378 + }, + { + "epoch": 0.468139788250412, + "grad_norm": 0.7307958602905273, + "learning_rate": 0.000768461452263559, + "loss": 0.5006, + "step": 8379 + }, + { + "epoch": 0.4681956588540939, + "grad_norm": 0.6133213639259338, + "learning_rate": 0.0007684334379202152, + "loss": 0.4476, + "step": 8380 + }, + { + "epoch": 0.4682515294577758, + "grad_norm": 0.536980926990509, + "learning_rate": 0.0007684054235768714, + "loss": 0.4378, + "step": 8381 + }, + { + "epoch": 0.46830740006145766, + "grad_norm": 2.020296096801758, + "learning_rate": 0.0007683774092335276, + "loss": 0.4776, + "step": 8382 + }, + { + "epoch": 0.46836327066513955, + "grad_norm": 0.6448720097541809, + "learning_rate": 0.0007683493948901837, + "loss": 0.5294, + "step": 8383 + }, + { + "epoch": 0.4684191412688214, + "grad_norm": 0.5758211612701416, + "learning_rate": 0.00076832138054684, + "loss": 0.496, + "step": 8384 + }, + { + "epoch": 0.4684750118725033, + "grad_norm": 0.7862688302993774, + "learning_rate": 0.0007682933662034962, + "loss": 0.3928, + "step": 8385 + }, + { + "epoch": 0.46853088247618513, + "grad_norm": 0.5686802268028259, + "learning_rate": 0.0007682653518601524, + "loss": 0.4516, + "step": 8386 + }, + { + "epoch": 0.468586753079867, + "grad_norm": 0.4234335720539093, + "learning_rate": 0.0007682373375168086, + "loss": 0.4544, + "step": 8387 + }, + { + "epoch": 0.4686426236835489, + "grad_norm": 1.9582819938659668, + "learning_rate": 0.0007682093231734648, + "loss": 0.4295, + "step": 8388 + }, + { + "epoch": 0.46869849428723076, + "grad_norm": 0.4618314206600189, + "learning_rate": 0.000768181308830121, + "loss": 0.444, + "step": 8389 + }, + { + "epoch": 0.46875436489091266, + "grad_norm": 0.4019697904586792, + "learning_rate": 0.0007681532944867772, + "loss": 0.3314, + "step": 8390 + }, + { + "epoch": 0.4688102354945945, + "grad_norm": 0.6245055198669434, + "learning_rate": 0.0007681252801434334, + "loss": 0.7434, + "step": 8391 + }, + { + "epoch": 0.4688661060982764, + "grad_norm": 0.3947533369064331, + "learning_rate": 0.0007680972658000896, + "loss": 0.45, + "step": 8392 + }, + { + "epoch": 0.4689219767019583, + "grad_norm": 0.4915502965450287, + "learning_rate": 0.0007680692514567459, + "loss": 0.385, + "step": 8393 + }, + { + "epoch": 0.4689778473056401, + "grad_norm": 0.6792342662811279, + "learning_rate": 0.0007680412371134022, + "loss": 0.4463, + "step": 8394 + }, + { + "epoch": 0.469033717909322, + "grad_norm": 0.4749889075756073, + "learning_rate": 0.0007680132227700583, + "loss": 0.5568, + "step": 8395 + }, + { + "epoch": 0.46908958851300386, + "grad_norm": 0.43437322974205017, + "learning_rate": 0.0007679852084267145, + "loss": 0.4248, + "step": 8396 + }, + { + "epoch": 0.46914545911668576, + "grad_norm": 0.6187331080436707, + "learning_rate": 0.0007679571940833707, + "loss": 0.4206, + "step": 8397 + }, + { + "epoch": 0.46920132972036765, + "grad_norm": 0.5554185509681702, + "learning_rate": 0.000767929179740027, + "loss": 0.5717, + "step": 8398 + }, + { + "epoch": 0.4692572003240495, + "grad_norm": 0.534995973110199, + "learning_rate": 0.0007679011653966831, + "loss": 0.4585, + "step": 8399 + }, + { + "epoch": 0.4693130709277314, + "grad_norm": 0.48531222343444824, + "learning_rate": 0.0007678731510533393, + "loss": 0.4891, + "step": 8400 + }, + { + "epoch": 0.46936894153141323, + "grad_norm": 0.9147448539733887, + "learning_rate": 0.0007678451367099956, + "loss": 0.395, + "step": 8401 + }, + { + "epoch": 0.4694248121350951, + "grad_norm": 0.821662425994873, + "learning_rate": 0.0007678171223666517, + "loss": 0.5459, + "step": 8402 + }, + { + "epoch": 0.46948068273877697, + "grad_norm": 0.43632882833480835, + "learning_rate": 0.0007677891080233079, + "loss": 0.4717, + "step": 8403 + }, + { + "epoch": 0.46953655334245886, + "grad_norm": 0.47147393226623535, + "learning_rate": 0.0007677610936799641, + "loss": 0.3629, + "step": 8404 + }, + { + "epoch": 0.46959242394614076, + "grad_norm": 0.95846027135849, + "learning_rate": 0.0007677330793366204, + "loss": 0.6309, + "step": 8405 + }, + { + "epoch": 0.4696482945498226, + "grad_norm": 0.591511607170105, + "learning_rate": 0.0007677050649932765, + "loss": 0.5092, + "step": 8406 + }, + { + "epoch": 0.4697041651535045, + "grad_norm": 0.48178425431251526, + "learning_rate": 0.0007676770506499327, + "loss": 0.3019, + "step": 8407 + }, + { + "epoch": 0.46976003575718633, + "grad_norm": 0.5246779918670654, + "learning_rate": 0.000767649036306589, + "loss": 0.3487, + "step": 8408 + }, + { + "epoch": 0.46981590636086823, + "grad_norm": 2.250732660293579, + "learning_rate": 0.0007676210219632453, + "loss": 0.4829, + "step": 8409 + }, + { + "epoch": 0.4698717769645501, + "grad_norm": 0.43436700105667114, + "learning_rate": 0.0007675930076199014, + "loss": 0.5058, + "step": 8410 + }, + { + "epoch": 0.46992764756823197, + "grad_norm": 0.4927273392677307, + "learning_rate": 0.0007675649932765576, + "loss": 0.4939, + "step": 8411 + }, + { + "epoch": 0.46998351817191386, + "grad_norm": 0.7655931711196899, + "learning_rate": 0.0007675369789332139, + "loss": 0.4811, + "step": 8412 + }, + { + "epoch": 0.4700393887755957, + "grad_norm": 0.3459693193435669, + "learning_rate": 0.0007675089645898701, + "loss": 0.3715, + "step": 8413 + }, + { + "epoch": 0.4700952593792776, + "grad_norm": 0.5446482300758362, + "learning_rate": 0.0007674809502465262, + "loss": 0.4464, + "step": 8414 + }, + { + "epoch": 0.4701511299829595, + "grad_norm": 0.5783968567848206, + "learning_rate": 0.0007674529359031824, + "loss": 0.4283, + "step": 8415 + }, + { + "epoch": 0.47020700058664133, + "grad_norm": 0.7465180158615112, + "learning_rate": 0.0007674249215598387, + "loss": 0.4631, + "step": 8416 + }, + { + "epoch": 0.47026287119032323, + "grad_norm": 0.532645046710968, + "learning_rate": 0.0007673969072164949, + "loss": 0.4822, + "step": 8417 + }, + { + "epoch": 0.47031874179400507, + "grad_norm": 0.7534663677215576, + "learning_rate": 0.000767368892873151, + "loss": 0.518, + "step": 8418 + }, + { + "epoch": 0.47037461239768696, + "grad_norm": 0.4904281198978424, + "learning_rate": 0.0007673408785298073, + "loss": 0.4338, + "step": 8419 + }, + { + "epoch": 0.4704304830013688, + "grad_norm": 0.8658162951469421, + "learning_rate": 0.0007673128641864635, + "loss": 0.5605, + "step": 8420 + }, + { + "epoch": 0.4704863536050507, + "grad_norm": 2.222447633743286, + "learning_rate": 0.0007672848498431197, + "loss": 0.4915, + "step": 8421 + }, + { + "epoch": 0.4705422242087326, + "grad_norm": 0.5620232820510864, + "learning_rate": 0.0007672568354997758, + "loss": 0.502, + "step": 8422 + }, + { + "epoch": 0.47059809481241444, + "grad_norm": 1.1731997728347778, + "learning_rate": 0.0007672288211564321, + "loss": 0.5993, + "step": 8423 + }, + { + "epoch": 0.47065396541609633, + "grad_norm": 0.5277357697486877, + "learning_rate": 0.0007672008068130883, + "loss": 0.5334, + "step": 8424 + }, + { + "epoch": 0.47070983601977817, + "grad_norm": 0.498995840549469, + "learning_rate": 0.0007671727924697444, + "loss": 0.5609, + "step": 8425 + }, + { + "epoch": 0.47076570662346007, + "grad_norm": 0.6883918642997742, + "learning_rate": 0.0007671447781264008, + "loss": 0.7171, + "step": 8426 + }, + { + "epoch": 0.47082157722714196, + "grad_norm": 1.3873242139816284, + "learning_rate": 0.000767116763783057, + "loss": 0.4485, + "step": 8427 + }, + { + "epoch": 0.4708774478308238, + "grad_norm": 0.40618687868118286, + "learning_rate": 0.0007670887494397132, + "loss": 0.473, + "step": 8428 + }, + { + "epoch": 0.4709333184345057, + "grad_norm": 0.8522046804428101, + "learning_rate": 0.0007670607350963693, + "loss": 0.6165, + "step": 8429 + }, + { + "epoch": 0.47098918903818754, + "grad_norm": 0.8077607154846191, + "learning_rate": 0.0007670327207530256, + "loss": 0.4463, + "step": 8430 + }, + { + "epoch": 0.47104505964186943, + "grad_norm": 0.4944069981575012, + "learning_rate": 0.0007670047064096818, + "loss": 0.4047, + "step": 8431 + }, + { + "epoch": 0.47110093024555133, + "grad_norm": 0.7058014869689941, + "learning_rate": 0.000766976692066338, + "loss": 0.441, + "step": 8432 + }, + { + "epoch": 0.47115680084923317, + "grad_norm": 2.537108898162842, + "learning_rate": 0.0007669486777229942, + "loss": 0.6935, + "step": 8433 + }, + { + "epoch": 0.47121267145291507, + "grad_norm": 0.89910888671875, + "learning_rate": 0.0007669206633796504, + "loss": 0.5286, + "step": 8434 + }, + { + "epoch": 0.4712685420565969, + "grad_norm": 2.671358346939087, + "learning_rate": 0.0007668926490363066, + "loss": 0.4782, + "step": 8435 + }, + { + "epoch": 0.4713244126602788, + "grad_norm": 0.7468029260635376, + "learning_rate": 0.0007668646346929628, + "loss": 0.5067, + "step": 8436 + }, + { + "epoch": 0.47138028326396064, + "grad_norm": 0.7785238027572632, + "learning_rate": 0.000766836620349619, + "loss": 0.5036, + "step": 8437 + }, + { + "epoch": 0.47143615386764254, + "grad_norm": 0.45879125595092773, + "learning_rate": 0.0007668086060062752, + "loss": 0.4709, + "step": 8438 + }, + { + "epoch": 0.47149202447132443, + "grad_norm": 0.4992526173591614, + "learning_rate": 0.0007667805916629314, + "loss": 0.4344, + "step": 8439 + }, + { + "epoch": 0.4715478950750063, + "grad_norm": 0.44825735688209534, + "learning_rate": 0.0007667525773195877, + "loss": 0.4336, + "step": 8440 + }, + { + "epoch": 0.47160376567868817, + "grad_norm": 0.5018333196640015, + "learning_rate": 0.0007667245629762438, + "loss": 0.4194, + "step": 8441 + }, + { + "epoch": 0.47165963628237, + "grad_norm": 0.7088776230812073, + "learning_rate": 0.0007666965486329, + "loss": 0.435, + "step": 8442 + }, + { + "epoch": 0.4717155068860519, + "grad_norm": 0.8174096941947937, + "learning_rate": 0.0007666685342895563, + "loss": 0.4974, + "step": 8443 + }, + { + "epoch": 0.4717713774897338, + "grad_norm": 0.42082467675209045, + "learning_rate": 0.0007666405199462125, + "loss": 0.3554, + "step": 8444 + }, + { + "epoch": 0.47182724809341564, + "grad_norm": 0.425809383392334, + "learning_rate": 0.0007666125056028687, + "loss": 0.5145, + "step": 8445 + }, + { + "epoch": 0.47188311869709754, + "grad_norm": 3.06546950340271, + "learning_rate": 0.0007665844912595249, + "loss": 0.4656, + "step": 8446 + }, + { + "epoch": 0.4719389893007794, + "grad_norm": 0.5696642398834229, + "learning_rate": 0.0007665564769161812, + "loss": 0.415, + "step": 8447 + }, + { + "epoch": 0.4719948599044613, + "grad_norm": 0.42671993374824524, + "learning_rate": 0.0007665284625728373, + "loss": 0.4481, + "step": 8448 + }, + { + "epoch": 0.47205073050814317, + "grad_norm": 0.5789678692817688, + "learning_rate": 0.0007665004482294935, + "loss": 0.501, + "step": 8449 + }, + { + "epoch": 0.472106601111825, + "grad_norm": 0.5643612742424011, + "learning_rate": 0.0007664724338861497, + "loss": 0.4893, + "step": 8450 + }, + { + "epoch": 0.4721624717155069, + "grad_norm": 0.45125335454940796, + "learning_rate": 0.000766444419542806, + "loss": 0.4472, + "step": 8451 + }, + { + "epoch": 0.47221834231918874, + "grad_norm": 0.4460381269454956, + "learning_rate": 0.0007664164051994621, + "loss": 0.4107, + "step": 8452 + }, + { + "epoch": 0.47227421292287064, + "grad_norm": 0.6163467168807983, + "learning_rate": 0.0007663883908561183, + "loss": 0.4596, + "step": 8453 + }, + { + "epoch": 0.4723300835265525, + "grad_norm": 0.8895230293273926, + "learning_rate": 0.0007663603765127745, + "loss": 0.4589, + "step": 8454 + }, + { + "epoch": 0.4723859541302344, + "grad_norm": 0.5911558866500854, + "learning_rate": 0.0007663323621694308, + "loss": 0.5294, + "step": 8455 + }, + { + "epoch": 0.47244182473391627, + "grad_norm": 0.9282490611076355, + "learning_rate": 0.0007663043478260869, + "loss": 0.4107, + "step": 8456 + }, + { + "epoch": 0.4724976953375981, + "grad_norm": 0.5446785688400269, + "learning_rate": 0.0007662763334827431, + "loss": 0.5089, + "step": 8457 + }, + { + "epoch": 0.47255356594128, + "grad_norm": 0.830390453338623, + "learning_rate": 0.0007662483191393995, + "loss": 0.4108, + "step": 8458 + }, + { + "epoch": 0.47260943654496185, + "grad_norm": 0.8505272269248962, + "learning_rate": 0.0007662203047960557, + "loss": 0.4582, + "step": 8459 + }, + { + "epoch": 0.47266530714864374, + "grad_norm": 0.6206179261207581, + "learning_rate": 0.0007661922904527118, + "loss": 0.4802, + "step": 8460 + }, + { + "epoch": 0.47272117775232564, + "grad_norm": 7.829445838928223, + "learning_rate": 0.000766164276109368, + "loss": 0.4512, + "step": 8461 + }, + { + "epoch": 0.4727770483560075, + "grad_norm": 0.9527490139007568, + "learning_rate": 0.0007661362617660243, + "loss": 0.4193, + "step": 8462 + }, + { + "epoch": 0.4728329189596894, + "grad_norm": 0.4292425513267517, + "learning_rate": 0.0007661082474226805, + "loss": 0.4841, + "step": 8463 + }, + { + "epoch": 0.4728887895633712, + "grad_norm": 0.5508646368980408, + "learning_rate": 0.0007660802330793366, + "loss": 0.3905, + "step": 8464 + }, + { + "epoch": 0.4729446601670531, + "grad_norm": 0.5645879507064819, + "learning_rate": 0.0007660522187359929, + "loss": 0.3641, + "step": 8465 + }, + { + "epoch": 0.47300053077073495, + "grad_norm": 0.4410681426525116, + "learning_rate": 0.0007660242043926491, + "loss": 0.4462, + "step": 8466 + }, + { + "epoch": 0.47305640137441685, + "grad_norm": 0.6589208841323853, + "learning_rate": 0.0007659961900493052, + "loss": 0.5057, + "step": 8467 + }, + { + "epoch": 0.47311227197809874, + "grad_norm": 0.7441514134407043, + "learning_rate": 0.0007659681757059614, + "loss": 0.5292, + "step": 8468 + }, + { + "epoch": 0.4731681425817806, + "grad_norm": 0.7831038236618042, + "learning_rate": 0.0007659401613626177, + "loss": 0.4326, + "step": 8469 + }, + { + "epoch": 0.4732240131854625, + "grad_norm": 0.8970457315444946, + "learning_rate": 0.0007659121470192739, + "loss": 0.4843, + "step": 8470 + }, + { + "epoch": 0.4732798837891443, + "grad_norm": 0.5688298940658569, + "learning_rate": 0.00076588413267593, + "loss": 0.3649, + "step": 8471 + }, + { + "epoch": 0.4733357543928262, + "grad_norm": 0.6533032059669495, + "learning_rate": 0.0007658561183325863, + "loss": 0.5044, + "step": 8472 + }, + { + "epoch": 0.4733916249965081, + "grad_norm": 0.47742390632629395, + "learning_rate": 0.0007658281039892425, + "loss": 0.4345, + "step": 8473 + }, + { + "epoch": 0.47344749560018995, + "grad_norm": 0.42188218235969543, + "learning_rate": 0.0007658000896458987, + "loss": 0.4636, + "step": 8474 + }, + { + "epoch": 0.47350336620387184, + "grad_norm": 0.475547194480896, + "learning_rate": 0.0007657720753025548, + "loss": 0.5201, + "step": 8475 + }, + { + "epoch": 0.4735592368075537, + "grad_norm": 0.6101295948028564, + "learning_rate": 0.0007657440609592112, + "loss": 0.479, + "step": 8476 + }, + { + "epoch": 0.4736151074112356, + "grad_norm": 0.7063676714897156, + "learning_rate": 0.0007657160466158674, + "loss": 0.4469, + "step": 8477 + }, + { + "epoch": 0.4736709780149175, + "grad_norm": 0.41180846095085144, + "learning_rate": 0.0007656880322725236, + "loss": 0.4455, + "step": 8478 + }, + { + "epoch": 0.4737268486185993, + "grad_norm": 0.37695619463920593, + "learning_rate": 0.0007656600179291798, + "loss": 0.4112, + "step": 8479 + }, + { + "epoch": 0.4737827192222812, + "grad_norm": 0.6603962182998657, + "learning_rate": 0.000765632003585836, + "loss": 0.475, + "step": 8480 + }, + { + "epoch": 0.47383858982596305, + "grad_norm": 0.6363667249679565, + "learning_rate": 0.0007656039892424922, + "loss": 0.4702, + "step": 8481 + }, + { + "epoch": 0.47389446042964495, + "grad_norm": 0.47928422689437866, + "learning_rate": 0.0007655759748991484, + "loss": 0.4725, + "step": 8482 + }, + { + "epoch": 0.4739503310333268, + "grad_norm": 1.726090908050537, + "learning_rate": 0.0007655479605558046, + "loss": 0.414, + "step": 8483 + }, + { + "epoch": 0.4740062016370087, + "grad_norm": 0.49291160702705383, + "learning_rate": 0.0007655199462124608, + "loss": 0.3805, + "step": 8484 + }, + { + "epoch": 0.4740620722406906, + "grad_norm": 1.1211094856262207, + "learning_rate": 0.000765491931869117, + "loss": 0.4509, + "step": 8485 + }, + { + "epoch": 0.4741179428443724, + "grad_norm": 0.4985153377056122, + "learning_rate": 0.0007654639175257731, + "loss": 0.4655, + "step": 8486 + }, + { + "epoch": 0.4741738134480543, + "grad_norm": 0.9617908000946045, + "learning_rate": 0.0007654359031824294, + "loss": 0.5555, + "step": 8487 + }, + { + "epoch": 0.47422968405173616, + "grad_norm": 6.598029136657715, + "learning_rate": 0.0007654078888390856, + "loss": 0.5353, + "step": 8488 + }, + { + "epoch": 0.47428555465541805, + "grad_norm": 0.5514321327209473, + "learning_rate": 0.0007653798744957418, + "loss": 0.5134, + "step": 8489 + }, + { + "epoch": 0.47434142525909995, + "grad_norm": 0.5084773898124695, + "learning_rate": 0.000765351860152398, + "loss": 0.4079, + "step": 8490 + }, + { + "epoch": 0.4743972958627818, + "grad_norm": 0.8965837955474854, + "learning_rate": 0.0007653238458090542, + "loss": 0.5276, + "step": 8491 + }, + { + "epoch": 0.4744531664664637, + "grad_norm": 0.47885167598724365, + "learning_rate": 0.0007652958314657104, + "loss": 0.4047, + "step": 8492 + }, + { + "epoch": 0.4745090370701455, + "grad_norm": 0.5450509190559387, + "learning_rate": 0.0007652678171223667, + "loss": 0.5296, + "step": 8493 + }, + { + "epoch": 0.4745649076738274, + "grad_norm": 0.37675222754478455, + "learning_rate": 0.0007652398027790229, + "loss": 0.4468, + "step": 8494 + }, + { + "epoch": 0.4746207782775093, + "grad_norm": 0.5022895336151123, + "learning_rate": 0.0007652117884356791, + "loss": 0.5199, + "step": 8495 + }, + { + "epoch": 0.47467664888119115, + "grad_norm": 0.569957435131073, + "learning_rate": 0.0007651837740923353, + "loss": 0.4852, + "step": 8496 + }, + { + "epoch": 0.47473251948487305, + "grad_norm": 0.47054752707481384, + "learning_rate": 0.0007651557597489916, + "loss": 0.4919, + "step": 8497 + }, + { + "epoch": 0.4747883900885549, + "grad_norm": 0.4552503526210785, + "learning_rate": 0.0007651277454056477, + "loss": 0.4839, + "step": 8498 + }, + { + "epoch": 0.4748442606922368, + "grad_norm": 0.6597632765769958, + "learning_rate": 0.0007650997310623039, + "loss": 0.5302, + "step": 8499 + }, + { + "epoch": 0.4749001312959186, + "grad_norm": 1.08214271068573, + "learning_rate": 0.0007650717167189601, + "loss": 0.6661, + "step": 8500 + }, + { + "epoch": 0.4749001312959186, + "eval_cer": 0.09111540266003296, + "eval_loss": 0.3452911972999573, + "eval_runtime": 56.2896, + "eval_samples_per_second": 80.619, + "eval_steps_per_second": 5.045, + "eval_wer": 0.35965475824697635, + "step": 8500 + }, + { + "epoch": 0.4749560018996005, + "grad_norm": 0.5763117074966431, + "learning_rate": 0.0007650437023756164, + "loss": 0.4964, + "step": 8501 + }, + { + "epoch": 0.4750118725032824, + "grad_norm": 0.6471214294433594, + "learning_rate": 0.0007650156880322725, + "loss": 0.5475, + "step": 8502 + }, + { + "epoch": 0.47506774310696426, + "grad_norm": 0.6939160823822021, + "learning_rate": 0.0007649876736889287, + "loss": 0.4909, + "step": 8503 + }, + { + "epoch": 0.47512361371064615, + "grad_norm": 0.8922209739685059, + "learning_rate": 0.000764959659345585, + "loss": 0.4641, + "step": 8504 + }, + { + "epoch": 0.475179484314328, + "grad_norm": 1.1093851327896118, + "learning_rate": 0.0007649316450022412, + "loss": 0.5628, + "step": 8505 + }, + { + "epoch": 0.4752353549180099, + "grad_norm": 0.4415801167488098, + "learning_rate": 0.0007649036306588973, + "loss": 0.3671, + "step": 8506 + }, + { + "epoch": 0.4752912255216918, + "grad_norm": 0.5722789168357849, + "learning_rate": 0.0007648756163155535, + "loss": 0.5695, + "step": 8507 + }, + { + "epoch": 0.4753470961253736, + "grad_norm": 0.4475419819355011, + "learning_rate": 0.0007648476019722098, + "loss": 0.4348, + "step": 8508 + }, + { + "epoch": 0.4754029667290555, + "grad_norm": 0.5316697359085083, + "learning_rate": 0.000764819587628866, + "loss": 0.4388, + "step": 8509 + }, + { + "epoch": 0.47545883733273736, + "grad_norm": 0.468132883310318, + "learning_rate": 0.0007647915732855222, + "loss": 0.4297, + "step": 8510 + }, + { + "epoch": 0.47551470793641926, + "grad_norm": 4.597579479217529, + "learning_rate": 0.0007647635589421785, + "loss": 0.3974, + "step": 8511 + }, + { + "epoch": 0.47557057854010115, + "grad_norm": 0.5459191203117371, + "learning_rate": 0.0007647355445988347, + "loss": 0.443, + "step": 8512 + }, + { + "epoch": 0.475626449143783, + "grad_norm": 3.9948084354400635, + "learning_rate": 0.0007647075302554908, + "loss": 0.5469, + "step": 8513 + }, + { + "epoch": 0.4756823197474649, + "grad_norm": 0.6383848786354065, + "learning_rate": 0.000764679515912147, + "loss": 0.4191, + "step": 8514 + }, + { + "epoch": 0.47573819035114673, + "grad_norm": 0.5636700987815857, + "learning_rate": 0.0007646515015688033, + "loss": 0.4137, + "step": 8515 + }, + { + "epoch": 0.4757940609548286, + "grad_norm": 0.7684242129325867, + "learning_rate": 0.0007646234872254595, + "loss": 0.4621, + "step": 8516 + }, + { + "epoch": 0.47584993155851046, + "grad_norm": 0.44358566403388977, + "learning_rate": 0.0007645954728821156, + "loss": 0.3701, + "step": 8517 + }, + { + "epoch": 0.47590580216219236, + "grad_norm": 0.4585936963558197, + "learning_rate": 0.0007645674585387719, + "loss": 0.6165, + "step": 8518 + }, + { + "epoch": 0.47596167276587426, + "grad_norm": 0.5164802074432373, + "learning_rate": 0.0007645394441954281, + "loss": 0.5026, + "step": 8519 + }, + { + "epoch": 0.4760175433695561, + "grad_norm": 0.47487038373947144, + "learning_rate": 0.0007645114298520843, + "loss": 0.3549, + "step": 8520 + }, + { + "epoch": 0.476073413973238, + "grad_norm": 5.048899173736572, + "learning_rate": 0.0007644834155087404, + "loss": 0.4835, + "step": 8521 + }, + { + "epoch": 0.47612928457691983, + "grad_norm": 0.5057005286216736, + "learning_rate": 0.0007644554011653967, + "loss": 0.4554, + "step": 8522 + }, + { + "epoch": 0.4761851551806017, + "grad_norm": 2.0464820861816406, + "learning_rate": 0.0007644273868220529, + "loss": 0.4194, + "step": 8523 + }, + { + "epoch": 0.4762410257842836, + "grad_norm": 0.6821114420890808, + "learning_rate": 0.0007643993724787091, + "loss": 0.5585, + "step": 8524 + }, + { + "epoch": 0.47629689638796546, + "grad_norm": 0.5724107027053833, + "learning_rate": 0.0007643713581353652, + "loss": 0.4725, + "step": 8525 + }, + { + "epoch": 0.47635276699164736, + "grad_norm": 0.42167454957962036, + "learning_rate": 0.0007643433437920216, + "loss": 0.4351, + "step": 8526 + }, + { + "epoch": 0.4764086375953292, + "grad_norm": 0.5396450161933899, + "learning_rate": 0.0007643153294486778, + "loss": 0.7271, + "step": 8527 + }, + { + "epoch": 0.4764645081990111, + "grad_norm": 0.5320107936859131, + "learning_rate": 0.000764287315105334, + "loss": 0.4442, + "step": 8528 + }, + { + "epoch": 0.476520378802693, + "grad_norm": 0.6971487402915955, + "learning_rate": 0.0007642593007619902, + "loss": 0.5866, + "step": 8529 + }, + { + "epoch": 0.47657624940637483, + "grad_norm": 1.207384467124939, + "learning_rate": 0.0007642312864186464, + "loss": 0.4328, + "step": 8530 + }, + { + "epoch": 0.4766321200100567, + "grad_norm": 0.5551571846008301, + "learning_rate": 0.0007642032720753026, + "loss": 0.4612, + "step": 8531 + }, + { + "epoch": 0.47668799061373857, + "grad_norm": 0.3762427866458893, + "learning_rate": 0.0007641752577319587, + "loss": 0.4178, + "step": 8532 + }, + { + "epoch": 0.47674386121742046, + "grad_norm": 0.6759915947914124, + "learning_rate": 0.000764147243388615, + "loss": 0.4882, + "step": 8533 + }, + { + "epoch": 0.4767997318211023, + "grad_norm": 0.6774831414222717, + "learning_rate": 0.0007641192290452712, + "loss": 0.45, + "step": 8534 + }, + { + "epoch": 0.4768556024247842, + "grad_norm": 0.5513893365859985, + "learning_rate": 0.0007640912147019274, + "loss": 0.4068, + "step": 8535 + }, + { + "epoch": 0.4769114730284661, + "grad_norm": 0.3892998993396759, + "learning_rate": 0.0007640632003585836, + "loss": 0.4145, + "step": 8536 + }, + { + "epoch": 0.47696734363214793, + "grad_norm": 0.5123004913330078, + "learning_rate": 0.0007640351860152398, + "loss": 0.5374, + "step": 8537 + }, + { + "epoch": 0.47702321423582983, + "grad_norm": 0.46781933307647705, + "learning_rate": 0.000764007171671896, + "loss": 0.4903, + "step": 8538 + }, + { + "epoch": 0.47707908483951167, + "grad_norm": 0.4760531485080719, + "learning_rate": 0.0007639791573285522, + "loss": 0.3829, + "step": 8539 + }, + { + "epoch": 0.47713495544319356, + "grad_norm": 0.8311558961868286, + "learning_rate": 0.0007639511429852084, + "loss": 0.4165, + "step": 8540 + }, + { + "epoch": 0.47719082604687546, + "grad_norm": 0.46650058031082153, + "learning_rate": 0.0007639231286418646, + "loss": 0.4638, + "step": 8541 + }, + { + "epoch": 0.4772466966505573, + "grad_norm": 2.7046992778778076, + "learning_rate": 0.0007638951142985208, + "loss": 0.5144, + "step": 8542 + }, + { + "epoch": 0.4773025672542392, + "grad_norm": 0.5467882752418518, + "learning_rate": 0.0007638670999551772, + "loss": 0.5878, + "step": 8543 + }, + { + "epoch": 0.47735843785792104, + "grad_norm": 0.5160366892814636, + "learning_rate": 0.0007638390856118333, + "loss": 0.4595, + "step": 8544 + }, + { + "epoch": 0.47741430846160293, + "grad_norm": 0.7942737936973572, + "learning_rate": 0.0007638110712684895, + "loss": 0.6231, + "step": 8545 + }, + { + "epoch": 0.4774701790652848, + "grad_norm": 1.8017168045043945, + "learning_rate": 0.0007637830569251457, + "loss": 0.399, + "step": 8546 + }, + { + "epoch": 0.47752604966896667, + "grad_norm": 0.5743169188499451, + "learning_rate": 0.000763755042581802, + "loss": 0.5285, + "step": 8547 + }, + { + "epoch": 0.47758192027264856, + "grad_norm": 0.5287247896194458, + "learning_rate": 0.0007637270282384581, + "loss": 0.4595, + "step": 8548 + }, + { + "epoch": 0.4776377908763304, + "grad_norm": 0.4828709363937378, + "learning_rate": 0.0007636990138951143, + "loss": 0.455, + "step": 8549 + }, + { + "epoch": 0.4776936614800123, + "grad_norm": 0.5566802024841309, + "learning_rate": 0.0007636709995517706, + "loss": 0.527, + "step": 8550 + }, + { + "epoch": 0.47774953208369414, + "grad_norm": 0.5271446108818054, + "learning_rate": 0.0007636429852084267, + "loss": 0.3682, + "step": 8551 + }, + { + "epoch": 0.47780540268737604, + "grad_norm": 0.7068520188331604, + "learning_rate": 0.0007636149708650829, + "loss": 0.4087, + "step": 8552 + }, + { + "epoch": 0.47786127329105793, + "grad_norm": 0.5282551646232605, + "learning_rate": 0.0007635869565217391, + "loss": 0.38, + "step": 8553 + }, + { + "epoch": 0.47791714389473977, + "grad_norm": 0.6435890197753906, + "learning_rate": 0.0007635589421783954, + "loss": 0.4646, + "step": 8554 + }, + { + "epoch": 0.47797301449842167, + "grad_norm": 1.2176438570022583, + "learning_rate": 0.0007635309278350515, + "loss": 0.5368, + "step": 8555 + }, + { + "epoch": 0.4780288851021035, + "grad_norm": 0.688869297504425, + "learning_rate": 0.0007635029134917077, + "loss": 0.4987, + "step": 8556 + }, + { + "epoch": 0.4780847557057854, + "grad_norm": 0.5361025333404541, + "learning_rate": 0.0007634748991483639, + "loss": 0.4684, + "step": 8557 + }, + { + "epoch": 0.4781406263094673, + "grad_norm": 0.6246371269226074, + "learning_rate": 0.0007634468848050202, + "loss": 0.4561, + "step": 8558 + }, + { + "epoch": 0.47819649691314914, + "grad_norm": 0.8262721300125122, + "learning_rate": 0.0007634188704616763, + "loss": 0.5413, + "step": 8559 + }, + { + "epoch": 0.47825236751683103, + "grad_norm": 0.8252934813499451, + "learning_rate": 0.0007633908561183326, + "loss": 0.5216, + "step": 8560 + }, + { + "epoch": 0.4783082381205129, + "grad_norm": 0.4097241759300232, + "learning_rate": 0.0007633628417749889, + "loss": 0.4533, + "step": 8561 + }, + { + "epoch": 0.47836410872419477, + "grad_norm": 0.632544755935669, + "learning_rate": 0.0007633348274316451, + "loss": 0.4721, + "step": 8562 + }, + { + "epoch": 0.4784199793278766, + "grad_norm": 1.6563345193862915, + "learning_rate": 0.0007633068130883012, + "loss": 0.4364, + "step": 8563 + }, + { + "epoch": 0.4784758499315585, + "grad_norm": 0.8653427362442017, + "learning_rate": 0.0007632787987449574, + "loss": 0.4614, + "step": 8564 + }, + { + "epoch": 0.4785317205352404, + "grad_norm": 0.7559834718704224, + "learning_rate": 0.0007632507844016137, + "loss": 0.4783, + "step": 8565 + }, + { + "epoch": 0.47858759113892224, + "grad_norm": 0.39020806550979614, + "learning_rate": 0.0007632227700582699, + "loss": 0.4314, + "step": 8566 + }, + { + "epoch": 0.47864346174260414, + "grad_norm": 4.129411697387695, + "learning_rate": 0.000763194755714926, + "loss": 0.4494, + "step": 8567 + }, + { + "epoch": 0.478699332346286, + "grad_norm": 0.5992353558540344, + "learning_rate": 0.0007631667413715823, + "loss": 0.5059, + "step": 8568 + }, + { + "epoch": 0.4787552029499679, + "grad_norm": 0.9497689604759216, + "learning_rate": 0.0007631387270282385, + "loss": 0.4369, + "step": 8569 + }, + { + "epoch": 0.47881107355364977, + "grad_norm": 0.6580318212509155, + "learning_rate": 0.0007631107126848947, + "loss": 0.5396, + "step": 8570 + }, + { + "epoch": 0.4788669441573316, + "grad_norm": 0.8042258024215698, + "learning_rate": 0.0007630826983415508, + "loss": 0.4064, + "step": 8571 + }, + { + "epoch": 0.4789228147610135, + "grad_norm": 1.2070605754852295, + "learning_rate": 0.0007630546839982071, + "loss": 0.3704, + "step": 8572 + }, + { + "epoch": 0.47897868536469534, + "grad_norm": 0.7461993098258972, + "learning_rate": 0.0007630266696548633, + "loss": 0.3986, + "step": 8573 + }, + { + "epoch": 0.47903455596837724, + "grad_norm": 0.49116384983062744, + "learning_rate": 0.0007629986553115194, + "loss": 0.4055, + "step": 8574 + }, + { + "epoch": 0.47909042657205914, + "grad_norm": 0.5769818425178528, + "learning_rate": 0.0007629706409681757, + "loss": 0.6005, + "step": 8575 + }, + { + "epoch": 0.479146297175741, + "grad_norm": 0.44770190119743347, + "learning_rate": 0.000762942626624832, + "loss": 0.4954, + "step": 8576 + }, + { + "epoch": 0.47920216777942287, + "grad_norm": 0.5402330160140991, + "learning_rate": 0.0007629146122814882, + "loss": 0.453, + "step": 8577 + }, + { + "epoch": 0.4792580383831047, + "grad_norm": 0.3983466327190399, + "learning_rate": 0.0007628865979381443, + "loss": 0.4521, + "step": 8578 + }, + { + "epoch": 0.4793139089867866, + "grad_norm": 3.075326442718506, + "learning_rate": 0.0007628585835948006, + "loss": 0.4018, + "step": 8579 + }, + { + "epoch": 0.47936977959046845, + "grad_norm": 0.374856561422348, + "learning_rate": 0.0007628305692514568, + "loss": 0.3755, + "step": 8580 + }, + { + "epoch": 0.47942565019415034, + "grad_norm": 0.5993520617485046, + "learning_rate": 0.000762802554908113, + "loss": 0.5069, + "step": 8581 + }, + { + "epoch": 0.47948152079783224, + "grad_norm": 0.49952030181884766, + "learning_rate": 0.0007627745405647692, + "loss": 0.5366, + "step": 8582 + }, + { + "epoch": 0.4795373914015141, + "grad_norm": 0.710914134979248, + "learning_rate": 0.0007627465262214254, + "loss": 0.3208, + "step": 8583 + }, + { + "epoch": 0.479593262005196, + "grad_norm": 1.3201864957809448, + "learning_rate": 0.0007627185118780816, + "loss": 0.526, + "step": 8584 + }, + { + "epoch": 0.4796491326088778, + "grad_norm": 1.3332087993621826, + "learning_rate": 0.0007626904975347378, + "loss": 0.4026, + "step": 8585 + }, + { + "epoch": 0.4797050032125597, + "grad_norm": 0.5893604755401611, + "learning_rate": 0.000762662483191394, + "loss": 0.4445, + "step": 8586 + }, + { + "epoch": 0.4797608738162416, + "grad_norm": 0.6138483881950378, + "learning_rate": 0.0007626344688480502, + "loss": 0.5317, + "step": 8587 + }, + { + "epoch": 0.47981674441992345, + "grad_norm": 0.7711845636367798, + "learning_rate": 0.0007626064545047064, + "loss": 0.4289, + "step": 8588 + }, + { + "epoch": 0.47987261502360534, + "grad_norm": 0.4187864065170288, + "learning_rate": 0.0007625784401613627, + "loss": 0.4047, + "step": 8589 + }, + { + "epoch": 0.4799284856272872, + "grad_norm": 0.5332260727882385, + "learning_rate": 0.0007625504258180188, + "loss": 0.4055, + "step": 8590 + }, + { + "epoch": 0.4799843562309691, + "grad_norm": 0.6528044939041138, + "learning_rate": 0.000762522411474675, + "loss": 0.4047, + "step": 8591 + }, + { + "epoch": 0.480040226834651, + "grad_norm": 0.5286265015602112, + "learning_rate": 0.0007624943971313312, + "loss": 0.4371, + "step": 8592 + }, + { + "epoch": 0.4800960974383328, + "grad_norm": 0.48917216062545776, + "learning_rate": 0.0007624663827879875, + "loss": 0.3877, + "step": 8593 + }, + { + "epoch": 0.4801519680420147, + "grad_norm": 0.6614208221435547, + "learning_rate": 0.0007624383684446437, + "loss": 0.4589, + "step": 8594 + }, + { + "epoch": 0.48020783864569655, + "grad_norm": 0.4674195945262909, + "learning_rate": 0.0007624103541012999, + "loss": 0.4736, + "step": 8595 + }, + { + "epoch": 0.48026370924937845, + "grad_norm": 0.43961238861083984, + "learning_rate": 0.0007623823397579561, + "loss": 0.4507, + "step": 8596 + }, + { + "epoch": 0.4803195798530603, + "grad_norm": 0.42322567105293274, + "learning_rate": 0.0007623543254146123, + "loss": 0.39, + "step": 8597 + }, + { + "epoch": 0.4803754504567422, + "grad_norm": 0.7019840478897095, + "learning_rate": 0.0007623263110712685, + "loss": 0.5287, + "step": 8598 + }, + { + "epoch": 0.4804313210604241, + "grad_norm": 1.35270357131958, + "learning_rate": 0.0007622982967279247, + "loss": 0.4325, + "step": 8599 + }, + { + "epoch": 0.4804871916641059, + "grad_norm": 0.8553147315979004, + "learning_rate": 0.000762270282384581, + "loss": 0.4514, + "step": 8600 + }, + { + "epoch": 0.4805430622677878, + "grad_norm": 0.5061778426170349, + "learning_rate": 0.0007622422680412371, + "loss": 0.4543, + "step": 8601 + }, + { + "epoch": 0.48059893287146965, + "grad_norm": 0.5754634737968445, + "learning_rate": 0.0007622142536978933, + "loss": 0.3958, + "step": 8602 + }, + { + "epoch": 0.48065480347515155, + "grad_norm": 0.5297070145606995, + "learning_rate": 0.0007621862393545495, + "loss": 0.4636, + "step": 8603 + }, + { + "epoch": 0.48071067407883344, + "grad_norm": 1.9844098091125488, + "learning_rate": 0.0007621582250112058, + "loss": 0.4471, + "step": 8604 + }, + { + "epoch": 0.4807665446825153, + "grad_norm": 0.48687267303466797, + "learning_rate": 0.0007621302106678619, + "loss": 0.3978, + "step": 8605 + }, + { + "epoch": 0.4808224152861972, + "grad_norm": 1.2758585214614868, + "learning_rate": 0.0007621021963245181, + "loss": 0.4766, + "step": 8606 + }, + { + "epoch": 0.480878285889879, + "grad_norm": 0.8563536405563354, + "learning_rate": 0.0007620741819811744, + "loss": 0.6618, + "step": 8607 + }, + { + "epoch": 0.4809341564935609, + "grad_norm": 0.5017459392547607, + "learning_rate": 0.0007620461676378306, + "loss": 0.455, + "step": 8608 + }, + { + "epoch": 0.4809900270972428, + "grad_norm": 0.43233123421669006, + "learning_rate": 0.0007620181532944867, + "loss": 0.493, + "step": 8609 + }, + { + "epoch": 0.48104589770092465, + "grad_norm": 0.4563996493816376, + "learning_rate": 0.000761990138951143, + "loss": 0.3825, + "step": 8610 + }, + { + "epoch": 0.48110176830460655, + "grad_norm": 0.641373872756958, + "learning_rate": 0.0007619621246077993, + "loss": 0.4819, + "step": 8611 + }, + { + "epoch": 0.4811576389082884, + "grad_norm": 0.7680355310440063, + "learning_rate": 0.0007619341102644555, + "loss": 0.5647, + "step": 8612 + }, + { + "epoch": 0.4812135095119703, + "grad_norm": 0.5190313458442688, + "learning_rate": 0.0007619060959211116, + "loss": 0.5115, + "step": 8613 + }, + { + "epoch": 0.4812693801156521, + "grad_norm": 0.5553613305091858, + "learning_rate": 0.0007618780815777679, + "loss": 0.5724, + "step": 8614 + }, + { + "epoch": 0.481325250719334, + "grad_norm": 0.9102928042411804, + "learning_rate": 0.0007618500672344241, + "loss": 0.5836, + "step": 8615 + }, + { + "epoch": 0.4813811213230159, + "grad_norm": 0.7251793146133423, + "learning_rate": 0.0007618220528910802, + "loss": 0.4209, + "step": 8616 + }, + { + "epoch": 0.48143699192669775, + "grad_norm": 0.5892723798751831, + "learning_rate": 0.0007617940385477364, + "loss": 0.4321, + "step": 8617 + }, + { + "epoch": 0.48149286253037965, + "grad_norm": 0.6403515338897705, + "learning_rate": 0.0007617660242043927, + "loss": 0.3843, + "step": 8618 + }, + { + "epoch": 0.4815487331340615, + "grad_norm": 0.49221181869506836, + "learning_rate": 0.0007617380098610489, + "loss": 0.4918, + "step": 8619 + }, + { + "epoch": 0.4816046037377434, + "grad_norm": 0.36031702160835266, + "learning_rate": 0.000761709995517705, + "loss": 0.3406, + "step": 8620 + }, + { + "epoch": 0.4816604743414253, + "grad_norm": 0.5376152396202087, + "learning_rate": 0.0007616819811743613, + "loss": 0.5066, + "step": 8621 + }, + { + "epoch": 0.4817163449451071, + "grad_norm": 1.0386210680007935, + "learning_rate": 0.0007616539668310175, + "loss": 0.6901, + "step": 8622 + }, + { + "epoch": 0.481772215548789, + "grad_norm": 3.949735403060913, + "learning_rate": 0.0007616259524876737, + "loss": 0.4025, + "step": 8623 + }, + { + "epoch": 0.48182808615247086, + "grad_norm": 0.5257871150970459, + "learning_rate": 0.0007615979381443298, + "loss": 0.407, + "step": 8624 + }, + { + "epoch": 0.48188395675615275, + "grad_norm": 0.3843136429786682, + "learning_rate": 0.0007615699238009861, + "loss": 0.423, + "step": 8625 + }, + { + "epoch": 0.48193982735983465, + "grad_norm": 0.6408377885818481, + "learning_rate": 0.0007615419094576424, + "loss": 0.5079, + "step": 8626 + }, + { + "epoch": 0.4819956979635165, + "grad_norm": 0.5908397436141968, + "learning_rate": 0.0007615138951142986, + "loss": 0.4296, + "step": 8627 + }, + { + "epoch": 0.4820515685671984, + "grad_norm": 0.47330552339553833, + "learning_rate": 0.0007614858807709547, + "loss": 0.4068, + "step": 8628 + }, + { + "epoch": 0.4821074391708802, + "grad_norm": 0.4865715205669403, + "learning_rate": 0.000761457866427611, + "loss": 0.4528, + "step": 8629 + }, + { + "epoch": 0.4821633097745621, + "grad_norm": 9.611160278320312, + "learning_rate": 0.0007614298520842672, + "loss": 0.493, + "step": 8630 + }, + { + "epoch": 0.48221918037824396, + "grad_norm": 0.47474992275238037, + "learning_rate": 0.0007614018377409234, + "loss": 0.3942, + "step": 8631 + }, + { + "epoch": 0.48227505098192586, + "grad_norm": 1.02047860622406, + "learning_rate": 0.0007613738233975796, + "loss": 0.5434, + "step": 8632 + }, + { + "epoch": 0.48233092158560775, + "grad_norm": 0.6391206383705139, + "learning_rate": 0.0007613458090542358, + "loss": 0.3749, + "step": 8633 + }, + { + "epoch": 0.4823867921892896, + "grad_norm": 0.4269672632217407, + "learning_rate": 0.000761317794710892, + "loss": 0.4448, + "step": 8634 + }, + { + "epoch": 0.4824426627929715, + "grad_norm": 0.47639724612236023, + "learning_rate": 0.0007612897803675481, + "loss": 0.3959, + "step": 8635 + }, + { + "epoch": 0.48249853339665333, + "grad_norm": 0.7114033102989197, + "learning_rate": 0.0007612617660242044, + "loss": 0.506, + "step": 8636 + }, + { + "epoch": 0.4825544040003352, + "grad_norm": 0.9550843238830566, + "learning_rate": 0.0007612337516808606, + "loss": 0.5042, + "step": 8637 + }, + { + "epoch": 0.4826102746040171, + "grad_norm": 0.7466159462928772, + "learning_rate": 0.0007612057373375168, + "loss": 0.5095, + "step": 8638 + }, + { + "epoch": 0.48266614520769896, + "grad_norm": 0.4760816693305969, + "learning_rate": 0.000761177722994173, + "loss": 0.4191, + "step": 8639 + }, + { + "epoch": 0.48272201581138086, + "grad_norm": 0.4545590877532959, + "learning_rate": 0.0007611497086508292, + "loss": 0.5696, + "step": 8640 + }, + { + "epoch": 0.4827778864150627, + "grad_norm": 1.4396181106567383, + "learning_rate": 0.0007611216943074854, + "loss": 0.4985, + "step": 8641 + }, + { + "epoch": 0.4828337570187446, + "grad_norm": 0.5959773063659668, + "learning_rate": 0.0007610936799641416, + "loss": 0.4663, + "step": 8642 + }, + { + "epoch": 0.4828896276224265, + "grad_norm": 1.0295697450637817, + "learning_rate": 0.0007610656656207979, + "loss": 0.4919, + "step": 8643 + }, + { + "epoch": 0.4829454982261083, + "grad_norm": 0.7854496240615845, + "learning_rate": 0.0007610376512774541, + "loss": 0.5488, + "step": 8644 + }, + { + "epoch": 0.4830013688297902, + "grad_norm": 0.48476651310920715, + "learning_rate": 0.0007610096369341103, + "loss": 0.3613, + "step": 8645 + }, + { + "epoch": 0.48305723943347206, + "grad_norm": 0.42086437344551086, + "learning_rate": 0.0007609816225907666, + "loss": 0.4707, + "step": 8646 + }, + { + "epoch": 0.48311311003715396, + "grad_norm": 1.2639470100402832, + "learning_rate": 0.0007609536082474227, + "loss": 0.4986, + "step": 8647 + }, + { + "epoch": 0.4831689806408358, + "grad_norm": 0.8861746191978455, + "learning_rate": 0.0007609255939040789, + "loss": 0.5203, + "step": 8648 + }, + { + "epoch": 0.4832248512445177, + "grad_norm": 0.49344703555107117, + "learning_rate": 0.0007608975795607351, + "loss": 0.4257, + "step": 8649 + }, + { + "epoch": 0.4832807218481996, + "grad_norm": 6.351654529571533, + "learning_rate": 0.0007608695652173914, + "loss": 0.4564, + "step": 8650 + }, + { + "epoch": 0.48333659245188143, + "grad_norm": 0.6986609697341919, + "learning_rate": 0.0007608415508740475, + "loss": 0.4704, + "step": 8651 + }, + { + "epoch": 0.4833924630555633, + "grad_norm": 0.797666072845459, + "learning_rate": 0.0007608135365307037, + "loss": 0.4779, + "step": 8652 + }, + { + "epoch": 0.48344833365924517, + "grad_norm": 0.49833446741104126, + "learning_rate": 0.00076078552218736, + "loss": 0.4197, + "step": 8653 + }, + { + "epoch": 0.48350420426292706, + "grad_norm": 0.5637015104293823, + "learning_rate": 0.0007607575078440162, + "loss": 0.4372, + "step": 8654 + }, + { + "epoch": 0.48356007486660896, + "grad_norm": 0.7185378670692444, + "learning_rate": 0.0007607294935006723, + "loss": 0.5468, + "step": 8655 + }, + { + "epoch": 0.4836159454702908, + "grad_norm": 0.9171081185340881, + "learning_rate": 0.0007607014791573285, + "loss": 0.4813, + "step": 8656 + }, + { + "epoch": 0.4836718160739727, + "grad_norm": 0.5369975566864014, + "learning_rate": 0.0007606734648139848, + "loss": 0.5511, + "step": 8657 + }, + { + "epoch": 0.48372768667765453, + "grad_norm": 11.43269157409668, + "learning_rate": 0.0007606454504706409, + "loss": 0.3809, + "step": 8658 + }, + { + "epoch": 0.48378355728133643, + "grad_norm": 0.40529292821884155, + "learning_rate": 0.0007606174361272971, + "loss": 0.4506, + "step": 8659 + }, + { + "epoch": 0.4838394278850183, + "grad_norm": 0.7748820185661316, + "learning_rate": 0.0007605894217839535, + "loss": 0.4614, + "step": 8660 + }, + { + "epoch": 0.48389529848870017, + "grad_norm": 0.6861974596977234, + "learning_rate": 0.0007605614074406097, + "loss": 0.4723, + "step": 8661 + }, + { + "epoch": 0.48395116909238206, + "grad_norm": 0.5426914095878601, + "learning_rate": 0.0007605333930972658, + "loss": 0.5006, + "step": 8662 + }, + { + "epoch": 0.4840070396960639, + "grad_norm": 0.9186028242111206, + "learning_rate": 0.000760505378753922, + "loss": 0.4117, + "step": 8663 + }, + { + "epoch": 0.4840629102997458, + "grad_norm": 0.36504268646240234, + "learning_rate": 0.0007604773644105783, + "loss": 0.4446, + "step": 8664 + }, + { + "epoch": 0.48411878090342764, + "grad_norm": 2.459787130355835, + "learning_rate": 0.0007604493500672345, + "loss": 0.4218, + "step": 8665 + }, + { + "epoch": 0.48417465150710953, + "grad_norm": 0.4741845726966858, + "learning_rate": 0.0007604213357238906, + "loss": 0.3555, + "step": 8666 + }, + { + "epoch": 0.48423052211079143, + "grad_norm": 0.6841551661491394, + "learning_rate": 0.0007603933213805468, + "loss": 0.47, + "step": 8667 + }, + { + "epoch": 0.48428639271447327, + "grad_norm": 2.2929959297180176, + "learning_rate": 0.0007603653070372031, + "loss": 0.4624, + "step": 8668 + }, + { + "epoch": 0.48434226331815516, + "grad_norm": 0.46403202414512634, + "learning_rate": 0.0007603372926938593, + "loss": 0.429, + "step": 8669 + }, + { + "epoch": 0.484398133921837, + "grad_norm": 0.8585380911827087, + "learning_rate": 0.0007603092783505154, + "loss": 0.4193, + "step": 8670 + }, + { + "epoch": 0.4844540045255189, + "grad_norm": 0.7862564921379089, + "learning_rate": 0.0007602812640071717, + "loss": 0.4914, + "step": 8671 + }, + { + "epoch": 0.4845098751292008, + "grad_norm": 1.0834660530090332, + "learning_rate": 0.0007602532496638279, + "loss": 0.4153, + "step": 8672 + }, + { + "epoch": 0.48456574573288264, + "grad_norm": 0.5580418109893799, + "learning_rate": 0.0007602252353204841, + "loss": 0.4217, + "step": 8673 + }, + { + "epoch": 0.48462161633656453, + "grad_norm": 0.4062054455280304, + "learning_rate": 0.0007601972209771402, + "loss": 0.4152, + "step": 8674 + }, + { + "epoch": 0.48467748694024637, + "grad_norm": 0.5420044660568237, + "learning_rate": 0.0007601692066337965, + "loss": 0.4184, + "step": 8675 + }, + { + "epoch": 0.48473335754392827, + "grad_norm": 0.4235580861568451, + "learning_rate": 0.0007601411922904528, + "loss": 0.3037, + "step": 8676 + }, + { + "epoch": 0.4847892281476101, + "grad_norm": 0.49304473400115967, + "learning_rate": 0.0007601131779471088, + "loss": 0.4807, + "step": 8677 + }, + { + "epoch": 0.484845098751292, + "grad_norm": 0.6457188129425049, + "learning_rate": 0.0007600851636037652, + "loss": 0.4463, + "step": 8678 + }, + { + "epoch": 0.4849009693549739, + "grad_norm": 0.8899171352386475, + "learning_rate": 0.0007600571492604214, + "loss": 0.415, + "step": 8679 + }, + { + "epoch": 0.48495683995865574, + "grad_norm": 0.6433651447296143, + "learning_rate": 0.0007600291349170776, + "loss": 0.4527, + "step": 8680 + }, + { + "epoch": 0.48501271056233763, + "grad_norm": 0.5287381410598755, + "learning_rate": 0.0007600011205737337, + "loss": 0.4294, + "step": 8681 + }, + { + "epoch": 0.4850685811660195, + "grad_norm": 0.597300112247467, + "learning_rate": 0.00075997310623039, + "loss": 0.4707, + "step": 8682 + }, + { + "epoch": 0.48512445176970137, + "grad_norm": 0.5814911723136902, + "learning_rate": 0.0007599450918870462, + "loss": 0.4988, + "step": 8683 + }, + { + "epoch": 0.48518032237338327, + "grad_norm": 0.4991615116596222, + "learning_rate": 0.0007599170775437024, + "loss": 0.4857, + "step": 8684 + }, + { + "epoch": 0.4852361929770651, + "grad_norm": 1.6155718564987183, + "learning_rate": 0.0007598890632003586, + "loss": 0.3922, + "step": 8685 + }, + { + "epoch": 0.485292063580747, + "grad_norm": 0.4555036127567291, + "learning_rate": 0.0007598610488570148, + "loss": 0.4841, + "step": 8686 + }, + { + "epoch": 0.48534793418442884, + "grad_norm": 0.49474722146987915, + "learning_rate": 0.000759833034513671, + "loss": 0.5423, + "step": 8687 + }, + { + "epoch": 0.48540380478811074, + "grad_norm": 0.7960287928581238, + "learning_rate": 0.0007598050201703272, + "loss": 0.4456, + "step": 8688 + }, + { + "epoch": 0.48545967539179263, + "grad_norm": 3.7326138019561768, + "learning_rate": 0.0007597770058269834, + "loss": 0.3742, + "step": 8689 + }, + { + "epoch": 0.4855155459954745, + "grad_norm": 0.4632548689842224, + "learning_rate": 0.0007597489914836396, + "loss": 0.4924, + "step": 8690 + }, + { + "epoch": 0.48557141659915637, + "grad_norm": 0.5833700299263, + "learning_rate": 0.0007597209771402958, + "loss": 0.4435, + "step": 8691 + }, + { + "epoch": 0.4856272872028382, + "grad_norm": 0.42902863025665283, + "learning_rate": 0.0007596929627969522, + "loss": 0.4555, + "step": 8692 + }, + { + "epoch": 0.4856831578065201, + "grad_norm": 0.6478224396705627, + "learning_rate": 0.0007596649484536083, + "loss": 0.4366, + "step": 8693 + }, + { + "epoch": 0.48573902841020195, + "grad_norm": 0.659077525138855, + "learning_rate": 0.0007596369341102645, + "loss": 0.478, + "step": 8694 + }, + { + "epoch": 0.48579489901388384, + "grad_norm": 0.4467069208621979, + "learning_rate": 0.0007596089197669207, + "loss": 0.4327, + "step": 8695 + }, + { + "epoch": 0.48585076961756574, + "grad_norm": 1.0016745328903198, + "learning_rate": 0.000759580905423577, + "loss": 0.4296, + "step": 8696 + }, + { + "epoch": 0.4859066402212476, + "grad_norm": 0.8027512431144714, + "learning_rate": 0.0007595528910802331, + "loss": 0.5607, + "step": 8697 + }, + { + "epoch": 0.4859625108249295, + "grad_norm": 0.463395357131958, + "learning_rate": 0.0007595248767368893, + "loss": 0.4722, + "step": 8698 + }, + { + "epoch": 0.4860183814286113, + "grad_norm": 3.58368182182312, + "learning_rate": 0.0007594968623935456, + "loss": 0.5653, + "step": 8699 + }, + { + "epoch": 0.4860742520322932, + "grad_norm": 0.7608360648155212, + "learning_rate": 0.0007594688480502017, + "loss": 0.5239, + "step": 8700 + }, + { + "epoch": 0.4861301226359751, + "grad_norm": 0.46880650520324707, + "learning_rate": 0.0007594408337068579, + "loss": 0.5301, + "step": 8701 + }, + { + "epoch": 0.48618599323965694, + "grad_norm": 0.5723884105682373, + "learning_rate": 0.0007594128193635141, + "loss": 0.4805, + "step": 8702 + }, + { + "epoch": 0.48624186384333884, + "grad_norm": 0.5185813307762146, + "learning_rate": 0.0007593848050201704, + "loss": 0.5178, + "step": 8703 + }, + { + "epoch": 0.4862977344470207, + "grad_norm": 0.5995446443557739, + "learning_rate": 0.0007593567906768265, + "loss": 0.4421, + "step": 8704 + }, + { + "epoch": 0.4863536050507026, + "grad_norm": 0.938341498374939, + "learning_rate": 0.0007593287763334827, + "loss": 0.4988, + "step": 8705 + }, + { + "epoch": 0.48640947565438447, + "grad_norm": 0.5025823712348938, + "learning_rate": 0.0007593007619901389, + "loss": 0.4664, + "step": 8706 + }, + { + "epoch": 0.4864653462580663, + "grad_norm": 0.4653874635696411, + "learning_rate": 0.0007592727476467952, + "loss": 0.5592, + "step": 8707 + }, + { + "epoch": 0.4865212168617482, + "grad_norm": 1.1325112581253052, + "learning_rate": 0.0007592447333034513, + "loss": 0.4737, + "step": 8708 + }, + { + "epoch": 0.48657708746543005, + "grad_norm": 0.4870409369468689, + "learning_rate": 0.0007592167189601075, + "loss": 0.4271, + "step": 8709 + }, + { + "epoch": 0.48663295806911194, + "grad_norm": 0.5664789080619812, + "learning_rate": 0.0007591887046167639, + "loss": 0.452, + "step": 8710 + }, + { + "epoch": 0.4866888286727938, + "grad_norm": 0.6413889527320862, + "learning_rate": 0.0007591606902734201, + "loss": 0.5115, + "step": 8711 + }, + { + "epoch": 0.4867446992764757, + "grad_norm": 0.446461021900177, + "learning_rate": 0.0007591326759300762, + "loss": 0.459, + "step": 8712 + }, + { + "epoch": 0.4868005698801576, + "grad_norm": 2.348949670791626, + "learning_rate": 0.0007591046615867324, + "loss": 0.3835, + "step": 8713 + }, + { + "epoch": 0.4868564404838394, + "grad_norm": 0.5329850316047668, + "learning_rate": 0.0007590766472433887, + "loss": 0.3672, + "step": 8714 + }, + { + "epoch": 0.4869123110875213, + "grad_norm": 0.691714346408844, + "learning_rate": 0.0007590486329000449, + "loss": 0.4151, + "step": 8715 + }, + { + "epoch": 0.48696818169120315, + "grad_norm": 0.5502071976661682, + "learning_rate": 0.000759020618556701, + "loss": 0.3375, + "step": 8716 + }, + { + "epoch": 0.48702405229488505, + "grad_norm": 3.7882936000823975, + "learning_rate": 0.0007589926042133573, + "loss": 0.5179, + "step": 8717 + }, + { + "epoch": 0.48707992289856694, + "grad_norm": 0.6309230327606201, + "learning_rate": 0.0007589645898700135, + "loss": 0.4125, + "step": 8718 + }, + { + "epoch": 0.4871357935022488, + "grad_norm": 0.9361518025398254, + "learning_rate": 0.0007589365755266697, + "loss": 0.4149, + "step": 8719 + }, + { + "epoch": 0.4871916641059307, + "grad_norm": 13.227106094360352, + "learning_rate": 0.0007589085611833258, + "loss": 0.4275, + "step": 8720 + }, + { + "epoch": 0.4872475347096125, + "grad_norm": 0.4950893819332123, + "learning_rate": 0.0007588805468399821, + "loss": 0.4321, + "step": 8721 + }, + { + "epoch": 0.4873034053132944, + "grad_norm": 0.4208502769470215, + "learning_rate": 0.0007588525324966383, + "loss": 0.4748, + "step": 8722 + }, + { + "epoch": 0.4873592759169763, + "grad_norm": 0.8061321973800659, + "learning_rate": 0.0007588245181532944, + "loss": 0.6209, + "step": 8723 + }, + { + "epoch": 0.48741514652065815, + "grad_norm": 0.509686291217804, + "learning_rate": 0.0007587965038099507, + "loss": 0.5244, + "step": 8724 + }, + { + "epoch": 0.48747101712434004, + "grad_norm": 1.027457594871521, + "learning_rate": 0.0007587684894666069, + "loss": 0.4277, + "step": 8725 + }, + { + "epoch": 0.4875268877280219, + "grad_norm": 0.43829774856567383, + "learning_rate": 0.0007587404751232632, + "loss": 0.3731, + "step": 8726 + }, + { + "epoch": 0.4875827583317038, + "grad_norm": 0.8542407751083374, + "learning_rate": 0.0007587124607799192, + "loss": 0.5143, + "step": 8727 + }, + { + "epoch": 0.4876386289353856, + "grad_norm": 0.5564186573028564, + "learning_rate": 0.0007586844464365756, + "loss": 0.5744, + "step": 8728 + }, + { + "epoch": 0.4876944995390675, + "grad_norm": 2.987553119659424, + "learning_rate": 0.0007586564320932318, + "loss": 0.4415, + "step": 8729 + }, + { + "epoch": 0.4877503701427494, + "grad_norm": 0.6064049005508423, + "learning_rate": 0.000758628417749888, + "loss": 0.3628, + "step": 8730 + }, + { + "epoch": 0.48780624074643125, + "grad_norm": 1.3109406232833862, + "learning_rate": 0.0007586004034065442, + "loss": 0.4626, + "step": 8731 + }, + { + "epoch": 0.48786211135011315, + "grad_norm": 0.6986375451087952, + "learning_rate": 0.0007585723890632004, + "loss": 0.5121, + "step": 8732 + }, + { + "epoch": 0.487917981953795, + "grad_norm": 0.6275233626365662, + "learning_rate": 0.0007585443747198566, + "loss": 0.4145, + "step": 8733 + }, + { + "epoch": 0.4879738525574769, + "grad_norm": 0.479203999042511, + "learning_rate": 0.0007585163603765128, + "loss": 0.3135, + "step": 8734 + }, + { + "epoch": 0.4880297231611588, + "grad_norm": 0.46070173382759094, + "learning_rate": 0.000758488346033169, + "loss": 0.3682, + "step": 8735 + }, + { + "epoch": 0.4880855937648406, + "grad_norm": 7.160121917724609, + "learning_rate": 0.0007584603316898252, + "loss": 0.4051, + "step": 8736 + }, + { + "epoch": 0.4881414643685225, + "grad_norm": 0.5462408065795898, + "learning_rate": 0.0007584323173464814, + "loss": 0.4465, + "step": 8737 + }, + { + "epoch": 0.48819733497220436, + "grad_norm": 0.5301185846328735, + "learning_rate": 0.0007584043030031376, + "loss": 0.3938, + "step": 8738 + }, + { + "epoch": 0.48825320557588625, + "grad_norm": 0.44413378834724426, + "learning_rate": 0.0007583762886597938, + "loss": 0.4034, + "step": 8739 + }, + { + "epoch": 0.48830907617956815, + "grad_norm": 0.45678335428237915, + "learning_rate": 0.00075834827431645, + "loss": 0.4251, + "step": 8740 + }, + { + "epoch": 0.48836494678325, + "grad_norm": 0.4279894232749939, + "learning_rate": 0.0007583202599731062, + "loss": 0.4063, + "step": 8741 + }, + { + "epoch": 0.4884208173869319, + "grad_norm": 0.9239768981933594, + "learning_rate": 0.0007582922456297624, + "loss": 0.4549, + "step": 8742 + }, + { + "epoch": 0.4884766879906137, + "grad_norm": 0.39075472950935364, + "learning_rate": 0.0007582642312864186, + "loss": 0.4045, + "step": 8743 + }, + { + "epoch": 0.4885325585942956, + "grad_norm": 0.597169041633606, + "learning_rate": 0.0007582362169430749, + "loss": 0.453, + "step": 8744 + }, + { + "epoch": 0.48858842919797746, + "grad_norm": 0.5329403281211853, + "learning_rate": 0.0007582082025997311, + "loss": 0.5518, + "step": 8745 + }, + { + "epoch": 0.48864429980165935, + "grad_norm": 4.893217086791992, + "learning_rate": 0.0007581801882563873, + "loss": 0.4395, + "step": 8746 + }, + { + "epoch": 0.48870017040534125, + "grad_norm": 1.4497296810150146, + "learning_rate": 0.0007581521739130435, + "loss": 0.6846, + "step": 8747 + }, + { + "epoch": 0.4887560410090231, + "grad_norm": 1.0280510187149048, + "learning_rate": 0.0007581241595696997, + "loss": 0.5894, + "step": 8748 + }, + { + "epoch": 0.488811911612705, + "grad_norm": 0.5828646421432495, + "learning_rate": 0.000758096145226356, + "loss": 0.4659, + "step": 8749 + }, + { + "epoch": 0.4888677822163868, + "grad_norm": 0.7532014846801758, + "learning_rate": 0.0007580681308830121, + "loss": 0.5214, + "step": 8750 + }, + { + "epoch": 0.4889236528200687, + "grad_norm": 0.6834123134613037, + "learning_rate": 0.0007580401165396683, + "loss": 0.5586, + "step": 8751 + }, + { + "epoch": 0.4889795234237506, + "grad_norm": 0.4973972737789154, + "learning_rate": 0.0007580121021963245, + "loss": 0.5051, + "step": 8752 + }, + { + "epoch": 0.48903539402743246, + "grad_norm": 1.1209625005722046, + "learning_rate": 0.0007579840878529808, + "loss": 0.5324, + "step": 8753 + }, + { + "epoch": 0.48909126463111435, + "grad_norm": 0.8376035690307617, + "learning_rate": 0.0007579560735096369, + "loss": 0.4851, + "step": 8754 + }, + { + "epoch": 0.4891471352347962, + "grad_norm": 0.807746410369873, + "learning_rate": 0.0007579280591662931, + "loss": 0.427, + "step": 8755 + }, + { + "epoch": 0.4892030058384781, + "grad_norm": 0.5414220094680786, + "learning_rate": 0.0007579000448229494, + "loss": 0.4682, + "step": 8756 + }, + { + "epoch": 0.48925887644216, + "grad_norm": 0.40343761444091797, + "learning_rate": 0.0007578720304796056, + "loss": 0.3451, + "step": 8757 + }, + { + "epoch": 0.4893147470458418, + "grad_norm": 0.4600204825401306, + "learning_rate": 0.0007578440161362617, + "loss": 0.4484, + "step": 8758 + }, + { + "epoch": 0.4893706176495237, + "grad_norm": 0.7554013729095459, + "learning_rate": 0.0007578160017929179, + "loss": 0.4494, + "step": 8759 + }, + { + "epoch": 0.48942648825320556, + "grad_norm": 0.7938094735145569, + "learning_rate": 0.0007577879874495743, + "loss": 0.5234, + "step": 8760 + }, + { + "epoch": 0.48948235885688746, + "grad_norm": 0.5839000940322876, + "learning_rate": 0.0007577599731062305, + "loss": 0.545, + "step": 8761 + }, + { + "epoch": 0.4895382294605693, + "grad_norm": 1.0542652606964111, + "learning_rate": 0.0007577319587628866, + "loss": 0.4856, + "step": 8762 + }, + { + "epoch": 0.4895941000642512, + "grad_norm": 0.7609893679618835, + "learning_rate": 0.0007577039444195429, + "loss": 0.4057, + "step": 8763 + }, + { + "epoch": 0.4896499706679331, + "grad_norm": 0.5585260391235352, + "learning_rate": 0.0007576759300761991, + "loss": 0.4522, + "step": 8764 + }, + { + "epoch": 0.48970584127161493, + "grad_norm": 0.66910320520401, + "learning_rate": 0.0007576479157328552, + "loss": 0.3924, + "step": 8765 + }, + { + "epoch": 0.4897617118752968, + "grad_norm": 0.495817095041275, + "learning_rate": 0.0007576199013895114, + "loss": 0.4784, + "step": 8766 + }, + { + "epoch": 0.48981758247897866, + "grad_norm": 0.45259177684783936, + "learning_rate": 0.0007575918870461677, + "loss": 0.4417, + "step": 8767 + }, + { + "epoch": 0.48987345308266056, + "grad_norm": 0.7163907289505005, + "learning_rate": 0.0007575638727028239, + "loss": 0.41, + "step": 8768 + }, + { + "epoch": 0.48992932368634245, + "grad_norm": 0.39949265122413635, + "learning_rate": 0.00075753585835948, + "loss": 0.4403, + "step": 8769 + }, + { + "epoch": 0.4899851942900243, + "grad_norm": 0.3841201364994049, + "learning_rate": 0.0007575078440161362, + "loss": 0.4871, + "step": 8770 + }, + { + "epoch": 0.4900410648937062, + "grad_norm": 0.7097846269607544, + "learning_rate": 0.0007574798296727925, + "loss": 0.3859, + "step": 8771 + }, + { + "epoch": 0.49009693549738803, + "grad_norm": 0.4706135392189026, + "learning_rate": 0.0007574518153294487, + "loss": 0.4628, + "step": 8772 + }, + { + "epoch": 0.4901528061010699, + "grad_norm": 0.46622800827026367, + "learning_rate": 0.0007574238009861048, + "loss": 0.4711, + "step": 8773 + }, + { + "epoch": 0.4902086767047518, + "grad_norm": 0.6570150256156921, + "learning_rate": 0.0007573957866427611, + "loss": 0.5111, + "step": 8774 + }, + { + "epoch": 0.49026454730843366, + "grad_norm": 0.45598670840263367, + "learning_rate": 0.0007573677722994173, + "loss": 0.4945, + "step": 8775 + }, + { + "epoch": 0.49032041791211556, + "grad_norm": 0.4793655574321747, + "learning_rate": 0.0007573397579560735, + "loss": 0.5491, + "step": 8776 + }, + { + "epoch": 0.4903762885157974, + "grad_norm": 2.223771572113037, + "learning_rate": 0.0007573117436127296, + "loss": 0.5185, + "step": 8777 + }, + { + "epoch": 0.4904321591194793, + "grad_norm": 0.5929295420646667, + "learning_rate": 0.000757283729269386, + "loss": 0.4342, + "step": 8778 + }, + { + "epoch": 0.49048802972316113, + "grad_norm": 0.5071588754653931, + "learning_rate": 0.0007572557149260422, + "loss": 0.5559, + "step": 8779 + }, + { + "epoch": 0.49054390032684303, + "grad_norm": 0.5095340609550476, + "learning_rate": 0.0007572277005826984, + "loss": 0.413, + "step": 8780 + }, + { + "epoch": 0.4905997709305249, + "grad_norm": 0.48880529403686523, + "learning_rate": 0.0007571996862393546, + "loss": 0.3579, + "step": 8781 + }, + { + "epoch": 0.49065564153420677, + "grad_norm": 0.8182757496833801, + "learning_rate": 0.0007571716718960108, + "loss": 0.6445, + "step": 8782 + }, + { + "epoch": 0.49071151213788866, + "grad_norm": 8.332427978515625, + "learning_rate": 0.000757143657552667, + "loss": 0.4706, + "step": 8783 + }, + { + "epoch": 0.4907673827415705, + "grad_norm": 0.412352979183197, + "learning_rate": 0.0007571156432093231, + "loss": 0.3599, + "step": 8784 + }, + { + "epoch": 0.4908232533452524, + "grad_norm": 0.40969452261924744, + "learning_rate": 0.0007570876288659794, + "loss": 0.5152, + "step": 8785 + }, + { + "epoch": 0.4908791239489343, + "grad_norm": 0.4832536578178406, + "learning_rate": 0.0007570596145226356, + "loss": 0.4055, + "step": 8786 + }, + { + "epoch": 0.49093499455261613, + "grad_norm": 0.4725615978240967, + "learning_rate": 0.0007570316001792918, + "loss": 0.4874, + "step": 8787 + }, + { + "epoch": 0.49099086515629803, + "grad_norm": 0.6335259079933167, + "learning_rate": 0.000757003585835948, + "loss": 0.4614, + "step": 8788 + }, + { + "epoch": 0.49104673575997987, + "grad_norm": 0.5354874134063721, + "learning_rate": 0.0007569755714926042, + "loss": 0.4648, + "step": 8789 + }, + { + "epoch": 0.49110260636366176, + "grad_norm": 0.5246819853782654, + "learning_rate": 0.0007569475571492604, + "loss": 0.5319, + "step": 8790 + }, + { + "epoch": 0.4911584769673436, + "grad_norm": 7.138073921203613, + "learning_rate": 0.0007569195428059166, + "loss": 0.4862, + "step": 8791 + }, + { + "epoch": 0.4912143475710255, + "grad_norm": 0.4583141803741455, + "learning_rate": 0.0007568915284625728, + "loss": 0.3704, + "step": 8792 + }, + { + "epoch": 0.4912702181747074, + "grad_norm": 0.4477171301841736, + "learning_rate": 0.000756863514119229, + "loss": 0.3213, + "step": 8793 + }, + { + "epoch": 0.49132608877838924, + "grad_norm": 2.7276968955993652, + "learning_rate": 0.0007568354997758853, + "loss": 0.4603, + "step": 8794 + }, + { + "epoch": 0.49138195938207113, + "grad_norm": 0.4219051003456116, + "learning_rate": 0.0007568074854325416, + "loss": 0.3618, + "step": 8795 + }, + { + "epoch": 0.49143782998575297, + "grad_norm": 0.8890396356582642, + "learning_rate": 0.0007567794710891977, + "loss": 0.4351, + "step": 8796 + }, + { + "epoch": 0.49149370058943487, + "grad_norm": 0.4105888307094574, + "learning_rate": 0.0007567514567458539, + "loss": 0.5966, + "step": 8797 + }, + { + "epoch": 0.49154957119311676, + "grad_norm": 0.5141633152961731, + "learning_rate": 0.0007567234424025101, + "loss": 0.4754, + "step": 8798 + }, + { + "epoch": 0.4916054417967986, + "grad_norm": 0.38025036454200745, + "learning_rate": 0.0007566954280591664, + "loss": 0.4611, + "step": 8799 + }, + { + "epoch": 0.4916613124004805, + "grad_norm": 0.4973037838935852, + "learning_rate": 0.0007566674137158225, + "loss": 0.3884, + "step": 8800 + }, + { + "epoch": 0.49171718300416234, + "grad_norm": 0.6534000635147095, + "learning_rate": 0.0007566393993724787, + "loss": 0.4821, + "step": 8801 + }, + { + "epoch": 0.49177305360784423, + "grad_norm": 0.6527429819107056, + "learning_rate": 0.000756611385029135, + "loss": 0.5106, + "step": 8802 + }, + { + "epoch": 0.49182892421152613, + "grad_norm": 0.5991565585136414, + "learning_rate": 0.0007565833706857912, + "loss": 0.4267, + "step": 8803 + }, + { + "epoch": 0.49188479481520797, + "grad_norm": 0.5205228328704834, + "learning_rate": 0.0007565553563424473, + "loss": 0.4813, + "step": 8804 + }, + { + "epoch": 0.49194066541888987, + "grad_norm": 2.069337844848633, + "learning_rate": 0.0007565273419991035, + "loss": 0.5437, + "step": 8805 + }, + { + "epoch": 0.4919965360225717, + "grad_norm": 0.6874570846557617, + "learning_rate": 0.0007564993276557598, + "loss": 0.3778, + "step": 8806 + }, + { + "epoch": 0.4920524066262536, + "grad_norm": 0.5654933452606201, + "learning_rate": 0.0007564713133124159, + "loss": 0.3993, + "step": 8807 + }, + { + "epoch": 0.49210827722993544, + "grad_norm": 0.7499202489852905, + "learning_rate": 0.0007564432989690721, + "loss": 0.4035, + "step": 8808 + }, + { + "epoch": 0.49216414783361734, + "grad_norm": 0.5103409290313721, + "learning_rate": 0.0007564152846257283, + "loss": 0.4927, + "step": 8809 + }, + { + "epoch": 0.49222001843729923, + "grad_norm": 1.3699233531951904, + "learning_rate": 0.0007563872702823847, + "loss": 0.4261, + "step": 8810 + }, + { + "epoch": 0.4922758890409811, + "grad_norm": 0.7152618765830994, + "learning_rate": 0.0007563592559390408, + "loss": 0.5315, + "step": 8811 + }, + { + "epoch": 0.49233175964466297, + "grad_norm": 0.5998836159706116, + "learning_rate": 0.000756331241595697, + "loss": 0.5243, + "step": 8812 + }, + { + "epoch": 0.4923876302483448, + "grad_norm": 0.47151196002960205, + "learning_rate": 0.0007563032272523533, + "loss": 0.388, + "step": 8813 + }, + { + "epoch": 0.4924435008520267, + "grad_norm": 0.6020907759666443, + "learning_rate": 0.0007562752129090095, + "loss": 0.4932, + "step": 8814 + }, + { + "epoch": 0.4924993714557086, + "grad_norm": 0.6177471280097961, + "learning_rate": 0.0007562471985656656, + "loss": 0.5098, + "step": 8815 + }, + { + "epoch": 0.49255524205939044, + "grad_norm": 3.209350109100342, + "learning_rate": 0.0007562191842223218, + "loss": 0.473, + "step": 8816 + }, + { + "epoch": 0.49261111266307234, + "grad_norm": 0.8462221622467041, + "learning_rate": 0.0007561911698789781, + "loss": 0.5896, + "step": 8817 + }, + { + "epoch": 0.4926669832667542, + "grad_norm": 0.542738676071167, + "learning_rate": 0.0007561631555356343, + "loss": 0.535, + "step": 8818 + }, + { + "epoch": 0.4927228538704361, + "grad_norm": 2.1922268867492676, + "learning_rate": 0.0007561351411922904, + "loss": 0.3996, + "step": 8819 + }, + { + "epoch": 0.49277872447411797, + "grad_norm": 0.45895734429359436, + "learning_rate": 0.0007561071268489467, + "loss": 0.4153, + "step": 8820 + }, + { + "epoch": 0.4928345950777998, + "grad_norm": 0.6438764929771423, + "learning_rate": 0.0007560791125056029, + "loss": 0.5383, + "step": 8821 + }, + { + "epoch": 0.4928904656814817, + "grad_norm": 0.4777509868144989, + "learning_rate": 0.0007560510981622591, + "loss": 0.4339, + "step": 8822 + }, + { + "epoch": 0.49294633628516354, + "grad_norm": 1.6530466079711914, + "learning_rate": 0.0007560230838189152, + "loss": 0.4556, + "step": 8823 + }, + { + "epoch": 0.49300220688884544, + "grad_norm": 0.6238105297088623, + "learning_rate": 0.0007559950694755715, + "loss": 0.4353, + "step": 8824 + }, + { + "epoch": 0.4930580774925273, + "grad_norm": 0.46071451902389526, + "learning_rate": 0.0007559670551322277, + "loss": 0.4856, + "step": 8825 + }, + { + "epoch": 0.4931139480962092, + "grad_norm": 0.45509377121925354, + "learning_rate": 0.0007559390407888838, + "loss": 0.3598, + "step": 8826 + }, + { + "epoch": 0.49316981869989107, + "grad_norm": 0.4176367521286011, + "learning_rate": 0.0007559110264455402, + "loss": 0.4257, + "step": 8827 + }, + { + "epoch": 0.4932256893035729, + "grad_norm": 0.639230489730835, + "learning_rate": 0.0007558830121021964, + "loss": 0.5104, + "step": 8828 + }, + { + "epoch": 0.4932815599072548, + "grad_norm": 0.6461355686187744, + "learning_rate": 0.0007558549977588526, + "loss": 0.5676, + "step": 8829 + }, + { + "epoch": 0.49333743051093665, + "grad_norm": 0.5837709903717041, + "learning_rate": 0.0007558269834155087, + "loss": 0.3936, + "step": 8830 + }, + { + "epoch": 0.49339330111461854, + "grad_norm": 0.7174801826477051, + "learning_rate": 0.000755798969072165, + "loss": 0.5396, + "step": 8831 + }, + { + "epoch": 0.49344917171830044, + "grad_norm": 5.201401233673096, + "learning_rate": 0.0007557709547288212, + "loss": 0.4206, + "step": 8832 + }, + { + "epoch": 0.4935050423219823, + "grad_norm": 0.4709891676902771, + "learning_rate": 0.0007557429403854774, + "loss": 0.4447, + "step": 8833 + }, + { + "epoch": 0.4935609129256642, + "grad_norm": 0.6299450397491455, + "learning_rate": 0.0007557149260421336, + "loss": 0.4895, + "step": 8834 + }, + { + "epoch": 0.493616783529346, + "grad_norm": 0.6696729063987732, + "learning_rate": 0.0007556869116987898, + "loss": 0.4093, + "step": 8835 + }, + { + "epoch": 0.4936726541330279, + "grad_norm": 0.33379676938056946, + "learning_rate": 0.000755658897355446, + "loss": 0.3511, + "step": 8836 + }, + { + "epoch": 0.4937285247367098, + "grad_norm": 0.45555776357650757, + "learning_rate": 0.0007556308830121022, + "loss": 0.3892, + "step": 8837 + }, + { + "epoch": 0.49378439534039165, + "grad_norm": 0.5049326419830322, + "learning_rate": 0.0007556028686687584, + "loss": 0.5367, + "step": 8838 + }, + { + "epoch": 0.49384026594407354, + "grad_norm": 0.5028725862503052, + "learning_rate": 0.0007555748543254146, + "loss": 0.4846, + "step": 8839 + }, + { + "epoch": 0.4938961365477554, + "grad_norm": 0.8569187521934509, + "learning_rate": 0.0007555468399820708, + "loss": 0.5603, + "step": 8840 + }, + { + "epoch": 0.4939520071514373, + "grad_norm": 0.7303178310394287, + "learning_rate": 0.0007555188256387271, + "loss": 0.5308, + "step": 8841 + }, + { + "epoch": 0.4940078777551191, + "grad_norm": 0.9213618636131287, + "learning_rate": 0.0007554908112953832, + "loss": 0.465, + "step": 8842 + }, + { + "epoch": 0.494063748358801, + "grad_norm": 0.5721760988235474, + "learning_rate": 0.0007554627969520394, + "loss": 0.5234, + "step": 8843 + }, + { + "epoch": 0.4941196189624829, + "grad_norm": 0.7005208730697632, + "learning_rate": 0.0007554347826086957, + "loss": 0.8783, + "step": 8844 + }, + { + "epoch": 0.49417548956616475, + "grad_norm": 0.4730110466480255, + "learning_rate": 0.000755406768265352, + "loss": 0.3934, + "step": 8845 + }, + { + "epoch": 0.49423136016984665, + "grad_norm": 6.558583736419678, + "learning_rate": 0.0007553787539220081, + "loss": 0.4736, + "step": 8846 + }, + { + "epoch": 0.4942872307735285, + "grad_norm": 1.4648594856262207, + "learning_rate": 0.0007553507395786643, + "loss": 0.412, + "step": 8847 + }, + { + "epoch": 0.4943431013772104, + "grad_norm": 1.754983901977539, + "learning_rate": 0.0007553227252353205, + "loss": 0.4567, + "step": 8848 + }, + { + "epoch": 0.4943989719808923, + "grad_norm": 0.44418618083000183, + "learning_rate": 0.0007552947108919767, + "loss": 0.4494, + "step": 8849 + }, + { + "epoch": 0.4944548425845741, + "grad_norm": 0.6316025257110596, + "learning_rate": 0.0007552666965486329, + "loss": 0.4381, + "step": 8850 + }, + { + "epoch": 0.494510713188256, + "grad_norm": 0.6860252618789673, + "learning_rate": 0.0007552386822052891, + "loss": 0.4838, + "step": 8851 + }, + { + "epoch": 0.49456658379193785, + "grad_norm": 0.5782914161682129, + "learning_rate": 0.0007552106678619454, + "loss": 0.4367, + "step": 8852 + }, + { + "epoch": 0.49462245439561975, + "grad_norm": 0.4780147671699524, + "learning_rate": 0.0007551826535186015, + "loss": 0.471, + "step": 8853 + }, + { + "epoch": 0.49467832499930164, + "grad_norm": 0.8324599862098694, + "learning_rate": 0.0007551546391752577, + "loss": 0.5892, + "step": 8854 + }, + { + "epoch": 0.4947341956029835, + "grad_norm": 0.4458690285682678, + "learning_rate": 0.0007551266248319139, + "loss": 0.4667, + "step": 8855 + }, + { + "epoch": 0.4947900662066654, + "grad_norm": 0.4935368299484253, + "learning_rate": 0.0007550986104885702, + "loss": 0.4767, + "step": 8856 + }, + { + "epoch": 0.4948459368103472, + "grad_norm": 0.9124104380607605, + "learning_rate": 0.0007550705961452263, + "loss": 0.5241, + "step": 8857 + }, + { + "epoch": 0.4949018074140291, + "grad_norm": 0.4056267738342285, + "learning_rate": 0.0007550425818018825, + "loss": 0.4222, + "step": 8858 + }, + { + "epoch": 0.49495767801771096, + "grad_norm": 0.7023693323135376, + "learning_rate": 0.0007550145674585388, + "loss": 0.3755, + "step": 8859 + }, + { + "epoch": 0.49501354862139285, + "grad_norm": 0.7250544428825378, + "learning_rate": 0.0007549865531151951, + "loss": 0.4221, + "step": 8860 + }, + { + "epoch": 0.49506941922507475, + "grad_norm": 0.598080039024353, + "learning_rate": 0.0007549585387718512, + "loss": 0.4954, + "step": 8861 + }, + { + "epoch": 0.4951252898287566, + "grad_norm": 0.4906323552131653, + "learning_rate": 0.0007549305244285074, + "loss": 0.4695, + "step": 8862 + }, + { + "epoch": 0.4951811604324385, + "grad_norm": 0.43408647179603577, + "learning_rate": 0.0007549025100851637, + "loss": 0.4066, + "step": 8863 + }, + { + "epoch": 0.4952370310361203, + "grad_norm": 0.5591174960136414, + "learning_rate": 0.0007548744957418199, + "loss": 0.3814, + "step": 8864 + }, + { + "epoch": 0.4952929016398022, + "grad_norm": 0.6671318411827087, + "learning_rate": 0.000754846481398476, + "loss": 0.4491, + "step": 8865 + }, + { + "epoch": 0.4953487722434841, + "grad_norm": 0.6056296825408936, + "learning_rate": 0.0007548184670551323, + "loss": 0.3848, + "step": 8866 + }, + { + "epoch": 0.49540464284716595, + "grad_norm": 1.3360919952392578, + "learning_rate": 0.0007547904527117885, + "loss": 0.4369, + "step": 8867 + }, + { + "epoch": 0.49546051345084785, + "grad_norm": 0.6478291749954224, + "learning_rate": 0.0007547624383684447, + "loss": 0.4857, + "step": 8868 + }, + { + "epoch": 0.4955163840545297, + "grad_norm": 0.619520902633667, + "learning_rate": 0.0007547344240251008, + "loss": 0.3272, + "step": 8869 + }, + { + "epoch": 0.4955722546582116, + "grad_norm": 0.5731959939002991, + "learning_rate": 0.0007547064096817571, + "loss": 0.5309, + "step": 8870 + }, + { + "epoch": 0.4956281252618935, + "grad_norm": 0.4557150602340698, + "learning_rate": 0.0007546783953384133, + "loss": 0.365, + "step": 8871 + }, + { + "epoch": 0.4956839958655753, + "grad_norm": 1.6877540349960327, + "learning_rate": 0.0007546503809950694, + "loss": 0.6481, + "step": 8872 + }, + { + "epoch": 0.4957398664692572, + "grad_norm": 0.5934830904006958, + "learning_rate": 0.0007546223666517257, + "loss": 0.5417, + "step": 8873 + }, + { + "epoch": 0.49579573707293906, + "grad_norm": 0.7078590989112854, + "learning_rate": 0.0007545943523083819, + "loss": 0.6508, + "step": 8874 + }, + { + "epoch": 0.49585160767662095, + "grad_norm": 0.3711836338043213, + "learning_rate": 0.0007545663379650381, + "loss": 0.3909, + "step": 8875 + }, + { + "epoch": 0.4959074782803028, + "grad_norm": 0.43117526173591614, + "learning_rate": 0.0007545383236216942, + "loss": 0.4014, + "step": 8876 + }, + { + "epoch": 0.4959633488839847, + "grad_norm": 0.4971355199813843, + "learning_rate": 0.0007545103092783506, + "loss": 0.5611, + "step": 8877 + }, + { + "epoch": 0.4960192194876666, + "grad_norm": 1.011406421661377, + "learning_rate": 0.0007544822949350068, + "loss": 0.465, + "step": 8878 + }, + { + "epoch": 0.4960750900913484, + "grad_norm": 0.67709881067276, + "learning_rate": 0.000754454280591663, + "loss": 0.3346, + "step": 8879 + }, + { + "epoch": 0.4961309606950303, + "grad_norm": 0.7010869383811951, + "learning_rate": 0.0007544262662483191, + "loss": 0.4596, + "step": 8880 + }, + { + "epoch": 0.49618683129871216, + "grad_norm": 1.2497836351394653, + "learning_rate": 0.0007543982519049754, + "loss": 0.4374, + "step": 8881 + }, + { + "epoch": 0.49624270190239406, + "grad_norm": 0.5943089127540588, + "learning_rate": 0.0007543702375616316, + "loss": 0.6425, + "step": 8882 + }, + { + "epoch": 0.49629857250607595, + "grad_norm": 0.40556105971336365, + "learning_rate": 0.0007543422232182878, + "loss": 0.434, + "step": 8883 + }, + { + "epoch": 0.4963544431097578, + "grad_norm": 0.5466732382774353, + "learning_rate": 0.000754314208874944, + "loss": 0.4738, + "step": 8884 + }, + { + "epoch": 0.4964103137134397, + "grad_norm": 0.44328513741493225, + "learning_rate": 0.0007542861945316002, + "loss": 0.4218, + "step": 8885 + }, + { + "epoch": 0.49646618431712153, + "grad_norm": 0.7219173312187195, + "learning_rate": 0.0007542581801882564, + "loss": 0.4496, + "step": 8886 + }, + { + "epoch": 0.4965220549208034, + "grad_norm": 0.6492074131965637, + "learning_rate": 0.0007542301658449126, + "loss": 0.5229, + "step": 8887 + }, + { + "epoch": 0.4965779255244853, + "grad_norm": 0.572975218296051, + "learning_rate": 0.0007542021515015688, + "loss": 0.4563, + "step": 8888 + }, + { + "epoch": 0.49663379612816716, + "grad_norm": 0.8426114916801453, + "learning_rate": 0.000754174137158225, + "loss": 0.6752, + "step": 8889 + }, + { + "epoch": 0.49668966673184906, + "grad_norm": 0.4448506534099579, + "learning_rate": 0.0007541461228148812, + "loss": 0.4044, + "step": 8890 + }, + { + "epoch": 0.4967455373355309, + "grad_norm": 0.933932363986969, + "learning_rate": 0.0007541181084715374, + "loss": 0.4108, + "step": 8891 + }, + { + "epoch": 0.4968014079392128, + "grad_norm": 2.2656023502349854, + "learning_rate": 0.0007540900941281936, + "loss": 0.3575, + "step": 8892 + }, + { + "epoch": 0.49685727854289463, + "grad_norm": 0.42158663272857666, + "learning_rate": 0.0007540620797848498, + "loss": 0.4604, + "step": 8893 + }, + { + "epoch": 0.4969131491465765, + "grad_norm": 0.7655993700027466, + "learning_rate": 0.000754034065441506, + "loss": 0.3936, + "step": 8894 + }, + { + "epoch": 0.4969690197502584, + "grad_norm": 1.5037869215011597, + "learning_rate": 0.0007540060510981623, + "loss": 0.4243, + "step": 8895 + }, + { + "epoch": 0.49702489035394026, + "grad_norm": 0.699792206287384, + "learning_rate": 0.0007539780367548185, + "loss": 0.4853, + "step": 8896 + }, + { + "epoch": 0.49708076095762216, + "grad_norm": 0.6091386079788208, + "learning_rate": 0.0007539500224114747, + "loss": 0.5542, + "step": 8897 + }, + { + "epoch": 0.497136631561304, + "grad_norm": 1.0791720151901245, + "learning_rate": 0.000753922008068131, + "loss": 0.485, + "step": 8898 + }, + { + "epoch": 0.4971925021649859, + "grad_norm": 0.6273187398910522, + "learning_rate": 0.0007538939937247871, + "loss": 0.3986, + "step": 8899 + }, + { + "epoch": 0.4972483727686678, + "grad_norm": 0.7797093987464905, + "learning_rate": 0.0007538659793814433, + "loss": 0.3704, + "step": 8900 + }, + { + "epoch": 0.49730424337234963, + "grad_norm": 0.9462474584579468, + "learning_rate": 0.0007538379650380995, + "loss": 0.7305, + "step": 8901 + }, + { + "epoch": 0.4973601139760315, + "grad_norm": 0.9756222367286682, + "learning_rate": 0.0007538099506947558, + "loss": 0.4693, + "step": 8902 + }, + { + "epoch": 0.49741598457971337, + "grad_norm": 0.6383190751075745, + "learning_rate": 0.0007537819363514119, + "loss": 0.4363, + "step": 8903 + }, + { + "epoch": 0.49747185518339526, + "grad_norm": 0.46564242243766785, + "learning_rate": 0.0007537539220080681, + "loss": 0.5091, + "step": 8904 + }, + { + "epoch": 0.4975277257870771, + "grad_norm": 0.465059757232666, + "learning_rate": 0.0007537259076647244, + "loss": 0.4965, + "step": 8905 + }, + { + "epoch": 0.497583596390759, + "grad_norm": 0.41991034150123596, + "learning_rate": 0.0007536978933213806, + "loss": 0.4525, + "step": 8906 + }, + { + "epoch": 0.4976394669944409, + "grad_norm": 0.737754225730896, + "learning_rate": 0.0007536698789780367, + "loss": 0.476, + "step": 8907 + }, + { + "epoch": 0.49769533759812273, + "grad_norm": 0.5868479609489441, + "learning_rate": 0.0007536418646346929, + "loss": 0.4809, + "step": 8908 + }, + { + "epoch": 0.49775120820180463, + "grad_norm": 0.5189733505249023, + "learning_rate": 0.0007536138502913492, + "loss": 0.3787, + "step": 8909 + }, + { + "epoch": 0.49780707880548647, + "grad_norm": 1.3675035238265991, + "learning_rate": 0.0007535858359480055, + "loss": 0.3453, + "step": 8910 + }, + { + "epoch": 0.49786294940916836, + "grad_norm": 0.3994930684566498, + "learning_rate": 0.0007535578216046616, + "loss": 0.3847, + "step": 8911 + }, + { + "epoch": 0.49791882001285026, + "grad_norm": 0.4370059072971344, + "learning_rate": 0.0007535298072613179, + "loss": 0.4235, + "step": 8912 + }, + { + "epoch": 0.4979746906165321, + "grad_norm": 0.567828893661499, + "learning_rate": 0.0007535017929179741, + "loss": 0.5432, + "step": 8913 + }, + { + "epoch": 0.498030561220214, + "grad_norm": 0.9793415069580078, + "learning_rate": 0.0007534737785746302, + "loss": 0.4337, + "step": 8914 + }, + { + "epoch": 0.49808643182389584, + "grad_norm": 0.9503465890884399, + "learning_rate": 0.0007534457642312864, + "loss": 0.489, + "step": 8915 + }, + { + "epoch": 0.49814230242757773, + "grad_norm": 0.6921616196632385, + "learning_rate": 0.0007534177498879427, + "loss": 0.3602, + "step": 8916 + }, + { + "epoch": 0.49819817303125963, + "grad_norm": 0.6372008919715881, + "learning_rate": 0.0007533897355445989, + "loss": 0.6307, + "step": 8917 + }, + { + "epoch": 0.49825404363494147, + "grad_norm": 0.9209797978401184, + "learning_rate": 0.000753361721201255, + "loss": 0.4415, + "step": 8918 + }, + { + "epoch": 0.49830991423862336, + "grad_norm": 0.40760940313339233, + "learning_rate": 0.0007533337068579112, + "loss": 0.4299, + "step": 8919 + }, + { + "epoch": 0.4983657848423052, + "grad_norm": 0.9915983080863953, + "learning_rate": 0.0007533056925145675, + "loss": 0.423, + "step": 8920 + }, + { + "epoch": 0.4984216554459871, + "grad_norm": 5.671023845672607, + "learning_rate": 0.0007532776781712237, + "loss": 0.5073, + "step": 8921 + }, + { + "epoch": 0.49847752604966894, + "grad_norm": 1.037713646888733, + "learning_rate": 0.0007532496638278798, + "loss": 0.4723, + "step": 8922 + }, + { + "epoch": 0.49853339665335084, + "grad_norm": 0.8411863446235657, + "learning_rate": 0.0007532216494845361, + "loss": 0.4612, + "step": 8923 + }, + { + "epoch": 0.49858926725703273, + "grad_norm": 0.6228397488594055, + "learning_rate": 0.0007531936351411923, + "loss": 0.4687, + "step": 8924 + }, + { + "epoch": 0.49864513786071457, + "grad_norm": 0.5396058559417725, + "learning_rate": 0.0007531656207978485, + "loss": 0.4121, + "step": 8925 + }, + { + "epoch": 0.49870100846439647, + "grad_norm": 0.5469385385513306, + "learning_rate": 0.0007531376064545046, + "loss": 0.4947, + "step": 8926 + }, + { + "epoch": 0.4987568790680783, + "grad_norm": 0.6040098071098328, + "learning_rate": 0.000753109592111161, + "loss": 0.635, + "step": 8927 + }, + { + "epoch": 0.4988127496717602, + "grad_norm": 1.6710044145584106, + "learning_rate": 0.0007530815777678172, + "loss": 0.3934, + "step": 8928 + }, + { + "epoch": 0.4988686202754421, + "grad_norm": 0.4942648708820343, + "learning_rate": 0.0007530535634244734, + "loss": 0.469, + "step": 8929 + }, + { + "epoch": 0.49892449087912394, + "grad_norm": 0.4672550857067108, + "learning_rate": 0.0007530255490811296, + "loss": 0.3296, + "step": 8930 + }, + { + "epoch": 0.49898036148280583, + "grad_norm": 0.5676379799842834, + "learning_rate": 0.0007529975347377858, + "loss": 0.4572, + "step": 8931 + }, + { + "epoch": 0.4990362320864877, + "grad_norm": 2.140721559524536, + "learning_rate": 0.000752969520394442, + "loss": 0.4273, + "step": 8932 + }, + { + "epoch": 0.49909210269016957, + "grad_norm": 0.6029700040817261, + "learning_rate": 0.0007529415060510981, + "loss": 0.5476, + "step": 8933 + }, + { + "epoch": 0.49914797329385147, + "grad_norm": 0.8704038262367249, + "learning_rate": 0.0007529134917077544, + "loss": 0.5322, + "step": 8934 + }, + { + "epoch": 0.4992038438975333, + "grad_norm": 0.5501376390457153, + "learning_rate": 0.0007528854773644106, + "loss": 0.4965, + "step": 8935 + }, + { + "epoch": 0.4992597145012152, + "grad_norm": 16.146209716796875, + "learning_rate": 0.0007528574630210668, + "loss": 0.5982, + "step": 8936 + }, + { + "epoch": 0.49931558510489704, + "grad_norm": 0.7697179913520813, + "learning_rate": 0.000752829448677723, + "loss": 0.4674, + "step": 8937 + }, + { + "epoch": 0.49937145570857894, + "grad_norm": 0.498274028301239, + "learning_rate": 0.0007528014343343792, + "loss": 0.4989, + "step": 8938 + }, + { + "epoch": 0.4994273263122608, + "grad_norm": 0.5341829657554626, + "learning_rate": 0.0007527734199910354, + "loss": 0.4274, + "step": 8939 + }, + { + "epoch": 0.4994831969159427, + "grad_norm": 0.538296639919281, + "learning_rate": 0.0007527454056476916, + "loss": 0.367, + "step": 8940 + }, + { + "epoch": 0.49953906751962457, + "grad_norm": 0.44971245527267456, + "learning_rate": 0.0007527173913043478, + "loss": 0.3744, + "step": 8941 + }, + { + "epoch": 0.4995949381233064, + "grad_norm": 0.4127325713634491, + "learning_rate": 0.000752689376961004, + "loss": 0.5666, + "step": 8942 + }, + { + "epoch": 0.4996508087269883, + "grad_norm": 0.39309367537498474, + "learning_rate": 0.0007526613626176602, + "loss": 0.4782, + "step": 8943 + }, + { + "epoch": 0.49970667933067014, + "grad_norm": 1.5005213022232056, + "learning_rate": 0.0007526333482743166, + "loss": 0.4426, + "step": 8944 + }, + { + "epoch": 0.49976254993435204, + "grad_norm": 0.7323582172393799, + "learning_rate": 0.0007526053339309727, + "loss": 0.4598, + "step": 8945 + }, + { + "epoch": 0.49981842053803394, + "grad_norm": 0.5529208183288574, + "learning_rate": 0.0007525773195876289, + "loss": 0.4259, + "step": 8946 + }, + { + "epoch": 0.4998742911417158, + "grad_norm": 15.37838077545166, + "learning_rate": 0.0007525493052442851, + "loss": 0.5896, + "step": 8947 + }, + { + "epoch": 0.49993016174539767, + "grad_norm": 0.5495063066482544, + "learning_rate": 0.0007525212909009414, + "loss": 0.5617, + "step": 8948 + }, + { + "epoch": 0.4999860323490795, + "grad_norm": 0.5304117202758789, + "learning_rate": 0.0007524932765575975, + "loss": 0.5028, + "step": 8949 + }, + { + "epoch": 0.5000419029527614, + "grad_norm": 0.7372130155563354, + "learning_rate": 0.0007524652622142537, + "loss": 0.4345, + "step": 8950 + }, + { + "epoch": 0.5000977735564432, + "grad_norm": 0.6499703526496887, + "learning_rate": 0.0007524372478709099, + "loss": 0.4385, + "step": 8951 + }, + { + "epoch": 0.5001536441601252, + "grad_norm": 0.4391862750053406, + "learning_rate": 0.0007524092335275662, + "loss": 0.3974, + "step": 8952 + }, + { + "epoch": 0.500209514763807, + "grad_norm": 0.5091111660003662, + "learning_rate": 0.0007523812191842223, + "loss": 0.3784, + "step": 8953 + }, + { + "epoch": 0.5002653853674889, + "grad_norm": 0.5741579532623291, + "learning_rate": 0.0007523532048408785, + "loss": 0.412, + "step": 8954 + }, + { + "epoch": 0.5003212559711707, + "grad_norm": 3.531191349029541, + "learning_rate": 0.0007523251904975348, + "loss": 0.526, + "step": 8955 + }, + { + "epoch": 0.5003771265748527, + "grad_norm": 0.4376312494277954, + "learning_rate": 0.0007522971761541909, + "loss": 0.4766, + "step": 8956 + }, + { + "epoch": 0.5004329971785345, + "grad_norm": 0.5312947034835815, + "learning_rate": 0.0007522691618108471, + "loss": 0.451, + "step": 8957 + }, + { + "epoch": 0.5004888677822164, + "grad_norm": 0.8048197627067566, + "learning_rate": 0.0007522411474675033, + "loss": 0.3984, + "step": 8958 + }, + { + "epoch": 0.5005447383858983, + "grad_norm": 0.4842889904975891, + "learning_rate": 0.0007522131331241596, + "loss": 0.3869, + "step": 8959 + }, + { + "epoch": 0.5006006089895801, + "grad_norm": 2.326453924179077, + "learning_rate": 0.0007521851187808157, + "loss": 0.4269, + "step": 8960 + }, + { + "epoch": 0.500656479593262, + "grad_norm": 1.0559864044189453, + "learning_rate": 0.000752157104437472, + "loss": 0.3788, + "step": 8961 + }, + { + "epoch": 0.5007123501969438, + "grad_norm": 0.4452824890613556, + "learning_rate": 0.0007521290900941283, + "loss": 0.4784, + "step": 8962 + }, + { + "epoch": 0.5007682208006258, + "grad_norm": 0.41248419880867004, + "learning_rate": 0.0007521010757507845, + "loss": 0.5119, + "step": 8963 + }, + { + "epoch": 0.5008240914043076, + "grad_norm": 0.5250813961029053, + "learning_rate": 0.0007520730614074406, + "loss": 0.4192, + "step": 8964 + }, + { + "epoch": 0.5008799620079895, + "grad_norm": 0.675563395023346, + "learning_rate": 0.0007520450470640968, + "loss": 0.6065, + "step": 8965 + }, + { + "epoch": 0.5009358326116714, + "grad_norm": 2.049633264541626, + "learning_rate": 0.0007520170327207531, + "loss": 0.7014, + "step": 8966 + }, + { + "epoch": 0.5009917032153532, + "grad_norm": 0.4691646695137024, + "learning_rate": 0.0007519890183774093, + "loss": 0.4966, + "step": 8967 + }, + { + "epoch": 0.5010475738190351, + "grad_norm": 0.5056072473526001, + "learning_rate": 0.0007519610040340654, + "loss": 0.4233, + "step": 8968 + }, + { + "epoch": 0.501103444422717, + "grad_norm": 0.6054498553276062, + "learning_rate": 0.0007519329896907217, + "loss": 0.3963, + "step": 8969 + }, + { + "epoch": 0.5011593150263989, + "grad_norm": 0.4178951680660248, + "learning_rate": 0.0007519049753473779, + "loss": 0.4426, + "step": 8970 + }, + { + "epoch": 0.5012151856300807, + "grad_norm": 0.4452814757823944, + "learning_rate": 0.0007518769610040341, + "loss": 0.4957, + "step": 8971 + }, + { + "epoch": 0.5012710562337626, + "grad_norm": 0.40991753339767456, + "learning_rate": 0.0007518489466606902, + "loss": 0.4372, + "step": 8972 + }, + { + "epoch": 0.5013269268374445, + "grad_norm": 0.4223722219467163, + "learning_rate": 0.0007518209323173465, + "loss": 0.4789, + "step": 8973 + }, + { + "epoch": 0.5013827974411263, + "grad_norm": 0.459852933883667, + "learning_rate": 0.0007517929179740027, + "loss": 0.3574, + "step": 8974 + }, + { + "epoch": 0.5014386680448082, + "grad_norm": 0.5394478440284729, + "learning_rate": 0.0007517649036306588, + "loss": 0.5638, + "step": 8975 + }, + { + "epoch": 0.5014945386484901, + "grad_norm": 0.7521314024925232, + "learning_rate": 0.0007517368892873151, + "loss": 0.4898, + "step": 8976 + }, + { + "epoch": 0.501550409252172, + "grad_norm": 0.5465013980865479, + "learning_rate": 0.0007517088749439714, + "loss": 0.4882, + "step": 8977 + }, + { + "epoch": 0.5016062798558538, + "grad_norm": 0.4685167372226715, + "learning_rate": 0.0007516808606006276, + "loss": 0.6173, + "step": 8978 + }, + { + "epoch": 0.5016621504595357, + "grad_norm": 0.5135801434516907, + "learning_rate": 0.0007516528462572837, + "loss": 0.6463, + "step": 8979 + }, + { + "epoch": 0.5017180210632176, + "grad_norm": 1.0483543872833252, + "learning_rate": 0.00075162483191394, + "loss": 0.4141, + "step": 8980 + }, + { + "epoch": 0.5017738916668995, + "grad_norm": 0.5255373120307922, + "learning_rate": 0.0007515968175705962, + "loss": 0.457, + "step": 8981 + }, + { + "epoch": 0.5018297622705813, + "grad_norm": 0.5174113512039185, + "learning_rate": 0.0007515688032272524, + "loss": 0.4066, + "step": 8982 + }, + { + "epoch": 0.5018856328742632, + "grad_norm": 0.5306497812271118, + "learning_rate": 0.0007515407888839085, + "loss": 0.5286, + "step": 8983 + }, + { + "epoch": 0.5019415034779451, + "grad_norm": 6.995298385620117, + "learning_rate": 0.0007515127745405648, + "loss": 0.469, + "step": 8984 + }, + { + "epoch": 0.5019973740816269, + "grad_norm": 0.7084858417510986, + "learning_rate": 0.000751484760197221, + "loss": 0.5209, + "step": 8985 + }, + { + "epoch": 0.5020532446853089, + "grad_norm": 1.0606426000595093, + "learning_rate": 0.0007514567458538772, + "loss": 0.558, + "step": 8986 + }, + { + "epoch": 0.5021091152889907, + "grad_norm": 4.042196750640869, + "learning_rate": 0.0007514287315105334, + "loss": 0.4421, + "step": 8987 + }, + { + "epoch": 0.5021649858926726, + "grad_norm": 0.3457331359386444, + "learning_rate": 0.0007514007171671896, + "loss": 0.3997, + "step": 8988 + }, + { + "epoch": 0.5022208564963544, + "grad_norm": 0.48976829648017883, + "learning_rate": 0.0007513727028238458, + "loss": 0.4691, + "step": 8989 + }, + { + "epoch": 0.5022767271000363, + "grad_norm": 0.7311646938323975, + "learning_rate": 0.000751344688480502, + "loss": 0.55, + "step": 8990 + }, + { + "epoch": 0.5023325977037182, + "grad_norm": 0.4838597774505615, + "learning_rate": 0.0007513166741371582, + "loss": 0.4465, + "step": 8991 + }, + { + "epoch": 0.5023884683074, + "grad_norm": 0.45388442277908325, + "learning_rate": 0.0007512886597938144, + "loss": 0.4889, + "step": 8992 + }, + { + "epoch": 0.502444338911082, + "grad_norm": 0.5266108512878418, + "learning_rate": 0.0007512606454504706, + "loss": 0.3866, + "step": 8993 + }, + { + "epoch": 0.5025002095147638, + "grad_norm": 0.575739324092865, + "learning_rate": 0.000751232631107127, + "loss": 0.5357, + "step": 8994 + }, + { + "epoch": 0.5025560801184457, + "grad_norm": 0.5436972975730896, + "learning_rate": 0.0007512046167637831, + "loss": 0.4404, + "step": 8995 + }, + { + "epoch": 0.5026119507221275, + "grad_norm": 1.1651039123535156, + "learning_rate": 0.0007511766024204393, + "loss": 0.4526, + "step": 8996 + }, + { + "epoch": 0.5026678213258094, + "grad_norm": 0.761415958404541, + "learning_rate": 0.0007511485880770955, + "loss": 0.4724, + "step": 8997 + }, + { + "epoch": 0.5027236919294913, + "grad_norm": 0.701293408870697, + "learning_rate": 0.0007511205737337517, + "loss": 0.3966, + "step": 8998 + }, + { + "epoch": 0.5027795625331731, + "grad_norm": 0.8774993419647217, + "learning_rate": 0.0007510925593904079, + "loss": 0.4641, + "step": 8999 + }, + { + "epoch": 0.5028354331368551, + "grad_norm": 0.9009296894073486, + "learning_rate": 0.0007510645450470641, + "loss": 0.4702, + "step": 9000 + }, + { + "epoch": 0.5028354331368551, + "eval_cer": 0.09293749249888165, + "eval_loss": 0.34716886281967163, + "eval_runtime": 56.276, + "eval_samples_per_second": 80.638, + "eval_steps_per_second": 5.047, + "eval_wer": 0.36722437920728473, + "step": 9000 + }, + { + "epoch": 0.5028913037405369, + "grad_norm": 0.31639334559440613, + "learning_rate": 0.0007510365307037204, + "loss": 0.3464, + "step": 9001 + }, + { + "epoch": 0.5029471743442188, + "grad_norm": 0.6824215054512024, + "learning_rate": 0.0007510085163603765, + "loss": 0.5078, + "step": 9002 + }, + { + "epoch": 0.5030030449479007, + "grad_norm": 0.6243409514427185, + "learning_rate": 0.0007509805020170327, + "loss": 0.4157, + "step": 9003 + }, + { + "epoch": 0.5030589155515826, + "grad_norm": 0.552995502948761, + "learning_rate": 0.0007509524876736889, + "loss": 0.6455, + "step": 9004 + }, + { + "epoch": 0.5031147861552644, + "grad_norm": 0.6879466772079468, + "learning_rate": 0.0007509244733303452, + "loss": 0.5511, + "step": 9005 + }, + { + "epoch": 0.5031706567589462, + "grad_norm": 0.3948444128036499, + "learning_rate": 0.0007508964589870013, + "loss": 0.4121, + "step": 9006 + }, + { + "epoch": 0.5032265273626282, + "grad_norm": 0.4749002754688263, + "learning_rate": 0.0007508684446436575, + "loss": 0.5112, + "step": 9007 + }, + { + "epoch": 0.50328239796631, + "grad_norm": 0.8731310367584229, + "learning_rate": 0.0007508404303003138, + "loss": 0.4499, + "step": 9008 + }, + { + "epoch": 0.5033382685699919, + "grad_norm": 4.981750011444092, + "learning_rate": 0.00075081241595697, + "loss": 0.4592, + "step": 9009 + }, + { + "epoch": 0.5033941391736738, + "grad_norm": 0.7657502293586731, + "learning_rate": 0.0007507844016136261, + "loss": 0.5754, + "step": 9010 + }, + { + "epoch": 0.5034500097773557, + "grad_norm": 0.6794905662536621, + "learning_rate": 0.0007507563872702824, + "loss": 0.4255, + "step": 9011 + }, + { + "epoch": 0.5035058803810375, + "grad_norm": 1.9186272621154785, + "learning_rate": 0.0007507283729269387, + "loss": 0.408, + "step": 9012 + }, + { + "epoch": 0.5035617509847193, + "grad_norm": 0.7236736416816711, + "learning_rate": 0.0007507003585835949, + "loss": 0.4978, + "step": 9013 + }, + { + "epoch": 0.5036176215884013, + "grad_norm": 0.5106865763664246, + "learning_rate": 0.000750672344240251, + "loss": 0.5016, + "step": 9014 + }, + { + "epoch": 0.5036734921920831, + "grad_norm": 0.4583682417869568, + "learning_rate": 0.0007506443298969073, + "loss": 0.4095, + "step": 9015 + }, + { + "epoch": 0.503729362795765, + "grad_norm": 0.5606458783149719, + "learning_rate": 0.0007506163155535635, + "loss": 0.4535, + "step": 9016 + }, + { + "epoch": 0.5037852333994469, + "grad_norm": 0.5734314918518066, + "learning_rate": 0.0007505883012102196, + "loss": 0.4409, + "step": 9017 + }, + { + "epoch": 0.5038411040031288, + "grad_norm": 0.538230836391449, + "learning_rate": 0.0007505602868668758, + "loss": 0.4788, + "step": 9018 + }, + { + "epoch": 0.5038969746068106, + "grad_norm": 0.5263999104499817, + "learning_rate": 0.0007505322725235321, + "loss": 0.3943, + "step": 9019 + }, + { + "epoch": 0.5039528452104926, + "grad_norm": 0.48352116346359253, + "learning_rate": 0.0007505042581801883, + "loss": 0.481, + "step": 9020 + }, + { + "epoch": 0.5040087158141744, + "grad_norm": 0.6678733229637146, + "learning_rate": 0.0007504762438368444, + "loss": 0.4807, + "step": 9021 + }, + { + "epoch": 0.5040645864178562, + "grad_norm": 0.49985361099243164, + "learning_rate": 0.0007504482294935006, + "loss": 0.4488, + "step": 9022 + }, + { + "epoch": 0.5041204570215381, + "grad_norm": 0.5101481676101685, + "learning_rate": 0.0007504202151501569, + "loss": 0.4738, + "step": 9023 + }, + { + "epoch": 0.50417632762522, + "grad_norm": 0.7749202847480774, + "learning_rate": 0.0007503922008068131, + "loss": 0.4229, + "step": 9024 + }, + { + "epoch": 0.5042321982289019, + "grad_norm": 1.2515226602554321, + "learning_rate": 0.0007503641864634692, + "loss": 0.4536, + "step": 9025 + }, + { + "epoch": 0.5042880688325837, + "grad_norm": 0.5758801102638245, + "learning_rate": 0.0007503361721201255, + "loss": 0.5343, + "step": 9026 + }, + { + "epoch": 0.5043439394362657, + "grad_norm": 1.8745170831680298, + "learning_rate": 0.0007503081577767818, + "loss": 0.3925, + "step": 9027 + }, + { + "epoch": 0.5043998100399475, + "grad_norm": 0.6487520933151245, + "learning_rate": 0.000750280143433438, + "loss": 0.453, + "step": 9028 + }, + { + "epoch": 0.5044556806436293, + "grad_norm": 0.48408031463623047, + "learning_rate": 0.0007502521290900941, + "loss": 0.4327, + "step": 9029 + }, + { + "epoch": 0.5045115512473112, + "grad_norm": 0.48120614886283875, + "learning_rate": 0.0007502241147467504, + "loss": 0.4204, + "step": 9030 + }, + { + "epoch": 0.5045674218509931, + "grad_norm": 0.4210514724254608, + "learning_rate": 0.0007501961004034066, + "loss": 0.3991, + "step": 9031 + }, + { + "epoch": 0.504623292454675, + "grad_norm": 0.5204105973243713, + "learning_rate": 0.0007501680860600628, + "loss": 0.3927, + "step": 9032 + }, + { + "epoch": 0.5046791630583568, + "grad_norm": 0.7705742716789246, + "learning_rate": 0.000750140071716719, + "loss": 0.516, + "step": 9033 + }, + { + "epoch": 0.5047350336620388, + "grad_norm": 0.4307080805301666, + "learning_rate": 0.0007501120573733752, + "loss": 0.6163, + "step": 9034 + }, + { + "epoch": 0.5047909042657206, + "grad_norm": 0.48713746666908264, + "learning_rate": 0.0007500840430300314, + "loss": 0.3934, + "step": 9035 + }, + { + "epoch": 0.5048467748694024, + "grad_norm": 0.5014331340789795, + "learning_rate": 0.0007500560286866876, + "loss": 0.5555, + "step": 9036 + }, + { + "epoch": 0.5049026454730844, + "grad_norm": 1.0431311130523682, + "learning_rate": 0.0007500280143433438, + "loss": 0.4002, + "step": 9037 + }, + { + "epoch": 0.5049585160767662, + "grad_norm": 0.5512183308601379, + "learning_rate": 0.00075, + "loss": 0.4485, + "step": 9038 + }, + { + "epoch": 0.5050143866804481, + "grad_norm": 0.45037466287612915, + "learning_rate": 0.0007499719856566562, + "loss": 0.5152, + "step": 9039 + }, + { + "epoch": 0.5050702572841299, + "grad_norm": 2.4718542098999023, + "learning_rate": 0.0007499439713133124, + "loss": 0.469, + "step": 9040 + }, + { + "epoch": 0.5051261278878119, + "grad_norm": 0.5508730411529541, + "learning_rate": 0.0007499159569699686, + "loss": 0.4924, + "step": 9041 + }, + { + "epoch": 0.5051819984914937, + "grad_norm": 0.6217725872993469, + "learning_rate": 0.0007498879426266248, + "loss": 0.5987, + "step": 9042 + }, + { + "epoch": 0.5052378690951755, + "grad_norm": 0.5148295164108276, + "learning_rate": 0.000749859928283281, + "loss": 0.5244, + "step": 9043 + }, + { + "epoch": 0.5052937396988575, + "grad_norm": 0.4555060863494873, + "learning_rate": 0.0007498319139399373, + "loss": 0.4046, + "step": 9044 + }, + { + "epoch": 0.5053496103025393, + "grad_norm": 0.4472290277481079, + "learning_rate": 0.0007498038995965935, + "loss": 0.4727, + "step": 9045 + }, + { + "epoch": 0.5054054809062212, + "grad_norm": 0.5192309617996216, + "learning_rate": 0.0007497758852532497, + "loss": 0.4958, + "step": 9046 + }, + { + "epoch": 0.505461351509903, + "grad_norm": 0.45290178060531616, + "learning_rate": 0.000749747870909906, + "loss": 0.5058, + "step": 9047 + }, + { + "epoch": 0.505517222113585, + "grad_norm": 0.7093987464904785, + "learning_rate": 0.0007497198565665621, + "loss": 0.5008, + "step": 9048 + }, + { + "epoch": 0.5055730927172668, + "grad_norm": 0.6077860593795776, + "learning_rate": 0.0007496918422232183, + "loss": 0.6271, + "step": 9049 + }, + { + "epoch": 0.5056289633209486, + "grad_norm": 0.6476551294326782, + "learning_rate": 0.0007496638278798745, + "loss": 0.4996, + "step": 9050 + }, + { + "epoch": 0.5056848339246306, + "grad_norm": 0.4350220859050751, + "learning_rate": 0.0007496358135365308, + "loss": 0.5145, + "step": 9051 + }, + { + "epoch": 0.5057407045283124, + "grad_norm": 0.3993472754955292, + "learning_rate": 0.0007496077991931869, + "loss": 0.3659, + "step": 9052 + }, + { + "epoch": 0.5057965751319943, + "grad_norm": 0.40965625643730164, + "learning_rate": 0.0007495797848498431, + "loss": 0.435, + "step": 9053 + }, + { + "epoch": 0.5058524457356762, + "grad_norm": 0.6535290479660034, + "learning_rate": 0.0007495517705064994, + "loss": 0.4693, + "step": 9054 + }, + { + "epoch": 0.5059083163393581, + "grad_norm": 0.7330450415611267, + "learning_rate": 0.0007495237561631556, + "loss": 0.4198, + "step": 9055 + }, + { + "epoch": 0.5059641869430399, + "grad_norm": 0.3705599009990692, + "learning_rate": 0.0007494957418198117, + "loss": 0.4277, + "step": 9056 + }, + { + "epoch": 0.5060200575467217, + "grad_norm": 0.6754529476165771, + "learning_rate": 0.0007494677274764679, + "loss": 0.3536, + "step": 9057 + }, + { + "epoch": 0.5060759281504037, + "grad_norm": 0.3983204960823059, + "learning_rate": 0.0007494397131331242, + "loss": 0.4799, + "step": 9058 + }, + { + "epoch": 0.5061317987540855, + "grad_norm": 0.5801072716712952, + "learning_rate": 0.0007494116987897804, + "loss": 0.505, + "step": 9059 + }, + { + "epoch": 0.5061876693577674, + "grad_norm": 0.8511034250259399, + "learning_rate": 0.0007493836844464365, + "loss": 0.5059, + "step": 9060 + }, + { + "epoch": 0.5062435399614493, + "grad_norm": 0.4586620628833771, + "learning_rate": 0.0007493556701030927, + "loss": 0.4169, + "step": 9061 + }, + { + "epoch": 0.5062994105651312, + "grad_norm": 0.5315724611282349, + "learning_rate": 0.0007493276557597491, + "loss": 0.4327, + "step": 9062 + }, + { + "epoch": 0.506355281168813, + "grad_norm": 0.40734949707984924, + "learning_rate": 0.0007492996414164052, + "loss": 0.3297, + "step": 9063 + }, + { + "epoch": 0.5064111517724948, + "grad_norm": 0.4848923981189728, + "learning_rate": 0.0007492716270730614, + "loss": 0.4394, + "step": 9064 + }, + { + "epoch": 0.5064670223761768, + "grad_norm": 12.479020118713379, + "learning_rate": 0.0007492436127297177, + "loss": 0.4758, + "step": 9065 + }, + { + "epoch": 0.5065228929798586, + "grad_norm": 0.7141255736351013, + "learning_rate": 0.0007492155983863739, + "loss": 0.5295, + "step": 9066 + }, + { + "epoch": 0.5065787635835405, + "grad_norm": 0.5766344666481018, + "learning_rate": 0.00074918758404303, + "loss": 0.5522, + "step": 9067 + }, + { + "epoch": 0.5066346341872224, + "grad_norm": 0.6207360625267029, + "learning_rate": 0.0007491595696996862, + "loss": 0.4775, + "step": 9068 + }, + { + "epoch": 0.5066905047909043, + "grad_norm": 0.5432892441749573, + "learning_rate": 0.0007491315553563425, + "loss": 0.4433, + "step": 9069 + }, + { + "epoch": 0.5067463753945861, + "grad_norm": 0.5206831097602844, + "learning_rate": 0.0007491035410129987, + "loss": 0.4521, + "step": 9070 + }, + { + "epoch": 0.5068022459982681, + "grad_norm": 0.5576988458633423, + "learning_rate": 0.0007490755266696548, + "loss": 0.5085, + "step": 9071 + }, + { + "epoch": 0.5068581166019499, + "grad_norm": 0.9290946125984192, + "learning_rate": 0.0007490475123263111, + "loss": 0.5814, + "step": 9072 + }, + { + "epoch": 0.5069139872056317, + "grad_norm": 0.4656401574611664, + "learning_rate": 0.0007490194979829673, + "loss": 0.3925, + "step": 9073 + }, + { + "epoch": 0.5069698578093136, + "grad_norm": 0.8648470044136047, + "learning_rate": 0.0007489914836396235, + "loss": 0.5013, + "step": 9074 + }, + { + "epoch": 0.5070257284129955, + "grad_norm": 0.5871793031692505, + "learning_rate": 0.0007489634692962796, + "loss": 0.4503, + "step": 9075 + }, + { + "epoch": 0.5070815990166774, + "grad_norm": 1.9396966695785522, + "learning_rate": 0.0007489354549529359, + "loss": 0.4989, + "step": 9076 + }, + { + "epoch": 0.5071374696203592, + "grad_norm": 0.8231956958770752, + "learning_rate": 0.0007489074406095922, + "loss": 0.6104, + "step": 9077 + }, + { + "epoch": 0.5071933402240412, + "grad_norm": 0.41291332244873047, + "learning_rate": 0.0007488794262662484, + "loss": 0.5003, + "step": 9078 + }, + { + "epoch": 0.507249210827723, + "grad_norm": 0.8650103211402893, + "learning_rate": 0.0007488514119229046, + "loss": 0.4379, + "step": 9079 + }, + { + "epoch": 0.5073050814314048, + "grad_norm": 0.4782946705818176, + "learning_rate": 0.0007488233975795608, + "loss": 0.4518, + "step": 9080 + }, + { + "epoch": 0.5073609520350867, + "grad_norm": 0.4415775537490845, + "learning_rate": 0.000748795383236217, + "loss": 0.4486, + "step": 9081 + }, + { + "epoch": 0.5074168226387686, + "grad_norm": 0.7125979065895081, + "learning_rate": 0.0007487673688928731, + "loss": 0.5166, + "step": 9082 + }, + { + "epoch": 0.5074726932424505, + "grad_norm": 0.4385545551776886, + "learning_rate": 0.0007487393545495294, + "loss": 0.4446, + "step": 9083 + }, + { + "epoch": 0.5075285638461323, + "grad_norm": 2.2506163120269775, + "learning_rate": 0.0007487113402061856, + "loss": 0.4562, + "step": 9084 + }, + { + "epoch": 0.5075844344498143, + "grad_norm": 0.7057067155838013, + "learning_rate": 0.0007486833258628418, + "loss": 0.3667, + "step": 9085 + }, + { + "epoch": 0.5076403050534961, + "grad_norm": 0.48509570956230164, + "learning_rate": 0.000748655311519498, + "loss": 0.4646, + "step": 9086 + }, + { + "epoch": 0.507696175657178, + "grad_norm": 0.45894232392311096, + "learning_rate": 0.0007486272971761542, + "loss": 0.4922, + "step": 9087 + }, + { + "epoch": 0.5077520462608599, + "grad_norm": 0.7865738272666931, + "learning_rate": 0.0007485992828328104, + "loss": 0.4083, + "step": 9088 + }, + { + "epoch": 0.5078079168645417, + "grad_norm": 0.5167746543884277, + "learning_rate": 0.0007485712684894666, + "loss": 0.4205, + "step": 9089 + }, + { + "epoch": 0.5078637874682236, + "grad_norm": 0.7160714268684387, + "learning_rate": 0.0007485432541461228, + "loss": 0.3924, + "step": 9090 + }, + { + "epoch": 0.5079196580719054, + "grad_norm": 0.9789039492607117, + "learning_rate": 0.000748515239802779, + "loss": 0.448, + "step": 9091 + }, + { + "epoch": 0.5079755286755874, + "grad_norm": 0.49711865186691284, + "learning_rate": 0.0007484872254594352, + "loss": 0.5739, + "step": 9092 + }, + { + "epoch": 0.5080313992792692, + "grad_norm": 0.5165126323699951, + "learning_rate": 0.0007484592111160916, + "loss": 0.4329, + "step": 9093 + }, + { + "epoch": 0.508087269882951, + "grad_norm": 0.6042811274528503, + "learning_rate": 0.0007484311967727476, + "loss": 0.5375, + "step": 9094 + }, + { + "epoch": 0.508143140486633, + "grad_norm": 2.036393642425537, + "learning_rate": 0.0007484031824294039, + "loss": 0.4958, + "step": 9095 + }, + { + "epoch": 0.5081990110903148, + "grad_norm": 1.0100491046905518, + "learning_rate": 0.0007483751680860601, + "loss": 0.5942, + "step": 9096 + }, + { + "epoch": 0.5082548816939967, + "grad_norm": 0.43437275290489197, + "learning_rate": 0.0007483471537427164, + "loss": 0.5518, + "step": 9097 + }, + { + "epoch": 0.5083107522976785, + "grad_norm": 0.7763335704803467, + "learning_rate": 0.0007483191393993725, + "loss": 0.3861, + "step": 9098 + }, + { + "epoch": 0.5083666229013605, + "grad_norm": 0.7658848166465759, + "learning_rate": 0.0007482911250560287, + "loss": 0.5001, + "step": 9099 + }, + { + "epoch": 0.5084224935050423, + "grad_norm": 0.6465545296669006, + "learning_rate": 0.0007482631107126849, + "loss": 0.4466, + "step": 9100 + }, + { + "epoch": 0.5084783641087242, + "grad_norm": 0.5193779468536377, + "learning_rate": 0.0007482350963693412, + "loss": 0.4094, + "step": 9101 + }, + { + "epoch": 0.5085342347124061, + "grad_norm": 7.09721040725708, + "learning_rate": 0.0007482070820259973, + "loss": 0.3935, + "step": 9102 + }, + { + "epoch": 0.508590105316088, + "grad_norm": 0.5787192583084106, + "learning_rate": 0.0007481790676826535, + "loss": 0.4989, + "step": 9103 + }, + { + "epoch": 0.5086459759197698, + "grad_norm": 0.4730759263038635, + "learning_rate": 0.0007481510533393098, + "loss": 0.4557, + "step": 9104 + }, + { + "epoch": 0.5087018465234517, + "grad_norm": 0.5495299696922302, + "learning_rate": 0.0007481230389959659, + "loss": 0.5152, + "step": 9105 + }, + { + "epoch": 0.5087577171271336, + "grad_norm": 0.481289803981781, + "learning_rate": 0.0007480950246526221, + "loss": 0.3746, + "step": 9106 + }, + { + "epoch": 0.5088135877308154, + "grad_norm": 0.831794798374176, + "learning_rate": 0.0007480670103092783, + "loss": 0.4543, + "step": 9107 + }, + { + "epoch": 0.5088694583344973, + "grad_norm": 0.6448431015014648, + "learning_rate": 0.0007480389959659346, + "loss": 0.5055, + "step": 9108 + }, + { + "epoch": 0.5089253289381792, + "grad_norm": 1.4708207845687866, + "learning_rate": 0.0007480109816225907, + "loss": 0.5693, + "step": 9109 + }, + { + "epoch": 0.508981199541861, + "grad_norm": 0.7427549958229065, + "learning_rate": 0.0007479829672792469, + "loss": 0.5927, + "step": 9110 + }, + { + "epoch": 0.5090370701455429, + "grad_norm": 0.44472432136535645, + "learning_rate": 0.0007479549529359033, + "loss": 0.2722, + "step": 9111 + }, + { + "epoch": 0.5090929407492248, + "grad_norm": 0.7680208086967468, + "learning_rate": 0.0007479269385925595, + "loss": 0.5171, + "step": 9112 + }, + { + "epoch": 0.5091488113529067, + "grad_norm": 0.678170382976532, + "learning_rate": 0.0007478989242492156, + "loss": 0.5466, + "step": 9113 + }, + { + "epoch": 0.5092046819565885, + "grad_norm": 0.9387204051017761, + "learning_rate": 0.0007478709099058718, + "loss": 0.5522, + "step": 9114 + }, + { + "epoch": 0.5092605525602704, + "grad_norm": 0.6104002594947815, + "learning_rate": 0.0007478428955625281, + "loss": 0.4196, + "step": 9115 + }, + { + "epoch": 0.5093164231639523, + "grad_norm": 1.2222832441329956, + "learning_rate": 0.0007478148812191843, + "loss": 0.3774, + "step": 9116 + }, + { + "epoch": 0.5093722937676342, + "grad_norm": 0.749352753162384, + "learning_rate": 0.0007477868668758404, + "loss": 0.6363, + "step": 9117 + }, + { + "epoch": 0.509428164371316, + "grad_norm": 0.3635939657688141, + "learning_rate": 0.0007477588525324967, + "loss": 0.4419, + "step": 9118 + }, + { + "epoch": 0.509484034974998, + "grad_norm": 0.5241103768348694, + "learning_rate": 0.0007477308381891529, + "loss": 0.4128, + "step": 9119 + }, + { + "epoch": 0.5095399055786798, + "grad_norm": 0.43841710686683655, + "learning_rate": 0.0007477028238458091, + "loss": 0.5253, + "step": 9120 + }, + { + "epoch": 0.5095957761823616, + "grad_norm": 0.4182918667793274, + "learning_rate": 0.0007476748095024652, + "loss": 0.5287, + "step": 9121 + }, + { + "epoch": 0.5096516467860436, + "grad_norm": 0.9066117405891418, + "learning_rate": 0.0007476467951591215, + "loss": 0.4841, + "step": 9122 + }, + { + "epoch": 0.5097075173897254, + "grad_norm": 1.5104665756225586, + "learning_rate": 0.0007476187808157777, + "loss": 0.5009, + "step": 9123 + }, + { + "epoch": 0.5097633879934073, + "grad_norm": 0.4759330153465271, + "learning_rate": 0.0007475907664724338, + "loss": 0.4685, + "step": 9124 + }, + { + "epoch": 0.5098192585970891, + "grad_norm": 1.2254196405410767, + "learning_rate": 0.0007475627521290901, + "loss": 0.5231, + "step": 9125 + }, + { + "epoch": 0.509875129200771, + "grad_norm": 0.6128411293029785, + "learning_rate": 0.0007475347377857463, + "loss": 0.5667, + "step": 9126 + }, + { + "epoch": 0.5099309998044529, + "grad_norm": 0.41445907950401306, + "learning_rate": 0.0007475067234424026, + "loss": 0.4013, + "step": 9127 + }, + { + "epoch": 0.5099868704081347, + "grad_norm": 0.4642314612865448, + "learning_rate": 0.0007474787090990586, + "loss": 0.4409, + "step": 9128 + }, + { + "epoch": 0.5100427410118167, + "grad_norm": 2.423947334289551, + "learning_rate": 0.000747450694755715, + "loss": 0.4282, + "step": 9129 + }, + { + "epoch": 0.5100986116154985, + "grad_norm": 0.4488952159881592, + "learning_rate": 0.0007474226804123712, + "loss": 0.531, + "step": 9130 + }, + { + "epoch": 0.5101544822191804, + "grad_norm": 0.5888784527778625, + "learning_rate": 0.0007473946660690274, + "loss": 0.4076, + "step": 9131 + }, + { + "epoch": 0.5102103528228622, + "grad_norm": 0.44491657614707947, + "learning_rate": 0.0007473666517256835, + "loss": 0.3472, + "step": 9132 + }, + { + "epoch": 0.5102662234265442, + "grad_norm": 0.5955325961112976, + "learning_rate": 0.0007473386373823398, + "loss": 0.4049, + "step": 9133 + }, + { + "epoch": 0.510322094030226, + "grad_norm": 0.7408579587936401, + "learning_rate": 0.000747310623038996, + "loss": 0.5076, + "step": 9134 + }, + { + "epoch": 0.5103779646339078, + "grad_norm": 0.6036206483840942, + "learning_rate": 0.0007472826086956522, + "loss": 0.4293, + "step": 9135 + }, + { + "epoch": 0.5104338352375898, + "grad_norm": 0.3539084196090698, + "learning_rate": 0.0007472545943523084, + "loss": 0.4687, + "step": 9136 + }, + { + "epoch": 0.5104897058412716, + "grad_norm": 0.507975697517395, + "learning_rate": 0.0007472265800089646, + "loss": 0.4734, + "step": 9137 + }, + { + "epoch": 0.5105455764449535, + "grad_norm": 0.7144582867622375, + "learning_rate": 0.0007471985656656208, + "loss": 0.5327, + "step": 9138 + }, + { + "epoch": 0.5106014470486354, + "grad_norm": 0.3763120770454407, + "learning_rate": 0.000747170551322277, + "loss": 0.3965, + "step": 9139 + }, + { + "epoch": 0.5106573176523173, + "grad_norm": 0.6727028489112854, + "learning_rate": 0.0007471425369789332, + "loss": 0.4231, + "step": 9140 + }, + { + "epoch": 0.5107131882559991, + "grad_norm": 0.5797296762466431, + "learning_rate": 0.0007471145226355894, + "loss": 0.586, + "step": 9141 + }, + { + "epoch": 0.5107690588596809, + "grad_norm": 0.3933301866054535, + "learning_rate": 0.0007470865082922456, + "loss": 0.357, + "step": 9142 + }, + { + "epoch": 0.5108249294633629, + "grad_norm": 1.127754807472229, + "learning_rate": 0.000747058493948902, + "loss": 0.4097, + "step": 9143 + }, + { + "epoch": 0.5108808000670447, + "grad_norm": 0.509846031665802, + "learning_rate": 0.000747030479605558, + "loss": 0.5025, + "step": 9144 + }, + { + "epoch": 0.5109366706707266, + "grad_norm": 0.34741249680519104, + "learning_rate": 0.0007470024652622143, + "loss": 0.3492, + "step": 9145 + }, + { + "epoch": 0.5109925412744085, + "grad_norm": 0.9220803380012512, + "learning_rate": 0.0007469744509188705, + "loss": 0.5449, + "step": 9146 + }, + { + "epoch": 0.5110484118780904, + "grad_norm": 0.5026488900184631, + "learning_rate": 0.0007469464365755267, + "loss": 0.3832, + "step": 9147 + }, + { + "epoch": 0.5111042824817722, + "grad_norm": 0.5100843906402588, + "learning_rate": 0.0007469184222321829, + "loss": 0.3684, + "step": 9148 + }, + { + "epoch": 0.511160153085454, + "grad_norm": 0.4366042912006378, + "learning_rate": 0.0007468904078888391, + "loss": 0.4186, + "step": 9149 + }, + { + "epoch": 0.511216023689136, + "grad_norm": 0.603803813457489, + "learning_rate": 0.0007468623935454954, + "loss": 0.4355, + "step": 9150 + }, + { + "epoch": 0.5112718942928178, + "grad_norm": 0.5869777202606201, + "learning_rate": 0.0007468343792021515, + "loss": 0.396, + "step": 9151 + }, + { + "epoch": 0.5113277648964997, + "grad_norm": 0.43160250782966614, + "learning_rate": 0.0007468063648588077, + "loss": 0.4405, + "step": 9152 + }, + { + "epoch": 0.5113836355001816, + "grad_norm": 0.8816807270050049, + "learning_rate": 0.0007467783505154639, + "loss": 0.5423, + "step": 9153 + }, + { + "epoch": 0.5114395061038635, + "grad_norm": 0.42430758476257324, + "learning_rate": 0.0007467503361721202, + "loss": 0.4505, + "step": 9154 + }, + { + "epoch": 0.5114953767075453, + "grad_norm": 1.2792296409606934, + "learning_rate": 0.0007467223218287763, + "loss": 0.4469, + "step": 9155 + }, + { + "epoch": 0.5115512473112273, + "grad_norm": 0.4372537434101105, + "learning_rate": 0.0007466943074854325, + "loss": 0.4192, + "step": 9156 + }, + { + "epoch": 0.5116071179149091, + "grad_norm": 0.5786083340644836, + "learning_rate": 0.0007466662931420888, + "loss": 0.4309, + "step": 9157 + }, + { + "epoch": 0.5116629885185909, + "grad_norm": 1.0649404525756836, + "learning_rate": 0.000746638278798745, + "loss": 0.3692, + "step": 9158 + }, + { + "epoch": 0.5117188591222728, + "grad_norm": 0.38742125034332275, + "learning_rate": 0.0007466102644554011, + "loss": 0.3839, + "step": 9159 + }, + { + "epoch": 0.5117747297259547, + "grad_norm": 0.5804023742675781, + "learning_rate": 0.0007465822501120573, + "loss": 0.4394, + "step": 9160 + }, + { + "epoch": 0.5118306003296366, + "grad_norm": 0.39864376187324524, + "learning_rate": 0.0007465542357687137, + "loss": 0.3469, + "step": 9161 + }, + { + "epoch": 0.5118864709333184, + "grad_norm": 0.599768877029419, + "learning_rate": 0.0007465262214253699, + "loss": 0.3887, + "step": 9162 + }, + { + "epoch": 0.5119423415370004, + "grad_norm": 0.6298130750656128, + "learning_rate": 0.000746498207082026, + "loss": 0.4366, + "step": 9163 + }, + { + "epoch": 0.5119982121406822, + "grad_norm": 1.935418725013733, + "learning_rate": 0.0007464701927386822, + "loss": 0.5189, + "step": 9164 + }, + { + "epoch": 0.512054082744364, + "grad_norm": 0.4836953580379486, + "learning_rate": 0.0007464421783953385, + "loss": 0.537, + "step": 9165 + }, + { + "epoch": 0.5121099533480459, + "grad_norm": 0.5404619574546814, + "learning_rate": 0.0007464141640519946, + "loss": 0.4306, + "step": 9166 + }, + { + "epoch": 0.5121658239517278, + "grad_norm": 0.36676859855651855, + "learning_rate": 0.0007463861497086508, + "loss": 0.4321, + "step": 9167 + }, + { + "epoch": 0.5122216945554097, + "grad_norm": 0.4224381148815155, + "learning_rate": 0.0007463581353653071, + "loss": 0.4434, + "step": 9168 + }, + { + "epoch": 0.5122775651590915, + "grad_norm": 0.5317745804786682, + "learning_rate": 0.0007463301210219633, + "loss": 0.449, + "step": 9169 + }, + { + "epoch": 0.5123334357627735, + "grad_norm": 0.5397883653640747, + "learning_rate": 0.0007463021066786194, + "loss": 0.4625, + "step": 9170 + }, + { + "epoch": 0.5123893063664553, + "grad_norm": 0.5229135751724243, + "learning_rate": 0.0007462740923352756, + "loss": 0.3618, + "step": 9171 + }, + { + "epoch": 0.5124451769701371, + "grad_norm": 0.9107275009155273, + "learning_rate": 0.0007462460779919319, + "loss": 0.4494, + "step": 9172 + }, + { + "epoch": 0.512501047573819, + "grad_norm": 1.1656450033187866, + "learning_rate": 0.0007462180636485881, + "loss": 0.4162, + "step": 9173 + }, + { + "epoch": 0.5125569181775009, + "grad_norm": 9.012630462646484, + "learning_rate": 0.0007461900493052442, + "loss": 0.4307, + "step": 9174 + }, + { + "epoch": 0.5126127887811828, + "grad_norm": 0.5180276036262512, + "learning_rate": 0.0007461620349619005, + "loss": 0.4387, + "step": 9175 + }, + { + "epoch": 0.5126686593848646, + "grad_norm": 0.6101803183555603, + "learning_rate": 0.0007461340206185567, + "loss": 0.6246, + "step": 9176 + }, + { + "epoch": 0.5127245299885466, + "grad_norm": 0.584950864315033, + "learning_rate": 0.000746106006275213, + "loss": 0.416, + "step": 9177 + }, + { + "epoch": 0.5127804005922284, + "grad_norm": 1.2467504739761353, + "learning_rate": 0.000746077991931869, + "loss": 0.4106, + "step": 9178 + }, + { + "epoch": 0.5128362711959102, + "grad_norm": 0.816985547542572, + "learning_rate": 0.0007460499775885254, + "loss": 0.4721, + "step": 9179 + }, + { + "epoch": 0.5128921417995922, + "grad_norm": 0.6043537855148315, + "learning_rate": 0.0007460219632451816, + "loss": 0.3474, + "step": 9180 + }, + { + "epoch": 0.512948012403274, + "grad_norm": 0.41068020462989807, + "learning_rate": 0.0007459939489018378, + "loss": 0.4096, + "step": 9181 + }, + { + "epoch": 0.5130038830069559, + "grad_norm": 1.4595378637313843, + "learning_rate": 0.000745965934558494, + "loss": 0.5126, + "step": 9182 + }, + { + "epoch": 0.5130597536106377, + "grad_norm": 0.5306618213653564, + "learning_rate": 0.0007459379202151502, + "loss": 0.5372, + "step": 9183 + }, + { + "epoch": 0.5131156242143197, + "grad_norm": 0.5676290392875671, + "learning_rate": 0.0007459099058718064, + "loss": 0.477, + "step": 9184 + }, + { + "epoch": 0.5131714948180015, + "grad_norm": 0.4620334208011627, + "learning_rate": 0.0007458818915284626, + "loss": 0.4261, + "step": 9185 + }, + { + "epoch": 0.5132273654216833, + "grad_norm": 0.7002170085906982, + "learning_rate": 0.0007458538771851188, + "loss": 0.4944, + "step": 9186 + }, + { + "epoch": 0.5132832360253653, + "grad_norm": 0.5264235138893127, + "learning_rate": 0.000745825862841775, + "loss": 0.6436, + "step": 9187 + }, + { + "epoch": 0.5133391066290471, + "grad_norm": 0.6905840635299683, + "learning_rate": 0.0007457978484984312, + "loss": 0.4855, + "step": 9188 + }, + { + "epoch": 0.513394977232729, + "grad_norm": 0.4593878984451294, + "learning_rate": 0.0007457698341550874, + "loss": 0.4328, + "step": 9189 + }, + { + "epoch": 0.5134508478364108, + "grad_norm": 0.6013474464416504, + "learning_rate": 0.0007457418198117436, + "loss": 0.4316, + "step": 9190 + }, + { + "epoch": 0.5135067184400928, + "grad_norm": 0.46668121218681335, + "learning_rate": 0.0007457138054683998, + "loss": 0.4591, + "step": 9191 + }, + { + "epoch": 0.5135625890437746, + "grad_norm": 0.39027294516563416, + "learning_rate": 0.000745685791125056, + "loss": 0.3897, + "step": 9192 + }, + { + "epoch": 0.5136184596474564, + "grad_norm": 0.5999705195426941, + "learning_rate": 0.0007456577767817122, + "loss": 0.4153, + "step": 9193 + }, + { + "epoch": 0.5136743302511384, + "grad_norm": 0.5127982497215271, + "learning_rate": 0.0007456297624383684, + "loss": 0.3528, + "step": 9194 + }, + { + "epoch": 0.5137302008548202, + "grad_norm": 0.6416000127792358, + "learning_rate": 0.0007456017480950247, + "loss": 0.6049, + "step": 9195 + }, + { + "epoch": 0.5137860714585021, + "grad_norm": 0.5215746164321899, + "learning_rate": 0.000745573733751681, + "loss": 0.4622, + "step": 9196 + }, + { + "epoch": 0.513841942062184, + "grad_norm": 0.9436953067779541, + "learning_rate": 0.0007455457194083371, + "loss": 0.3908, + "step": 9197 + }, + { + "epoch": 0.5138978126658659, + "grad_norm": 1.5480756759643555, + "learning_rate": 0.0007455177050649933, + "loss": 0.607, + "step": 9198 + }, + { + "epoch": 0.5139536832695477, + "grad_norm": 0.9811853170394897, + "learning_rate": 0.0007454896907216495, + "loss": 0.3769, + "step": 9199 + }, + { + "epoch": 0.5140095538732296, + "grad_norm": 0.6165327429771423, + "learning_rate": 0.0007454616763783058, + "loss": 0.4316, + "step": 9200 + }, + { + "epoch": 0.5140654244769115, + "grad_norm": 0.4890907406806946, + "learning_rate": 0.0007454336620349619, + "loss": 0.4585, + "step": 9201 + }, + { + "epoch": 0.5141212950805933, + "grad_norm": 1.1280238628387451, + "learning_rate": 0.0007454056476916181, + "loss": 0.4557, + "step": 9202 + }, + { + "epoch": 0.5141771656842752, + "grad_norm": 0.5477803945541382, + "learning_rate": 0.0007453776333482743, + "loss": 0.4438, + "step": 9203 + }, + { + "epoch": 0.5142330362879571, + "grad_norm": 0.762059211730957, + "learning_rate": 0.0007453496190049306, + "loss": 0.3713, + "step": 9204 + }, + { + "epoch": 0.514288906891639, + "grad_norm": 0.4498588442802429, + "learning_rate": 0.0007453216046615867, + "loss": 0.3812, + "step": 9205 + }, + { + "epoch": 0.5143447774953208, + "grad_norm": 0.6232873797416687, + "learning_rate": 0.0007452935903182429, + "loss": 0.4105, + "step": 9206 + }, + { + "epoch": 0.5144006480990027, + "grad_norm": 0.9728793501853943, + "learning_rate": 0.0007452655759748992, + "loss": 0.5921, + "step": 9207 + }, + { + "epoch": 0.5144565187026846, + "grad_norm": 0.5305212140083313, + "learning_rate": 0.0007452375616315553, + "loss": 0.4957, + "step": 9208 + }, + { + "epoch": 0.5145123893063664, + "grad_norm": 0.4598933160305023, + "learning_rate": 0.0007452095472882115, + "loss": 0.4816, + "step": 9209 + }, + { + "epoch": 0.5145682599100483, + "grad_norm": 1.8803457021713257, + "learning_rate": 0.0007451815329448677, + "loss": 0.4895, + "step": 9210 + }, + { + "epoch": 0.5146241305137302, + "grad_norm": 0.8470715284347534, + "learning_rate": 0.0007451535186015241, + "loss": 0.543, + "step": 9211 + }, + { + "epoch": 0.5146800011174121, + "grad_norm": 5.696910381317139, + "learning_rate": 0.0007451255042581802, + "loss": 0.4438, + "step": 9212 + }, + { + "epoch": 0.5147358717210939, + "grad_norm": 0.6539350748062134, + "learning_rate": 0.0007450974899148364, + "loss": 0.4107, + "step": 9213 + }, + { + "epoch": 0.5147917423247759, + "grad_norm": 0.666713297367096, + "learning_rate": 0.0007450694755714927, + "loss": 0.4801, + "step": 9214 + }, + { + "epoch": 0.5148476129284577, + "grad_norm": 0.4722334146499634, + "learning_rate": 0.0007450414612281489, + "loss": 0.4288, + "step": 9215 + }, + { + "epoch": 0.5149034835321395, + "grad_norm": 0.503851592540741, + "learning_rate": 0.000745013446884805, + "loss": 0.3712, + "step": 9216 + }, + { + "epoch": 0.5149593541358214, + "grad_norm": 1.993861198425293, + "learning_rate": 0.0007449854325414612, + "loss": 0.5647, + "step": 9217 + }, + { + "epoch": 0.5150152247395033, + "grad_norm": 0.5173690915107727, + "learning_rate": 0.0007449574181981175, + "loss": 0.48, + "step": 9218 + }, + { + "epoch": 0.5150710953431852, + "grad_norm": 5.9558610916137695, + "learning_rate": 0.0007449294038547737, + "loss": 0.606, + "step": 9219 + }, + { + "epoch": 0.515126965946867, + "grad_norm": 1.183558464050293, + "learning_rate": 0.0007449013895114298, + "loss": 0.6385, + "step": 9220 + }, + { + "epoch": 0.515182836550549, + "grad_norm": 0.6979782581329346, + "learning_rate": 0.0007448733751680861, + "loss": 0.4996, + "step": 9221 + }, + { + "epoch": 0.5152387071542308, + "grad_norm": 0.45150768756866455, + "learning_rate": 0.0007448453608247423, + "loss": 0.4667, + "step": 9222 + }, + { + "epoch": 0.5152945777579127, + "grad_norm": 1.747374415397644, + "learning_rate": 0.0007448173464813985, + "loss": 0.5347, + "step": 9223 + }, + { + "epoch": 0.5153504483615945, + "grad_norm": 2.570775270462036, + "learning_rate": 0.0007447893321380546, + "loss": 0.5601, + "step": 9224 + }, + { + "epoch": 0.5154063189652764, + "grad_norm": 0.5129666924476624, + "learning_rate": 0.0007447613177947109, + "loss": 0.3873, + "step": 9225 + }, + { + "epoch": 0.5154621895689583, + "grad_norm": 0.6170926094055176, + "learning_rate": 0.0007447333034513671, + "loss": 0.4218, + "step": 9226 + }, + { + "epoch": 0.5155180601726401, + "grad_norm": 0.5330318212509155, + "learning_rate": 0.0007447052891080233, + "loss": 0.431, + "step": 9227 + }, + { + "epoch": 0.5155739307763221, + "grad_norm": 0.5266491770744324, + "learning_rate": 0.0007446772747646796, + "loss": 0.4894, + "step": 9228 + }, + { + "epoch": 0.5156298013800039, + "grad_norm": 0.7661254405975342, + "learning_rate": 0.0007446492604213358, + "loss": 0.4358, + "step": 9229 + }, + { + "epoch": 0.5156856719836858, + "grad_norm": 0.5561255216598511, + "learning_rate": 0.000744621246077992, + "loss": 0.4408, + "step": 9230 + }, + { + "epoch": 0.5157415425873677, + "grad_norm": 0.594172477722168, + "learning_rate": 0.0007445932317346481, + "loss": 0.4367, + "step": 9231 + }, + { + "epoch": 0.5157974131910495, + "grad_norm": 0.6599130034446716, + "learning_rate": 0.0007445652173913044, + "loss": 0.3951, + "step": 9232 + }, + { + "epoch": 0.5158532837947314, + "grad_norm": 0.378416508436203, + "learning_rate": 0.0007445372030479606, + "loss": 0.3839, + "step": 9233 + }, + { + "epoch": 0.5159091543984132, + "grad_norm": 0.47902482748031616, + "learning_rate": 0.0007445091887046168, + "loss": 0.4301, + "step": 9234 + }, + { + "epoch": 0.5159650250020952, + "grad_norm": 0.4276255667209625, + "learning_rate": 0.0007444811743612729, + "loss": 0.4379, + "step": 9235 + }, + { + "epoch": 0.516020895605777, + "grad_norm": 0.6580808758735657, + "learning_rate": 0.0007444531600179292, + "loss": 0.4839, + "step": 9236 + }, + { + "epoch": 0.5160767662094589, + "grad_norm": 0.6042422652244568, + "learning_rate": 0.0007444251456745854, + "loss": 0.4676, + "step": 9237 + }, + { + "epoch": 0.5161326368131408, + "grad_norm": 0.4504204988479614, + "learning_rate": 0.0007443971313312416, + "loss": 0.3764, + "step": 9238 + }, + { + "epoch": 0.5161885074168227, + "grad_norm": 0.3751535713672638, + "learning_rate": 0.0007443691169878978, + "loss": 0.3481, + "step": 9239 + }, + { + "epoch": 0.5162443780205045, + "grad_norm": 0.4950685501098633, + "learning_rate": 0.000744341102644554, + "loss": 0.4259, + "step": 9240 + }, + { + "epoch": 0.5163002486241863, + "grad_norm": 0.43320131301879883, + "learning_rate": 0.0007443130883012102, + "loss": 0.3796, + "step": 9241 + }, + { + "epoch": 0.5163561192278683, + "grad_norm": 0.625272810459137, + "learning_rate": 0.0007442850739578664, + "loss": 0.4353, + "step": 9242 + }, + { + "epoch": 0.5164119898315501, + "grad_norm": 0.4912653863430023, + "learning_rate": 0.0007442570596145226, + "loss": 0.4101, + "step": 9243 + }, + { + "epoch": 0.516467860435232, + "grad_norm": 0.48046931624412537, + "learning_rate": 0.0007442290452711788, + "loss": 0.5332, + "step": 9244 + }, + { + "epoch": 0.5165237310389139, + "grad_norm": 0.7915756106376648, + "learning_rate": 0.000744201030927835, + "loss": 0.7359, + "step": 9245 + }, + { + "epoch": 0.5165796016425958, + "grad_norm": 0.49785909056663513, + "learning_rate": 0.0007441730165844914, + "loss": 0.4862, + "step": 9246 + }, + { + "epoch": 0.5166354722462776, + "grad_norm": 0.5645042061805725, + "learning_rate": 0.0007441450022411475, + "loss": 0.4472, + "step": 9247 + }, + { + "epoch": 0.5166913428499595, + "grad_norm": 0.47228819131851196, + "learning_rate": 0.0007441169878978037, + "loss": 0.3855, + "step": 9248 + }, + { + "epoch": 0.5167472134536414, + "grad_norm": 0.9623945355415344, + "learning_rate": 0.0007440889735544599, + "loss": 0.4511, + "step": 9249 + }, + { + "epoch": 0.5168030840573232, + "grad_norm": 0.636593759059906, + "learning_rate": 0.0007440609592111162, + "loss": 0.7466, + "step": 9250 + }, + { + "epoch": 0.5168589546610051, + "grad_norm": 0.5028451085090637, + "learning_rate": 0.0007440329448677723, + "loss": 0.3422, + "step": 9251 + }, + { + "epoch": 0.516914825264687, + "grad_norm": 0.6975560784339905, + "learning_rate": 0.0007440049305244285, + "loss": 0.3779, + "step": 9252 + }, + { + "epoch": 0.5169706958683689, + "grad_norm": 1.085620641708374, + "learning_rate": 0.0007439769161810848, + "loss": 0.423, + "step": 9253 + }, + { + "epoch": 0.5170265664720507, + "grad_norm": 0.7073306441307068, + "learning_rate": 0.0007439489018377409, + "loss": 0.4295, + "step": 9254 + }, + { + "epoch": 0.5170824370757326, + "grad_norm": 0.7864713668823242, + "learning_rate": 0.0007439208874943971, + "loss": 0.4474, + "step": 9255 + }, + { + "epoch": 0.5171383076794145, + "grad_norm": 0.48626649379730225, + "learning_rate": 0.0007438928731510533, + "loss": 0.5252, + "step": 9256 + }, + { + "epoch": 0.5171941782830963, + "grad_norm": 0.46149757504463196, + "learning_rate": 0.0007438648588077096, + "loss": 0.4479, + "step": 9257 + }, + { + "epoch": 0.5172500488867782, + "grad_norm": 0.4213576912879944, + "learning_rate": 0.0007438368444643657, + "loss": 0.4715, + "step": 9258 + }, + { + "epoch": 0.5173059194904601, + "grad_norm": 0.4125516712665558, + "learning_rate": 0.0007438088301210219, + "loss": 0.4022, + "step": 9259 + }, + { + "epoch": 0.517361790094142, + "grad_norm": 0.7152969241142273, + "learning_rate": 0.0007437808157776782, + "loss": 0.4607, + "step": 9260 + }, + { + "epoch": 0.5174176606978238, + "grad_norm": 0.872535228729248, + "learning_rate": 0.0007437528014343345, + "loss": 0.4496, + "step": 9261 + }, + { + "epoch": 0.5174735313015058, + "grad_norm": 0.4538310468196869, + "learning_rate": 0.0007437247870909906, + "loss": 0.405, + "step": 9262 + }, + { + "epoch": 0.5175294019051876, + "grad_norm": 0.742912769317627, + "learning_rate": 0.0007436967727476468, + "loss": 0.3815, + "step": 9263 + }, + { + "epoch": 0.5175852725088694, + "grad_norm": 0.4003710150718689, + "learning_rate": 0.0007436687584043031, + "loss": 0.5519, + "step": 9264 + }, + { + "epoch": 0.5176411431125514, + "grad_norm": 0.37874332070350647, + "learning_rate": 0.0007436407440609593, + "loss": 0.3902, + "step": 9265 + }, + { + "epoch": 0.5176970137162332, + "grad_norm": 0.7697202563285828, + "learning_rate": 0.0007436127297176154, + "loss": 0.4317, + "step": 9266 + }, + { + "epoch": 0.5177528843199151, + "grad_norm": 0.41875508427619934, + "learning_rate": 0.0007435847153742717, + "loss": 0.4239, + "step": 9267 + }, + { + "epoch": 0.5178087549235969, + "grad_norm": 0.6377653479576111, + "learning_rate": 0.0007435567010309279, + "loss": 0.3933, + "step": 9268 + }, + { + "epoch": 0.5178646255272789, + "grad_norm": 0.5031879544258118, + "learning_rate": 0.0007435286866875841, + "loss": 0.4903, + "step": 9269 + }, + { + "epoch": 0.5179204961309607, + "grad_norm": 0.4192863702774048, + "learning_rate": 0.0007435006723442402, + "loss": 0.5729, + "step": 9270 + }, + { + "epoch": 0.5179763667346425, + "grad_norm": 0.5134372711181641, + "learning_rate": 0.0007434726580008965, + "loss": 0.4483, + "step": 9271 + }, + { + "epoch": 0.5180322373383245, + "grad_norm": 0.49694207310676575, + "learning_rate": 0.0007434446436575527, + "loss": 0.4388, + "step": 9272 + }, + { + "epoch": 0.5180881079420063, + "grad_norm": 2.0856289863586426, + "learning_rate": 0.0007434166293142088, + "loss": 0.3944, + "step": 9273 + }, + { + "epoch": 0.5181439785456882, + "grad_norm": 4.562323093414307, + "learning_rate": 0.000743388614970865, + "loss": 0.5452, + "step": 9274 + }, + { + "epoch": 0.51819984914937, + "grad_norm": 2.019547700881958, + "learning_rate": 0.0007433606006275213, + "loss": 0.5405, + "step": 9275 + }, + { + "epoch": 0.518255719753052, + "grad_norm": 0.5072323679924011, + "learning_rate": 0.0007433325862841775, + "loss": 0.4117, + "step": 9276 + }, + { + "epoch": 0.5183115903567338, + "grad_norm": 0.4326562285423279, + "learning_rate": 0.0007433045719408336, + "loss": 0.396, + "step": 9277 + }, + { + "epoch": 0.5183674609604156, + "grad_norm": 0.6768496036529541, + "learning_rate": 0.00074327655759749, + "loss": 0.551, + "step": 9278 + }, + { + "epoch": 0.5184233315640976, + "grad_norm": 0.6069522500038147, + "learning_rate": 0.0007432485432541462, + "loss": 0.5494, + "step": 9279 + }, + { + "epoch": 0.5184792021677794, + "grad_norm": 0.4644950330257416, + "learning_rate": 0.0007432205289108024, + "loss": 0.341, + "step": 9280 + }, + { + "epoch": 0.5185350727714613, + "grad_norm": 0.5338155627250671, + "learning_rate": 0.0007431925145674585, + "loss": 0.4418, + "step": 9281 + }, + { + "epoch": 0.5185909433751432, + "grad_norm": 6.268515586853027, + "learning_rate": 0.0007431645002241148, + "loss": 0.4543, + "step": 9282 + }, + { + "epoch": 0.5186468139788251, + "grad_norm": 0.5553454160690308, + "learning_rate": 0.000743136485880771, + "loss": 0.4521, + "step": 9283 + }, + { + "epoch": 0.5187026845825069, + "grad_norm": 0.6292113661766052, + "learning_rate": 0.0007431084715374272, + "loss": 0.4096, + "step": 9284 + }, + { + "epoch": 0.5187585551861887, + "grad_norm": 1.5575902462005615, + "learning_rate": 0.0007430804571940834, + "loss": 0.5016, + "step": 9285 + }, + { + "epoch": 0.5188144257898707, + "grad_norm": 0.5681557059288025, + "learning_rate": 0.0007430524428507396, + "loss": 0.3605, + "step": 9286 + }, + { + "epoch": 0.5188702963935525, + "grad_norm": 0.3998614549636841, + "learning_rate": 0.0007430244285073958, + "loss": 0.3833, + "step": 9287 + }, + { + "epoch": 0.5189261669972344, + "grad_norm": 0.4541720747947693, + "learning_rate": 0.000742996414164052, + "loss": 0.37, + "step": 9288 + }, + { + "epoch": 0.5189820376009163, + "grad_norm": 0.43403342366218567, + "learning_rate": 0.0007429683998207082, + "loss": 0.397, + "step": 9289 + }, + { + "epoch": 0.5190379082045982, + "grad_norm": 0.47690102458000183, + "learning_rate": 0.0007429403854773644, + "loss": 0.4028, + "step": 9290 + }, + { + "epoch": 0.51909377880828, + "grad_norm": 0.6916494965553284, + "learning_rate": 0.0007429123711340206, + "loss": 0.4107, + "step": 9291 + }, + { + "epoch": 0.5191496494119618, + "grad_norm": 0.7887623310089111, + "learning_rate": 0.0007428843567906769, + "loss": 0.5869, + "step": 9292 + }, + { + "epoch": 0.5192055200156438, + "grad_norm": 0.5489433407783508, + "learning_rate": 0.000742856342447333, + "loss": 0.4642, + "step": 9293 + }, + { + "epoch": 0.5192613906193256, + "grad_norm": 0.5217433571815491, + "learning_rate": 0.0007428283281039892, + "loss": 0.4526, + "step": 9294 + }, + { + "epoch": 0.5193172612230075, + "grad_norm": 0.6170317530632019, + "learning_rate": 0.0007428003137606455, + "loss": 0.3899, + "step": 9295 + }, + { + "epoch": 0.5193731318266894, + "grad_norm": 1.5950547456741333, + "learning_rate": 0.0007427722994173017, + "loss": 0.4, + "step": 9296 + }, + { + "epoch": 0.5194290024303713, + "grad_norm": 0.8839400410652161, + "learning_rate": 0.0007427442850739579, + "loss": 0.4263, + "step": 9297 + }, + { + "epoch": 0.5194848730340531, + "grad_norm": 0.5274655818939209, + "learning_rate": 0.0007427162707306141, + "loss": 0.4182, + "step": 9298 + }, + { + "epoch": 0.5195407436377351, + "grad_norm": 1.38511323928833, + "learning_rate": 0.0007426882563872704, + "loss": 0.3956, + "step": 9299 + }, + { + "epoch": 0.5195966142414169, + "grad_norm": 0.7631036043167114, + "learning_rate": 0.0007426602420439265, + "loss": 0.4983, + "step": 9300 + }, + { + "epoch": 0.5196524848450987, + "grad_norm": 0.5770743489265442, + "learning_rate": 0.0007426322277005827, + "loss": 0.4656, + "step": 9301 + }, + { + "epoch": 0.5197083554487806, + "grad_norm": 0.5458757877349854, + "learning_rate": 0.0007426042133572389, + "loss": 0.3715, + "step": 9302 + }, + { + "epoch": 0.5197642260524625, + "grad_norm": 0.5163546204566956, + "learning_rate": 0.0007425761990138952, + "loss": 0.3564, + "step": 9303 + }, + { + "epoch": 0.5198200966561444, + "grad_norm": 1.5534257888793945, + "learning_rate": 0.0007425481846705513, + "loss": 0.5305, + "step": 9304 + }, + { + "epoch": 0.5198759672598262, + "grad_norm": 0.3339133858680725, + "learning_rate": 0.0007425201703272075, + "loss": 0.398, + "step": 9305 + }, + { + "epoch": 0.5199318378635082, + "grad_norm": 0.5200069546699524, + "learning_rate": 0.0007424921559838638, + "loss": 0.5196, + "step": 9306 + }, + { + "epoch": 0.51998770846719, + "grad_norm": 0.948495090007782, + "learning_rate": 0.00074246414164052, + "loss": 0.4253, + "step": 9307 + }, + { + "epoch": 0.5200435790708718, + "grad_norm": 0.643720805644989, + "learning_rate": 0.0007424361272971761, + "loss": 0.5947, + "step": 9308 + }, + { + "epoch": 0.5200994496745537, + "grad_norm": 0.41436949372291565, + "learning_rate": 0.0007424081129538323, + "loss": 0.3257, + "step": 9309 + }, + { + "epoch": 0.5201553202782356, + "grad_norm": 1.181365728378296, + "learning_rate": 0.0007423800986104886, + "loss": 0.4767, + "step": 9310 + }, + { + "epoch": 0.5202111908819175, + "grad_norm": 0.6265082955360413, + "learning_rate": 0.0007423520842671449, + "loss": 0.4559, + "step": 9311 + }, + { + "epoch": 0.5202670614855993, + "grad_norm": 0.7281716465950012, + "learning_rate": 0.000742324069923801, + "loss": 0.4883, + "step": 9312 + }, + { + "epoch": 0.5203229320892813, + "grad_norm": 0.4987582862377167, + "learning_rate": 0.0007422960555804572, + "loss": 0.3828, + "step": 9313 + }, + { + "epoch": 0.5203788026929631, + "grad_norm": 0.5249767899513245, + "learning_rate": 0.0007422680412371135, + "loss": 0.4524, + "step": 9314 + }, + { + "epoch": 0.520434673296645, + "grad_norm": 0.5780797600746155, + "learning_rate": 0.0007422400268937696, + "loss": 0.4678, + "step": 9315 + }, + { + "epoch": 0.5204905439003269, + "grad_norm": 0.557775616645813, + "learning_rate": 0.0007422120125504258, + "loss": 0.4611, + "step": 9316 + }, + { + "epoch": 0.5205464145040087, + "grad_norm": 0.6490108966827393, + "learning_rate": 0.0007421839982070821, + "loss": 0.509, + "step": 9317 + }, + { + "epoch": 0.5206022851076906, + "grad_norm": 0.4269646406173706, + "learning_rate": 0.0007421559838637383, + "loss": 0.4656, + "step": 9318 + }, + { + "epoch": 0.5206581557113724, + "grad_norm": 0.6968739032745361, + "learning_rate": 0.0007421279695203944, + "loss": 0.3924, + "step": 9319 + }, + { + "epoch": 0.5207140263150544, + "grad_norm": 0.7367920279502869, + "learning_rate": 0.0007420999551770506, + "loss": 0.463, + "step": 9320 + }, + { + "epoch": 0.5207698969187362, + "grad_norm": 0.4890534579753876, + "learning_rate": 0.0007420719408337069, + "loss": 0.4558, + "step": 9321 + }, + { + "epoch": 0.520825767522418, + "grad_norm": 0.5631798505783081, + "learning_rate": 0.0007420439264903631, + "loss": 0.51, + "step": 9322 + }, + { + "epoch": 0.5208816381261, + "grad_norm": 0.4977385997772217, + "learning_rate": 0.0007420159121470192, + "loss": 0.4166, + "step": 9323 + }, + { + "epoch": 0.5209375087297818, + "grad_norm": 0.4634439945220947, + "learning_rate": 0.0007419878978036755, + "loss": 0.4326, + "step": 9324 + }, + { + "epoch": 0.5209933793334637, + "grad_norm": 0.5203236937522888, + "learning_rate": 0.0007419598834603317, + "loss": 0.4494, + "step": 9325 + }, + { + "epoch": 0.5210492499371455, + "grad_norm": 0.9395164847373962, + "learning_rate": 0.0007419318691169879, + "loss": 0.5321, + "step": 9326 + }, + { + "epoch": 0.5211051205408275, + "grad_norm": 0.5615613460540771, + "learning_rate": 0.000741903854773644, + "loss": 0.4218, + "step": 9327 + }, + { + "epoch": 0.5211609911445093, + "grad_norm": 0.4155277609825134, + "learning_rate": 0.0007418758404303004, + "loss": 0.4765, + "step": 9328 + }, + { + "epoch": 0.5212168617481912, + "grad_norm": 0.40703529119491577, + "learning_rate": 0.0007418478260869566, + "loss": 0.4217, + "step": 9329 + }, + { + "epoch": 0.5212727323518731, + "grad_norm": 0.6499472856521606, + "learning_rate": 0.0007418198117436128, + "loss": 0.4059, + "step": 9330 + }, + { + "epoch": 0.5213286029555549, + "grad_norm": 0.8443948030471802, + "learning_rate": 0.000741791797400269, + "loss": 0.5239, + "step": 9331 + }, + { + "epoch": 0.5213844735592368, + "grad_norm": 1.664968729019165, + "learning_rate": 0.0007417637830569252, + "loss": 0.4394, + "step": 9332 + }, + { + "epoch": 0.5214403441629187, + "grad_norm": 0.49589231610298157, + "learning_rate": 0.0007417357687135814, + "loss": 0.4634, + "step": 9333 + }, + { + "epoch": 0.5214962147666006, + "grad_norm": 0.4229677617549896, + "learning_rate": 0.0007417077543702376, + "loss": 0.3699, + "step": 9334 + }, + { + "epoch": 0.5215520853702824, + "grad_norm": 1.1321911811828613, + "learning_rate": 0.0007416797400268938, + "loss": 0.4931, + "step": 9335 + }, + { + "epoch": 0.5216079559739643, + "grad_norm": 0.9047557711601257, + "learning_rate": 0.00074165172568355, + "loss": 0.4473, + "step": 9336 + }, + { + "epoch": 0.5216638265776462, + "grad_norm": 0.636295735836029, + "learning_rate": 0.0007416237113402062, + "loss": 0.5067, + "step": 9337 + }, + { + "epoch": 0.521719697181328, + "grad_norm": 1.1422438621520996, + "learning_rate": 0.0007415956969968624, + "loss": 0.4782, + "step": 9338 + }, + { + "epoch": 0.5217755677850099, + "grad_norm": 0.8750447034835815, + "learning_rate": 0.0007415676826535186, + "loss": 0.609, + "step": 9339 + }, + { + "epoch": 0.5218314383886918, + "grad_norm": 0.5644298791885376, + "learning_rate": 0.0007415396683101748, + "loss": 0.4408, + "step": 9340 + }, + { + "epoch": 0.5218873089923737, + "grad_norm": 0.7472765445709229, + "learning_rate": 0.000741511653966831, + "loss": 0.4805, + "step": 9341 + }, + { + "epoch": 0.5219431795960555, + "grad_norm": 0.46158286929130554, + "learning_rate": 0.0007414836396234872, + "loss": 0.4422, + "step": 9342 + }, + { + "epoch": 0.5219990501997374, + "grad_norm": 0.428722083568573, + "learning_rate": 0.0007414556252801434, + "loss": 0.4232, + "step": 9343 + }, + { + "epoch": 0.5220549208034193, + "grad_norm": 0.43222540616989136, + "learning_rate": 0.0007414276109367996, + "loss": 0.4809, + "step": 9344 + }, + { + "epoch": 0.5221107914071011, + "grad_norm": 0.4993489682674408, + "learning_rate": 0.0007413995965934559, + "loss": 0.4163, + "step": 9345 + }, + { + "epoch": 0.522166662010783, + "grad_norm": 0.9876783490180969, + "learning_rate": 0.0007413715822501121, + "loss": 0.395, + "step": 9346 + }, + { + "epoch": 0.5222225326144649, + "grad_norm": 0.6898161768913269, + "learning_rate": 0.0007413435679067683, + "loss": 0.4924, + "step": 9347 + }, + { + "epoch": 0.5222784032181468, + "grad_norm": 0.5186506509780884, + "learning_rate": 0.0007413155535634245, + "loss": 0.436, + "step": 9348 + }, + { + "epoch": 0.5223342738218286, + "grad_norm": 0.5028510689735413, + "learning_rate": 0.0007412875392200808, + "loss": 0.5265, + "step": 9349 + }, + { + "epoch": 0.5223901444255106, + "grad_norm": 0.45152124762535095, + "learning_rate": 0.0007412595248767369, + "loss": 0.3786, + "step": 9350 + }, + { + "epoch": 0.5224460150291924, + "grad_norm": 0.8115715384483337, + "learning_rate": 0.0007412315105333931, + "loss": 0.3764, + "step": 9351 + }, + { + "epoch": 0.5225018856328743, + "grad_norm": 0.5280886888504028, + "learning_rate": 0.0007412034961900493, + "loss": 0.433, + "step": 9352 + }, + { + "epoch": 0.5225577562365561, + "grad_norm": 1.9092200994491577, + "learning_rate": 0.0007411754818467056, + "loss": 0.38, + "step": 9353 + }, + { + "epoch": 0.522613626840238, + "grad_norm": 0.7007749080657959, + "learning_rate": 0.0007411474675033617, + "loss": 0.52, + "step": 9354 + }, + { + "epoch": 0.5226694974439199, + "grad_norm": 1.24805748462677, + "learning_rate": 0.0007411194531600179, + "loss": 0.3653, + "step": 9355 + }, + { + "epoch": 0.5227253680476017, + "grad_norm": 0.5950300097465515, + "learning_rate": 0.0007410914388166742, + "loss": 0.4678, + "step": 9356 + }, + { + "epoch": 0.5227812386512837, + "grad_norm": 0.5235185623168945, + "learning_rate": 0.0007410634244733303, + "loss": 0.4617, + "step": 9357 + }, + { + "epoch": 0.5228371092549655, + "grad_norm": 0.44873523712158203, + "learning_rate": 0.0007410354101299865, + "loss": 0.4442, + "step": 9358 + }, + { + "epoch": 0.5228929798586474, + "grad_norm": 2.4830918312072754, + "learning_rate": 0.0007410073957866427, + "loss": 0.518, + "step": 9359 + }, + { + "epoch": 0.5229488504623292, + "grad_norm": 0.5538445711135864, + "learning_rate": 0.000740979381443299, + "loss": 0.3922, + "step": 9360 + }, + { + "epoch": 0.5230047210660111, + "grad_norm": 0.511539876461029, + "learning_rate": 0.0007409513670999551, + "loss": 0.4468, + "step": 9361 + }, + { + "epoch": 0.523060591669693, + "grad_norm": 0.49246132373809814, + "learning_rate": 0.0007409233527566114, + "loss": 0.5232, + "step": 9362 + }, + { + "epoch": 0.5231164622733748, + "grad_norm": 0.5165978670120239, + "learning_rate": 0.0007408953384132677, + "loss": 0.4218, + "step": 9363 + }, + { + "epoch": 0.5231723328770568, + "grad_norm": 0.5026726126670837, + "learning_rate": 0.0007408673240699239, + "loss": 0.4882, + "step": 9364 + }, + { + "epoch": 0.5232282034807386, + "grad_norm": 0.4965970814228058, + "learning_rate": 0.00074083930972658, + "loss": 0.4226, + "step": 9365 + }, + { + "epoch": 0.5232840740844205, + "grad_norm": 0.8533666729927063, + "learning_rate": 0.0007408112953832362, + "loss": 0.4893, + "step": 9366 + }, + { + "epoch": 0.5233399446881024, + "grad_norm": 2.0836799144744873, + "learning_rate": 0.0007407832810398925, + "loss": 0.4068, + "step": 9367 + }, + { + "epoch": 0.5233958152917842, + "grad_norm": 0.40958452224731445, + "learning_rate": 0.0007407552666965487, + "loss": 0.4585, + "step": 9368 + }, + { + "epoch": 0.5234516858954661, + "grad_norm": 0.5068880915641785, + "learning_rate": 0.0007407272523532048, + "loss": 0.4754, + "step": 9369 + }, + { + "epoch": 0.5235075564991479, + "grad_norm": 0.4894161820411682, + "learning_rate": 0.0007406992380098611, + "loss": 0.4189, + "step": 9370 + }, + { + "epoch": 0.5235634271028299, + "grad_norm": 1.0890203714370728, + "learning_rate": 0.0007406712236665173, + "loss": 0.3852, + "step": 9371 + }, + { + "epoch": 0.5236192977065117, + "grad_norm": 6.822014331817627, + "learning_rate": 0.0007406432093231735, + "loss": 0.4657, + "step": 9372 + }, + { + "epoch": 0.5236751683101936, + "grad_norm": 4.126667499542236, + "learning_rate": 0.0007406151949798296, + "loss": 0.4507, + "step": 9373 + }, + { + "epoch": 0.5237310389138755, + "grad_norm": 0.5409882664680481, + "learning_rate": 0.0007405871806364859, + "loss": 0.4311, + "step": 9374 + }, + { + "epoch": 0.5237869095175574, + "grad_norm": 0.5983648300170898, + "learning_rate": 0.0007405591662931421, + "loss": 0.6234, + "step": 9375 + }, + { + "epoch": 0.5238427801212392, + "grad_norm": 1.401875376701355, + "learning_rate": 0.0007405311519497983, + "loss": 0.441, + "step": 9376 + }, + { + "epoch": 0.523898650724921, + "grad_norm": 0.6748141646385193, + "learning_rate": 0.0007405031376064544, + "loss": 0.4322, + "step": 9377 + }, + { + "epoch": 0.523954521328603, + "grad_norm": 0.3625546991825104, + "learning_rate": 0.0007404751232631108, + "loss": 0.3814, + "step": 9378 + }, + { + "epoch": 0.5240103919322848, + "grad_norm": 0.6215754747390747, + "learning_rate": 0.000740447108919767, + "loss": 0.5834, + "step": 9379 + }, + { + "epoch": 0.5240662625359667, + "grad_norm": 14.170745849609375, + "learning_rate": 0.0007404190945764231, + "loss": 0.456, + "step": 9380 + }, + { + "epoch": 0.5241221331396486, + "grad_norm": 0.717950165271759, + "learning_rate": 0.0007403910802330794, + "loss": 0.3945, + "step": 9381 + }, + { + "epoch": 0.5241780037433305, + "grad_norm": 0.5516665577888489, + "learning_rate": 0.0007403630658897356, + "loss": 0.4212, + "step": 9382 + }, + { + "epoch": 0.5242338743470123, + "grad_norm": 0.4441826045513153, + "learning_rate": 0.0007403350515463918, + "loss": 0.353, + "step": 9383 + }, + { + "epoch": 0.5242897449506942, + "grad_norm": 0.5165998935699463, + "learning_rate": 0.0007403070372030479, + "loss": 0.4447, + "step": 9384 + }, + { + "epoch": 0.5243456155543761, + "grad_norm": 0.47876980900764465, + "learning_rate": 0.0007402790228597042, + "loss": 0.552, + "step": 9385 + }, + { + "epoch": 0.5244014861580579, + "grad_norm": 0.495816707611084, + "learning_rate": 0.0007402510085163604, + "loss": 0.4287, + "step": 9386 + }, + { + "epoch": 0.5244573567617398, + "grad_norm": 0.37802961468696594, + "learning_rate": 0.0007402229941730166, + "loss": 0.3716, + "step": 9387 + }, + { + "epoch": 0.5245132273654217, + "grad_norm": 0.5903489589691162, + "learning_rate": 0.0007401949798296728, + "loss": 0.428, + "step": 9388 + }, + { + "epoch": 0.5245690979691036, + "grad_norm": 0.589214026927948, + "learning_rate": 0.000740166965486329, + "loss": 0.4568, + "step": 9389 + }, + { + "epoch": 0.5246249685727854, + "grad_norm": 2.567993640899658, + "learning_rate": 0.0007401389511429852, + "loss": 0.4681, + "step": 9390 + }, + { + "epoch": 0.5246808391764674, + "grad_norm": 0.3737889528274536, + "learning_rate": 0.0007401109367996414, + "loss": 0.4233, + "step": 9391 + }, + { + "epoch": 0.5247367097801492, + "grad_norm": 0.40438681840896606, + "learning_rate": 0.0007400829224562976, + "loss": 0.4179, + "step": 9392 + }, + { + "epoch": 0.524792580383831, + "grad_norm": 0.5059916377067566, + "learning_rate": 0.0007400549081129538, + "loss": 0.4301, + "step": 9393 + }, + { + "epoch": 0.5248484509875129, + "grad_norm": 0.4639037251472473, + "learning_rate": 0.00074002689376961, + "loss": 0.5587, + "step": 9394 + }, + { + "epoch": 0.5249043215911948, + "grad_norm": 0.4327906370162964, + "learning_rate": 0.0007399988794262664, + "loss": 0.4007, + "step": 9395 + }, + { + "epoch": 0.5249601921948767, + "grad_norm": 0.4765983521938324, + "learning_rate": 0.0007399708650829225, + "loss": 0.5818, + "step": 9396 + }, + { + "epoch": 0.5250160627985585, + "grad_norm": 1.5188112258911133, + "learning_rate": 0.0007399428507395787, + "loss": 0.4228, + "step": 9397 + }, + { + "epoch": 0.5250719334022405, + "grad_norm": 1.450356364250183, + "learning_rate": 0.0007399148363962349, + "loss": 0.4627, + "step": 9398 + }, + { + "epoch": 0.5251278040059223, + "grad_norm": 0.5601702928543091, + "learning_rate": 0.0007398868220528912, + "loss": 0.4862, + "step": 9399 + }, + { + "epoch": 0.5251836746096041, + "grad_norm": 0.6950417757034302, + "learning_rate": 0.0007398588077095473, + "loss": 0.4324, + "step": 9400 + }, + { + "epoch": 0.525239545213286, + "grad_norm": 0.9604664444923401, + "learning_rate": 0.0007398307933662035, + "loss": 0.4771, + "step": 9401 + }, + { + "epoch": 0.5252954158169679, + "grad_norm": 0.7345626950263977, + "learning_rate": 0.0007398027790228598, + "loss": 0.4276, + "step": 9402 + }, + { + "epoch": 0.5253512864206498, + "grad_norm": 0.7002467513084412, + "learning_rate": 0.0007397747646795159, + "loss": 0.5476, + "step": 9403 + }, + { + "epoch": 0.5254071570243316, + "grad_norm": 0.4793983995914459, + "learning_rate": 0.0007397467503361721, + "loss": 0.5556, + "step": 9404 + }, + { + "epoch": 0.5254630276280136, + "grad_norm": 0.5861102938652039, + "learning_rate": 0.0007397187359928283, + "loss": 0.4961, + "step": 9405 + }, + { + "epoch": 0.5255188982316954, + "grad_norm": 0.5279638171195984, + "learning_rate": 0.0007396907216494846, + "loss": 0.3887, + "step": 9406 + }, + { + "epoch": 0.5255747688353772, + "grad_norm": 0.6432577967643738, + "learning_rate": 0.0007396627073061407, + "loss": 0.5567, + "step": 9407 + }, + { + "epoch": 0.5256306394390592, + "grad_norm": 0.5580750107765198, + "learning_rate": 0.0007396346929627969, + "loss": 0.4208, + "step": 9408 + }, + { + "epoch": 0.525686510042741, + "grad_norm": 0.6874804496765137, + "learning_rate": 0.0007396066786194532, + "loss": 0.4107, + "step": 9409 + }, + { + "epoch": 0.5257423806464229, + "grad_norm": 0.5909602046012878, + "learning_rate": 0.0007395786642761094, + "loss": 0.3771, + "step": 9410 + }, + { + "epoch": 0.5257982512501047, + "grad_norm": 0.896121084690094, + "learning_rate": 0.0007395506499327655, + "loss": 0.4771, + "step": 9411 + }, + { + "epoch": 0.5258541218537867, + "grad_norm": 0.47372305393218994, + "learning_rate": 0.0007395226355894217, + "loss": 0.4583, + "step": 9412 + }, + { + "epoch": 0.5259099924574685, + "grad_norm": 2.7406415939331055, + "learning_rate": 0.0007394946212460781, + "loss": 0.4138, + "step": 9413 + }, + { + "epoch": 0.5259658630611503, + "grad_norm": 0.5190346240997314, + "learning_rate": 0.0007394666069027343, + "loss": 0.4731, + "step": 9414 + }, + { + "epoch": 0.5260217336648323, + "grad_norm": 0.4361165463924408, + "learning_rate": 0.0007394385925593904, + "loss": 0.4125, + "step": 9415 + }, + { + "epoch": 0.5260776042685141, + "grad_norm": 23.199901580810547, + "learning_rate": 0.0007394105782160466, + "loss": 0.5264, + "step": 9416 + }, + { + "epoch": 0.526133474872196, + "grad_norm": 0.47927114367485046, + "learning_rate": 0.0007393825638727029, + "loss": 0.5127, + "step": 9417 + }, + { + "epoch": 0.5261893454758778, + "grad_norm": 0.6060893535614014, + "learning_rate": 0.0007393545495293591, + "loss": 0.5041, + "step": 9418 + }, + { + "epoch": 0.5262452160795598, + "grad_norm": 0.5497305989265442, + "learning_rate": 0.0007393265351860152, + "loss": 0.3396, + "step": 9419 + }, + { + "epoch": 0.5263010866832416, + "grad_norm": 0.38399648666381836, + "learning_rate": 0.0007392985208426715, + "loss": 0.4133, + "step": 9420 + }, + { + "epoch": 0.5263569572869234, + "grad_norm": 0.6668453216552734, + "learning_rate": 0.0007392705064993277, + "loss": 0.4425, + "step": 9421 + }, + { + "epoch": 0.5264128278906054, + "grad_norm": 0.5116440653800964, + "learning_rate": 0.0007392424921559838, + "loss": 0.5468, + "step": 9422 + }, + { + "epoch": 0.5264686984942872, + "grad_norm": 0.47654691338539124, + "learning_rate": 0.00073921447781264, + "loss": 0.4218, + "step": 9423 + }, + { + "epoch": 0.5265245690979691, + "grad_norm": 0.42173099517822266, + "learning_rate": 0.0007391864634692963, + "loss": 0.5557, + "step": 9424 + }, + { + "epoch": 0.526580439701651, + "grad_norm": 0.48297858238220215, + "learning_rate": 0.0007391584491259525, + "loss": 0.404, + "step": 9425 + }, + { + "epoch": 0.5266363103053329, + "grad_norm": 0.5308840274810791, + "learning_rate": 0.0007391304347826086, + "loss": 0.3655, + "step": 9426 + }, + { + "epoch": 0.5266921809090147, + "grad_norm": 0.4613935649394989, + "learning_rate": 0.0007391024204392649, + "loss": 0.4903, + "step": 9427 + }, + { + "epoch": 0.5267480515126965, + "grad_norm": 2.308619260787964, + "learning_rate": 0.0007390744060959212, + "loss": 0.5059, + "step": 9428 + }, + { + "epoch": 0.5268039221163785, + "grad_norm": 0.5871858596801758, + "learning_rate": 0.0007390463917525774, + "loss": 0.4826, + "step": 9429 + }, + { + "epoch": 0.5268597927200603, + "grad_norm": 0.92790687084198, + "learning_rate": 0.0007390183774092335, + "loss": 0.3813, + "step": 9430 + }, + { + "epoch": 0.5269156633237422, + "grad_norm": 1.0030665397644043, + "learning_rate": 0.0007389903630658898, + "loss": 0.5701, + "step": 9431 + }, + { + "epoch": 0.5269715339274241, + "grad_norm": 0.6908155083656311, + "learning_rate": 0.000738962348722546, + "loss": 0.5692, + "step": 9432 + }, + { + "epoch": 0.527027404531106, + "grad_norm": 0.5540923476219177, + "learning_rate": 0.0007389343343792022, + "loss": 0.5592, + "step": 9433 + }, + { + "epoch": 0.5270832751347878, + "grad_norm": 0.41123712062835693, + "learning_rate": 0.0007389063200358584, + "loss": 0.4555, + "step": 9434 + }, + { + "epoch": 0.5271391457384696, + "grad_norm": 0.4622998833656311, + "learning_rate": 0.0007388783056925146, + "loss": 0.4088, + "step": 9435 + }, + { + "epoch": 0.5271950163421516, + "grad_norm": 0.5787147879600525, + "learning_rate": 0.0007388502913491708, + "loss": 0.5269, + "step": 9436 + }, + { + "epoch": 0.5272508869458334, + "grad_norm": 5.317632675170898, + "learning_rate": 0.000738822277005827, + "loss": 0.5952, + "step": 9437 + }, + { + "epoch": 0.5273067575495153, + "grad_norm": 0.5284690856933594, + "learning_rate": 0.0007387942626624832, + "loss": 0.3768, + "step": 9438 + }, + { + "epoch": 0.5273626281531972, + "grad_norm": 0.5017854571342468, + "learning_rate": 0.0007387662483191394, + "loss": 0.4988, + "step": 9439 + }, + { + "epoch": 0.5274184987568791, + "grad_norm": 0.6054083108901978, + "learning_rate": 0.0007387382339757956, + "loss": 0.4181, + "step": 9440 + }, + { + "epoch": 0.5274743693605609, + "grad_norm": 0.5037269592285156, + "learning_rate": 0.0007387102196324519, + "loss": 0.4102, + "step": 9441 + }, + { + "epoch": 0.5275302399642429, + "grad_norm": 0.6569815278053284, + "learning_rate": 0.000738682205289108, + "loss": 0.4608, + "step": 9442 + }, + { + "epoch": 0.5275861105679247, + "grad_norm": 0.526621401309967, + "learning_rate": 0.0007386541909457642, + "loss": 0.426, + "step": 9443 + }, + { + "epoch": 0.5276419811716065, + "grad_norm": 0.5006901621818542, + "learning_rate": 0.0007386261766024204, + "loss": 0.4626, + "step": 9444 + }, + { + "epoch": 0.5276978517752884, + "grad_norm": 1.1367123126983643, + "learning_rate": 0.0007385981622590766, + "loss": 0.5264, + "step": 9445 + }, + { + "epoch": 0.5277537223789703, + "grad_norm": 0.8447065949440002, + "learning_rate": 0.0007385701479157329, + "loss": 0.4456, + "step": 9446 + }, + { + "epoch": 0.5278095929826522, + "grad_norm": 0.5148995518684387, + "learning_rate": 0.0007385421335723891, + "loss": 0.5096, + "step": 9447 + }, + { + "epoch": 0.527865463586334, + "grad_norm": 1.6712063550949097, + "learning_rate": 0.0007385141192290454, + "loss": 0.3905, + "step": 9448 + }, + { + "epoch": 0.527921334190016, + "grad_norm": 0.48053422570228577, + "learning_rate": 0.0007384861048857015, + "loss": 0.413, + "step": 9449 + }, + { + "epoch": 0.5279772047936978, + "grad_norm": 0.6537254452705383, + "learning_rate": 0.0007384580905423577, + "loss": 0.5573, + "step": 9450 + }, + { + "epoch": 0.5280330753973796, + "grad_norm": 0.5284618735313416, + "learning_rate": 0.0007384300761990139, + "loss": 0.4158, + "step": 9451 + }, + { + "epoch": 0.5280889460010615, + "grad_norm": 0.49365121126174927, + "learning_rate": 0.0007384020618556702, + "loss": 0.4116, + "step": 9452 + }, + { + "epoch": 0.5281448166047434, + "grad_norm": 0.48897650837898254, + "learning_rate": 0.0007383740475123263, + "loss": 0.4294, + "step": 9453 + }, + { + "epoch": 0.5282006872084253, + "grad_norm": 0.6362665891647339, + "learning_rate": 0.0007383460331689825, + "loss": 0.5531, + "step": 9454 + }, + { + "epoch": 0.5282565578121071, + "grad_norm": 1.6269398927688599, + "learning_rate": 0.0007383180188256387, + "loss": 0.5102, + "step": 9455 + }, + { + "epoch": 0.5283124284157891, + "grad_norm": 0.34774184226989746, + "learning_rate": 0.000738290004482295, + "loss": 0.4244, + "step": 9456 + }, + { + "epoch": 0.5283682990194709, + "grad_norm": 0.43652045726776123, + "learning_rate": 0.0007382619901389511, + "loss": 0.529, + "step": 9457 + }, + { + "epoch": 0.5284241696231528, + "grad_norm": 0.7325999736785889, + "learning_rate": 0.0007382339757956073, + "loss": 0.4913, + "step": 9458 + }, + { + "epoch": 0.5284800402268347, + "grad_norm": 1.218981385231018, + "learning_rate": 0.0007382059614522636, + "loss": 0.3633, + "step": 9459 + }, + { + "epoch": 0.5285359108305165, + "grad_norm": 0.5096985697746277, + "learning_rate": 0.0007381779471089198, + "loss": 0.4625, + "step": 9460 + }, + { + "epoch": 0.5285917814341984, + "grad_norm": 0.4567103087902069, + "learning_rate": 0.0007381499327655759, + "loss": 0.4763, + "step": 9461 + }, + { + "epoch": 0.5286476520378802, + "grad_norm": 0.9527206420898438, + "learning_rate": 0.0007381219184222321, + "loss": 0.394, + "step": 9462 + }, + { + "epoch": 0.5287035226415622, + "grad_norm": 0.4318629503250122, + "learning_rate": 0.0007380939040788885, + "loss": 0.5832, + "step": 9463 + }, + { + "epoch": 0.528759393245244, + "grad_norm": 0.5763278603553772, + "learning_rate": 0.0007380658897355446, + "loss": 0.43, + "step": 9464 + }, + { + "epoch": 0.5288152638489259, + "grad_norm": 0.4750012755393982, + "learning_rate": 0.0007380378753922008, + "loss": 0.4095, + "step": 9465 + }, + { + "epoch": 0.5288711344526078, + "grad_norm": 0.47209981083869934, + "learning_rate": 0.0007380098610488571, + "loss": 0.477, + "step": 9466 + }, + { + "epoch": 0.5289270050562896, + "grad_norm": 0.3692709505558014, + "learning_rate": 0.0007379818467055133, + "loss": 0.3721, + "step": 9467 + }, + { + "epoch": 0.5289828756599715, + "grad_norm": 0.5342313647270203, + "learning_rate": 0.0007379538323621694, + "loss": 0.5634, + "step": 9468 + }, + { + "epoch": 0.5290387462636533, + "grad_norm": 0.6114898324012756, + "learning_rate": 0.0007379258180188256, + "loss": 0.4043, + "step": 9469 + }, + { + "epoch": 0.5290946168673353, + "grad_norm": 0.6616738438606262, + "learning_rate": 0.0007378978036754819, + "loss": 0.4655, + "step": 9470 + }, + { + "epoch": 0.5291504874710171, + "grad_norm": 0.6606064438819885, + "learning_rate": 0.0007378697893321381, + "loss": 0.4418, + "step": 9471 + }, + { + "epoch": 0.529206358074699, + "grad_norm": 0.7014811038970947, + "learning_rate": 0.0007378417749887942, + "loss": 0.4358, + "step": 9472 + }, + { + "epoch": 0.5292622286783809, + "grad_norm": 0.7999378442764282, + "learning_rate": 0.0007378137606454505, + "loss": 0.4944, + "step": 9473 + }, + { + "epoch": 0.5293180992820627, + "grad_norm": 0.749316394329071, + "learning_rate": 0.0007377857463021067, + "loss": 0.5088, + "step": 9474 + }, + { + "epoch": 0.5293739698857446, + "grad_norm": 0.4496496021747589, + "learning_rate": 0.0007377577319587629, + "loss": 0.4157, + "step": 9475 + }, + { + "epoch": 0.5294298404894265, + "grad_norm": 0.5182341933250427, + "learning_rate": 0.000737729717615419, + "loss": 0.382, + "step": 9476 + }, + { + "epoch": 0.5294857110931084, + "grad_norm": 0.7425380349159241, + "learning_rate": 0.0007377017032720753, + "loss": 0.4675, + "step": 9477 + }, + { + "epoch": 0.5295415816967902, + "grad_norm": 0.4183203876018524, + "learning_rate": 0.0007376736889287316, + "loss": 0.4443, + "step": 9478 + }, + { + "epoch": 0.5295974523004721, + "grad_norm": 0.5040174126625061, + "learning_rate": 0.0007376456745853878, + "loss": 0.4597, + "step": 9479 + }, + { + "epoch": 0.529653322904154, + "grad_norm": 0.612956702709198, + "learning_rate": 0.000737617660242044, + "loss": 0.5474, + "step": 9480 + }, + { + "epoch": 0.5297091935078359, + "grad_norm": 0.3999241590499878, + "learning_rate": 0.0007375896458987002, + "loss": 0.3773, + "step": 9481 + }, + { + "epoch": 0.5297650641115177, + "grad_norm": 1.00253164768219, + "learning_rate": 0.0007375616315553564, + "loss": 0.3961, + "step": 9482 + }, + { + "epoch": 0.5298209347151996, + "grad_norm": 0.47388598322868347, + "learning_rate": 0.0007375336172120126, + "loss": 0.4068, + "step": 9483 + }, + { + "epoch": 0.5298768053188815, + "grad_norm": 0.4934338927268982, + "learning_rate": 0.0007375056028686688, + "loss": 0.4823, + "step": 9484 + }, + { + "epoch": 0.5299326759225633, + "grad_norm": 0.7496746182441711, + "learning_rate": 0.000737477588525325, + "loss": 0.4172, + "step": 9485 + }, + { + "epoch": 0.5299885465262452, + "grad_norm": 0.3807942569255829, + "learning_rate": 0.0007374495741819812, + "loss": 0.3916, + "step": 9486 + }, + { + "epoch": 0.5300444171299271, + "grad_norm": 3.995964765548706, + "learning_rate": 0.0007374215598386373, + "loss": 0.427, + "step": 9487 + }, + { + "epoch": 0.530100287733609, + "grad_norm": 0.594178318977356, + "learning_rate": 0.0007373935454952936, + "loss": 0.4472, + "step": 9488 + }, + { + "epoch": 0.5301561583372908, + "grad_norm": 1.1062265634536743, + "learning_rate": 0.0007373655311519498, + "loss": 0.5812, + "step": 9489 + }, + { + "epoch": 0.5302120289409727, + "grad_norm": 0.4838773012161255, + "learning_rate": 0.000737337516808606, + "loss": 0.2987, + "step": 9490 + }, + { + "epoch": 0.5302678995446546, + "grad_norm": 0.5900713801383972, + "learning_rate": 0.0007373095024652622, + "loss": 0.5451, + "step": 9491 + }, + { + "epoch": 0.5303237701483364, + "grad_norm": 0.9036344289779663, + "learning_rate": 0.0007372814881219184, + "loss": 0.6319, + "step": 9492 + }, + { + "epoch": 0.5303796407520184, + "grad_norm": 0.9367061853408813, + "learning_rate": 0.0007372534737785746, + "loss": 0.525, + "step": 9493 + }, + { + "epoch": 0.5304355113557002, + "grad_norm": 0.4523260295391083, + "learning_rate": 0.0007372254594352308, + "loss": 0.4837, + "step": 9494 + }, + { + "epoch": 0.5304913819593821, + "grad_norm": 0.7822661995887756, + "learning_rate": 0.000737197445091887, + "loss": 0.3771, + "step": 9495 + }, + { + "epoch": 0.5305472525630639, + "grad_norm": 0.8009644746780396, + "learning_rate": 0.0007371694307485433, + "loss": 0.4594, + "step": 9496 + }, + { + "epoch": 0.5306031231667458, + "grad_norm": 0.8880871534347534, + "learning_rate": 0.0007371414164051995, + "loss": 0.6399, + "step": 9497 + }, + { + "epoch": 0.5306589937704277, + "grad_norm": 0.8859409093856812, + "learning_rate": 0.0007371134020618558, + "loss": 0.5705, + "step": 9498 + }, + { + "epoch": 0.5307148643741095, + "grad_norm": 0.8645036220550537, + "learning_rate": 0.0007370853877185119, + "loss": 0.4294, + "step": 9499 + }, + { + "epoch": 0.5307707349777915, + "grad_norm": 0.4508734345436096, + "learning_rate": 0.0007370573733751681, + "loss": 0.3877, + "step": 9500 + }, + { + "epoch": 0.5307707349777915, + "eval_cer": 0.09515236817125462, + "eval_loss": 0.34672826528549194, + "eval_runtime": 56.5733, + "eval_samples_per_second": 80.215, + "eval_steps_per_second": 5.02, + "eval_wer": 0.37719616770481273, + "step": 9500 + }, + { + "epoch": 0.5308266055814733, + "grad_norm": 0.6640086770057678, + "learning_rate": 0.0007370293590318243, + "loss": 0.438, + "step": 9501 + }, + { + "epoch": 0.5308824761851552, + "grad_norm": 0.7739949226379395, + "learning_rate": 0.0007370013446884806, + "loss": 0.5081, + "step": 9502 + }, + { + "epoch": 0.530938346788837, + "grad_norm": 0.7713280320167542, + "learning_rate": 0.0007369733303451367, + "loss": 0.3514, + "step": 9503 + }, + { + "epoch": 0.530994217392519, + "grad_norm": 0.427277535200119, + "learning_rate": 0.0007369453160017929, + "loss": 0.4078, + "step": 9504 + }, + { + "epoch": 0.5310500879962008, + "grad_norm": 0.6514362096786499, + "learning_rate": 0.0007369173016584492, + "loss": 0.5255, + "step": 9505 + }, + { + "epoch": 0.5311059585998826, + "grad_norm": 0.40397799015045166, + "learning_rate": 0.0007368892873151053, + "loss": 0.3664, + "step": 9506 + }, + { + "epoch": 0.5311618292035646, + "grad_norm": 0.5158389806747437, + "learning_rate": 0.0007368612729717615, + "loss": 0.3929, + "step": 9507 + }, + { + "epoch": 0.5312176998072464, + "grad_norm": 6.709324836730957, + "learning_rate": 0.0007368332586284177, + "loss": 0.4217, + "step": 9508 + }, + { + "epoch": 0.5312735704109283, + "grad_norm": 5.369687080383301, + "learning_rate": 0.000736805244285074, + "loss": 0.5215, + "step": 9509 + }, + { + "epoch": 0.5313294410146102, + "grad_norm": 0.6254154443740845, + "learning_rate": 0.0007367772299417301, + "loss": 0.5018, + "step": 9510 + }, + { + "epoch": 0.5313853116182921, + "grad_norm": 0.7239257097244263, + "learning_rate": 0.0007367492155983863, + "loss": 0.5132, + "step": 9511 + }, + { + "epoch": 0.5314411822219739, + "grad_norm": 0.5064093470573425, + "learning_rate": 0.0007367212012550427, + "loss": 0.509, + "step": 9512 + }, + { + "epoch": 0.5314970528256557, + "grad_norm": 0.34267404675483704, + "learning_rate": 0.0007366931869116989, + "loss": 0.4121, + "step": 9513 + }, + { + "epoch": 0.5315529234293377, + "grad_norm": 3.0746190547943115, + "learning_rate": 0.000736665172568355, + "loss": 0.5343, + "step": 9514 + }, + { + "epoch": 0.5316087940330195, + "grad_norm": 0.6060262322425842, + "learning_rate": 0.0007366371582250112, + "loss": 0.5508, + "step": 9515 + }, + { + "epoch": 0.5316646646367014, + "grad_norm": 0.3788052797317505, + "learning_rate": 0.0007366091438816675, + "loss": 0.4894, + "step": 9516 + }, + { + "epoch": 0.5317205352403833, + "grad_norm": 0.6785464286804199, + "learning_rate": 0.0007365811295383237, + "loss": 0.4587, + "step": 9517 + }, + { + "epoch": 0.5317764058440652, + "grad_norm": 0.8826349973678589, + "learning_rate": 0.0007365531151949798, + "loss": 0.3645, + "step": 9518 + }, + { + "epoch": 0.531832276447747, + "grad_norm": 2.0791447162628174, + "learning_rate": 0.0007365251008516361, + "loss": 0.6785, + "step": 9519 + }, + { + "epoch": 0.5318881470514288, + "grad_norm": 0.48994508385658264, + "learning_rate": 0.0007364970865082923, + "loss": 0.5029, + "step": 9520 + }, + { + "epoch": 0.5319440176551108, + "grad_norm": 0.49983519315719604, + "learning_rate": 0.0007364690721649485, + "loss": 0.5102, + "step": 9521 + }, + { + "epoch": 0.5319998882587926, + "grad_norm": 0.8459694385528564, + "learning_rate": 0.0007364410578216046, + "loss": 0.439, + "step": 9522 + }, + { + "epoch": 0.5320557588624745, + "grad_norm": 0.628628134727478, + "learning_rate": 0.0007364130434782609, + "loss": 0.4625, + "step": 9523 + }, + { + "epoch": 0.5321116294661564, + "grad_norm": 1.0802357196807861, + "learning_rate": 0.0007363850291349171, + "loss": 0.4897, + "step": 9524 + }, + { + "epoch": 0.5321675000698383, + "grad_norm": 0.6003199815750122, + "learning_rate": 0.0007363570147915733, + "loss": 0.354, + "step": 9525 + }, + { + "epoch": 0.5322233706735201, + "grad_norm": 0.5316800475120544, + "learning_rate": 0.0007363290004482294, + "loss": 0.5319, + "step": 9526 + }, + { + "epoch": 0.532279241277202, + "grad_norm": 1.3710542917251587, + "learning_rate": 0.0007363009861048857, + "loss": 0.4042, + "step": 9527 + }, + { + "epoch": 0.5323351118808839, + "grad_norm": 0.5625563263893127, + "learning_rate": 0.000736272971761542, + "loss": 0.4547, + "step": 9528 + }, + { + "epoch": 0.5323909824845657, + "grad_norm": 0.7523877024650574, + "learning_rate": 0.000736244957418198, + "loss": 0.46, + "step": 9529 + }, + { + "epoch": 0.5324468530882476, + "grad_norm": 0.7049440741539001, + "learning_rate": 0.0007362169430748544, + "loss": 0.4658, + "step": 9530 + }, + { + "epoch": 0.5325027236919295, + "grad_norm": 0.5364850163459778, + "learning_rate": 0.0007361889287315106, + "loss": 0.5438, + "step": 9531 + }, + { + "epoch": 0.5325585942956114, + "grad_norm": 0.42239829897880554, + "learning_rate": 0.0007361609143881668, + "loss": 0.4922, + "step": 9532 + }, + { + "epoch": 0.5326144648992932, + "grad_norm": 0.612972617149353, + "learning_rate": 0.0007361329000448229, + "loss": 0.4738, + "step": 9533 + }, + { + "epoch": 0.5326703355029752, + "grad_norm": 0.5847073793411255, + "learning_rate": 0.0007361048857014792, + "loss": 0.3897, + "step": 9534 + }, + { + "epoch": 0.532726206106657, + "grad_norm": 0.4557243287563324, + "learning_rate": 0.0007360768713581354, + "loss": 0.4028, + "step": 9535 + }, + { + "epoch": 0.5327820767103388, + "grad_norm": 0.5331316590309143, + "learning_rate": 0.0007360488570147916, + "loss": 0.4198, + "step": 9536 + }, + { + "epoch": 0.5328379473140207, + "grad_norm": 1.388588786125183, + "learning_rate": 0.0007360208426714478, + "loss": 0.4544, + "step": 9537 + }, + { + "epoch": 0.5328938179177026, + "grad_norm": 1.2249032258987427, + "learning_rate": 0.000735992828328104, + "loss": 0.4875, + "step": 9538 + }, + { + "epoch": 0.5329496885213845, + "grad_norm": 0.6199700236320496, + "learning_rate": 0.0007359648139847602, + "loss": 0.5778, + "step": 9539 + }, + { + "epoch": 0.5330055591250663, + "grad_norm": 0.7522794008255005, + "learning_rate": 0.0007359367996414164, + "loss": 0.4489, + "step": 9540 + }, + { + "epoch": 0.5330614297287483, + "grad_norm": 0.8172647953033447, + "learning_rate": 0.0007359087852980726, + "loss": 0.3641, + "step": 9541 + }, + { + "epoch": 0.5331173003324301, + "grad_norm": 0.887337863445282, + "learning_rate": 0.0007358807709547288, + "loss": 0.5067, + "step": 9542 + }, + { + "epoch": 0.5331731709361119, + "grad_norm": 0.610144317150116, + "learning_rate": 0.000735852756611385, + "loss": 0.3945, + "step": 9543 + }, + { + "epoch": 0.5332290415397939, + "grad_norm": 1.1674920320510864, + "learning_rate": 0.0007358247422680414, + "loss": 0.6741, + "step": 9544 + }, + { + "epoch": 0.5332849121434757, + "grad_norm": 0.5802062153816223, + "learning_rate": 0.0007357967279246974, + "loss": 0.4652, + "step": 9545 + }, + { + "epoch": 0.5333407827471576, + "grad_norm": 0.8443202972412109, + "learning_rate": 0.0007357687135813537, + "loss": 0.5309, + "step": 9546 + }, + { + "epoch": 0.5333966533508394, + "grad_norm": 0.5263758897781372, + "learning_rate": 0.0007357406992380099, + "loss": 0.5648, + "step": 9547 + }, + { + "epoch": 0.5334525239545214, + "grad_norm": 1.6626850366592407, + "learning_rate": 0.0007357126848946661, + "loss": 0.5236, + "step": 9548 + }, + { + "epoch": 0.5335083945582032, + "grad_norm": 0.4968186616897583, + "learning_rate": 0.0007356846705513223, + "loss": 0.3626, + "step": 9549 + }, + { + "epoch": 0.533564265161885, + "grad_norm": 1.281759262084961, + "learning_rate": 0.0007356566562079785, + "loss": 0.5166, + "step": 9550 + }, + { + "epoch": 0.533620135765567, + "grad_norm": 0.834347665309906, + "learning_rate": 0.0007356286418646348, + "loss": 0.4554, + "step": 9551 + }, + { + "epoch": 0.5336760063692488, + "grad_norm": 0.4597853720188141, + "learning_rate": 0.0007356006275212909, + "loss": 0.4513, + "step": 9552 + }, + { + "epoch": 0.5337318769729307, + "grad_norm": 0.7161881327629089, + "learning_rate": 0.0007355726131779471, + "loss": 0.4597, + "step": 9553 + }, + { + "epoch": 0.5337877475766125, + "grad_norm": 0.6686603426933289, + "learning_rate": 0.0007355445988346033, + "loss": 0.4549, + "step": 9554 + }, + { + "epoch": 0.5338436181802945, + "grad_norm": 0.7642550468444824, + "learning_rate": 0.0007355165844912596, + "loss": 0.6114, + "step": 9555 + }, + { + "epoch": 0.5338994887839763, + "grad_norm": 0.8337416648864746, + "learning_rate": 0.0007354885701479157, + "loss": 0.6253, + "step": 9556 + }, + { + "epoch": 0.5339553593876581, + "grad_norm": 2.7806122303009033, + "learning_rate": 0.0007354605558045719, + "loss": 0.4809, + "step": 9557 + }, + { + "epoch": 0.5340112299913401, + "grad_norm": 0.5733252763748169, + "learning_rate": 0.0007354325414612281, + "loss": 0.4308, + "step": 9558 + }, + { + "epoch": 0.5340671005950219, + "grad_norm": 0.5921027660369873, + "learning_rate": 0.0007354045271178844, + "loss": 0.4282, + "step": 9559 + }, + { + "epoch": 0.5341229711987038, + "grad_norm": 0.5893519520759583, + "learning_rate": 0.0007353765127745405, + "loss": 0.3692, + "step": 9560 + }, + { + "epoch": 0.5341788418023857, + "grad_norm": 0.8226804733276367, + "learning_rate": 0.0007353484984311967, + "loss": 0.3851, + "step": 9561 + }, + { + "epoch": 0.5342347124060676, + "grad_norm": 0.4966617226600647, + "learning_rate": 0.0007353204840878531, + "loss": 0.4413, + "step": 9562 + }, + { + "epoch": 0.5342905830097494, + "grad_norm": 0.5057656168937683, + "learning_rate": 0.0007352924697445093, + "loss": 0.4159, + "step": 9563 + }, + { + "epoch": 0.5343464536134312, + "grad_norm": 0.646847665309906, + "learning_rate": 0.0007352644554011654, + "loss": 0.4404, + "step": 9564 + }, + { + "epoch": 0.5344023242171132, + "grad_norm": 0.4193251132965088, + "learning_rate": 0.0007352364410578216, + "loss": 0.5215, + "step": 9565 + }, + { + "epoch": 0.534458194820795, + "grad_norm": 0.8601469993591309, + "learning_rate": 0.0007352084267144779, + "loss": 0.5922, + "step": 9566 + }, + { + "epoch": 0.5345140654244769, + "grad_norm": 0.8029796481132507, + "learning_rate": 0.0007351804123711341, + "loss": 0.4754, + "step": 9567 + }, + { + "epoch": 0.5345699360281588, + "grad_norm": 0.8710438013076782, + "learning_rate": 0.0007351523980277902, + "loss": 0.3867, + "step": 9568 + }, + { + "epoch": 0.5346258066318407, + "grad_norm": 0.6149936318397522, + "learning_rate": 0.0007351243836844465, + "loss": 0.5183, + "step": 9569 + }, + { + "epoch": 0.5346816772355225, + "grad_norm": 0.4845406711101532, + "learning_rate": 0.0007350963693411027, + "loss": 0.4547, + "step": 9570 + }, + { + "epoch": 0.5347375478392044, + "grad_norm": 1.321524739265442, + "learning_rate": 0.0007350683549977588, + "loss": 0.3682, + "step": 9571 + }, + { + "epoch": 0.5347934184428863, + "grad_norm": 0.6401252150535583, + "learning_rate": 0.000735040340654415, + "loss": 0.5283, + "step": 9572 + }, + { + "epoch": 0.5348492890465681, + "grad_norm": 3.8739120960235596, + "learning_rate": 0.0007350123263110713, + "loss": 0.6241, + "step": 9573 + }, + { + "epoch": 0.53490515965025, + "grad_norm": 1.3839811086654663, + "learning_rate": 0.0007349843119677275, + "loss": 0.4396, + "step": 9574 + }, + { + "epoch": 0.5349610302539319, + "grad_norm": 0.4583531320095062, + "learning_rate": 0.0007349562976243836, + "loss": 0.3603, + "step": 9575 + }, + { + "epoch": 0.5350169008576138, + "grad_norm": 0.6845083236694336, + "learning_rate": 0.0007349282832810399, + "loss": 0.5748, + "step": 9576 + }, + { + "epoch": 0.5350727714612956, + "grad_norm": 0.42667698860168457, + "learning_rate": 0.0007349002689376961, + "loss": 0.4291, + "step": 9577 + }, + { + "epoch": 0.5351286420649776, + "grad_norm": 0.6249213814735413, + "learning_rate": 0.0007348722545943523, + "loss": 0.4613, + "step": 9578 + }, + { + "epoch": 0.5351845126686594, + "grad_norm": 0.4961353838443756, + "learning_rate": 0.0007348442402510084, + "loss": 0.3604, + "step": 9579 + }, + { + "epoch": 0.5352403832723412, + "grad_norm": 0.6090753674507141, + "learning_rate": 0.0007348162259076648, + "loss": 0.5017, + "step": 9580 + }, + { + "epoch": 0.5352962538760231, + "grad_norm": 0.4203108549118042, + "learning_rate": 0.000734788211564321, + "loss": 0.4189, + "step": 9581 + }, + { + "epoch": 0.535352124479705, + "grad_norm": 2.2869441509246826, + "learning_rate": 0.0007347601972209772, + "loss": 0.4775, + "step": 9582 + }, + { + "epoch": 0.5354079950833869, + "grad_norm": 0.4856134057044983, + "learning_rate": 0.0007347321828776334, + "loss": 0.5462, + "step": 9583 + }, + { + "epoch": 0.5354638656870687, + "grad_norm": 1.4967466592788696, + "learning_rate": 0.0007347041685342896, + "loss": 0.4451, + "step": 9584 + }, + { + "epoch": 0.5355197362907507, + "grad_norm": 0.4881626069545746, + "learning_rate": 0.0007346761541909458, + "loss": 0.452, + "step": 9585 + }, + { + "epoch": 0.5355756068944325, + "grad_norm": 0.4292442202568054, + "learning_rate": 0.000734648139847602, + "loss": 0.4161, + "step": 9586 + }, + { + "epoch": 0.5356314774981144, + "grad_norm": 0.43770313262939453, + "learning_rate": 0.0007346201255042582, + "loss": 0.4617, + "step": 9587 + }, + { + "epoch": 0.5356873481017962, + "grad_norm": 0.48553022742271423, + "learning_rate": 0.0007345921111609144, + "loss": 0.4576, + "step": 9588 + }, + { + "epoch": 0.5357432187054781, + "grad_norm": 0.7372949719429016, + "learning_rate": 0.0007345640968175706, + "loss": 0.5059, + "step": 9589 + }, + { + "epoch": 0.53579908930916, + "grad_norm": 0.7224258184432983, + "learning_rate": 0.0007345360824742269, + "loss": 0.4849, + "step": 9590 + }, + { + "epoch": 0.5358549599128418, + "grad_norm": 0.4852941632270813, + "learning_rate": 0.000734508068130883, + "loss": 0.394, + "step": 9591 + }, + { + "epoch": 0.5359108305165238, + "grad_norm": 1.0977864265441895, + "learning_rate": 0.0007344800537875392, + "loss": 0.4399, + "step": 9592 + }, + { + "epoch": 0.5359667011202056, + "grad_norm": 0.6219632625579834, + "learning_rate": 0.0007344520394441954, + "loss": 0.4092, + "step": 9593 + }, + { + "epoch": 0.5360225717238875, + "grad_norm": 1.0537599325180054, + "learning_rate": 0.0007344240251008516, + "loss": 0.3335, + "step": 9594 + }, + { + "epoch": 0.5360784423275694, + "grad_norm": 0.8569068908691406, + "learning_rate": 0.0007343960107575078, + "loss": 0.4127, + "step": 9595 + }, + { + "epoch": 0.5361343129312512, + "grad_norm": 0.3713802099227905, + "learning_rate": 0.000734367996414164, + "loss": 0.4586, + "step": 9596 + }, + { + "epoch": 0.5361901835349331, + "grad_norm": 0.9644632935523987, + "learning_rate": 0.0007343399820708203, + "loss": 0.4824, + "step": 9597 + }, + { + "epoch": 0.5362460541386149, + "grad_norm": 0.7779699563980103, + "learning_rate": 0.0007343119677274765, + "loss": 0.529, + "step": 9598 + }, + { + "epoch": 0.5363019247422969, + "grad_norm": 0.4850205183029175, + "learning_rate": 0.0007342839533841327, + "loss": 0.4834, + "step": 9599 + }, + { + "epoch": 0.5363577953459787, + "grad_norm": 0.44751763343811035, + "learning_rate": 0.0007342559390407889, + "loss": 0.4055, + "step": 9600 + }, + { + "epoch": 0.5364136659496606, + "grad_norm": 0.4143773019313812, + "learning_rate": 0.0007342279246974452, + "loss": 0.5074, + "step": 9601 + }, + { + "epoch": 0.5364695365533425, + "grad_norm": 0.4499647915363312, + "learning_rate": 0.0007341999103541013, + "loss": 0.4517, + "step": 9602 + }, + { + "epoch": 0.5365254071570243, + "grad_norm": 0.5708054304122925, + "learning_rate": 0.0007341718960107575, + "loss": 0.4087, + "step": 9603 + }, + { + "epoch": 0.5365812777607062, + "grad_norm": 1.2631261348724365, + "learning_rate": 0.0007341438816674137, + "loss": 0.436, + "step": 9604 + }, + { + "epoch": 0.536637148364388, + "grad_norm": 0.8372437357902527, + "learning_rate": 0.00073411586732407, + "loss": 0.4777, + "step": 9605 + }, + { + "epoch": 0.53669301896807, + "grad_norm": 0.5422977805137634, + "learning_rate": 0.0007340878529807261, + "loss": 0.4688, + "step": 9606 + }, + { + "epoch": 0.5367488895717518, + "grad_norm": 1.3958781957626343, + "learning_rate": 0.0007340598386373823, + "loss": 0.4996, + "step": 9607 + }, + { + "epoch": 0.5368047601754337, + "grad_norm": 0.5873874425888062, + "learning_rate": 0.0007340318242940386, + "loss": 0.4015, + "step": 9608 + }, + { + "epoch": 0.5368606307791156, + "grad_norm": 0.48345863819122314, + "learning_rate": 0.0007340038099506948, + "loss": 0.4077, + "step": 9609 + }, + { + "epoch": 0.5369165013827975, + "grad_norm": 0.5441824793815613, + "learning_rate": 0.0007339757956073509, + "loss": 0.6609, + "step": 9610 + }, + { + "epoch": 0.5369723719864793, + "grad_norm": 0.5437590479850769, + "learning_rate": 0.0007339477812640071, + "loss": 0.4448, + "step": 9611 + }, + { + "epoch": 0.5370282425901612, + "grad_norm": 1.125412940979004, + "learning_rate": 0.0007339197669206635, + "loss": 0.5503, + "step": 9612 + }, + { + "epoch": 0.5370841131938431, + "grad_norm": 0.4116196930408478, + "learning_rate": 0.0007338917525773196, + "loss": 0.4776, + "step": 9613 + }, + { + "epoch": 0.5371399837975249, + "grad_norm": 3.037252187728882, + "learning_rate": 0.0007338637382339758, + "loss": 0.593, + "step": 9614 + }, + { + "epoch": 0.5371958544012068, + "grad_norm": 0.5102656483650208, + "learning_rate": 0.0007338357238906321, + "loss": 0.3849, + "step": 9615 + }, + { + "epoch": 0.5372517250048887, + "grad_norm": 0.7266536355018616, + "learning_rate": 0.0007338077095472883, + "loss": 0.4761, + "step": 9616 + }, + { + "epoch": 0.5373075956085706, + "grad_norm": 0.6717156171798706, + "learning_rate": 0.0007337796952039444, + "loss": 0.4501, + "step": 9617 + }, + { + "epoch": 0.5373634662122524, + "grad_norm": 0.9842637777328491, + "learning_rate": 0.0007337516808606006, + "loss": 0.5147, + "step": 9618 + }, + { + "epoch": 0.5374193368159343, + "grad_norm": 0.6293283104896545, + "learning_rate": 0.0007337236665172569, + "loss": 0.3899, + "step": 9619 + }, + { + "epoch": 0.5374752074196162, + "grad_norm": 0.7908117175102234, + "learning_rate": 0.0007336956521739131, + "loss": 0.6289, + "step": 9620 + }, + { + "epoch": 0.537531078023298, + "grad_norm": 0.5345439314842224, + "learning_rate": 0.0007336676378305692, + "loss": 0.4268, + "step": 9621 + }, + { + "epoch": 0.5375869486269799, + "grad_norm": 1.0073155164718628, + "learning_rate": 0.0007336396234872255, + "loss": 0.5505, + "step": 9622 + }, + { + "epoch": 0.5376428192306618, + "grad_norm": 0.5015206933021545, + "learning_rate": 0.0007336116091438817, + "loss": 0.3587, + "step": 9623 + }, + { + "epoch": 0.5376986898343437, + "grad_norm": 0.6704979538917542, + "learning_rate": 0.0007335835948005379, + "loss": 0.5556, + "step": 9624 + }, + { + "epoch": 0.5377545604380255, + "grad_norm": 0.6410689353942871, + "learning_rate": 0.000733555580457194, + "loss": 0.4068, + "step": 9625 + }, + { + "epoch": 0.5378104310417074, + "grad_norm": 0.6539160013198853, + "learning_rate": 0.0007335275661138503, + "loss": 0.3324, + "step": 9626 + }, + { + "epoch": 0.5378663016453893, + "grad_norm": 0.6330994367599487, + "learning_rate": 0.0007334995517705065, + "loss": 0.4829, + "step": 9627 + }, + { + "epoch": 0.5379221722490711, + "grad_norm": 0.6577051281929016, + "learning_rate": 0.0007334715374271627, + "loss": 0.5033, + "step": 9628 + }, + { + "epoch": 0.537978042852753, + "grad_norm": 0.8891093134880066, + "learning_rate": 0.0007334435230838188, + "loss": 0.4689, + "step": 9629 + }, + { + "epoch": 0.5380339134564349, + "grad_norm": 0.7907425761222839, + "learning_rate": 0.0007334155087404752, + "loss": 0.5073, + "step": 9630 + }, + { + "epoch": 0.5380897840601168, + "grad_norm": 0.749832272529602, + "learning_rate": 0.0007333874943971314, + "loss": 0.4162, + "step": 9631 + }, + { + "epoch": 0.5381456546637986, + "grad_norm": 4.270244121551514, + "learning_rate": 0.0007333594800537876, + "loss": 0.4699, + "step": 9632 + }, + { + "epoch": 0.5382015252674806, + "grad_norm": 0.5053085088729858, + "learning_rate": 0.0007333314657104438, + "loss": 0.4928, + "step": 9633 + }, + { + "epoch": 0.5382573958711624, + "grad_norm": 1.0502394437789917, + "learning_rate": 0.0007333034513671, + "loss": 0.3782, + "step": 9634 + }, + { + "epoch": 0.5383132664748442, + "grad_norm": 0.6174446940422058, + "learning_rate": 0.0007332754370237562, + "loss": 0.3971, + "step": 9635 + }, + { + "epoch": 0.5383691370785262, + "grad_norm": 0.4950055181980133, + "learning_rate": 0.0007332474226804123, + "loss": 0.48, + "step": 9636 + }, + { + "epoch": 0.538425007682208, + "grad_norm": 0.519260048866272, + "learning_rate": 0.0007332194083370686, + "loss": 0.4755, + "step": 9637 + }, + { + "epoch": 0.5384808782858899, + "grad_norm": 0.45063525438308716, + "learning_rate": 0.0007331913939937248, + "loss": 0.4787, + "step": 9638 + }, + { + "epoch": 0.5385367488895717, + "grad_norm": 0.7486417293548584, + "learning_rate": 0.000733163379650381, + "loss": 0.4602, + "step": 9639 + }, + { + "epoch": 0.5385926194932537, + "grad_norm": 0.39729058742523193, + "learning_rate": 0.0007331353653070372, + "loss": 0.3798, + "step": 9640 + }, + { + "epoch": 0.5386484900969355, + "grad_norm": 0.5883775353431702, + "learning_rate": 0.0007331073509636934, + "loss": 0.4994, + "step": 9641 + }, + { + "epoch": 0.5387043607006173, + "grad_norm": 0.5966442823410034, + "learning_rate": 0.0007330793366203496, + "loss": 0.5235, + "step": 9642 + }, + { + "epoch": 0.5387602313042993, + "grad_norm": 0.5644200444221497, + "learning_rate": 0.0007330513222770058, + "loss": 0.5114, + "step": 9643 + }, + { + "epoch": 0.5388161019079811, + "grad_norm": 0.4425967335700989, + "learning_rate": 0.000733023307933662, + "loss": 0.4209, + "step": 9644 + }, + { + "epoch": 0.538871972511663, + "grad_norm": 1.5599433183670044, + "learning_rate": 0.0007329952935903182, + "loss": 0.464, + "step": 9645 + }, + { + "epoch": 0.5389278431153448, + "grad_norm": 0.48797833919525146, + "learning_rate": 0.0007329672792469745, + "loss": 0.4703, + "step": 9646 + }, + { + "epoch": 0.5389837137190268, + "grad_norm": 0.4840639531612396, + "learning_rate": 0.0007329392649036308, + "loss": 0.5472, + "step": 9647 + }, + { + "epoch": 0.5390395843227086, + "grad_norm": 0.3956967294216156, + "learning_rate": 0.0007329112505602869, + "loss": 0.4347, + "step": 9648 + }, + { + "epoch": 0.5390954549263904, + "grad_norm": 0.6399507522583008, + "learning_rate": 0.0007328832362169431, + "loss": 0.4208, + "step": 9649 + }, + { + "epoch": 0.5391513255300724, + "grad_norm": 0.4924982190132141, + "learning_rate": 0.0007328552218735993, + "loss": 0.3718, + "step": 9650 + }, + { + "epoch": 0.5392071961337542, + "grad_norm": 0.4584347903728485, + "learning_rate": 0.0007328272075302556, + "loss": 0.42, + "step": 9651 + }, + { + "epoch": 0.5392630667374361, + "grad_norm": 0.5087452530860901, + "learning_rate": 0.0007327991931869117, + "loss": 0.4477, + "step": 9652 + }, + { + "epoch": 0.539318937341118, + "grad_norm": 0.48496773838996887, + "learning_rate": 0.0007327711788435679, + "loss": 0.4781, + "step": 9653 + }, + { + "epoch": 0.5393748079447999, + "grad_norm": 0.5072957277297974, + "learning_rate": 0.0007327431645002242, + "loss": 0.4615, + "step": 9654 + }, + { + "epoch": 0.5394306785484817, + "grad_norm": 0.47968149185180664, + "learning_rate": 0.0007327151501568803, + "loss": 0.5313, + "step": 9655 + }, + { + "epoch": 0.5394865491521635, + "grad_norm": 0.5403556227684021, + "learning_rate": 0.0007326871358135365, + "loss": 0.4431, + "step": 9656 + }, + { + "epoch": 0.5395424197558455, + "grad_norm": 2.584275722503662, + "learning_rate": 0.0007326591214701927, + "loss": 0.5409, + "step": 9657 + }, + { + "epoch": 0.5395982903595273, + "grad_norm": 0.7629426717758179, + "learning_rate": 0.000732631107126849, + "loss": 0.6915, + "step": 9658 + }, + { + "epoch": 0.5396541609632092, + "grad_norm": 2.3765835762023926, + "learning_rate": 0.0007326030927835051, + "loss": 0.4437, + "step": 9659 + }, + { + "epoch": 0.5397100315668911, + "grad_norm": 0.4232288599014282, + "learning_rate": 0.0007325750784401613, + "loss": 0.3709, + "step": 9660 + }, + { + "epoch": 0.539765902170573, + "grad_norm": 0.5440918803215027, + "learning_rate": 0.0007325470640968176, + "loss": 0.3768, + "step": 9661 + }, + { + "epoch": 0.5398217727742548, + "grad_norm": 2.1216044425964355, + "learning_rate": 0.0007325190497534739, + "loss": 0.3846, + "step": 9662 + }, + { + "epoch": 0.5398776433779366, + "grad_norm": 1.2588558197021484, + "learning_rate": 0.00073249103541013, + "loss": 0.4145, + "step": 9663 + }, + { + "epoch": 0.5399335139816186, + "grad_norm": 0.5828195810317993, + "learning_rate": 0.0007324630210667862, + "loss": 0.5409, + "step": 9664 + }, + { + "epoch": 0.5399893845853004, + "grad_norm": 14.872206687927246, + "learning_rate": 0.0007324350067234425, + "loss": 0.3969, + "step": 9665 + }, + { + "epoch": 0.5400452551889823, + "grad_norm": 0.4147075414657593, + "learning_rate": 0.0007324069923800987, + "loss": 0.4671, + "step": 9666 + }, + { + "epoch": 0.5401011257926642, + "grad_norm": 1.13008451461792, + "learning_rate": 0.0007323789780367548, + "loss": 0.3767, + "step": 9667 + }, + { + "epoch": 0.5401569963963461, + "grad_norm": 0.8963695168495178, + "learning_rate": 0.000732350963693411, + "loss": 0.5149, + "step": 9668 + }, + { + "epoch": 0.5402128670000279, + "grad_norm": 2.1477346420288086, + "learning_rate": 0.0007323229493500673, + "loss": 0.4627, + "step": 9669 + }, + { + "epoch": 0.5402687376037099, + "grad_norm": 0.5303489565849304, + "learning_rate": 0.0007322949350067235, + "loss": 0.5291, + "step": 9670 + }, + { + "epoch": 0.5403246082073917, + "grad_norm": 1.327365756034851, + "learning_rate": 0.0007322669206633796, + "loss": 0.3871, + "step": 9671 + }, + { + "epoch": 0.5403804788110735, + "grad_norm": 0.9205843210220337, + "learning_rate": 0.0007322389063200359, + "loss": 0.5336, + "step": 9672 + }, + { + "epoch": 0.5404363494147554, + "grad_norm": 0.4792768955230713, + "learning_rate": 0.0007322108919766921, + "loss": 0.4234, + "step": 9673 + }, + { + "epoch": 0.5404922200184373, + "grad_norm": 0.64832603931427, + "learning_rate": 0.0007321828776333483, + "loss": 0.3784, + "step": 9674 + }, + { + "epoch": 0.5405480906221192, + "grad_norm": 0.5785941481590271, + "learning_rate": 0.0007321548632900044, + "loss": 0.4572, + "step": 9675 + }, + { + "epoch": 0.540603961225801, + "grad_norm": 0.6081535220146179, + "learning_rate": 0.0007321268489466607, + "loss": 0.3607, + "step": 9676 + }, + { + "epoch": 0.540659831829483, + "grad_norm": 0.41339385509490967, + "learning_rate": 0.0007320988346033169, + "loss": 0.5128, + "step": 9677 + }, + { + "epoch": 0.5407157024331648, + "grad_norm": 0.41932544112205505, + "learning_rate": 0.000732070820259973, + "loss": 0.4557, + "step": 9678 + }, + { + "epoch": 0.5407715730368466, + "grad_norm": 0.4364529252052307, + "learning_rate": 0.0007320428059166294, + "loss": 0.4657, + "step": 9679 + }, + { + "epoch": 0.5408274436405285, + "grad_norm": 0.8164024353027344, + "learning_rate": 0.0007320147915732856, + "loss": 0.4601, + "step": 9680 + }, + { + "epoch": 0.5408833142442104, + "grad_norm": 0.677200198173523, + "learning_rate": 0.0007319867772299418, + "loss": 0.4882, + "step": 9681 + }, + { + "epoch": 0.5409391848478923, + "grad_norm": 0.5248569250106812, + "learning_rate": 0.0007319587628865979, + "loss": 0.481, + "step": 9682 + }, + { + "epoch": 0.5409950554515741, + "grad_norm": 0.37538301944732666, + "learning_rate": 0.0007319307485432542, + "loss": 0.4614, + "step": 9683 + }, + { + "epoch": 0.5410509260552561, + "grad_norm": 0.47397226095199585, + "learning_rate": 0.0007319027341999104, + "loss": 0.471, + "step": 9684 + }, + { + "epoch": 0.5411067966589379, + "grad_norm": 0.7922878861427307, + "learning_rate": 0.0007318747198565666, + "loss": 0.4731, + "step": 9685 + }, + { + "epoch": 0.5411626672626197, + "grad_norm": 0.5008073449134827, + "learning_rate": 0.0007318467055132228, + "loss": 0.5304, + "step": 9686 + }, + { + "epoch": 0.5412185378663017, + "grad_norm": 1.019485592842102, + "learning_rate": 0.000731818691169879, + "loss": 0.4642, + "step": 9687 + }, + { + "epoch": 0.5412744084699835, + "grad_norm": 0.5183514356613159, + "learning_rate": 0.0007317906768265352, + "loss": 0.6758, + "step": 9688 + }, + { + "epoch": 0.5413302790736654, + "grad_norm": 0.4795229136943817, + "learning_rate": 0.0007317626624831914, + "loss": 0.4042, + "step": 9689 + }, + { + "epoch": 0.5413861496773472, + "grad_norm": 0.5264679789543152, + "learning_rate": 0.0007317346481398476, + "loss": 0.5381, + "step": 9690 + }, + { + "epoch": 0.5414420202810292, + "grad_norm": 0.7487430572509766, + "learning_rate": 0.0007317066337965038, + "loss": 0.3746, + "step": 9691 + }, + { + "epoch": 0.541497890884711, + "grad_norm": 0.45973315834999084, + "learning_rate": 0.00073167861945316, + "loss": 0.5236, + "step": 9692 + }, + { + "epoch": 0.5415537614883928, + "grad_norm": 0.3754914402961731, + "learning_rate": 0.0007316506051098163, + "loss": 0.4234, + "step": 9693 + }, + { + "epoch": 0.5416096320920748, + "grad_norm": 5.0544657707214355, + "learning_rate": 0.0007316225907664724, + "loss": 0.5555, + "step": 9694 + }, + { + "epoch": 0.5416655026957566, + "grad_norm": 0.8883426785469055, + "learning_rate": 0.0007315945764231286, + "loss": 0.4457, + "step": 9695 + }, + { + "epoch": 0.5417213732994385, + "grad_norm": 0.45912158489227295, + "learning_rate": 0.0007315665620797849, + "loss": 0.5075, + "step": 9696 + }, + { + "epoch": 0.5417772439031203, + "grad_norm": 0.5846559405326843, + "learning_rate": 0.0007315385477364411, + "loss": 0.4968, + "step": 9697 + }, + { + "epoch": 0.5418331145068023, + "grad_norm": 0.532818078994751, + "learning_rate": 0.0007315105333930973, + "loss": 0.6212, + "step": 9698 + }, + { + "epoch": 0.5418889851104841, + "grad_norm": 0.4118671417236328, + "learning_rate": 0.0007314825190497535, + "loss": 0.5491, + "step": 9699 + }, + { + "epoch": 0.541944855714166, + "grad_norm": 0.6494531631469727, + "learning_rate": 0.0007314545047064098, + "loss": 0.7313, + "step": 9700 + }, + { + "epoch": 0.5420007263178479, + "grad_norm": 0.43631088733673096, + "learning_rate": 0.0007314264903630659, + "loss": 0.4481, + "step": 9701 + }, + { + "epoch": 0.5420565969215297, + "grad_norm": 0.6695535778999329, + "learning_rate": 0.0007313984760197221, + "loss": 0.5015, + "step": 9702 + }, + { + "epoch": 0.5421124675252116, + "grad_norm": 1.2024284601211548, + "learning_rate": 0.0007313704616763783, + "loss": 0.5247, + "step": 9703 + }, + { + "epoch": 0.5421683381288935, + "grad_norm": 0.39276546239852905, + "learning_rate": 0.0007313424473330346, + "loss": 0.4394, + "step": 9704 + }, + { + "epoch": 0.5422242087325754, + "grad_norm": 0.5582042932510376, + "learning_rate": 0.0007313144329896907, + "loss": 0.5448, + "step": 9705 + }, + { + "epoch": 0.5422800793362572, + "grad_norm": 0.4468725621700287, + "learning_rate": 0.0007312864186463469, + "loss": 0.4039, + "step": 9706 + }, + { + "epoch": 0.542335949939939, + "grad_norm": 0.6082244515419006, + "learning_rate": 0.0007312584043030031, + "loss": 0.4602, + "step": 9707 + }, + { + "epoch": 0.542391820543621, + "grad_norm": 0.5273591876029968, + "learning_rate": 0.0007312303899596594, + "loss": 0.435, + "step": 9708 + }, + { + "epoch": 0.5424476911473028, + "grad_norm": 0.4063739776611328, + "learning_rate": 0.0007312023756163155, + "loss": 0.3748, + "step": 9709 + }, + { + "epoch": 0.5425035617509847, + "grad_norm": 0.42780888080596924, + "learning_rate": 0.0007311743612729717, + "loss": 0.4742, + "step": 9710 + }, + { + "epoch": 0.5425594323546666, + "grad_norm": 0.6500921249389648, + "learning_rate": 0.000731146346929628, + "loss": 0.4724, + "step": 9711 + }, + { + "epoch": 0.5426153029583485, + "grad_norm": 0.5993912816047668, + "learning_rate": 0.0007311183325862843, + "loss": 0.4698, + "step": 9712 + }, + { + "epoch": 0.5426711735620303, + "grad_norm": 1.0188837051391602, + "learning_rate": 0.0007310903182429404, + "loss": 0.4517, + "step": 9713 + }, + { + "epoch": 0.5427270441657122, + "grad_norm": 0.4840168356895447, + "learning_rate": 0.0007310623038995966, + "loss": 0.4449, + "step": 9714 + }, + { + "epoch": 0.5427829147693941, + "grad_norm": 0.4852624535560608, + "learning_rate": 0.0007310342895562529, + "loss": 0.4756, + "step": 9715 + }, + { + "epoch": 0.542838785373076, + "grad_norm": 0.40125879645347595, + "learning_rate": 0.0007310062752129091, + "loss": 0.3747, + "step": 9716 + }, + { + "epoch": 0.5428946559767578, + "grad_norm": 0.4310374855995178, + "learning_rate": 0.0007309782608695652, + "loss": 0.4836, + "step": 9717 + }, + { + "epoch": 0.5429505265804397, + "grad_norm": 0.45206668972969055, + "learning_rate": 0.0007309502465262215, + "loss": 0.4101, + "step": 9718 + }, + { + "epoch": 0.5430063971841216, + "grad_norm": 1.0091686248779297, + "learning_rate": 0.0007309222321828777, + "loss": 0.495, + "step": 9719 + }, + { + "epoch": 0.5430622677878034, + "grad_norm": 0.33111631870269775, + "learning_rate": 0.0007308942178395338, + "loss": 0.3575, + "step": 9720 + }, + { + "epoch": 0.5431181383914854, + "grad_norm": 1.0076470375061035, + "learning_rate": 0.00073086620349619, + "loss": 0.3998, + "step": 9721 + }, + { + "epoch": 0.5431740089951672, + "grad_norm": 1.3642774820327759, + "learning_rate": 0.0007308381891528463, + "loss": 0.5956, + "step": 9722 + }, + { + "epoch": 0.543229879598849, + "grad_norm": 0.5961556434631348, + "learning_rate": 0.0007308101748095025, + "loss": 0.4924, + "step": 9723 + }, + { + "epoch": 0.5432857502025309, + "grad_norm": 0.889056384563446, + "learning_rate": 0.0007307821604661586, + "loss": 0.4445, + "step": 9724 + }, + { + "epoch": 0.5433416208062128, + "grad_norm": 0.6498591303825378, + "learning_rate": 0.0007307541461228149, + "loss": 0.5477, + "step": 9725 + }, + { + "epoch": 0.5433974914098947, + "grad_norm": 1.078582763671875, + "learning_rate": 0.0007307261317794711, + "loss": 0.4401, + "step": 9726 + }, + { + "epoch": 0.5434533620135765, + "grad_norm": 0.6110192537307739, + "learning_rate": 0.0007306981174361273, + "loss": 0.6944, + "step": 9727 + }, + { + "epoch": 0.5435092326172585, + "grad_norm": 0.5857904553413391, + "learning_rate": 0.0007306701030927834, + "loss": 0.4742, + "step": 9728 + }, + { + "epoch": 0.5435651032209403, + "grad_norm": 0.42945531010627747, + "learning_rate": 0.0007306420887494398, + "loss": 0.3222, + "step": 9729 + }, + { + "epoch": 0.5436209738246222, + "grad_norm": 1.0346059799194336, + "learning_rate": 0.000730614074406096, + "loss": 0.4113, + "step": 9730 + }, + { + "epoch": 0.543676844428304, + "grad_norm": 0.6773257851600647, + "learning_rate": 0.0007305860600627522, + "loss": 0.4812, + "step": 9731 + }, + { + "epoch": 0.543732715031986, + "grad_norm": 0.4756525754928589, + "learning_rate": 0.0007305580457194084, + "loss": 0.3997, + "step": 9732 + }, + { + "epoch": 0.5437885856356678, + "grad_norm": 0.4849086105823517, + "learning_rate": 0.0007305300313760646, + "loss": 0.339, + "step": 9733 + }, + { + "epoch": 0.5438444562393496, + "grad_norm": 0.4701670706272125, + "learning_rate": 0.0007305020170327208, + "loss": 0.4493, + "step": 9734 + }, + { + "epoch": 0.5439003268430316, + "grad_norm": 0.9758315682411194, + "learning_rate": 0.000730474002689377, + "loss": 0.354, + "step": 9735 + }, + { + "epoch": 0.5439561974467134, + "grad_norm": 1.1000313758850098, + "learning_rate": 0.0007304459883460332, + "loss": 0.5003, + "step": 9736 + }, + { + "epoch": 0.5440120680503953, + "grad_norm": 0.5134649872779846, + "learning_rate": 0.0007304179740026894, + "loss": 0.4179, + "step": 9737 + }, + { + "epoch": 0.5440679386540772, + "grad_norm": 0.48212847113609314, + "learning_rate": 0.0007303899596593456, + "loss": 0.513, + "step": 9738 + }, + { + "epoch": 0.544123809257759, + "grad_norm": 0.4516012668609619, + "learning_rate": 0.0007303619453160018, + "loss": 0.403, + "step": 9739 + }, + { + "epoch": 0.5441796798614409, + "grad_norm": 0.8619877099990845, + "learning_rate": 0.000730333930972658, + "loss": 0.6157, + "step": 9740 + }, + { + "epoch": 0.5442355504651227, + "grad_norm": 0.5663276314735413, + "learning_rate": 0.0007303059166293142, + "loss": 0.4692, + "step": 9741 + }, + { + "epoch": 0.5442914210688047, + "grad_norm": 0.9644789099693298, + "learning_rate": 0.0007302779022859704, + "loss": 0.4737, + "step": 9742 + }, + { + "epoch": 0.5443472916724865, + "grad_norm": 0.49254584312438965, + "learning_rate": 0.0007302498879426266, + "loss": 0.4698, + "step": 9743 + }, + { + "epoch": 0.5444031622761684, + "grad_norm": 0.3280447721481323, + "learning_rate": 0.0007302218735992828, + "loss": 0.3626, + "step": 9744 + }, + { + "epoch": 0.5444590328798503, + "grad_norm": 0.5738497972488403, + "learning_rate": 0.000730193859255939, + "loss": 0.4853, + "step": 9745 + }, + { + "epoch": 0.5445149034835322, + "grad_norm": 0.5151926875114441, + "learning_rate": 0.0007301658449125953, + "loss": 0.4333, + "step": 9746 + }, + { + "epoch": 0.544570774087214, + "grad_norm": 0.5143014788627625, + "learning_rate": 0.0007301378305692515, + "loss": 0.6346, + "step": 9747 + }, + { + "epoch": 0.5446266446908958, + "grad_norm": 0.6831324696540833, + "learning_rate": 0.0007301098162259077, + "loss": 0.4514, + "step": 9748 + }, + { + "epoch": 0.5446825152945778, + "grad_norm": 0.4314040243625641, + "learning_rate": 0.0007300818018825639, + "loss": 0.4691, + "step": 9749 + }, + { + "epoch": 0.5447383858982596, + "grad_norm": 0.6449495553970337, + "learning_rate": 0.0007300537875392202, + "loss": 0.5107, + "step": 9750 + }, + { + "epoch": 0.5447942565019415, + "grad_norm": 0.4242788255214691, + "learning_rate": 0.0007300257731958763, + "loss": 0.4118, + "step": 9751 + }, + { + "epoch": 0.5448501271056234, + "grad_norm": 0.4563763439655304, + "learning_rate": 0.0007299977588525325, + "loss": 0.4521, + "step": 9752 + }, + { + "epoch": 0.5449059977093053, + "grad_norm": 0.46802380681037903, + "learning_rate": 0.0007299697445091887, + "loss": 0.5006, + "step": 9753 + }, + { + "epoch": 0.5449618683129871, + "grad_norm": 0.5624966621398926, + "learning_rate": 0.000729941730165845, + "loss": 0.539, + "step": 9754 + }, + { + "epoch": 0.545017738916669, + "grad_norm": 0.4624473750591278, + "learning_rate": 0.0007299137158225011, + "loss": 0.4584, + "step": 9755 + }, + { + "epoch": 0.5450736095203509, + "grad_norm": 0.667576253414154, + "learning_rate": 0.0007298857014791573, + "loss": 0.7307, + "step": 9756 + }, + { + "epoch": 0.5451294801240327, + "grad_norm": 0.6000552773475647, + "learning_rate": 0.0007298576871358136, + "loss": 0.4092, + "step": 9757 + }, + { + "epoch": 0.5451853507277146, + "grad_norm": 0.42138195037841797, + "learning_rate": 0.0007298296727924698, + "loss": 0.3961, + "step": 9758 + }, + { + "epoch": 0.5452412213313965, + "grad_norm": 0.7471389174461365, + "learning_rate": 0.0007298016584491259, + "loss": 0.471, + "step": 9759 + }, + { + "epoch": 0.5452970919350784, + "grad_norm": 0.49219438433647156, + "learning_rate": 0.0007297736441057821, + "loss": 0.4106, + "step": 9760 + }, + { + "epoch": 0.5453529625387602, + "grad_norm": 0.8161259293556213, + "learning_rate": 0.0007297456297624384, + "loss": 0.6895, + "step": 9761 + }, + { + "epoch": 0.5454088331424422, + "grad_norm": 0.503624677658081, + "learning_rate": 0.0007297176154190945, + "loss": 0.4054, + "step": 9762 + }, + { + "epoch": 0.545464703746124, + "grad_norm": 0.3934236168861389, + "learning_rate": 0.0007296896010757507, + "loss": 0.3627, + "step": 9763 + }, + { + "epoch": 0.5455205743498058, + "grad_norm": 2.1070053577423096, + "learning_rate": 0.0007296615867324071, + "loss": 0.3901, + "step": 9764 + }, + { + "epoch": 0.5455764449534877, + "grad_norm": 0.42272287607192993, + "learning_rate": 0.0007296335723890633, + "loss": 0.3958, + "step": 9765 + }, + { + "epoch": 0.5456323155571696, + "grad_norm": 0.4605497717857361, + "learning_rate": 0.0007296055580457194, + "loss": 0.4619, + "step": 9766 + }, + { + "epoch": 0.5456881861608515, + "grad_norm": 0.5920350551605225, + "learning_rate": 0.0007295775437023756, + "loss": 0.5272, + "step": 9767 + }, + { + "epoch": 0.5457440567645333, + "grad_norm": 0.5344292521476746, + "learning_rate": 0.0007295495293590319, + "loss": 0.5499, + "step": 9768 + }, + { + "epoch": 0.5457999273682153, + "grad_norm": 0.5737155079841614, + "learning_rate": 0.0007295215150156881, + "loss": 0.4729, + "step": 9769 + }, + { + "epoch": 0.5458557979718971, + "grad_norm": 0.4556988775730133, + "learning_rate": 0.0007294935006723442, + "loss": 0.4654, + "step": 9770 + }, + { + "epoch": 0.5459116685755789, + "grad_norm": 0.5105224847793579, + "learning_rate": 0.0007294654863290004, + "loss": 0.4783, + "step": 9771 + }, + { + "epoch": 0.5459675391792609, + "grad_norm": 0.6510880589485168, + "learning_rate": 0.0007294374719856567, + "loss": 0.4323, + "step": 9772 + }, + { + "epoch": 0.5460234097829427, + "grad_norm": 1.6172579526901245, + "learning_rate": 0.0007294094576423129, + "loss": 0.4346, + "step": 9773 + }, + { + "epoch": 0.5460792803866246, + "grad_norm": 0.5065052509307861, + "learning_rate": 0.000729381443298969, + "loss": 0.3702, + "step": 9774 + }, + { + "epoch": 0.5461351509903064, + "grad_norm": 0.32364770770072937, + "learning_rate": 0.0007293534289556253, + "loss": 0.4422, + "step": 9775 + }, + { + "epoch": 0.5461910215939884, + "grad_norm": 0.4460287392139435, + "learning_rate": 0.0007293254146122815, + "loss": 0.4228, + "step": 9776 + }, + { + "epoch": 0.5462468921976702, + "grad_norm": 2.514692544937134, + "learning_rate": 0.0007292974002689377, + "loss": 0.5253, + "step": 9777 + }, + { + "epoch": 0.546302762801352, + "grad_norm": 0.48712798953056335, + "learning_rate": 0.0007292693859255938, + "loss": 0.3642, + "step": 9778 + }, + { + "epoch": 0.546358633405034, + "grad_norm": 0.6475629210472107, + "learning_rate": 0.0007292413715822502, + "loss": 0.5096, + "step": 9779 + }, + { + "epoch": 0.5464145040087158, + "grad_norm": 0.52589350938797, + "learning_rate": 0.0007292133572389064, + "loss": 0.4873, + "step": 9780 + }, + { + "epoch": 0.5464703746123977, + "grad_norm": 0.6169883608818054, + "learning_rate": 0.0007291853428955626, + "loss": 0.5443, + "step": 9781 + }, + { + "epoch": 0.5465262452160795, + "grad_norm": 0.6212354302406311, + "learning_rate": 0.0007291573285522188, + "loss": 0.4738, + "step": 9782 + }, + { + "epoch": 0.5465821158197615, + "grad_norm": 0.6799402832984924, + "learning_rate": 0.000729129314208875, + "loss": 0.4156, + "step": 9783 + }, + { + "epoch": 0.5466379864234433, + "grad_norm": 0.5863766074180603, + "learning_rate": 0.0007291012998655312, + "loss": 0.4223, + "step": 9784 + }, + { + "epoch": 0.5466938570271251, + "grad_norm": 0.50301593542099, + "learning_rate": 0.0007290732855221873, + "loss": 0.3989, + "step": 9785 + }, + { + "epoch": 0.5467497276308071, + "grad_norm": 2.552992105484009, + "learning_rate": 0.0007290452711788436, + "loss": 0.5023, + "step": 9786 + }, + { + "epoch": 0.5468055982344889, + "grad_norm": 0.48118123412132263, + "learning_rate": 0.0007290172568354998, + "loss": 0.6074, + "step": 9787 + }, + { + "epoch": 0.5468614688381708, + "grad_norm": 0.5041949152946472, + "learning_rate": 0.000728989242492156, + "loss": 0.4695, + "step": 9788 + }, + { + "epoch": 0.5469173394418527, + "grad_norm": 0.48799315094947815, + "learning_rate": 0.0007289612281488122, + "loss": 0.5175, + "step": 9789 + }, + { + "epoch": 0.5469732100455346, + "grad_norm": 0.6362625956535339, + "learning_rate": 0.0007289332138054684, + "loss": 0.5829, + "step": 9790 + }, + { + "epoch": 0.5470290806492164, + "grad_norm": 0.41305965185165405, + "learning_rate": 0.0007289051994621246, + "loss": 0.4047, + "step": 9791 + }, + { + "epoch": 0.5470849512528982, + "grad_norm": 0.5298073291778564, + "learning_rate": 0.0007288771851187808, + "loss": 0.5736, + "step": 9792 + }, + { + "epoch": 0.5471408218565802, + "grad_norm": 0.6537981629371643, + "learning_rate": 0.000728849170775437, + "loss": 0.429, + "step": 9793 + }, + { + "epoch": 0.547196692460262, + "grad_norm": 0.843245267868042, + "learning_rate": 0.0007288211564320932, + "loss": 0.4575, + "step": 9794 + }, + { + "epoch": 0.5472525630639439, + "grad_norm": 0.6171890497207642, + "learning_rate": 0.0007287931420887494, + "loss": 0.511, + "step": 9795 + }, + { + "epoch": 0.5473084336676258, + "grad_norm": 0.40177857875823975, + "learning_rate": 0.0007287651277454058, + "loss": 0.4346, + "step": 9796 + }, + { + "epoch": 0.5473643042713077, + "grad_norm": 0.6373304128646851, + "learning_rate": 0.0007287371134020619, + "loss": 0.5031, + "step": 9797 + }, + { + "epoch": 0.5474201748749895, + "grad_norm": 0.5751506686210632, + "learning_rate": 0.0007287090990587181, + "loss": 0.5631, + "step": 9798 + }, + { + "epoch": 0.5474760454786713, + "grad_norm": 0.7778580188751221, + "learning_rate": 0.0007286810847153743, + "loss": 0.5356, + "step": 9799 + }, + { + "epoch": 0.5475319160823533, + "grad_norm": 0.4758369028568268, + "learning_rate": 0.0007286530703720306, + "loss": 0.3779, + "step": 9800 + }, + { + "epoch": 0.5475877866860351, + "grad_norm": 0.6106384992599487, + "learning_rate": 0.0007286250560286867, + "loss": 0.4241, + "step": 9801 + }, + { + "epoch": 0.547643657289717, + "grad_norm": 2.9507312774658203, + "learning_rate": 0.0007285970416853429, + "loss": 0.3552, + "step": 9802 + }, + { + "epoch": 0.5476995278933989, + "grad_norm": 0.8345327377319336, + "learning_rate": 0.0007285690273419992, + "loss": 0.6654, + "step": 9803 + }, + { + "epoch": 0.5477553984970808, + "grad_norm": 0.44680818915367126, + "learning_rate": 0.0007285410129986553, + "loss": 0.4942, + "step": 9804 + }, + { + "epoch": 0.5478112691007626, + "grad_norm": 0.8575248718261719, + "learning_rate": 0.0007285129986553115, + "loss": 0.432, + "step": 9805 + }, + { + "epoch": 0.5478671397044446, + "grad_norm": 0.9744383096694946, + "learning_rate": 0.0007284849843119677, + "loss": 0.5957, + "step": 9806 + }, + { + "epoch": 0.5479230103081264, + "grad_norm": 0.533794105052948, + "learning_rate": 0.000728456969968624, + "loss": 0.4378, + "step": 9807 + }, + { + "epoch": 0.5479788809118082, + "grad_norm": 0.4961017966270447, + "learning_rate": 0.0007284289556252801, + "loss": 0.5317, + "step": 9808 + }, + { + "epoch": 0.5480347515154901, + "grad_norm": 0.6119482517242432, + "learning_rate": 0.0007284009412819363, + "loss": 0.4266, + "step": 9809 + }, + { + "epoch": 0.548090622119172, + "grad_norm": 1.2448484897613525, + "learning_rate": 0.0007283729269385925, + "loss": 0.5047, + "step": 9810 + }, + { + "epoch": 0.5481464927228539, + "grad_norm": 0.6708424687385559, + "learning_rate": 0.0007283449125952488, + "loss": 0.4812, + "step": 9811 + }, + { + "epoch": 0.5482023633265357, + "grad_norm": 0.4309360682964325, + "learning_rate": 0.0007283168982519049, + "loss": 0.4643, + "step": 9812 + }, + { + "epoch": 0.5482582339302177, + "grad_norm": 0.984844446182251, + "learning_rate": 0.0007282888839085611, + "loss": 0.6491, + "step": 9813 + }, + { + "epoch": 0.5483141045338995, + "grad_norm": 0.770213782787323, + "learning_rate": 0.0007282608695652175, + "loss": 0.2937, + "step": 9814 + }, + { + "epoch": 0.5483699751375813, + "grad_norm": 1.7408465147018433, + "learning_rate": 0.0007282328552218737, + "loss": 0.5606, + "step": 9815 + }, + { + "epoch": 0.5484258457412632, + "grad_norm": 1.494553804397583, + "learning_rate": 0.0007282048408785298, + "loss": 0.4494, + "step": 9816 + }, + { + "epoch": 0.5484817163449451, + "grad_norm": 0.6715970039367676, + "learning_rate": 0.000728176826535186, + "loss": 0.4312, + "step": 9817 + }, + { + "epoch": 0.548537586948627, + "grad_norm": 0.66514652967453, + "learning_rate": 0.0007281488121918423, + "loss": 0.3376, + "step": 9818 + }, + { + "epoch": 0.5485934575523088, + "grad_norm": 0.7303655743598938, + "learning_rate": 0.0007281207978484985, + "loss": 0.4282, + "step": 9819 + }, + { + "epoch": 0.5486493281559908, + "grad_norm": 0.45677444338798523, + "learning_rate": 0.0007280927835051546, + "loss": 0.4714, + "step": 9820 + }, + { + "epoch": 0.5487051987596726, + "grad_norm": 0.4749726951122284, + "learning_rate": 0.0007280647691618109, + "loss": 0.4839, + "step": 9821 + }, + { + "epoch": 0.5487610693633544, + "grad_norm": 0.591911792755127, + "learning_rate": 0.0007280367548184671, + "loss": 0.4636, + "step": 9822 + }, + { + "epoch": 0.5488169399670364, + "grad_norm": 1.8650214672088623, + "learning_rate": 0.0007280087404751233, + "loss": 0.5678, + "step": 9823 + }, + { + "epoch": 0.5488728105707182, + "grad_norm": 0.7053783535957336, + "learning_rate": 0.0007279807261317794, + "loss": 0.4669, + "step": 9824 + }, + { + "epoch": 0.5489286811744001, + "grad_norm": 0.4603267312049866, + "learning_rate": 0.0007279527117884357, + "loss": 0.4321, + "step": 9825 + }, + { + "epoch": 0.5489845517780819, + "grad_norm": 0.5262329578399658, + "learning_rate": 0.0007279246974450919, + "loss": 0.4508, + "step": 9826 + }, + { + "epoch": 0.5490404223817639, + "grad_norm": 0.5568184852600098, + "learning_rate": 0.000727896683101748, + "loss": 0.531, + "step": 9827 + }, + { + "epoch": 0.5490962929854457, + "grad_norm": 2.6121222972869873, + "learning_rate": 0.0007278686687584043, + "loss": 0.6203, + "step": 9828 + }, + { + "epoch": 0.5491521635891276, + "grad_norm": 1.2325553894042969, + "learning_rate": 0.0007278406544150606, + "loss": 0.4441, + "step": 9829 + }, + { + "epoch": 0.5492080341928095, + "grad_norm": 1.326191782951355, + "learning_rate": 0.0007278126400717168, + "loss": 0.3837, + "step": 9830 + }, + { + "epoch": 0.5492639047964913, + "grad_norm": 1.8238914012908936, + "learning_rate": 0.0007277846257283729, + "loss": 0.5026, + "step": 9831 + }, + { + "epoch": 0.5493197754001732, + "grad_norm": 0.4605728089809418, + "learning_rate": 0.0007277566113850292, + "loss": 0.4617, + "step": 9832 + }, + { + "epoch": 0.549375646003855, + "grad_norm": 0.4448632597923279, + "learning_rate": 0.0007277285970416854, + "loss": 0.4281, + "step": 9833 + }, + { + "epoch": 0.549431516607537, + "grad_norm": 1.00554358959198, + "learning_rate": 0.0007277005826983416, + "loss": 0.4719, + "step": 9834 + }, + { + "epoch": 0.5494873872112188, + "grad_norm": 0.5083926320075989, + "learning_rate": 0.0007276725683549978, + "loss": 0.4678, + "step": 9835 + }, + { + "epoch": 0.5495432578149007, + "grad_norm": 0.3770988881587982, + "learning_rate": 0.000727644554011654, + "loss": 0.3476, + "step": 9836 + }, + { + "epoch": 0.5495991284185826, + "grad_norm": 0.5761573910713196, + "learning_rate": 0.0007276165396683102, + "loss": 0.5224, + "step": 9837 + }, + { + "epoch": 0.5496549990222644, + "grad_norm": 1.7116999626159668, + "learning_rate": 0.0007275885253249664, + "loss": 0.5924, + "step": 9838 + }, + { + "epoch": 0.5497108696259463, + "grad_norm": 1.166099190711975, + "learning_rate": 0.0007275605109816226, + "loss": 0.5475, + "step": 9839 + }, + { + "epoch": 0.5497667402296281, + "grad_norm": 1.0780997276306152, + "learning_rate": 0.0007275324966382788, + "loss": 0.3415, + "step": 9840 + }, + { + "epoch": 0.5498226108333101, + "grad_norm": 0.5179852247238159, + "learning_rate": 0.000727504482294935, + "loss": 0.5109, + "step": 9841 + }, + { + "epoch": 0.5498784814369919, + "grad_norm": 0.4621276259422302, + "learning_rate": 0.0007274764679515913, + "loss": 0.434, + "step": 9842 + }, + { + "epoch": 0.5499343520406738, + "grad_norm": 0.7363477945327759, + "learning_rate": 0.0007274484536082474, + "loss": 0.4912, + "step": 9843 + }, + { + "epoch": 0.5499902226443557, + "grad_norm": 0.4029199779033661, + "learning_rate": 0.0007274204392649036, + "loss": 0.369, + "step": 9844 + }, + { + "epoch": 0.5500460932480375, + "grad_norm": 0.48415523767471313, + "learning_rate": 0.0007273924249215598, + "loss": 0.438, + "step": 9845 + }, + { + "epoch": 0.5501019638517194, + "grad_norm": 0.6887811422348022, + "learning_rate": 0.000727364410578216, + "loss": 0.5116, + "step": 9846 + }, + { + "epoch": 0.5501578344554013, + "grad_norm": 0.4610520005226135, + "learning_rate": 0.0007273363962348723, + "loss": 0.4494, + "step": 9847 + }, + { + "epoch": 0.5502137050590832, + "grad_norm": 0.4653385579586029, + "learning_rate": 0.0007273083818915285, + "loss": 0.3621, + "step": 9848 + }, + { + "epoch": 0.550269575662765, + "grad_norm": 0.41656070947647095, + "learning_rate": 0.0007272803675481847, + "loss": 0.4849, + "step": 9849 + }, + { + "epoch": 0.5503254462664469, + "grad_norm": 1.816779613494873, + "learning_rate": 0.0007272523532048409, + "loss": 0.4708, + "step": 9850 + }, + { + "epoch": 0.5503813168701288, + "grad_norm": 0.5495056509971619, + "learning_rate": 0.0007272243388614971, + "loss": 0.4832, + "step": 9851 + }, + { + "epoch": 0.5504371874738107, + "grad_norm": 0.3510303497314453, + "learning_rate": 0.0007271963245181533, + "loss": 0.3989, + "step": 9852 + }, + { + "epoch": 0.5504930580774925, + "grad_norm": 0.5764368772506714, + "learning_rate": 0.0007271683101748096, + "loss": 0.5253, + "step": 9853 + }, + { + "epoch": 0.5505489286811744, + "grad_norm": 0.43032360076904297, + "learning_rate": 0.0007271402958314657, + "loss": 0.4471, + "step": 9854 + }, + { + "epoch": 0.5506047992848563, + "grad_norm": 0.3864692151546478, + "learning_rate": 0.0007271122814881219, + "loss": 0.3749, + "step": 9855 + }, + { + "epoch": 0.5506606698885381, + "grad_norm": 0.4942381978034973, + "learning_rate": 0.0007270842671447781, + "loss": 0.4739, + "step": 9856 + }, + { + "epoch": 0.55071654049222, + "grad_norm": 0.5141517519950867, + "learning_rate": 0.0007270562528014344, + "loss": 0.4637, + "step": 9857 + }, + { + "epoch": 0.5507724110959019, + "grad_norm": 0.5845747590065002, + "learning_rate": 0.0007270282384580905, + "loss": 0.4432, + "step": 9858 + }, + { + "epoch": 0.5508282816995838, + "grad_norm": 0.47151535749435425, + "learning_rate": 0.0007270002241147467, + "loss": 0.4026, + "step": 9859 + }, + { + "epoch": 0.5508841523032656, + "grad_norm": 0.537710964679718, + "learning_rate": 0.000726972209771403, + "loss": 0.5146, + "step": 9860 + }, + { + "epoch": 0.5509400229069475, + "grad_norm": 0.39292383193969727, + "learning_rate": 0.0007269441954280592, + "loss": 0.3124, + "step": 9861 + }, + { + "epoch": 0.5509958935106294, + "grad_norm": 0.5549148321151733, + "learning_rate": 0.0007269161810847153, + "loss": 0.4059, + "step": 9862 + }, + { + "epoch": 0.5510517641143112, + "grad_norm": 3.743234872817993, + "learning_rate": 0.0007268881667413715, + "loss": 0.7519, + "step": 9863 + }, + { + "epoch": 0.5511076347179932, + "grad_norm": 0.9088842868804932, + "learning_rate": 0.0007268601523980279, + "loss": 0.448, + "step": 9864 + }, + { + "epoch": 0.551163505321675, + "grad_norm": 0.5757520794868469, + "learning_rate": 0.0007268321380546841, + "loss": 0.4054, + "step": 9865 + }, + { + "epoch": 0.5512193759253569, + "grad_norm": 1.3318277597427368, + "learning_rate": 0.0007268041237113402, + "loss": 0.5201, + "step": 9866 + }, + { + "epoch": 0.5512752465290387, + "grad_norm": 0.5328368544578552, + "learning_rate": 0.0007267761093679965, + "loss": 0.4954, + "step": 9867 + }, + { + "epoch": 0.5513311171327206, + "grad_norm": 0.4341873824596405, + "learning_rate": 0.0007267480950246527, + "loss": 0.4354, + "step": 9868 + }, + { + "epoch": 0.5513869877364025, + "grad_norm": 0.4187398850917816, + "learning_rate": 0.0007267200806813088, + "loss": 0.4372, + "step": 9869 + }, + { + "epoch": 0.5514428583400843, + "grad_norm": 0.5710111260414124, + "learning_rate": 0.000726692066337965, + "loss": 0.5621, + "step": 9870 + }, + { + "epoch": 0.5514987289437663, + "grad_norm": 0.8056668043136597, + "learning_rate": 0.0007266640519946213, + "loss": 0.3587, + "step": 9871 + }, + { + "epoch": 0.5515545995474481, + "grad_norm": 0.7718157172203064, + "learning_rate": 0.0007266360376512775, + "loss": 0.5185, + "step": 9872 + }, + { + "epoch": 0.55161047015113, + "grad_norm": 0.4899265468120575, + "learning_rate": 0.0007266080233079336, + "loss": 0.5048, + "step": 9873 + }, + { + "epoch": 0.5516663407548118, + "grad_norm": 0.8002147078514099, + "learning_rate": 0.0007265800089645899, + "loss": 0.3711, + "step": 9874 + }, + { + "epoch": 0.5517222113584938, + "grad_norm": 0.4433348476886749, + "learning_rate": 0.0007265519946212461, + "loss": 0.4658, + "step": 9875 + }, + { + "epoch": 0.5517780819621756, + "grad_norm": 1.8628441095352173, + "learning_rate": 0.0007265239802779023, + "loss": 0.392, + "step": 9876 + }, + { + "epoch": 0.5518339525658574, + "grad_norm": 3.88790225982666, + "learning_rate": 0.0007264959659345584, + "loss": 0.5278, + "step": 9877 + }, + { + "epoch": 0.5518898231695394, + "grad_norm": 0.4782730042934418, + "learning_rate": 0.0007264679515912147, + "loss": 0.4328, + "step": 9878 + }, + { + "epoch": 0.5519456937732212, + "grad_norm": 0.700740396976471, + "learning_rate": 0.000726439937247871, + "loss": 0.454, + "step": 9879 + }, + { + "epoch": 0.5520015643769031, + "grad_norm": 1.0143311023712158, + "learning_rate": 0.0007264119229045272, + "loss": 0.506, + "step": 9880 + }, + { + "epoch": 0.552057434980585, + "grad_norm": 0.461122989654541, + "learning_rate": 0.0007263839085611833, + "loss": 0.4915, + "step": 9881 + }, + { + "epoch": 0.5521133055842669, + "grad_norm": 1.4650269746780396, + "learning_rate": 0.0007263558942178396, + "loss": 0.5026, + "step": 9882 + }, + { + "epoch": 0.5521691761879487, + "grad_norm": 0.44180789589881897, + "learning_rate": 0.0007263278798744958, + "loss": 0.4246, + "step": 9883 + }, + { + "epoch": 0.5522250467916305, + "grad_norm": 0.5413908958435059, + "learning_rate": 0.000726299865531152, + "loss": 0.5016, + "step": 9884 + }, + { + "epoch": 0.5522809173953125, + "grad_norm": 0.587480366230011, + "learning_rate": 0.0007262718511878082, + "loss": 0.4137, + "step": 9885 + }, + { + "epoch": 0.5523367879989943, + "grad_norm": 0.41121259331703186, + "learning_rate": 0.0007262438368444644, + "loss": 0.4881, + "step": 9886 + }, + { + "epoch": 0.5523926586026762, + "grad_norm": 0.3621082901954651, + "learning_rate": 0.0007262158225011206, + "loss": 0.3418, + "step": 9887 + }, + { + "epoch": 0.5524485292063581, + "grad_norm": 0.4883413016796112, + "learning_rate": 0.0007261878081577767, + "loss": 0.3933, + "step": 9888 + }, + { + "epoch": 0.55250439981004, + "grad_norm": 0.6519923210144043, + "learning_rate": 0.000726159793814433, + "loss": 0.3878, + "step": 9889 + }, + { + "epoch": 0.5525602704137218, + "grad_norm": 0.49083226919174194, + "learning_rate": 0.0007261317794710892, + "loss": 0.4734, + "step": 9890 + }, + { + "epoch": 0.5526161410174036, + "grad_norm": 0.4407425820827484, + "learning_rate": 0.0007261037651277454, + "loss": 0.4633, + "step": 9891 + }, + { + "epoch": 0.5526720116210856, + "grad_norm": 0.7256817817687988, + "learning_rate": 0.0007260757507844016, + "loss": 0.5545, + "step": 9892 + }, + { + "epoch": 0.5527278822247674, + "grad_norm": 0.5588866472244263, + "learning_rate": 0.0007260477364410578, + "loss": 0.4993, + "step": 9893 + }, + { + "epoch": 0.5527837528284493, + "grad_norm": 1.104570746421814, + "learning_rate": 0.000726019722097714, + "loss": 0.415, + "step": 9894 + }, + { + "epoch": 0.5528396234321312, + "grad_norm": 0.5324800610542297, + "learning_rate": 0.0007259917077543702, + "loss": 0.3802, + "step": 9895 + }, + { + "epoch": 0.5528954940358131, + "grad_norm": 3.8082263469696045, + "learning_rate": 0.0007259636934110264, + "loss": 0.4293, + "step": 9896 + }, + { + "epoch": 0.5529513646394949, + "grad_norm": 0.704369843006134, + "learning_rate": 0.0007259356790676827, + "loss": 0.4928, + "step": 9897 + }, + { + "epoch": 0.5530072352431769, + "grad_norm": 0.6472663879394531, + "learning_rate": 0.0007259076647243389, + "loss": 0.4814, + "step": 9898 + }, + { + "epoch": 0.5530631058468587, + "grad_norm": 0.5148155689239502, + "learning_rate": 0.0007258796503809952, + "loss": 0.4598, + "step": 9899 + }, + { + "epoch": 0.5531189764505405, + "grad_norm": 0.823942244052887, + "learning_rate": 0.0007258516360376513, + "loss": 0.5071, + "step": 9900 + }, + { + "epoch": 0.5531748470542224, + "grad_norm": 0.8869868516921997, + "learning_rate": 0.0007258236216943075, + "loss": 0.4135, + "step": 9901 + }, + { + "epoch": 0.5532307176579043, + "grad_norm": 0.4735725224018097, + "learning_rate": 0.0007257956073509637, + "loss": 0.4637, + "step": 9902 + }, + { + "epoch": 0.5532865882615862, + "grad_norm": 0.4970542788505554, + "learning_rate": 0.00072576759300762, + "loss": 0.4039, + "step": 9903 + }, + { + "epoch": 0.553342458865268, + "grad_norm": 2.3057875633239746, + "learning_rate": 0.0007257395786642761, + "loss": 0.5674, + "step": 9904 + }, + { + "epoch": 0.55339832946895, + "grad_norm": 0.703972578048706, + "learning_rate": 0.0007257115643209323, + "loss": 0.4847, + "step": 9905 + }, + { + "epoch": 0.5534542000726318, + "grad_norm": 0.6732432246208191, + "learning_rate": 0.0007256835499775886, + "loss": 0.5243, + "step": 9906 + }, + { + "epoch": 0.5535100706763136, + "grad_norm": 0.41702887415885925, + "learning_rate": 0.0007256555356342448, + "loss": 0.3909, + "step": 9907 + }, + { + "epoch": 0.5535659412799955, + "grad_norm": 0.5144908428192139, + "learning_rate": 0.0007256275212909009, + "loss": 0.4393, + "step": 9908 + }, + { + "epoch": 0.5536218118836774, + "grad_norm": 0.5285742282867432, + "learning_rate": 0.0007255995069475571, + "loss": 0.4877, + "step": 9909 + }, + { + "epoch": 0.5536776824873593, + "grad_norm": 0.39467111229896545, + "learning_rate": 0.0007255714926042134, + "loss": 0.4815, + "step": 9910 + }, + { + "epoch": 0.5537335530910411, + "grad_norm": 0.7021602988243103, + "learning_rate": 0.0007255434782608695, + "loss": 0.6569, + "step": 9911 + }, + { + "epoch": 0.5537894236947231, + "grad_norm": 0.5013124942779541, + "learning_rate": 0.0007255154639175257, + "loss": 0.5219, + "step": 9912 + }, + { + "epoch": 0.5538452942984049, + "grad_norm": 1.5304629802703857, + "learning_rate": 0.000725487449574182, + "loss": 0.5015, + "step": 9913 + }, + { + "epoch": 0.5539011649020867, + "grad_norm": 0.5397960543632507, + "learning_rate": 0.0007254594352308383, + "loss": 0.5085, + "step": 9914 + }, + { + "epoch": 0.5539570355057687, + "grad_norm": 1.195281744003296, + "learning_rate": 0.0007254314208874944, + "loss": 0.511, + "step": 9915 + }, + { + "epoch": 0.5540129061094505, + "grad_norm": 0.4987823963165283, + "learning_rate": 0.0007254034065441506, + "loss": 0.5487, + "step": 9916 + }, + { + "epoch": 0.5540687767131324, + "grad_norm": 0.42418384552001953, + "learning_rate": 0.0007253753922008069, + "loss": 0.4371, + "step": 9917 + }, + { + "epoch": 0.5541246473168142, + "grad_norm": 0.579489529132843, + "learning_rate": 0.0007253473778574631, + "loss": 0.403, + "step": 9918 + }, + { + "epoch": 0.5541805179204962, + "grad_norm": 0.5557631850242615, + "learning_rate": 0.0007253193635141192, + "loss": 0.3722, + "step": 9919 + }, + { + "epoch": 0.554236388524178, + "grad_norm": 0.8017315864562988, + "learning_rate": 0.0007252913491707754, + "loss": 0.6121, + "step": 9920 + }, + { + "epoch": 0.5542922591278598, + "grad_norm": 4.215581893920898, + "learning_rate": 0.0007252633348274317, + "loss": 0.4468, + "step": 9921 + }, + { + "epoch": 0.5543481297315418, + "grad_norm": 0.497445285320282, + "learning_rate": 0.0007252353204840879, + "loss": 0.3771, + "step": 9922 + }, + { + "epoch": 0.5544040003352236, + "grad_norm": 0.5483517646789551, + "learning_rate": 0.000725207306140744, + "loss": 0.4351, + "step": 9923 + }, + { + "epoch": 0.5544598709389055, + "grad_norm": 0.7252764701843262, + "learning_rate": 0.0007251792917974003, + "loss": 0.673, + "step": 9924 + }, + { + "epoch": 0.5545157415425873, + "grad_norm": 0.4583989679813385, + "learning_rate": 0.0007251512774540565, + "loss": 0.4357, + "step": 9925 + }, + { + "epoch": 0.5545716121462693, + "grad_norm": 0.818831741809845, + "learning_rate": 0.0007251232631107127, + "loss": 0.5967, + "step": 9926 + }, + { + "epoch": 0.5546274827499511, + "grad_norm": 0.5691839456558228, + "learning_rate": 0.0007250952487673688, + "loss": 0.3451, + "step": 9927 + }, + { + "epoch": 0.554683353353633, + "grad_norm": 0.6864220499992371, + "learning_rate": 0.0007250672344240251, + "loss": 0.6234, + "step": 9928 + }, + { + "epoch": 0.5547392239573149, + "grad_norm": 0.6954836845397949, + "learning_rate": 0.0007250392200806813, + "loss": 0.5909, + "step": 9929 + }, + { + "epoch": 0.5547950945609967, + "grad_norm": 0.7870447635650635, + "learning_rate": 0.0007250112057373376, + "loss": 0.3428, + "step": 9930 + }, + { + "epoch": 0.5548509651646786, + "grad_norm": 0.44840216636657715, + "learning_rate": 0.0007249831913939938, + "loss": 0.4363, + "step": 9931 + }, + { + "epoch": 0.5549068357683605, + "grad_norm": 0.45312732458114624, + "learning_rate": 0.00072495517705065, + "loss": 0.484, + "step": 9932 + }, + { + "epoch": 0.5549627063720424, + "grad_norm": 0.5259836912155151, + "learning_rate": 0.0007249271627073062, + "loss": 0.4705, + "step": 9933 + }, + { + "epoch": 0.5550185769757242, + "grad_norm": 0.5031245946884155, + "learning_rate": 0.0007248991483639623, + "loss": 0.4629, + "step": 9934 + }, + { + "epoch": 0.555074447579406, + "grad_norm": 0.5759528279304504, + "learning_rate": 0.0007248711340206186, + "loss": 0.489, + "step": 9935 + }, + { + "epoch": 0.555130318183088, + "grad_norm": 0.545530378818512, + "learning_rate": 0.0007248431196772748, + "loss": 0.394, + "step": 9936 + }, + { + "epoch": 0.5551861887867698, + "grad_norm": 0.42244166135787964, + "learning_rate": 0.000724815105333931, + "loss": 0.4251, + "step": 9937 + }, + { + "epoch": 0.5552420593904517, + "grad_norm": 0.6812238693237305, + "learning_rate": 0.0007247870909905872, + "loss": 0.471, + "step": 9938 + }, + { + "epoch": 0.5552979299941336, + "grad_norm": 0.6979041695594788, + "learning_rate": 0.0007247590766472434, + "loss": 0.5834, + "step": 9939 + }, + { + "epoch": 0.5553538005978155, + "grad_norm": 0.5384186506271362, + "learning_rate": 0.0007247310623038996, + "loss": 0.5013, + "step": 9940 + }, + { + "epoch": 0.5554096712014973, + "grad_norm": 0.41319090127944946, + "learning_rate": 0.0007247030479605558, + "loss": 0.4233, + "step": 9941 + }, + { + "epoch": 0.5554655418051792, + "grad_norm": 0.7267858386039734, + "learning_rate": 0.000724675033617212, + "loss": 0.3396, + "step": 9942 + }, + { + "epoch": 0.5555214124088611, + "grad_norm": 0.6083621382713318, + "learning_rate": 0.0007246470192738682, + "loss": 0.6684, + "step": 9943 + }, + { + "epoch": 0.5555772830125429, + "grad_norm": 1.9772441387176514, + "learning_rate": 0.0007246190049305244, + "loss": 0.4104, + "step": 9944 + }, + { + "epoch": 0.5556331536162248, + "grad_norm": 0.49388474225997925, + "learning_rate": 0.0007245909905871808, + "loss": 0.4661, + "step": 9945 + }, + { + "epoch": 0.5556890242199067, + "grad_norm": 0.4986811578273773, + "learning_rate": 0.0007245629762438368, + "loss": 0.5298, + "step": 9946 + }, + { + "epoch": 0.5557448948235886, + "grad_norm": 1.3128538131713867, + "learning_rate": 0.000724534961900493, + "loss": 0.4763, + "step": 9947 + }, + { + "epoch": 0.5558007654272704, + "grad_norm": 0.8648236989974976, + "learning_rate": 0.0007245069475571493, + "loss": 0.6257, + "step": 9948 + }, + { + "epoch": 0.5558566360309524, + "grad_norm": 1.8598947525024414, + "learning_rate": 0.0007244789332138056, + "loss": 0.456, + "step": 9949 + }, + { + "epoch": 0.5559125066346342, + "grad_norm": 0.8586663007736206, + "learning_rate": 0.0007244509188704617, + "loss": 0.3367, + "step": 9950 + }, + { + "epoch": 0.555968377238316, + "grad_norm": 0.6067330837249756, + "learning_rate": 0.0007244229045271179, + "loss": 0.4963, + "step": 9951 + }, + { + "epoch": 0.5560242478419979, + "grad_norm": 0.6674944758415222, + "learning_rate": 0.0007243948901837741, + "loss": 0.4228, + "step": 9952 + }, + { + "epoch": 0.5560801184456798, + "grad_norm": 0.5622943639755249, + "learning_rate": 0.0007243668758404303, + "loss": 0.5236, + "step": 9953 + }, + { + "epoch": 0.5561359890493617, + "grad_norm": 0.9979636669158936, + "learning_rate": 0.0007243388614970865, + "loss": 0.4357, + "step": 9954 + }, + { + "epoch": 0.5561918596530435, + "grad_norm": 0.5174521803855896, + "learning_rate": 0.0007243108471537427, + "loss": 0.4779, + "step": 9955 + }, + { + "epoch": 0.5562477302567255, + "grad_norm": 0.37243643403053284, + "learning_rate": 0.000724282832810399, + "loss": 0.2899, + "step": 9956 + }, + { + "epoch": 0.5563036008604073, + "grad_norm": 0.5598762035369873, + "learning_rate": 0.0007242548184670551, + "loss": 0.4505, + "step": 9957 + }, + { + "epoch": 0.5563594714640892, + "grad_norm": 0.6168156862258911, + "learning_rate": 0.0007242268041237113, + "loss": 0.3568, + "step": 9958 + }, + { + "epoch": 0.556415342067771, + "grad_norm": 0.4670690894126892, + "learning_rate": 0.0007241987897803675, + "loss": 0.4841, + "step": 9959 + }, + { + "epoch": 0.5564712126714529, + "grad_norm": 0.3493165075778961, + "learning_rate": 0.0007241707754370238, + "loss": 0.3817, + "step": 9960 + }, + { + "epoch": 0.5565270832751348, + "grad_norm": 0.5114171504974365, + "learning_rate": 0.0007241427610936799, + "loss": 0.5596, + "step": 9961 + }, + { + "epoch": 0.5565829538788166, + "grad_norm": 4.46920919418335, + "learning_rate": 0.0007241147467503361, + "loss": 0.397, + "step": 9962 + }, + { + "epoch": 0.5566388244824986, + "grad_norm": 0.41902801394462585, + "learning_rate": 0.0007240867324069925, + "loss": 0.4626, + "step": 9963 + }, + { + "epoch": 0.5566946950861804, + "grad_norm": 0.5118802189826965, + "learning_rate": 0.0007240587180636487, + "loss": 0.4791, + "step": 9964 + }, + { + "epoch": 0.5567505656898623, + "grad_norm": 0.39250606298446655, + "learning_rate": 0.0007240307037203048, + "loss": 0.3622, + "step": 9965 + }, + { + "epoch": 0.5568064362935442, + "grad_norm": 0.5961163640022278, + "learning_rate": 0.000724002689376961, + "loss": 0.2988, + "step": 9966 + }, + { + "epoch": 0.556862306897226, + "grad_norm": 1.5597953796386719, + "learning_rate": 0.0007239746750336173, + "loss": 0.5566, + "step": 9967 + }, + { + "epoch": 0.5569181775009079, + "grad_norm": 0.659925639629364, + "learning_rate": 0.0007239466606902735, + "loss": 0.5449, + "step": 9968 + }, + { + "epoch": 0.5569740481045897, + "grad_norm": 0.48996180295944214, + "learning_rate": 0.0007239186463469296, + "loss": 0.3747, + "step": 9969 + }, + { + "epoch": 0.5570299187082717, + "grad_norm": 0.6632741093635559, + "learning_rate": 0.0007238906320035859, + "loss": 0.4826, + "step": 9970 + }, + { + "epoch": 0.5570857893119535, + "grad_norm": 0.4711921513080597, + "learning_rate": 0.0007238626176602421, + "loss": 0.4396, + "step": 9971 + }, + { + "epoch": 0.5571416599156354, + "grad_norm": 3.912112236022949, + "learning_rate": 0.0007238346033168983, + "loss": 0.4712, + "step": 9972 + }, + { + "epoch": 0.5571975305193173, + "grad_norm": 4.826351165771484, + "learning_rate": 0.0007238065889735544, + "loss": 0.444, + "step": 9973 + }, + { + "epoch": 0.5572534011229991, + "grad_norm": 0.5131835341453552, + "learning_rate": 0.0007237785746302107, + "loss": 0.4658, + "step": 9974 + }, + { + "epoch": 0.557309271726681, + "grad_norm": 0.4947875738143921, + "learning_rate": 0.0007237505602868669, + "loss": 0.4615, + "step": 9975 + }, + { + "epoch": 0.5573651423303628, + "grad_norm": 2.0987648963928223, + "learning_rate": 0.000723722545943523, + "loss": 0.498, + "step": 9976 + }, + { + "epoch": 0.5574210129340448, + "grad_norm": 0.48486408591270447, + "learning_rate": 0.0007236945316001793, + "loss": 0.4406, + "step": 9977 + }, + { + "epoch": 0.5574768835377266, + "grad_norm": 0.7327888607978821, + "learning_rate": 0.0007236665172568355, + "loss": 0.549, + "step": 9978 + }, + { + "epoch": 0.5575327541414085, + "grad_norm": 0.5146079063415527, + "learning_rate": 0.0007236385029134917, + "loss": 0.3884, + "step": 9979 + }, + { + "epoch": 0.5575886247450904, + "grad_norm": 0.5052695870399475, + "learning_rate": 0.0007236104885701478, + "loss": 0.3911, + "step": 9980 + }, + { + "epoch": 0.5576444953487723, + "grad_norm": 0.5994778275489807, + "learning_rate": 0.0007235824742268042, + "loss": 0.4613, + "step": 9981 + }, + { + "epoch": 0.5577003659524541, + "grad_norm": 0.4906054437160492, + "learning_rate": 0.0007235544598834604, + "loss": 0.462, + "step": 9982 + }, + { + "epoch": 0.557756236556136, + "grad_norm": 0.4248184263706207, + "learning_rate": 0.0007235264455401166, + "loss": 0.3256, + "step": 9983 + }, + { + "epoch": 0.5578121071598179, + "grad_norm": 1.3460263013839722, + "learning_rate": 0.0007234984311967727, + "loss": 0.405, + "step": 9984 + }, + { + "epoch": 0.5578679777634997, + "grad_norm": 0.3758232593536377, + "learning_rate": 0.000723470416853429, + "loss": 0.4355, + "step": 9985 + }, + { + "epoch": 0.5579238483671816, + "grad_norm": 0.7077274918556213, + "learning_rate": 0.0007234424025100852, + "loss": 0.4536, + "step": 9986 + }, + { + "epoch": 0.5579797189708635, + "grad_norm": 0.47109174728393555, + "learning_rate": 0.0007234143881667414, + "loss": 0.3694, + "step": 9987 + }, + { + "epoch": 0.5580355895745454, + "grad_norm": 2.611955165863037, + "learning_rate": 0.0007233863738233976, + "loss": 0.5309, + "step": 9988 + }, + { + "epoch": 0.5580914601782272, + "grad_norm": 0.49059534072875977, + "learning_rate": 0.0007233583594800538, + "loss": 0.4683, + "step": 9989 + }, + { + "epoch": 0.5581473307819091, + "grad_norm": 0.5528339147567749, + "learning_rate": 0.00072333034513671, + "loss": 0.4536, + "step": 9990 + }, + { + "epoch": 0.558203201385591, + "grad_norm": 0.6858622431755066, + "learning_rate": 0.0007233023307933662, + "loss": 0.4604, + "step": 9991 + }, + { + "epoch": 0.5582590719892728, + "grad_norm": 1.3289003372192383, + "learning_rate": 0.0007232743164500224, + "loss": 0.605, + "step": 9992 + }, + { + "epoch": 0.5583149425929547, + "grad_norm": 1.362453818321228, + "learning_rate": 0.0007232463021066786, + "loss": 0.483, + "step": 9993 + }, + { + "epoch": 0.5583708131966366, + "grad_norm": 0.7851483225822449, + "learning_rate": 0.0007232182877633348, + "loss": 0.5482, + "step": 9994 + }, + { + "epoch": 0.5584266838003185, + "grad_norm": 0.7497435808181763, + "learning_rate": 0.000723190273419991, + "loss": 0.5379, + "step": 9995 + }, + { + "epoch": 0.5584825544040003, + "grad_norm": 0.5124725699424744, + "learning_rate": 0.0007231622590766472, + "loss": 0.4626, + "step": 9996 + }, + { + "epoch": 0.5585384250076822, + "grad_norm": 3.6004581451416016, + "learning_rate": 0.0007231342447333035, + "loss": 0.4299, + "step": 9997 + }, + { + "epoch": 0.5585942956113641, + "grad_norm": 10.910412788391113, + "learning_rate": 0.0007231062303899597, + "loss": 0.4969, + "step": 9998 + }, + { + "epoch": 0.5586501662150459, + "grad_norm": 0.9458263516426086, + "learning_rate": 0.0007230782160466159, + "loss": 0.5231, + "step": 9999 + }, + { + "epoch": 0.5587060368187279, + "grad_norm": 0.5028937458992004, + "learning_rate": 0.0007230502017032721, + "loss": 0.5848, + "step": 10000 + }, + { + "epoch": 0.5587060368187279, + "eval_cer": 0.0920155368618594, + "eval_loss": 0.34216660261154175, + "eval_runtime": 56.2776, + "eval_samples_per_second": 80.636, + "eval_steps_per_second": 5.046, + "eval_wer": 0.36515739783804924, + "step": 10000 + }, + { + "epoch": 0.5587619074224097, + "grad_norm": 0.45057550072669983, + "learning_rate": 0.0007230221873599283, + "loss": 0.4348, + "step": 10001 + }, + { + "epoch": 0.5588177780260916, + "grad_norm": 0.444240927696228, + "learning_rate": 0.0007229941730165846, + "loss": 0.4488, + "step": 10002 + }, + { + "epoch": 0.5588736486297734, + "grad_norm": 0.6431654095649719, + "learning_rate": 0.0007229661586732407, + "loss": 0.4108, + "step": 10003 + }, + { + "epoch": 0.5589295192334554, + "grad_norm": 0.7787125706672668, + "learning_rate": 0.0007229381443298969, + "loss": 0.4106, + "step": 10004 + }, + { + "epoch": 0.5589853898371372, + "grad_norm": 0.48144540190696716, + "learning_rate": 0.0007229101299865531, + "loss": 0.5172, + "step": 10005 + }, + { + "epoch": 0.559041260440819, + "grad_norm": 0.49830248951911926, + "learning_rate": 0.0007228821156432094, + "loss": 0.3771, + "step": 10006 + }, + { + "epoch": 0.559097131044501, + "grad_norm": 0.766761839389801, + "learning_rate": 0.0007228541012998655, + "loss": 0.4684, + "step": 10007 + }, + { + "epoch": 0.5591530016481828, + "grad_norm": 0.9281830787658691, + "learning_rate": 0.0007228260869565217, + "loss": 0.4969, + "step": 10008 + }, + { + "epoch": 0.5592088722518647, + "grad_norm": 0.5067682862281799, + "learning_rate": 0.000722798072613178, + "loss": 0.4058, + "step": 10009 + }, + { + "epoch": 0.5592647428555465, + "grad_norm": 0.5489600896835327, + "learning_rate": 0.0007227700582698342, + "loss": 0.4467, + "step": 10010 + }, + { + "epoch": 0.5593206134592285, + "grad_norm": 0.5160000920295715, + "learning_rate": 0.0007227420439264903, + "loss": 0.3721, + "step": 10011 + }, + { + "epoch": 0.5593764840629103, + "grad_norm": 4.604032039642334, + "learning_rate": 0.0007227140295831465, + "loss": 0.4158, + "step": 10012 + }, + { + "epoch": 0.5594323546665921, + "grad_norm": 1.5370848178863525, + "learning_rate": 0.0007226860152398029, + "loss": 0.4928, + "step": 10013 + }, + { + "epoch": 0.5594882252702741, + "grad_norm": 2.057692050933838, + "learning_rate": 0.0007226580008964591, + "loss": 0.3847, + "step": 10014 + }, + { + "epoch": 0.5595440958739559, + "grad_norm": 0.5383592247962952, + "learning_rate": 0.0007226299865531152, + "loss": 0.3906, + "step": 10015 + }, + { + "epoch": 0.5595999664776378, + "grad_norm": 0.7538604736328125, + "learning_rate": 0.0007226019722097715, + "loss": 0.509, + "step": 10016 + }, + { + "epoch": 0.5596558370813197, + "grad_norm": 0.9098837971687317, + "learning_rate": 0.0007225739578664277, + "loss": 0.4152, + "step": 10017 + }, + { + "epoch": 0.5597117076850016, + "grad_norm": 0.7096293568611145, + "learning_rate": 0.0007225459435230838, + "loss": 0.5587, + "step": 10018 + }, + { + "epoch": 0.5597675782886834, + "grad_norm": 0.8480110764503479, + "learning_rate": 0.00072251792917974, + "loss": 0.5174, + "step": 10019 + }, + { + "epoch": 0.5598234488923652, + "grad_norm": 0.5157456994056702, + "learning_rate": 0.0007224899148363963, + "loss": 0.4586, + "step": 10020 + }, + { + "epoch": 0.5598793194960472, + "grad_norm": 0.5780086517333984, + "learning_rate": 0.0007224619004930525, + "loss": 0.4804, + "step": 10021 + }, + { + "epoch": 0.559935190099729, + "grad_norm": 0.7091310620307922, + "learning_rate": 0.0007224338861497086, + "loss": 0.4848, + "step": 10022 + }, + { + "epoch": 0.5599910607034109, + "grad_norm": 0.5144706964492798, + "learning_rate": 0.0007224058718063648, + "loss": 0.5679, + "step": 10023 + }, + { + "epoch": 0.5600469313070928, + "grad_norm": 0.4794108271598816, + "learning_rate": 0.0007223778574630211, + "loss": 0.4376, + "step": 10024 + }, + { + "epoch": 0.5601028019107747, + "grad_norm": 0.6669219136238098, + "learning_rate": 0.0007223498431196773, + "loss": 0.4522, + "step": 10025 + }, + { + "epoch": 0.5601586725144565, + "grad_norm": 0.6301625370979309, + "learning_rate": 0.0007223218287763334, + "loss": 0.4797, + "step": 10026 + }, + { + "epoch": 0.5602145431181383, + "grad_norm": 0.9963908791542053, + "learning_rate": 0.0007222938144329897, + "loss": 0.483, + "step": 10027 + }, + { + "epoch": 0.5602704137218203, + "grad_norm": 0.7279836535453796, + "learning_rate": 0.0007222658000896459, + "loss": 0.4491, + "step": 10028 + }, + { + "epoch": 0.5603262843255021, + "grad_norm": 0.8949409127235413, + "learning_rate": 0.0007222377857463021, + "loss": 0.5772, + "step": 10029 + }, + { + "epoch": 0.560382154929184, + "grad_norm": 1.6472995281219482, + "learning_rate": 0.0007222097714029582, + "loss": 0.6709, + "step": 10030 + }, + { + "epoch": 0.5604380255328659, + "grad_norm": 1.5446020364761353, + "learning_rate": 0.0007221817570596146, + "loss": 0.7681, + "step": 10031 + }, + { + "epoch": 0.5604938961365478, + "grad_norm": 1.2962969541549683, + "learning_rate": 0.0007221537427162708, + "loss": 0.5074, + "step": 10032 + }, + { + "epoch": 0.5605497667402296, + "grad_norm": 1.3529280424118042, + "learning_rate": 0.000722125728372927, + "loss": 0.4533, + "step": 10033 + }, + { + "epoch": 0.5606056373439116, + "grad_norm": 1.1083950996398926, + "learning_rate": 0.0007220977140295832, + "loss": 0.5336, + "step": 10034 + }, + { + "epoch": 0.5606615079475934, + "grad_norm": 3.6962783336639404, + "learning_rate": 0.0007220696996862394, + "loss": 0.4284, + "step": 10035 + }, + { + "epoch": 0.5607173785512752, + "grad_norm": 2.2344400882720947, + "learning_rate": 0.0007220416853428956, + "loss": 0.5206, + "step": 10036 + }, + { + "epoch": 0.5607732491549571, + "grad_norm": 0.6140164732933044, + "learning_rate": 0.0007220136709995517, + "loss": 0.5663, + "step": 10037 + }, + { + "epoch": 0.560829119758639, + "grad_norm": 0.9345272183418274, + "learning_rate": 0.000721985656656208, + "loss": 0.5434, + "step": 10038 + }, + { + "epoch": 0.5608849903623209, + "grad_norm": 0.611477792263031, + "learning_rate": 0.0007219576423128642, + "loss": 0.4266, + "step": 10039 + }, + { + "epoch": 0.5609408609660027, + "grad_norm": 0.8785183429718018, + "learning_rate": 0.0007219296279695204, + "loss": 0.2917, + "step": 10040 + }, + { + "epoch": 0.5609967315696847, + "grad_norm": 0.5165649056434631, + "learning_rate": 0.0007219016136261766, + "loss": 0.4223, + "step": 10041 + }, + { + "epoch": 0.5610526021733665, + "grad_norm": 1.601319432258606, + "learning_rate": 0.0007218735992828328, + "loss": 0.3789, + "step": 10042 + }, + { + "epoch": 0.5611084727770483, + "grad_norm": 0.763471245765686, + "learning_rate": 0.000721845584939489, + "loss": 0.3541, + "step": 10043 + }, + { + "epoch": 0.5611643433807302, + "grad_norm": 0.5761995911598206, + "learning_rate": 0.0007218175705961452, + "loss": 0.415, + "step": 10044 + }, + { + "epoch": 0.5612202139844121, + "grad_norm": 1.5224334001541138, + "learning_rate": 0.0007217895562528014, + "loss": 0.5597, + "step": 10045 + }, + { + "epoch": 0.561276084588094, + "grad_norm": 0.6699972152709961, + "learning_rate": 0.0007217615419094576, + "loss": 0.5994, + "step": 10046 + }, + { + "epoch": 0.5613319551917758, + "grad_norm": 0.42376166582107544, + "learning_rate": 0.0007217335275661139, + "loss": 0.411, + "step": 10047 + }, + { + "epoch": 0.5613878257954578, + "grad_norm": 0.5899167656898499, + "learning_rate": 0.0007217055132227702, + "loss": 0.4018, + "step": 10048 + }, + { + "epoch": 0.5614436963991396, + "grad_norm": 0.43348658084869385, + "learning_rate": 0.0007216774988794263, + "loss": 0.4442, + "step": 10049 + }, + { + "epoch": 0.5614995670028214, + "grad_norm": 0.38596105575561523, + "learning_rate": 0.0007216494845360825, + "loss": 0.3228, + "step": 10050 + }, + { + "epoch": 0.5615554376065034, + "grad_norm": 0.5156455636024475, + "learning_rate": 0.0007216214701927387, + "loss": 0.4862, + "step": 10051 + }, + { + "epoch": 0.5616113082101852, + "grad_norm": 1.0238555669784546, + "learning_rate": 0.000721593455849395, + "loss": 0.4761, + "step": 10052 + }, + { + "epoch": 0.5616671788138671, + "grad_norm": 0.3682181239128113, + "learning_rate": 0.0007215654415060511, + "loss": 0.3875, + "step": 10053 + }, + { + "epoch": 0.5617230494175489, + "grad_norm": 0.5656912922859192, + "learning_rate": 0.0007215374271627073, + "loss": 0.5731, + "step": 10054 + }, + { + "epoch": 0.5617789200212309, + "grad_norm": 0.4092146158218384, + "learning_rate": 0.0007215094128193636, + "loss": 0.3825, + "step": 10055 + }, + { + "epoch": 0.5618347906249127, + "grad_norm": 0.49394965171813965, + "learning_rate": 0.0007214813984760198, + "loss": 0.4898, + "step": 10056 + }, + { + "epoch": 0.5618906612285945, + "grad_norm": 0.6778452396392822, + "learning_rate": 0.0007214533841326759, + "loss": 0.3878, + "step": 10057 + }, + { + "epoch": 0.5619465318322765, + "grad_norm": 0.7057639360427856, + "learning_rate": 0.0007214253697893321, + "loss": 0.4379, + "step": 10058 + }, + { + "epoch": 0.5620024024359583, + "grad_norm": 0.48926064372062683, + "learning_rate": 0.0007213973554459884, + "loss": 0.322, + "step": 10059 + }, + { + "epoch": 0.5620582730396402, + "grad_norm": 0.6252299547195435, + "learning_rate": 0.0007213693411026445, + "loss": 0.4932, + "step": 10060 + }, + { + "epoch": 0.562114143643322, + "grad_norm": 0.8332555890083313, + "learning_rate": 0.0007213413267593007, + "loss": 0.4773, + "step": 10061 + }, + { + "epoch": 0.562170014247004, + "grad_norm": 0.42078274488449097, + "learning_rate": 0.0007213133124159569, + "loss": 0.3795, + "step": 10062 + }, + { + "epoch": 0.5622258848506858, + "grad_norm": 1.2126647233963013, + "learning_rate": 0.0007212852980726133, + "loss": 0.4239, + "step": 10063 + }, + { + "epoch": 0.5622817554543676, + "grad_norm": 0.4859369099140167, + "learning_rate": 0.0007212572837292694, + "loss": 0.3839, + "step": 10064 + }, + { + "epoch": 0.5623376260580496, + "grad_norm": 0.5036464333534241, + "learning_rate": 0.0007212292693859256, + "loss": 0.3889, + "step": 10065 + }, + { + "epoch": 0.5623934966617314, + "grad_norm": 0.449126660823822, + "learning_rate": 0.0007212012550425819, + "loss": 0.514, + "step": 10066 + }, + { + "epoch": 0.5624493672654133, + "grad_norm": 0.4280398488044739, + "learning_rate": 0.0007211732406992381, + "loss": 0.3613, + "step": 10067 + }, + { + "epoch": 0.5625052378690951, + "grad_norm": 0.4540652632713318, + "learning_rate": 0.0007211452263558942, + "loss": 0.5058, + "step": 10068 + }, + { + "epoch": 0.5625611084727771, + "grad_norm": 0.579265296459198, + "learning_rate": 0.0007211172120125504, + "loss": 0.4415, + "step": 10069 + }, + { + "epoch": 0.5626169790764589, + "grad_norm": 0.767410397529602, + "learning_rate": 0.0007210891976692067, + "loss": 0.4674, + "step": 10070 + }, + { + "epoch": 0.5626728496801408, + "grad_norm": 0.9634297490119934, + "learning_rate": 0.0007210611833258629, + "loss": 0.6046, + "step": 10071 + }, + { + "epoch": 0.5627287202838227, + "grad_norm": 0.4385506510734558, + "learning_rate": 0.000721033168982519, + "loss": 0.502, + "step": 10072 + }, + { + "epoch": 0.5627845908875045, + "grad_norm": 0.46144312620162964, + "learning_rate": 0.0007210051546391753, + "loss": 0.4505, + "step": 10073 + }, + { + "epoch": 0.5628404614911864, + "grad_norm": 0.5899154543876648, + "learning_rate": 0.0007209771402958315, + "loss": 0.4956, + "step": 10074 + }, + { + "epoch": 0.5628963320948683, + "grad_norm": 0.7589423656463623, + "learning_rate": 0.0007209491259524877, + "loss": 0.4233, + "step": 10075 + }, + { + "epoch": 0.5629522026985502, + "grad_norm": 2.7256364822387695, + "learning_rate": 0.0007209211116091438, + "loss": 0.5767, + "step": 10076 + }, + { + "epoch": 0.563008073302232, + "grad_norm": 0.6137312650680542, + "learning_rate": 0.0007208930972658001, + "loss": 0.4955, + "step": 10077 + }, + { + "epoch": 0.5630639439059139, + "grad_norm": 0.5357343554496765, + "learning_rate": 0.0007208650829224563, + "loss": 0.4177, + "step": 10078 + }, + { + "epoch": 0.5631198145095958, + "grad_norm": 0.48477911949157715, + "learning_rate": 0.0007208370685791125, + "loss": 0.3789, + "step": 10079 + }, + { + "epoch": 0.5631756851132776, + "grad_norm": 0.45042914152145386, + "learning_rate": 0.0007208090542357688, + "loss": 0.5995, + "step": 10080 + }, + { + "epoch": 0.5632315557169595, + "grad_norm": 1.1643003225326538, + "learning_rate": 0.000720781039892425, + "loss": 0.5285, + "step": 10081 + }, + { + "epoch": 0.5632874263206414, + "grad_norm": 1.4648374319076538, + "learning_rate": 0.0007207530255490812, + "loss": 0.48, + "step": 10082 + }, + { + "epoch": 0.5633432969243233, + "grad_norm": 1.119872808456421, + "learning_rate": 0.0007207250112057373, + "loss": 0.3435, + "step": 10083 + }, + { + "epoch": 0.5633991675280051, + "grad_norm": 0.6773856282234192, + "learning_rate": 0.0007206969968623936, + "loss": 0.5861, + "step": 10084 + }, + { + "epoch": 0.563455038131687, + "grad_norm": 0.8726876974105835, + "learning_rate": 0.0007206689825190498, + "loss": 0.4392, + "step": 10085 + }, + { + "epoch": 0.5635109087353689, + "grad_norm": 0.7897850871086121, + "learning_rate": 0.000720640968175706, + "loss": 0.6472, + "step": 10086 + }, + { + "epoch": 0.5635667793390507, + "grad_norm": 0.5093886852264404, + "learning_rate": 0.0007206129538323622, + "loss": 0.494, + "step": 10087 + }, + { + "epoch": 0.5636226499427326, + "grad_norm": 0.561976969242096, + "learning_rate": 0.0007205849394890184, + "loss": 0.4465, + "step": 10088 + }, + { + "epoch": 0.5636785205464145, + "grad_norm": 0.4285147786140442, + "learning_rate": 0.0007205569251456746, + "loss": 0.4479, + "step": 10089 + }, + { + "epoch": 0.5637343911500964, + "grad_norm": 0.42515793442726135, + "learning_rate": 0.0007205289108023308, + "loss": 0.4344, + "step": 10090 + }, + { + "epoch": 0.5637902617537782, + "grad_norm": 0.4281598627567291, + "learning_rate": 0.000720500896458987, + "loss": 0.3977, + "step": 10091 + }, + { + "epoch": 0.5638461323574602, + "grad_norm": 0.4705718457698822, + "learning_rate": 0.0007204728821156432, + "loss": 0.3976, + "step": 10092 + }, + { + "epoch": 0.563902002961142, + "grad_norm": 0.5340166091918945, + "learning_rate": 0.0007204448677722994, + "loss": 0.4489, + "step": 10093 + }, + { + "epoch": 0.5639578735648239, + "grad_norm": 0.5144967436790466, + "learning_rate": 0.0007204168534289556, + "loss": 0.463, + "step": 10094 + }, + { + "epoch": 0.5640137441685057, + "grad_norm": 0.5640775561332703, + "learning_rate": 0.0007203888390856118, + "loss": 0.5121, + "step": 10095 + }, + { + "epoch": 0.5640696147721876, + "grad_norm": 0.5347294807434082, + "learning_rate": 0.000720360824742268, + "loss": 0.4663, + "step": 10096 + }, + { + "epoch": 0.5641254853758695, + "grad_norm": 0.45263612270355225, + "learning_rate": 0.0007203328103989243, + "loss": 0.4142, + "step": 10097 + }, + { + "epoch": 0.5641813559795513, + "grad_norm": 0.5335516929626465, + "learning_rate": 0.0007203047960555806, + "loss": 0.3939, + "step": 10098 + }, + { + "epoch": 0.5642372265832333, + "grad_norm": 0.49971187114715576, + "learning_rate": 0.0007202767817122367, + "loss": 0.4121, + "step": 10099 + }, + { + "epoch": 0.5642930971869151, + "grad_norm": 0.38997524976730347, + "learning_rate": 0.0007202487673688929, + "loss": 0.4639, + "step": 10100 + }, + { + "epoch": 0.564348967790597, + "grad_norm": 0.6200942397117615, + "learning_rate": 0.0007202207530255491, + "loss": 0.5139, + "step": 10101 + }, + { + "epoch": 0.5644048383942788, + "grad_norm": 0.5036097168922424, + "learning_rate": 0.0007201927386822053, + "loss": 0.409, + "step": 10102 + }, + { + "epoch": 0.5644607089979607, + "grad_norm": 0.5657359957695007, + "learning_rate": 0.0007201647243388615, + "loss": 0.4165, + "step": 10103 + }, + { + "epoch": 0.5645165796016426, + "grad_norm": 0.590039074420929, + "learning_rate": 0.0007201367099955177, + "loss": 0.3653, + "step": 10104 + }, + { + "epoch": 0.5645724502053244, + "grad_norm": 0.5966876745223999, + "learning_rate": 0.000720108695652174, + "loss": 0.5986, + "step": 10105 + }, + { + "epoch": 0.5646283208090064, + "grad_norm": 0.5030028820037842, + "learning_rate": 0.0007200806813088301, + "loss": 0.4363, + "step": 10106 + }, + { + "epoch": 0.5646841914126882, + "grad_norm": 0.688234806060791, + "learning_rate": 0.0007200526669654863, + "loss": 0.4992, + "step": 10107 + }, + { + "epoch": 0.5647400620163701, + "grad_norm": 14.769930839538574, + "learning_rate": 0.0007200246526221425, + "loss": 0.5075, + "step": 10108 + }, + { + "epoch": 0.564795932620052, + "grad_norm": 0.4652042090892792, + "learning_rate": 0.0007199966382787988, + "loss": 0.4082, + "step": 10109 + }, + { + "epoch": 0.5648518032237339, + "grad_norm": 0.4128246605396271, + "learning_rate": 0.0007199686239354549, + "loss": 0.4765, + "step": 10110 + }, + { + "epoch": 0.5649076738274157, + "grad_norm": 3.713712215423584, + "learning_rate": 0.0007199406095921111, + "loss": 0.4444, + "step": 10111 + }, + { + "epoch": 0.5649635444310975, + "grad_norm": 0.5619252920150757, + "learning_rate": 0.0007199125952487674, + "loss": 0.4784, + "step": 10112 + }, + { + "epoch": 0.5650194150347795, + "grad_norm": 0.5670107007026672, + "learning_rate": 0.0007198845809054237, + "loss": 0.4785, + "step": 10113 + }, + { + "epoch": 0.5650752856384613, + "grad_norm": 0.47824397683143616, + "learning_rate": 0.0007198565665620797, + "loss": 0.458, + "step": 10114 + }, + { + "epoch": 0.5651311562421432, + "grad_norm": 0.40383726358413696, + "learning_rate": 0.000719828552218736, + "loss": 0.3526, + "step": 10115 + }, + { + "epoch": 0.5651870268458251, + "grad_norm": 1.376578688621521, + "learning_rate": 0.0007198005378753923, + "loss": 0.4849, + "step": 10116 + }, + { + "epoch": 0.565242897449507, + "grad_norm": 0.4814205467700958, + "learning_rate": 0.0007197725235320485, + "loss": 0.4786, + "step": 10117 + }, + { + "epoch": 0.5652987680531888, + "grad_norm": 0.5742153525352478, + "learning_rate": 0.0007197445091887046, + "loss": 0.5307, + "step": 10118 + }, + { + "epoch": 0.5653546386568706, + "grad_norm": 0.4222102463245392, + "learning_rate": 0.0007197164948453609, + "loss": 0.3925, + "step": 10119 + }, + { + "epoch": 0.5654105092605526, + "grad_norm": 0.6138949990272522, + "learning_rate": 0.0007196884805020171, + "loss": 0.4603, + "step": 10120 + }, + { + "epoch": 0.5654663798642344, + "grad_norm": 0.9461135268211365, + "learning_rate": 0.0007196604661586733, + "loss": 0.4816, + "step": 10121 + }, + { + "epoch": 0.5655222504679163, + "grad_norm": 0.5235634446144104, + "learning_rate": 0.0007196324518153294, + "loss": 0.4168, + "step": 10122 + }, + { + "epoch": 0.5655781210715982, + "grad_norm": 0.6622493863105774, + "learning_rate": 0.0007196044374719857, + "loss": 0.4465, + "step": 10123 + }, + { + "epoch": 0.5656339916752801, + "grad_norm": 0.5273646116256714, + "learning_rate": 0.0007195764231286419, + "loss": 0.4146, + "step": 10124 + }, + { + "epoch": 0.5656898622789619, + "grad_norm": 0.8709349632263184, + "learning_rate": 0.000719548408785298, + "loss": 0.5147, + "step": 10125 + }, + { + "epoch": 0.5657457328826438, + "grad_norm": 0.5598995089530945, + "learning_rate": 0.0007195203944419542, + "loss": 0.3992, + "step": 10126 + }, + { + "epoch": 0.5658016034863257, + "grad_norm": 1.8765383958816528, + "learning_rate": 0.0007194923800986105, + "loss": 0.4226, + "step": 10127 + }, + { + "epoch": 0.5658574740900075, + "grad_norm": 0.5031315684318542, + "learning_rate": 0.0007194643657552667, + "loss": 0.6574, + "step": 10128 + }, + { + "epoch": 0.5659133446936894, + "grad_norm": 1.1307653188705444, + "learning_rate": 0.0007194363514119228, + "loss": 0.431, + "step": 10129 + }, + { + "epoch": 0.5659692152973713, + "grad_norm": 0.4390008747577667, + "learning_rate": 0.0007194083370685792, + "loss": 0.3939, + "step": 10130 + }, + { + "epoch": 0.5660250859010532, + "grad_norm": 4.446183204650879, + "learning_rate": 0.0007193803227252354, + "loss": 0.4105, + "step": 10131 + }, + { + "epoch": 0.566080956504735, + "grad_norm": 0.7809885144233704, + "learning_rate": 0.0007193523083818916, + "loss": 0.3881, + "step": 10132 + }, + { + "epoch": 0.566136827108417, + "grad_norm": 0.5374835729598999, + "learning_rate": 0.0007193242940385477, + "loss": 0.4282, + "step": 10133 + }, + { + "epoch": 0.5661926977120988, + "grad_norm": 0.43285810947418213, + "learning_rate": 0.000719296279695204, + "loss": 0.3846, + "step": 10134 + }, + { + "epoch": 0.5662485683157806, + "grad_norm": 0.5977165102958679, + "learning_rate": 0.0007192682653518602, + "loss": 0.4703, + "step": 10135 + }, + { + "epoch": 0.5663044389194625, + "grad_norm": 0.6795068383216858, + "learning_rate": 0.0007192402510085164, + "loss": 0.5388, + "step": 10136 + }, + { + "epoch": 0.5663603095231444, + "grad_norm": 0.3788456916809082, + "learning_rate": 0.0007192122366651726, + "loss": 0.5029, + "step": 10137 + }, + { + "epoch": 0.5664161801268263, + "grad_norm": 2.36832332611084, + "learning_rate": 0.0007191842223218288, + "loss": 0.4918, + "step": 10138 + }, + { + "epoch": 0.5664720507305081, + "grad_norm": 0.47309717535972595, + "learning_rate": 0.000719156207978485, + "loss": 0.5669, + "step": 10139 + }, + { + "epoch": 0.56652792133419, + "grad_norm": 0.5047085881233215, + "learning_rate": 0.0007191281936351412, + "loss": 0.4489, + "step": 10140 + }, + { + "epoch": 0.5665837919378719, + "grad_norm": 0.479474276304245, + "learning_rate": 0.0007191001792917974, + "loss": 0.4661, + "step": 10141 + }, + { + "epoch": 0.5666396625415537, + "grad_norm": 0.5490813851356506, + "learning_rate": 0.0007190721649484536, + "loss": 0.5206, + "step": 10142 + }, + { + "epoch": 0.5666955331452357, + "grad_norm": 0.5063911080360413, + "learning_rate": 0.0007190441506051098, + "loss": 0.4864, + "step": 10143 + }, + { + "epoch": 0.5667514037489175, + "grad_norm": 1.473293662071228, + "learning_rate": 0.000719016136261766, + "loss": 0.4287, + "step": 10144 + }, + { + "epoch": 0.5668072743525994, + "grad_norm": 0.736222505569458, + "learning_rate": 0.0007189881219184222, + "loss": 0.5994, + "step": 10145 + }, + { + "epoch": 0.5668631449562812, + "grad_norm": 0.4753133952617645, + "learning_rate": 0.0007189601075750784, + "loss": 0.4778, + "step": 10146 + }, + { + "epoch": 0.5669190155599632, + "grad_norm": 0.7524417042732239, + "learning_rate": 0.0007189320932317347, + "loss": 0.4704, + "step": 10147 + }, + { + "epoch": 0.566974886163645, + "grad_norm": 1.2533557415008545, + "learning_rate": 0.0007189040788883909, + "loss": 0.4699, + "step": 10148 + }, + { + "epoch": 0.5670307567673268, + "grad_norm": 0.7509532570838928, + "learning_rate": 0.0007188760645450471, + "loss": 0.4877, + "step": 10149 + }, + { + "epoch": 0.5670866273710088, + "grad_norm": 0.6325125098228455, + "learning_rate": 0.0007188480502017033, + "loss": 0.5031, + "step": 10150 + }, + { + "epoch": 0.5671424979746906, + "grad_norm": 0.6895074844360352, + "learning_rate": 0.0007188200358583596, + "loss": 0.5198, + "step": 10151 + }, + { + "epoch": 0.5671983685783725, + "grad_norm": 0.6906082630157471, + "learning_rate": 0.0007187920215150157, + "loss": 0.5106, + "step": 10152 + }, + { + "epoch": 0.5672542391820543, + "grad_norm": 0.5413280129432678, + "learning_rate": 0.0007187640071716719, + "loss": 0.5065, + "step": 10153 + }, + { + "epoch": 0.5673101097857363, + "grad_norm": 0.5054965615272522, + "learning_rate": 0.0007187359928283281, + "loss": 0.482, + "step": 10154 + }, + { + "epoch": 0.5673659803894181, + "grad_norm": 0.4665525555610657, + "learning_rate": 0.0007187079784849844, + "loss": 0.6166, + "step": 10155 + }, + { + "epoch": 0.5674218509930999, + "grad_norm": 0.5381871461868286, + "learning_rate": 0.0007186799641416405, + "loss": 0.5302, + "step": 10156 + }, + { + "epoch": 0.5674777215967819, + "grad_norm": 0.6179479360580444, + "learning_rate": 0.0007186519497982967, + "loss": 0.4405, + "step": 10157 + }, + { + "epoch": 0.5675335922004637, + "grad_norm": 0.6155729293823242, + "learning_rate": 0.000718623935454953, + "loss": 0.4876, + "step": 10158 + }, + { + "epoch": 0.5675894628041456, + "grad_norm": 0.616601288318634, + "learning_rate": 0.0007185959211116092, + "loss": 0.4531, + "step": 10159 + }, + { + "epoch": 0.5676453334078275, + "grad_norm": 0.4742949903011322, + "learning_rate": 0.0007185679067682653, + "loss": 0.4804, + "step": 10160 + }, + { + "epoch": 0.5677012040115094, + "grad_norm": 0.7935799360275269, + "learning_rate": 0.0007185398924249215, + "loss": 0.3769, + "step": 10161 + }, + { + "epoch": 0.5677570746151912, + "grad_norm": 0.8506180047988892, + "learning_rate": 0.0007185118780815778, + "loss": 0.5116, + "step": 10162 + }, + { + "epoch": 0.567812945218873, + "grad_norm": 0.5120320320129395, + "learning_rate": 0.000718483863738234, + "loss": 0.5515, + "step": 10163 + }, + { + "epoch": 0.567868815822555, + "grad_norm": 0.7421508431434631, + "learning_rate": 0.0007184558493948901, + "loss": 0.5924, + "step": 10164 + }, + { + "epoch": 0.5679246864262368, + "grad_norm": 1.3043012619018555, + "learning_rate": 0.0007184278350515464, + "loss": 0.4097, + "step": 10165 + }, + { + "epoch": 0.5679805570299187, + "grad_norm": 0.6972129940986633, + "learning_rate": 0.0007183998207082027, + "loss": 0.5306, + "step": 10166 + }, + { + "epoch": 0.5680364276336006, + "grad_norm": 0.5984460711479187, + "learning_rate": 0.0007183718063648588, + "loss": 0.4149, + "step": 10167 + }, + { + "epoch": 0.5680922982372825, + "grad_norm": 0.3899592161178589, + "learning_rate": 0.000718343792021515, + "loss": 0.395, + "step": 10168 + }, + { + "epoch": 0.5681481688409643, + "grad_norm": 0.49292251467704773, + "learning_rate": 0.0007183157776781713, + "loss": 0.5632, + "step": 10169 + }, + { + "epoch": 0.5682040394446461, + "grad_norm": 0.4856055676937103, + "learning_rate": 0.0007182877633348275, + "loss": 0.453, + "step": 10170 + }, + { + "epoch": 0.5682599100483281, + "grad_norm": 0.6655503511428833, + "learning_rate": 0.0007182597489914836, + "loss": 0.3962, + "step": 10171 + }, + { + "epoch": 0.5683157806520099, + "grad_norm": 0.37886449694633484, + "learning_rate": 0.0007182317346481398, + "loss": 0.3869, + "step": 10172 + }, + { + "epoch": 0.5683716512556918, + "grad_norm": 0.5053687691688538, + "learning_rate": 0.0007182037203047961, + "loss": 0.3885, + "step": 10173 + }, + { + "epoch": 0.5684275218593737, + "grad_norm": 0.8909807801246643, + "learning_rate": 0.0007181757059614523, + "loss": 0.4646, + "step": 10174 + }, + { + "epoch": 0.5684833924630556, + "grad_norm": 0.7329044938087463, + "learning_rate": 0.0007181476916181084, + "loss": 0.5772, + "step": 10175 + }, + { + "epoch": 0.5685392630667374, + "grad_norm": 2.9428915977478027, + "learning_rate": 0.0007181196772747647, + "loss": 0.4535, + "step": 10176 + }, + { + "epoch": 0.5685951336704194, + "grad_norm": 0.5532366633415222, + "learning_rate": 0.0007180916629314209, + "loss": 0.5196, + "step": 10177 + }, + { + "epoch": 0.5686510042741012, + "grad_norm": 0.43562814593315125, + "learning_rate": 0.0007180636485880771, + "loss": 0.4642, + "step": 10178 + }, + { + "epoch": 0.568706874877783, + "grad_norm": 0.4175507724285126, + "learning_rate": 0.0007180356342447332, + "loss": 0.3855, + "step": 10179 + }, + { + "epoch": 0.5687627454814649, + "grad_norm": 1.0882407426834106, + "learning_rate": 0.0007180076199013896, + "loss": 0.4464, + "step": 10180 + }, + { + "epoch": 0.5688186160851468, + "grad_norm": 0.48965102434158325, + "learning_rate": 0.0007179796055580458, + "loss": 0.4608, + "step": 10181 + }, + { + "epoch": 0.5688744866888287, + "grad_norm": 0.7793510556221008, + "learning_rate": 0.000717951591214702, + "loss": 0.3921, + "step": 10182 + }, + { + "epoch": 0.5689303572925105, + "grad_norm": 0.5073941946029663, + "learning_rate": 0.0007179235768713582, + "loss": 0.3443, + "step": 10183 + }, + { + "epoch": 0.5689862278961925, + "grad_norm": 0.8972160816192627, + "learning_rate": 0.0007178955625280144, + "loss": 0.57, + "step": 10184 + }, + { + "epoch": 0.5690420984998743, + "grad_norm": 0.5302116274833679, + "learning_rate": 0.0007178675481846706, + "loss": 0.3283, + "step": 10185 + }, + { + "epoch": 0.5690979691035561, + "grad_norm": 0.9759556651115417, + "learning_rate": 0.0007178395338413267, + "loss": 0.5134, + "step": 10186 + }, + { + "epoch": 0.569153839707238, + "grad_norm": 0.4213533401489258, + "learning_rate": 0.000717811519497983, + "loss": 0.3872, + "step": 10187 + }, + { + "epoch": 0.5692097103109199, + "grad_norm": 0.5770381093025208, + "learning_rate": 0.0007177835051546392, + "loss": 0.4758, + "step": 10188 + }, + { + "epoch": 0.5692655809146018, + "grad_norm": 0.4610619843006134, + "learning_rate": 0.0007177554908112954, + "loss": 0.2989, + "step": 10189 + }, + { + "epoch": 0.5693214515182836, + "grad_norm": 0.6897260546684265, + "learning_rate": 0.0007177274764679516, + "loss": 0.5185, + "step": 10190 + }, + { + "epoch": 0.5693773221219656, + "grad_norm": 0.6436983942985535, + "learning_rate": 0.0007176994621246078, + "loss": 0.5929, + "step": 10191 + }, + { + "epoch": 0.5694331927256474, + "grad_norm": 1.0138683319091797, + "learning_rate": 0.000717671447781264, + "loss": 0.5349, + "step": 10192 + }, + { + "epoch": 0.5694890633293292, + "grad_norm": 0.865609347820282, + "learning_rate": 0.0007176434334379202, + "loss": 0.5753, + "step": 10193 + }, + { + "epoch": 0.5695449339330112, + "grad_norm": 2.4042091369628906, + "learning_rate": 0.0007176154190945764, + "loss": 0.385, + "step": 10194 + }, + { + "epoch": 0.569600804536693, + "grad_norm": 0.44594264030456543, + "learning_rate": 0.0007175874047512326, + "loss": 0.4766, + "step": 10195 + }, + { + "epoch": 0.5696566751403749, + "grad_norm": 0.3858219385147095, + "learning_rate": 0.0007175593904078888, + "loss": 0.3643, + "step": 10196 + }, + { + "epoch": 0.5697125457440567, + "grad_norm": 2.6011481285095215, + "learning_rate": 0.0007175313760645452, + "loss": 0.5212, + "step": 10197 + }, + { + "epoch": 0.5697684163477387, + "grad_norm": 0.46844252943992615, + "learning_rate": 0.0007175033617212013, + "loss": 0.4331, + "step": 10198 + }, + { + "epoch": 0.5698242869514205, + "grad_norm": 0.7922723293304443, + "learning_rate": 0.0007174753473778575, + "loss": 0.4861, + "step": 10199 + }, + { + "epoch": 0.5698801575551024, + "grad_norm": 0.504944384098053, + "learning_rate": 0.0007174473330345137, + "loss": 0.417, + "step": 10200 + }, + { + "epoch": 0.5699360281587843, + "grad_norm": 0.9095560312271118, + "learning_rate": 0.00071741931869117, + "loss": 0.5564, + "step": 10201 + }, + { + "epoch": 0.5699918987624661, + "grad_norm": 0.4781077206134796, + "learning_rate": 0.0007173913043478261, + "loss": 0.5233, + "step": 10202 + }, + { + "epoch": 0.570047769366148, + "grad_norm": 0.7526348233222961, + "learning_rate": 0.0007173632900044823, + "loss": 0.4169, + "step": 10203 + }, + { + "epoch": 0.5701036399698298, + "grad_norm": 0.8232529163360596, + "learning_rate": 0.0007173352756611385, + "loss": 0.4404, + "step": 10204 + }, + { + "epoch": 0.5701595105735118, + "grad_norm": 1.0068408250808716, + "learning_rate": 0.0007173072613177948, + "loss": 0.4393, + "step": 10205 + }, + { + "epoch": 0.5702153811771936, + "grad_norm": 0.43606632947921753, + "learning_rate": 0.0007172792469744509, + "loss": 0.5023, + "step": 10206 + }, + { + "epoch": 0.5702712517808755, + "grad_norm": 0.6271001696586609, + "learning_rate": 0.0007172512326311071, + "loss": 0.5191, + "step": 10207 + }, + { + "epoch": 0.5703271223845574, + "grad_norm": 1.8885998725891113, + "learning_rate": 0.0007172232182877634, + "loss": 0.6356, + "step": 10208 + }, + { + "epoch": 0.5703829929882392, + "grad_norm": 0.9968698024749756, + "learning_rate": 0.0007171952039444195, + "loss": 0.344, + "step": 10209 + }, + { + "epoch": 0.5704388635919211, + "grad_norm": 3.6872174739837646, + "learning_rate": 0.0007171671896010757, + "loss": 0.477, + "step": 10210 + }, + { + "epoch": 0.570494734195603, + "grad_norm": 0.4603792130947113, + "learning_rate": 0.0007171391752577319, + "loss": 0.3364, + "step": 10211 + }, + { + "epoch": 0.5705506047992849, + "grad_norm": 0.4302327334880829, + "learning_rate": 0.0007171111609143882, + "loss": 0.3907, + "step": 10212 + }, + { + "epoch": 0.5706064754029667, + "grad_norm": 0.8307905793190002, + "learning_rate": 0.0007170831465710443, + "loss": 0.5927, + "step": 10213 + }, + { + "epoch": 0.5706623460066486, + "grad_norm": 0.6413657069206238, + "learning_rate": 0.0007170551322277005, + "loss": 0.6792, + "step": 10214 + }, + { + "epoch": 0.5707182166103305, + "grad_norm": 1.218487024307251, + "learning_rate": 0.0007170271178843569, + "loss": 0.4923, + "step": 10215 + }, + { + "epoch": 0.5707740872140123, + "grad_norm": 2.6757187843322754, + "learning_rate": 0.0007169991035410131, + "loss": 0.5975, + "step": 10216 + }, + { + "epoch": 0.5708299578176942, + "grad_norm": 0.5649046897888184, + "learning_rate": 0.0007169710891976692, + "loss": 0.4386, + "step": 10217 + }, + { + "epoch": 0.5708858284213761, + "grad_norm": 0.6312761902809143, + "learning_rate": 0.0007169430748543254, + "loss": 0.4322, + "step": 10218 + }, + { + "epoch": 0.570941699025058, + "grad_norm": 0.8670374751091003, + "learning_rate": 0.0007169150605109817, + "loss": 0.527, + "step": 10219 + }, + { + "epoch": 0.5709975696287398, + "grad_norm": 0.4801878333091736, + "learning_rate": 0.0007168870461676379, + "loss": 0.4262, + "step": 10220 + }, + { + "epoch": 0.5710534402324217, + "grad_norm": 0.5641873478889465, + "learning_rate": 0.000716859031824294, + "loss": 0.3935, + "step": 10221 + }, + { + "epoch": 0.5711093108361036, + "grad_norm": 0.733172595500946, + "learning_rate": 0.0007168310174809503, + "loss": 0.4888, + "step": 10222 + }, + { + "epoch": 0.5711651814397855, + "grad_norm": 0.6340610384941101, + "learning_rate": 0.0007168030031376065, + "loss": 0.6749, + "step": 10223 + }, + { + "epoch": 0.5712210520434673, + "grad_norm": 0.44112056493759155, + "learning_rate": 0.0007167749887942627, + "loss": 0.5135, + "step": 10224 + }, + { + "epoch": 0.5712769226471492, + "grad_norm": 0.42149612307548523, + "learning_rate": 0.0007167469744509188, + "loss": 0.4527, + "step": 10225 + }, + { + "epoch": 0.5713327932508311, + "grad_norm": 0.5048412680625916, + "learning_rate": 0.0007167189601075751, + "loss": 0.6748, + "step": 10226 + }, + { + "epoch": 0.5713886638545129, + "grad_norm": 1.4579925537109375, + "learning_rate": 0.0007166909457642313, + "loss": 0.4797, + "step": 10227 + }, + { + "epoch": 0.5714445344581949, + "grad_norm": 0.5557900071144104, + "learning_rate": 0.0007166629314208874, + "loss": 0.384, + "step": 10228 + }, + { + "epoch": 0.5715004050618767, + "grad_norm": 1.9850332736968994, + "learning_rate": 0.0007166349170775437, + "loss": 0.4335, + "step": 10229 + }, + { + "epoch": 0.5715562756655586, + "grad_norm": 0.891612708568573, + "learning_rate": 0.0007166069027342, + "loss": 0.3574, + "step": 10230 + }, + { + "epoch": 0.5716121462692404, + "grad_norm": 0.5195823311805725, + "learning_rate": 0.0007165788883908562, + "loss": 0.4677, + "step": 10231 + }, + { + "epoch": 0.5716680168729223, + "grad_norm": 0.7752794623374939, + "learning_rate": 0.0007165508740475123, + "loss": 0.4717, + "step": 10232 + }, + { + "epoch": 0.5717238874766042, + "grad_norm": 0.5758737325668335, + "learning_rate": 0.0007165228597041686, + "loss": 0.456, + "step": 10233 + }, + { + "epoch": 0.571779758080286, + "grad_norm": 0.53157639503479, + "learning_rate": 0.0007164948453608248, + "loss": 0.4213, + "step": 10234 + }, + { + "epoch": 0.571835628683968, + "grad_norm": 0.3654715120792389, + "learning_rate": 0.000716466831017481, + "loss": 0.4255, + "step": 10235 + }, + { + "epoch": 0.5718914992876498, + "grad_norm": 0.5911012291908264, + "learning_rate": 0.0007164388166741371, + "loss": 0.5193, + "step": 10236 + }, + { + "epoch": 0.5719473698913317, + "grad_norm": 0.5398606061935425, + "learning_rate": 0.0007164108023307934, + "loss": 0.4401, + "step": 10237 + }, + { + "epoch": 0.5720032404950135, + "grad_norm": 0.4369382858276367, + "learning_rate": 0.0007163827879874496, + "loss": 0.3721, + "step": 10238 + }, + { + "epoch": 0.5720591110986954, + "grad_norm": 0.7940000295639038, + "learning_rate": 0.0007163547736441058, + "loss": 0.5043, + "step": 10239 + }, + { + "epoch": 0.5721149817023773, + "grad_norm": 0.7947934865951538, + "learning_rate": 0.000716326759300762, + "loss": 0.4033, + "step": 10240 + }, + { + "epoch": 0.5721708523060591, + "grad_norm": 0.6155095100402832, + "learning_rate": 0.0007162987449574182, + "loss": 0.359, + "step": 10241 + }, + { + "epoch": 0.5722267229097411, + "grad_norm": 0.4576118290424347, + "learning_rate": 0.0007162707306140744, + "loss": 0.4284, + "step": 10242 + }, + { + "epoch": 0.5722825935134229, + "grad_norm": 0.6658133864402771, + "learning_rate": 0.0007162427162707306, + "loss": 0.4304, + "step": 10243 + }, + { + "epoch": 0.5723384641171048, + "grad_norm": 0.928941011428833, + "learning_rate": 0.0007162147019273868, + "loss": 0.5104, + "step": 10244 + }, + { + "epoch": 0.5723943347207867, + "grad_norm": 0.5664893984794617, + "learning_rate": 0.000716186687584043, + "loss": 0.5701, + "step": 10245 + }, + { + "epoch": 0.5724502053244686, + "grad_norm": 0.5139405131340027, + "learning_rate": 0.0007161586732406992, + "loss": 0.4685, + "step": 10246 + }, + { + "epoch": 0.5725060759281504, + "grad_norm": 0.6884979605674744, + "learning_rate": 0.0007161306588973556, + "loss": 0.4396, + "step": 10247 + }, + { + "epoch": 0.5725619465318322, + "grad_norm": 0.5265318751335144, + "learning_rate": 0.0007161026445540117, + "loss": 0.396, + "step": 10248 + }, + { + "epoch": 0.5726178171355142, + "grad_norm": 0.9082959890365601, + "learning_rate": 0.0007160746302106679, + "loss": 0.5733, + "step": 10249 + }, + { + "epoch": 0.572673687739196, + "grad_norm": 0.4789411425590515, + "learning_rate": 0.0007160466158673241, + "loss": 0.4325, + "step": 10250 + }, + { + "epoch": 0.5727295583428779, + "grad_norm": 1.1113015413284302, + "learning_rate": 0.0007160186015239803, + "loss": 0.4948, + "step": 10251 + }, + { + "epoch": 0.5727854289465598, + "grad_norm": 0.8175991177558899, + "learning_rate": 0.0007159905871806365, + "loss": 0.4315, + "step": 10252 + }, + { + "epoch": 0.5728412995502417, + "grad_norm": 0.6284119486808777, + "learning_rate": 0.0007159625728372927, + "loss": 0.4855, + "step": 10253 + }, + { + "epoch": 0.5728971701539235, + "grad_norm": 0.5312758684158325, + "learning_rate": 0.000715934558493949, + "loss": 0.5561, + "step": 10254 + }, + { + "epoch": 0.5729530407576053, + "grad_norm": 1.1118340492248535, + "learning_rate": 0.0007159065441506051, + "loss": 0.4504, + "step": 10255 + }, + { + "epoch": 0.5730089113612873, + "grad_norm": 0.4794607162475586, + "learning_rate": 0.0007158785298072613, + "loss": 0.4964, + "step": 10256 + }, + { + "epoch": 0.5730647819649691, + "grad_norm": 2.9418208599090576, + "learning_rate": 0.0007158505154639175, + "loss": 0.4243, + "step": 10257 + }, + { + "epoch": 0.573120652568651, + "grad_norm": 0.7264583110809326, + "learning_rate": 0.0007158225011205738, + "loss": 0.6078, + "step": 10258 + }, + { + "epoch": 0.5731765231723329, + "grad_norm": 0.6680260300636292, + "learning_rate": 0.0007157944867772299, + "loss": 0.4343, + "step": 10259 + }, + { + "epoch": 0.5732323937760148, + "grad_norm": 1.030440092086792, + "learning_rate": 0.0007157664724338861, + "loss": 0.4464, + "step": 10260 + }, + { + "epoch": 0.5732882643796966, + "grad_norm": 0.41079169511795044, + "learning_rate": 0.0007157384580905424, + "loss": 0.4114, + "step": 10261 + }, + { + "epoch": 0.5733441349833786, + "grad_norm": 0.4535685181617737, + "learning_rate": 0.0007157104437471986, + "loss": 0.3776, + "step": 10262 + }, + { + "epoch": 0.5734000055870604, + "grad_norm": 0.5336297154426575, + "learning_rate": 0.0007156824294038547, + "loss": 0.4586, + "step": 10263 + }, + { + "epoch": 0.5734558761907422, + "grad_norm": 0.5448347926139832, + "learning_rate": 0.000715654415060511, + "loss": 0.5145, + "step": 10264 + }, + { + "epoch": 0.5735117467944241, + "grad_norm": 0.4457460343837738, + "learning_rate": 0.0007156264007171673, + "loss": 0.3177, + "step": 10265 + }, + { + "epoch": 0.573567617398106, + "grad_norm": 0.5926575064659119, + "learning_rate": 0.0007155983863738235, + "loss": 0.4604, + "step": 10266 + }, + { + "epoch": 0.5736234880017879, + "grad_norm": 0.7129864692687988, + "learning_rate": 0.0007155703720304796, + "loss": 0.5155, + "step": 10267 + }, + { + "epoch": 0.5736793586054697, + "grad_norm": 0.4710366129875183, + "learning_rate": 0.0007155423576871359, + "loss": 0.344, + "step": 10268 + }, + { + "epoch": 0.5737352292091517, + "grad_norm": 1.256500005722046, + "learning_rate": 0.0007155143433437921, + "loss": 0.4144, + "step": 10269 + }, + { + "epoch": 0.5737910998128335, + "grad_norm": 0.4798011779785156, + "learning_rate": 0.0007154863290004483, + "loss": 0.3729, + "step": 10270 + }, + { + "epoch": 0.5738469704165153, + "grad_norm": 0.48284655809402466, + "learning_rate": 0.0007154583146571044, + "loss": 0.4894, + "step": 10271 + }, + { + "epoch": 0.5739028410201972, + "grad_norm": 0.8092394471168518, + "learning_rate": 0.0007154303003137607, + "loss": 0.4898, + "step": 10272 + }, + { + "epoch": 0.5739587116238791, + "grad_norm": 0.37767428159713745, + "learning_rate": 0.0007154022859704169, + "loss": 0.3421, + "step": 10273 + }, + { + "epoch": 0.574014582227561, + "grad_norm": 0.670428454875946, + "learning_rate": 0.000715374271627073, + "loss": 0.5419, + "step": 10274 + }, + { + "epoch": 0.5740704528312428, + "grad_norm": 0.695173442363739, + "learning_rate": 0.0007153462572837292, + "loss": 0.5305, + "step": 10275 + }, + { + "epoch": 0.5741263234349248, + "grad_norm": 0.4634600579738617, + "learning_rate": 0.0007153182429403855, + "loss": 0.5042, + "step": 10276 + }, + { + "epoch": 0.5741821940386066, + "grad_norm": 11.775195121765137, + "learning_rate": 0.0007152902285970417, + "loss": 0.498, + "step": 10277 + }, + { + "epoch": 0.5742380646422884, + "grad_norm": 0.7471457719802856, + "learning_rate": 0.0007152622142536978, + "loss": 0.6177, + "step": 10278 + }, + { + "epoch": 0.5742939352459703, + "grad_norm": 0.4956466555595398, + "learning_rate": 0.0007152341999103541, + "loss": 0.3624, + "step": 10279 + }, + { + "epoch": 0.5743498058496522, + "grad_norm": 0.5829969644546509, + "learning_rate": 0.0007152061855670103, + "loss": 0.5419, + "step": 10280 + }, + { + "epoch": 0.5744056764533341, + "grad_norm": 1.9719575643539429, + "learning_rate": 0.0007151781712236666, + "loss": 0.4946, + "step": 10281 + }, + { + "epoch": 0.5744615470570159, + "grad_norm": 1.2453199625015259, + "learning_rate": 0.0007151501568803227, + "loss": 0.4397, + "step": 10282 + }, + { + "epoch": 0.5745174176606979, + "grad_norm": 0.727525532245636, + "learning_rate": 0.000715122142536979, + "loss": 0.4605, + "step": 10283 + }, + { + "epoch": 0.5745732882643797, + "grad_norm": 1.1344470977783203, + "learning_rate": 0.0007150941281936352, + "loss": 0.4263, + "step": 10284 + }, + { + "epoch": 0.5746291588680615, + "grad_norm": 0.5869541168212891, + "learning_rate": 0.0007150661138502914, + "loss": 0.4838, + "step": 10285 + }, + { + "epoch": 0.5746850294717435, + "grad_norm": 0.528438150882721, + "learning_rate": 0.0007150380995069476, + "loss": 0.4074, + "step": 10286 + }, + { + "epoch": 0.5747409000754253, + "grad_norm": 0.5859910249710083, + "learning_rate": 0.0007150100851636038, + "loss": 0.4848, + "step": 10287 + }, + { + "epoch": 0.5747967706791072, + "grad_norm": 1.1897056102752686, + "learning_rate": 0.00071498207082026, + "loss": 0.4235, + "step": 10288 + }, + { + "epoch": 0.574852641282789, + "grad_norm": 0.38150250911712646, + "learning_rate": 0.0007149540564769162, + "loss": 0.4222, + "step": 10289 + }, + { + "epoch": 0.574908511886471, + "grad_norm": 3.1833395957946777, + "learning_rate": 0.0007149260421335724, + "loss": 0.3776, + "step": 10290 + }, + { + "epoch": 0.5749643824901528, + "grad_norm": 0.5225669741630554, + "learning_rate": 0.0007148980277902286, + "loss": 0.4286, + "step": 10291 + }, + { + "epoch": 0.5750202530938346, + "grad_norm": 0.6481263637542725, + "learning_rate": 0.0007148700134468848, + "loss": 0.4671, + "step": 10292 + }, + { + "epoch": 0.5750761236975166, + "grad_norm": 0.6424452066421509, + "learning_rate": 0.000714841999103541, + "loss": 0.5407, + "step": 10293 + }, + { + "epoch": 0.5751319943011984, + "grad_norm": 4.941681861877441, + "learning_rate": 0.0007148139847601972, + "loss": 0.5061, + "step": 10294 + }, + { + "epoch": 0.5751878649048803, + "grad_norm": 0.5071628093719482, + "learning_rate": 0.0007147859704168534, + "loss": 0.4249, + "step": 10295 + }, + { + "epoch": 0.5752437355085621, + "grad_norm": 0.7469527125358582, + "learning_rate": 0.0007147579560735096, + "loss": 0.4495, + "step": 10296 + }, + { + "epoch": 0.5752996061122441, + "grad_norm": 0.6257029175758362, + "learning_rate": 0.0007147299417301658, + "loss": 0.4527, + "step": 10297 + }, + { + "epoch": 0.5753554767159259, + "grad_norm": 1.1284128427505493, + "learning_rate": 0.000714701927386822, + "loss": 0.503, + "step": 10298 + }, + { + "epoch": 0.5754113473196077, + "grad_norm": 0.501481294631958, + "learning_rate": 0.0007146739130434783, + "loss": 0.429, + "step": 10299 + }, + { + "epoch": 0.5754672179232897, + "grad_norm": 0.5414264798164368, + "learning_rate": 0.0007146458987001346, + "loss": 0.4402, + "step": 10300 + }, + { + "epoch": 0.5755230885269715, + "grad_norm": 0.5607287287712097, + "learning_rate": 0.0007146178843567907, + "loss": 0.5386, + "step": 10301 + }, + { + "epoch": 0.5755789591306534, + "grad_norm": 0.618623673915863, + "learning_rate": 0.0007145898700134469, + "loss": 0.4338, + "step": 10302 + }, + { + "epoch": 0.5756348297343353, + "grad_norm": 0.492512971162796, + "learning_rate": 0.0007145618556701031, + "loss": 0.4697, + "step": 10303 + }, + { + "epoch": 0.5756907003380172, + "grad_norm": 2.528923749923706, + "learning_rate": 0.0007145338413267594, + "loss": 0.3772, + "step": 10304 + }, + { + "epoch": 0.575746570941699, + "grad_norm": 0.38832637667655945, + "learning_rate": 0.0007145058269834155, + "loss": 0.4338, + "step": 10305 + }, + { + "epoch": 0.5758024415453808, + "grad_norm": 0.5237044095993042, + "learning_rate": 0.0007144778126400717, + "loss": 0.5347, + "step": 10306 + }, + { + "epoch": 0.5758583121490628, + "grad_norm": 0.4636237621307373, + "learning_rate": 0.0007144497982967279, + "loss": 0.442, + "step": 10307 + }, + { + "epoch": 0.5759141827527446, + "grad_norm": 0.4816652238368988, + "learning_rate": 0.0007144217839533842, + "loss": 0.4176, + "step": 10308 + }, + { + "epoch": 0.5759700533564265, + "grad_norm": 0.5621851086616516, + "learning_rate": 0.0007143937696100403, + "loss": 0.4348, + "step": 10309 + }, + { + "epoch": 0.5760259239601084, + "grad_norm": 0.557892382144928, + "learning_rate": 0.0007143657552666965, + "loss": 0.4218, + "step": 10310 + }, + { + "epoch": 0.5760817945637903, + "grad_norm": 0.6112344264984131, + "learning_rate": 0.0007143377409233528, + "loss": 0.4095, + "step": 10311 + }, + { + "epoch": 0.5761376651674721, + "grad_norm": 1.0893585681915283, + "learning_rate": 0.000714309726580009, + "loss": 0.3691, + "step": 10312 + }, + { + "epoch": 0.576193535771154, + "grad_norm": 0.9230806827545166, + "learning_rate": 0.0007142817122366651, + "loss": 0.3728, + "step": 10313 + }, + { + "epoch": 0.5762494063748359, + "grad_norm": 0.42749056220054626, + "learning_rate": 0.0007142536978933213, + "loss": 0.3989, + "step": 10314 + }, + { + "epoch": 0.5763052769785177, + "grad_norm": 0.5090134739875793, + "learning_rate": 0.0007142256835499777, + "loss": 0.4822, + "step": 10315 + }, + { + "epoch": 0.5763611475821996, + "grad_norm": 0.5424376130104065, + "learning_rate": 0.0007141976692066338, + "loss": 0.4128, + "step": 10316 + }, + { + "epoch": 0.5764170181858815, + "grad_norm": 0.9199063181877136, + "learning_rate": 0.00071416965486329, + "loss": 0.6427, + "step": 10317 + }, + { + "epoch": 0.5764728887895634, + "grad_norm": 0.5182110071182251, + "learning_rate": 0.0007141416405199463, + "loss": 0.511, + "step": 10318 + }, + { + "epoch": 0.5765287593932452, + "grad_norm": 0.48280903697013855, + "learning_rate": 0.0007141136261766025, + "loss": 0.4213, + "step": 10319 + }, + { + "epoch": 0.5765846299969272, + "grad_norm": 0.38309675455093384, + "learning_rate": 0.0007140856118332586, + "loss": 0.4256, + "step": 10320 + }, + { + "epoch": 0.576640500600609, + "grad_norm": 0.8973721861839294, + "learning_rate": 0.0007140575974899148, + "loss": 0.4536, + "step": 10321 + }, + { + "epoch": 0.5766963712042908, + "grad_norm": 0.5180665254592896, + "learning_rate": 0.0007140295831465711, + "loss": 0.5384, + "step": 10322 + }, + { + "epoch": 0.5767522418079727, + "grad_norm": 1.350711464881897, + "learning_rate": 0.0007140015688032273, + "loss": 0.4241, + "step": 10323 + }, + { + "epoch": 0.5768081124116546, + "grad_norm": 0.6836056113243103, + "learning_rate": 0.0007139735544598834, + "loss": 0.4463, + "step": 10324 + }, + { + "epoch": 0.5768639830153365, + "grad_norm": 0.9659157395362854, + "learning_rate": 0.0007139455401165397, + "loss": 0.5822, + "step": 10325 + }, + { + "epoch": 0.5769198536190183, + "grad_norm": 0.6166980862617493, + "learning_rate": 0.0007139175257731959, + "loss": 0.4219, + "step": 10326 + }, + { + "epoch": 0.5769757242227003, + "grad_norm": 0.5400338768959045, + "learning_rate": 0.0007138895114298521, + "loss": 0.39, + "step": 10327 + }, + { + "epoch": 0.5770315948263821, + "grad_norm": 0.44763219356536865, + "learning_rate": 0.0007138614970865082, + "loss": 0.3977, + "step": 10328 + }, + { + "epoch": 0.577087465430064, + "grad_norm": 1.3662537336349487, + "learning_rate": 0.0007138334827431645, + "loss": 0.3695, + "step": 10329 + }, + { + "epoch": 0.5771433360337458, + "grad_norm": 1.7581779956817627, + "learning_rate": 0.0007138054683998207, + "loss": 0.4499, + "step": 10330 + }, + { + "epoch": 0.5771992066374277, + "grad_norm": 0.5704188346862793, + "learning_rate": 0.000713777454056477, + "loss": 0.3696, + "step": 10331 + }, + { + "epoch": 0.5772550772411096, + "grad_norm": 0.474879652261734, + "learning_rate": 0.0007137494397131332, + "loss": 0.4363, + "step": 10332 + }, + { + "epoch": 0.5773109478447914, + "grad_norm": 0.5255080461502075, + "learning_rate": 0.0007137214253697894, + "loss": 0.4038, + "step": 10333 + }, + { + "epoch": 0.5773668184484734, + "grad_norm": 0.7671663165092468, + "learning_rate": 0.0007136934110264456, + "loss": 0.4483, + "step": 10334 + }, + { + "epoch": 0.5774226890521552, + "grad_norm": 2.010115146636963, + "learning_rate": 0.0007136653966831017, + "loss": 0.4707, + "step": 10335 + }, + { + "epoch": 0.577478559655837, + "grad_norm": 0.44805994629859924, + "learning_rate": 0.000713637382339758, + "loss": 0.4992, + "step": 10336 + }, + { + "epoch": 0.577534430259519, + "grad_norm": 0.9622441530227661, + "learning_rate": 0.0007136093679964142, + "loss": 0.4856, + "step": 10337 + }, + { + "epoch": 0.5775903008632008, + "grad_norm": 0.4773328900337219, + "learning_rate": 0.0007135813536530704, + "loss": 0.4963, + "step": 10338 + }, + { + "epoch": 0.5776461714668827, + "grad_norm": 0.4553998112678528, + "learning_rate": 0.0007135533393097265, + "loss": 0.3872, + "step": 10339 + }, + { + "epoch": 0.5777020420705645, + "grad_norm": 0.4789580702781677, + "learning_rate": 0.0007135253249663828, + "loss": 0.4536, + "step": 10340 + }, + { + "epoch": 0.5777579126742465, + "grad_norm": 1.1802852153778076, + "learning_rate": 0.000713497310623039, + "loss": 0.4364, + "step": 10341 + }, + { + "epoch": 0.5778137832779283, + "grad_norm": 0.6158331036567688, + "learning_rate": 0.0007134692962796952, + "loss": 0.4303, + "step": 10342 + }, + { + "epoch": 0.5778696538816102, + "grad_norm": 0.5059067606925964, + "learning_rate": 0.0007134412819363514, + "loss": 0.4636, + "step": 10343 + }, + { + "epoch": 0.5779255244852921, + "grad_norm": 0.4362029731273651, + "learning_rate": 0.0007134132675930076, + "loss": 0.4884, + "step": 10344 + }, + { + "epoch": 0.577981395088974, + "grad_norm": 0.3722774386405945, + "learning_rate": 0.0007133852532496638, + "loss": 0.3694, + "step": 10345 + }, + { + "epoch": 0.5780372656926558, + "grad_norm": 0.8048056364059448, + "learning_rate": 0.00071335723890632, + "loss": 0.4022, + "step": 10346 + }, + { + "epoch": 0.5780931362963376, + "grad_norm": 0.5522124767303467, + "learning_rate": 0.0007133292245629762, + "loss": 0.5375, + "step": 10347 + }, + { + "epoch": 0.5781490069000196, + "grad_norm": 0.4463343620300293, + "learning_rate": 0.0007133012102196325, + "loss": 0.3861, + "step": 10348 + }, + { + "epoch": 0.5782048775037014, + "grad_norm": 0.48629772663116455, + "learning_rate": 0.0007132731958762887, + "loss": 0.4721, + "step": 10349 + }, + { + "epoch": 0.5782607481073833, + "grad_norm": 0.7460613250732422, + "learning_rate": 0.000713245181532945, + "loss": 0.4824, + "step": 10350 + }, + { + "epoch": 0.5783166187110652, + "grad_norm": 1.9668742418289185, + "learning_rate": 0.0007132171671896011, + "loss": 0.5007, + "step": 10351 + }, + { + "epoch": 0.578372489314747, + "grad_norm": 1.196803331375122, + "learning_rate": 0.0007131891528462573, + "loss": 0.5807, + "step": 10352 + }, + { + "epoch": 0.5784283599184289, + "grad_norm": 0.445200651884079, + "learning_rate": 0.0007131611385029135, + "loss": 0.5723, + "step": 10353 + }, + { + "epoch": 0.5784842305221108, + "grad_norm": 0.540052056312561, + "learning_rate": 0.0007131331241595698, + "loss": 0.3904, + "step": 10354 + }, + { + "epoch": 0.5785401011257927, + "grad_norm": 0.48920106887817383, + "learning_rate": 0.0007131051098162259, + "loss": 0.5984, + "step": 10355 + }, + { + "epoch": 0.5785959717294745, + "grad_norm": 4.638386249542236, + "learning_rate": 0.0007130770954728821, + "loss": 0.3939, + "step": 10356 + }, + { + "epoch": 0.5786518423331564, + "grad_norm": 5.258081912994385, + "learning_rate": 0.0007130490811295384, + "loss": 0.3634, + "step": 10357 + }, + { + "epoch": 0.5787077129368383, + "grad_norm": 0.5569864511489868, + "learning_rate": 0.0007130210667861945, + "loss": 0.4612, + "step": 10358 + }, + { + "epoch": 0.5787635835405202, + "grad_norm": 0.526405394077301, + "learning_rate": 0.0007129930524428507, + "loss": 0.4913, + "step": 10359 + }, + { + "epoch": 0.578819454144202, + "grad_norm": 0.664033830165863, + "learning_rate": 0.0007129650380995069, + "loss": 0.6255, + "step": 10360 + }, + { + "epoch": 0.578875324747884, + "grad_norm": 0.4463587999343872, + "learning_rate": 0.0007129370237561632, + "loss": 0.4643, + "step": 10361 + }, + { + "epoch": 0.5789311953515658, + "grad_norm": 0.7499536871910095, + "learning_rate": 0.0007129090094128193, + "loss": 0.55, + "step": 10362 + }, + { + "epoch": 0.5789870659552476, + "grad_norm": 0.4626516103744507, + "learning_rate": 0.0007128809950694755, + "loss": 0.366, + "step": 10363 + }, + { + "epoch": 0.5790429365589295, + "grad_norm": 0.3885541558265686, + "learning_rate": 0.0007128529807261319, + "loss": 0.3853, + "step": 10364 + }, + { + "epoch": 0.5790988071626114, + "grad_norm": 0.5110954642295837, + "learning_rate": 0.0007128249663827881, + "loss": 0.3836, + "step": 10365 + }, + { + "epoch": 0.5791546777662933, + "grad_norm": 1.902910828590393, + "learning_rate": 0.0007127969520394442, + "loss": 0.4932, + "step": 10366 + }, + { + "epoch": 0.5792105483699751, + "grad_norm": 0.8305098414421082, + "learning_rate": 0.0007127689376961004, + "loss": 0.5072, + "step": 10367 + }, + { + "epoch": 0.579266418973657, + "grad_norm": 0.7023335099220276, + "learning_rate": 0.0007127409233527567, + "loss": 0.4778, + "step": 10368 + }, + { + "epoch": 0.5793222895773389, + "grad_norm": 0.7065083980560303, + "learning_rate": 0.0007127129090094129, + "loss": 0.4254, + "step": 10369 + }, + { + "epoch": 0.5793781601810207, + "grad_norm": 0.5374351739883423, + "learning_rate": 0.000712684894666069, + "loss": 0.4342, + "step": 10370 + }, + { + "epoch": 0.5794340307847027, + "grad_norm": 0.48333775997161865, + "learning_rate": 0.0007126568803227253, + "loss": 0.474, + "step": 10371 + }, + { + "epoch": 0.5794899013883845, + "grad_norm": 0.48219993710517883, + "learning_rate": 0.0007126288659793815, + "loss": 0.5374, + "step": 10372 + }, + { + "epoch": 0.5795457719920664, + "grad_norm": 0.4672122001647949, + "learning_rate": 0.0007126008516360377, + "loss": 0.3826, + "step": 10373 + }, + { + "epoch": 0.5796016425957482, + "grad_norm": 0.6266250014305115, + "learning_rate": 0.0007125728372926938, + "loss": 0.4381, + "step": 10374 + }, + { + "epoch": 0.5796575131994302, + "grad_norm": 1.1360177993774414, + "learning_rate": 0.0007125448229493501, + "loss": 0.4387, + "step": 10375 + }, + { + "epoch": 0.579713383803112, + "grad_norm": 0.4878767728805542, + "learning_rate": 0.0007125168086060063, + "loss": 0.4527, + "step": 10376 + }, + { + "epoch": 0.5797692544067938, + "grad_norm": 0.741028904914856, + "learning_rate": 0.0007124887942626624, + "loss": 0.3812, + "step": 10377 + }, + { + "epoch": 0.5798251250104758, + "grad_norm": 0.43169498443603516, + "learning_rate": 0.0007124607799193186, + "loss": 0.4359, + "step": 10378 + }, + { + "epoch": 0.5798809956141576, + "grad_norm": 0.4134253263473511, + "learning_rate": 0.0007124327655759749, + "loss": 0.4125, + "step": 10379 + }, + { + "epoch": 0.5799368662178395, + "grad_norm": 0.4489768147468567, + "learning_rate": 0.0007124047512326311, + "loss": 0.3977, + "step": 10380 + }, + { + "epoch": 0.5799927368215213, + "grad_norm": 0.9099529385566711, + "learning_rate": 0.0007123767368892872, + "loss": 0.3666, + "step": 10381 + }, + { + "epoch": 0.5800486074252033, + "grad_norm": 0.43378764390945435, + "learning_rate": 0.0007123487225459436, + "loss": 0.3738, + "step": 10382 + }, + { + "epoch": 0.5801044780288851, + "grad_norm": 0.5279557704925537, + "learning_rate": 0.0007123207082025998, + "loss": 0.4219, + "step": 10383 + }, + { + "epoch": 0.5801603486325669, + "grad_norm": 0.5167741179466248, + "learning_rate": 0.000712292693859256, + "loss": 0.3799, + "step": 10384 + }, + { + "epoch": 0.5802162192362489, + "grad_norm": 0.4591062366962433, + "learning_rate": 0.0007122646795159121, + "loss": 0.4822, + "step": 10385 + }, + { + "epoch": 0.5802720898399307, + "grad_norm": 0.41829991340637207, + "learning_rate": 0.0007122366651725684, + "loss": 0.4247, + "step": 10386 + }, + { + "epoch": 0.5803279604436126, + "grad_norm": 0.8200260996818542, + "learning_rate": 0.0007122086508292246, + "loss": 0.6607, + "step": 10387 + }, + { + "epoch": 0.5803838310472945, + "grad_norm": 0.7788015604019165, + "learning_rate": 0.0007121806364858808, + "loss": 0.4306, + "step": 10388 + }, + { + "epoch": 0.5804397016509764, + "grad_norm": 1.5087569952011108, + "learning_rate": 0.000712152622142537, + "loss": 0.4358, + "step": 10389 + }, + { + "epoch": 0.5804955722546582, + "grad_norm": 0.6222716569900513, + "learning_rate": 0.0007121246077991932, + "loss": 0.4234, + "step": 10390 + }, + { + "epoch": 0.58055144285834, + "grad_norm": 0.7910364270210266, + "learning_rate": 0.0007120965934558494, + "loss": 0.4488, + "step": 10391 + }, + { + "epoch": 0.580607313462022, + "grad_norm": 5.514947891235352, + "learning_rate": 0.0007120685791125056, + "loss": 0.4021, + "step": 10392 + }, + { + "epoch": 0.5806631840657038, + "grad_norm": 0.6676428914070129, + "learning_rate": 0.0007120405647691618, + "loss": 0.5858, + "step": 10393 + }, + { + "epoch": 0.5807190546693857, + "grad_norm": 2.5711894035339355, + "learning_rate": 0.000712012550425818, + "loss": 0.4671, + "step": 10394 + }, + { + "epoch": 0.5807749252730676, + "grad_norm": 1.3241738080978394, + "learning_rate": 0.0007119845360824742, + "loss": 0.7605, + "step": 10395 + }, + { + "epoch": 0.5808307958767495, + "grad_norm": 0.4819618761539459, + "learning_rate": 0.0007119565217391305, + "loss": 0.5467, + "step": 10396 + }, + { + "epoch": 0.5808866664804313, + "grad_norm": 0.49792003631591797, + "learning_rate": 0.0007119285073957866, + "loss": 0.4113, + "step": 10397 + }, + { + "epoch": 0.5809425370841131, + "grad_norm": 0.46255436539649963, + "learning_rate": 0.0007119004930524429, + "loss": 0.4118, + "step": 10398 + }, + { + "epoch": 0.5809984076877951, + "grad_norm": 0.5191859006881714, + "learning_rate": 0.0007118724787090991, + "loss": 0.4811, + "step": 10399 + }, + { + "epoch": 0.5810542782914769, + "grad_norm": 0.42386484146118164, + "learning_rate": 0.0007118444643657553, + "loss": 0.3534, + "step": 10400 + }, + { + "epoch": 0.5811101488951588, + "grad_norm": 0.789578914642334, + "learning_rate": 0.0007118164500224115, + "loss": 0.7066, + "step": 10401 + }, + { + "epoch": 0.5811660194988407, + "grad_norm": 0.5437617897987366, + "learning_rate": 0.0007117884356790677, + "loss": 0.5473, + "step": 10402 + }, + { + "epoch": 0.5812218901025226, + "grad_norm": 0.4598684310913086, + "learning_rate": 0.000711760421335724, + "loss": 0.4403, + "step": 10403 + }, + { + "epoch": 0.5812777607062044, + "grad_norm": 0.871187150478363, + "learning_rate": 0.0007117324069923801, + "loss": 0.5903, + "step": 10404 + }, + { + "epoch": 0.5813336313098864, + "grad_norm": 0.6397603750228882, + "learning_rate": 0.0007117043926490363, + "loss": 0.5376, + "step": 10405 + }, + { + "epoch": 0.5813895019135682, + "grad_norm": 0.8317081928253174, + "learning_rate": 0.0007116763783056925, + "loss": 0.4272, + "step": 10406 + }, + { + "epoch": 0.58144537251725, + "grad_norm": 4.521035194396973, + "learning_rate": 0.0007116483639623488, + "loss": 0.3665, + "step": 10407 + }, + { + "epoch": 0.5815012431209319, + "grad_norm": 0.47313788533210754, + "learning_rate": 0.0007116203496190049, + "loss": 0.3961, + "step": 10408 + }, + { + "epoch": 0.5815571137246138, + "grad_norm": 0.5657212138175964, + "learning_rate": 0.0007115923352756611, + "loss": 0.5574, + "step": 10409 + }, + { + "epoch": 0.5816129843282957, + "grad_norm": 0.8840219974517822, + "learning_rate": 0.0007115643209323174, + "loss": 0.5641, + "step": 10410 + }, + { + "epoch": 0.5816688549319775, + "grad_norm": 0.6096137762069702, + "learning_rate": 0.0007115363065889736, + "loss": 0.3439, + "step": 10411 + }, + { + "epoch": 0.5817247255356595, + "grad_norm": 0.40740329027175903, + "learning_rate": 0.0007115082922456297, + "loss": 0.4494, + "step": 10412 + }, + { + "epoch": 0.5817805961393413, + "grad_norm": 0.4956085681915283, + "learning_rate": 0.0007114802779022859, + "loss": 0.4725, + "step": 10413 + }, + { + "epoch": 0.5818364667430231, + "grad_norm": 0.5981189012527466, + "learning_rate": 0.0007114522635589423, + "loss": 0.3682, + "step": 10414 + }, + { + "epoch": 0.581892337346705, + "grad_norm": 1.0511904954910278, + "learning_rate": 0.0007114242492155985, + "loss": 0.449, + "step": 10415 + }, + { + "epoch": 0.5819482079503869, + "grad_norm": 0.44727277755737305, + "learning_rate": 0.0007113962348722546, + "loss": 0.4295, + "step": 10416 + }, + { + "epoch": 0.5820040785540688, + "grad_norm": 0.9529737234115601, + "learning_rate": 0.0007113682205289108, + "loss": 0.3398, + "step": 10417 + }, + { + "epoch": 0.5820599491577506, + "grad_norm": 1.7675217390060425, + "learning_rate": 0.0007113402061855671, + "loss": 0.4698, + "step": 10418 + }, + { + "epoch": 0.5821158197614326, + "grad_norm": 0.5289393663406372, + "learning_rate": 0.0007113121918422233, + "loss": 0.4274, + "step": 10419 + }, + { + "epoch": 0.5821716903651144, + "grad_norm": 0.7507824897766113, + "learning_rate": 0.0007112841774988794, + "loss": 0.4109, + "step": 10420 + }, + { + "epoch": 0.5822275609687962, + "grad_norm": 0.5017775893211365, + "learning_rate": 0.0007112561631555357, + "loss": 0.4818, + "step": 10421 + }, + { + "epoch": 0.5822834315724782, + "grad_norm": 0.5254977345466614, + "learning_rate": 0.0007112281488121919, + "loss": 0.3999, + "step": 10422 + }, + { + "epoch": 0.58233930217616, + "grad_norm": 0.5582948327064514, + "learning_rate": 0.000711200134468848, + "loss": 0.3671, + "step": 10423 + }, + { + "epoch": 0.5823951727798419, + "grad_norm": 2.452111005783081, + "learning_rate": 0.0007111721201255042, + "loss": 0.5062, + "step": 10424 + }, + { + "epoch": 0.5824510433835237, + "grad_norm": 0.5343247652053833, + "learning_rate": 0.0007111441057821605, + "loss": 0.5493, + "step": 10425 + }, + { + "epoch": 0.5825069139872057, + "grad_norm": 0.8896372318267822, + "learning_rate": 0.0007111160914388167, + "loss": 0.4127, + "step": 10426 + }, + { + "epoch": 0.5825627845908875, + "grad_norm": 1.3722203969955444, + "learning_rate": 0.0007110880770954728, + "loss": 0.3917, + "step": 10427 + }, + { + "epoch": 0.5826186551945693, + "grad_norm": 3.3964765071868896, + "learning_rate": 0.0007110600627521291, + "loss": 0.4613, + "step": 10428 + }, + { + "epoch": 0.5826745257982513, + "grad_norm": 0.5284232497215271, + "learning_rate": 0.0007110320484087853, + "loss": 0.5329, + "step": 10429 + }, + { + "epoch": 0.5827303964019331, + "grad_norm": 0.7681365013122559, + "learning_rate": 0.0007110040340654415, + "loss": 0.3883, + "step": 10430 + }, + { + "epoch": 0.582786267005615, + "grad_norm": 1.2938379049301147, + "learning_rate": 0.0007109760197220976, + "loss": 0.4584, + "step": 10431 + }, + { + "epoch": 0.5828421376092968, + "grad_norm": 0.4414912462234497, + "learning_rate": 0.000710948005378754, + "loss": 0.4261, + "step": 10432 + }, + { + "epoch": 0.5828980082129788, + "grad_norm": 0.5072042942047119, + "learning_rate": 0.0007109199910354102, + "loss": 0.3914, + "step": 10433 + }, + { + "epoch": 0.5829538788166606, + "grad_norm": 0.5476480722427368, + "learning_rate": 0.0007108919766920664, + "loss": 0.353, + "step": 10434 + }, + { + "epoch": 0.5830097494203424, + "grad_norm": 0.4134286642074585, + "learning_rate": 0.0007108639623487226, + "loss": 0.3397, + "step": 10435 + }, + { + "epoch": 0.5830656200240244, + "grad_norm": 0.5597545504570007, + "learning_rate": 0.0007108359480053788, + "loss": 0.4519, + "step": 10436 + }, + { + "epoch": 0.5831214906277062, + "grad_norm": 0.4564744234085083, + "learning_rate": 0.000710807933662035, + "loss": 0.461, + "step": 10437 + }, + { + "epoch": 0.5831773612313881, + "grad_norm": 0.5072201490402222, + "learning_rate": 0.0007107799193186912, + "loss": 0.3548, + "step": 10438 + }, + { + "epoch": 0.58323323183507, + "grad_norm": 0.6778005957603455, + "learning_rate": 0.0007107519049753474, + "loss": 0.3696, + "step": 10439 + }, + { + "epoch": 0.5832891024387519, + "grad_norm": 0.5806835889816284, + "learning_rate": 0.0007107238906320036, + "loss": 0.4209, + "step": 10440 + }, + { + "epoch": 0.5833449730424337, + "grad_norm": 0.539847731590271, + "learning_rate": 0.0007106958762886598, + "loss": 0.3925, + "step": 10441 + }, + { + "epoch": 0.5834008436461156, + "grad_norm": 0.5126388669013977, + "learning_rate": 0.000710667861945316, + "loss": 0.4614, + "step": 10442 + }, + { + "epoch": 0.5834567142497975, + "grad_norm": 0.36958837509155273, + "learning_rate": 0.0007106398476019722, + "loss": 0.3696, + "step": 10443 + }, + { + "epoch": 0.5835125848534793, + "grad_norm": 0.6057869791984558, + "learning_rate": 0.0007106118332586284, + "loss": 0.343, + "step": 10444 + }, + { + "epoch": 0.5835684554571612, + "grad_norm": 0.8230266571044922, + "learning_rate": 0.0007105838189152846, + "loss": 0.4307, + "step": 10445 + }, + { + "epoch": 0.5836243260608431, + "grad_norm": 1.1856259107589722, + "learning_rate": 0.0007105558045719408, + "loss": 0.4926, + "step": 10446 + }, + { + "epoch": 0.583680196664525, + "grad_norm": 1.2864859104156494, + "learning_rate": 0.000710527790228597, + "loss": 0.4445, + "step": 10447 + }, + { + "epoch": 0.5837360672682068, + "grad_norm": 0.8265144228935242, + "learning_rate": 0.0007104997758852533, + "loss": 0.5748, + "step": 10448 + }, + { + "epoch": 0.5837919378718887, + "grad_norm": 0.4856041371822357, + "learning_rate": 0.0007104717615419096, + "loss": 0.4461, + "step": 10449 + }, + { + "epoch": 0.5838478084755706, + "grad_norm": 0.4631437361240387, + "learning_rate": 0.0007104437471985657, + "loss": 0.4539, + "step": 10450 + }, + { + "epoch": 0.5839036790792524, + "grad_norm": 0.3622188866138458, + "learning_rate": 0.0007104157328552219, + "loss": 0.3329, + "step": 10451 + }, + { + "epoch": 0.5839595496829343, + "grad_norm": 0.5753214955329895, + "learning_rate": 0.0007103877185118781, + "loss": 0.4181, + "step": 10452 + }, + { + "epoch": 0.5840154202866162, + "grad_norm": 0.5772531032562256, + "learning_rate": 0.0007103597041685344, + "loss": 0.467, + "step": 10453 + }, + { + "epoch": 0.5840712908902981, + "grad_norm": 4.041547775268555, + "learning_rate": 0.0007103316898251905, + "loss": 0.3682, + "step": 10454 + }, + { + "epoch": 0.5841271614939799, + "grad_norm": 0.36424747109413147, + "learning_rate": 0.0007103036754818467, + "loss": 0.4333, + "step": 10455 + }, + { + "epoch": 0.5841830320976619, + "grad_norm": 0.5251452922821045, + "learning_rate": 0.0007102756611385029, + "loss": 0.5191, + "step": 10456 + }, + { + "epoch": 0.5842389027013437, + "grad_norm": 0.6493807435035706, + "learning_rate": 0.0007102476467951592, + "loss": 0.3944, + "step": 10457 + }, + { + "epoch": 0.5842947733050255, + "grad_norm": 0.5333901047706604, + "learning_rate": 0.0007102196324518153, + "loss": 0.4691, + "step": 10458 + }, + { + "epoch": 0.5843506439087074, + "grad_norm": 0.5327048301696777, + "learning_rate": 0.0007101916181084715, + "loss": 0.4274, + "step": 10459 + }, + { + "epoch": 0.5844065145123893, + "grad_norm": 1.0474193096160889, + "learning_rate": 0.0007101636037651278, + "loss": 0.4702, + "step": 10460 + }, + { + "epoch": 0.5844623851160712, + "grad_norm": 0.5701755881309509, + "learning_rate": 0.000710135589421784, + "loss": 0.4198, + "step": 10461 + }, + { + "epoch": 0.584518255719753, + "grad_norm": 0.6537631750106812, + "learning_rate": 0.0007101075750784401, + "loss": 0.5382, + "step": 10462 + }, + { + "epoch": 0.584574126323435, + "grad_norm": 0.624996542930603, + "learning_rate": 0.0007100795607350963, + "loss": 0.4648, + "step": 10463 + }, + { + "epoch": 0.5846299969271168, + "grad_norm": 1.0407466888427734, + "learning_rate": 0.0007100515463917527, + "loss": 0.5543, + "step": 10464 + }, + { + "epoch": 0.5846858675307987, + "grad_norm": 0.7280141115188599, + "learning_rate": 0.0007100235320484087, + "loss": 0.4206, + "step": 10465 + }, + { + "epoch": 0.5847417381344805, + "grad_norm": 0.4898303747177124, + "learning_rate": 0.000709995517705065, + "loss": 0.519, + "step": 10466 + }, + { + "epoch": 0.5847976087381624, + "grad_norm": 0.7544603943824768, + "learning_rate": 0.0007099675033617213, + "loss": 0.5429, + "step": 10467 + }, + { + "epoch": 0.5848534793418443, + "grad_norm": 0.888749897480011, + "learning_rate": 0.0007099394890183775, + "loss": 0.8934, + "step": 10468 + }, + { + "epoch": 0.5849093499455261, + "grad_norm": 0.6428008675575256, + "learning_rate": 0.0007099114746750336, + "loss": 0.5121, + "step": 10469 + }, + { + "epoch": 0.5849652205492081, + "grad_norm": 0.6439871191978455, + "learning_rate": 0.0007098834603316898, + "loss": 0.4484, + "step": 10470 + }, + { + "epoch": 0.5850210911528899, + "grad_norm": 0.48000970482826233, + "learning_rate": 0.0007098554459883461, + "loss": 0.4893, + "step": 10471 + }, + { + "epoch": 0.5850769617565718, + "grad_norm": 0.8577825427055359, + "learning_rate": 0.0007098274316450023, + "loss": 0.4336, + "step": 10472 + }, + { + "epoch": 0.5851328323602537, + "grad_norm": 0.9751829504966736, + "learning_rate": 0.0007097994173016584, + "loss": 0.4558, + "step": 10473 + }, + { + "epoch": 0.5851887029639355, + "grad_norm": 1.3155993223190308, + "learning_rate": 0.0007097714029583147, + "loss": 0.4169, + "step": 10474 + }, + { + "epoch": 0.5852445735676174, + "grad_norm": 0.51240473985672, + "learning_rate": 0.0007097433886149709, + "loss": 0.4741, + "step": 10475 + }, + { + "epoch": 0.5853004441712992, + "grad_norm": 0.5476670861244202, + "learning_rate": 0.0007097153742716271, + "loss": 0.5692, + "step": 10476 + }, + { + "epoch": 0.5853563147749812, + "grad_norm": 0.4675905108451843, + "learning_rate": 0.0007096873599282832, + "loss": 0.434, + "step": 10477 + }, + { + "epoch": 0.585412185378663, + "grad_norm": 0.7967668175697327, + "learning_rate": 0.0007096593455849395, + "loss": 0.5978, + "step": 10478 + }, + { + "epoch": 0.5854680559823449, + "grad_norm": 5.975982189178467, + "learning_rate": 0.0007096313312415957, + "loss": 0.5252, + "step": 10479 + }, + { + "epoch": 0.5855239265860268, + "grad_norm": 0.6371981501579285, + "learning_rate": 0.0007096033168982519, + "loss": 0.5811, + "step": 10480 + }, + { + "epoch": 0.5855797971897087, + "grad_norm": 12.299577713012695, + "learning_rate": 0.0007095753025549082, + "loss": 0.493, + "step": 10481 + }, + { + "epoch": 0.5856356677933905, + "grad_norm": 0.8576894998550415, + "learning_rate": 0.0007095472882115644, + "loss": 0.3973, + "step": 10482 + }, + { + "epoch": 0.5856915383970723, + "grad_norm": 0.5336518287658691, + "learning_rate": 0.0007095192738682206, + "loss": 0.3294, + "step": 10483 + }, + { + "epoch": 0.5857474090007543, + "grad_norm": 0.8593136072158813, + "learning_rate": 0.0007094912595248767, + "loss": 0.4205, + "step": 10484 + }, + { + "epoch": 0.5858032796044361, + "grad_norm": 1.0719473361968994, + "learning_rate": 0.000709463245181533, + "loss": 0.4995, + "step": 10485 + }, + { + "epoch": 0.585859150208118, + "grad_norm": 0.4097357392311096, + "learning_rate": 0.0007094352308381892, + "loss": 0.4595, + "step": 10486 + }, + { + "epoch": 0.5859150208117999, + "grad_norm": 0.45105215907096863, + "learning_rate": 0.0007094072164948454, + "loss": 0.426, + "step": 10487 + }, + { + "epoch": 0.5859708914154818, + "grad_norm": 0.6869425773620605, + "learning_rate": 0.0007093792021515015, + "loss": 0.4221, + "step": 10488 + }, + { + "epoch": 0.5860267620191636, + "grad_norm": 0.522962749004364, + "learning_rate": 0.0007093511878081578, + "loss": 0.4798, + "step": 10489 + }, + { + "epoch": 0.5860826326228455, + "grad_norm": 0.5609034895896912, + "learning_rate": 0.000709323173464814, + "loss": 0.4546, + "step": 10490 + }, + { + "epoch": 0.5861385032265274, + "grad_norm": 0.5600711107254028, + "learning_rate": 0.0007092951591214702, + "loss": 0.5345, + "step": 10491 + }, + { + "epoch": 0.5861943738302092, + "grad_norm": 13.792756080627441, + "learning_rate": 0.0007092671447781264, + "loss": 0.5713, + "step": 10492 + }, + { + "epoch": 0.5862502444338911, + "grad_norm": 0.3860197067260742, + "learning_rate": 0.0007092391304347826, + "loss": 0.4626, + "step": 10493 + }, + { + "epoch": 0.586306115037573, + "grad_norm": 0.5321462750434875, + "learning_rate": 0.0007092111160914388, + "loss": 0.4489, + "step": 10494 + }, + { + "epoch": 0.5863619856412549, + "grad_norm": 0.41418349742889404, + "learning_rate": 0.000709183101748095, + "loss": 0.417, + "step": 10495 + }, + { + "epoch": 0.5864178562449367, + "grad_norm": 0.4913509786128998, + "learning_rate": 0.0007091550874047512, + "loss": 0.4267, + "step": 10496 + }, + { + "epoch": 0.5864737268486186, + "grad_norm": 0.7091698050498962, + "learning_rate": 0.0007091270730614074, + "loss": 0.4527, + "step": 10497 + }, + { + "epoch": 0.5865295974523005, + "grad_norm": 0.6013452410697937, + "learning_rate": 0.0007090990587180637, + "loss": 0.4484, + "step": 10498 + }, + { + "epoch": 0.5865854680559823, + "grad_norm": 0.4284360408782959, + "learning_rate": 0.00070907104437472, + "loss": 0.5056, + "step": 10499 + }, + { + "epoch": 0.5866413386596642, + "grad_norm": 0.45183828473091125, + "learning_rate": 0.0007090430300313761, + "loss": 0.4943, + "step": 10500 + }, + { + "epoch": 0.5866413386596642, + "eval_cer": 0.09257743881815107, + "eval_loss": 0.3443659543991089, + "eval_runtime": 56.2113, + "eval_samples_per_second": 80.731, + "eval_steps_per_second": 5.052, + "eval_wer": 0.3701013938158152, + "step": 10500 + }, + { + "epoch": 0.5866972092633461, + "grad_norm": 0.5457750558853149, + "learning_rate": 0.0007090150156880323, + "loss": 0.4935, + "step": 10501 + }, + { + "epoch": 0.586753079867028, + "grad_norm": 0.7841604948043823, + "learning_rate": 0.0007089870013446885, + "loss": 0.6175, + "step": 10502 + }, + { + "epoch": 0.5868089504707098, + "grad_norm": 1.0254507064819336, + "learning_rate": 0.0007089589870013448, + "loss": 0.4889, + "step": 10503 + }, + { + "epoch": 0.5868648210743918, + "grad_norm": 0.4320913553237915, + "learning_rate": 0.0007089309726580009, + "loss": 0.4737, + "step": 10504 + }, + { + "epoch": 0.5869206916780736, + "grad_norm": 3.8105900287628174, + "learning_rate": 0.0007089029583146571, + "loss": 0.456, + "step": 10505 + }, + { + "epoch": 0.5869765622817554, + "grad_norm": 0.5584713816642761, + "learning_rate": 0.0007088749439713134, + "loss": 0.4388, + "step": 10506 + }, + { + "epoch": 0.5870324328854373, + "grad_norm": 0.5289224982261658, + "learning_rate": 0.0007088469296279695, + "loss": 0.3742, + "step": 10507 + }, + { + "epoch": 0.5870883034891192, + "grad_norm": 0.5687105655670166, + "learning_rate": 0.0007088189152846257, + "loss": 0.5806, + "step": 10508 + }, + { + "epoch": 0.5871441740928011, + "grad_norm": 0.4095568060874939, + "learning_rate": 0.0007087909009412819, + "loss": 0.5361, + "step": 10509 + }, + { + "epoch": 0.5872000446964829, + "grad_norm": 1.3863614797592163, + "learning_rate": 0.0007087628865979382, + "loss": 0.3794, + "step": 10510 + }, + { + "epoch": 0.5872559153001649, + "grad_norm": 1.2127840518951416, + "learning_rate": 0.0007087348722545943, + "loss": 0.4443, + "step": 10511 + }, + { + "epoch": 0.5873117859038467, + "grad_norm": 0.6346824169158936, + "learning_rate": 0.0007087068579112505, + "loss": 0.4406, + "step": 10512 + }, + { + "epoch": 0.5873676565075285, + "grad_norm": 0.6538674831390381, + "learning_rate": 0.0007086788435679068, + "loss": 0.4511, + "step": 10513 + }, + { + "epoch": 0.5874235271112105, + "grad_norm": 0.597610592842102, + "learning_rate": 0.000708650829224563, + "loss": 0.5939, + "step": 10514 + }, + { + "epoch": 0.5874793977148923, + "grad_norm": 0.585281252861023, + "learning_rate": 0.0007086228148812191, + "loss": 0.4105, + "step": 10515 + }, + { + "epoch": 0.5875352683185742, + "grad_norm": 0.48217594623565674, + "learning_rate": 0.0007085948005378754, + "loss": 0.4712, + "step": 10516 + }, + { + "epoch": 0.587591138922256, + "grad_norm": 0.3446861505508423, + "learning_rate": 0.0007085667861945317, + "loss": 0.4426, + "step": 10517 + }, + { + "epoch": 0.587647009525938, + "grad_norm": 0.9893266558647156, + "learning_rate": 0.0007085387718511879, + "loss": 0.5376, + "step": 10518 + }, + { + "epoch": 0.5877028801296198, + "grad_norm": 0.44841891527175903, + "learning_rate": 0.000708510757507844, + "loss": 0.4552, + "step": 10519 + }, + { + "epoch": 0.5877587507333016, + "grad_norm": 0.408349871635437, + "learning_rate": 0.0007084827431645002, + "loss": 0.4712, + "step": 10520 + }, + { + "epoch": 0.5878146213369836, + "grad_norm": 0.6875362396240234, + "learning_rate": 0.0007084547288211565, + "loss": 0.469, + "step": 10521 + }, + { + "epoch": 0.5878704919406654, + "grad_norm": 0.5102987289428711, + "learning_rate": 0.0007084267144778127, + "loss": 0.3781, + "step": 10522 + }, + { + "epoch": 0.5879263625443473, + "grad_norm": 0.820102870464325, + "learning_rate": 0.0007083987001344688, + "loss": 0.6231, + "step": 10523 + }, + { + "epoch": 0.5879822331480291, + "grad_norm": 0.5872384309768677, + "learning_rate": 0.0007083706857911251, + "loss": 0.4515, + "step": 10524 + }, + { + "epoch": 0.5880381037517111, + "grad_norm": 1.9699509143829346, + "learning_rate": 0.0007083426714477813, + "loss": 0.6221, + "step": 10525 + }, + { + "epoch": 0.5880939743553929, + "grad_norm": 1.0874755382537842, + "learning_rate": 0.0007083146571044374, + "loss": 0.4582, + "step": 10526 + }, + { + "epoch": 0.5881498449590747, + "grad_norm": 0.34314876794815063, + "learning_rate": 0.0007082866427610936, + "loss": 0.267, + "step": 10527 + }, + { + "epoch": 0.5882057155627567, + "grad_norm": 1.2932288646697998, + "learning_rate": 0.0007082586284177499, + "loss": 0.4939, + "step": 10528 + }, + { + "epoch": 0.5882615861664385, + "grad_norm": 0.4819102883338928, + "learning_rate": 0.0007082306140744061, + "loss": 0.4776, + "step": 10529 + }, + { + "epoch": 0.5883174567701204, + "grad_norm": 0.4819238781929016, + "learning_rate": 0.0007082025997310622, + "loss": 0.4394, + "step": 10530 + }, + { + "epoch": 0.5883733273738023, + "grad_norm": 1.2381831407546997, + "learning_rate": 0.0007081745853877186, + "loss": 0.497, + "step": 10531 + }, + { + "epoch": 0.5884291979774842, + "grad_norm": 0.5975236892700195, + "learning_rate": 0.0007081465710443748, + "loss": 0.4142, + "step": 10532 + }, + { + "epoch": 0.588485068581166, + "grad_norm": 0.472127765417099, + "learning_rate": 0.000708118556701031, + "loss": 0.4541, + "step": 10533 + }, + { + "epoch": 0.5885409391848478, + "grad_norm": 3.301687240600586, + "learning_rate": 0.0007080905423576871, + "loss": 0.4853, + "step": 10534 + }, + { + "epoch": 0.5885968097885298, + "grad_norm": 0.9417245388031006, + "learning_rate": 0.0007080625280143434, + "loss": 0.4356, + "step": 10535 + }, + { + "epoch": 0.5886526803922116, + "grad_norm": 0.49267831444740295, + "learning_rate": 0.0007080345136709996, + "loss": 0.4472, + "step": 10536 + }, + { + "epoch": 0.5887085509958935, + "grad_norm": 0.5449758172035217, + "learning_rate": 0.0007080064993276558, + "loss": 0.4622, + "step": 10537 + }, + { + "epoch": 0.5887644215995754, + "grad_norm": 0.7734169363975525, + "learning_rate": 0.000707978484984312, + "loss": 0.4349, + "step": 10538 + }, + { + "epoch": 0.5888202922032573, + "grad_norm": 1.3837463855743408, + "learning_rate": 0.0007079504706409682, + "loss": 0.4265, + "step": 10539 + }, + { + "epoch": 0.5888761628069391, + "grad_norm": 0.48832234740257263, + "learning_rate": 0.0007079224562976244, + "loss": 0.5259, + "step": 10540 + }, + { + "epoch": 0.588932033410621, + "grad_norm": 0.4604237675666809, + "learning_rate": 0.0007078944419542806, + "loss": 0.3925, + "step": 10541 + }, + { + "epoch": 0.5889879040143029, + "grad_norm": 0.6220524907112122, + "learning_rate": 0.0007078664276109368, + "loss": 0.5956, + "step": 10542 + }, + { + "epoch": 0.5890437746179847, + "grad_norm": 0.3266134560108185, + "learning_rate": 0.000707838413267593, + "loss": 0.2893, + "step": 10543 + }, + { + "epoch": 0.5890996452216666, + "grad_norm": 0.566933274269104, + "learning_rate": 0.0007078103989242492, + "loss": 0.3469, + "step": 10544 + }, + { + "epoch": 0.5891555158253485, + "grad_norm": 0.558926522731781, + "learning_rate": 0.0007077823845809055, + "loss": 0.4687, + "step": 10545 + }, + { + "epoch": 0.5892113864290304, + "grad_norm": 0.6042711138725281, + "learning_rate": 0.0007077543702375616, + "loss": 0.4782, + "step": 10546 + }, + { + "epoch": 0.5892672570327122, + "grad_norm": 0.8021321296691895, + "learning_rate": 0.0007077263558942178, + "loss": 0.4836, + "step": 10547 + }, + { + "epoch": 0.5893231276363942, + "grad_norm": 0.398113876581192, + "learning_rate": 0.000707698341550874, + "loss": 0.3496, + "step": 10548 + }, + { + "epoch": 0.589378998240076, + "grad_norm": 2.818457841873169, + "learning_rate": 0.0007076703272075303, + "loss": 0.3994, + "step": 10549 + }, + { + "epoch": 0.5894348688437578, + "grad_norm": 0.6505283713340759, + "learning_rate": 0.0007076423128641865, + "loss": 0.4994, + "step": 10550 + }, + { + "epoch": 0.5894907394474397, + "grad_norm": 0.5240305662155151, + "learning_rate": 0.0007076142985208427, + "loss": 0.371, + "step": 10551 + }, + { + "epoch": 0.5895466100511216, + "grad_norm": 0.6182552576065063, + "learning_rate": 0.000707586284177499, + "loss": 0.4162, + "step": 10552 + }, + { + "epoch": 0.5896024806548035, + "grad_norm": 0.5719916820526123, + "learning_rate": 0.0007075582698341551, + "loss": 0.4487, + "step": 10553 + }, + { + "epoch": 0.5896583512584853, + "grad_norm": 0.6066009402275085, + "learning_rate": 0.0007075302554908113, + "loss": 0.5182, + "step": 10554 + }, + { + "epoch": 0.5897142218621673, + "grad_norm": 0.433051735162735, + "learning_rate": 0.0007075022411474675, + "loss": 0.3518, + "step": 10555 + }, + { + "epoch": 0.5897700924658491, + "grad_norm": 0.4246828556060791, + "learning_rate": 0.0007074742268041238, + "loss": 0.4659, + "step": 10556 + }, + { + "epoch": 0.589825963069531, + "grad_norm": 4.291890621185303, + "learning_rate": 0.0007074462124607799, + "loss": 1.6421, + "step": 10557 + }, + { + "epoch": 0.5898818336732128, + "grad_norm": 0.42692527174949646, + "learning_rate": 0.0007074181981174361, + "loss": 0.421, + "step": 10558 + }, + { + "epoch": 0.5899377042768947, + "grad_norm": 0.4082728326320648, + "learning_rate": 0.0007073901837740923, + "loss": 0.4015, + "step": 10559 + }, + { + "epoch": 0.5899935748805766, + "grad_norm": 0.5416547060012817, + "learning_rate": 0.0007073621694307486, + "loss": 0.4226, + "step": 10560 + }, + { + "epoch": 0.5900494454842584, + "grad_norm": 0.368289589881897, + "learning_rate": 0.0007073341550874047, + "loss": 0.3795, + "step": 10561 + }, + { + "epoch": 0.5901053160879404, + "grad_norm": 1.856859564781189, + "learning_rate": 0.0007073061407440609, + "loss": 0.413, + "step": 10562 + }, + { + "epoch": 0.5901611866916222, + "grad_norm": 1.170412540435791, + "learning_rate": 0.0007072781264007172, + "loss": 0.5559, + "step": 10563 + }, + { + "epoch": 0.590217057295304, + "grad_norm": 0.3994736671447754, + "learning_rate": 0.0007072501120573735, + "loss": 0.4413, + "step": 10564 + }, + { + "epoch": 0.590272927898986, + "grad_norm": 0.37837135791778564, + "learning_rate": 0.0007072220977140295, + "loss": 0.3275, + "step": 10565 + }, + { + "epoch": 0.5903287985026678, + "grad_norm": 0.5266129970550537, + "learning_rate": 0.0007071940833706858, + "loss": 0.5281, + "step": 10566 + }, + { + "epoch": 0.5903846691063497, + "grad_norm": 0.9741831421852112, + "learning_rate": 0.0007071660690273421, + "loss": 0.4494, + "step": 10567 + }, + { + "epoch": 0.5904405397100315, + "grad_norm": 0.4782985746860504, + "learning_rate": 0.0007071380546839982, + "loss": 0.4478, + "step": 10568 + }, + { + "epoch": 0.5904964103137135, + "grad_norm": Infinity, + "learning_rate": 0.0007071380546839982, + "loss": 0.5585, + "step": 10569 + }, + { + "epoch": 0.5905522809173953, + "grad_norm": 0.5587042570114136, + "learning_rate": 0.0007071100403406544, + "loss": 0.391, + "step": 10570 + }, + { + "epoch": 0.5906081515210772, + "grad_norm": 0.4625883996486664, + "learning_rate": 0.0007070820259973107, + "loss": 0.4573, + "step": 10571 + }, + { + "epoch": 0.5906640221247591, + "grad_norm": 0.6065558791160583, + "learning_rate": 0.0007070540116539669, + "loss": 0.4263, + "step": 10572 + }, + { + "epoch": 0.5907198927284409, + "grad_norm": 0.4138694405555725, + "learning_rate": 0.000707025997310623, + "loss": 0.4048, + "step": 10573 + }, + { + "epoch": 0.5907757633321228, + "grad_norm": 0.8945753574371338, + "learning_rate": 0.0007069979829672792, + "loss": 0.5746, + "step": 10574 + }, + { + "epoch": 0.5908316339358046, + "grad_norm": 2.1406073570251465, + "learning_rate": 0.0007069699686239355, + "loss": 0.5151, + "step": 10575 + }, + { + "epoch": 0.5908875045394866, + "grad_norm": 0.5993004441261292, + "learning_rate": 0.0007069419542805917, + "loss": 0.4299, + "step": 10576 + }, + { + "epoch": 0.5909433751431684, + "grad_norm": 0.4434584975242615, + "learning_rate": 0.0007069139399372478, + "loss": 0.4059, + "step": 10577 + }, + { + "epoch": 0.5909992457468503, + "grad_norm": 0.5570957660675049, + "learning_rate": 0.0007068859255939041, + "loss": 0.4953, + "step": 10578 + }, + { + "epoch": 0.5910551163505322, + "grad_norm": 1.338035225868225, + "learning_rate": 0.0007068579112505603, + "loss": 0.3481, + "step": 10579 + }, + { + "epoch": 0.591110986954214, + "grad_norm": 0.6354560852050781, + "learning_rate": 0.0007068298969072165, + "loss": 0.5526, + "step": 10580 + }, + { + "epoch": 0.5911668575578959, + "grad_norm": 0.507318913936615, + "learning_rate": 0.0007068018825638726, + "loss": 0.505, + "step": 10581 + }, + { + "epoch": 0.5912227281615778, + "grad_norm": 0.6350831389427185, + "learning_rate": 0.000706773868220529, + "loss": 0.4113, + "step": 10582 + }, + { + "epoch": 0.5912785987652597, + "grad_norm": 0.44030919671058655, + "learning_rate": 0.0007067458538771852, + "loss": 0.3735, + "step": 10583 + }, + { + "epoch": 0.5913344693689415, + "grad_norm": 1.8448177576065063, + "learning_rate": 0.0007067178395338414, + "loss": 0.658, + "step": 10584 + }, + { + "epoch": 0.5913903399726234, + "grad_norm": 0.6296502947807312, + "learning_rate": 0.0007066898251904976, + "loss": 0.3615, + "step": 10585 + }, + { + "epoch": 0.5914462105763053, + "grad_norm": 0.35576528310775757, + "learning_rate": 0.0007066618108471538, + "loss": 0.3422, + "step": 10586 + }, + { + "epoch": 0.5915020811799871, + "grad_norm": 0.7760498523712158, + "learning_rate": 0.00070663379650381, + "loss": 0.5832, + "step": 10587 + }, + { + "epoch": 0.591557951783669, + "grad_norm": 1.066068172454834, + "learning_rate": 0.0007066057821604662, + "loss": 0.3251, + "step": 10588 + }, + { + "epoch": 0.5916138223873509, + "grad_norm": 0.5221807360649109, + "learning_rate": 0.0007065777678171224, + "loss": 0.4293, + "step": 10589 + }, + { + "epoch": 0.5916696929910328, + "grad_norm": 0.5692546963691711, + "learning_rate": 0.0007065497534737786, + "loss": 0.5256, + "step": 10590 + }, + { + "epoch": 0.5917255635947146, + "grad_norm": 0.5060479640960693, + "learning_rate": 0.0007065217391304348, + "loss": 0.482, + "step": 10591 + }, + { + "epoch": 0.5917814341983965, + "grad_norm": 0.6106531620025635, + "learning_rate": 0.0007064937247870909, + "loss": 0.5149, + "step": 10592 + }, + { + "epoch": 0.5918373048020784, + "grad_norm": 2.383341073989868, + "learning_rate": 0.0007064657104437472, + "loss": 0.4872, + "step": 10593 + }, + { + "epoch": 0.5918931754057603, + "grad_norm": 0.8947480320930481, + "learning_rate": 0.0007064376961004034, + "loss": 0.4412, + "step": 10594 + }, + { + "epoch": 0.5919490460094421, + "grad_norm": 0.8103718161582947, + "learning_rate": 0.0007064096817570596, + "loss": 0.3765, + "step": 10595 + }, + { + "epoch": 0.592004916613124, + "grad_norm": 0.39758867025375366, + "learning_rate": 0.0007063816674137158, + "loss": 0.5033, + "step": 10596 + }, + { + "epoch": 0.5920607872168059, + "grad_norm": 0.5993728637695312, + "learning_rate": 0.000706353653070372, + "loss": 0.4926, + "step": 10597 + }, + { + "epoch": 0.5921166578204877, + "grad_norm": 0.6696208119392395, + "learning_rate": 0.0007063256387270282, + "loss": 0.502, + "step": 10598 + }, + { + "epoch": 0.5921725284241697, + "grad_norm": 0.5079922676086426, + "learning_rate": 0.0007062976243836844, + "loss": 0.3972, + "step": 10599 + }, + { + "epoch": 0.5922283990278515, + "grad_norm": 0.40260428190231323, + "learning_rate": 0.0007062696100403407, + "loss": 0.3926, + "step": 10600 + }, + { + "epoch": 0.5922842696315334, + "grad_norm": 0.9660550355911255, + "learning_rate": 0.0007062415956969969, + "loss": 0.6712, + "step": 10601 + }, + { + "epoch": 0.5923401402352152, + "grad_norm": 2.0948963165283203, + "learning_rate": 0.0007062135813536531, + "loss": 0.5109, + "step": 10602 + }, + { + "epoch": 0.5923960108388971, + "grad_norm": 0.4967329800128937, + "learning_rate": 0.0007061855670103094, + "loss": 0.4903, + "step": 10603 + }, + { + "epoch": 0.592451881442579, + "grad_norm": 1.5557384490966797, + "learning_rate": 0.0007061575526669655, + "loss": 0.6135, + "step": 10604 + }, + { + "epoch": 0.5925077520462608, + "grad_norm": 0.5643572807312012, + "learning_rate": 0.0007061295383236217, + "loss": 0.4887, + "step": 10605 + }, + { + "epoch": 0.5925636226499428, + "grad_norm": 2.764357566833496, + "learning_rate": 0.0007061015239802779, + "loss": 0.4158, + "step": 10606 + }, + { + "epoch": 0.5926194932536246, + "grad_norm": 0.40161803364753723, + "learning_rate": 0.0007060735096369342, + "loss": 0.3735, + "step": 10607 + }, + { + "epoch": 0.5926753638573065, + "grad_norm": 0.4654372036457062, + "learning_rate": 0.0007060454952935903, + "loss": 0.375, + "step": 10608 + }, + { + "epoch": 0.5927312344609883, + "grad_norm": 0.4442000985145569, + "learning_rate": 0.0007060174809502465, + "loss": 0.3759, + "step": 10609 + }, + { + "epoch": 0.5927871050646702, + "grad_norm": 1.0673935413360596, + "learning_rate": 0.0007059894666069028, + "loss": 0.5633, + "step": 10610 + }, + { + "epoch": 0.5928429756683521, + "grad_norm": 0.410057932138443, + "learning_rate": 0.000705961452263559, + "loss": 0.4353, + "step": 10611 + }, + { + "epoch": 0.5928988462720339, + "grad_norm": 0.4069942533969879, + "learning_rate": 0.0007059334379202151, + "loss": 0.4853, + "step": 10612 + }, + { + "epoch": 0.5929547168757159, + "grad_norm": 0.44976329803466797, + "learning_rate": 0.0007059054235768713, + "loss": 0.3339, + "step": 10613 + }, + { + "epoch": 0.5930105874793977, + "grad_norm": 1.0745095014572144, + "learning_rate": 0.0007058774092335276, + "loss": 0.4169, + "step": 10614 + }, + { + "epoch": 0.5930664580830796, + "grad_norm": 0.7884443998336792, + "learning_rate": 0.0007058493948901837, + "loss": 0.4218, + "step": 10615 + }, + { + "epoch": 0.5931223286867615, + "grad_norm": 1.664429783821106, + "learning_rate": 0.00070582138054684, + "loss": 0.4963, + "step": 10616 + }, + { + "epoch": 0.5931781992904434, + "grad_norm": 0.5740304589271545, + "learning_rate": 0.0007057933662034963, + "loss": 0.509, + "step": 10617 + }, + { + "epoch": 0.5932340698941252, + "grad_norm": 0.5977592468261719, + "learning_rate": 0.0007057653518601525, + "loss": 0.3456, + "step": 10618 + }, + { + "epoch": 0.593289940497807, + "grad_norm": 0.6304298043251038, + "learning_rate": 0.0007057373375168086, + "loss": 0.4369, + "step": 10619 + }, + { + "epoch": 0.593345811101489, + "grad_norm": 0.5095614194869995, + "learning_rate": 0.0007057093231734648, + "loss": 0.417, + "step": 10620 + }, + { + "epoch": 0.5934016817051708, + "grad_norm": 0.9613871574401855, + "learning_rate": 0.0007056813088301211, + "loss": 0.6028, + "step": 10621 + }, + { + "epoch": 0.5934575523088527, + "grad_norm": 0.502708375453949, + "learning_rate": 0.0007056532944867773, + "loss": 0.3985, + "step": 10622 + }, + { + "epoch": 0.5935134229125346, + "grad_norm": 0.5241557359695435, + "learning_rate": 0.0007056252801434334, + "loss": 0.4736, + "step": 10623 + }, + { + "epoch": 0.5935692935162165, + "grad_norm": 0.4307381212711334, + "learning_rate": 0.0007055972658000897, + "loss": 0.4645, + "step": 10624 + }, + { + "epoch": 0.5936251641198983, + "grad_norm": 0.6144028306007385, + "learning_rate": 0.0007055692514567459, + "loss": 0.4939, + "step": 10625 + }, + { + "epoch": 0.5936810347235801, + "grad_norm": 0.39793047308921814, + "learning_rate": 0.0007055412371134021, + "loss": 0.3522, + "step": 10626 + }, + { + "epoch": 0.5937369053272621, + "grad_norm": 1.1511204242706299, + "learning_rate": 0.0007055132227700582, + "loss": 0.3487, + "step": 10627 + }, + { + "epoch": 0.5937927759309439, + "grad_norm": 0.5012922286987305, + "learning_rate": 0.0007054852084267145, + "loss": 0.4253, + "step": 10628 + }, + { + "epoch": 0.5938486465346258, + "grad_norm": 0.4291864037513733, + "learning_rate": 0.0007054571940833707, + "loss": 0.463, + "step": 10629 + }, + { + "epoch": 0.5939045171383077, + "grad_norm": 0.9241259098052979, + "learning_rate": 0.0007054291797400269, + "loss": 0.6492, + "step": 10630 + }, + { + "epoch": 0.5939603877419896, + "grad_norm": 0.666739284992218, + "learning_rate": 0.000705401165396683, + "loss": 0.5375, + "step": 10631 + }, + { + "epoch": 0.5940162583456714, + "grad_norm": 0.827488899230957, + "learning_rate": 0.0007053731510533393, + "loss": 0.5251, + "step": 10632 + }, + { + "epoch": 0.5940721289493534, + "grad_norm": 0.6772411465644836, + "learning_rate": 0.0007053451367099956, + "loss": 0.4579, + "step": 10633 + }, + { + "epoch": 0.5941279995530352, + "grad_norm": 1.4262694120407104, + "learning_rate": 0.0007053171223666517, + "loss": 0.577, + "step": 10634 + }, + { + "epoch": 0.594183870156717, + "grad_norm": 2.61434268951416, + "learning_rate": 0.000705289108023308, + "loss": 0.5072, + "step": 10635 + }, + { + "epoch": 0.5942397407603989, + "grad_norm": 3.682220458984375, + "learning_rate": 0.0007052610936799642, + "loss": 0.4048, + "step": 10636 + }, + { + "epoch": 0.5942956113640808, + "grad_norm": 1.1742042303085327, + "learning_rate": 0.0007052330793366204, + "loss": 0.4847, + "step": 10637 + }, + { + "epoch": 0.5943514819677627, + "grad_norm": 0.49727070331573486, + "learning_rate": 0.0007052050649932765, + "loss": 0.4736, + "step": 10638 + }, + { + "epoch": 0.5944073525714445, + "grad_norm": 0.5396834015846252, + "learning_rate": 0.0007051770506499328, + "loss": 0.5338, + "step": 10639 + }, + { + "epoch": 0.5944632231751265, + "grad_norm": 0.5238367319107056, + "learning_rate": 0.000705149036306589, + "loss": 0.3201, + "step": 10640 + }, + { + "epoch": 0.5945190937788083, + "grad_norm": 0.9191264510154724, + "learning_rate": 0.0007051210219632452, + "loss": 0.452, + "step": 10641 + }, + { + "epoch": 0.5945749643824901, + "grad_norm": 0.880526065826416, + "learning_rate": 0.0007050930076199014, + "loss": 0.4715, + "step": 10642 + }, + { + "epoch": 0.594630834986172, + "grad_norm": 0.5678358674049377, + "learning_rate": 0.0007050649932765576, + "loss": 0.4456, + "step": 10643 + }, + { + "epoch": 0.5946867055898539, + "grad_norm": 0.3863213062286377, + "learning_rate": 0.0007050369789332138, + "loss": 0.3908, + "step": 10644 + }, + { + "epoch": 0.5947425761935358, + "grad_norm": 1.152246356010437, + "learning_rate": 0.00070500896458987, + "loss": 0.4143, + "step": 10645 + }, + { + "epoch": 0.5947984467972176, + "grad_norm": 0.6201992034912109, + "learning_rate": 0.0007049809502465262, + "loss": 0.3455, + "step": 10646 + }, + { + "epoch": 0.5948543174008996, + "grad_norm": 0.6000551581382751, + "learning_rate": 0.0007049529359031824, + "loss": 0.477, + "step": 10647 + }, + { + "epoch": 0.5949101880045814, + "grad_norm": 0.46386829018592834, + "learning_rate": 0.0007049249215598386, + "loss": 0.3781, + "step": 10648 + }, + { + "epoch": 0.5949660586082632, + "grad_norm": 0.4015580713748932, + "learning_rate": 0.000704896907216495, + "loss": 0.4196, + "step": 10649 + }, + { + "epoch": 0.5950219292119452, + "grad_norm": 1.0267517566680908, + "learning_rate": 0.000704868892873151, + "loss": 0.396, + "step": 10650 + }, + { + "epoch": 0.595077799815627, + "grad_norm": 0.4482877254486084, + "learning_rate": 0.0007048408785298073, + "loss": 0.4066, + "step": 10651 + }, + { + "epoch": 0.5951336704193089, + "grad_norm": 0.5090351104736328, + "learning_rate": 0.0007048128641864635, + "loss": 0.5097, + "step": 10652 + }, + { + "epoch": 0.5951895410229907, + "grad_norm": 0.8994319438934326, + "learning_rate": 0.0007047848498431198, + "loss": 0.4807, + "step": 10653 + }, + { + "epoch": 0.5952454116266727, + "grad_norm": 1.1345936059951782, + "learning_rate": 0.0007047568354997759, + "loss": 0.4147, + "step": 10654 + }, + { + "epoch": 0.5953012822303545, + "grad_norm": 0.5302750468254089, + "learning_rate": 0.0007047288211564321, + "loss": 0.4204, + "step": 10655 + }, + { + "epoch": 0.5953571528340363, + "grad_norm": 0.7031237483024597, + "learning_rate": 0.0007047008068130884, + "loss": 0.3767, + "step": 10656 + }, + { + "epoch": 0.5954130234377183, + "grad_norm": 1.5943266153335571, + "learning_rate": 0.0007046727924697445, + "loss": 0.427, + "step": 10657 + }, + { + "epoch": 0.5954688940414001, + "grad_norm": 0.7673460841178894, + "learning_rate": 0.0007046447781264007, + "loss": 0.547, + "step": 10658 + }, + { + "epoch": 0.595524764645082, + "grad_norm": 0.47304996848106384, + "learning_rate": 0.0007046167637830569, + "loss": 0.4571, + "step": 10659 + }, + { + "epoch": 0.5955806352487638, + "grad_norm": 1.64436674118042, + "learning_rate": 0.0007045887494397132, + "loss": 0.3833, + "step": 10660 + }, + { + "epoch": 0.5956365058524458, + "grad_norm": 0.40201833844184875, + "learning_rate": 0.0007045607350963693, + "loss": 0.4598, + "step": 10661 + }, + { + "epoch": 0.5956923764561276, + "grad_norm": 0.42890164256095886, + "learning_rate": 0.0007045327207530255, + "loss": 0.4803, + "step": 10662 + }, + { + "epoch": 0.5957482470598094, + "grad_norm": 2.023137331008911, + "learning_rate": 0.0007045047064096818, + "loss": 0.5255, + "step": 10663 + }, + { + "epoch": 0.5958041176634914, + "grad_norm": 0.44804203510284424, + "learning_rate": 0.000704476692066338, + "loss": 0.4957, + "step": 10664 + }, + { + "epoch": 0.5958599882671732, + "grad_norm": 0.7331030368804932, + "learning_rate": 0.0007044486777229941, + "loss": 0.5467, + "step": 10665 + }, + { + "epoch": 0.5959158588708551, + "grad_norm": 0.9253422617912292, + "learning_rate": 0.0007044206633796503, + "loss": 0.4851, + "step": 10666 + }, + { + "epoch": 0.595971729474537, + "grad_norm": 0.4835907816886902, + "learning_rate": 0.0007043926490363067, + "loss": 0.3394, + "step": 10667 + }, + { + "epoch": 0.5960276000782189, + "grad_norm": 0.4381670355796814, + "learning_rate": 0.0007043646346929629, + "loss": 0.4229, + "step": 10668 + }, + { + "epoch": 0.5960834706819007, + "grad_norm": 6.002093315124512, + "learning_rate": 0.000704336620349619, + "loss": 0.348, + "step": 10669 + }, + { + "epoch": 0.5961393412855825, + "grad_norm": 0.9109649062156677, + "learning_rate": 0.0007043086060062752, + "loss": 0.3963, + "step": 10670 + }, + { + "epoch": 0.5961952118892645, + "grad_norm": 0.6678275465965271, + "learning_rate": 0.0007042805916629315, + "loss": 0.4023, + "step": 10671 + }, + { + "epoch": 0.5962510824929463, + "grad_norm": 0.9679526090621948, + "learning_rate": 0.0007042525773195877, + "loss": 0.4758, + "step": 10672 + }, + { + "epoch": 0.5963069530966282, + "grad_norm": 1.0836491584777832, + "learning_rate": 0.0007042245629762438, + "loss": 0.6888, + "step": 10673 + }, + { + "epoch": 0.5963628237003101, + "grad_norm": 0.3845500946044922, + "learning_rate": 0.0007041965486329001, + "loss": 0.3709, + "step": 10674 + }, + { + "epoch": 0.596418694303992, + "grad_norm": 0.40340712666511536, + "learning_rate": 0.0007041685342895563, + "loss": 0.3997, + "step": 10675 + }, + { + "epoch": 0.5964745649076738, + "grad_norm": 0.5983710289001465, + "learning_rate": 0.0007041405199462124, + "loss": 0.5948, + "step": 10676 + }, + { + "epoch": 0.5965304355113557, + "grad_norm": 2.056506633758545, + "learning_rate": 0.0007041125056028686, + "loss": 0.3745, + "step": 10677 + }, + { + "epoch": 0.5965863061150376, + "grad_norm": 0.6320216059684753, + "learning_rate": 0.0007040844912595249, + "loss": 0.4356, + "step": 10678 + }, + { + "epoch": 0.5966421767187194, + "grad_norm": 0.7727453112602234, + "learning_rate": 0.0007040564769161811, + "loss": 0.4164, + "step": 10679 + }, + { + "epoch": 0.5966980473224013, + "grad_norm": 0.6336807012557983, + "learning_rate": 0.0007040284625728372, + "loss": 0.5325, + "step": 10680 + }, + { + "epoch": 0.5967539179260832, + "grad_norm": 2.953831195831299, + "learning_rate": 0.0007040004482294935, + "loss": 0.4469, + "step": 10681 + }, + { + "epoch": 0.5968097885297651, + "grad_norm": 2.717120409011841, + "learning_rate": 0.0007039724338861497, + "loss": 0.4263, + "step": 10682 + }, + { + "epoch": 0.5968656591334469, + "grad_norm": 0.7408727407455444, + "learning_rate": 0.000703944419542806, + "loss": 0.4564, + "step": 10683 + }, + { + "epoch": 0.5969215297371289, + "grad_norm": 0.49386051297187805, + "learning_rate": 0.000703916405199462, + "loss": 0.4608, + "step": 10684 + }, + { + "epoch": 0.5969774003408107, + "grad_norm": 0.9177375435829163, + "learning_rate": 0.0007038883908561184, + "loss": 0.4967, + "step": 10685 + }, + { + "epoch": 0.5970332709444925, + "grad_norm": 0.46739310026168823, + "learning_rate": 0.0007038603765127746, + "loss": 0.455, + "step": 10686 + }, + { + "epoch": 0.5970891415481744, + "grad_norm": 0.49177825450897217, + "learning_rate": 0.0007038323621694308, + "loss": 0.4426, + "step": 10687 + }, + { + "epoch": 0.5971450121518563, + "grad_norm": 0.3878510296344757, + "learning_rate": 0.000703804347826087, + "loss": 0.4344, + "step": 10688 + }, + { + "epoch": 0.5972008827555382, + "grad_norm": 0.3814642131328583, + "learning_rate": 0.0007037763334827432, + "loss": 0.4247, + "step": 10689 + }, + { + "epoch": 0.59725675335922, + "grad_norm": 0.8399043083190918, + "learning_rate": 0.0007037483191393994, + "loss": 0.5441, + "step": 10690 + }, + { + "epoch": 0.597312623962902, + "grad_norm": 0.9777029752731323, + "learning_rate": 0.0007037203047960556, + "loss": 0.458, + "step": 10691 + }, + { + "epoch": 0.5973684945665838, + "grad_norm": 0.48881974816322327, + "learning_rate": 0.0007036922904527118, + "loss": 0.44, + "step": 10692 + }, + { + "epoch": 0.5974243651702656, + "grad_norm": 1.0806488990783691, + "learning_rate": 0.000703664276109368, + "loss": 0.4718, + "step": 10693 + }, + { + "epoch": 0.5974802357739475, + "grad_norm": 0.5673818588256836, + "learning_rate": 0.0007036362617660242, + "loss": 0.4824, + "step": 10694 + }, + { + "epoch": 0.5975361063776294, + "grad_norm": 0.8995773196220398, + "learning_rate": 0.0007036082474226805, + "loss": 0.493, + "step": 10695 + }, + { + "epoch": 0.5975919769813113, + "grad_norm": 0.4830895662307739, + "learning_rate": 0.0007035802330793366, + "loss": 0.639, + "step": 10696 + }, + { + "epoch": 0.5976478475849931, + "grad_norm": 0.6560522317886353, + "learning_rate": 0.0007035522187359928, + "loss": 0.4698, + "step": 10697 + }, + { + "epoch": 0.5977037181886751, + "grad_norm": 0.6611362099647522, + "learning_rate": 0.000703524204392649, + "loss": 0.3472, + "step": 10698 + }, + { + "epoch": 0.5977595887923569, + "grad_norm": 0.6183484196662903, + "learning_rate": 0.0007034961900493052, + "loss": 0.4345, + "step": 10699 + }, + { + "epoch": 0.5978154593960388, + "grad_norm": 0.5718401074409485, + "learning_rate": 0.0007034681757059615, + "loss": 0.3001, + "step": 10700 + }, + { + "epoch": 0.5978713299997207, + "grad_norm": 0.46827420592308044, + "learning_rate": 0.0007034401613626177, + "loss": 0.4594, + "step": 10701 + }, + { + "epoch": 0.5979272006034025, + "grad_norm": 0.6339165568351746, + "learning_rate": 0.0007034121470192739, + "loss": 0.4463, + "step": 10702 + }, + { + "epoch": 0.5979830712070844, + "grad_norm": 1.3793516159057617, + "learning_rate": 0.0007033841326759301, + "loss": 0.6961, + "step": 10703 + }, + { + "epoch": 0.5980389418107662, + "grad_norm": 1.0683624744415283, + "learning_rate": 0.0007033561183325863, + "loss": 0.5372, + "step": 10704 + }, + { + "epoch": 0.5980948124144482, + "grad_norm": 0.36534950137138367, + "learning_rate": 0.0007033281039892425, + "loss": 0.3682, + "step": 10705 + }, + { + "epoch": 0.59815068301813, + "grad_norm": 0.5291146039962769, + "learning_rate": 0.0007033000896458988, + "loss": 0.4808, + "step": 10706 + }, + { + "epoch": 0.5982065536218119, + "grad_norm": 0.7014515995979309, + "learning_rate": 0.0007032720753025549, + "loss": 0.471, + "step": 10707 + }, + { + "epoch": 0.5982624242254938, + "grad_norm": 0.6202278137207031, + "learning_rate": 0.0007032440609592111, + "loss": 0.6781, + "step": 10708 + }, + { + "epoch": 0.5983182948291756, + "grad_norm": 0.521587073802948, + "learning_rate": 0.0007032160466158673, + "loss": 0.4511, + "step": 10709 + }, + { + "epoch": 0.5983741654328575, + "grad_norm": 0.495892733335495, + "learning_rate": 0.0007031880322725236, + "loss": 0.3914, + "step": 10710 + }, + { + "epoch": 0.5984300360365393, + "grad_norm": 0.703321635723114, + "learning_rate": 0.0007031600179291797, + "loss": 0.4189, + "step": 10711 + }, + { + "epoch": 0.5984859066402213, + "grad_norm": 0.3838154375553131, + "learning_rate": 0.0007031320035858359, + "loss": 0.3503, + "step": 10712 + }, + { + "epoch": 0.5985417772439031, + "grad_norm": 0.6379182934761047, + "learning_rate": 0.0007031039892424922, + "loss": 0.4535, + "step": 10713 + }, + { + "epoch": 0.598597647847585, + "grad_norm": 0.5247781276702881, + "learning_rate": 0.0007030759748991484, + "loss": 0.5109, + "step": 10714 + }, + { + "epoch": 0.5986535184512669, + "grad_norm": 2.453977108001709, + "learning_rate": 0.0007030479605558045, + "loss": 0.5652, + "step": 10715 + }, + { + "epoch": 0.5987093890549487, + "grad_norm": 0.4516439437866211, + "learning_rate": 0.0007030199462124607, + "loss": 0.4457, + "step": 10716 + }, + { + "epoch": 0.5987652596586306, + "grad_norm": 1.7648907899856567, + "learning_rate": 0.0007029919318691171, + "loss": 0.4301, + "step": 10717 + }, + { + "epoch": 0.5988211302623124, + "grad_norm": 0.6366819739341736, + "learning_rate": 0.0007029639175257732, + "loss": 0.4753, + "step": 10718 + }, + { + "epoch": 0.5988770008659944, + "grad_norm": 6.045609951019287, + "learning_rate": 0.0007029359031824294, + "loss": 0.4873, + "step": 10719 + }, + { + "epoch": 0.5989328714696762, + "grad_norm": 1.6230803728103638, + "learning_rate": 0.0007029078888390857, + "loss": 0.4639, + "step": 10720 + }, + { + "epoch": 0.5989887420733581, + "grad_norm": 0.5354899168014526, + "learning_rate": 0.0007028798744957419, + "loss": 0.3523, + "step": 10721 + }, + { + "epoch": 0.59904461267704, + "grad_norm": 0.6042068600654602, + "learning_rate": 0.000702851860152398, + "loss": 0.4867, + "step": 10722 + }, + { + "epoch": 0.5991004832807219, + "grad_norm": 1.1678177118301392, + "learning_rate": 0.0007028238458090542, + "loss": 0.3554, + "step": 10723 + }, + { + "epoch": 0.5991563538844037, + "grad_norm": 0.7739242911338806, + "learning_rate": 0.0007027958314657105, + "loss": 0.551, + "step": 10724 + }, + { + "epoch": 0.5992122244880856, + "grad_norm": 0.4626062512397766, + "learning_rate": 0.0007027678171223667, + "loss": 0.422, + "step": 10725 + }, + { + "epoch": 0.5992680950917675, + "grad_norm": 0.5082313418388367, + "learning_rate": 0.0007027398027790228, + "loss": 0.4751, + "step": 10726 + }, + { + "epoch": 0.5993239656954493, + "grad_norm": 0.7273955941200256, + "learning_rate": 0.0007027117884356791, + "loss": 0.5116, + "step": 10727 + }, + { + "epoch": 0.5993798362991312, + "grad_norm": 2.1017580032348633, + "learning_rate": 0.0007026837740923353, + "loss": 0.9009, + "step": 10728 + }, + { + "epoch": 0.5994357069028131, + "grad_norm": 0.8967925310134888, + "learning_rate": 0.0007026557597489915, + "loss": 0.483, + "step": 10729 + }, + { + "epoch": 0.599491577506495, + "grad_norm": 0.6895310282707214, + "learning_rate": 0.0007026277454056476, + "loss": 0.4245, + "step": 10730 + }, + { + "epoch": 0.5995474481101768, + "grad_norm": 0.41783490777015686, + "learning_rate": 0.0007025997310623039, + "loss": 0.4371, + "step": 10731 + }, + { + "epoch": 0.5996033187138587, + "grad_norm": 0.6271681785583496, + "learning_rate": 0.0007025717167189601, + "loss": 0.4542, + "step": 10732 + }, + { + "epoch": 0.5996591893175406, + "grad_norm": 2.8926126956939697, + "learning_rate": 0.0007025437023756164, + "loss": 0.4653, + "step": 10733 + }, + { + "epoch": 0.5997150599212224, + "grad_norm": 0.4959912598133087, + "learning_rate": 0.0007025156880322725, + "loss": 0.4676, + "step": 10734 + }, + { + "epoch": 0.5997709305249043, + "grad_norm": 0.5203404426574707, + "learning_rate": 0.0007024876736889288, + "loss": 0.5347, + "step": 10735 + }, + { + "epoch": 0.5998268011285862, + "grad_norm": 0.980491042137146, + "learning_rate": 0.000702459659345585, + "loss": 0.6162, + "step": 10736 + }, + { + "epoch": 0.5998826717322681, + "grad_norm": 1.0293254852294922, + "learning_rate": 0.0007024316450022412, + "loss": 0.4867, + "step": 10737 + }, + { + "epoch": 0.5999385423359499, + "grad_norm": 2.913088798522949, + "learning_rate": 0.0007024036306588974, + "loss": 0.5232, + "step": 10738 + }, + { + "epoch": 0.5999944129396318, + "grad_norm": 0.440929114818573, + "learning_rate": 0.0007023756163155536, + "loss": 0.4461, + "step": 10739 + }, + { + "epoch": 0.6000502835433137, + "grad_norm": 0.5049867630004883, + "learning_rate": 0.0007023476019722098, + "loss": 0.5367, + "step": 10740 + }, + { + "epoch": 0.6001061541469955, + "grad_norm": 0.9609411358833313, + "learning_rate": 0.0007023195876288659, + "loss": 0.5233, + "step": 10741 + }, + { + "epoch": 0.6001620247506775, + "grad_norm": 0.5591278672218323, + "learning_rate": 0.0007022915732855222, + "loss": 0.441, + "step": 10742 + }, + { + "epoch": 0.6002178953543593, + "grad_norm": 4.963044166564941, + "learning_rate": 0.0007022635589421784, + "loss": 0.4034, + "step": 10743 + }, + { + "epoch": 0.6002737659580412, + "grad_norm": 0.5401554703712463, + "learning_rate": 0.0007022355445988346, + "loss": 0.6562, + "step": 10744 + }, + { + "epoch": 0.600329636561723, + "grad_norm": 0.6769577264785767, + "learning_rate": 0.0007022075302554908, + "loss": 0.4582, + "step": 10745 + }, + { + "epoch": 0.600385507165405, + "grad_norm": 0.6239258050918579, + "learning_rate": 0.000702179515912147, + "loss": 0.5938, + "step": 10746 + }, + { + "epoch": 0.6004413777690868, + "grad_norm": 1.3697011470794678, + "learning_rate": 0.0007021515015688032, + "loss": 0.7781, + "step": 10747 + }, + { + "epoch": 0.6004972483727686, + "grad_norm": 0.5094802975654602, + "learning_rate": 0.0007021234872254594, + "loss": 0.3357, + "step": 10748 + }, + { + "epoch": 0.6005531189764506, + "grad_norm": 2.0784542560577393, + "learning_rate": 0.0007020954728821156, + "loss": 0.461, + "step": 10749 + }, + { + "epoch": 0.6006089895801324, + "grad_norm": 0.523945689201355, + "learning_rate": 0.0007020674585387719, + "loss": 0.4659, + "step": 10750 + }, + { + "epoch": 0.6006648601838143, + "grad_norm": 0.4673360586166382, + "learning_rate": 0.0007020394441954281, + "loss": 0.5138, + "step": 10751 + }, + { + "epoch": 0.6007207307874961, + "grad_norm": 0.6866123676300049, + "learning_rate": 0.0007020114298520844, + "loss": 0.4562, + "step": 10752 + }, + { + "epoch": 0.6007766013911781, + "grad_norm": 0.7821410894393921, + "learning_rate": 0.0007019834155087405, + "loss": 0.4928, + "step": 10753 + }, + { + "epoch": 0.6008324719948599, + "grad_norm": 0.3024638593196869, + "learning_rate": 0.0007019554011653967, + "loss": 0.3705, + "step": 10754 + }, + { + "epoch": 0.6008883425985417, + "grad_norm": 0.4206867516040802, + "learning_rate": 0.0007019273868220529, + "loss": 0.437, + "step": 10755 + }, + { + "epoch": 0.6009442132022237, + "grad_norm": 0.7161227464675903, + "learning_rate": 0.0007018993724787092, + "loss": 0.4711, + "step": 10756 + }, + { + "epoch": 0.6010000838059055, + "grad_norm": 0.49148017168045044, + "learning_rate": 0.0007018713581353653, + "loss": 0.5198, + "step": 10757 + }, + { + "epoch": 0.6010559544095874, + "grad_norm": 0.6148686408996582, + "learning_rate": 0.0007018433437920215, + "loss": 0.4805, + "step": 10758 + }, + { + "epoch": 0.6011118250132693, + "grad_norm": 0.4678902328014374, + "learning_rate": 0.0007018153294486778, + "loss": 0.5052, + "step": 10759 + }, + { + "epoch": 0.6011676956169512, + "grad_norm": 0.5336238145828247, + "learning_rate": 0.000701787315105334, + "loss": 0.4237, + "step": 10760 + }, + { + "epoch": 0.601223566220633, + "grad_norm": 0.5325070023536682, + "learning_rate": 0.0007017593007619901, + "loss": 0.376, + "step": 10761 + }, + { + "epoch": 0.6012794368243148, + "grad_norm": 0.5435318946838379, + "learning_rate": 0.0007017312864186463, + "loss": 0.5348, + "step": 10762 + }, + { + "epoch": 0.6013353074279968, + "grad_norm": 0.46791818737983704, + "learning_rate": 0.0007017032720753026, + "loss": 0.4236, + "step": 10763 + }, + { + "epoch": 0.6013911780316786, + "grad_norm": 0.48111358284950256, + "learning_rate": 0.0007016752577319587, + "loss": 0.4976, + "step": 10764 + }, + { + "epoch": 0.6014470486353605, + "grad_norm": 0.5517467260360718, + "learning_rate": 0.0007016472433886149, + "loss": 0.4432, + "step": 10765 + }, + { + "epoch": 0.6015029192390424, + "grad_norm": 0.7193959355354309, + "learning_rate": 0.0007016192290452713, + "loss": 0.4327, + "step": 10766 + }, + { + "epoch": 0.6015587898427243, + "grad_norm": 0.4420635402202606, + "learning_rate": 0.0007015912147019275, + "loss": 0.446, + "step": 10767 + }, + { + "epoch": 0.6016146604464061, + "grad_norm": 0.8064660429954529, + "learning_rate": 0.0007015632003585836, + "loss": 0.5417, + "step": 10768 + }, + { + "epoch": 0.6016705310500879, + "grad_norm": 0.8224947452545166, + "learning_rate": 0.0007015351860152398, + "loss": 0.3942, + "step": 10769 + }, + { + "epoch": 0.6017264016537699, + "grad_norm": 0.42773669958114624, + "learning_rate": 0.0007015071716718961, + "loss": 0.4366, + "step": 10770 + }, + { + "epoch": 0.6017822722574517, + "grad_norm": 0.581246554851532, + "learning_rate": 0.0007014791573285523, + "loss": 0.3714, + "step": 10771 + }, + { + "epoch": 0.6018381428611336, + "grad_norm": 0.7085760235786438, + "learning_rate": 0.0007014511429852084, + "loss": 0.4155, + "step": 10772 + }, + { + "epoch": 0.6018940134648155, + "grad_norm": 0.491563081741333, + "learning_rate": 0.0007014231286418646, + "loss": 0.4522, + "step": 10773 + }, + { + "epoch": 0.6019498840684974, + "grad_norm": 0.40800485014915466, + "learning_rate": 0.0007013951142985209, + "loss": 0.3759, + "step": 10774 + }, + { + "epoch": 0.6020057546721792, + "grad_norm": 0.4582277238368988, + "learning_rate": 0.0007013670999551771, + "loss": 0.387, + "step": 10775 + }, + { + "epoch": 0.6020616252758612, + "grad_norm": 0.6016603112220764, + "learning_rate": 0.0007013390856118332, + "loss": 0.5081, + "step": 10776 + }, + { + "epoch": 0.602117495879543, + "grad_norm": 0.3883710503578186, + "learning_rate": 0.0007013110712684895, + "loss": 0.4726, + "step": 10777 + }, + { + "epoch": 0.6021733664832248, + "grad_norm": 0.4064609706401825, + "learning_rate": 0.0007012830569251457, + "loss": 0.4401, + "step": 10778 + }, + { + "epoch": 0.6022292370869067, + "grad_norm": 0.4226343333721161, + "learning_rate": 0.0007012550425818019, + "loss": 0.458, + "step": 10779 + }, + { + "epoch": 0.6022851076905886, + "grad_norm": 1.7223536968231201, + "learning_rate": 0.000701227028238458, + "loss": 0.4512, + "step": 10780 + }, + { + "epoch": 0.6023409782942705, + "grad_norm": 0.6902680993080139, + "learning_rate": 0.0007011990138951143, + "loss": 0.4834, + "step": 10781 + }, + { + "epoch": 0.6023968488979523, + "grad_norm": 0.46517351269721985, + "learning_rate": 0.0007011709995517705, + "loss": 0.4419, + "step": 10782 + }, + { + "epoch": 0.6024527195016343, + "grad_norm": 0.6185488700866699, + "learning_rate": 0.0007011429852084266, + "loss": 0.4346, + "step": 10783 + }, + { + "epoch": 0.6025085901053161, + "grad_norm": 0.8254278898239136, + "learning_rate": 0.000701114970865083, + "loss": 0.3464, + "step": 10784 + }, + { + "epoch": 0.6025644607089979, + "grad_norm": 0.5473043918609619, + "learning_rate": 0.0007010869565217392, + "loss": 0.5468, + "step": 10785 + }, + { + "epoch": 0.6026203313126798, + "grad_norm": 0.6181616187095642, + "learning_rate": 0.0007010589421783954, + "loss": 0.4319, + "step": 10786 + }, + { + "epoch": 0.6026762019163617, + "grad_norm": 0.5484591722488403, + "learning_rate": 0.0007010309278350515, + "loss": 0.442, + "step": 10787 + }, + { + "epoch": 0.6027320725200436, + "grad_norm": 0.499711811542511, + "learning_rate": 0.0007010029134917078, + "loss": 0.4783, + "step": 10788 + }, + { + "epoch": 0.6027879431237254, + "grad_norm": 0.6180328130722046, + "learning_rate": 0.000700974899148364, + "loss": 0.4706, + "step": 10789 + }, + { + "epoch": 0.6028438137274074, + "grad_norm": 0.6218631267547607, + "learning_rate": 0.0007009468848050202, + "loss": 0.5652, + "step": 10790 + }, + { + "epoch": 0.6028996843310892, + "grad_norm": 0.7320855259895325, + "learning_rate": 0.0007009188704616764, + "loss": 0.6505, + "step": 10791 + }, + { + "epoch": 0.602955554934771, + "grad_norm": 0.6277717351913452, + "learning_rate": 0.0007008908561183326, + "loss": 0.3948, + "step": 10792 + }, + { + "epoch": 0.603011425538453, + "grad_norm": 1.4014288187026978, + "learning_rate": 0.0007008628417749888, + "loss": 0.5247, + "step": 10793 + }, + { + "epoch": 0.6030672961421348, + "grad_norm": 1.4914559125900269, + "learning_rate": 0.000700834827431645, + "loss": 0.3825, + "step": 10794 + }, + { + "epoch": 0.6031231667458167, + "grad_norm": 0.47788023948669434, + "learning_rate": 0.0007008068130883012, + "loss": 0.5756, + "step": 10795 + }, + { + "epoch": 0.6031790373494985, + "grad_norm": 0.4510924220085144, + "learning_rate": 0.0007007787987449574, + "loss": 0.386, + "step": 10796 + }, + { + "epoch": 0.6032349079531805, + "grad_norm": 0.38995233178138733, + "learning_rate": 0.0007007507844016136, + "loss": 0.3352, + "step": 10797 + }, + { + "epoch": 0.6032907785568623, + "grad_norm": 0.4332106113433838, + "learning_rate": 0.00070072277005827, + "loss": 0.4744, + "step": 10798 + }, + { + "epoch": 0.6033466491605441, + "grad_norm": 0.510684609413147, + "learning_rate": 0.000700694755714926, + "loss": 0.4159, + "step": 10799 + }, + { + "epoch": 0.6034025197642261, + "grad_norm": 0.4838373363018036, + "learning_rate": 0.0007006667413715823, + "loss": 0.3973, + "step": 10800 + }, + { + "epoch": 0.6034583903679079, + "grad_norm": 0.48856696486473083, + "learning_rate": 0.0007006387270282385, + "loss": 0.5147, + "step": 10801 + }, + { + "epoch": 0.6035142609715898, + "grad_norm": 3.363673448562622, + "learning_rate": 0.0007006107126848948, + "loss": 0.6084, + "step": 10802 + }, + { + "epoch": 0.6035701315752716, + "grad_norm": 0.9888228178024292, + "learning_rate": 0.0007005826983415509, + "loss": 0.5008, + "step": 10803 + }, + { + "epoch": 0.6036260021789536, + "grad_norm": 0.49211588501930237, + "learning_rate": 0.0007005546839982071, + "loss": 0.4843, + "step": 10804 + }, + { + "epoch": 0.6036818727826354, + "grad_norm": 0.4502878785133362, + "learning_rate": 0.0007005266696548634, + "loss": 0.3986, + "step": 10805 + }, + { + "epoch": 0.6037377433863172, + "grad_norm": 0.7306734323501587, + "learning_rate": 0.0007004986553115195, + "loss": 0.4333, + "step": 10806 + }, + { + "epoch": 0.6037936139899992, + "grad_norm": 1.1187536716461182, + "learning_rate": 0.0007004706409681757, + "loss": 0.4029, + "step": 10807 + }, + { + "epoch": 0.603849484593681, + "grad_norm": 0.45482927560806274, + "learning_rate": 0.0007004426266248319, + "loss": 0.5841, + "step": 10808 + }, + { + "epoch": 0.6039053551973629, + "grad_norm": 0.4856034219264984, + "learning_rate": 0.0007004146122814882, + "loss": 0.4136, + "step": 10809 + }, + { + "epoch": 0.6039612258010448, + "grad_norm": 0.9522944688796997, + "learning_rate": 0.0007003865979381443, + "loss": 0.4628, + "step": 10810 + }, + { + "epoch": 0.6040170964047267, + "grad_norm": 1.0689826011657715, + "learning_rate": 0.0007003585835948005, + "loss": 0.4759, + "step": 10811 + }, + { + "epoch": 0.6040729670084085, + "grad_norm": 1.774666666984558, + "learning_rate": 0.0007003305692514567, + "loss": 0.514, + "step": 10812 + }, + { + "epoch": 0.6041288376120904, + "grad_norm": 0.4887818694114685, + "learning_rate": 0.000700302554908113, + "loss": 0.4681, + "step": 10813 + }, + { + "epoch": 0.6041847082157723, + "grad_norm": 0.8642081618309021, + "learning_rate": 0.0007002745405647691, + "loss": 0.4151, + "step": 10814 + }, + { + "epoch": 0.6042405788194541, + "grad_norm": 0.3760095536708832, + "learning_rate": 0.0007002465262214253, + "loss": 0.4888, + "step": 10815 + }, + { + "epoch": 0.604296449423136, + "grad_norm": 0.44090336561203003, + "learning_rate": 0.0007002185118780817, + "loss": 0.4185, + "step": 10816 + }, + { + "epoch": 0.6043523200268179, + "grad_norm": 0.42825159430503845, + "learning_rate": 0.0007001904975347379, + "loss": 0.451, + "step": 10817 + }, + { + "epoch": 0.6044081906304998, + "grad_norm": 0.41711801290512085, + "learning_rate": 0.000700162483191394, + "loss": 0.3338, + "step": 10818 + }, + { + "epoch": 0.6044640612341816, + "grad_norm": 0.3988780677318573, + "learning_rate": 0.0007001344688480502, + "loss": 0.467, + "step": 10819 + }, + { + "epoch": 0.6045199318378635, + "grad_norm": 0.49119582772254944, + "learning_rate": 0.0007001064545047065, + "loss": 0.4779, + "step": 10820 + }, + { + "epoch": 0.6045758024415454, + "grad_norm": 1.43997061252594, + "learning_rate": 0.0007000784401613627, + "loss": 0.4632, + "step": 10821 + }, + { + "epoch": 0.6046316730452272, + "grad_norm": 6.609170913696289, + "learning_rate": 0.0007000504258180188, + "loss": 0.4183, + "step": 10822 + }, + { + "epoch": 0.6046875436489091, + "grad_norm": 0.8663058280944824, + "learning_rate": 0.0007000224114746751, + "loss": 0.4512, + "step": 10823 + }, + { + "epoch": 0.604743414252591, + "grad_norm": 0.5412480235099792, + "learning_rate": 0.0006999943971313313, + "loss": 0.4831, + "step": 10824 + }, + { + "epoch": 0.6047992848562729, + "grad_norm": 0.4516329765319824, + "learning_rate": 0.0006999663827879874, + "loss": 0.4173, + "step": 10825 + }, + { + "epoch": 0.6048551554599547, + "grad_norm": 0.7876225709915161, + "learning_rate": 0.0006999383684446436, + "loss": 0.5211, + "step": 10826 + }, + { + "epoch": 0.6049110260636367, + "grad_norm": 0.5225306153297424, + "learning_rate": 0.0006999103541012999, + "loss": 0.375, + "step": 10827 + }, + { + "epoch": 0.6049668966673185, + "grad_norm": 24.635576248168945, + "learning_rate": 0.0006998823397579561, + "loss": 0.3364, + "step": 10828 + }, + { + "epoch": 0.6050227672710004, + "grad_norm": 0.839431881904602, + "learning_rate": 0.0006998543254146122, + "loss": 0.4972, + "step": 10829 + }, + { + "epoch": 0.6050786378746822, + "grad_norm": 0.36285218596458435, + "learning_rate": 0.0006998263110712685, + "loss": 0.3509, + "step": 10830 + }, + { + "epoch": 0.6051345084783641, + "grad_norm": 0.6331982016563416, + "learning_rate": 0.0006997982967279247, + "loss": 0.5879, + "step": 10831 + }, + { + "epoch": 0.605190379082046, + "grad_norm": 0.5041036009788513, + "learning_rate": 0.0006997702823845809, + "loss": 0.4322, + "step": 10832 + }, + { + "epoch": 0.6052462496857278, + "grad_norm": 0.622738778591156, + "learning_rate": 0.000699742268041237, + "loss": 0.4063, + "step": 10833 + }, + { + "epoch": 0.6053021202894098, + "grad_norm": 0.6318166255950928, + "learning_rate": 0.0006997142536978934, + "loss": 0.4634, + "step": 10834 + }, + { + "epoch": 0.6053579908930916, + "grad_norm": 0.7400081753730774, + "learning_rate": 0.0006996862393545496, + "loss": 0.5121, + "step": 10835 + }, + { + "epoch": 0.6054138614967735, + "grad_norm": 1.587043046951294, + "learning_rate": 0.0006996582250112058, + "loss": 0.447, + "step": 10836 + }, + { + "epoch": 0.6054697321004553, + "grad_norm": 0.44833675026893616, + "learning_rate": 0.000699630210667862, + "loss": 0.4271, + "step": 10837 + }, + { + "epoch": 0.6055256027041372, + "grad_norm": 0.4454409182071686, + "learning_rate": 0.0006996021963245182, + "loss": 0.3995, + "step": 10838 + }, + { + "epoch": 0.6055814733078191, + "grad_norm": 0.6134569644927979, + "learning_rate": 0.0006995741819811744, + "loss": 0.447, + "step": 10839 + }, + { + "epoch": 0.6056373439115009, + "grad_norm": 0.5053538084030151, + "learning_rate": 0.0006995461676378306, + "loss": 0.4436, + "step": 10840 + }, + { + "epoch": 0.6056932145151829, + "grad_norm": 0.5637103319168091, + "learning_rate": 0.0006995181532944868, + "loss": 0.4472, + "step": 10841 + }, + { + "epoch": 0.6057490851188647, + "grad_norm": 0.4254436194896698, + "learning_rate": 0.000699490138951143, + "loss": 0.4786, + "step": 10842 + }, + { + "epoch": 0.6058049557225466, + "grad_norm": 0.5658349990844727, + "learning_rate": 0.0006994621246077992, + "loss": 0.2974, + "step": 10843 + }, + { + "epoch": 0.6058608263262285, + "grad_norm": 1.2448679208755493, + "learning_rate": 0.0006994341102644555, + "loss": 0.4274, + "step": 10844 + }, + { + "epoch": 0.6059166969299103, + "grad_norm": 0.3773098289966583, + "learning_rate": 0.0006994060959211116, + "loss": 0.408, + "step": 10845 + }, + { + "epoch": 0.6059725675335922, + "grad_norm": 0.47492921352386475, + "learning_rate": 0.0006993780815777678, + "loss": 0.3985, + "step": 10846 + }, + { + "epoch": 0.606028438137274, + "grad_norm": 0.5155929923057556, + "learning_rate": 0.000699350067234424, + "loss": 0.5109, + "step": 10847 + }, + { + "epoch": 0.606084308740956, + "grad_norm": 0.4165742099285126, + "learning_rate": 0.0006993220528910802, + "loss": 0.375, + "step": 10848 + }, + { + "epoch": 0.6061401793446378, + "grad_norm": 0.48419514298439026, + "learning_rate": 0.0006992940385477364, + "loss": 0.4402, + "step": 10849 + }, + { + "epoch": 0.6061960499483197, + "grad_norm": 0.5024129152297974, + "learning_rate": 0.0006992660242043927, + "loss": 0.3658, + "step": 10850 + }, + { + "epoch": 0.6062519205520016, + "grad_norm": 1.8765443563461304, + "learning_rate": 0.0006992380098610489, + "loss": 0.4357, + "step": 10851 + }, + { + "epoch": 0.6063077911556835, + "grad_norm": 0.40016937255859375, + "learning_rate": 0.0006992099955177051, + "loss": 0.4296, + "step": 10852 + }, + { + "epoch": 0.6063636617593653, + "grad_norm": 0.7513602375984192, + "learning_rate": 0.0006991819811743613, + "loss": 0.5975, + "step": 10853 + }, + { + "epoch": 0.6064195323630471, + "grad_norm": 0.6686567664146423, + "learning_rate": 0.0006991539668310175, + "loss": 0.407, + "step": 10854 + }, + { + "epoch": 0.6064754029667291, + "grad_norm": 1.6638398170471191, + "learning_rate": 0.0006991259524876738, + "loss": 0.6571, + "step": 10855 + }, + { + "epoch": 0.6065312735704109, + "grad_norm": 0.5215296149253845, + "learning_rate": 0.0006990979381443299, + "loss": 0.3909, + "step": 10856 + }, + { + "epoch": 0.6065871441740928, + "grad_norm": 9.0654296875, + "learning_rate": 0.0006990699238009861, + "loss": 0.4206, + "step": 10857 + }, + { + "epoch": 0.6066430147777747, + "grad_norm": 0.9241636991500854, + "learning_rate": 0.0006990419094576423, + "loss": 0.4761, + "step": 10858 + }, + { + "epoch": 0.6066988853814566, + "grad_norm": 0.7309237718582153, + "learning_rate": 0.0006990138951142986, + "loss": 0.5202, + "step": 10859 + }, + { + "epoch": 0.6067547559851384, + "grad_norm": 0.4638083577156067, + "learning_rate": 0.0006989858807709547, + "loss": 0.4551, + "step": 10860 + }, + { + "epoch": 0.6068106265888203, + "grad_norm": 1.1662614345550537, + "learning_rate": 0.0006989578664276109, + "loss": 0.4206, + "step": 10861 + }, + { + "epoch": 0.6068664971925022, + "grad_norm": 0.6447190046310425, + "learning_rate": 0.0006989298520842672, + "loss": 0.7557, + "step": 10862 + }, + { + "epoch": 0.606922367796184, + "grad_norm": 0.6600208878517151, + "learning_rate": 0.0006989018377409234, + "loss": 0.5982, + "step": 10863 + }, + { + "epoch": 0.6069782383998659, + "grad_norm": 1.099826455116272, + "learning_rate": 0.0006988738233975795, + "loss": 0.5259, + "step": 10864 + }, + { + "epoch": 0.6070341090035478, + "grad_norm": 0.3981029987335205, + "learning_rate": 0.0006988458090542357, + "loss": 0.4926, + "step": 10865 + }, + { + "epoch": 0.6070899796072297, + "grad_norm": 0.7078547477722168, + "learning_rate": 0.000698817794710892, + "loss": 0.4509, + "step": 10866 + }, + { + "epoch": 0.6071458502109115, + "grad_norm": 0.6348114609718323, + "learning_rate": 0.0006987897803675481, + "loss": 0.4571, + "step": 10867 + }, + { + "epoch": 0.6072017208145934, + "grad_norm": 0.7448760867118835, + "learning_rate": 0.0006987617660242044, + "loss": 0.4064, + "step": 10868 + }, + { + "epoch": 0.6072575914182753, + "grad_norm": 1.6054201126098633, + "learning_rate": 0.0006987337516808607, + "loss": 0.4455, + "step": 10869 + }, + { + "epoch": 0.6073134620219571, + "grad_norm": 0.42748793959617615, + "learning_rate": 0.0006987057373375169, + "loss": 0.4228, + "step": 10870 + }, + { + "epoch": 0.607369332625639, + "grad_norm": 1.3938915729522705, + "learning_rate": 0.000698677722994173, + "loss": 0.3648, + "step": 10871 + }, + { + "epoch": 0.6074252032293209, + "grad_norm": 0.6557050347328186, + "learning_rate": 0.0006986497086508292, + "loss": 0.5458, + "step": 10872 + }, + { + "epoch": 0.6074810738330028, + "grad_norm": 0.7501401305198669, + "learning_rate": 0.0006986216943074855, + "loss": 0.4095, + "step": 10873 + }, + { + "epoch": 0.6075369444366846, + "grad_norm": 0.5292900204658508, + "learning_rate": 0.0006985936799641417, + "loss": 0.4394, + "step": 10874 + }, + { + "epoch": 0.6075928150403666, + "grad_norm": 1.8074007034301758, + "learning_rate": 0.0006985656656207978, + "loss": 0.3639, + "step": 10875 + }, + { + "epoch": 0.6076486856440484, + "grad_norm": 0.5218722820281982, + "learning_rate": 0.0006985376512774541, + "loss": 0.3861, + "step": 10876 + }, + { + "epoch": 0.6077045562477302, + "grad_norm": 0.48415708541870117, + "learning_rate": 0.0006985096369341103, + "loss": 0.3462, + "step": 10877 + }, + { + "epoch": 0.6077604268514122, + "grad_norm": 0.732953667640686, + "learning_rate": 0.0006984816225907665, + "loss": 0.462, + "step": 10878 + }, + { + "epoch": 0.607816297455094, + "grad_norm": 0.4412771463394165, + "learning_rate": 0.0006984536082474226, + "loss": 0.4334, + "step": 10879 + }, + { + "epoch": 0.6078721680587759, + "grad_norm": 0.4791337549686432, + "learning_rate": 0.0006984255939040789, + "loss": 0.4031, + "step": 10880 + }, + { + "epoch": 0.6079280386624577, + "grad_norm": 0.5995904803276062, + "learning_rate": 0.0006983975795607351, + "loss": 0.5016, + "step": 10881 + }, + { + "epoch": 0.6079839092661397, + "grad_norm": 0.9440083503723145, + "learning_rate": 0.0006983695652173913, + "loss": 0.4931, + "step": 10882 + }, + { + "epoch": 0.6080397798698215, + "grad_norm": 0.9279505610466003, + "learning_rate": 0.0006983415508740474, + "loss": 0.459, + "step": 10883 + }, + { + "epoch": 0.6080956504735033, + "grad_norm": 0.5577465295791626, + "learning_rate": 0.0006983135365307038, + "loss": 0.3795, + "step": 10884 + }, + { + "epoch": 0.6081515210771853, + "grad_norm": 0.9137718677520752, + "learning_rate": 0.00069828552218736, + "loss": 0.5568, + "step": 10885 + }, + { + "epoch": 0.6082073916808671, + "grad_norm": 0.9247826933860779, + "learning_rate": 0.0006982575078440162, + "loss": 0.4967, + "step": 10886 + }, + { + "epoch": 0.608263262284549, + "grad_norm": 0.9960607886314392, + "learning_rate": 0.0006982294935006724, + "loss": 0.5051, + "step": 10887 + }, + { + "epoch": 0.6083191328882308, + "grad_norm": 0.5004391670227051, + "learning_rate": 0.0006982014791573286, + "loss": 0.4265, + "step": 10888 + }, + { + "epoch": 0.6083750034919128, + "grad_norm": 0.7375860214233398, + "learning_rate": 0.0006981734648139848, + "loss": 0.4514, + "step": 10889 + }, + { + "epoch": 0.6084308740955946, + "grad_norm": 1.5579235553741455, + "learning_rate": 0.0006981454504706409, + "loss": 0.4837, + "step": 10890 + }, + { + "epoch": 0.6084867446992764, + "grad_norm": 0.8194553256034851, + "learning_rate": 0.0006981174361272972, + "loss": 0.4089, + "step": 10891 + }, + { + "epoch": 0.6085426153029584, + "grad_norm": 0.44341611862182617, + "learning_rate": 0.0006980894217839534, + "loss": 0.4602, + "step": 10892 + }, + { + "epoch": 0.6085984859066402, + "grad_norm": 6.702724456787109, + "learning_rate": 0.0006980614074406096, + "loss": 0.3978, + "step": 10893 + }, + { + "epoch": 0.6086543565103221, + "grad_norm": 0.7225988507270813, + "learning_rate": 0.0006980333930972658, + "loss": 0.4199, + "step": 10894 + }, + { + "epoch": 0.608710227114004, + "grad_norm": 0.4699893891811371, + "learning_rate": 0.000698005378753922, + "loss": 0.4035, + "step": 10895 + }, + { + "epoch": 0.6087660977176859, + "grad_norm": 1.332767367362976, + "learning_rate": 0.0006979773644105782, + "loss": 0.3855, + "step": 10896 + }, + { + "epoch": 0.6088219683213677, + "grad_norm": 0.5776852369308472, + "learning_rate": 0.0006979493500672344, + "loss": 0.433, + "step": 10897 + }, + { + "epoch": 0.6088778389250495, + "grad_norm": 0.6432585716247559, + "learning_rate": 0.0006979213357238906, + "loss": 0.5445, + "step": 10898 + }, + { + "epoch": 0.6089337095287315, + "grad_norm": 0.6219654083251953, + "learning_rate": 0.0006978933213805468, + "loss": 0.3271, + "step": 10899 + }, + { + "epoch": 0.6089895801324133, + "grad_norm": 0.571088433265686, + "learning_rate": 0.000697865307037203, + "loss": 0.5447, + "step": 10900 + }, + { + "epoch": 0.6090454507360952, + "grad_norm": 3.3319194316864014, + "learning_rate": 0.0006978372926938594, + "loss": 0.4444, + "step": 10901 + }, + { + "epoch": 0.6091013213397771, + "grad_norm": 0.6118518114089966, + "learning_rate": 0.0006978092783505155, + "loss": 0.4434, + "step": 10902 + }, + { + "epoch": 0.609157191943459, + "grad_norm": 0.47292619943618774, + "learning_rate": 0.0006977812640071717, + "loss": 0.5408, + "step": 10903 + }, + { + "epoch": 0.6092130625471408, + "grad_norm": 1.5404787063598633, + "learning_rate": 0.0006977532496638279, + "loss": 0.4092, + "step": 10904 + }, + { + "epoch": 0.6092689331508226, + "grad_norm": 0.39497682452201843, + "learning_rate": 0.0006977252353204842, + "loss": 0.4806, + "step": 10905 + }, + { + "epoch": 0.6093248037545046, + "grad_norm": 0.4069972038269043, + "learning_rate": 0.0006976972209771403, + "loss": 0.467, + "step": 10906 + }, + { + "epoch": 0.6093806743581864, + "grad_norm": 0.6865409016609192, + "learning_rate": 0.0006976692066337965, + "loss": 0.3223, + "step": 10907 + }, + { + "epoch": 0.6094365449618683, + "grad_norm": 0.8140150904655457, + "learning_rate": 0.0006976411922904528, + "loss": 0.4227, + "step": 10908 + }, + { + "epoch": 0.6094924155655502, + "grad_norm": 0.6379975080490112, + "learning_rate": 0.0006976131779471089, + "loss": 0.4979, + "step": 10909 + }, + { + "epoch": 0.6095482861692321, + "grad_norm": 0.40799611806869507, + "learning_rate": 0.0006975851636037651, + "loss": 0.4022, + "step": 10910 + }, + { + "epoch": 0.6096041567729139, + "grad_norm": 1.647381067276001, + "learning_rate": 0.0006975571492604213, + "loss": 0.4534, + "step": 10911 + }, + { + "epoch": 0.6096600273765959, + "grad_norm": 0.537512481212616, + "learning_rate": 0.0006975291349170776, + "loss": 0.4439, + "step": 10912 + }, + { + "epoch": 0.6097158979802777, + "grad_norm": 0.5926315784454346, + "learning_rate": 0.0006975011205737337, + "loss": 0.4996, + "step": 10913 + }, + { + "epoch": 0.6097717685839595, + "grad_norm": 0.6692612767219543, + "learning_rate": 0.0006974731062303899, + "loss": 0.4813, + "step": 10914 + }, + { + "epoch": 0.6098276391876414, + "grad_norm": 0.44668567180633545, + "learning_rate": 0.0006974450918870461, + "loss": 0.4226, + "step": 10915 + }, + { + "epoch": 0.6098835097913233, + "grad_norm": 0.5556468367576599, + "learning_rate": 0.0006974170775437025, + "loss": 0.4566, + "step": 10916 + }, + { + "epoch": 0.6099393803950052, + "grad_norm": 0.6892062425613403, + "learning_rate": 0.0006973890632003585, + "loss": 0.4026, + "step": 10917 + }, + { + "epoch": 0.609995250998687, + "grad_norm": 0.8187916874885559, + "learning_rate": 0.0006973610488570148, + "loss": 0.5705, + "step": 10918 + }, + { + "epoch": 0.610051121602369, + "grad_norm": 6.455482006072998, + "learning_rate": 0.0006973330345136711, + "loss": 0.4931, + "step": 10919 + }, + { + "epoch": 0.6101069922060508, + "grad_norm": 0.4251543879508972, + "learning_rate": 0.0006973050201703273, + "loss": 0.428, + "step": 10920 + }, + { + "epoch": 0.6101628628097326, + "grad_norm": 2.255570411682129, + "learning_rate": 0.0006972770058269834, + "loss": 0.7503, + "step": 10921 + }, + { + "epoch": 0.6102187334134145, + "grad_norm": 0.6039485931396484, + "learning_rate": 0.0006972489914836396, + "loss": 0.3888, + "step": 10922 + }, + { + "epoch": 0.6102746040170964, + "grad_norm": 0.4805282652378082, + "learning_rate": 0.0006972209771402959, + "loss": 0.4502, + "step": 10923 + }, + { + "epoch": 0.6103304746207783, + "grad_norm": 0.49019482731819153, + "learning_rate": 0.0006971929627969521, + "loss": 0.5369, + "step": 10924 + }, + { + "epoch": 0.6103863452244601, + "grad_norm": 0.4538500905036926, + "learning_rate": 0.0006971649484536082, + "loss": 0.4859, + "step": 10925 + }, + { + "epoch": 0.6104422158281421, + "grad_norm": 0.469007670879364, + "learning_rate": 0.0006971369341102645, + "loss": 0.4746, + "step": 10926 + }, + { + "epoch": 0.6104980864318239, + "grad_norm": 0.5524870157241821, + "learning_rate": 0.0006971089197669207, + "loss": 0.5709, + "step": 10927 + }, + { + "epoch": 0.6105539570355057, + "grad_norm": 1.4263373613357544, + "learning_rate": 0.0006970809054235769, + "loss": 0.4888, + "step": 10928 + }, + { + "epoch": 0.6106098276391877, + "grad_norm": 0.4480831027030945, + "learning_rate": 0.000697052891080233, + "loss": 0.4555, + "step": 10929 + }, + { + "epoch": 0.6106656982428695, + "grad_norm": 0.49150988459587097, + "learning_rate": 0.0006970248767368893, + "loss": 0.3473, + "step": 10930 + }, + { + "epoch": 0.6107215688465514, + "grad_norm": 1.2766716480255127, + "learning_rate": 0.0006969968623935455, + "loss": 0.4929, + "step": 10931 + }, + { + "epoch": 0.6107774394502332, + "grad_norm": 0.5195226669311523, + "learning_rate": 0.0006969688480502016, + "loss": 0.4759, + "step": 10932 + }, + { + "epoch": 0.6108333100539152, + "grad_norm": 0.7987514138221741, + "learning_rate": 0.000696940833706858, + "loss": 0.3959, + "step": 10933 + }, + { + "epoch": 0.610889180657597, + "grad_norm": 0.8689250349998474, + "learning_rate": 0.0006969128193635142, + "loss": 0.4192, + "step": 10934 + }, + { + "epoch": 0.6109450512612788, + "grad_norm": 0.5640870332717896, + "learning_rate": 0.0006968848050201704, + "loss": 0.5816, + "step": 10935 + }, + { + "epoch": 0.6110009218649608, + "grad_norm": 0.3610035479068756, + "learning_rate": 0.0006968567906768265, + "loss": 0.4171, + "step": 10936 + }, + { + "epoch": 0.6110567924686426, + "grad_norm": 0.4541049301624298, + "learning_rate": 0.0006968287763334828, + "loss": 0.477, + "step": 10937 + }, + { + "epoch": 0.6111126630723245, + "grad_norm": 0.6329872012138367, + "learning_rate": 0.000696800761990139, + "loss": 0.4497, + "step": 10938 + }, + { + "epoch": 0.6111685336760063, + "grad_norm": 0.43217381834983826, + "learning_rate": 0.0006967727476467952, + "loss": 0.4083, + "step": 10939 + }, + { + "epoch": 0.6112244042796883, + "grad_norm": 0.5640842318534851, + "learning_rate": 0.0006967447333034514, + "loss": 0.5913, + "step": 10940 + }, + { + "epoch": 0.6112802748833701, + "grad_norm": 0.43353933095932007, + "learning_rate": 0.0006967167189601076, + "loss": 0.4621, + "step": 10941 + }, + { + "epoch": 0.611336145487052, + "grad_norm": 0.5792891383171082, + "learning_rate": 0.0006966887046167638, + "loss": 0.409, + "step": 10942 + }, + { + "epoch": 0.6113920160907339, + "grad_norm": 0.5155993700027466, + "learning_rate": 0.00069666069027342, + "loss": 0.4874, + "step": 10943 + }, + { + "epoch": 0.6114478866944157, + "grad_norm": 0.7079447507858276, + "learning_rate": 0.0006966326759300762, + "loss": 0.4046, + "step": 10944 + }, + { + "epoch": 0.6115037572980976, + "grad_norm": 0.3959319293498993, + "learning_rate": 0.0006966046615867324, + "loss": 0.4837, + "step": 10945 + }, + { + "epoch": 0.6115596279017794, + "grad_norm": 0.48974311351776123, + "learning_rate": 0.0006965766472433886, + "loss": 0.3908, + "step": 10946 + }, + { + "epoch": 0.6116154985054614, + "grad_norm": 0.6370267271995544, + "learning_rate": 0.0006965486329000449, + "loss": 0.5665, + "step": 10947 + }, + { + "epoch": 0.6116713691091432, + "grad_norm": 0.4947304427623749, + "learning_rate": 0.000696520618556701, + "loss": 0.5079, + "step": 10948 + }, + { + "epoch": 0.611727239712825, + "grad_norm": 0.5393973588943481, + "learning_rate": 0.0006964926042133572, + "loss": 0.4136, + "step": 10949 + }, + { + "epoch": 0.611783110316507, + "grad_norm": 0.4965747892856598, + "learning_rate": 0.0006964645898700134, + "loss": 0.3792, + "step": 10950 + }, + { + "epoch": 0.6118389809201888, + "grad_norm": 0.4697798788547516, + "learning_rate": 0.0006964365755266698, + "loss": 0.4291, + "step": 10951 + }, + { + "epoch": 0.6118948515238707, + "grad_norm": 0.5711991786956787, + "learning_rate": 0.0006964085611833259, + "loss": 0.4247, + "step": 10952 + }, + { + "epoch": 0.6119507221275526, + "grad_norm": 0.8259242177009583, + "learning_rate": 0.0006963805468399821, + "loss": 0.4586, + "step": 10953 + }, + { + "epoch": 0.6120065927312345, + "grad_norm": 0.39672979712486267, + "learning_rate": 0.0006963525324966383, + "loss": 0.4599, + "step": 10954 + }, + { + "epoch": 0.6120624633349163, + "grad_norm": 4.2349724769592285, + "learning_rate": 0.0006963245181532945, + "loss": 0.4786, + "step": 10955 + }, + { + "epoch": 0.6121183339385982, + "grad_norm": 0.4299297034740448, + "learning_rate": 0.0006962965038099507, + "loss": 0.4399, + "step": 10956 + }, + { + "epoch": 0.6121742045422801, + "grad_norm": 0.4782244563102722, + "learning_rate": 0.0006962684894666069, + "loss": 0.3989, + "step": 10957 + }, + { + "epoch": 0.612230075145962, + "grad_norm": 0.44280490279197693, + "learning_rate": 0.0006962404751232632, + "loss": 0.3986, + "step": 10958 + }, + { + "epoch": 0.6122859457496438, + "grad_norm": 0.5937649011611938, + "learning_rate": 0.0006962124607799193, + "loss": 0.5007, + "step": 10959 + }, + { + "epoch": 0.6123418163533257, + "grad_norm": 1.9391570091247559, + "learning_rate": 0.0006961844464365755, + "loss": 0.5977, + "step": 10960 + }, + { + "epoch": 0.6123976869570076, + "grad_norm": 0.46384742856025696, + "learning_rate": 0.0006961564320932317, + "loss": 0.392, + "step": 10961 + }, + { + "epoch": 0.6124535575606894, + "grad_norm": 0.4626814126968384, + "learning_rate": 0.000696128417749888, + "loss": 0.5098, + "step": 10962 + }, + { + "epoch": 0.6125094281643713, + "grad_norm": 0.3600527048110962, + "learning_rate": 0.0006961004034065441, + "loss": 0.3959, + "step": 10963 + }, + { + "epoch": 0.6125652987680532, + "grad_norm": 0.7628322839736938, + "learning_rate": 0.0006960723890632003, + "loss": 0.4151, + "step": 10964 + }, + { + "epoch": 0.612621169371735, + "grad_norm": 0.597402036190033, + "learning_rate": 0.0006960443747198566, + "loss": 0.4331, + "step": 10965 + }, + { + "epoch": 0.6126770399754169, + "grad_norm": 0.4972659647464752, + "learning_rate": 0.0006960163603765129, + "loss": 0.4254, + "step": 10966 + }, + { + "epoch": 0.6127329105790988, + "grad_norm": 0.7882716655731201, + "learning_rate": 0.000695988346033169, + "loss": 0.5641, + "step": 10967 + }, + { + "epoch": 0.6127887811827807, + "grad_norm": 0.5656225681304932, + "learning_rate": 0.0006959603316898252, + "loss": 0.4982, + "step": 10968 + }, + { + "epoch": 0.6128446517864625, + "grad_norm": 0.5417853593826294, + "learning_rate": 0.0006959323173464815, + "loss": 0.3804, + "step": 10969 + }, + { + "epoch": 0.6129005223901445, + "grad_norm": 0.3930934965610504, + "learning_rate": 0.0006959043030031377, + "loss": 0.3677, + "step": 10970 + }, + { + "epoch": 0.6129563929938263, + "grad_norm": 0.5083939433097839, + "learning_rate": 0.0006958762886597938, + "loss": 0.5755, + "step": 10971 + }, + { + "epoch": 0.6130122635975082, + "grad_norm": 2.953263759613037, + "learning_rate": 0.0006958482743164501, + "loss": 0.4156, + "step": 10972 + }, + { + "epoch": 0.61306813420119, + "grad_norm": 0.6896535158157349, + "learning_rate": 0.0006958202599731063, + "loss": 0.35, + "step": 10973 + }, + { + "epoch": 0.613124004804872, + "grad_norm": 5.467078685760498, + "learning_rate": 0.0006957922456297624, + "loss": 0.477, + "step": 10974 + }, + { + "epoch": 0.6131798754085538, + "grad_norm": 0.5780688524246216, + "learning_rate": 0.0006957642312864186, + "loss": 0.5813, + "step": 10975 + }, + { + "epoch": 0.6132357460122356, + "grad_norm": 0.9960601925849915, + "learning_rate": 0.0006957362169430749, + "loss": 0.4089, + "step": 10976 + }, + { + "epoch": 0.6132916166159176, + "grad_norm": 0.7866194248199463, + "learning_rate": 0.0006957082025997311, + "loss": 0.4687, + "step": 10977 + }, + { + "epoch": 0.6133474872195994, + "grad_norm": 0.4371885061264038, + "learning_rate": 0.0006956801882563872, + "loss": 0.4402, + "step": 10978 + }, + { + "epoch": 0.6134033578232813, + "grad_norm": 0.8668546080589294, + "learning_rate": 0.0006956521739130435, + "loss": 0.4986, + "step": 10979 + }, + { + "epoch": 0.6134592284269631, + "grad_norm": 0.4797297716140747, + "learning_rate": 0.0006956241595696997, + "loss": 0.5272, + "step": 10980 + }, + { + "epoch": 0.613515099030645, + "grad_norm": 0.641409695148468, + "learning_rate": 0.0006955961452263559, + "loss": 0.6005, + "step": 10981 + }, + { + "epoch": 0.6135709696343269, + "grad_norm": 0.5335143804550171, + "learning_rate": 0.000695568130883012, + "loss": 0.4632, + "step": 10982 + }, + { + "epoch": 0.6136268402380087, + "grad_norm": 0.6781131625175476, + "learning_rate": 0.0006955401165396683, + "loss": 0.5589, + "step": 10983 + }, + { + "epoch": 0.6136827108416907, + "grad_norm": 0.4954073429107666, + "learning_rate": 0.0006955121021963246, + "loss": 0.4589, + "step": 10984 + }, + { + "epoch": 0.6137385814453725, + "grad_norm": 1.046730875968933, + "learning_rate": 0.0006954840878529808, + "loss": 0.6182, + "step": 10985 + }, + { + "epoch": 0.6137944520490544, + "grad_norm": 0.5065287947654724, + "learning_rate": 0.0006954560735096369, + "loss": 0.445, + "step": 10986 + }, + { + "epoch": 0.6138503226527363, + "grad_norm": 40.6158561706543, + "learning_rate": 0.0006954280591662932, + "loss": 0.5445, + "step": 10987 + }, + { + "epoch": 0.6139061932564182, + "grad_norm": 0.45153161883354187, + "learning_rate": 0.0006954000448229494, + "loss": 0.4138, + "step": 10988 + }, + { + "epoch": 0.6139620638601, + "grad_norm": 1.3783748149871826, + "learning_rate": 0.0006953720304796056, + "loss": 0.4598, + "step": 10989 + }, + { + "epoch": 0.6140179344637818, + "grad_norm": 0.5812196135520935, + "learning_rate": 0.0006953440161362618, + "loss": 0.4239, + "step": 10990 + }, + { + "epoch": 0.6140738050674638, + "grad_norm": 0.4081328213214874, + "learning_rate": 0.000695316001792918, + "loss": 0.5149, + "step": 10991 + }, + { + "epoch": 0.6141296756711456, + "grad_norm": 0.350934773683548, + "learning_rate": 0.0006952879874495742, + "loss": 0.3737, + "step": 10992 + }, + { + "epoch": 0.6141855462748275, + "grad_norm": 0.5066013336181641, + "learning_rate": 0.0006952599731062304, + "loss": 0.4024, + "step": 10993 + }, + { + "epoch": 0.6142414168785094, + "grad_norm": 0.5362482070922852, + "learning_rate": 0.0006952319587628866, + "loss": 0.5104, + "step": 10994 + }, + { + "epoch": 0.6142972874821913, + "grad_norm": 0.43415412306785583, + "learning_rate": 0.0006952039444195428, + "loss": 0.4701, + "step": 10995 + }, + { + "epoch": 0.6143531580858731, + "grad_norm": 0.6797946095466614, + "learning_rate": 0.000695175930076199, + "loss": 0.5311, + "step": 10996 + }, + { + "epoch": 0.6144090286895549, + "grad_norm": 0.4768713712692261, + "learning_rate": 0.0006951479157328552, + "loss": 0.3779, + "step": 10997 + }, + { + "epoch": 0.6144648992932369, + "grad_norm": 0.5099359154701233, + "learning_rate": 0.0006951199013895114, + "loss": 0.5008, + "step": 10998 + }, + { + "epoch": 0.6145207698969187, + "grad_norm": 0.3818591237068176, + "learning_rate": 0.0006950918870461676, + "loss": 0.3781, + "step": 10999 + }, + { + "epoch": 0.6145766405006006, + "grad_norm": 0.9262353181838989, + "learning_rate": 0.0006950638727028238, + "loss": 0.5451, + "step": 11000 + }, + { + "epoch": 0.6145766405006006, + "eval_cer": 0.0908044472084929, + "eval_loss": 0.3440577983856201, + "eval_runtime": 56.2517, + "eval_samples_per_second": 80.673, + "eval_steps_per_second": 5.049, + "eval_wer": 0.35834194575570516, + "step": 11000 + }, + { + "epoch": 0.6146325111042825, + "grad_norm": 0.8845785856246948, + "learning_rate": 0.00069503585835948, + "loss": 0.5296, + "step": 11001 + }, + { + "epoch": 0.6146883817079644, + "grad_norm": 0.4024244546890259, + "learning_rate": 0.0006950078440161363, + "loss": 0.4016, + "step": 11002 + }, + { + "epoch": 0.6147442523116462, + "grad_norm": 0.9227713942527771, + "learning_rate": 0.0006949798296727925, + "loss": 0.368, + "step": 11003 + }, + { + "epoch": 0.6148001229153282, + "grad_norm": 0.8236423134803772, + "learning_rate": 0.0006949518153294488, + "loss": 0.5616, + "step": 11004 + }, + { + "epoch": 0.61485599351901, + "grad_norm": 0.9554377198219299, + "learning_rate": 0.0006949238009861049, + "loss": 0.6976, + "step": 11005 + }, + { + "epoch": 0.6149118641226918, + "grad_norm": 0.7624560594558716, + "learning_rate": 0.0006948957866427611, + "loss": 0.5954, + "step": 11006 + }, + { + "epoch": 0.6149677347263737, + "grad_norm": 0.5328403115272522, + "learning_rate": 0.0006948677722994173, + "loss": 0.3826, + "step": 11007 + }, + { + "epoch": 0.6150236053300556, + "grad_norm": 0.6098049879074097, + "learning_rate": 0.0006948397579560736, + "loss": 0.4312, + "step": 11008 + }, + { + "epoch": 0.6150794759337375, + "grad_norm": 0.34902772307395935, + "learning_rate": 0.0006948117436127297, + "loss": 0.3943, + "step": 11009 + }, + { + "epoch": 0.6151353465374193, + "grad_norm": 0.7988739609718323, + "learning_rate": 0.0006947837292693859, + "loss": 0.4286, + "step": 11010 + }, + { + "epoch": 0.6151912171411013, + "grad_norm": 0.5671432018280029, + "learning_rate": 0.0006947557149260422, + "loss": 0.3875, + "step": 11011 + }, + { + "epoch": 0.6152470877447831, + "grad_norm": 0.9224414825439453, + "learning_rate": 0.0006947277005826984, + "loss": 0.4567, + "step": 11012 + }, + { + "epoch": 0.6153029583484649, + "grad_norm": 0.5870524644851685, + "learning_rate": 0.0006946996862393545, + "loss": 0.5061, + "step": 11013 + }, + { + "epoch": 0.6153588289521468, + "grad_norm": 0.9653826355934143, + "learning_rate": 0.0006946716718960107, + "loss": 0.3593, + "step": 11014 + }, + { + "epoch": 0.6154146995558287, + "grad_norm": 0.9053806662559509, + "learning_rate": 0.000694643657552667, + "loss": 0.4029, + "step": 11015 + }, + { + "epoch": 0.6154705701595106, + "grad_norm": 0.7190251350402832, + "learning_rate": 0.0006946156432093231, + "loss": 0.5395, + "step": 11016 + }, + { + "epoch": 0.6155264407631924, + "grad_norm": 0.7906932830810547, + "learning_rate": 0.0006945876288659793, + "loss": 0.3535, + "step": 11017 + }, + { + "epoch": 0.6155823113668744, + "grad_norm": 0.571099579334259, + "learning_rate": 0.0006945596145226357, + "loss": 0.3826, + "step": 11018 + }, + { + "epoch": 0.6156381819705562, + "grad_norm": 0.4973728656768799, + "learning_rate": 0.0006945316001792919, + "loss": 0.4536, + "step": 11019 + }, + { + "epoch": 0.615694052574238, + "grad_norm": 0.4841751754283905, + "learning_rate": 0.000694503585835948, + "loss": 0.3797, + "step": 11020 + }, + { + "epoch": 0.61574992317792, + "grad_norm": 1.6200263500213623, + "learning_rate": 0.0006944755714926042, + "loss": 0.6006, + "step": 11021 + }, + { + "epoch": 0.6158057937816018, + "grad_norm": 0.7835395336151123, + "learning_rate": 0.0006944475571492605, + "loss": 0.6266, + "step": 11022 + }, + { + "epoch": 0.6158616643852837, + "grad_norm": 1.3697069883346558, + "learning_rate": 0.0006944195428059167, + "loss": 0.4512, + "step": 11023 + }, + { + "epoch": 0.6159175349889655, + "grad_norm": 0.4329674541950226, + "learning_rate": 0.0006943915284625728, + "loss": 0.3697, + "step": 11024 + }, + { + "epoch": 0.6159734055926475, + "grad_norm": 0.43835216760635376, + "learning_rate": 0.000694363514119229, + "loss": 0.3622, + "step": 11025 + }, + { + "epoch": 0.6160292761963293, + "grad_norm": 0.534782350063324, + "learning_rate": 0.0006943354997758853, + "loss": 0.3841, + "step": 11026 + }, + { + "epoch": 0.6160851468000111, + "grad_norm": 0.4641934633255005, + "learning_rate": 0.0006943074854325415, + "loss": 0.4855, + "step": 11027 + }, + { + "epoch": 0.6161410174036931, + "grad_norm": 0.5260438323020935, + "learning_rate": 0.0006942794710891976, + "loss": 0.5148, + "step": 11028 + }, + { + "epoch": 0.6161968880073749, + "grad_norm": 0.4230198264122009, + "learning_rate": 0.0006942514567458539, + "loss": 0.5276, + "step": 11029 + }, + { + "epoch": 0.6162527586110568, + "grad_norm": 0.456572562456131, + "learning_rate": 0.0006942234424025101, + "loss": 0.3749, + "step": 11030 + }, + { + "epoch": 0.6163086292147386, + "grad_norm": 0.46221446990966797, + "learning_rate": 0.0006941954280591663, + "loss": 0.4186, + "step": 11031 + }, + { + "epoch": 0.6163644998184206, + "grad_norm": 0.46335458755493164, + "learning_rate": 0.0006941674137158224, + "loss": 0.5502, + "step": 11032 + }, + { + "epoch": 0.6164203704221024, + "grad_norm": 4.279612064361572, + "learning_rate": 0.0006941393993724787, + "loss": 0.3883, + "step": 11033 + }, + { + "epoch": 0.6164762410257842, + "grad_norm": 0.7432553768157959, + "learning_rate": 0.000694111385029135, + "loss": 0.4261, + "step": 11034 + }, + { + "epoch": 0.6165321116294662, + "grad_norm": 0.7563694715499878, + "learning_rate": 0.0006940833706857912, + "loss": 0.4468, + "step": 11035 + }, + { + "epoch": 0.616587982233148, + "grad_norm": 0.4429987370967865, + "learning_rate": 0.0006940553563424474, + "loss": 0.4409, + "step": 11036 + }, + { + "epoch": 0.6166438528368299, + "grad_norm": 0.580010175704956, + "learning_rate": 0.0006940273419991036, + "loss": 0.3898, + "step": 11037 + }, + { + "epoch": 0.6166997234405118, + "grad_norm": 0.7709540128707886, + "learning_rate": 0.0006939993276557598, + "loss": 0.6364, + "step": 11038 + }, + { + "epoch": 0.6167555940441937, + "grad_norm": 1.1324424743652344, + "learning_rate": 0.0006939713133124159, + "loss": 0.4552, + "step": 11039 + }, + { + "epoch": 0.6168114646478755, + "grad_norm": 0.39821892976760864, + "learning_rate": 0.0006939432989690722, + "loss": 0.499, + "step": 11040 + }, + { + "epoch": 0.6168673352515573, + "grad_norm": 0.3835356533527374, + "learning_rate": 0.0006939152846257284, + "loss": 0.4026, + "step": 11041 + }, + { + "epoch": 0.6169232058552393, + "grad_norm": 1.8794928789138794, + "learning_rate": 0.0006938872702823846, + "loss": 0.4524, + "step": 11042 + }, + { + "epoch": 0.6169790764589211, + "grad_norm": 0.7022536993026733, + "learning_rate": 0.0006938592559390408, + "loss": 0.4776, + "step": 11043 + }, + { + "epoch": 0.617034947062603, + "grad_norm": 0.5183675289154053, + "learning_rate": 0.000693831241595697, + "loss": 0.5952, + "step": 11044 + }, + { + "epoch": 0.6170908176662849, + "grad_norm": 0.37328019738197327, + "learning_rate": 0.0006938032272523532, + "loss": 0.4036, + "step": 11045 + }, + { + "epoch": 0.6171466882699668, + "grad_norm": 0.664216935634613, + "learning_rate": 0.0006937752129090094, + "loss": 0.4708, + "step": 11046 + }, + { + "epoch": 0.6172025588736486, + "grad_norm": 0.5656876564025879, + "learning_rate": 0.0006937471985656656, + "loss": 0.3852, + "step": 11047 + }, + { + "epoch": 0.6172584294773305, + "grad_norm": 0.6418405175209045, + "learning_rate": 0.0006937191842223218, + "loss": 0.391, + "step": 11048 + }, + { + "epoch": 0.6173143000810124, + "grad_norm": 0.6026923656463623, + "learning_rate": 0.000693691169878978, + "loss": 0.5049, + "step": 11049 + }, + { + "epoch": 0.6173701706846942, + "grad_norm": 0.5105898976325989, + "learning_rate": 0.0006936631555356344, + "loss": 0.4304, + "step": 11050 + }, + { + "epoch": 0.6174260412883761, + "grad_norm": 6.242321968078613, + "learning_rate": 0.0006936351411922905, + "loss": 0.3911, + "step": 11051 + }, + { + "epoch": 0.617481911892058, + "grad_norm": 0.3466965854167938, + "learning_rate": 0.0006936071268489467, + "loss": 0.4809, + "step": 11052 + }, + { + "epoch": 0.6175377824957399, + "grad_norm": 0.6630131602287292, + "learning_rate": 0.0006935791125056029, + "loss": 0.4733, + "step": 11053 + }, + { + "epoch": 0.6175936530994217, + "grad_norm": 0.49306240677833557, + "learning_rate": 0.0006935510981622592, + "loss": 0.4239, + "step": 11054 + }, + { + "epoch": 0.6176495237031037, + "grad_norm": 0.4366578161716461, + "learning_rate": 0.0006935230838189153, + "loss": 0.4329, + "step": 11055 + }, + { + "epoch": 0.6177053943067855, + "grad_norm": 0.5243264436721802, + "learning_rate": 0.0006934950694755715, + "loss": 0.4052, + "step": 11056 + }, + { + "epoch": 0.6177612649104673, + "grad_norm": 0.4206082224845886, + "learning_rate": 0.0006934670551322278, + "loss": 0.6411, + "step": 11057 + }, + { + "epoch": 0.6178171355141492, + "grad_norm": 0.5498961806297302, + "learning_rate": 0.0006934390407888839, + "loss": 0.3731, + "step": 11058 + }, + { + "epoch": 0.6178730061178311, + "grad_norm": 0.4922085404396057, + "learning_rate": 0.0006934110264455401, + "loss": 0.4432, + "step": 11059 + }, + { + "epoch": 0.617928876721513, + "grad_norm": 0.5025709271430969, + "learning_rate": 0.0006933830121021963, + "loss": 0.4052, + "step": 11060 + }, + { + "epoch": 0.6179847473251948, + "grad_norm": 0.49691078066825867, + "learning_rate": 0.0006933549977588526, + "loss": 0.5017, + "step": 11061 + }, + { + "epoch": 0.6180406179288768, + "grad_norm": 1.1282058954238892, + "learning_rate": 0.0006933269834155087, + "loss": 0.4252, + "step": 11062 + }, + { + "epoch": 0.6180964885325586, + "grad_norm": 0.5865066051483154, + "learning_rate": 0.0006932989690721649, + "loss": 0.5862, + "step": 11063 + }, + { + "epoch": 0.6181523591362404, + "grad_norm": 0.42592713236808777, + "learning_rate": 0.0006932709547288211, + "loss": 0.3923, + "step": 11064 + }, + { + "epoch": 0.6182082297399223, + "grad_norm": 0.37372148036956787, + "learning_rate": 0.0006932429403854774, + "loss": 0.3723, + "step": 11065 + }, + { + "epoch": 0.6182641003436042, + "grad_norm": 0.5701408386230469, + "learning_rate": 0.0006932149260421335, + "loss": 0.4687, + "step": 11066 + }, + { + "epoch": 0.6183199709472861, + "grad_norm": 0.5134482383728027, + "learning_rate": 0.0006931869116987897, + "loss": 0.4574, + "step": 11067 + }, + { + "epoch": 0.6183758415509679, + "grad_norm": 0.37663978338241577, + "learning_rate": 0.0006931588973554461, + "loss": 0.3757, + "step": 11068 + }, + { + "epoch": 0.6184317121546499, + "grad_norm": 0.5840036869049072, + "learning_rate": 0.0006931308830121023, + "loss": 0.493, + "step": 11069 + }, + { + "epoch": 0.6184875827583317, + "grad_norm": 0.4429190158843994, + "learning_rate": 0.0006931028686687584, + "loss": 0.3696, + "step": 11070 + }, + { + "epoch": 0.6185434533620136, + "grad_norm": 0.5419504046440125, + "learning_rate": 0.0006930748543254146, + "loss": 0.3846, + "step": 11071 + }, + { + "epoch": 0.6185993239656955, + "grad_norm": 0.5614458918571472, + "learning_rate": 0.0006930468399820709, + "loss": 0.5358, + "step": 11072 + }, + { + "epoch": 0.6186551945693773, + "grad_norm": 0.5360759496688843, + "learning_rate": 0.0006930188256387271, + "loss": 0.4653, + "step": 11073 + }, + { + "epoch": 0.6187110651730592, + "grad_norm": 0.5503902435302734, + "learning_rate": 0.0006929908112953832, + "loss": 0.5244, + "step": 11074 + }, + { + "epoch": 0.618766935776741, + "grad_norm": 0.6099270582199097, + "learning_rate": 0.0006929627969520395, + "loss": 0.4806, + "step": 11075 + }, + { + "epoch": 0.618822806380423, + "grad_norm": 0.4945969581604004, + "learning_rate": 0.0006929347826086957, + "loss": 0.4325, + "step": 11076 + }, + { + "epoch": 0.6188786769841048, + "grad_norm": 0.6666768193244934, + "learning_rate": 0.0006929067682653519, + "loss": 0.5351, + "step": 11077 + }, + { + "epoch": 0.6189345475877867, + "grad_norm": 0.49068892002105713, + "learning_rate": 0.000692878753922008, + "loss": 0.4038, + "step": 11078 + }, + { + "epoch": 0.6189904181914686, + "grad_norm": 0.6131870746612549, + "learning_rate": 0.0006928507395786643, + "loss": 0.4338, + "step": 11079 + }, + { + "epoch": 0.6190462887951504, + "grad_norm": 14.639059066772461, + "learning_rate": 0.0006928227252353205, + "loss": 0.3864, + "step": 11080 + }, + { + "epoch": 0.6191021593988323, + "grad_norm": 1.1200779676437378, + "learning_rate": 0.0006927947108919766, + "loss": 0.4149, + "step": 11081 + }, + { + "epoch": 0.6191580300025141, + "grad_norm": 1.0797783136367798, + "learning_rate": 0.0006927666965486329, + "loss": 0.5401, + "step": 11082 + }, + { + "epoch": 0.6192139006061961, + "grad_norm": 0.4015829861164093, + "learning_rate": 0.0006927386822052891, + "loss": 0.4512, + "step": 11083 + }, + { + "epoch": 0.6192697712098779, + "grad_norm": 0.7268348932266235, + "learning_rate": 0.0006927106678619454, + "loss": 0.3566, + "step": 11084 + }, + { + "epoch": 0.6193256418135598, + "grad_norm": 0.5738482475280762, + "learning_rate": 0.0006926826535186015, + "loss": 0.4166, + "step": 11085 + }, + { + "epoch": 0.6193815124172417, + "grad_norm": 0.6557052135467529, + "learning_rate": 0.0006926546391752578, + "loss": 0.4961, + "step": 11086 + }, + { + "epoch": 0.6194373830209235, + "grad_norm": 0.6675397753715515, + "learning_rate": 0.000692626624831914, + "loss": 0.4896, + "step": 11087 + }, + { + "epoch": 0.6194932536246054, + "grad_norm": 0.5960477590560913, + "learning_rate": 0.0006925986104885702, + "loss": 0.408, + "step": 11088 + }, + { + "epoch": 0.6195491242282873, + "grad_norm": 0.6909528970718384, + "learning_rate": 0.0006925705961452264, + "loss": 0.4215, + "step": 11089 + }, + { + "epoch": 0.6196049948319692, + "grad_norm": 0.34164005517959595, + "learning_rate": 0.0006925425818018826, + "loss": 0.3737, + "step": 11090 + }, + { + "epoch": 0.619660865435651, + "grad_norm": 3.3743388652801514, + "learning_rate": 0.0006925145674585388, + "loss": 0.4502, + "step": 11091 + }, + { + "epoch": 0.6197167360393329, + "grad_norm": 0.40733346343040466, + "learning_rate": 0.000692486553115195, + "loss": 0.4904, + "step": 11092 + }, + { + "epoch": 0.6197726066430148, + "grad_norm": 1.3853651285171509, + "learning_rate": 0.0006924585387718512, + "loss": 0.518, + "step": 11093 + }, + { + "epoch": 0.6198284772466967, + "grad_norm": 0.7750471830368042, + "learning_rate": 0.0006924305244285074, + "loss": 0.3696, + "step": 11094 + }, + { + "epoch": 0.6198843478503785, + "grad_norm": 0.5101208090782166, + "learning_rate": 0.0006924025100851636, + "loss": 0.4139, + "step": 11095 + }, + { + "epoch": 0.6199402184540604, + "grad_norm": 0.7800289392471313, + "learning_rate": 0.0006923744957418198, + "loss": 0.487, + "step": 11096 + }, + { + "epoch": 0.6199960890577423, + "grad_norm": 0.4912012815475464, + "learning_rate": 0.000692346481398476, + "loss": 0.432, + "step": 11097 + }, + { + "epoch": 0.6200519596614241, + "grad_norm": 0.39764535427093506, + "learning_rate": 0.0006923184670551322, + "loss": 0.4072, + "step": 11098 + }, + { + "epoch": 0.620107830265106, + "grad_norm": 0.8579317927360535, + "learning_rate": 0.0006922904527117884, + "loss": 0.5656, + "step": 11099 + }, + { + "epoch": 0.6201637008687879, + "grad_norm": 0.5959371328353882, + "learning_rate": 0.0006922624383684448, + "loss": 0.4776, + "step": 11100 + }, + { + "epoch": 0.6202195714724698, + "grad_norm": 0.38905373215675354, + "learning_rate": 0.0006922344240251009, + "loss": 0.4404, + "step": 11101 + }, + { + "epoch": 0.6202754420761516, + "grad_norm": 0.32555627822875977, + "learning_rate": 0.0006922064096817571, + "loss": 0.4244, + "step": 11102 + }, + { + "epoch": 0.6203313126798335, + "grad_norm": 0.7344619035720825, + "learning_rate": 0.0006921783953384133, + "loss": 0.4511, + "step": 11103 + }, + { + "epoch": 0.6203871832835154, + "grad_norm": 0.8948274254798889, + "learning_rate": 0.0006921503809950695, + "loss": 0.5294, + "step": 11104 + }, + { + "epoch": 0.6204430538871972, + "grad_norm": 0.44325584173202515, + "learning_rate": 0.0006921223666517257, + "loss": 0.4657, + "step": 11105 + }, + { + "epoch": 0.6204989244908792, + "grad_norm": 1.115774154663086, + "learning_rate": 0.0006920943523083819, + "loss": 0.5296, + "step": 11106 + }, + { + "epoch": 0.620554795094561, + "grad_norm": 0.5755539536476135, + "learning_rate": 0.0006920663379650382, + "loss": 0.4567, + "step": 11107 + }, + { + "epoch": 0.6206106656982429, + "grad_norm": 0.8404194116592407, + "learning_rate": 0.0006920383236216943, + "loss": 0.5189, + "step": 11108 + }, + { + "epoch": 0.6206665363019247, + "grad_norm": 0.8287814259529114, + "learning_rate": 0.0006920103092783505, + "loss": 0.4232, + "step": 11109 + }, + { + "epoch": 0.6207224069056066, + "grad_norm": 4.02254581451416, + "learning_rate": 0.0006919822949350067, + "loss": 0.4023, + "step": 11110 + }, + { + "epoch": 0.6207782775092885, + "grad_norm": 0.6081405282020569, + "learning_rate": 0.000691954280591663, + "loss": 0.4559, + "step": 11111 + }, + { + "epoch": 0.6208341481129703, + "grad_norm": 0.4459034502506256, + "learning_rate": 0.0006919262662483191, + "loss": 0.4201, + "step": 11112 + }, + { + "epoch": 0.6208900187166523, + "grad_norm": 0.858077347278595, + "learning_rate": 0.0006918982519049753, + "loss": 0.574, + "step": 11113 + }, + { + "epoch": 0.6209458893203341, + "grad_norm": 0.5370938777923584, + "learning_rate": 0.0006918702375616316, + "loss": 0.4504, + "step": 11114 + }, + { + "epoch": 0.621001759924016, + "grad_norm": 0.523332417011261, + "learning_rate": 0.0006918422232182878, + "loss": 0.5274, + "step": 11115 + }, + { + "epoch": 0.6210576305276978, + "grad_norm": 0.6763775944709778, + "learning_rate": 0.0006918142088749439, + "loss": 0.4144, + "step": 11116 + }, + { + "epoch": 0.6211135011313798, + "grad_norm": 0.544579803943634, + "learning_rate": 0.0006917861945316001, + "loss": 0.4295, + "step": 11117 + }, + { + "epoch": 0.6211693717350616, + "grad_norm": 0.5962921977043152, + "learning_rate": 0.0006917581801882565, + "loss": 0.4319, + "step": 11118 + }, + { + "epoch": 0.6212252423387434, + "grad_norm": 0.5330449342727661, + "learning_rate": 0.0006917301658449127, + "loss": 0.4685, + "step": 11119 + }, + { + "epoch": 0.6212811129424254, + "grad_norm": 0.6566351652145386, + "learning_rate": 0.0006917021515015688, + "loss": 0.4201, + "step": 11120 + }, + { + "epoch": 0.6213369835461072, + "grad_norm": 5.203031539916992, + "learning_rate": 0.0006916741371582251, + "loss": 0.5, + "step": 11121 + }, + { + "epoch": 0.6213928541497891, + "grad_norm": 0.4839022159576416, + "learning_rate": 0.0006916461228148813, + "loss": 0.4391, + "step": 11122 + }, + { + "epoch": 0.621448724753471, + "grad_norm": 0.5662132501602173, + "learning_rate": 0.0006916181084715374, + "loss": 0.4729, + "step": 11123 + }, + { + "epoch": 0.6215045953571529, + "grad_norm": 0.5771980881690979, + "learning_rate": 0.0006915900941281936, + "loss": 0.5076, + "step": 11124 + }, + { + "epoch": 0.6215604659608347, + "grad_norm": 0.7676430940628052, + "learning_rate": 0.0006915620797848499, + "loss": 0.4901, + "step": 11125 + }, + { + "epoch": 0.6216163365645165, + "grad_norm": 0.6716554760932922, + "learning_rate": 0.0006915340654415061, + "loss": 0.4322, + "step": 11126 + }, + { + "epoch": 0.6216722071681985, + "grad_norm": 3.386824369430542, + "learning_rate": 0.0006915060510981622, + "loss": 0.4769, + "step": 11127 + }, + { + "epoch": 0.6217280777718803, + "grad_norm": 0.5425902605056763, + "learning_rate": 0.0006914780367548184, + "loss": 0.4999, + "step": 11128 + }, + { + "epoch": 0.6217839483755622, + "grad_norm": 0.7169960737228394, + "learning_rate": 0.0006914500224114747, + "loss": 0.5643, + "step": 11129 + }, + { + "epoch": 0.6218398189792441, + "grad_norm": 1.0737022161483765, + "learning_rate": 0.0006914220080681309, + "loss": 0.5866, + "step": 11130 + }, + { + "epoch": 0.621895689582926, + "grad_norm": 0.5591928958892822, + "learning_rate": 0.000691393993724787, + "loss": 0.3925, + "step": 11131 + }, + { + "epoch": 0.6219515601866078, + "grad_norm": 0.6984981298446655, + "learning_rate": 0.0006913659793814433, + "loss": 0.4908, + "step": 11132 + }, + { + "epoch": 0.6220074307902896, + "grad_norm": 0.7122300863265991, + "learning_rate": 0.0006913379650380995, + "loss": 0.4367, + "step": 11133 + }, + { + "epoch": 0.6220633013939716, + "grad_norm": 0.460123211145401, + "learning_rate": 0.0006913099506947558, + "loss": 0.5091, + "step": 11134 + }, + { + "epoch": 0.6221191719976534, + "grad_norm": 0.5631390810012817, + "learning_rate": 0.0006912819363514118, + "loss": 0.4353, + "step": 11135 + }, + { + "epoch": 0.6221750426013353, + "grad_norm": 1.4653749465942383, + "learning_rate": 0.0006912539220080682, + "loss": 0.5916, + "step": 11136 + }, + { + "epoch": 0.6222309132050172, + "grad_norm": 0.6223687529563904, + "learning_rate": 0.0006912259076647244, + "loss": 0.3823, + "step": 11137 + }, + { + "epoch": 0.6222867838086991, + "grad_norm": 0.5337184071540833, + "learning_rate": 0.0006911978933213806, + "loss": 0.5697, + "step": 11138 + }, + { + "epoch": 0.6223426544123809, + "grad_norm": 0.5040651559829712, + "learning_rate": 0.0006911698789780368, + "loss": 0.4929, + "step": 11139 + }, + { + "epoch": 0.6223985250160629, + "grad_norm": 2.0280349254608154, + "learning_rate": 0.000691141864634693, + "loss": 0.7051, + "step": 11140 + }, + { + "epoch": 0.6224543956197447, + "grad_norm": 0.5087584853172302, + "learning_rate": 0.0006911138502913492, + "loss": 0.4604, + "step": 11141 + }, + { + "epoch": 0.6225102662234265, + "grad_norm": 0.7333350777626038, + "learning_rate": 0.0006910858359480054, + "loss": 0.5004, + "step": 11142 + }, + { + "epoch": 0.6225661368271084, + "grad_norm": 24.370113372802734, + "learning_rate": 0.0006910578216046616, + "loss": 0.4621, + "step": 11143 + }, + { + "epoch": 0.6226220074307903, + "grad_norm": 0.46089476346969604, + "learning_rate": 0.0006910298072613178, + "loss": 0.4007, + "step": 11144 + }, + { + "epoch": 0.6226778780344722, + "grad_norm": 0.4180925488471985, + "learning_rate": 0.000691001792917974, + "loss": 0.4778, + "step": 11145 + }, + { + "epoch": 0.622733748638154, + "grad_norm": 0.44453659653663635, + "learning_rate": 0.0006909737785746302, + "loss": 0.3983, + "step": 11146 + }, + { + "epoch": 0.622789619241836, + "grad_norm": 1.3886038064956665, + "learning_rate": 0.0006909457642312864, + "loss": 0.523, + "step": 11147 + }, + { + "epoch": 0.6228454898455178, + "grad_norm": 1.6370371580123901, + "learning_rate": 0.0006909177498879426, + "loss": 0.4041, + "step": 11148 + }, + { + "epoch": 0.6229013604491996, + "grad_norm": 0.6705982685089111, + "learning_rate": 0.0006908897355445988, + "loss": 0.3707, + "step": 11149 + }, + { + "epoch": 0.6229572310528815, + "grad_norm": 1.1650927066802979, + "learning_rate": 0.000690861721201255, + "loss": 0.4301, + "step": 11150 + }, + { + "epoch": 0.6230131016565634, + "grad_norm": 0.7787575721740723, + "learning_rate": 0.0006908337068579113, + "loss": 0.5153, + "step": 11151 + }, + { + "epoch": 0.6230689722602453, + "grad_norm": 0.6707040667533875, + "learning_rate": 0.0006908056925145675, + "loss": 0.4985, + "step": 11152 + }, + { + "epoch": 0.6231248428639271, + "grad_norm": 3.1200602054595947, + "learning_rate": 0.0006907776781712238, + "loss": 0.4826, + "step": 11153 + }, + { + "epoch": 0.6231807134676091, + "grad_norm": 0.6712022423744202, + "learning_rate": 0.0006907496638278799, + "loss": 0.6259, + "step": 11154 + }, + { + "epoch": 0.6232365840712909, + "grad_norm": 1.9263728857040405, + "learning_rate": 0.0006907216494845361, + "loss": 0.4665, + "step": 11155 + }, + { + "epoch": 0.6232924546749727, + "grad_norm": 0.46651774644851685, + "learning_rate": 0.0006906936351411923, + "loss": 0.4606, + "step": 11156 + }, + { + "epoch": 0.6233483252786547, + "grad_norm": 0.3989298939704895, + "learning_rate": 0.0006906656207978486, + "loss": 0.5245, + "step": 11157 + }, + { + "epoch": 0.6234041958823365, + "grad_norm": 0.44109058380126953, + "learning_rate": 0.0006906376064545047, + "loss": 0.4122, + "step": 11158 + }, + { + "epoch": 0.6234600664860184, + "grad_norm": 0.49251511693000793, + "learning_rate": 0.0006906095921111609, + "loss": 0.5599, + "step": 11159 + }, + { + "epoch": 0.6235159370897002, + "grad_norm": 0.753513514995575, + "learning_rate": 0.0006905815777678172, + "loss": 0.4539, + "step": 11160 + }, + { + "epoch": 0.6235718076933822, + "grad_norm": 0.38246509432792664, + "learning_rate": 0.0006905535634244734, + "loss": 0.4756, + "step": 11161 + }, + { + "epoch": 0.623627678297064, + "grad_norm": 0.6388275027275085, + "learning_rate": 0.0006905255490811295, + "loss": 0.6296, + "step": 11162 + }, + { + "epoch": 0.6236835489007458, + "grad_norm": 1.3513450622558594, + "learning_rate": 0.0006904975347377857, + "loss": 0.6395, + "step": 11163 + }, + { + "epoch": 0.6237394195044278, + "grad_norm": 0.8291516900062561, + "learning_rate": 0.000690469520394442, + "loss": 0.4874, + "step": 11164 + }, + { + "epoch": 0.6237952901081096, + "grad_norm": 0.7319660782814026, + "learning_rate": 0.0006904415060510981, + "loss": 0.5834, + "step": 11165 + }, + { + "epoch": 0.6238511607117915, + "grad_norm": 0.5216976404190063, + "learning_rate": 0.0006904134917077543, + "loss": 0.511, + "step": 11166 + }, + { + "epoch": 0.6239070313154733, + "grad_norm": 0.4558706283569336, + "learning_rate": 0.0006903854773644105, + "loss": 0.4645, + "step": 11167 + }, + { + "epoch": 0.6239629019191553, + "grad_norm": 0.6305792331695557, + "learning_rate": 0.0006903574630210669, + "loss": 0.5847, + "step": 11168 + }, + { + "epoch": 0.6240187725228371, + "grad_norm": 0.4531366527080536, + "learning_rate": 0.000690329448677723, + "loss": 0.4501, + "step": 11169 + }, + { + "epoch": 0.624074643126519, + "grad_norm": 0.5270293354988098, + "learning_rate": 0.0006903014343343792, + "loss": 0.4375, + "step": 11170 + }, + { + "epoch": 0.6241305137302009, + "grad_norm": 0.5696723461151123, + "learning_rate": 0.0006902734199910355, + "loss": 0.441, + "step": 11171 + }, + { + "epoch": 0.6241863843338827, + "grad_norm": 0.6201713681221008, + "learning_rate": 0.0006902454056476917, + "loss": 0.4905, + "step": 11172 + }, + { + "epoch": 0.6242422549375646, + "grad_norm": 0.3825182020664215, + "learning_rate": 0.0006902173913043478, + "loss": 0.4474, + "step": 11173 + }, + { + "epoch": 0.6242981255412464, + "grad_norm": 0.8311170935630798, + "learning_rate": 0.000690189376961004, + "loss": 0.3581, + "step": 11174 + }, + { + "epoch": 0.6243539961449284, + "grad_norm": 0.661284327507019, + "learning_rate": 0.0006901613626176603, + "loss": 0.4595, + "step": 11175 + }, + { + "epoch": 0.6244098667486102, + "grad_norm": 0.5374329686164856, + "learning_rate": 0.0006901333482743165, + "loss": 0.404, + "step": 11176 + }, + { + "epoch": 0.624465737352292, + "grad_norm": 1.0885533094406128, + "learning_rate": 0.0006901053339309726, + "loss": 0.4525, + "step": 11177 + }, + { + "epoch": 0.624521607955974, + "grad_norm": 0.4526996314525604, + "learning_rate": 0.0006900773195876289, + "loss": 0.3738, + "step": 11178 + }, + { + "epoch": 0.6245774785596558, + "grad_norm": 0.6197134256362915, + "learning_rate": 0.0006900493052442851, + "loss": 0.508, + "step": 11179 + }, + { + "epoch": 0.6246333491633377, + "grad_norm": 0.5670042037963867, + "learning_rate": 0.0006900212909009413, + "loss": 0.4288, + "step": 11180 + }, + { + "epoch": 0.6246892197670196, + "grad_norm": 0.5839502215385437, + "learning_rate": 0.0006899932765575974, + "loss": 0.568, + "step": 11181 + }, + { + "epoch": 0.6247450903707015, + "grad_norm": 0.46006760001182556, + "learning_rate": 0.0006899652622142537, + "loss": 0.457, + "step": 11182 + }, + { + "epoch": 0.6248009609743833, + "grad_norm": 0.7435041069984436, + "learning_rate": 0.0006899372478709099, + "loss": 0.495, + "step": 11183 + }, + { + "epoch": 0.6248568315780652, + "grad_norm": 0.6444823145866394, + "learning_rate": 0.0006899092335275662, + "loss": 0.4178, + "step": 11184 + }, + { + "epoch": 0.6249127021817471, + "grad_norm": 0.35974007844924927, + "learning_rate": 0.0006898812191842224, + "loss": 0.4256, + "step": 11185 + }, + { + "epoch": 0.6249685727854289, + "grad_norm": 1.1975343227386475, + "learning_rate": 0.0006898532048408786, + "loss": 0.4675, + "step": 11186 + }, + { + "epoch": 0.6250244433891108, + "grad_norm": 0.40562328696250916, + "learning_rate": 0.0006898251904975348, + "loss": 0.3714, + "step": 11187 + }, + { + "epoch": 0.6250803139927927, + "grad_norm": 0.5043787360191345, + "learning_rate": 0.0006897971761541909, + "loss": 0.5044, + "step": 11188 + }, + { + "epoch": 0.6251361845964746, + "grad_norm": 1.679274082183838, + "learning_rate": 0.0006897691618108472, + "loss": 0.4693, + "step": 11189 + }, + { + "epoch": 0.6251920552001564, + "grad_norm": 1.1691844463348389, + "learning_rate": 0.0006897411474675034, + "loss": 0.4404, + "step": 11190 + }, + { + "epoch": 0.6252479258038383, + "grad_norm": 0.37647613883018494, + "learning_rate": 0.0006897131331241596, + "loss": 0.4835, + "step": 11191 + }, + { + "epoch": 0.6253037964075202, + "grad_norm": 0.5773559212684631, + "learning_rate": 0.0006896851187808158, + "loss": 0.4527, + "step": 11192 + }, + { + "epoch": 0.625359667011202, + "grad_norm": 0.3882405757904053, + "learning_rate": 0.000689657104437472, + "loss": 0.3961, + "step": 11193 + }, + { + "epoch": 0.6254155376148839, + "grad_norm": 0.5241730809211731, + "learning_rate": 0.0006896290900941282, + "loss": 0.4775, + "step": 11194 + }, + { + "epoch": 0.6254714082185658, + "grad_norm": 0.44323790073394775, + "learning_rate": 0.0006896010757507844, + "loss": 0.3735, + "step": 11195 + }, + { + "epoch": 0.6255272788222477, + "grad_norm": 2.642230272293091, + "learning_rate": 0.0006895730614074406, + "loss": 0.385, + "step": 11196 + }, + { + "epoch": 0.6255831494259295, + "grad_norm": 0.8481842875480652, + "learning_rate": 0.0006895450470640968, + "loss": 0.5321, + "step": 11197 + }, + { + "epoch": 0.6256390200296115, + "grad_norm": 0.44455012679100037, + "learning_rate": 0.000689517032720753, + "loss": 0.424, + "step": 11198 + }, + { + "epoch": 0.6256948906332933, + "grad_norm": 0.8271108269691467, + "learning_rate": 0.0006894890183774093, + "loss": 0.4888, + "step": 11199 + }, + { + "epoch": 0.6257507612369752, + "grad_norm": 0.7884595394134521, + "learning_rate": 0.0006894610040340654, + "loss": 0.5049, + "step": 11200 + }, + { + "epoch": 0.625806631840657, + "grad_norm": 0.8092600107192993, + "learning_rate": 0.0006894329896907217, + "loss": 0.4962, + "step": 11201 + }, + { + "epoch": 0.6258625024443389, + "grad_norm": 0.6914955377578735, + "learning_rate": 0.0006894049753473779, + "loss": 0.5654, + "step": 11202 + }, + { + "epoch": 0.6259183730480208, + "grad_norm": 0.5157763361930847, + "learning_rate": 0.0006893769610040342, + "loss": 0.4513, + "step": 11203 + }, + { + "epoch": 0.6259742436517026, + "grad_norm": 0.43041035532951355, + "learning_rate": 0.0006893489466606903, + "loss": 0.3011, + "step": 11204 + }, + { + "epoch": 0.6260301142553846, + "grad_norm": 0.5464975237846375, + "learning_rate": 0.0006893209323173465, + "loss": 0.468, + "step": 11205 + }, + { + "epoch": 0.6260859848590664, + "grad_norm": 0.4933786392211914, + "learning_rate": 0.0006892929179740027, + "loss": 0.4085, + "step": 11206 + }, + { + "epoch": 0.6261418554627483, + "grad_norm": 0.5566145777702332, + "learning_rate": 0.0006892649036306589, + "loss": 0.6454, + "step": 11207 + }, + { + "epoch": 0.6261977260664301, + "grad_norm": 0.4396577477455139, + "learning_rate": 0.0006892368892873151, + "loss": 0.4407, + "step": 11208 + }, + { + "epoch": 0.626253596670112, + "grad_norm": 0.4702457785606384, + "learning_rate": 0.0006892088749439713, + "loss": 0.4901, + "step": 11209 + }, + { + "epoch": 0.6263094672737939, + "grad_norm": 0.6385652422904968, + "learning_rate": 0.0006891808606006276, + "loss": 0.3813, + "step": 11210 + }, + { + "epoch": 0.6263653378774757, + "grad_norm": 0.49488747119903564, + "learning_rate": 0.0006891528462572837, + "loss": 0.4442, + "step": 11211 + }, + { + "epoch": 0.6264212084811577, + "grad_norm": 2.8739495277404785, + "learning_rate": 0.0006891248319139399, + "loss": 0.6555, + "step": 11212 + }, + { + "epoch": 0.6264770790848395, + "grad_norm": 0.5680075883865356, + "learning_rate": 0.0006890968175705961, + "loss": 0.6314, + "step": 11213 + }, + { + "epoch": 0.6265329496885214, + "grad_norm": 0.5800068378448486, + "learning_rate": 0.0006890688032272524, + "loss": 0.4218, + "step": 11214 + }, + { + "epoch": 0.6265888202922033, + "grad_norm": 0.6050625443458557, + "learning_rate": 0.0006890407888839085, + "loss": 0.6125, + "step": 11215 + }, + { + "epoch": 0.6266446908958851, + "grad_norm": 0.6976779103279114, + "learning_rate": 0.0006890127745405647, + "loss": 0.462, + "step": 11216 + }, + { + "epoch": 0.626700561499567, + "grad_norm": 1.5251508951187134, + "learning_rate": 0.000688984760197221, + "loss": 0.4206, + "step": 11217 + }, + { + "epoch": 0.6267564321032488, + "grad_norm": 3.378981351852417, + "learning_rate": 0.0006889567458538773, + "loss": 0.4895, + "step": 11218 + }, + { + "epoch": 0.6268123027069308, + "grad_norm": 0.7220731973648071, + "learning_rate": 0.0006889287315105334, + "loss": 0.4634, + "step": 11219 + }, + { + "epoch": 0.6268681733106126, + "grad_norm": 0.5666050910949707, + "learning_rate": 0.0006889007171671896, + "loss": 0.5359, + "step": 11220 + }, + { + "epoch": 0.6269240439142945, + "grad_norm": 0.43356752395629883, + "learning_rate": 0.0006888727028238459, + "loss": 0.5064, + "step": 11221 + }, + { + "epoch": 0.6269799145179764, + "grad_norm": 2.8219189643859863, + "learning_rate": 0.0006888446884805021, + "loss": 0.4363, + "step": 11222 + }, + { + "epoch": 0.6270357851216583, + "grad_norm": 0.37588122487068176, + "learning_rate": 0.0006888166741371582, + "loss": 0.3028, + "step": 11223 + }, + { + "epoch": 0.6270916557253401, + "grad_norm": 0.7014046311378479, + "learning_rate": 0.0006887886597938145, + "loss": 0.4248, + "step": 11224 + }, + { + "epoch": 0.6271475263290219, + "grad_norm": 0.8695746660232544, + "learning_rate": 0.0006887606454504707, + "loss": 0.6964, + "step": 11225 + }, + { + "epoch": 0.6272033969327039, + "grad_norm": 0.8878431916236877, + "learning_rate": 0.0006887326311071269, + "loss": 0.4735, + "step": 11226 + }, + { + "epoch": 0.6272592675363857, + "grad_norm": 0.41875511407852173, + "learning_rate": 0.000688704616763783, + "loss": 0.4509, + "step": 11227 + }, + { + "epoch": 0.6273151381400676, + "grad_norm": 2.0032553672790527, + "learning_rate": 0.0006886766024204393, + "loss": 0.5502, + "step": 11228 + }, + { + "epoch": 0.6273710087437495, + "grad_norm": 0.6498506665229797, + "learning_rate": 0.0006886485880770955, + "loss": 0.5174, + "step": 11229 + }, + { + "epoch": 0.6274268793474314, + "grad_norm": 0.5153188109397888, + "learning_rate": 0.0006886205737337516, + "loss": 0.3475, + "step": 11230 + }, + { + "epoch": 0.6274827499511132, + "grad_norm": 0.5849705934524536, + "learning_rate": 0.0006885925593904079, + "loss": 0.5821, + "step": 11231 + }, + { + "epoch": 0.6275386205547951, + "grad_norm": 0.4256785809993744, + "learning_rate": 0.0006885645450470641, + "loss": 0.452, + "step": 11232 + }, + { + "epoch": 0.627594491158477, + "grad_norm": 5.78323221206665, + "learning_rate": 0.0006885365307037203, + "loss": 0.4167, + "step": 11233 + }, + { + "epoch": 0.6276503617621588, + "grad_norm": 1.5527552366256714, + "learning_rate": 0.0006885085163603764, + "loss": 0.4478, + "step": 11234 + }, + { + "epoch": 0.6277062323658407, + "grad_norm": 0.9094555377960205, + "learning_rate": 0.0006884805020170328, + "loss": 0.4021, + "step": 11235 + }, + { + "epoch": 0.6277621029695226, + "grad_norm": 0.47777706384658813, + "learning_rate": 0.000688452487673689, + "loss": 0.3876, + "step": 11236 + }, + { + "epoch": 0.6278179735732045, + "grad_norm": 1.5650432109832764, + "learning_rate": 0.0006884244733303452, + "loss": 0.4054, + "step": 11237 + }, + { + "epoch": 0.6278738441768863, + "grad_norm": 0.46649202704429626, + "learning_rate": 0.0006883964589870013, + "loss": 0.3898, + "step": 11238 + }, + { + "epoch": 0.6279297147805682, + "grad_norm": 0.5785773992538452, + "learning_rate": 0.0006883684446436576, + "loss": 0.4848, + "step": 11239 + }, + { + "epoch": 0.6279855853842501, + "grad_norm": 1.6625165939331055, + "learning_rate": 0.0006883404303003138, + "loss": 0.458, + "step": 11240 + }, + { + "epoch": 0.6280414559879319, + "grad_norm": 4.811558723449707, + "learning_rate": 0.00068831241595697, + "loss": 0.4291, + "step": 11241 + }, + { + "epoch": 0.6280973265916138, + "grad_norm": 0.5498399138450623, + "learning_rate": 0.0006882844016136262, + "loss": 0.4091, + "step": 11242 + }, + { + "epoch": 0.6281531971952957, + "grad_norm": 0.5162630081176758, + "learning_rate": 0.0006882563872702824, + "loss": 0.3867, + "step": 11243 + }, + { + "epoch": 0.6282090677989776, + "grad_norm": 0.5528560280799866, + "learning_rate": 0.0006882283729269386, + "loss": 0.4629, + "step": 11244 + }, + { + "epoch": 0.6282649384026594, + "grad_norm": 0.3657792806625366, + "learning_rate": 0.0006882003585835948, + "loss": 0.4043, + "step": 11245 + }, + { + "epoch": 0.6283208090063414, + "grad_norm": 0.7096413969993591, + "learning_rate": 0.000688172344240251, + "loss": 0.4411, + "step": 11246 + }, + { + "epoch": 0.6283766796100232, + "grad_norm": 0.462246298789978, + "learning_rate": 0.0006881443298969072, + "loss": 0.4423, + "step": 11247 + }, + { + "epoch": 0.628432550213705, + "grad_norm": 0.6578935980796814, + "learning_rate": 0.0006881163155535634, + "loss": 0.6077, + "step": 11248 + }, + { + "epoch": 0.628488420817387, + "grad_norm": 0.5089840888977051, + "learning_rate": 0.0006880883012102196, + "loss": 0.4453, + "step": 11249 + }, + { + "epoch": 0.6285442914210688, + "grad_norm": 0.4601778984069824, + "learning_rate": 0.0006880602868668758, + "loss": 0.5494, + "step": 11250 + }, + { + "epoch": 0.6286001620247507, + "grad_norm": 0.4280807077884674, + "learning_rate": 0.000688032272523532, + "loss": 0.3703, + "step": 11251 + }, + { + "epoch": 0.6286560326284325, + "grad_norm": 0.9105669856071472, + "learning_rate": 0.0006880042581801883, + "loss": 0.6544, + "step": 11252 + }, + { + "epoch": 0.6287119032321145, + "grad_norm": 0.6301037073135376, + "learning_rate": 0.0006879762438368445, + "loss": 0.6037, + "step": 11253 + }, + { + "epoch": 0.6287677738357963, + "grad_norm": 0.4217209815979004, + "learning_rate": 0.0006879482294935007, + "loss": 0.4282, + "step": 11254 + }, + { + "epoch": 0.6288236444394781, + "grad_norm": 1.034014105796814, + "learning_rate": 0.0006879202151501569, + "loss": 0.385, + "step": 11255 + }, + { + "epoch": 0.6288795150431601, + "grad_norm": 1.4069578647613525, + "learning_rate": 0.0006878922008068132, + "loss": 0.3937, + "step": 11256 + }, + { + "epoch": 0.6289353856468419, + "grad_norm": 0.3930588662624359, + "learning_rate": 0.0006878641864634693, + "loss": 0.3772, + "step": 11257 + }, + { + "epoch": 0.6289912562505238, + "grad_norm": 0.46060752868652344, + "learning_rate": 0.0006878361721201255, + "loss": 0.4689, + "step": 11258 + }, + { + "epoch": 0.6290471268542056, + "grad_norm": 0.6815087199211121, + "learning_rate": 0.0006878081577767817, + "loss": 0.5011, + "step": 11259 + }, + { + "epoch": 0.6291029974578876, + "grad_norm": 2.521610975265503, + "learning_rate": 0.000687780143433438, + "loss": 0.4652, + "step": 11260 + }, + { + "epoch": 0.6291588680615694, + "grad_norm": 0.5843362212181091, + "learning_rate": 0.0006877521290900941, + "loss": 0.4305, + "step": 11261 + }, + { + "epoch": 0.6292147386652512, + "grad_norm": 0.4424970746040344, + "learning_rate": 0.0006877241147467503, + "loss": 0.495, + "step": 11262 + }, + { + "epoch": 0.6292706092689332, + "grad_norm": 0.40144726634025574, + "learning_rate": 0.0006876961004034066, + "loss": 0.3647, + "step": 11263 + }, + { + "epoch": 0.629326479872615, + "grad_norm": 0.5347927808761597, + "learning_rate": 0.0006876680860600628, + "loss": 0.4886, + "step": 11264 + }, + { + "epoch": 0.6293823504762969, + "grad_norm": 0.8418329358100891, + "learning_rate": 0.0006876400717167189, + "loss": 0.5298, + "step": 11265 + }, + { + "epoch": 0.6294382210799788, + "grad_norm": 0.4396285116672516, + "learning_rate": 0.0006876120573733751, + "loss": 0.5377, + "step": 11266 + }, + { + "epoch": 0.6294940916836607, + "grad_norm": 0.4389066696166992, + "learning_rate": 0.0006875840430300315, + "loss": 0.3391, + "step": 11267 + }, + { + "epoch": 0.6295499622873425, + "grad_norm": 2.383962869644165, + "learning_rate": 0.0006875560286866877, + "loss": 0.3966, + "step": 11268 + }, + { + "epoch": 0.6296058328910243, + "grad_norm": 0.6426882743835449, + "learning_rate": 0.0006875280143433438, + "loss": 0.4546, + "step": 11269 + }, + { + "epoch": 0.6296617034947063, + "grad_norm": 0.9331918358802795, + "learning_rate": 0.0006875, + "loss": 0.4454, + "step": 11270 + }, + { + "epoch": 0.6297175740983881, + "grad_norm": 0.8867982625961304, + "learning_rate": 0.0006874719856566563, + "loss": 0.4046, + "step": 11271 + }, + { + "epoch": 0.62977344470207, + "grad_norm": 0.534116268157959, + "learning_rate": 0.0006874439713133124, + "loss": 0.4723, + "step": 11272 + }, + { + "epoch": 0.6298293153057519, + "grad_norm": 0.550615131855011, + "learning_rate": 0.0006874159569699686, + "loss": 0.4939, + "step": 11273 + }, + { + "epoch": 0.6298851859094338, + "grad_norm": 0.42769092321395874, + "learning_rate": 0.0006873879426266249, + "loss": 0.4955, + "step": 11274 + }, + { + "epoch": 0.6299410565131156, + "grad_norm": 2.2492735385894775, + "learning_rate": 0.0006873599282832811, + "loss": 0.4091, + "step": 11275 + }, + { + "epoch": 0.6299969271167974, + "grad_norm": 0.7107661366462708, + "learning_rate": 0.0006873319139399372, + "loss": 0.4255, + "step": 11276 + }, + { + "epoch": 0.6300527977204794, + "grad_norm": 0.6113039255142212, + "learning_rate": 0.0006873038995965934, + "loss": 0.4137, + "step": 11277 + }, + { + "epoch": 0.6301086683241612, + "grad_norm": 0.669360339641571, + "learning_rate": 0.0006872758852532497, + "loss": 0.6103, + "step": 11278 + }, + { + "epoch": 0.6301645389278431, + "grad_norm": 0.7044636011123657, + "learning_rate": 0.0006872478709099059, + "loss": 0.4747, + "step": 11279 + }, + { + "epoch": 0.630220409531525, + "grad_norm": 0.4784722626209259, + "learning_rate": 0.000687219856566562, + "loss": 0.4908, + "step": 11280 + }, + { + "epoch": 0.6302762801352069, + "grad_norm": 0.4685162901878357, + "learning_rate": 0.0006871918422232183, + "loss": 0.45, + "step": 11281 + }, + { + "epoch": 0.6303321507388887, + "grad_norm": 0.575983464717865, + "learning_rate": 0.0006871638278798745, + "loss": 0.3294, + "step": 11282 + }, + { + "epoch": 0.6303880213425707, + "grad_norm": 0.8314383625984192, + "learning_rate": 0.0006871358135365307, + "loss": 0.5319, + "step": 11283 + }, + { + "epoch": 0.6304438919462525, + "grad_norm": 0.6484651565551758, + "learning_rate": 0.0006871077991931868, + "loss": 0.4294, + "step": 11284 + }, + { + "epoch": 0.6304997625499343, + "grad_norm": 0.6766480803489685, + "learning_rate": 0.0006870797848498432, + "loss": 0.3841, + "step": 11285 + }, + { + "epoch": 0.6305556331536162, + "grad_norm": 0.6213365197181702, + "learning_rate": 0.0006870517705064994, + "loss": 0.4577, + "step": 11286 + }, + { + "epoch": 0.6306115037572981, + "grad_norm": 0.6235005259513855, + "learning_rate": 0.0006870237561631556, + "loss": 0.4732, + "step": 11287 + }, + { + "epoch": 0.63066737436098, + "grad_norm": 0.42419198155403137, + "learning_rate": 0.0006869957418198118, + "loss": 0.4387, + "step": 11288 + }, + { + "epoch": 0.6307232449646618, + "grad_norm": 0.6855747699737549, + "learning_rate": 0.000686967727476468, + "loss": 0.3863, + "step": 11289 + }, + { + "epoch": 0.6307791155683438, + "grad_norm": 0.4105197787284851, + "learning_rate": 0.0006869397131331242, + "loss": 0.3746, + "step": 11290 + }, + { + "epoch": 0.6308349861720256, + "grad_norm": 0.5211912989616394, + "learning_rate": 0.0006869116987897804, + "loss": 0.4714, + "step": 11291 + }, + { + "epoch": 0.6308908567757074, + "grad_norm": 0.5455973744392395, + "learning_rate": 0.0006868836844464366, + "loss": 0.4719, + "step": 11292 + }, + { + "epoch": 0.6309467273793893, + "grad_norm": 1.2712677717208862, + "learning_rate": 0.0006868556701030928, + "loss": 0.3709, + "step": 11293 + }, + { + "epoch": 0.6310025979830712, + "grad_norm": 0.43031010031700134, + "learning_rate": 0.000686827655759749, + "loss": 0.4508, + "step": 11294 + }, + { + "epoch": 0.6310584685867531, + "grad_norm": 0.388359397649765, + "learning_rate": 0.0006867996414164052, + "loss": 0.4642, + "step": 11295 + }, + { + "epoch": 0.6311143391904349, + "grad_norm": 0.4975282549858093, + "learning_rate": 0.0006867716270730614, + "loss": 0.5013, + "step": 11296 + }, + { + "epoch": 0.6311702097941169, + "grad_norm": 0.5266387462615967, + "learning_rate": 0.0006867436127297176, + "loss": 0.4121, + "step": 11297 + }, + { + "epoch": 0.6312260803977987, + "grad_norm": 0.6671774983406067, + "learning_rate": 0.0006867155983863738, + "loss": 0.4026, + "step": 11298 + }, + { + "epoch": 0.6312819510014805, + "grad_norm": 0.7799452543258667, + "learning_rate": 0.00068668758404303, + "loss": 0.3706, + "step": 11299 + }, + { + "epoch": 0.6313378216051625, + "grad_norm": 0.5848537087440491, + "learning_rate": 0.0006866595696996862, + "loss": 0.3509, + "step": 11300 + }, + { + "epoch": 0.6313936922088443, + "grad_norm": 0.9516686201095581, + "learning_rate": 0.0006866315553563424, + "loss": 0.3797, + "step": 11301 + }, + { + "epoch": 0.6314495628125262, + "grad_norm": 0.43520450592041016, + "learning_rate": 0.0006866035410129988, + "loss": 0.4648, + "step": 11302 + }, + { + "epoch": 0.631505433416208, + "grad_norm": 0.7157742381095886, + "learning_rate": 0.0006865755266696549, + "loss": 0.5816, + "step": 11303 + }, + { + "epoch": 0.63156130401989, + "grad_norm": 0.4683733284473419, + "learning_rate": 0.0006865475123263111, + "loss": 0.522, + "step": 11304 + }, + { + "epoch": 0.6316171746235718, + "grad_norm": 0.5543689131736755, + "learning_rate": 0.0006865194979829673, + "loss": 0.3615, + "step": 11305 + }, + { + "epoch": 0.6316730452272536, + "grad_norm": 0.4939996600151062, + "learning_rate": 0.0006864914836396236, + "loss": 0.5208, + "step": 11306 + }, + { + "epoch": 0.6317289158309356, + "grad_norm": 0.6242522597312927, + "learning_rate": 0.0006864634692962797, + "loss": 0.3918, + "step": 11307 + }, + { + "epoch": 0.6317847864346174, + "grad_norm": 0.9270811080932617, + "learning_rate": 0.0006864354549529359, + "loss": 0.4167, + "step": 11308 + }, + { + "epoch": 0.6318406570382993, + "grad_norm": 0.36534780263900757, + "learning_rate": 0.0006864074406095921, + "loss": 0.4153, + "step": 11309 + }, + { + "epoch": 0.6318965276419811, + "grad_norm": 0.4874607026576996, + "learning_rate": 0.0006863794262662484, + "loss": 0.3695, + "step": 11310 + }, + { + "epoch": 0.6319523982456631, + "grad_norm": 0.9038450717926025, + "learning_rate": 0.0006863514119229045, + "loss": 0.4801, + "step": 11311 + }, + { + "epoch": 0.6320082688493449, + "grad_norm": 0.48685523867607117, + "learning_rate": 0.0006863233975795607, + "loss": 0.4051, + "step": 11312 + }, + { + "epoch": 0.6320641394530268, + "grad_norm": 0.7759826183319092, + "learning_rate": 0.000686295383236217, + "loss": 0.3502, + "step": 11313 + }, + { + "epoch": 0.6321200100567087, + "grad_norm": 0.7487497329711914, + "learning_rate": 0.0006862673688928731, + "loss": 0.3954, + "step": 11314 + }, + { + "epoch": 0.6321758806603905, + "grad_norm": 0.7539016008377075, + "learning_rate": 0.0006862393545495293, + "loss": 0.4287, + "step": 11315 + }, + { + "epoch": 0.6322317512640724, + "grad_norm": 0.5634896159172058, + "learning_rate": 0.0006862113402061855, + "loss": 0.546, + "step": 11316 + }, + { + "epoch": 0.6322876218677543, + "grad_norm": 0.4523473381996155, + "learning_rate": 0.0006861833258628419, + "loss": 0.4576, + "step": 11317 + }, + { + "epoch": 0.6323434924714362, + "grad_norm": 0.7494411468505859, + "learning_rate": 0.000686155311519498, + "loss": 0.4392, + "step": 11318 + }, + { + "epoch": 0.632399363075118, + "grad_norm": 1.0298991203308105, + "learning_rate": 0.0006861272971761542, + "loss": 0.3295, + "step": 11319 + }, + { + "epoch": 0.6324552336787999, + "grad_norm": 0.8398178815841675, + "learning_rate": 0.0006860992828328105, + "loss": 0.4266, + "step": 11320 + }, + { + "epoch": 0.6325111042824818, + "grad_norm": 0.7195974588394165, + "learning_rate": 0.0006860712684894667, + "loss": 0.4961, + "step": 11321 + }, + { + "epoch": 0.6325669748861636, + "grad_norm": 0.564642071723938, + "learning_rate": 0.0006860432541461228, + "loss": 0.4365, + "step": 11322 + }, + { + "epoch": 0.6326228454898455, + "grad_norm": 0.5113140344619751, + "learning_rate": 0.000686015239802779, + "loss": 0.6243, + "step": 11323 + }, + { + "epoch": 0.6326787160935274, + "grad_norm": 0.6191045641899109, + "learning_rate": 0.0006859872254594353, + "loss": 0.4959, + "step": 11324 + }, + { + "epoch": 0.6327345866972093, + "grad_norm": 0.4462256133556366, + "learning_rate": 0.0006859592111160915, + "loss": 0.4261, + "step": 11325 + }, + { + "epoch": 0.6327904573008911, + "grad_norm": 1.7149739265441895, + "learning_rate": 0.0006859311967727476, + "loss": 0.4819, + "step": 11326 + }, + { + "epoch": 0.632846327904573, + "grad_norm": 0.46564558148384094, + "learning_rate": 0.0006859031824294039, + "loss": 0.4229, + "step": 11327 + }, + { + "epoch": 0.6329021985082549, + "grad_norm": 0.5726218223571777, + "learning_rate": 0.0006858751680860601, + "loss": 0.5606, + "step": 11328 + }, + { + "epoch": 0.6329580691119367, + "grad_norm": 0.6679570078849792, + "learning_rate": 0.0006858471537427163, + "loss": 0.4245, + "step": 11329 + }, + { + "epoch": 0.6330139397156186, + "grad_norm": 0.6007128357887268, + "learning_rate": 0.0006858191393993724, + "loss": 0.3147, + "step": 11330 + }, + { + "epoch": 0.6330698103193005, + "grad_norm": 0.42979490756988525, + "learning_rate": 0.0006857911250560287, + "loss": 0.4646, + "step": 11331 + }, + { + "epoch": 0.6331256809229824, + "grad_norm": 3.117032289505005, + "learning_rate": 0.0006857631107126849, + "loss": 0.5123, + "step": 11332 + }, + { + "epoch": 0.6331815515266642, + "grad_norm": 0.5216575264930725, + "learning_rate": 0.0006857350963693411, + "loss": 0.439, + "step": 11333 + }, + { + "epoch": 0.6332374221303462, + "grad_norm": 0.5814629793167114, + "learning_rate": 0.0006857070820259973, + "loss": 0.6494, + "step": 11334 + }, + { + "epoch": 0.633293292734028, + "grad_norm": 0.39899349212646484, + "learning_rate": 0.0006856790676826536, + "loss": 0.4142, + "step": 11335 + }, + { + "epoch": 0.6333491633377099, + "grad_norm": 0.5199178457260132, + "learning_rate": 0.0006856510533393098, + "loss": 0.4677, + "step": 11336 + }, + { + "epoch": 0.6334050339413917, + "grad_norm": 0.459079384803772, + "learning_rate": 0.0006856230389959659, + "loss": 0.4783, + "step": 11337 + }, + { + "epoch": 0.6334609045450736, + "grad_norm": 0.5516748428344727, + "learning_rate": 0.0006855950246526222, + "loss": 0.4398, + "step": 11338 + }, + { + "epoch": 0.6335167751487555, + "grad_norm": 0.9427421689033508, + "learning_rate": 0.0006855670103092784, + "loss": 0.5201, + "step": 11339 + }, + { + "epoch": 0.6335726457524373, + "grad_norm": 0.8120189309120178, + "learning_rate": 0.0006855389959659346, + "loss": 0.4578, + "step": 11340 + }, + { + "epoch": 0.6336285163561193, + "grad_norm": 0.5029048919677734, + "learning_rate": 0.0006855109816225907, + "loss": 0.5289, + "step": 11341 + }, + { + "epoch": 0.6336843869598011, + "grad_norm": 1.6795085668563843, + "learning_rate": 0.000685482967279247, + "loss": 0.4709, + "step": 11342 + }, + { + "epoch": 0.633740257563483, + "grad_norm": 0.4794639050960541, + "learning_rate": 0.0006854549529359032, + "loss": 0.4015, + "step": 11343 + }, + { + "epoch": 0.6337961281671648, + "grad_norm": 0.7330037355422974, + "learning_rate": 0.0006854269385925594, + "loss": 0.6604, + "step": 11344 + }, + { + "epoch": 0.6338519987708467, + "grad_norm": 0.4604896605014801, + "learning_rate": 0.0006853989242492156, + "loss": 0.3924, + "step": 11345 + }, + { + "epoch": 0.6339078693745286, + "grad_norm": 0.5954636931419373, + "learning_rate": 0.0006853709099058718, + "loss": 0.4239, + "step": 11346 + }, + { + "epoch": 0.6339637399782104, + "grad_norm": 0.9639409780502319, + "learning_rate": 0.000685342895562528, + "loss": 0.3723, + "step": 11347 + }, + { + "epoch": 0.6340196105818924, + "grad_norm": 0.41257143020629883, + "learning_rate": 0.0006853148812191842, + "loss": 0.3805, + "step": 11348 + }, + { + "epoch": 0.6340754811855742, + "grad_norm": 0.426636278629303, + "learning_rate": 0.0006852868668758404, + "loss": 0.3927, + "step": 11349 + }, + { + "epoch": 0.6341313517892561, + "grad_norm": 0.5230627655982971, + "learning_rate": 0.0006852588525324966, + "loss": 0.3962, + "step": 11350 + }, + { + "epoch": 0.634187222392938, + "grad_norm": 0.7488715648651123, + "learning_rate": 0.0006852308381891528, + "loss": 0.4631, + "step": 11351 + }, + { + "epoch": 0.6342430929966199, + "grad_norm": 0.4802359640598297, + "learning_rate": 0.0006852028238458092, + "loss": 0.4897, + "step": 11352 + }, + { + "epoch": 0.6342989636003017, + "grad_norm": 0.5725406408309937, + "learning_rate": 0.0006851748095024653, + "loss": 0.4425, + "step": 11353 + }, + { + "epoch": 0.6343548342039835, + "grad_norm": 0.8926483988761902, + "learning_rate": 0.0006851467951591215, + "loss": 0.4596, + "step": 11354 + }, + { + "epoch": 0.6344107048076655, + "grad_norm": 0.7824434041976929, + "learning_rate": 0.0006851187808157777, + "loss": 0.3157, + "step": 11355 + }, + { + "epoch": 0.6344665754113473, + "grad_norm": 1.3196061849594116, + "learning_rate": 0.0006850907664724339, + "loss": 0.4555, + "step": 11356 + }, + { + "epoch": 0.6345224460150292, + "grad_norm": 0.37529924511909485, + "learning_rate": 0.0006850627521290901, + "loss": 0.3849, + "step": 11357 + }, + { + "epoch": 0.6345783166187111, + "grad_norm": 1.8950897455215454, + "learning_rate": 0.0006850347377857463, + "loss": 0.4398, + "step": 11358 + }, + { + "epoch": 0.634634187222393, + "grad_norm": 1.072999119758606, + "learning_rate": 0.0006850067234424026, + "loss": 0.398, + "step": 11359 + }, + { + "epoch": 0.6346900578260748, + "grad_norm": 0.39809224009513855, + "learning_rate": 0.0006849787090990587, + "loss": 0.4112, + "step": 11360 + }, + { + "epoch": 0.6347459284297566, + "grad_norm": 0.7783305048942566, + "learning_rate": 0.0006849506947557149, + "loss": 0.4179, + "step": 11361 + }, + { + "epoch": 0.6348017990334386, + "grad_norm": 0.4006766080856323, + "learning_rate": 0.0006849226804123711, + "loss": 0.4542, + "step": 11362 + }, + { + "epoch": 0.6348576696371204, + "grad_norm": 0.48224249482154846, + "learning_rate": 0.0006848946660690274, + "loss": 0.4675, + "step": 11363 + }, + { + "epoch": 0.6349135402408023, + "grad_norm": 0.39772602915763855, + "learning_rate": 0.0006848666517256835, + "loss": 0.4212, + "step": 11364 + }, + { + "epoch": 0.6349694108444842, + "grad_norm": 0.5913202166557312, + "learning_rate": 0.0006848386373823397, + "loss": 0.4668, + "step": 11365 + }, + { + "epoch": 0.6350252814481661, + "grad_norm": 0.9626784324645996, + "learning_rate": 0.000684810623038996, + "loss": 0.5415, + "step": 11366 + }, + { + "epoch": 0.6350811520518479, + "grad_norm": 3.1712942123413086, + "learning_rate": 0.0006847826086956522, + "loss": 0.4224, + "step": 11367 + }, + { + "epoch": 0.6351370226555298, + "grad_norm": 0.41800785064697266, + "learning_rate": 0.0006847545943523083, + "loss": 0.3035, + "step": 11368 + }, + { + "epoch": 0.6351928932592117, + "grad_norm": 0.7797258496284485, + "learning_rate": 0.0006847265800089646, + "loss": 0.4146, + "step": 11369 + }, + { + "epoch": 0.6352487638628935, + "grad_norm": 0.5877987146377563, + "learning_rate": 0.0006846985656656209, + "loss": 0.5549, + "step": 11370 + }, + { + "epoch": 0.6353046344665754, + "grad_norm": 1.281429648399353, + "learning_rate": 0.0006846705513222771, + "loss": 0.5682, + "step": 11371 + }, + { + "epoch": 0.6353605050702573, + "grad_norm": 1.793413758277893, + "learning_rate": 0.0006846425369789332, + "loss": 0.3339, + "step": 11372 + }, + { + "epoch": 0.6354163756739392, + "grad_norm": 4.225500583648682, + "learning_rate": 0.0006846145226355895, + "loss": 0.5681, + "step": 11373 + }, + { + "epoch": 0.635472246277621, + "grad_norm": 0.7669020295143127, + "learning_rate": 0.0006845865082922457, + "loss": 0.4142, + "step": 11374 + }, + { + "epoch": 0.635528116881303, + "grad_norm": 0.8747077584266663, + "learning_rate": 0.0006845584939489019, + "loss": 0.5944, + "step": 11375 + }, + { + "epoch": 0.6355839874849848, + "grad_norm": 0.886133074760437, + "learning_rate": 0.000684530479605558, + "loss": 0.4537, + "step": 11376 + }, + { + "epoch": 0.6356398580886666, + "grad_norm": 0.6638587117195129, + "learning_rate": 0.0006845024652622143, + "loss": 0.3542, + "step": 11377 + }, + { + "epoch": 0.6356957286923485, + "grad_norm": 0.39902305603027344, + "learning_rate": 0.0006844744509188705, + "loss": 0.385, + "step": 11378 + }, + { + "epoch": 0.6357515992960304, + "grad_norm": 0.5712414383888245, + "learning_rate": 0.0006844464365755266, + "loss": 0.5439, + "step": 11379 + }, + { + "epoch": 0.6358074698997123, + "grad_norm": 0.619327962398529, + "learning_rate": 0.0006844184222321828, + "loss": 0.3727, + "step": 11380 + }, + { + "epoch": 0.6358633405033941, + "grad_norm": 0.6807682514190674, + "learning_rate": 0.0006843904078888391, + "loss": 0.4713, + "step": 11381 + }, + { + "epoch": 0.635919211107076, + "grad_norm": 1.0232585668563843, + "learning_rate": 0.0006843623935454953, + "loss": 0.4874, + "step": 11382 + }, + { + "epoch": 0.6359750817107579, + "grad_norm": 0.7262845039367676, + "learning_rate": 0.0006843343792021514, + "loss": 0.4944, + "step": 11383 + }, + { + "epoch": 0.6360309523144397, + "grad_norm": 0.4873013496398926, + "learning_rate": 0.0006843063648588077, + "loss": 0.5015, + "step": 11384 + }, + { + "epoch": 0.6360868229181216, + "grad_norm": 0.8490318655967712, + "learning_rate": 0.000684278350515464, + "loss": 0.4154, + "step": 11385 + }, + { + "epoch": 0.6361426935218035, + "grad_norm": 0.6251857280731201, + "learning_rate": 0.0006842503361721202, + "loss": 0.4523, + "step": 11386 + }, + { + "epoch": 0.6361985641254854, + "grad_norm": 0.5266708135604858, + "learning_rate": 0.0006842223218287763, + "loss": 0.4726, + "step": 11387 + }, + { + "epoch": 0.6362544347291672, + "grad_norm": 0.5620631575584412, + "learning_rate": 0.0006841943074854326, + "loss": 0.41, + "step": 11388 + }, + { + "epoch": 0.6363103053328492, + "grad_norm": 3.440857172012329, + "learning_rate": 0.0006841662931420888, + "loss": 0.3597, + "step": 11389 + }, + { + "epoch": 0.636366175936531, + "grad_norm": 0.4863125681877136, + "learning_rate": 0.000684138278798745, + "loss": 0.4505, + "step": 11390 + }, + { + "epoch": 0.6364220465402128, + "grad_norm": 0.5482548475265503, + "learning_rate": 0.0006841102644554012, + "loss": 0.4599, + "step": 11391 + }, + { + "epoch": 0.6364779171438948, + "grad_norm": 0.5396884679794312, + "learning_rate": 0.0006840822501120574, + "loss": 0.5262, + "step": 11392 + }, + { + "epoch": 0.6365337877475766, + "grad_norm": 0.6125249266624451, + "learning_rate": 0.0006840542357687136, + "loss": 0.4823, + "step": 11393 + }, + { + "epoch": 0.6365896583512585, + "grad_norm": 0.5398811101913452, + "learning_rate": 0.0006840262214253698, + "loss": 0.442, + "step": 11394 + }, + { + "epoch": 0.6366455289549403, + "grad_norm": 0.5847583413124084, + "learning_rate": 0.000683998207082026, + "loss": 0.3959, + "step": 11395 + }, + { + "epoch": 0.6367013995586223, + "grad_norm": 1.7209343910217285, + "learning_rate": 0.0006839701927386822, + "loss": 0.3777, + "step": 11396 + }, + { + "epoch": 0.6367572701623041, + "grad_norm": 2.095736265182495, + "learning_rate": 0.0006839421783953384, + "loss": 0.4556, + "step": 11397 + }, + { + "epoch": 0.6368131407659859, + "grad_norm": 0.43395957350730896, + "learning_rate": 0.0006839141640519946, + "loss": 0.4313, + "step": 11398 + }, + { + "epoch": 0.6368690113696679, + "grad_norm": 0.5285685062408447, + "learning_rate": 0.0006838861497086508, + "loss": 0.4661, + "step": 11399 + }, + { + "epoch": 0.6369248819733497, + "grad_norm": 0.4932166337966919, + "learning_rate": 0.000683858135365307, + "loss": 0.5157, + "step": 11400 + }, + { + "epoch": 0.6369807525770316, + "grad_norm": 0.6883057355880737, + "learning_rate": 0.0006838301210219632, + "loss": 0.6094, + "step": 11401 + }, + { + "epoch": 0.6370366231807134, + "grad_norm": 0.46947935223579407, + "learning_rate": 0.0006838021066786195, + "loss": 0.4621, + "step": 11402 + }, + { + "epoch": 0.6370924937843954, + "grad_norm": 0.5205413103103638, + "learning_rate": 0.0006837740923352757, + "loss": 0.4414, + "step": 11403 + }, + { + "epoch": 0.6371483643880772, + "grad_norm": 0.3725862205028534, + "learning_rate": 0.0006837460779919319, + "loss": 0.3556, + "step": 11404 + }, + { + "epoch": 0.637204234991759, + "grad_norm": 1.061444878578186, + "learning_rate": 0.0006837180636485882, + "loss": 0.3907, + "step": 11405 + }, + { + "epoch": 0.637260105595441, + "grad_norm": 0.4347083270549774, + "learning_rate": 0.0006836900493052443, + "loss": 0.4536, + "step": 11406 + }, + { + "epoch": 0.6373159761991228, + "grad_norm": 10.393534660339355, + "learning_rate": 0.0006836620349619005, + "loss": 0.5847, + "step": 11407 + }, + { + "epoch": 0.6373718468028047, + "grad_norm": 0.7491111159324646, + "learning_rate": 0.0006836340206185567, + "loss": 0.4494, + "step": 11408 + }, + { + "epoch": 0.6374277174064866, + "grad_norm": 0.43870675563812256, + "learning_rate": 0.000683606006275213, + "loss": 0.4236, + "step": 11409 + }, + { + "epoch": 0.6374835880101685, + "grad_norm": 0.40753766894340515, + "learning_rate": 0.0006835779919318691, + "loss": 0.3879, + "step": 11410 + }, + { + "epoch": 0.6375394586138503, + "grad_norm": 0.40884214639663696, + "learning_rate": 0.0006835499775885253, + "loss": 0.3866, + "step": 11411 + }, + { + "epoch": 0.6375953292175321, + "grad_norm": 0.36223751306533813, + "learning_rate": 0.0006835219632451816, + "loss": 0.3386, + "step": 11412 + }, + { + "epoch": 0.6376511998212141, + "grad_norm": 0.5323598384857178, + "learning_rate": 0.0006834939489018378, + "loss": 0.5063, + "step": 11413 + }, + { + "epoch": 0.6377070704248959, + "grad_norm": 8.66077995300293, + "learning_rate": 0.0006834659345584939, + "loss": 0.442, + "step": 11414 + }, + { + "epoch": 0.6377629410285778, + "grad_norm": 1.6373034715652466, + "learning_rate": 0.0006834379202151501, + "loss": 0.3883, + "step": 11415 + }, + { + "epoch": 0.6378188116322597, + "grad_norm": 3.745107889175415, + "learning_rate": 0.0006834099058718064, + "loss": 0.5037, + "step": 11416 + }, + { + "epoch": 0.6378746822359416, + "grad_norm": 0.4921225309371948, + "learning_rate": 0.0006833818915284626, + "loss": 0.5102, + "step": 11417 + }, + { + "epoch": 0.6379305528396234, + "grad_norm": 0.7116353511810303, + "learning_rate": 0.0006833538771851187, + "loss": 0.538, + "step": 11418 + }, + { + "epoch": 0.6379864234433053, + "grad_norm": 0.4648634195327759, + "learning_rate": 0.000683325862841775, + "loss": 0.5295, + "step": 11419 + }, + { + "epoch": 0.6380422940469872, + "grad_norm": 0.54286789894104, + "learning_rate": 0.0006832978484984313, + "loss": 0.583, + "step": 11420 + }, + { + "epoch": 0.638098164650669, + "grad_norm": 3.3969194889068604, + "learning_rate": 0.0006832698341550874, + "loss": 0.4613, + "step": 11421 + }, + { + "epoch": 0.6381540352543509, + "grad_norm": 0.333391010761261, + "learning_rate": 0.0006832418198117436, + "loss": 0.412, + "step": 11422 + }, + { + "epoch": 0.6382099058580328, + "grad_norm": 0.6688084602355957, + "learning_rate": 0.0006832138054683999, + "loss": 0.4609, + "step": 11423 + }, + { + "epoch": 0.6382657764617147, + "grad_norm": 0.4769790470600128, + "learning_rate": 0.0006831857911250561, + "loss": 0.5887, + "step": 11424 + }, + { + "epoch": 0.6383216470653965, + "grad_norm": 0.5707235336303711, + "learning_rate": 0.0006831577767817122, + "loss": 0.4742, + "step": 11425 + }, + { + "epoch": 0.6383775176690785, + "grad_norm": 0.5748907327651978, + "learning_rate": 0.0006831297624383684, + "loss": 0.5339, + "step": 11426 + }, + { + "epoch": 0.6384333882727603, + "grad_norm": 0.4521685540676117, + "learning_rate": 0.0006831017480950247, + "loss": 0.4533, + "step": 11427 + }, + { + "epoch": 0.6384892588764421, + "grad_norm": 0.37167099118232727, + "learning_rate": 0.0006830737337516809, + "loss": 0.4429, + "step": 11428 + }, + { + "epoch": 0.638545129480124, + "grad_norm": 0.49489647150039673, + "learning_rate": 0.000683045719408337, + "loss": 0.4116, + "step": 11429 + }, + { + "epoch": 0.6386010000838059, + "grad_norm": 0.6384615302085876, + "learning_rate": 0.0006830177050649933, + "loss": 0.5842, + "step": 11430 + }, + { + "epoch": 0.6386568706874878, + "grad_norm": 0.5502079725265503, + "learning_rate": 0.0006829896907216495, + "loss": 0.3812, + "step": 11431 + }, + { + "epoch": 0.6387127412911696, + "grad_norm": 0.4828428626060486, + "learning_rate": 0.0006829616763783057, + "loss": 0.4763, + "step": 11432 + }, + { + "epoch": 0.6387686118948516, + "grad_norm": 0.4553386867046356, + "learning_rate": 0.0006829336620349618, + "loss": 0.4543, + "step": 11433 + }, + { + "epoch": 0.6388244824985334, + "grad_norm": 1.9208201169967651, + "learning_rate": 0.0006829056476916181, + "loss": 0.4897, + "step": 11434 + }, + { + "epoch": 0.6388803531022152, + "grad_norm": 0.5216221213340759, + "learning_rate": 0.0006828776333482744, + "loss": 0.4835, + "step": 11435 + }, + { + "epoch": 0.6389362237058971, + "grad_norm": 0.35320231318473816, + "learning_rate": 0.0006828496190049306, + "loss": 0.4078, + "step": 11436 + }, + { + "epoch": 0.638992094309579, + "grad_norm": 0.8623009324073792, + "learning_rate": 0.0006828216046615868, + "loss": 0.478, + "step": 11437 + }, + { + "epoch": 0.6390479649132609, + "grad_norm": 0.5886081457138062, + "learning_rate": 0.000682793590318243, + "loss": 0.4767, + "step": 11438 + }, + { + "epoch": 0.6391038355169427, + "grad_norm": 0.6327439546585083, + "learning_rate": 0.0006827655759748992, + "loss": 0.506, + "step": 11439 + }, + { + "epoch": 0.6391597061206247, + "grad_norm": 0.5809468626976013, + "learning_rate": 0.0006827375616315553, + "loss": 0.4599, + "step": 11440 + }, + { + "epoch": 0.6392155767243065, + "grad_norm": 0.392841637134552, + "learning_rate": 0.0006827095472882116, + "loss": 0.3965, + "step": 11441 + }, + { + "epoch": 0.6392714473279884, + "grad_norm": 0.4870837330818176, + "learning_rate": 0.0006826815329448678, + "loss": 0.4064, + "step": 11442 + }, + { + "epoch": 0.6393273179316703, + "grad_norm": 1.0995746850967407, + "learning_rate": 0.000682653518601524, + "loss": 0.3786, + "step": 11443 + }, + { + "epoch": 0.6393831885353521, + "grad_norm": 0.43472641706466675, + "learning_rate": 0.0006826255042581802, + "loss": 0.4, + "step": 11444 + }, + { + "epoch": 0.639439059139034, + "grad_norm": 0.4373909831047058, + "learning_rate": 0.0006825974899148364, + "loss": 0.4302, + "step": 11445 + }, + { + "epoch": 0.6394949297427158, + "grad_norm": 0.4278801381587982, + "learning_rate": 0.0006825694755714926, + "loss": 0.5232, + "step": 11446 + }, + { + "epoch": 0.6395508003463978, + "grad_norm": 0.6221674084663391, + "learning_rate": 0.0006825414612281488, + "loss": 0.4976, + "step": 11447 + }, + { + "epoch": 0.6396066709500796, + "grad_norm": 0.6645322442054749, + "learning_rate": 0.000682513446884805, + "loss": 0.4976, + "step": 11448 + }, + { + "epoch": 0.6396625415537615, + "grad_norm": 0.4041202962398529, + "learning_rate": 0.0006824854325414612, + "loss": 0.4719, + "step": 11449 + }, + { + "epoch": 0.6397184121574434, + "grad_norm": 3.279813766479492, + "learning_rate": 0.0006824574181981174, + "loss": 0.6194, + "step": 11450 + }, + { + "epoch": 0.6397742827611252, + "grad_norm": 1.3836629390716553, + "learning_rate": 0.0006824294038547736, + "loss": 0.5678, + "step": 11451 + }, + { + "epoch": 0.6398301533648071, + "grad_norm": 0.4742598235607147, + "learning_rate": 0.0006824013895114299, + "loss": 0.4, + "step": 11452 + }, + { + "epoch": 0.6398860239684889, + "grad_norm": 1.7176811695098877, + "learning_rate": 0.0006823733751680861, + "loss": 0.4543, + "step": 11453 + }, + { + "epoch": 0.6399418945721709, + "grad_norm": 15.97020435333252, + "learning_rate": 0.0006823453608247423, + "loss": 0.4309, + "step": 11454 + }, + { + "epoch": 0.6399977651758527, + "grad_norm": 0.6796072125434875, + "learning_rate": 0.0006823173464813986, + "loss": 0.3632, + "step": 11455 + }, + { + "epoch": 0.6400536357795346, + "grad_norm": 0.5804921388626099, + "learning_rate": 0.0006822893321380547, + "loss": 0.4914, + "step": 11456 + }, + { + "epoch": 0.6401095063832165, + "grad_norm": 2.5291311740875244, + "learning_rate": 0.0006822613177947109, + "loss": 0.3781, + "step": 11457 + }, + { + "epoch": 0.6401653769868983, + "grad_norm": 0.5463931560516357, + "learning_rate": 0.0006822333034513671, + "loss": 0.4786, + "step": 11458 + }, + { + "epoch": 0.6402212475905802, + "grad_norm": 0.7863777279853821, + "learning_rate": 0.0006822052891080234, + "loss": 0.511, + "step": 11459 + }, + { + "epoch": 0.6402771181942621, + "grad_norm": 0.578152060508728, + "learning_rate": 0.0006821772747646795, + "loss": 0.4515, + "step": 11460 + }, + { + "epoch": 0.640332988797944, + "grad_norm": 0.41628527641296387, + "learning_rate": 0.0006821492604213357, + "loss": 0.517, + "step": 11461 + }, + { + "epoch": 0.6403888594016258, + "grad_norm": 0.48850908875465393, + "learning_rate": 0.000682121246077992, + "loss": 0.4929, + "step": 11462 + }, + { + "epoch": 0.6404447300053077, + "grad_norm": 0.3662755787372589, + "learning_rate": 0.0006820932317346481, + "loss": 0.4033, + "step": 11463 + }, + { + "epoch": 0.6405006006089896, + "grad_norm": 0.41410595178604126, + "learning_rate": 0.0006820652173913043, + "loss": 0.4695, + "step": 11464 + }, + { + "epoch": 0.6405564712126715, + "grad_norm": 0.6360518932342529, + "learning_rate": 0.0006820372030479605, + "loss": 0.4564, + "step": 11465 + }, + { + "epoch": 0.6406123418163533, + "grad_norm": 0.370284765958786, + "learning_rate": 0.0006820091887046168, + "loss": 0.3236, + "step": 11466 + }, + { + "epoch": 0.6406682124200352, + "grad_norm": 1.343434453010559, + "learning_rate": 0.0006819811743612729, + "loss": 0.5384, + "step": 11467 + }, + { + "epoch": 0.6407240830237171, + "grad_norm": 0.4672095477581024, + "learning_rate": 0.0006819531600179291, + "loss": 0.4045, + "step": 11468 + }, + { + "epoch": 0.6407799536273989, + "grad_norm": 0.4644332528114319, + "learning_rate": 0.0006819251456745855, + "loss": 0.4682, + "step": 11469 + }, + { + "epoch": 0.6408358242310808, + "grad_norm": 0.4923323690891266, + "learning_rate": 0.0006818971313312417, + "loss": 0.5293, + "step": 11470 + }, + { + "epoch": 0.6408916948347627, + "grad_norm": 0.41480791568756104, + "learning_rate": 0.0006818691169878978, + "loss": 0.5402, + "step": 11471 + }, + { + "epoch": 0.6409475654384446, + "grad_norm": 0.5922974348068237, + "learning_rate": 0.000681841102644554, + "loss": 0.4452, + "step": 11472 + }, + { + "epoch": 0.6410034360421264, + "grad_norm": 0.5942279100418091, + "learning_rate": 0.0006818130883012103, + "loss": 0.4149, + "step": 11473 + }, + { + "epoch": 0.6410593066458083, + "grad_norm": 0.5808907747268677, + "learning_rate": 0.0006817850739578665, + "loss": 0.3768, + "step": 11474 + }, + { + "epoch": 0.6411151772494902, + "grad_norm": 0.4868180751800537, + "learning_rate": 0.0006817570596145226, + "loss": 0.5355, + "step": 11475 + }, + { + "epoch": 0.641171047853172, + "grad_norm": 1.2850604057312012, + "learning_rate": 0.0006817290452711789, + "loss": 0.386, + "step": 11476 + }, + { + "epoch": 0.641226918456854, + "grad_norm": 0.49494677782058716, + "learning_rate": 0.0006817010309278351, + "loss": 0.3876, + "step": 11477 + }, + { + "epoch": 0.6412827890605358, + "grad_norm": 0.45027220249176025, + "learning_rate": 0.0006816730165844913, + "loss": 0.3865, + "step": 11478 + }, + { + "epoch": 0.6413386596642177, + "grad_norm": 0.4144449532032013, + "learning_rate": 0.0006816450022411474, + "loss": 0.3762, + "step": 11479 + }, + { + "epoch": 0.6413945302678995, + "grad_norm": 0.3629459738731384, + "learning_rate": 0.0006816169878978037, + "loss": 0.4336, + "step": 11480 + }, + { + "epoch": 0.6414504008715814, + "grad_norm": 0.7388653755187988, + "learning_rate": 0.0006815889735544599, + "loss": 0.6182, + "step": 11481 + }, + { + "epoch": 0.6415062714752633, + "grad_norm": 3.7114195823669434, + "learning_rate": 0.0006815609592111161, + "loss": 0.367, + "step": 11482 + }, + { + "epoch": 0.6415621420789451, + "grad_norm": 0.5018291473388672, + "learning_rate": 0.0006815329448677722, + "loss": 0.3751, + "step": 11483 + }, + { + "epoch": 0.6416180126826271, + "grad_norm": 0.4316920340061188, + "learning_rate": 0.0006815049305244285, + "loss": 0.4444, + "step": 11484 + }, + { + "epoch": 0.6416738832863089, + "grad_norm": 0.7858763933181763, + "learning_rate": 0.0006814769161810848, + "loss": 0.5251, + "step": 11485 + }, + { + "epoch": 0.6417297538899908, + "grad_norm": 0.4977187514305115, + "learning_rate": 0.0006814489018377408, + "loss": 0.366, + "step": 11486 + }, + { + "epoch": 0.6417856244936726, + "grad_norm": 0.6836873888969421, + "learning_rate": 0.0006814208874943972, + "loss": 0.382, + "step": 11487 + }, + { + "epoch": 0.6418414950973546, + "grad_norm": 0.4533426761627197, + "learning_rate": 0.0006813928731510534, + "loss": 0.4763, + "step": 11488 + }, + { + "epoch": 0.6418973657010364, + "grad_norm": 0.4315653145313263, + "learning_rate": 0.0006813648588077096, + "loss": 0.476, + "step": 11489 + }, + { + "epoch": 0.6419532363047182, + "grad_norm": 0.6401244401931763, + "learning_rate": 0.0006813368444643657, + "loss": 0.5409, + "step": 11490 + }, + { + "epoch": 0.6420091069084002, + "grad_norm": 0.695097029209137, + "learning_rate": 0.000681308830121022, + "loss": 0.3889, + "step": 11491 + }, + { + "epoch": 0.642064977512082, + "grad_norm": 0.6676113605499268, + "learning_rate": 0.0006812808157776782, + "loss": 0.4444, + "step": 11492 + }, + { + "epoch": 0.6421208481157639, + "grad_norm": 0.4801180064678192, + "learning_rate": 0.0006812528014343344, + "loss": 0.4224, + "step": 11493 + }, + { + "epoch": 0.6421767187194458, + "grad_norm": 0.5819891691207886, + "learning_rate": 0.0006812247870909906, + "loss": 0.4675, + "step": 11494 + }, + { + "epoch": 0.6422325893231277, + "grad_norm": 0.4830174148082733, + "learning_rate": 0.0006811967727476468, + "loss": 0.3991, + "step": 11495 + }, + { + "epoch": 0.6422884599268095, + "grad_norm": 1.7730536460876465, + "learning_rate": 0.000681168758404303, + "loss": 0.4222, + "step": 11496 + }, + { + "epoch": 0.6423443305304913, + "grad_norm": 0.43986138701438904, + "learning_rate": 0.0006811407440609592, + "loss": 0.4597, + "step": 11497 + }, + { + "epoch": 0.6424002011341733, + "grad_norm": 0.5206801891326904, + "learning_rate": 0.0006811127297176154, + "loss": 0.4574, + "step": 11498 + }, + { + "epoch": 0.6424560717378551, + "grad_norm": 0.5923295021057129, + "learning_rate": 0.0006810847153742716, + "loss": 0.4293, + "step": 11499 + }, + { + "epoch": 0.642511942341537, + "grad_norm": 0.9030653834342957, + "learning_rate": 0.0006810567010309278, + "loss": 0.4033, + "step": 11500 + }, + { + "epoch": 0.642511942341537, + "eval_cer": 0.0906789739561171, + "eval_loss": 0.342400461435318, + "eval_runtime": 55.9731, + "eval_samples_per_second": 81.075, + "eval_steps_per_second": 5.074, + "eval_wer": 0.35719672634842603, + "step": 11500 + }, + { + "epoch": 0.6425678129452189, + "grad_norm": 13.432381629943848, + "learning_rate": 0.0006810286866875842, + "loss": 0.3439, + "step": 11501 + }, + { + "epoch": 0.6426236835489008, + "grad_norm": 0.5672720074653625, + "learning_rate": 0.0006810006723442403, + "loss": 0.5775, + "step": 11502 + }, + { + "epoch": 0.6426795541525826, + "grad_norm": 0.5291237831115723, + "learning_rate": 0.0006809726580008965, + "loss": 0.4488, + "step": 11503 + }, + { + "epoch": 0.6427354247562644, + "grad_norm": 0.7949419021606445, + "learning_rate": 0.0006809446436575527, + "loss": 0.474, + "step": 11504 + }, + { + "epoch": 0.6427912953599464, + "grad_norm": 1.8241426944732666, + "learning_rate": 0.0006809166293142089, + "loss": 0.5879, + "step": 11505 + }, + { + "epoch": 0.6428471659636282, + "grad_norm": 0.3967645466327667, + "learning_rate": 0.0006808886149708651, + "loss": 0.3936, + "step": 11506 + }, + { + "epoch": 0.6429030365673101, + "grad_norm": 0.4404817819595337, + "learning_rate": 0.0006808606006275213, + "loss": 0.3894, + "step": 11507 + }, + { + "epoch": 0.642958907170992, + "grad_norm": 0.562120795249939, + "learning_rate": 0.0006808325862841776, + "loss": 0.4654, + "step": 11508 + }, + { + "epoch": 0.6430147777746739, + "grad_norm": 0.5470510125160217, + "learning_rate": 0.0006808045719408337, + "loss": 0.4211, + "step": 11509 + }, + { + "epoch": 0.6430706483783557, + "grad_norm": 0.549589216709137, + "learning_rate": 0.0006807765575974899, + "loss": 0.5675, + "step": 11510 + }, + { + "epoch": 0.6431265189820377, + "grad_norm": 0.43896055221557617, + "learning_rate": 0.0006807485432541461, + "loss": 0.3512, + "step": 11511 + }, + { + "epoch": 0.6431823895857195, + "grad_norm": 0.5348843336105347, + "learning_rate": 0.0006807205289108024, + "loss": 0.5959, + "step": 11512 + }, + { + "epoch": 0.6432382601894013, + "grad_norm": 0.6013202667236328, + "learning_rate": 0.0006806925145674585, + "loss": 0.4785, + "step": 11513 + }, + { + "epoch": 0.6432941307930832, + "grad_norm": 0.3849291205406189, + "learning_rate": 0.0006806645002241147, + "loss": 0.3997, + "step": 11514 + }, + { + "epoch": 0.6433500013967651, + "grad_norm": 0.3147357702255249, + "learning_rate": 0.000680636485880771, + "loss": 0.3883, + "step": 11515 + }, + { + "epoch": 0.643405872000447, + "grad_norm": 0.5475407242774963, + "learning_rate": 0.0006806084715374272, + "loss": 0.5176, + "step": 11516 + }, + { + "epoch": 0.6434617426041288, + "grad_norm": 0.8397700190544128, + "learning_rate": 0.0006805804571940833, + "loss": 0.4151, + "step": 11517 + }, + { + "epoch": 0.6435176132078108, + "grad_norm": 0.5640847086906433, + "learning_rate": 0.0006805524428507395, + "loss": 0.5131, + "step": 11518 + }, + { + "epoch": 0.6435734838114926, + "grad_norm": 0.6081070899963379, + "learning_rate": 0.0006805244285073959, + "loss": 0.4781, + "step": 11519 + }, + { + "epoch": 0.6436293544151744, + "grad_norm": 1.1746684312820435, + "learning_rate": 0.0006804964141640521, + "loss": 0.4926, + "step": 11520 + }, + { + "epoch": 0.6436852250188563, + "grad_norm": 1.2292348146438599, + "learning_rate": 0.0006804683998207082, + "loss": 0.3676, + "step": 11521 + }, + { + "epoch": 0.6437410956225382, + "grad_norm": 0.3144816756248474, + "learning_rate": 0.0006804403854773644, + "loss": 0.4029, + "step": 11522 + }, + { + "epoch": 0.6437969662262201, + "grad_norm": 0.4149528741836548, + "learning_rate": 0.0006804123711340207, + "loss": 0.5253, + "step": 11523 + }, + { + "epoch": 0.6438528368299019, + "grad_norm": 0.5927987694740295, + "learning_rate": 0.0006803843567906769, + "loss": 0.5256, + "step": 11524 + }, + { + "epoch": 0.6439087074335839, + "grad_norm": 0.7137500047683716, + "learning_rate": 0.000680356342447333, + "loss": 0.4887, + "step": 11525 + }, + { + "epoch": 0.6439645780372657, + "grad_norm": 0.5452092885971069, + "learning_rate": 0.0006803283281039893, + "loss": 0.4586, + "step": 11526 + }, + { + "epoch": 0.6440204486409475, + "grad_norm": 0.45550960302352905, + "learning_rate": 0.0006803003137606455, + "loss": 0.4319, + "step": 11527 + }, + { + "epoch": 0.6440763192446295, + "grad_norm": 0.3726610243320465, + "learning_rate": 0.0006802722994173016, + "loss": 0.3993, + "step": 11528 + }, + { + "epoch": 0.6441321898483113, + "grad_norm": 0.5474866628646851, + "learning_rate": 0.0006802442850739578, + "loss": 0.4782, + "step": 11529 + }, + { + "epoch": 0.6441880604519932, + "grad_norm": 0.4509713053703308, + "learning_rate": 0.0006802162707306141, + "loss": 0.4187, + "step": 11530 + }, + { + "epoch": 0.644243931055675, + "grad_norm": 0.7723603248596191, + "learning_rate": 0.0006801882563872703, + "loss": 0.3924, + "step": 11531 + }, + { + "epoch": 0.644299801659357, + "grad_norm": 0.6448555588722229, + "learning_rate": 0.0006801602420439264, + "loss": 0.4625, + "step": 11532 + }, + { + "epoch": 0.6443556722630388, + "grad_norm": 0.8401827216148376, + "learning_rate": 0.0006801322277005827, + "loss": 0.4291, + "step": 11533 + }, + { + "epoch": 0.6444115428667206, + "grad_norm": 0.47067803144454956, + "learning_rate": 0.000680104213357239, + "loss": 0.3853, + "step": 11534 + }, + { + "epoch": 0.6444674134704026, + "grad_norm": 3.071997880935669, + "learning_rate": 0.0006800761990138952, + "loss": 0.3697, + "step": 11535 + }, + { + "epoch": 0.6445232840740844, + "grad_norm": 0.5983738899230957, + "learning_rate": 0.0006800481846705512, + "loss": 0.4909, + "step": 11536 + }, + { + "epoch": 0.6445791546777663, + "grad_norm": 0.5735681653022766, + "learning_rate": 0.0006800201703272076, + "loss": 0.4888, + "step": 11537 + }, + { + "epoch": 0.6446350252814481, + "grad_norm": 0.4523080885410309, + "learning_rate": 0.0006799921559838638, + "loss": 0.4711, + "step": 11538 + }, + { + "epoch": 0.6446908958851301, + "grad_norm": 0.577050507068634, + "learning_rate": 0.00067996414164052, + "loss": 0.5296, + "step": 11539 + }, + { + "epoch": 0.6447467664888119, + "grad_norm": 0.5096551775932312, + "learning_rate": 0.0006799361272971762, + "loss": 0.4456, + "step": 11540 + }, + { + "epoch": 0.6448026370924937, + "grad_norm": 0.5811684131622314, + "learning_rate": 0.0006799081129538324, + "loss": 0.4907, + "step": 11541 + }, + { + "epoch": 0.6448585076961757, + "grad_norm": 5.926117420196533, + "learning_rate": 0.0006798800986104886, + "loss": 0.5042, + "step": 11542 + }, + { + "epoch": 0.6449143782998575, + "grad_norm": 1.8269685506820679, + "learning_rate": 0.0006798520842671448, + "loss": 0.4298, + "step": 11543 + }, + { + "epoch": 0.6449702489035394, + "grad_norm": 0.6519777178764343, + "learning_rate": 0.000679824069923801, + "loss": 0.4841, + "step": 11544 + }, + { + "epoch": 0.6450261195072213, + "grad_norm": 0.4324147403240204, + "learning_rate": 0.0006797960555804572, + "loss": 0.4608, + "step": 11545 + }, + { + "epoch": 0.6450819901109032, + "grad_norm": 0.6057810187339783, + "learning_rate": 0.0006797680412371134, + "loss": 0.4134, + "step": 11546 + }, + { + "epoch": 0.645137860714585, + "grad_norm": 0.4602072536945343, + "learning_rate": 0.0006797400268937696, + "loss": 0.3836, + "step": 11547 + }, + { + "epoch": 0.6451937313182668, + "grad_norm": 0.3763127624988556, + "learning_rate": 0.0006797120125504258, + "loss": 0.39, + "step": 11548 + }, + { + "epoch": 0.6452496019219488, + "grad_norm": 0.5089964866638184, + "learning_rate": 0.000679683998207082, + "loss": 0.3607, + "step": 11549 + }, + { + "epoch": 0.6453054725256306, + "grad_norm": 0.4514484703540802, + "learning_rate": 0.0006796559838637382, + "loss": 0.4167, + "step": 11550 + }, + { + "epoch": 0.6453613431293125, + "grad_norm": 1.4191609621047974, + "learning_rate": 0.0006796279695203944, + "loss": 0.4615, + "step": 11551 + }, + { + "epoch": 0.6454172137329944, + "grad_norm": 0.4958000183105469, + "learning_rate": 0.0006795999551770507, + "loss": 0.4689, + "step": 11552 + }, + { + "epoch": 0.6454730843366763, + "grad_norm": 0.5287027955055237, + "learning_rate": 0.0006795719408337069, + "loss": 0.36, + "step": 11553 + }, + { + "epoch": 0.6455289549403581, + "grad_norm": 0.6340475082397461, + "learning_rate": 0.0006795439264903632, + "loss": 0.4528, + "step": 11554 + }, + { + "epoch": 0.64558482554404, + "grad_norm": 0.6941174268722534, + "learning_rate": 0.0006795159121470193, + "loss": 0.4524, + "step": 11555 + }, + { + "epoch": 0.6456406961477219, + "grad_norm": 0.5152488946914673, + "learning_rate": 0.0006794878978036755, + "loss": 0.3809, + "step": 11556 + }, + { + "epoch": 0.6456965667514037, + "grad_norm": 0.5966008305549622, + "learning_rate": 0.0006794598834603317, + "loss": 0.4488, + "step": 11557 + }, + { + "epoch": 0.6457524373550856, + "grad_norm": 0.8994854092597961, + "learning_rate": 0.000679431869116988, + "loss": 0.4441, + "step": 11558 + }, + { + "epoch": 0.6458083079587675, + "grad_norm": 1.1048753261566162, + "learning_rate": 0.0006794038547736441, + "loss": 0.4799, + "step": 11559 + }, + { + "epoch": 0.6458641785624494, + "grad_norm": 0.541298508644104, + "learning_rate": 0.0006793758404303003, + "loss": 0.5042, + "step": 11560 + }, + { + "epoch": 0.6459200491661312, + "grad_norm": 1.525039553642273, + "learning_rate": 0.0006793478260869565, + "loss": 0.4274, + "step": 11561 + }, + { + "epoch": 0.6459759197698132, + "grad_norm": 0.40846577286720276, + "learning_rate": 0.0006793198117436128, + "loss": 0.5065, + "step": 11562 + }, + { + "epoch": 0.646031790373495, + "grad_norm": 2.7428998947143555, + "learning_rate": 0.0006792917974002689, + "loss": 0.4245, + "step": 11563 + }, + { + "epoch": 0.6460876609771768, + "grad_norm": 0.34091195464134216, + "learning_rate": 0.0006792637830569251, + "loss": 0.4191, + "step": 11564 + }, + { + "epoch": 0.6461435315808587, + "grad_norm": 0.5110710263252258, + "learning_rate": 0.0006792357687135814, + "loss": 0.4807, + "step": 11565 + }, + { + "epoch": 0.6461994021845406, + "grad_norm": 0.41156205534935, + "learning_rate": 0.0006792077543702376, + "loss": 0.3965, + "step": 11566 + }, + { + "epoch": 0.6462552727882225, + "grad_norm": 0.649430513381958, + "learning_rate": 0.0006791797400268937, + "loss": 0.3685, + "step": 11567 + }, + { + "epoch": 0.6463111433919043, + "grad_norm": 0.9859501123428345, + "learning_rate": 0.0006791517256835499, + "loss": 0.5551, + "step": 11568 + }, + { + "epoch": 0.6463670139955863, + "grad_norm": 1.2241342067718506, + "learning_rate": 0.0006791237113402063, + "loss": 0.4595, + "step": 11569 + }, + { + "epoch": 0.6464228845992681, + "grad_norm": 0.41051924228668213, + "learning_rate": 0.0006790956969968624, + "loss": 0.4715, + "step": 11570 + }, + { + "epoch": 0.64647875520295, + "grad_norm": 5.8693132400512695, + "learning_rate": 0.0006790676826535186, + "loss": 0.7774, + "step": 11571 + }, + { + "epoch": 0.6465346258066318, + "grad_norm": 1.4542211294174194, + "learning_rate": 0.0006790396683101749, + "loss": 0.4743, + "step": 11572 + }, + { + "epoch": 0.6465904964103137, + "grad_norm": 0.5339109897613525, + "learning_rate": 0.0006790116539668311, + "loss": 0.4051, + "step": 11573 + }, + { + "epoch": 0.6466463670139956, + "grad_norm": 0.5757443308830261, + "learning_rate": 0.0006789836396234872, + "loss": 0.4937, + "step": 11574 + }, + { + "epoch": 0.6467022376176774, + "grad_norm": 0.542805552482605, + "learning_rate": 0.0006789556252801434, + "loss": 0.5161, + "step": 11575 + }, + { + "epoch": 0.6467581082213594, + "grad_norm": 1.1883506774902344, + "learning_rate": 0.0006789276109367997, + "loss": 0.3519, + "step": 11576 + }, + { + "epoch": 0.6468139788250412, + "grad_norm": 4.72308874130249, + "learning_rate": 0.0006788995965934559, + "loss": 0.4152, + "step": 11577 + }, + { + "epoch": 0.646869849428723, + "grad_norm": 0.5821355581283569, + "learning_rate": 0.000678871582250112, + "loss": 0.5835, + "step": 11578 + }, + { + "epoch": 0.646925720032405, + "grad_norm": 0.7493256330490112, + "learning_rate": 0.0006788435679067683, + "loss": 0.543, + "step": 11579 + }, + { + "epoch": 0.6469815906360868, + "grad_norm": 0.47415658831596375, + "learning_rate": 0.0006788155535634245, + "loss": 0.4144, + "step": 11580 + }, + { + "epoch": 0.6470374612397687, + "grad_norm": 0.4380160868167877, + "learning_rate": 0.0006787875392200807, + "loss": 0.4079, + "step": 11581 + }, + { + "epoch": 0.6470933318434505, + "grad_norm": 0.8972997069358826, + "learning_rate": 0.0006787595248767368, + "loss": 0.4642, + "step": 11582 + }, + { + "epoch": 0.6471492024471325, + "grad_norm": 0.5211485624313354, + "learning_rate": 0.0006787315105333931, + "loss": 0.4639, + "step": 11583 + }, + { + "epoch": 0.6472050730508143, + "grad_norm": 0.7144274115562439, + "learning_rate": 0.0006787034961900493, + "loss": 0.5124, + "step": 11584 + }, + { + "epoch": 0.6472609436544962, + "grad_norm": 0.35405033826828003, + "learning_rate": 0.0006786754818467056, + "loss": 0.4514, + "step": 11585 + }, + { + "epoch": 0.6473168142581781, + "grad_norm": 0.5674266815185547, + "learning_rate": 0.0006786474675033618, + "loss": 0.4006, + "step": 11586 + }, + { + "epoch": 0.64737268486186, + "grad_norm": 0.5131620168685913, + "learning_rate": 0.000678619453160018, + "loss": 0.463, + "step": 11587 + }, + { + "epoch": 0.6474285554655418, + "grad_norm": 0.6190488934516907, + "learning_rate": 0.0006785914388166742, + "loss": 0.6817, + "step": 11588 + }, + { + "epoch": 0.6474844260692236, + "grad_norm": 1.0990500450134277, + "learning_rate": 0.0006785634244733303, + "loss": 0.5222, + "step": 11589 + }, + { + "epoch": 0.6475402966729056, + "grad_norm": 0.43771761655807495, + "learning_rate": 0.0006785354101299866, + "loss": 0.3828, + "step": 11590 + }, + { + "epoch": 0.6475961672765874, + "grad_norm": 4.657693386077881, + "learning_rate": 0.0006785073957866428, + "loss": 0.3674, + "step": 11591 + }, + { + "epoch": 0.6476520378802693, + "grad_norm": 0.6917026042938232, + "learning_rate": 0.000678479381443299, + "loss": 0.4612, + "step": 11592 + }, + { + "epoch": 0.6477079084839512, + "grad_norm": 0.37101882696151733, + "learning_rate": 0.0006784513670999551, + "loss": 0.3812, + "step": 11593 + }, + { + "epoch": 0.647763779087633, + "grad_norm": 0.5901873707771301, + "learning_rate": 0.0006784233527566114, + "loss": 0.3844, + "step": 11594 + }, + { + "epoch": 0.6478196496913149, + "grad_norm": 0.7292694449424744, + "learning_rate": 0.0006783953384132676, + "loss": 0.4675, + "step": 11595 + }, + { + "epoch": 0.6478755202949968, + "grad_norm": 0.4497532844543457, + "learning_rate": 0.0006783673240699238, + "loss": 0.4086, + "step": 11596 + }, + { + "epoch": 0.6479313908986787, + "grad_norm": 0.5530431866645813, + "learning_rate": 0.00067833930972658, + "loss": 0.5698, + "step": 11597 + }, + { + "epoch": 0.6479872615023605, + "grad_norm": 0.47854283452033997, + "learning_rate": 0.0006783112953832362, + "loss": 0.4519, + "step": 11598 + }, + { + "epoch": 0.6480431321060424, + "grad_norm": 0.5254963040351868, + "learning_rate": 0.0006782832810398924, + "loss": 0.4403, + "step": 11599 + }, + { + "epoch": 0.6480990027097243, + "grad_norm": 0.49434900283813477, + "learning_rate": 0.0006782552666965486, + "loss": 0.4827, + "step": 11600 + }, + { + "epoch": 0.6481548733134062, + "grad_norm": 0.3966476023197174, + "learning_rate": 0.0006782272523532048, + "loss": 0.3559, + "step": 11601 + }, + { + "epoch": 0.648210743917088, + "grad_norm": 0.7008979916572571, + "learning_rate": 0.000678199238009861, + "loss": 0.4986, + "step": 11602 + }, + { + "epoch": 0.64826661452077, + "grad_norm": 0.9415128231048584, + "learning_rate": 0.0006781712236665173, + "loss": 0.5292, + "step": 11603 + }, + { + "epoch": 0.6483224851244518, + "grad_norm": 2.2434961795806885, + "learning_rate": 0.0006781432093231736, + "loss": 0.5272, + "step": 11604 + }, + { + "epoch": 0.6483783557281336, + "grad_norm": 0.7087838053703308, + "learning_rate": 0.0006781151949798297, + "loss": 0.3403, + "step": 11605 + }, + { + "epoch": 0.6484342263318155, + "grad_norm": 0.46735942363739014, + "learning_rate": 0.0006780871806364859, + "loss": 0.441, + "step": 11606 + }, + { + "epoch": 0.6484900969354974, + "grad_norm": 0.4802722930908203, + "learning_rate": 0.0006780591662931421, + "loss": 0.4813, + "step": 11607 + }, + { + "epoch": 0.6485459675391793, + "grad_norm": 0.873013973236084, + "learning_rate": 0.0006780311519497984, + "loss": 0.4288, + "step": 11608 + }, + { + "epoch": 0.6486018381428611, + "grad_norm": 0.8049773573875427, + "learning_rate": 0.0006780031376064545, + "loss": 0.4684, + "step": 11609 + }, + { + "epoch": 0.648657708746543, + "grad_norm": 0.41690102219581604, + "learning_rate": 0.0006779751232631107, + "loss": 0.4073, + "step": 11610 + }, + { + "epoch": 0.6487135793502249, + "grad_norm": 0.38432732224464417, + "learning_rate": 0.000677947108919767, + "loss": 0.425, + "step": 11611 + }, + { + "epoch": 0.6487694499539067, + "grad_norm": 0.4993667006492615, + "learning_rate": 0.0006779190945764231, + "loss": 0.5597, + "step": 11612 + }, + { + "epoch": 0.6488253205575886, + "grad_norm": 4.291062355041504, + "learning_rate": 0.0006778910802330793, + "loss": 0.4875, + "step": 11613 + }, + { + "epoch": 0.6488811911612705, + "grad_norm": 1.009907841682434, + "learning_rate": 0.0006778630658897355, + "loss": 0.4186, + "step": 11614 + }, + { + "epoch": 0.6489370617649524, + "grad_norm": 0.7582849264144897, + "learning_rate": 0.0006778350515463918, + "loss": 0.5953, + "step": 11615 + }, + { + "epoch": 0.6489929323686342, + "grad_norm": 0.8332192897796631, + "learning_rate": 0.0006778070372030479, + "loss": 0.4497, + "step": 11616 + }, + { + "epoch": 0.6490488029723162, + "grad_norm": 0.5574273467063904, + "learning_rate": 0.0006777790228597041, + "loss": 0.4184, + "step": 11617 + }, + { + "epoch": 0.649104673575998, + "grad_norm": 0.7184653282165527, + "learning_rate": 0.0006777510085163605, + "loss": 0.4587, + "step": 11618 + }, + { + "epoch": 0.6491605441796798, + "grad_norm": 0.6840894222259521, + "learning_rate": 0.0006777229941730167, + "loss": 0.4502, + "step": 11619 + }, + { + "epoch": 0.6492164147833618, + "grad_norm": 0.6686516404151917, + "learning_rate": 0.0006776949798296728, + "loss": 0.3687, + "step": 11620 + }, + { + "epoch": 0.6492722853870436, + "grad_norm": 0.5127590894699097, + "learning_rate": 0.000677666965486329, + "loss": 0.3572, + "step": 11621 + }, + { + "epoch": 0.6493281559907255, + "grad_norm": 0.655938982963562, + "learning_rate": 0.0006776389511429853, + "loss": 0.8208, + "step": 11622 + }, + { + "epoch": 0.6493840265944073, + "grad_norm": 0.41722121834754944, + "learning_rate": 0.0006776109367996415, + "loss": 0.4477, + "step": 11623 + }, + { + "epoch": 0.6494398971980893, + "grad_norm": 0.5419352650642395, + "learning_rate": 0.0006775829224562976, + "loss": 0.4284, + "step": 11624 + }, + { + "epoch": 0.6494957678017711, + "grad_norm": 0.7010300159454346, + "learning_rate": 0.0006775549081129539, + "loss": 0.3686, + "step": 11625 + }, + { + "epoch": 0.6495516384054529, + "grad_norm": 2.842231512069702, + "learning_rate": 0.0006775268937696101, + "loss": 0.5991, + "step": 11626 + }, + { + "epoch": 0.6496075090091349, + "grad_norm": 0.7839992642402649, + "learning_rate": 0.0006774988794262663, + "loss": 0.4687, + "step": 11627 + }, + { + "epoch": 0.6496633796128167, + "grad_norm": 0.4659079611301422, + "learning_rate": 0.0006774708650829224, + "loss": 0.5913, + "step": 11628 + }, + { + "epoch": 0.6497192502164986, + "grad_norm": 0.4496075212955475, + "learning_rate": 0.0006774428507395787, + "loss": 0.3757, + "step": 11629 + }, + { + "epoch": 0.6497751208201804, + "grad_norm": 0.39058443903923035, + "learning_rate": 0.0006774148363962349, + "loss": 0.3892, + "step": 11630 + }, + { + "epoch": 0.6498309914238624, + "grad_norm": 0.5109968781471252, + "learning_rate": 0.0006773868220528911, + "loss": 0.4824, + "step": 11631 + }, + { + "epoch": 0.6498868620275442, + "grad_norm": 0.5463802814483643, + "learning_rate": 0.0006773588077095472, + "loss": 0.551, + "step": 11632 + }, + { + "epoch": 0.649942732631226, + "grad_norm": 0.7534076571464539, + "learning_rate": 0.0006773307933662035, + "loss": 0.571, + "step": 11633 + }, + { + "epoch": 0.649998603234908, + "grad_norm": 0.505126416683197, + "learning_rate": 0.0006773027790228597, + "loss": 0.3783, + "step": 11634 + }, + { + "epoch": 0.6500544738385898, + "grad_norm": 0.35712680220603943, + "learning_rate": 0.0006772747646795158, + "loss": 0.3551, + "step": 11635 + }, + { + "epoch": 0.6501103444422717, + "grad_norm": 0.551135241985321, + "learning_rate": 0.0006772467503361722, + "loss": 0.4188, + "step": 11636 + }, + { + "epoch": 0.6501662150459536, + "grad_norm": 0.430326908826828, + "learning_rate": 0.0006772187359928284, + "loss": 0.5052, + "step": 11637 + }, + { + "epoch": 0.6502220856496355, + "grad_norm": 0.5853392481803894, + "learning_rate": 0.0006771907216494846, + "loss": 0.5177, + "step": 11638 + }, + { + "epoch": 0.6502779562533173, + "grad_norm": 6.479478359222412, + "learning_rate": 0.0006771627073061407, + "loss": 0.4496, + "step": 11639 + }, + { + "epoch": 0.6503338268569991, + "grad_norm": 0.8059582710266113, + "learning_rate": 0.000677134692962797, + "loss": 0.4024, + "step": 11640 + }, + { + "epoch": 0.6503896974606811, + "grad_norm": 0.5701563954353333, + "learning_rate": 0.0006771066786194532, + "loss": 0.3673, + "step": 11641 + }, + { + "epoch": 0.6504455680643629, + "grad_norm": 1.8575098514556885, + "learning_rate": 0.0006770786642761094, + "loss": 0.4315, + "step": 11642 + }, + { + "epoch": 0.6505014386680448, + "grad_norm": 0.55864018201828, + "learning_rate": 0.0006770506499327656, + "loss": 0.5202, + "step": 11643 + }, + { + "epoch": 0.6505573092717267, + "grad_norm": 1.5040792226791382, + "learning_rate": 0.0006770226355894218, + "loss": 0.4021, + "step": 11644 + }, + { + "epoch": 0.6506131798754086, + "grad_norm": 0.4699106216430664, + "learning_rate": 0.000676994621246078, + "loss": 0.3577, + "step": 11645 + }, + { + "epoch": 0.6506690504790904, + "grad_norm": 0.5594823360443115, + "learning_rate": 0.0006769666069027342, + "loss": 0.3938, + "step": 11646 + }, + { + "epoch": 0.6507249210827722, + "grad_norm": 0.4256640076637268, + "learning_rate": 0.0006769385925593904, + "loss": 0.4853, + "step": 11647 + }, + { + "epoch": 0.6507807916864542, + "grad_norm": 17.20612335205078, + "learning_rate": 0.0006769105782160466, + "loss": 0.447, + "step": 11648 + }, + { + "epoch": 0.650836662290136, + "grad_norm": 0.5787219405174255, + "learning_rate": 0.0006768825638727028, + "loss": 0.5194, + "step": 11649 + }, + { + "epoch": 0.6508925328938179, + "grad_norm": 0.442265123128891, + "learning_rate": 0.0006768545495293591, + "loss": 0.4077, + "step": 11650 + }, + { + "epoch": 0.6509484034974998, + "grad_norm": 2.5445749759674072, + "learning_rate": 0.0006768265351860152, + "loss": 0.4851, + "step": 11651 + }, + { + "epoch": 0.6510042741011817, + "grad_norm": 0.5790731906890869, + "learning_rate": 0.0006767985208426714, + "loss": 0.4284, + "step": 11652 + }, + { + "epoch": 0.6510601447048635, + "grad_norm": 0.48015129566192627, + "learning_rate": 0.0006767705064993277, + "loss": 0.4647, + "step": 11653 + }, + { + "epoch": 0.6511160153085455, + "grad_norm": 0.5071241855621338, + "learning_rate": 0.0006767424921559839, + "loss": 0.4583, + "step": 11654 + }, + { + "epoch": 0.6511718859122273, + "grad_norm": 1.0982784032821655, + "learning_rate": 0.0006767144778126401, + "loss": 0.3731, + "step": 11655 + }, + { + "epoch": 0.6512277565159091, + "grad_norm": 0.4986317753791809, + "learning_rate": 0.0006766864634692963, + "loss": 0.5188, + "step": 11656 + }, + { + "epoch": 0.651283627119591, + "grad_norm": 0.8034142851829529, + "learning_rate": 0.0006766584491259526, + "loss": 0.517, + "step": 11657 + }, + { + "epoch": 0.6513394977232729, + "grad_norm": 0.43738511204719543, + "learning_rate": 0.0006766304347826087, + "loss": 0.5028, + "step": 11658 + }, + { + "epoch": 0.6513953683269548, + "grad_norm": 0.5498617887496948, + "learning_rate": 0.0006766024204392649, + "loss": 0.4581, + "step": 11659 + }, + { + "epoch": 0.6514512389306366, + "grad_norm": 0.4378119111061096, + "learning_rate": 0.0006765744060959211, + "loss": 0.4245, + "step": 11660 + }, + { + "epoch": 0.6515071095343186, + "grad_norm": 0.6957234144210815, + "learning_rate": 0.0006765463917525774, + "loss": 0.4558, + "step": 11661 + }, + { + "epoch": 0.6515629801380004, + "grad_norm": 0.566748321056366, + "learning_rate": 0.0006765183774092335, + "loss": 0.4893, + "step": 11662 + }, + { + "epoch": 0.6516188507416822, + "grad_norm": 0.5257377624511719, + "learning_rate": 0.0006764903630658897, + "loss": 0.3633, + "step": 11663 + }, + { + "epoch": 0.6516747213453641, + "grad_norm": 0.5210147500038147, + "learning_rate": 0.0006764623487225459, + "loss": 0.5152, + "step": 11664 + }, + { + "epoch": 0.651730591949046, + "grad_norm": 2.5857536792755127, + "learning_rate": 0.0006764343343792022, + "loss": 0.4974, + "step": 11665 + }, + { + "epoch": 0.6517864625527279, + "grad_norm": 1.3570845127105713, + "learning_rate": 0.0006764063200358583, + "loss": 0.5303, + "step": 11666 + }, + { + "epoch": 0.6518423331564097, + "grad_norm": 0.7648836970329285, + "learning_rate": 0.0006763783056925145, + "loss": 0.6069, + "step": 11667 + }, + { + "epoch": 0.6518982037600917, + "grad_norm": 0.5305622816085815, + "learning_rate": 0.0006763502913491709, + "loss": 0.3881, + "step": 11668 + }, + { + "epoch": 0.6519540743637735, + "grad_norm": 7.333127021789551, + "learning_rate": 0.0006763222770058271, + "loss": 0.4788, + "step": 11669 + }, + { + "epoch": 0.6520099449674553, + "grad_norm": 0.6176650524139404, + "learning_rate": 0.0006762942626624832, + "loss": 0.4463, + "step": 11670 + }, + { + "epoch": 0.6520658155711373, + "grad_norm": 0.7940816879272461, + "learning_rate": 0.0006762662483191394, + "loss": 0.419, + "step": 11671 + }, + { + "epoch": 0.6521216861748191, + "grad_norm": 0.48387500643730164, + "learning_rate": 0.0006762382339757957, + "loss": 0.4268, + "step": 11672 + }, + { + "epoch": 0.652177556778501, + "grad_norm": 0.691193163394928, + "learning_rate": 0.0006762102196324519, + "loss": 0.6214, + "step": 11673 + }, + { + "epoch": 0.6522334273821828, + "grad_norm": 0.4615857005119324, + "learning_rate": 0.000676182205289108, + "loss": 0.5187, + "step": 11674 + }, + { + "epoch": 0.6522892979858648, + "grad_norm": 0.529464066028595, + "learning_rate": 0.0006761541909457643, + "loss": 0.4847, + "step": 11675 + }, + { + "epoch": 0.6523451685895466, + "grad_norm": 0.5702305436134338, + "learning_rate": 0.0006761261766024205, + "loss": 0.4638, + "step": 11676 + }, + { + "epoch": 0.6524010391932284, + "grad_norm": 0.8011796474456787, + "learning_rate": 0.0006760981622590766, + "loss": 0.409, + "step": 11677 + }, + { + "epoch": 0.6524569097969104, + "grad_norm": 0.5641700625419617, + "learning_rate": 0.0006760701479157328, + "loss": 0.3805, + "step": 11678 + }, + { + "epoch": 0.6525127804005922, + "grad_norm": 0.4438215494155884, + "learning_rate": 0.0006760421335723891, + "loss": 0.4394, + "step": 11679 + }, + { + "epoch": 0.6525686510042741, + "grad_norm": 0.48501941561698914, + "learning_rate": 0.0006760141192290453, + "loss": 0.4572, + "step": 11680 + }, + { + "epoch": 0.6526245216079559, + "grad_norm": 0.45566120743751526, + "learning_rate": 0.0006759861048857014, + "loss": 0.388, + "step": 11681 + }, + { + "epoch": 0.6526803922116379, + "grad_norm": 0.39829838275909424, + "learning_rate": 0.0006759580905423577, + "loss": 0.4394, + "step": 11682 + }, + { + "epoch": 0.6527362628153197, + "grad_norm": 2.700564384460449, + "learning_rate": 0.0006759300761990139, + "loss": 0.4093, + "step": 11683 + }, + { + "epoch": 0.6527921334190016, + "grad_norm": 0.9816147685050964, + "learning_rate": 0.0006759020618556701, + "loss": 0.4399, + "step": 11684 + }, + { + "epoch": 0.6528480040226835, + "grad_norm": 0.5896447896957397, + "learning_rate": 0.0006758740475123262, + "loss": 0.46, + "step": 11685 + }, + { + "epoch": 0.6529038746263653, + "grad_norm": 1.2870469093322754, + "learning_rate": 0.0006758460331689826, + "loss": 0.5056, + "step": 11686 + }, + { + "epoch": 0.6529597452300472, + "grad_norm": 0.8708796501159668, + "learning_rate": 0.0006758180188256388, + "loss": 0.4695, + "step": 11687 + }, + { + "epoch": 0.6530156158337291, + "grad_norm": 0.46143224835395813, + "learning_rate": 0.000675790004482295, + "loss": 0.4569, + "step": 11688 + }, + { + "epoch": 0.653071486437411, + "grad_norm": 0.5002076029777527, + "learning_rate": 0.0006757619901389512, + "loss": 0.4879, + "step": 11689 + }, + { + "epoch": 0.6531273570410928, + "grad_norm": 0.4184543192386627, + "learning_rate": 0.0006757339757956074, + "loss": 0.4407, + "step": 11690 + }, + { + "epoch": 0.6531832276447747, + "grad_norm": 0.8082404732704163, + "learning_rate": 0.0006757059614522636, + "loss": 0.5654, + "step": 11691 + }, + { + "epoch": 0.6532390982484566, + "grad_norm": 0.571424126625061, + "learning_rate": 0.0006756779471089198, + "loss": 0.4575, + "step": 11692 + }, + { + "epoch": 0.6532949688521384, + "grad_norm": 0.8449896574020386, + "learning_rate": 0.000675649932765576, + "loss": 0.4079, + "step": 11693 + }, + { + "epoch": 0.6533508394558203, + "grad_norm": 0.4990605115890503, + "learning_rate": 0.0006756219184222322, + "loss": 0.4209, + "step": 11694 + }, + { + "epoch": 0.6534067100595022, + "grad_norm": 0.547896683216095, + "learning_rate": 0.0006755939040788884, + "loss": 0.5022, + "step": 11695 + }, + { + "epoch": 0.6534625806631841, + "grad_norm": 0.49053430557250977, + "learning_rate": 0.0006755658897355445, + "loss": 0.4349, + "step": 11696 + }, + { + "epoch": 0.6535184512668659, + "grad_norm": 0.41171562671661377, + "learning_rate": 0.0006755378753922008, + "loss": 0.4407, + "step": 11697 + }, + { + "epoch": 0.6535743218705478, + "grad_norm": 0.5072189569473267, + "learning_rate": 0.000675509861048857, + "loss": 0.4468, + "step": 11698 + }, + { + "epoch": 0.6536301924742297, + "grad_norm": 0.46094003319740295, + "learning_rate": 0.0006754818467055132, + "loss": 0.4111, + "step": 11699 + }, + { + "epoch": 0.6536860630779115, + "grad_norm": 1.7285726070404053, + "learning_rate": 0.0006754538323621694, + "loss": 0.7276, + "step": 11700 + }, + { + "epoch": 0.6537419336815934, + "grad_norm": 0.39473220705986023, + "learning_rate": 0.0006754258180188256, + "loss": 0.449, + "step": 11701 + }, + { + "epoch": 0.6537978042852753, + "grad_norm": 0.5438697934150696, + "learning_rate": 0.0006753978036754818, + "loss": 0.4606, + "step": 11702 + }, + { + "epoch": 0.6538536748889572, + "grad_norm": 0.487091064453125, + "learning_rate": 0.000675369789332138, + "loss": 0.4332, + "step": 11703 + }, + { + "epoch": 0.653909545492639, + "grad_norm": 0.6394351720809937, + "learning_rate": 0.0006753417749887943, + "loss": 0.4848, + "step": 11704 + }, + { + "epoch": 0.653965416096321, + "grad_norm": 0.6302346587181091, + "learning_rate": 0.0006753137606454505, + "loss": 0.4135, + "step": 11705 + }, + { + "epoch": 0.6540212867000028, + "grad_norm": 0.42579385638237, + "learning_rate": 0.0006752857463021067, + "loss": 0.3946, + "step": 11706 + }, + { + "epoch": 0.6540771573036847, + "grad_norm": 0.30358991026878357, + "learning_rate": 0.000675257731958763, + "loss": 0.4263, + "step": 11707 + }, + { + "epoch": 0.6541330279073665, + "grad_norm": 0.4223328232765198, + "learning_rate": 0.0006752297176154191, + "loss": 0.46, + "step": 11708 + }, + { + "epoch": 0.6541888985110484, + "grad_norm": 0.40694767236709595, + "learning_rate": 0.0006752017032720753, + "loss": 0.5276, + "step": 11709 + }, + { + "epoch": 0.6542447691147303, + "grad_norm": 0.8477859497070312, + "learning_rate": 0.0006751736889287315, + "loss": 0.6008, + "step": 11710 + }, + { + "epoch": 0.6543006397184121, + "grad_norm": 0.44860538840293884, + "learning_rate": 0.0006751456745853878, + "loss": 0.4163, + "step": 11711 + }, + { + "epoch": 0.6543565103220941, + "grad_norm": 0.5230246782302856, + "learning_rate": 0.0006751176602420439, + "loss": 0.3884, + "step": 11712 + }, + { + "epoch": 0.6544123809257759, + "grad_norm": 0.8050486445426941, + "learning_rate": 0.0006750896458987001, + "loss": 0.4651, + "step": 11713 + }, + { + "epoch": 0.6544682515294578, + "grad_norm": 0.30042198300361633, + "learning_rate": 0.0006750616315553564, + "loss": 0.4244, + "step": 11714 + }, + { + "epoch": 0.6545241221331396, + "grad_norm": 0.44785743951797485, + "learning_rate": 0.0006750336172120126, + "loss": 0.4933, + "step": 11715 + }, + { + "epoch": 0.6545799927368215, + "grad_norm": 0.5601733326911926, + "learning_rate": 0.0006750056028686687, + "loss": 0.4364, + "step": 11716 + }, + { + "epoch": 0.6546358633405034, + "grad_norm": 0.4032447040081024, + "learning_rate": 0.0006749775885253249, + "loss": 0.3383, + "step": 11717 + }, + { + "epoch": 0.6546917339441852, + "grad_norm": 0.4820249676704407, + "learning_rate": 0.0006749495741819812, + "loss": 0.4377, + "step": 11718 + }, + { + "epoch": 0.6547476045478672, + "grad_norm": 0.5275340676307678, + "learning_rate": 0.0006749215598386373, + "loss": 0.5252, + "step": 11719 + }, + { + "epoch": 0.654803475151549, + "grad_norm": 0.4519187808036804, + "learning_rate": 0.0006748935454952936, + "loss": 0.3964, + "step": 11720 + }, + { + "epoch": 0.6548593457552309, + "grad_norm": 0.460341215133667, + "learning_rate": 0.0006748655311519499, + "loss": 0.4857, + "step": 11721 + }, + { + "epoch": 0.6549152163589128, + "grad_norm": 0.5621837377548218, + "learning_rate": 0.0006748375168086061, + "loss": 0.4055, + "step": 11722 + }, + { + "epoch": 0.6549710869625947, + "grad_norm": 0.6465786695480347, + "learning_rate": 0.0006748095024652622, + "loss": 0.7188, + "step": 11723 + }, + { + "epoch": 0.6550269575662765, + "grad_norm": 0.4590943455696106, + "learning_rate": 0.0006747814881219184, + "loss": 0.4251, + "step": 11724 + }, + { + "epoch": 0.6550828281699583, + "grad_norm": 0.42339396476745605, + "learning_rate": 0.0006747534737785747, + "loss": 0.4612, + "step": 11725 + }, + { + "epoch": 0.6551386987736403, + "grad_norm": 0.7108662128448486, + "learning_rate": 0.0006747254594352309, + "loss": 0.6634, + "step": 11726 + }, + { + "epoch": 0.6551945693773221, + "grad_norm": 0.6381366848945618, + "learning_rate": 0.000674697445091887, + "loss": 0.5536, + "step": 11727 + }, + { + "epoch": 0.655250439981004, + "grad_norm": 7.709049701690674, + "learning_rate": 0.0006746694307485433, + "loss": 0.2979, + "step": 11728 + }, + { + "epoch": 0.6553063105846859, + "grad_norm": 0.5209701061248779, + "learning_rate": 0.0006746414164051995, + "loss": 0.4304, + "step": 11729 + }, + { + "epoch": 0.6553621811883678, + "grad_norm": 1.1189898252487183, + "learning_rate": 0.0006746134020618557, + "loss": 0.5489, + "step": 11730 + }, + { + "epoch": 0.6554180517920496, + "grad_norm": 1.7607414722442627, + "learning_rate": 0.0006745853877185118, + "loss": 0.4768, + "step": 11731 + }, + { + "epoch": 0.6554739223957314, + "grad_norm": 0.4467327296733856, + "learning_rate": 0.0006745573733751681, + "loss": 0.4838, + "step": 11732 + }, + { + "epoch": 0.6555297929994134, + "grad_norm": 1.5688869953155518, + "learning_rate": 0.0006745293590318243, + "loss": 0.4538, + "step": 11733 + }, + { + "epoch": 0.6555856636030952, + "grad_norm": 0.828205943107605, + "learning_rate": 0.0006745013446884805, + "loss": 0.4309, + "step": 11734 + }, + { + "epoch": 0.6556415342067771, + "grad_norm": 0.5029181241989136, + "learning_rate": 0.0006744733303451366, + "loss": 0.3872, + "step": 11735 + }, + { + "epoch": 0.655697404810459, + "grad_norm": 0.8659115433692932, + "learning_rate": 0.000674445316001793, + "loss": 0.4789, + "step": 11736 + }, + { + "epoch": 0.6557532754141409, + "grad_norm": 0.40636900067329407, + "learning_rate": 0.0006744173016584492, + "loss": 0.4357, + "step": 11737 + }, + { + "epoch": 0.6558091460178227, + "grad_norm": 1.275384783744812, + "learning_rate": 0.0006743892873151053, + "loss": 0.409, + "step": 11738 + }, + { + "epoch": 0.6558650166215046, + "grad_norm": 1.0581921339035034, + "learning_rate": 0.0006743612729717616, + "loss": 0.4375, + "step": 11739 + }, + { + "epoch": 0.6559208872251865, + "grad_norm": 0.8515260219573975, + "learning_rate": 0.0006743332586284178, + "loss": 0.5461, + "step": 11740 + }, + { + "epoch": 0.6559767578288683, + "grad_norm": 0.33981460332870483, + "learning_rate": 0.000674305244285074, + "loss": 0.3968, + "step": 11741 + }, + { + "epoch": 0.6560326284325502, + "grad_norm": 0.41219857335090637, + "learning_rate": 0.0006742772299417301, + "loss": 0.505, + "step": 11742 + }, + { + "epoch": 0.6560884990362321, + "grad_norm": 0.5041130185127258, + "learning_rate": 0.0006742492155983864, + "loss": 0.4173, + "step": 11743 + }, + { + "epoch": 0.656144369639914, + "grad_norm": 0.47870057821273804, + "learning_rate": 0.0006742212012550426, + "loss": 0.4985, + "step": 11744 + }, + { + "epoch": 0.6562002402435958, + "grad_norm": 0.46390220522880554, + "learning_rate": 0.0006741931869116988, + "loss": 0.3953, + "step": 11745 + }, + { + "epoch": 0.6562561108472778, + "grad_norm": 0.43491312861442566, + "learning_rate": 0.000674165172568355, + "loss": 0.4497, + "step": 11746 + }, + { + "epoch": 0.6563119814509596, + "grad_norm": 0.5778117775917053, + "learning_rate": 0.0006741371582250112, + "loss": 0.4961, + "step": 11747 + }, + { + "epoch": 0.6563678520546414, + "grad_norm": 0.7531816959381104, + "learning_rate": 0.0006741091438816674, + "loss": 0.5537, + "step": 11748 + }, + { + "epoch": 0.6564237226583233, + "grad_norm": 0.4223712980747223, + "learning_rate": 0.0006740811295383236, + "loss": 0.3713, + "step": 11749 + }, + { + "epoch": 0.6564795932620052, + "grad_norm": 0.6083646416664124, + "learning_rate": 0.0006740531151949798, + "loss": 0.4096, + "step": 11750 + }, + { + "epoch": 0.6565354638656871, + "grad_norm": 0.7185924053192139, + "learning_rate": 0.000674025100851636, + "loss": 0.625, + "step": 11751 + }, + { + "epoch": 0.6565913344693689, + "grad_norm": 0.5438894033432007, + "learning_rate": 0.0006739970865082922, + "loss": 0.434, + "step": 11752 + }, + { + "epoch": 0.6566472050730509, + "grad_norm": 2.8954379558563232, + "learning_rate": 0.0006739690721649486, + "loss": 0.4552, + "step": 11753 + }, + { + "epoch": 0.6567030756767327, + "grad_norm": 16.053600311279297, + "learning_rate": 0.0006739410578216047, + "loss": 0.5365, + "step": 11754 + }, + { + "epoch": 0.6567589462804145, + "grad_norm": 1.0137635469436646, + "learning_rate": 0.0006739130434782609, + "loss": 0.5111, + "step": 11755 + }, + { + "epoch": 0.6568148168840965, + "grad_norm": 0.4842285215854645, + "learning_rate": 0.0006738850291349171, + "loss": 0.4254, + "step": 11756 + }, + { + "epoch": 0.6568706874877783, + "grad_norm": 0.4265272915363312, + "learning_rate": 0.0006738570147915734, + "loss": 0.3705, + "step": 11757 + }, + { + "epoch": 0.6569265580914602, + "grad_norm": 2.1332855224609375, + "learning_rate": 0.0006738290004482295, + "loss": 0.4594, + "step": 11758 + }, + { + "epoch": 0.656982428695142, + "grad_norm": 0.6029937863349915, + "learning_rate": 0.0006738009861048857, + "loss": 0.4481, + "step": 11759 + }, + { + "epoch": 0.657038299298824, + "grad_norm": 0.40434303879737854, + "learning_rate": 0.000673772971761542, + "loss": 0.4082, + "step": 11760 + }, + { + "epoch": 0.6570941699025058, + "grad_norm": 0.45240846276283264, + "learning_rate": 0.0006737449574181981, + "loss": 0.4372, + "step": 11761 + }, + { + "epoch": 0.6571500405061876, + "grad_norm": 0.5713226199150085, + "learning_rate": 0.0006737169430748543, + "loss": 0.441, + "step": 11762 + }, + { + "epoch": 0.6572059111098696, + "grad_norm": 0.5519190430641174, + "learning_rate": 0.0006736889287315105, + "loss": 0.5189, + "step": 11763 + }, + { + "epoch": 0.6572617817135514, + "grad_norm": 0.5453360080718994, + "learning_rate": 0.0006736609143881668, + "loss": 0.3617, + "step": 11764 + }, + { + "epoch": 0.6573176523172333, + "grad_norm": 1.2162230014801025, + "learning_rate": 0.0006736329000448229, + "loss": 0.571, + "step": 11765 + }, + { + "epoch": 0.6573735229209151, + "grad_norm": 0.4584588408470154, + "learning_rate": 0.0006736048857014791, + "loss": 0.4702, + "step": 11766 + }, + { + "epoch": 0.6574293935245971, + "grad_norm": 0.7566846609115601, + "learning_rate": 0.0006735768713581354, + "loss": 0.4549, + "step": 11767 + }, + { + "epoch": 0.6574852641282789, + "grad_norm": 0.5044066309928894, + "learning_rate": 0.0006735488570147916, + "loss": 0.5217, + "step": 11768 + }, + { + "epoch": 0.6575411347319607, + "grad_norm": 1.2820409536361694, + "learning_rate": 0.0006735208426714477, + "loss": 0.5096, + "step": 11769 + }, + { + "epoch": 0.6575970053356427, + "grad_norm": 2.4694125652313232, + "learning_rate": 0.000673492828328104, + "loss": 0.4876, + "step": 11770 + }, + { + "epoch": 0.6576528759393245, + "grad_norm": 0.6453713178634644, + "learning_rate": 0.0006734648139847603, + "loss": 0.5344, + "step": 11771 + }, + { + "epoch": 0.6577087465430064, + "grad_norm": 0.5858170986175537, + "learning_rate": 0.0006734367996414165, + "loss": 0.5343, + "step": 11772 + }, + { + "epoch": 0.6577646171466883, + "grad_norm": 0.6531484127044678, + "learning_rate": 0.0006734087852980726, + "loss": 0.4471, + "step": 11773 + }, + { + "epoch": 0.6578204877503702, + "grad_norm": 0.963578462600708, + "learning_rate": 0.0006733807709547288, + "loss": 0.4472, + "step": 11774 + }, + { + "epoch": 0.657876358354052, + "grad_norm": 0.647685170173645, + "learning_rate": 0.0006733527566113851, + "loss": 0.4873, + "step": 11775 + }, + { + "epoch": 0.6579322289577338, + "grad_norm": 1.1863338947296143, + "learning_rate": 0.0006733247422680413, + "loss": 0.4469, + "step": 11776 + }, + { + "epoch": 0.6579880995614158, + "grad_norm": 0.39732134342193604, + "learning_rate": 0.0006732967279246974, + "loss": 0.3727, + "step": 11777 + }, + { + "epoch": 0.6580439701650976, + "grad_norm": 0.44603317975997925, + "learning_rate": 0.0006732687135813537, + "loss": 0.4693, + "step": 11778 + }, + { + "epoch": 0.6580998407687795, + "grad_norm": 2.5758180618286133, + "learning_rate": 0.0006732406992380099, + "loss": 0.4673, + "step": 11779 + }, + { + "epoch": 0.6581557113724614, + "grad_norm": 0.5336591601371765, + "learning_rate": 0.000673212684894666, + "loss": 0.3982, + "step": 11780 + }, + { + "epoch": 0.6582115819761433, + "grad_norm": 0.8727526068687439, + "learning_rate": 0.0006731846705513222, + "loss": 0.476, + "step": 11781 + }, + { + "epoch": 0.6582674525798251, + "grad_norm": 3.773280620574951, + "learning_rate": 0.0006731566562079785, + "loss": 0.5065, + "step": 11782 + }, + { + "epoch": 0.658323323183507, + "grad_norm": 1.3426311016082764, + "learning_rate": 0.0006731286418646347, + "loss": 0.5314, + "step": 11783 + }, + { + "epoch": 0.6583791937871889, + "grad_norm": 0.4700508117675781, + "learning_rate": 0.0006731006275212908, + "loss": 0.4506, + "step": 11784 + }, + { + "epoch": 0.6584350643908707, + "grad_norm": 0.5549445748329163, + "learning_rate": 0.0006730726131779471, + "loss": 0.4036, + "step": 11785 + }, + { + "epoch": 0.6584909349945526, + "grad_norm": 0.4769854247570038, + "learning_rate": 0.0006730445988346034, + "loss": 0.51, + "step": 11786 + }, + { + "epoch": 0.6585468055982345, + "grad_norm": 0.36181512475013733, + "learning_rate": 0.0006730165844912596, + "loss": 0.4672, + "step": 11787 + }, + { + "epoch": 0.6586026762019164, + "grad_norm": 0.4538766145706177, + "learning_rate": 0.0006729885701479157, + "loss": 0.4347, + "step": 11788 + }, + { + "epoch": 0.6586585468055982, + "grad_norm": 0.42258280515670776, + "learning_rate": 0.000672960555804572, + "loss": 0.4311, + "step": 11789 + }, + { + "epoch": 0.6587144174092802, + "grad_norm": 0.4768436849117279, + "learning_rate": 0.0006729325414612282, + "loss": 0.4643, + "step": 11790 + }, + { + "epoch": 0.658770288012962, + "grad_norm": 0.5184430480003357, + "learning_rate": 0.0006729045271178844, + "loss": 0.5837, + "step": 11791 + }, + { + "epoch": 0.6588261586166438, + "grad_norm": 0.4682547152042389, + "learning_rate": 0.0006728765127745406, + "loss": 0.5087, + "step": 11792 + }, + { + "epoch": 0.6588820292203257, + "grad_norm": 0.4459194839000702, + "learning_rate": 0.0006728484984311968, + "loss": 0.405, + "step": 11793 + }, + { + "epoch": 0.6589378998240076, + "grad_norm": 0.4544607102870941, + "learning_rate": 0.000672820484087853, + "loss": 0.3499, + "step": 11794 + }, + { + "epoch": 0.6589937704276895, + "grad_norm": 0.4842791259288788, + "learning_rate": 0.0006727924697445092, + "loss": 0.381, + "step": 11795 + }, + { + "epoch": 0.6590496410313713, + "grad_norm": 0.6901815533638, + "learning_rate": 0.0006727644554011654, + "loss": 0.3534, + "step": 11796 + }, + { + "epoch": 0.6591055116350533, + "grad_norm": 0.5284761190414429, + "learning_rate": 0.0006727364410578216, + "loss": 0.4428, + "step": 11797 + }, + { + "epoch": 0.6591613822387351, + "grad_norm": 0.8288255929946899, + "learning_rate": 0.0006727084267144778, + "loss": 0.5354, + "step": 11798 + }, + { + "epoch": 0.659217252842417, + "grad_norm": 1.0844848155975342, + "learning_rate": 0.0006726804123711341, + "loss": 0.4804, + "step": 11799 + }, + { + "epoch": 0.6592731234460988, + "grad_norm": 0.8856677412986755, + "learning_rate": 0.0006726523980277902, + "loss": 0.7297, + "step": 11800 + }, + { + "epoch": 0.6593289940497807, + "grad_norm": 0.7067208886146545, + "learning_rate": 0.0006726243836844464, + "loss": 0.4097, + "step": 11801 + }, + { + "epoch": 0.6593848646534626, + "grad_norm": 0.7555345892906189, + "learning_rate": 0.0006725963693411026, + "loss": 0.4721, + "step": 11802 + }, + { + "epoch": 0.6594407352571444, + "grad_norm": 1.621207594871521, + "learning_rate": 0.0006725683549977589, + "loss": 0.4997, + "step": 11803 + }, + { + "epoch": 0.6594966058608264, + "grad_norm": 9.852027893066406, + "learning_rate": 0.0006725403406544151, + "loss": 0.5942, + "step": 11804 + }, + { + "epoch": 0.6595524764645082, + "grad_norm": 0.7829749584197998, + "learning_rate": 0.0006725123263110713, + "loss": 0.4406, + "step": 11805 + }, + { + "epoch": 0.65960834706819, + "grad_norm": 0.6866810917854309, + "learning_rate": 0.0006724843119677276, + "loss": 0.3585, + "step": 11806 + }, + { + "epoch": 0.659664217671872, + "grad_norm": 2.9128754138946533, + "learning_rate": 0.0006724562976243837, + "loss": 0.5392, + "step": 11807 + }, + { + "epoch": 0.6597200882755538, + "grad_norm": 0.6439321041107178, + "learning_rate": 0.0006724282832810399, + "loss": 0.4875, + "step": 11808 + }, + { + "epoch": 0.6597759588792357, + "grad_norm": 0.4871627986431122, + "learning_rate": 0.0006724002689376961, + "loss": 0.6072, + "step": 11809 + }, + { + "epoch": 0.6598318294829175, + "grad_norm": 0.8324915170669556, + "learning_rate": 0.0006723722545943524, + "loss": 0.5473, + "step": 11810 + }, + { + "epoch": 0.6598877000865995, + "grad_norm": 0.584287166595459, + "learning_rate": 0.0006723442402510085, + "loss": 0.6835, + "step": 11811 + }, + { + "epoch": 0.6599435706902813, + "grad_norm": 0.7388282418251038, + "learning_rate": 0.0006723162259076647, + "loss": 0.5455, + "step": 11812 + }, + { + "epoch": 0.6599994412939632, + "grad_norm": 0.5744937658309937, + "learning_rate": 0.0006722882115643209, + "loss": 0.4503, + "step": 11813 + }, + { + "epoch": 0.6600553118976451, + "grad_norm": 0.6059413552284241, + "learning_rate": 0.0006722601972209772, + "loss": 0.4483, + "step": 11814 + }, + { + "epoch": 0.6601111825013269, + "grad_norm": 0.6324524879455566, + "learning_rate": 0.0006722321828776333, + "loss": 0.512, + "step": 11815 + }, + { + "epoch": 0.6601670531050088, + "grad_norm": 0.5293042659759521, + "learning_rate": 0.0006722041685342895, + "loss": 0.5598, + "step": 11816 + }, + { + "epoch": 0.6602229237086906, + "grad_norm": 0.6966378092765808, + "learning_rate": 0.0006721761541909458, + "loss": 0.5193, + "step": 11817 + }, + { + "epoch": 0.6602787943123726, + "grad_norm": 6.236052513122559, + "learning_rate": 0.000672148139847602, + "loss": 0.4549, + "step": 11818 + }, + { + "epoch": 0.6603346649160544, + "grad_norm": 0.7694763541221619, + "learning_rate": 0.0006721201255042581, + "loss": 0.4514, + "step": 11819 + }, + { + "epoch": 0.6603905355197363, + "grad_norm": 1.1963673830032349, + "learning_rate": 0.0006720921111609144, + "loss": 0.4567, + "step": 11820 + }, + { + "epoch": 0.6604464061234182, + "grad_norm": 0.8687496781349182, + "learning_rate": 0.0006720640968175707, + "loss": 0.4826, + "step": 11821 + }, + { + "epoch": 0.6605022767271, + "grad_norm": 0.5795978903770447, + "learning_rate": 0.0006720360824742269, + "loss": 0.3923, + "step": 11822 + }, + { + "epoch": 0.6605581473307819, + "grad_norm": 0.5040907859802246, + "learning_rate": 0.000672008068130883, + "loss": 0.5383, + "step": 11823 + }, + { + "epoch": 0.6606140179344637, + "grad_norm": 1.2145497798919678, + "learning_rate": 0.0006719800537875393, + "loss": 0.4113, + "step": 11824 + }, + { + "epoch": 0.6606698885381457, + "grad_norm": 0.4929317235946655, + "learning_rate": 0.0006719520394441955, + "loss": 0.383, + "step": 11825 + }, + { + "epoch": 0.6607257591418275, + "grad_norm": 0.5714660882949829, + "learning_rate": 0.0006719240251008516, + "loss": 0.4276, + "step": 11826 + }, + { + "epoch": 0.6607816297455094, + "grad_norm": 0.4131568968296051, + "learning_rate": 0.0006718960107575078, + "loss": 0.4215, + "step": 11827 + }, + { + "epoch": 0.6608375003491913, + "grad_norm": 1.4825621843338013, + "learning_rate": 0.0006718679964141641, + "loss": 0.7044, + "step": 11828 + }, + { + "epoch": 0.6608933709528731, + "grad_norm": 0.4202149510383606, + "learning_rate": 0.0006718399820708203, + "loss": 0.4031, + "step": 11829 + }, + { + "epoch": 0.660949241556555, + "grad_norm": 0.43239226937294006, + "learning_rate": 0.0006718119677274764, + "loss": 0.47, + "step": 11830 + }, + { + "epoch": 0.6610051121602369, + "grad_norm": 0.4283757507801056, + "learning_rate": 0.0006717839533841327, + "loss": 0.3837, + "step": 11831 + }, + { + "epoch": 0.6610609827639188, + "grad_norm": 0.5139572620391846, + "learning_rate": 0.0006717559390407889, + "loss": 0.3434, + "step": 11832 + }, + { + "epoch": 0.6611168533676006, + "grad_norm": 0.8711274862289429, + "learning_rate": 0.0006717279246974451, + "loss": 0.5298, + "step": 11833 + }, + { + "epoch": 0.6611727239712825, + "grad_norm": 0.5921872854232788, + "learning_rate": 0.0006716999103541012, + "loss": 0.4661, + "step": 11834 + }, + { + "epoch": 0.6612285945749644, + "grad_norm": 6.397389888763428, + "learning_rate": 0.0006716718960107575, + "loss": 0.3734, + "step": 11835 + }, + { + "epoch": 0.6612844651786463, + "grad_norm": 0.4228278696537018, + "learning_rate": 0.0006716438816674138, + "loss": 0.3665, + "step": 11836 + }, + { + "epoch": 0.6613403357823281, + "grad_norm": 0.6316782832145691, + "learning_rate": 0.00067161586732407, + "loss": 0.595, + "step": 11837 + }, + { + "epoch": 0.66139620638601, + "grad_norm": 1.2052218914031982, + "learning_rate": 0.0006715878529807262, + "loss": 0.4572, + "step": 11838 + }, + { + "epoch": 0.6614520769896919, + "grad_norm": 0.5019708871841431, + "learning_rate": 0.0006715598386373824, + "loss": 0.4422, + "step": 11839 + }, + { + "epoch": 0.6615079475933737, + "grad_norm": 1.0087063312530518, + "learning_rate": 0.0006715318242940386, + "loss": 0.5924, + "step": 11840 + }, + { + "epoch": 0.6615638181970556, + "grad_norm": 0.44480767846107483, + "learning_rate": 0.0006715038099506948, + "loss": 0.424, + "step": 11841 + }, + { + "epoch": 0.6616196888007375, + "grad_norm": 1.0693864822387695, + "learning_rate": 0.000671475795607351, + "loss": 0.4642, + "step": 11842 + }, + { + "epoch": 0.6616755594044194, + "grad_norm": 0.7961464524269104, + "learning_rate": 0.0006714477812640072, + "loss": 0.4875, + "step": 11843 + }, + { + "epoch": 0.6617314300081012, + "grad_norm": 0.5273467302322388, + "learning_rate": 0.0006714197669206634, + "loss": 0.5084, + "step": 11844 + }, + { + "epoch": 0.6617873006117831, + "grad_norm": 0.4064296782016754, + "learning_rate": 0.0006713917525773195, + "loss": 0.4019, + "step": 11845 + }, + { + "epoch": 0.661843171215465, + "grad_norm": 0.37351304292678833, + "learning_rate": 0.0006713637382339758, + "loss": 0.3888, + "step": 11846 + }, + { + "epoch": 0.6618990418191468, + "grad_norm": 0.396330863237381, + "learning_rate": 0.000671335723890632, + "loss": 0.3667, + "step": 11847 + }, + { + "epoch": 0.6619549124228288, + "grad_norm": 0.8460236191749573, + "learning_rate": 0.0006713077095472882, + "loss": 0.451, + "step": 11848 + }, + { + "epoch": 0.6620107830265106, + "grad_norm": 2.5272068977355957, + "learning_rate": 0.0006712796952039444, + "loss": 0.4119, + "step": 11849 + }, + { + "epoch": 0.6620666536301925, + "grad_norm": 0.7062801122665405, + "learning_rate": 0.0006712516808606006, + "loss": 0.5192, + "step": 11850 + }, + { + "epoch": 0.6621225242338743, + "grad_norm": 0.9363716840744019, + "learning_rate": 0.0006712236665172568, + "loss": 0.4229, + "step": 11851 + }, + { + "epoch": 0.6621783948375563, + "grad_norm": 0.45732787251472473, + "learning_rate": 0.000671195652173913, + "loss": 0.4242, + "step": 11852 + }, + { + "epoch": 0.6622342654412381, + "grad_norm": 0.4349896311759949, + "learning_rate": 0.0006711676378305693, + "loss": 0.4028, + "step": 11853 + }, + { + "epoch": 0.6622901360449199, + "grad_norm": 0.5854281187057495, + "learning_rate": 0.0006711396234872255, + "loss": 0.494, + "step": 11854 + }, + { + "epoch": 0.6623460066486019, + "grad_norm": 0.5321286916732788, + "learning_rate": 0.0006711116091438817, + "loss": 0.4016, + "step": 11855 + }, + { + "epoch": 0.6624018772522837, + "grad_norm": 2.325442314147949, + "learning_rate": 0.000671083594800538, + "loss": 0.5134, + "step": 11856 + }, + { + "epoch": 0.6624577478559656, + "grad_norm": 0.4739091694355011, + "learning_rate": 0.0006710555804571941, + "loss": 0.4108, + "step": 11857 + }, + { + "epoch": 0.6625136184596474, + "grad_norm": 2.248377561569214, + "learning_rate": 0.0006710275661138503, + "loss": 0.5097, + "step": 11858 + }, + { + "epoch": 0.6625694890633294, + "grad_norm": 0.6182782053947449, + "learning_rate": 0.0006709995517705065, + "loss": 0.4182, + "step": 11859 + }, + { + "epoch": 0.6626253596670112, + "grad_norm": 0.69984370470047, + "learning_rate": 0.0006709715374271628, + "loss": 0.4646, + "step": 11860 + }, + { + "epoch": 0.662681230270693, + "grad_norm": 0.4461551010608673, + "learning_rate": 0.0006709435230838189, + "loss": 0.4326, + "step": 11861 + }, + { + "epoch": 0.662737100874375, + "grad_norm": 0.4451746344566345, + "learning_rate": 0.0006709155087404751, + "loss": 0.4046, + "step": 11862 + }, + { + "epoch": 0.6627929714780568, + "grad_norm": 2.4580509662628174, + "learning_rate": 0.0006708874943971314, + "loss": 0.484, + "step": 11863 + }, + { + "epoch": 0.6628488420817387, + "grad_norm": 4.361410140991211, + "learning_rate": 0.0006708594800537876, + "loss": 0.4107, + "step": 11864 + }, + { + "epoch": 0.6629047126854206, + "grad_norm": 0.40710926055908203, + "learning_rate": 0.0006708314657104437, + "loss": 0.4204, + "step": 11865 + }, + { + "epoch": 0.6629605832891025, + "grad_norm": 0.630128800868988, + "learning_rate": 0.0006708034513670999, + "loss": 0.4354, + "step": 11866 + }, + { + "epoch": 0.6630164538927843, + "grad_norm": 0.6690042614936829, + "learning_rate": 0.0006707754370237562, + "loss": 0.3853, + "step": 11867 + }, + { + "epoch": 0.6630723244964661, + "grad_norm": 0.4581613540649414, + "learning_rate": 0.0006707474226804123, + "loss": 0.5155, + "step": 11868 + }, + { + "epoch": 0.6631281951001481, + "grad_norm": 1.836388111114502, + "learning_rate": 0.0006707194083370685, + "loss": 0.5607, + "step": 11869 + }, + { + "epoch": 0.6631840657038299, + "grad_norm": 0.604332685470581, + "learning_rate": 0.0006706913939937249, + "loss": 0.4202, + "step": 11870 + }, + { + "epoch": 0.6632399363075118, + "grad_norm": 0.48104244470596313, + "learning_rate": 0.0006706633796503811, + "loss": 0.4997, + "step": 11871 + }, + { + "epoch": 0.6632958069111937, + "grad_norm": 1.0321277379989624, + "learning_rate": 0.0006706353653070372, + "loss": 0.5746, + "step": 11872 + }, + { + "epoch": 0.6633516775148756, + "grad_norm": 21.718414306640625, + "learning_rate": 0.0006706073509636934, + "loss": 0.4569, + "step": 11873 + }, + { + "epoch": 0.6634075481185574, + "grad_norm": 1.3584234714508057, + "learning_rate": 0.0006705793366203497, + "loss": 0.4127, + "step": 11874 + }, + { + "epoch": 0.6634634187222392, + "grad_norm": 0.946448802947998, + "learning_rate": 0.0006705513222770059, + "loss": 0.6003, + "step": 11875 + }, + { + "epoch": 0.6635192893259212, + "grad_norm": 0.703381359577179, + "learning_rate": 0.000670523307933662, + "loss": 0.4934, + "step": 11876 + }, + { + "epoch": 0.663575159929603, + "grad_norm": 0.6596462726593018, + "learning_rate": 0.0006704952935903182, + "loss": 0.433, + "step": 11877 + }, + { + "epoch": 0.6636310305332849, + "grad_norm": 0.37781888246536255, + "learning_rate": 0.0006704672792469745, + "loss": 0.4017, + "step": 11878 + }, + { + "epoch": 0.6636869011369668, + "grad_norm": 0.4708905816078186, + "learning_rate": 0.0006704392649036307, + "loss": 0.4645, + "step": 11879 + }, + { + "epoch": 0.6637427717406487, + "grad_norm": 0.7120184898376465, + "learning_rate": 0.0006704112505602868, + "loss": 0.4806, + "step": 11880 + }, + { + "epoch": 0.6637986423443305, + "grad_norm": 0.5566036701202393, + "learning_rate": 0.0006703832362169431, + "loss": 0.5281, + "step": 11881 + }, + { + "epoch": 0.6638545129480125, + "grad_norm": 0.5441855788230896, + "learning_rate": 0.0006703552218735993, + "loss": 0.2924, + "step": 11882 + }, + { + "epoch": 0.6639103835516943, + "grad_norm": 0.5368353128433228, + "learning_rate": 0.0006703272075302555, + "loss": 0.419, + "step": 11883 + }, + { + "epoch": 0.6639662541553761, + "grad_norm": 0.508298397064209, + "learning_rate": 0.0006702991931869116, + "loss": 0.4694, + "step": 11884 + }, + { + "epoch": 0.664022124759058, + "grad_norm": 0.43259260058403015, + "learning_rate": 0.000670271178843568, + "loss": 0.3652, + "step": 11885 + }, + { + "epoch": 0.6640779953627399, + "grad_norm": 0.5383428931236267, + "learning_rate": 0.0006702431645002242, + "loss": 0.4938, + "step": 11886 + }, + { + "epoch": 0.6641338659664218, + "grad_norm": 0.5450477600097656, + "learning_rate": 0.0006702151501568802, + "loss": 0.4477, + "step": 11887 + }, + { + "epoch": 0.6641897365701036, + "grad_norm": 0.582770049571991, + "learning_rate": 0.0006701871358135366, + "loss": 0.4087, + "step": 11888 + }, + { + "epoch": 0.6642456071737856, + "grad_norm": 0.6235706210136414, + "learning_rate": 0.0006701591214701928, + "loss": 0.5338, + "step": 11889 + }, + { + "epoch": 0.6643014777774674, + "grad_norm": 1.0143810510635376, + "learning_rate": 0.000670131107126849, + "loss": 0.4541, + "step": 11890 + }, + { + "epoch": 0.6643573483811492, + "grad_norm": 0.5028754472732544, + "learning_rate": 0.0006701030927835051, + "loss": 0.4952, + "step": 11891 + }, + { + "epoch": 0.6644132189848311, + "grad_norm": 0.3932006359100342, + "learning_rate": 0.0006700750784401614, + "loss": 0.3924, + "step": 11892 + }, + { + "epoch": 0.664469089588513, + "grad_norm": 3.81634259223938, + "learning_rate": 0.0006700470640968176, + "loss": 0.5246, + "step": 11893 + }, + { + "epoch": 0.6645249601921949, + "grad_norm": 0.7755728960037231, + "learning_rate": 0.0006700190497534738, + "loss": 0.4577, + "step": 11894 + }, + { + "epoch": 0.6645808307958767, + "grad_norm": 0.5026862621307373, + "learning_rate": 0.00066999103541013, + "loss": 0.4656, + "step": 11895 + }, + { + "epoch": 0.6646367013995587, + "grad_norm": 1.086047887802124, + "learning_rate": 0.0006699630210667862, + "loss": 0.4507, + "step": 11896 + }, + { + "epoch": 0.6646925720032405, + "grad_norm": 1.931666374206543, + "learning_rate": 0.0006699350067234424, + "loss": 0.5646, + "step": 11897 + }, + { + "epoch": 0.6647484426069223, + "grad_norm": 0.49103766679763794, + "learning_rate": 0.0006699069923800986, + "loss": 0.3336, + "step": 11898 + }, + { + "epoch": 0.6648043132106043, + "grad_norm": 0.4837893843650818, + "learning_rate": 0.0006698789780367548, + "loss": 0.5218, + "step": 11899 + }, + { + "epoch": 0.6648601838142861, + "grad_norm": 0.6386026740074158, + "learning_rate": 0.000669850963693411, + "loss": 0.4235, + "step": 11900 + }, + { + "epoch": 0.664916054417968, + "grad_norm": 0.4180901348590851, + "learning_rate": 0.0006698229493500672, + "loss": 0.5227, + "step": 11901 + }, + { + "epoch": 0.6649719250216498, + "grad_norm": 1.1794902086257935, + "learning_rate": 0.0006697949350067236, + "loss": 0.4744, + "step": 11902 + }, + { + "epoch": 0.6650277956253318, + "grad_norm": 0.4242669641971588, + "learning_rate": 0.0006697669206633797, + "loss": 0.4402, + "step": 11903 + }, + { + "epoch": 0.6650836662290136, + "grad_norm": 0.412875771522522, + "learning_rate": 0.0006697389063200359, + "loss": 0.4151, + "step": 11904 + }, + { + "epoch": 0.6651395368326954, + "grad_norm": 0.46977564692497253, + "learning_rate": 0.0006697108919766921, + "loss": 0.4756, + "step": 11905 + }, + { + "epoch": 0.6651954074363774, + "grad_norm": 0.721066415309906, + "learning_rate": 0.0006696828776333484, + "loss": 0.5621, + "step": 11906 + }, + { + "epoch": 0.6652512780400592, + "grad_norm": 0.5401799082756042, + "learning_rate": 0.0006696548632900045, + "loss": 0.5285, + "step": 11907 + }, + { + "epoch": 0.6653071486437411, + "grad_norm": 2.2176625728607178, + "learning_rate": 0.0006696268489466607, + "loss": 0.3974, + "step": 11908 + }, + { + "epoch": 0.6653630192474229, + "grad_norm": 0.4804267883300781, + "learning_rate": 0.000669598834603317, + "loss": 0.4979, + "step": 11909 + }, + { + "epoch": 0.6654188898511049, + "grad_norm": 0.6513219475746155, + "learning_rate": 0.0006695708202599731, + "loss": 0.4901, + "step": 11910 + }, + { + "epoch": 0.6654747604547867, + "grad_norm": 0.5356496572494507, + "learning_rate": 0.0006695428059166293, + "loss": 0.5596, + "step": 11911 + }, + { + "epoch": 0.6655306310584685, + "grad_norm": 0.5836079120635986, + "learning_rate": 0.0006695147915732855, + "loss": 0.6737, + "step": 11912 + }, + { + "epoch": 0.6655865016621505, + "grad_norm": 3.1835527420043945, + "learning_rate": 0.0006694867772299418, + "loss": 0.5407, + "step": 11913 + }, + { + "epoch": 0.6656423722658323, + "grad_norm": 1.0737193822860718, + "learning_rate": 0.0006694587628865979, + "loss": 0.4326, + "step": 11914 + }, + { + "epoch": 0.6656982428695142, + "grad_norm": 0.5350149273872375, + "learning_rate": 0.0006694307485432541, + "loss": 0.4733, + "step": 11915 + }, + { + "epoch": 0.6657541134731961, + "grad_norm": 0.515341579914093, + "learning_rate": 0.0006694027341999103, + "loss": 0.4255, + "step": 11916 + }, + { + "epoch": 0.665809984076878, + "grad_norm": 0.5157885551452637, + "learning_rate": 0.0006693747198565666, + "loss": 0.4791, + "step": 11917 + }, + { + "epoch": 0.6658658546805598, + "grad_norm": 0.8459282517433167, + "learning_rate": 0.0006693467055132227, + "loss": 0.4665, + "step": 11918 + }, + { + "epoch": 0.6659217252842417, + "grad_norm": 0.4492494463920593, + "learning_rate": 0.0006693186911698789, + "loss": 0.4155, + "step": 11919 + }, + { + "epoch": 0.6659775958879236, + "grad_norm": 0.9152358770370483, + "learning_rate": 0.0006692906768265353, + "loss": 0.378, + "step": 11920 + }, + { + "epoch": 0.6660334664916054, + "grad_norm": 3.304091453552246, + "learning_rate": 0.0006692626624831915, + "loss": 0.4823, + "step": 11921 + }, + { + "epoch": 0.6660893370952873, + "grad_norm": 0.5508351922035217, + "learning_rate": 0.0006692346481398476, + "loss": 0.4116, + "step": 11922 + }, + { + "epoch": 0.6661452076989692, + "grad_norm": 0.33653154969215393, + "learning_rate": 0.0006692066337965038, + "loss": 0.3503, + "step": 11923 + }, + { + "epoch": 0.6662010783026511, + "grad_norm": 0.598733127117157, + "learning_rate": 0.0006691786194531601, + "loss": 0.4896, + "step": 11924 + }, + { + "epoch": 0.6662569489063329, + "grad_norm": 6.215941429138184, + "learning_rate": 0.0006691506051098163, + "loss": 0.5394, + "step": 11925 + }, + { + "epoch": 0.6663128195100148, + "grad_norm": 0.5925695300102234, + "learning_rate": 0.0006691225907664724, + "loss": 0.4023, + "step": 11926 + }, + { + "epoch": 0.6663686901136967, + "grad_norm": 0.4429542124271393, + "learning_rate": 0.0006690945764231287, + "loss": 0.4321, + "step": 11927 + }, + { + "epoch": 0.6664245607173785, + "grad_norm": 0.4263897240161896, + "learning_rate": 0.0006690665620797849, + "loss": 0.3738, + "step": 11928 + }, + { + "epoch": 0.6664804313210604, + "grad_norm": 0.5590676069259644, + "learning_rate": 0.000669038547736441, + "loss": 0.4069, + "step": 11929 + }, + { + "epoch": 0.6665363019247423, + "grad_norm": 0.3989647328853607, + "learning_rate": 0.0006690105333930972, + "loss": 0.4039, + "step": 11930 + }, + { + "epoch": 0.6665921725284242, + "grad_norm": 0.5598880052566528, + "learning_rate": 0.0006689825190497535, + "loss": 0.5177, + "step": 11931 + }, + { + "epoch": 0.666648043132106, + "grad_norm": 0.5538251996040344, + "learning_rate": 0.0006689545047064097, + "loss": 0.5125, + "step": 11932 + }, + { + "epoch": 0.666703913735788, + "grad_norm": 0.7919104695320129, + "learning_rate": 0.0006689264903630658, + "loss": 0.5888, + "step": 11933 + }, + { + "epoch": 0.6667597843394698, + "grad_norm": 0.41892197728157043, + "learning_rate": 0.0006688984760197221, + "loss": 0.4504, + "step": 11934 + }, + { + "epoch": 0.6668156549431516, + "grad_norm": 0.46386268734931946, + "learning_rate": 0.0006688704616763783, + "loss": 0.4603, + "step": 11935 + }, + { + "epoch": 0.6668715255468335, + "grad_norm": 0.6524155735969543, + "learning_rate": 0.0006688424473330346, + "loss": 0.5719, + "step": 11936 + }, + { + "epoch": 0.6669273961505154, + "grad_norm": 0.6862685084342957, + "learning_rate": 0.0006688144329896906, + "loss": 0.5495, + "step": 11937 + }, + { + "epoch": 0.6669832667541973, + "grad_norm": 0.4679696559906006, + "learning_rate": 0.000668786418646347, + "loss": 0.4515, + "step": 11938 + }, + { + "epoch": 0.6670391373578791, + "grad_norm": 0.7810059189796448, + "learning_rate": 0.0006687584043030032, + "loss": 0.3928, + "step": 11939 + }, + { + "epoch": 0.6670950079615611, + "grad_norm": 0.4288873076438904, + "learning_rate": 0.0006687303899596594, + "loss": 0.4681, + "step": 11940 + }, + { + "epoch": 0.6671508785652429, + "grad_norm": 1.7220063209533691, + "learning_rate": 0.0006687023756163156, + "loss": 0.4545, + "step": 11941 + }, + { + "epoch": 0.6672067491689248, + "grad_norm": 0.5674805045127869, + "learning_rate": 0.0006686743612729718, + "loss": 0.3556, + "step": 11942 + }, + { + "epoch": 0.6672626197726066, + "grad_norm": 1.5371547937393188, + "learning_rate": 0.000668646346929628, + "loss": 0.5296, + "step": 11943 + }, + { + "epoch": 0.6673184903762885, + "grad_norm": 0.4675791561603546, + "learning_rate": 0.0006686183325862842, + "loss": 0.4275, + "step": 11944 + }, + { + "epoch": 0.6673743609799704, + "grad_norm": 0.568248987197876, + "learning_rate": 0.0006685903182429404, + "loss": 0.5674, + "step": 11945 + }, + { + "epoch": 0.6674302315836522, + "grad_norm": 0.792906641960144, + "learning_rate": 0.0006685623038995966, + "loss": 0.4399, + "step": 11946 + }, + { + "epoch": 0.6674861021873342, + "grad_norm": 0.39106667041778564, + "learning_rate": 0.0006685342895562528, + "loss": 0.4806, + "step": 11947 + }, + { + "epoch": 0.667541972791016, + "grad_norm": 0.49943792819976807, + "learning_rate": 0.0006685062752129091, + "loss": 0.5131, + "step": 11948 + }, + { + "epoch": 0.6675978433946979, + "grad_norm": 2.161207914352417, + "learning_rate": 0.0006684782608695652, + "loss": 0.4553, + "step": 11949 + }, + { + "epoch": 0.6676537139983798, + "grad_norm": 2.3446831703186035, + "learning_rate": 0.0006684502465262214, + "loss": 0.4008, + "step": 11950 + }, + { + "epoch": 0.6677095846020616, + "grad_norm": 0.5329267382621765, + "learning_rate": 0.0006684222321828776, + "loss": 0.4922, + "step": 11951 + }, + { + "epoch": 0.6677654552057435, + "grad_norm": 0.45179006457328796, + "learning_rate": 0.0006683942178395338, + "loss": 0.372, + "step": 11952 + }, + { + "epoch": 0.6678213258094253, + "grad_norm": 1.4821993112564087, + "learning_rate": 0.00066836620349619, + "loss": 0.438, + "step": 11953 + }, + { + "epoch": 0.6678771964131073, + "grad_norm": 0.7107797265052795, + "learning_rate": 0.0006683381891528463, + "loss": 0.3712, + "step": 11954 + }, + { + "epoch": 0.6679330670167891, + "grad_norm": 0.4754236936569214, + "learning_rate": 0.0006683101748095025, + "loss": 0.4888, + "step": 11955 + }, + { + "epoch": 0.667988937620471, + "grad_norm": 0.5108724236488342, + "learning_rate": 0.0006682821604661587, + "loss": 0.3543, + "step": 11956 + }, + { + "epoch": 0.6680448082241529, + "grad_norm": 0.6899701952934265, + "learning_rate": 0.0006682541461228149, + "loss": 0.4775, + "step": 11957 + }, + { + "epoch": 0.6681006788278347, + "grad_norm": 0.4322168529033661, + "learning_rate": 0.0006682261317794711, + "loss": 0.4933, + "step": 11958 + }, + { + "epoch": 0.6681565494315166, + "grad_norm": 0.6655228734016418, + "learning_rate": 0.0006681981174361274, + "loss": 0.5064, + "step": 11959 + }, + { + "epoch": 0.6682124200351984, + "grad_norm": 0.47448399662971497, + "learning_rate": 0.0006681701030927835, + "loss": 0.4049, + "step": 11960 + }, + { + "epoch": 0.6682682906388804, + "grad_norm": 0.8102615475654602, + "learning_rate": 0.0006681420887494397, + "loss": 0.3894, + "step": 11961 + }, + { + "epoch": 0.6683241612425622, + "grad_norm": 0.558324933052063, + "learning_rate": 0.0006681140744060959, + "loss": 0.4067, + "step": 11962 + }, + { + "epoch": 0.6683800318462441, + "grad_norm": 0.8992058634757996, + "learning_rate": 0.0006680860600627522, + "loss": 0.5579, + "step": 11963 + }, + { + "epoch": 0.668435902449926, + "grad_norm": 1.4858407974243164, + "learning_rate": 0.0006680580457194083, + "loss": 0.4986, + "step": 11964 + }, + { + "epoch": 0.6684917730536079, + "grad_norm": 7.185491561889648, + "learning_rate": 0.0006680300313760645, + "loss": 0.4436, + "step": 11965 + }, + { + "epoch": 0.6685476436572897, + "grad_norm": 0.4849085211753845, + "learning_rate": 0.0006680020170327208, + "loss": 0.4206, + "step": 11966 + }, + { + "epoch": 0.6686035142609716, + "grad_norm": 0.8865067958831787, + "learning_rate": 0.000667974002689377, + "loss": 0.4177, + "step": 11967 + }, + { + "epoch": 0.6686593848646535, + "grad_norm": 1.4322865009307861, + "learning_rate": 0.0006679459883460331, + "loss": 0.724, + "step": 11968 + }, + { + "epoch": 0.6687152554683353, + "grad_norm": 0.36339297890663147, + "learning_rate": 0.0006679179740026893, + "loss": 0.4463, + "step": 11969 + }, + { + "epoch": 0.6687711260720172, + "grad_norm": 0.5684939622879028, + "learning_rate": 0.0006678899596593457, + "loss": 0.408, + "step": 11970 + }, + { + "epoch": 0.6688269966756991, + "grad_norm": 0.5236297845840454, + "learning_rate": 0.0006678619453160019, + "loss": 0.3389, + "step": 11971 + }, + { + "epoch": 0.668882867279381, + "grad_norm": 0.6459040641784668, + "learning_rate": 0.000667833930972658, + "loss": 0.5809, + "step": 11972 + }, + { + "epoch": 0.6689387378830628, + "grad_norm": 0.4581083357334137, + "learning_rate": 0.0006678059166293143, + "loss": 0.5078, + "step": 11973 + }, + { + "epoch": 0.6689946084867447, + "grad_norm": 0.5507989525794983, + "learning_rate": 0.0006677779022859705, + "loss": 0.4089, + "step": 11974 + }, + { + "epoch": 0.6690504790904266, + "grad_norm": 0.47498708963394165, + "learning_rate": 0.0006677498879426266, + "loss": 0.4979, + "step": 11975 + }, + { + "epoch": 0.6691063496941084, + "grad_norm": 0.6490650177001953, + "learning_rate": 0.0006677218735992828, + "loss": 0.3876, + "step": 11976 + }, + { + "epoch": 0.6691622202977903, + "grad_norm": 0.6967902779579163, + "learning_rate": 0.0006676938592559391, + "loss": 0.4521, + "step": 11977 + }, + { + "epoch": 0.6692180909014722, + "grad_norm": 0.44781142473220825, + "learning_rate": 0.0006676658449125953, + "loss": 0.3724, + "step": 11978 + }, + { + "epoch": 0.6692739615051541, + "grad_norm": 0.44736504554748535, + "learning_rate": 0.0006676378305692514, + "loss": 0.4399, + "step": 11979 + }, + { + "epoch": 0.6693298321088359, + "grad_norm": 0.8333998322486877, + "learning_rate": 0.0006676098162259077, + "loss": 0.5272, + "step": 11980 + }, + { + "epoch": 0.6693857027125178, + "grad_norm": 0.9536080360412598, + "learning_rate": 0.0006675818018825639, + "loss": 0.5373, + "step": 11981 + }, + { + "epoch": 0.6694415733161997, + "grad_norm": 0.4937134087085724, + "learning_rate": 0.0006675537875392201, + "loss": 0.456, + "step": 11982 + }, + { + "epoch": 0.6694974439198815, + "grad_norm": 0.5559743642807007, + "learning_rate": 0.0006675257731958762, + "loss": 0.4472, + "step": 11983 + }, + { + "epoch": 0.6695533145235635, + "grad_norm": 0.6172904372215271, + "learning_rate": 0.0006674977588525325, + "loss": 0.544, + "step": 11984 + }, + { + "epoch": 0.6696091851272453, + "grad_norm": 0.44793787598609924, + "learning_rate": 0.0006674697445091887, + "loss": 0.4472, + "step": 11985 + }, + { + "epoch": 0.6696650557309272, + "grad_norm": 0.65681391954422, + "learning_rate": 0.000667441730165845, + "loss": 0.4816, + "step": 11986 + }, + { + "epoch": 0.669720926334609, + "grad_norm": 1.989658236503601, + "learning_rate": 0.000667413715822501, + "loss": 0.4813, + "step": 11987 + }, + { + "epoch": 0.669776796938291, + "grad_norm": 0.5056659579277039, + "learning_rate": 0.0006673857014791574, + "loss": 0.3924, + "step": 11988 + }, + { + "epoch": 0.6698326675419728, + "grad_norm": 0.7013728022575378, + "learning_rate": 0.0006673576871358136, + "loss": 0.4752, + "step": 11989 + }, + { + "epoch": 0.6698885381456546, + "grad_norm": 0.8148704767227173, + "learning_rate": 0.0006673296727924698, + "loss": 0.4319, + "step": 11990 + }, + { + "epoch": 0.6699444087493366, + "grad_norm": 0.6683910489082336, + "learning_rate": 0.000667301658449126, + "loss": 0.388, + "step": 11991 + }, + { + "epoch": 0.6700002793530184, + "grad_norm": 8.329096794128418, + "learning_rate": 0.0006672736441057822, + "loss": 0.4696, + "step": 11992 + }, + { + "epoch": 0.6700561499567003, + "grad_norm": 0.5204866528511047, + "learning_rate": 0.0006672456297624384, + "loss": 0.3941, + "step": 11993 + }, + { + "epoch": 0.6701120205603821, + "grad_norm": 0.39093953371047974, + "learning_rate": 0.0006672176154190945, + "loss": 0.3711, + "step": 11994 + }, + { + "epoch": 0.6701678911640641, + "grad_norm": 0.5442216396331787, + "learning_rate": 0.0006671896010757508, + "loss": 0.479, + "step": 11995 + }, + { + "epoch": 0.6702237617677459, + "grad_norm": 0.4194156527519226, + "learning_rate": 0.000667161586732407, + "loss": 0.4281, + "step": 11996 + }, + { + "epoch": 0.6702796323714277, + "grad_norm": 0.5095095038414001, + "learning_rate": 0.0006671335723890632, + "loss": 0.5235, + "step": 11997 + }, + { + "epoch": 0.6703355029751097, + "grad_norm": 0.6097812056541443, + "learning_rate": 0.0006671055580457194, + "loss": 0.3996, + "step": 11998 + }, + { + "epoch": 0.6703913735787915, + "grad_norm": 0.46970075368881226, + "learning_rate": 0.0006670775437023756, + "loss": 0.4273, + "step": 11999 + }, + { + "epoch": 0.6704472441824734, + "grad_norm": 0.37489718198776245, + "learning_rate": 0.0006670495293590318, + "loss": 0.4437, + "step": 12000 + }, + { + "epoch": 0.6704472441824734, + "eval_cer": 0.09060805429173077, + "eval_loss": 0.3426876664161682, + "eval_runtime": 56.0591, + "eval_samples_per_second": 80.95, + "eval_steps_per_second": 5.066, + "eval_wer": 0.35764364123907155, + "step": 12000 + }, + { + "epoch": 0.6705031147861553, + "grad_norm": 0.45205971598625183, + "learning_rate": 0.000667021515015688, + "loss": 0.5957, + "step": 12001 + }, + { + "epoch": 0.6705589853898372, + "grad_norm": 0.4532722532749176, + "learning_rate": 0.0006669935006723442, + "loss": 0.4614, + "step": 12002 + }, + { + "epoch": 0.670614855993519, + "grad_norm": 0.7101292014122009, + "learning_rate": 0.0006669654863290004, + "loss": 0.4122, + "step": 12003 + }, + { + "epoch": 0.6706707265972008, + "grad_norm": 0.4864048361778259, + "learning_rate": 0.0006669374719856567, + "loss": 0.3746, + "step": 12004 + }, + { + "epoch": 0.6707265972008828, + "grad_norm": 0.59797602891922, + "learning_rate": 0.000666909457642313, + "loss": 0.4574, + "step": 12005 + }, + { + "epoch": 0.6707824678045646, + "grad_norm": 0.4772562086582184, + "learning_rate": 0.0006668814432989691, + "loss": 0.4864, + "step": 12006 + }, + { + "epoch": 0.6708383384082465, + "grad_norm": 0.3763349652290344, + "learning_rate": 0.0006668534289556253, + "loss": 0.4023, + "step": 12007 + }, + { + "epoch": 0.6708942090119284, + "grad_norm": 0.6340355277061462, + "learning_rate": 0.0006668254146122815, + "loss": 0.5351, + "step": 12008 + }, + { + "epoch": 0.6709500796156103, + "grad_norm": 0.6661238670349121, + "learning_rate": 0.0006667974002689378, + "loss": 0.422, + "step": 12009 + }, + { + "epoch": 0.6710059502192921, + "grad_norm": 1.567510962486267, + "learning_rate": 0.0006667693859255939, + "loss": 0.4218, + "step": 12010 + }, + { + "epoch": 0.6710618208229739, + "grad_norm": 0.9127955436706543, + "learning_rate": 0.0006667413715822501, + "loss": 0.5663, + "step": 12011 + }, + { + "epoch": 0.6711176914266559, + "grad_norm": 0.39525723457336426, + "learning_rate": 0.0006667133572389064, + "loss": 0.4897, + "step": 12012 + }, + { + "epoch": 0.6711735620303377, + "grad_norm": 0.5062659978866577, + "learning_rate": 0.0006666853428955626, + "loss": 0.3388, + "step": 12013 + }, + { + "epoch": 0.6712294326340196, + "grad_norm": 0.4742056429386139, + "learning_rate": 0.0006666573285522187, + "loss": 0.4679, + "step": 12014 + }, + { + "epoch": 0.6712853032377015, + "grad_norm": 0.48809874057769775, + "learning_rate": 0.0006666293142088749, + "loss": 0.6278, + "step": 12015 + }, + { + "epoch": 0.6713411738413834, + "grad_norm": 1.0428968667984009, + "learning_rate": 0.0006666012998655312, + "loss": 0.5339, + "step": 12016 + }, + { + "epoch": 0.6713970444450652, + "grad_norm": 0.3583538234233856, + "learning_rate": 0.0006665732855221873, + "loss": 0.3141, + "step": 12017 + }, + { + "epoch": 0.6714529150487472, + "grad_norm": 0.5583202242851257, + "learning_rate": 0.0006665452711788435, + "loss": 0.416, + "step": 12018 + }, + { + "epoch": 0.671508785652429, + "grad_norm": 1.990698218345642, + "learning_rate": 0.0006665172568354999, + "loss": 0.3768, + "step": 12019 + }, + { + "epoch": 0.6715646562561108, + "grad_norm": 0.4053715169429779, + "learning_rate": 0.0006664892424921561, + "loss": 0.4175, + "step": 12020 + }, + { + "epoch": 0.6716205268597927, + "grad_norm": 0.5364459753036499, + "learning_rate": 0.0006664612281488122, + "loss": 0.4985, + "step": 12021 + }, + { + "epoch": 0.6716763974634746, + "grad_norm": 3.560028314590454, + "learning_rate": 0.0006664332138054684, + "loss": 0.4531, + "step": 12022 + }, + { + "epoch": 0.6717322680671565, + "grad_norm": 0.7216856479644775, + "learning_rate": 0.0006664051994621247, + "loss": 0.574, + "step": 12023 + }, + { + "epoch": 0.6717881386708383, + "grad_norm": 0.4858110845088959, + "learning_rate": 0.0006663771851187809, + "loss": 0.3855, + "step": 12024 + }, + { + "epoch": 0.6718440092745203, + "grad_norm": 0.5756111145019531, + "learning_rate": 0.000666349170775437, + "loss": 0.4011, + "step": 12025 + }, + { + "epoch": 0.6718998798782021, + "grad_norm": 0.8296681642532349, + "learning_rate": 0.0006663211564320932, + "loss": 0.4374, + "step": 12026 + }, + { + "epoch": 0.6719557504818839, + "grad_norm": 0.4089227318763733, + "learning_rate": 0.0006662931420887495, + "loss": 0.3887, + "step": 12027 + }, + { + "epoch": 0.6720116210855658, + "grad_norm": 0.704284131526947, + "learning_rate": 0.0006662651277454057, + "loss": 0.5421, + "step": 12028 + }, + { + "epoch": 0.6720674916892477, + "grad_norm": 0.37553808093070984, + "learning_rate": 0.0006662371134020618, + "loss": 0.3948, + "step": 12029 + }, + { + "epoch": 0.6721233622929296, + "grad_norm": 0.5348711609840393, + "learning_rate": 0.0006662090990587181, + "loss": 0.4647, + "step": 12030 + }, + { + "epoch": 0.6721792328966114, + "grad_norm": 0.5512802004814148, + "learning_rate": 0.0006661810847153743, + "loss": 0.4125, + "step": 12031 + }, + { + "epoch": 0.6722351035002934, + "grad_norm": 0.5150562524795532, + "learning_rate": 0.0006661530703720305, + "loss": 0.4665, + "step": 12032 + }, + { + "epoch": 0.6722909741039752, + "grad_norm": 0.5676779747009277, + "learning_rate": 0.0006661250560286866, + "loss": 0.7544, + "step": 12033 + }, + { + "epoch": 0.672346844707657, + "grad_norm": 1.3371925354003906, + "learning_rate": 0.0006660970416853429, + "loss": 0.3609, + "step": 12034 + }, + { + "epoch": 0.672402715311339, + "grad_norm": 0.34091949462890625, + "learning_rate": 0.0006660690273419991, + "loss": 0.3907, + "step": 12035 + }, + { + "epoch": 0.6724585859150208, + "grad_norm": 1.2389219999313354, + "learning_rate": 0.0006660410129986552, + "loss": 0.4565, + "step": 12036 + }, + { + "epoch": 0.6725144565187027, + "grad_norm": 0.5887259244918823, + "learning_rate": 0.0006660129986553116, + "loss": 0.5978, + "step": 12037 + }, + { + "epoch": 0.6725703271223845, + "grad_norm": 0.4916236102581024, + "learning_rate": 0.0006659849843119678, + "loss": 0.4269, + "step": 12038 + }, + { + "epoch": 0.6726261977260665, + "grad_norm": 0.423195481300354, + "learning_rate": 0.000665956969968624, + "loss": 0.501, + "step": 12039 + }, + { + "epoch": 0.6726820683297483, + "grad_norm": 0.33990564942359924, + "learning_rate": 0.0006659289556252801, + "loss": 0.3179, + "step": 12040 + }, + { + "epoch": 0.6727379389334301, + "grad_norm": 0.5031219720840454, + "learning_rate": 0.0006659009412819364, + "loss": 0.552, + "step": 12041 + }, + { + "epoch": 0.6727938095371121, + "grad_norm": 0.4898545742034912, + "learning_rate": 0.0006658729269385926, + "loss": 0.4586, + "step": 12042 + }, + { + "epoch": 0.6728496801407939, + "grad_norm": 3.476188898086548, + "learning_rate": 0.0006658449125952488, + "loss": 0.513, + "step": 12043 + }, + { + "epoch": 0.6729055507444758, + "grad_norm": 0.48872169852256775, + "learning_rate": 0.000665816898251905, + "loss": 0.5018, + "step": 12044 + }, + { + "epoch": 0.6729614213481576, + "grad_norm": 0.4980540871620178, + "learning_rate": 0.0006657888839085612, + "loss": 0.5411, + "step": 12045 + }, + { + "epoch": 0.6730172919518396, + "grad_norm": 0.43007567524909973, + "learning_rate": 0.0006657608695652174, + "loss": 0.4157, + "step": 12046 + }, + { + "epoch": 0.6730731625555214, + "grad_norm": 0.43981626629829407, + "learning_rate": 0.0006657328552218736, + "loss": 0.4718, + "step": 12047 + }, + { + "epoch": 0.6731290331592032, + "grad_norm": 0.7485710382461548, + "learning_rate": 0.0006657048408785298, + "loss": 0.4691, + "step": 12048 + }, + { + "epoch": 0.6731849037628852, + "grad_norm": 1.0607640743255615, + "learning_rate": 0.000665676826535186, + "loss": 0.6052, + "step": 12049 + }, + { + "epoch": 0.673240774366567, + "grad_norm": 4.6706061363220215, + "learning_rate": 0.0006656488121918422, + "loss": 0.5464, + "step": 12050 + }, + { + "epoch": 0.6732966449702489, + "grad_norm": 0.4982675015926361, + "learning_rate": 0.0006656207978484985, + "loss": 0.5381, + "step": 12051 + }, + { + "epoch": 0.6733525155739307, + "grad_norm": 0.6218124032020569, + "learning_rate": 0.0006655927835051546, + "loss": 0.4534, + "step": 12052 + }, + { + "epoch": 0.6734083861776127, + "grad_norm": 0.45700135827064514, + "learning_rate": 0.0006655647691618108, + "loss": 0.3975, + "step": 12053 + }, + { + "epoch": 0.6734642567812945, + "grad_norm": 0.5723519325256348, + "learning_rate": 0.000665536754818467, + "loss": 0.4192, + "step": 12054 + }, + { + "epoch": 0.6735201273849764, + "grad_norm": 5.362627029418945, + "learning_rate": 0.0006655087404751234, + "loss": 0.456, + "step": 12055 + }, + { + "epoch": 0.6735759979886583, + "grad_norm": 0.3989286720752716, + "learning_rate": 0.0006654807261317795, + "loss": 0.4365, + "step": 12056 + }, + { + "epoch": 0.6736318685923401, + "grad_norm": 0.587829053401947, + "learning_rate": 0.0006654527117884357, + "loss": 0.6538, + "step": 12057 + }, + { + "epoch": 0.673687739196022, + "grad_norm": 1.2690942287445068, + "learning_rate": 0.0006654246974450919, + "loss": 0.4629, + "step": 12058 + }, + { + "epoch": 0.6737436097997039, + "grad_norm": 0.6632868647575378, + "learning_rate": 0.0006653966831017481, + "loss": 0.5352, + "step": 12059 + }, + { + "epoch": 0.6737994804033858, + "grad_norm": 0.6219850182533264, + "learning_rate": 0.0006653686687584043, + "loss": 0.4855, + "step": 12060 + }, + { + "epoch": 0.6738553510070676, + "grad_norm": 1.6683636903762817, + "learning_rate": 0.0006653406544150605, + "loss": 0.3237, + "step": 12061 + }, + { + "epoch": 0.6739112216107495, + "grad_norm": 1.3797463178634644, + "learning_rate": 0.0006653126400717168, + "loss": 0.5874, + "step": 12062 + }, + { + "epoch": 0.6739670922144314, + "grad_norm": 0.3609963059425354, + "learning_rate": 0.0006652846257283729, + "loss": 0.4248, + "step": 12063 + }, + { + "epoch": 0.6740229628181132, + "grad_norm": 0.565015971660614, + "learning_rate": 0.0006652566113850291, + "loss": 0.4104, + "step": 12064 + }, + { + "epoch": 0.6740788334217951, + "grad_norm": 0.665215790271759, + "learning_rate": 0.0006652285970416853, + "loss": 0.4463, + "step": 12065 + }, + { + "epoch": 0.674134704025477, + "grad_norm": 0.8716135025024414, + "learning_rate": 0.0006652005826983416, + "loss": 0.6785, + "step": 12066 + }, + { + "epoch": 0.6741905746291589, + "grad_norm": 0.4100133180618286, + "learning_rate": 0.0006651725683549977, + "loss": 0.4273, + "step": 12067 + }, + { + "epoch": 0.6742464452328407, + "grad_norm": 0.5582512617111206, + "learning_rate": 0.0006651445540116539, + "loss": 0.5019, + "step": 12068 + }, + { + "epoch": 0.6743023158365226, + "grad_norm": 0.9816803336143494, + "learning_rate": 0.0006651165396683102, + "loss": 0.4555, + "step": 12069 + }, + { + "epoch": 0.6743581864402045, + "grad_norm": 0.5383749604225159, + "learning_rate": 0.0006650885253249665, + "loss": 0.4632, + "step": 12070 + }, + { + "epoch": 0.6744140570438864, + "grad_norm": 1.3958255052566528, + "learning_rate": 0.0006650605109816226, + "loss": 0.369, + "step": 12071 + }, + { + "epoch": 0.6744699276475682, + "grad_norm": 0.5971868634223938, + "learning_rate": 0.0006650324966382788, + "loss": 0.3812, + "step": 12072 + }, + { + "epoch": 0.6745257982512501, + "grad_norm": 0.6539416313171387, + "learning_rate": 0.0006650044822949351, + "loss": 0.6775, + "step": 12073 + }, + { + "epoch": 0.674581668854932, + "grad_norm": 0.7821443676948547, + "learning_rate": 0.0006649764679515913, + "loss": 0.5043, + "step": 12074 + }, + { + "epoch": 0.6746375394586138, + "grad_norm": 0.6043152213096619, + "learning_rate": 0.0006649484536082474, + "loss": 0.6057, + "step": 12075 + }, + { + "epoch": 0.6746934100622958, + "grad_norm": 0.35482531785964966, + "learning_rate": 0.0006649204392649037, + "loss": 0.3621, + "step": 12076 + }, + { + "epoch": 0.6747492806659776, + "grad_norm": 0.668795108795166, + "learning_rate": 0.0006648924249215599, + "loss": 0.4496, + "step": 12077 + }, + { + "epoch": 0.6748051512696595, + "grad_norm": 0.686455249786377, + "learning_rate": 0.000664864410578216, + "loss": 0.4201, + "step": 12078 + }, + { + "epoch": 0.6748610218733413, + "grad_norm": 0.4387229382991791, + "learning_rate": 0.0006648363962348722, + "loss": 0.429, + "step": 12079 + }, + { + "epoch": 0.6749168924770232, + "grad_norm": 0.7037979960441589, + "learning_rate": 0.0006648083818915285, + "loss": 0.4392, + "step": 12080 + }, + { + "epoch": 0.6749727630807051, + "grad_norm": 0.5684598684310913, + "learning_rate": 0.0006647803675481847, + "loss": 0.4491, + "step": 12081 + }, + { + "epoch": 0.6750286336843869, + "grad_norm": 0.7674437165260315, + "learning_rate": 0.0006647523532048408, + "loss": 0.3732, + "step": 12082 + }, + { + "epoch": 0.6750845042880689, + "grad_norm": 0.5429156422615051, + "learning_rate": 0.0006647243388614971, + "loss": 0.4637, + "step": 12083 + }, + { + "epoch": 0.6751403748917507, + "grad_norm": 0.4526750147342682, + "learning_rate": 0.0006646963245181533, + "loss": 0.4036, + "step": 12084 + }, + { + "epoch": 0.6751962454954326, + "grad_norm": 0.8927469849586487, + "learning_rate": 0.0006646683101748095, + "loss": 0.5018, + "step": 12085 + }, + { + "epoch": 0.6752521160991144, + "grad_norm": 0.40133044123649597, + "learning_rate": 0.0006646402958314656, + "loss": 0.3737, + "step": 12086 + }, + { + "epoch": 0.6753079867027963, + "grad_norm": 0.5275247097015381, + "learning_rate": 0.000664612281488122, + "loss": 0.4604, + "step": 12087 + }, + { + "epoch": 0.6753638573064782, + "grad_norm": 0.8558287024497986, + "learning_rate": 0.0006645842671447782, + "loss": 0.4883, + "step": 12088 + }, + { + "epoch": 0.67541972791016, + "grad_norm": 0.5358519554138184, + "learning_rate": 0.0006645562528014344, + "loss": 0.5296, + "step": 12089 + }, + { + "epoch": 0.675475598513842, + "grad_norm": 0.7149332761764526, + "learning_rate": 0.0006645282384580905, + "loss": 0.3941, + "step": 12090 + }, + { + "epoch": 0.6755314691175238, + "grad_norm": 5.140353202819824, + "learning_rate": 0.0006645002241147468, + "loss": 0.3893, + "step": 12091 + }, + { + "epoch": 0.6755873397212057, + "grad_norm": 0.5173874497413635, + "learning_rate": 0.000664472209771403, + "loss": 0.4618, + "step": 12092 + }, + { + "epoch": 0.6756432103248876, + "grad_norm": 0.734495222568512, + "learning_rate": 0.0006644441954280592, + "loss": 0.4534, + "step": 12093 + }, + { + "epoch": 0.6756990809285695, + "grad_norm": 0.49917128682136536, + "learning_rate": 0.0006644161810847154, + "loss": 0.4547, + "step": 12094 + }, + { + "epoch": 0.6757549515322513, + "grad_norm": 0.48738420009613037, + "learning_rate": 0.0006643881667413716, + "loss": 0.4342, + "step": 12095 + }, + { + "epoch": 0.6758108221359331, + "grad_norm": 0.5017119646072388, + "learning_rate": 0.0006643601523980278, + "loss": 0.425, + "step": 12096 + }, + { + "epoch": 0.6758666927396151, + "grad_norm": 0.5653423070907593, + "learning_rate": 0.000664332138054684, + "loss": 0.5404, + "step": 12097 + }, + { + "epoch": 0.6759225633432969, + "grad_norm": 0.6185843348503113, + "learning_rate": 0.0006643041237113402, + "loss": 0.5957, + "step": 12098 + }, + { + "epoch": 0.6759784339469788, + "grad_norm": 1.8669347763061523, + "learning_rate": 0.0006642761093679964, + "loss": 0.4819, + "step": 12099 + }, + { + "epoch": 0.6760343045506607, + "grad_norm": 1.2889412641525269, + "learning_rate": 0.0006642480950246526, + "loss": 0.4175, + "step": 12100 + }, + { + "epoch": 0.6760901751543426, + "grad_norm": 0.4416337311267853, + "learning_rate": 0.0006642200806813088, + "loss": 0.4753, + "step": 12101 + }, + { + "epoch": 0.6761460457580244, + "grad_norm": 0.4668799936771393, + "learning_rate": 0.000664192066337965, + "loss": 0.4174, + "step": 12102 + }, + { + "epoch": 0.6762019163617062, + "grad_norm": 0.4461088478565216, + "learning_rate": 0.0006641640519946212, + "loss": 0.4044, + "step": 12103 + }, + { + "epoch": 0.6762577869653882, + "grad_norm": 0.5919915437698364, + "learning_rate": 0.0006641360376512775, + "loss": 0.5318, + "step": 12104 + }, + { + "epoch": 0.67631365756907, + "grad_norm": 0.4968312680721283, + "learning_rate": 0.0006641080233079337, + "loss": 0.5914, + "step": 12105 + }, + { + "epoch": 0.6763695281727519, + "grad_norm": 0.5419401526451111, + "learning_rate": 0.0006640800089645899, + "loss": 0.4648, + "step": 12106 + }, + { + "epoch": 0.6764253987764338, + "grad_norm": 0.5383129119873047, + "learning_rate": 0.0006640519946212461, + "loss": 0.5225, + "step": 12107 + }, + { + "epoch": 0.6764812693801157, + "grad_norm": 0.49332308769226074, + "learning_rate": 0.0006640239802779024, + "loss": 0.393, + "step": 12108 + }, + { + "epoch": 0.6765371399837975, + "grad_norm": 0.6018158793449402, + "learning_rate": 0.0006639959659345585, + "loss": 0.3101, + "step": 12109 + }, + { + "epoch": 0.6765930105874794, + "grad_norm": 0.5057805776596069, + "learning_rate": 0.0006639679515912147, + "loss": 0.5529, + "step": 12110 + }, + { + "epoch": 0.6766488811911613, + "grad_norm": 0.489611953496933, + "learning_rate": 0.0006639399372478709, + "loss": 0.385, + "step": 12111 + }, + { + "epoch": 0.6767047517948431, + "grad_norm": 0.6429499983787537, + "learning_rate": 0.0006639119229045272, + "loss": 0.4151, + "step": 12112 + }, + { + "epoch": 0.676760622398525, + "grad_norm": 0.4464999735355377, + "learning_rate": 0.0006638839085611833, + "loss": 0.4557, + "step": 12113 + }, + { + "epoch": 0.6768164930022069, + "grad_norm": 0.3610619604587555, + "learning_rate": 0.0006638558942178395, + "loss": 0.3592, + "step": 12114 + }, + { + "epoch": 0.6768723636058888, + "grad_norm": 0.8123363256454468, + "learning_rate": 0.0006638278798744958, + "loss": 0.379, + "step": 12115 + }, + { + "epoch": 0.6769282342095706, + "grad_norm": 0.38403958082199097, + "learning_rate": 0.000663799865531152, + "loss": 0.4113, + "step": 12116 + }, + { + "epoch": 0.6769841048132526, + "grad_norm": 0.5400655269622803, + "learning_rate": 0.0006637718511878081, + "loss": 0.468, + "step": 12117 + }, + { + "epoch": 0.6770399754169344, + "grad_norm": 0.8339864015579224, + "learning_rate": 0.0006637438368444643, + "loss": 0.5036, + "step": 12118 + }, + { + "epoch": 0.6770958460206162, + "grad_norm": 3.806410551071167, + "learning_rate": 0.0006637158225011206, + "loss": 0.3702, + "step": 12119 + }, + { + "epoch": 0.6771517166242981, + "grad_norm": 3.720856189727783, + "learning_rate": 0.0006636878081577767, + "loss": 0.3914, + "step": 12120 + }, + { + "epoch": 0.67720758722798, + "grad_norm": 0.7931181192398071, + "learning_rate": 0.000663659793814433, + "loss": 0.4002, + "step": 12121 + }, + { + "epoch": 0.6772634578316619, + "grad_norm": 0.3601086735725403, + "learning_rate": 0.0006636317794710893, + "loss": 0.3536, + "step": 12122 + }, + { + "epoch": 0.6773193284353437, + "grad_norm": 0.48901861906051636, + "learning_rate": 0.0006636037651277455, + "loss": 0.5132, + "step": 12123 + }, + { + "epoch": 0.6773751990390257, + "grad_norm": 0.5356402397155762, + "learning_rate": 0.0006635757507844016, + "loss": 0.4729, + "step": 12124 + }, + { + "epoch": 0.6774310696427075, + "grad_norm": 1.7545113563537598, + "learning_rate": 0.0006635477364410578, + "loss": 0.4042, + "step": 12125 + }, + { + "epoch": 0.6774869402463893, + "grad_norm": 0.4585729241371155, + "learning_rate": 0.0006635197220977141, + "loss": 0.4751, + "step": 12126 + }, + { + "epoch": 0.6775428108500713, + "grad_norm": 0.5439075231552124, + "learning_rate": 0.0006634917077543703, + "loss": 0.4476, + "step": 12127 + }, + { + "epoch": 0.6775986814537531, + "grad_norm": 2.827550172805786, + "learning_rate": 0.0006634636934110264, + "loss": 0.6387, + "step": 12128 + }, + { + "epoch": 0.677654552057435, + "grad_norm": 1.5471045970916748, + "learning_rate": 0.0006634356790676826, + "loss": 0.4725, + "step": 12129 + }, + { + "epoch": 0.6777104226611168, + "grad_norm": 0.7297206521034241, + "learning_rate": 0.0006634076647243389, + "loss": 0.472, + "step": 12130 + }, + { + "epoch": 0.6777662932647988, + "grad_norm": 0.6265151500701904, + "learning_rate": 0.0006633796503809951, + "loss": 0.4972, + "step": 12131 + }, + { + "epoch": 0.6778221638684806, + "grad_norm": 2.9177911281585693, + "learning_rate": 0.0006633516360376512, + "loss": 0.4447, + "step": 12132 + }, + { + "epoch": 0.6778780344721624, + "grad_norm": 0.5984437465667725, + "learning_rate": 0.0006633236216943075, + "loss": 0.4321, + "step": 12133 + }, + { + "epoch": 0.6779339050758444, + "grad_norm": 0.5108895897865295, + "learning_rate": 0.0006632956073509637, + "loss": 0.4755, + "step": 12134 + }, + { + "epoch": 0.6779897756795262, + "grad_norm": 0.6375856399536133, + "learning_rate": 0.0006632675930076199, + "loss": 0.5996, + "step": 12135 + }, + { + "epoch": 0.6780456462832081, + "grad_norm": 0.38159239292144775, + "learning_rate": 0.000663239578664276, + "loss": 0.4698, + "step": 12136 + }, + { + "epoch": 0.6781015168868899, + "grad_norm": 0.663817822933197, + "learning_rate": 0.0006632115643209324, + "loss": 0.5922, + "step": 12137 + }, + { + "epoch": 0.6781573874905719, + "grad_norm": 0.38931000232696533, + "learning_rate": 0.0006631835499775886, + "loss": 0.3856, + "step": 12138 + }, + { + "epoch": 0.6782132580942537, + "grad_norm": 2.1890618801116943, + "learning_rate": 0.0006631555356342448, + "loss": 0.4832, + "step": 12139 + }, + { + "epoch": 0.6782691286979355, + "grad_norm": 0.4320657551288605, + "learning_rate": 0.000663127521290901, + "loss": 0.4523, + "step": 12140 + }, + { + "epoch": 0.6783249993016175, + "grad_norm": 0.4799239933490753, + "learning_rate": 0.0006630995069475572, + "loss": 0.4947, + "step": 12141 + }, + { + "epoch": 0.6783808699052993, + "grad_norm": 0.5745739936828613, + "learning_rate": 0.0006630714926042134, + "loss": 0.5424, + "step": 12142 + }, + { + "epoch": 0.6784367405089812, + "grad_norm": 0.6328808665275574, + "learning_rate": 0.0006630434782608695, + "loss": 0.5895, + "step": 12143 + }, + { + "epoch": 0.6784926111126631, + "grad_norm": 3.1393280029296875, + "learning_rate": 0.0006630154639175258, + "loss": 0.3944, + "step": 12144 + }, + { + "epoch": 0.678548481716345, + "grad_norm": 0.49175360798835754, + "learning_rate": 0.000662987449574182, + "loss": 0.4823, + "step": 12145 + }, + { + "epoch": 0.6786043523200268, + "grad_norm": 0.7664163112640381, + "learning_rate": 0.0006629594352308382, + "loss": 0.4822, + "step": 12146 + }, + { + "epoch": 0.6786602229237086, + "grad_norm": 0.3466714918613434, + "learning_rate": 0.0006629314208874944, + "loss": 0.4246, + "step": 12147 + }, + { + "epoch": 0.6787160935273906, + "grad_norm": 0.480686217546463, + "learning_rate": 0.0006629034065441506, + "loss": 0.5044, + "step": 12148 + }, + { + "epoch": 0.6787719641310724, + "grad_norm": 0.42434561252593994, + "learning_rate": 0.0006628753922008068, + "loss": 0.3381, + "step": 12149 + }, + { + "epoch": 0.6788278347347543, + "grad_norm": 1.1494009494781494, + "learning_rate": 0.000662847377857463, + "loss": 0.4255, + "step": 12150 + }, + { + "epoch": 0.6788837053384362, + "grad_norm": 0.56053626537323, + "learning_rate": 0.0006628193635141192, + "loss": 0.4875, + "step": 12151 + }, + { + "epoch": 0.6789395759421181, + "grad_norm": 0.46690070629119873, + "learning_rate": 0.0006627913491707754, + "loss": 0.4627, + "step": 12152 + }, + { + "epoch": 0.6789954465457999, + "grad_norm": 0.7505831718444824, + "learning_rate": 0.0006627633348274316, + "loss": 0.4501, + "step": 12153 + }, + { + "epoch": 0.6790513171494817, + "grad_norm": 0.6304952502250671, + "learning_rate": 0.000662735320484088, + "loss": 0.488, + "step": 12154 + }, + { + "epoch": 0.6791071877531637, + "grad_norm": 0.376792311668396, + "learning_rate": 0.0006627073061407441, + "loss": 0.3675, + "step": 12155 + }, + { + "epoch": 0.6791630583568455, + "grad_norm": 0.9358426332473755, + "learning_rate": 0.0006626792917974003, + "loss": 0.5911, + "step": 12156 + }, + { + "epoch": 0.6792189289605274, + "grad_norm": 2.009202003479004, + "learning_rate": 0.0006626512774540565, + "loss": 0.4212, + "step": 12157 + }, + { + "epoch": 0.6792747995642093, + "grad_norm": 1.5568758249282837, + "learning_rate": 0.0006626232631107128, + "loss": 0.4044, + "step": 12158 + }, + { + "epoch": 0.6793306701678912, + "grad_norm": 0.9419112801551819, + "learning_rate": 0.0006625952487673689, + "loss": 0.4029, + "step": 12159 + }, + { + "epoch": 0.679386540771573, + "grad_norm": 0.4104714095592499, + "learning_rate": 0.0006625672344240251, + "loss": 0.4352, + "step": 12160 + }, + { + "epoch": 0.679442411375255, + "grad_norm": 0.781950831413269, + "learning_rate": 0.0006625392200806814, + "loss": 0.4893, + "step": 12161 + }, + { + "epoch": 0.6794982819789368, + "grad_norm": 0.7795931100845337, + "learning_rate": 0.0006625112057373376, + "loss": 0.5232, + "step": 12162 + }, + { + "epoch": 0.6795541525826186, + "grad_norm": 0.40902698040008545, + "learning_rate": 0.0006624831913939937, + "loss": 0.3444, + "step": 12163 + }, + { + "epoch": 0.6796100231863005, + "grad_norm": 1.8376938104629517, + "learning_rate": 0.0006624551770506499, + "loss": 0.4677, + "step": 12164 + }, + { + "epoch": 0.6796658937899824, + "grad_norm": 0.3876369297504425, + "learning_rate": 0.0006624271627073062, + "loss": 0.3896, + "step": 12165 + }, + { + "epoch": 0.6797217643936643, + "grad_norm": 0.9743863344192505, + "learning_rate": 0.0006623991483639623, + "loss": 0.5291, + "step": 12166 + }, + { + "epoch": 0.6797776349973461, + "grad_norm": 0.9888876676559448, + "learning_rate": 0.0006623711340206185, + "loss": 0.4652, + "step": 12167 + }, + { + "epoch": 0.6798335056010281, + "grad_norm": 0.4388604164123535, + "learning_rate": 0.0006623431196772747, + "loss": 0.4265, + "step": 12168 + }, + { + "epoch": 0.6798893762047099, + "grad_norm": 0.4221736490726471, + "learning_rate": 0.000662315105333931, + "loss": 0.3607, + "step": 12169 + }, + { + "epoch": 0.6799452468083917, + "grad_norm": 0.3795951008796692, + "learning_rate": 0.0006622870909905871, + "loss": 0.3685, + "step": 12170 + }, + { + "epoch": 0.6800011174120736, + "grad_norm": 0.584857702255249, + "learning_rate": 0.0006622590766472434, + "loss": 0.3867, + "step": 12171 + }, + { + "epoch": 0.6800569880157555, + "grad_norm": 0.5684229731559753, + "learning_rate": 0.0006622310623038997, + "loss": 0.5112, + "step": 12172 + }, + { + "epoch": 0.6801128586194374, + "grad_norm": 24.732563018798828, + "learning_rate": 0.0006622030479605559, + "loss": 0.3574, + "step": 12173 + }, + { + "epoch": 0.6801687292231192, + "grad_norm": 0.9245827198028564, + "learning_rate": 0.000662175033617212, + "loss": 0.4057, + "step": 12174 + }, + { + "epoch": 0.6802245998268012, + "grad_norm": 0.5891727805137634, + "learning_rate": 0.0006621470192738682, + "loss": 0.4572, + "step": 12175 + }, + { + "epoch": 0.680280470430483, + "grad_norm": 0.7587485313415527, + "learning_rate": 0.0006621190049305245, + "loss": 0.438, + "step": 12176 + }, + { + "epoch": 0.6803363410341648, + "grad_norm": 0.45359718799591064, + "learning_rate": 0.0006620909905871807, + "loss": 0.3404, + "step": 12177 + }, + { + "epoch": 0.6803922116378468, + "grad_norm": 0.49607688188552856, + "learning_rate": 0.0006620629762438368, + "loss": 0.4651, + "step": 12178 + }, + { + "epoch": 0.6804480822415286, + "grad_norm": 0.6163540482521057, + "learning_rate": 0.0006620349619004931, + "loss": 0.4101, + "step": 12179 + }, + { + "epoch": 0.6805039528452105, + "grad_norm": 0.5999632477760315, + "learning_rate": 0.0006620069475571493, + "loss": 0.3639, + "step": 12180 + }, + { + "epoch": 0.6805598234488923, + "grad_norm": 0.5860523581504822, + "learning_rate": 0.0006619789332138055, + "loss": 0.3693, + "step": 12181 + }, + { + "epoch": 0.6806156940525743, + "grad_norm": 0.5594480037689209, + "learning_rate": 0.0006619509188704616, + "loss": 0.3711, + "step": 12182 + }, + { + "epoch": 0.6806715646562561, + "grad_norm": 0.7717341780662537, + "learning_rate": 0.0006619229045271179, + "loss": 0.5688, + "step": 12183 + }, + { + "epoch": 0.680727435259938, + "grad_norm": 0.5354214906692505, + "learning_rate": 0.0006618948901837741, + "loss": 0.3839, + "step": 12184 + }, + { + "epoch": 0.6807833058636199, + "grad_norm": 0.8268642425537109, + "learning_rate": 0.0006618668758404302, + "loss": 0.4389, + "step": 12185 + }, + { + "epoch": 0.6808391764673017, + "grad_norm": 0.7277330160140991, + "learning_rate": 0.0006618388614970865, + "loss": 0.4925, + "step": 12186 + }, + { + "epoch": 0.6808950470709836, + "grad_norm": 0.7579416036605835, + "learning_rate": 0.0006618108471537428, + "loss": 0.533, + "step": 12187 + }, + { + "epoch": 0.6809509176746654, + "grad_norm": 0.36419478058815, + "learning_rate": 0.000661782832810399, + "loss": 0.3414, + "step": 12188 + }, + { + "epoch": 0.6810067882783474, + "grad_norm": 0.31273210048675537, + "learning_rate": 0.0006617548184670551, + "loss": 0.3147, + "step": 12189 + }, + { + "epoch": 0.6810626588820292, + "grad_norm": 0.7339096069335938, + "learning_rate": 0.0006617268041237114, + "loss": 0.561, + "step": 12190 + }, + { + "epoch": 0.681118529485711, + "grad_norm": 0.6273639798164368, + "learning_rate": 0.0006616987897803676, + "loss": 0.3992, + "step": 12191 + }, + { + "epoch": 0.681174400089393, + "grad_norm": 0.47251683473587036, + "learning_rate": 0.0006616707754370238, + "loss": 0.3463, + "step": 12192 + }, + { + "epoch": 0.6812302706930748, + "grad_norm": 0.5845441222190857, + "learning_rate": 0.00066164276109368, + "loss": 0.5191, + "step": 12193 + }, + { + "epoch": 0.6812861412967567, + "grad_norm": 0.5446106195449829, + "learning_rate": 0.0006616147467503362, + "loss": 0.6195, + "step": 12194 + }, + { + "epoch": 0.6813420119004386, + "grad_norm": 1.653283953666687, + "learning_rate": 0.0006615867324069924, + "loss": 0.5487, + "step": 12195 + }, + { + "epoch": 0.6813978825041205, + "grad_norm": 0.9641126990318298, + "learning_rate": 0.0006615587180636486, + "loss": 0.4226, + "step": 12196 + }, + { + "epoch": 0.6814537531078023, + "grad_norm": 1.1229079961776733, + "learning_rate": 0.0006615307037203048, + "loss": 0.4869, + "step": 12197 + }, + { + "epoch": 0.6815096237114842, + "grad_norm": 0.536135196685791, + "learning_rate": 0.000661502689376961, + "loss": 0.4795, + "step": 12198 + }, + { + "epoch": 0.6815654943151661, + "grad_norm": 0.6087976694107056, + "learning_rate": 0.0006614746750336172, + "loss": 0.4435, + "step": 12199 + }, + { + "epoch": 0.681621364918848, + "grad_norm": 0.6635640263557434, + "learning_rate": 0.0006614466606902735, + "loss": 0.4483, + "step": 12200 + }, + { + "epoch": 0.6816772355225298, + "grad_norm": 0.5035842657089233, + "learning_rate": 0.0006614186463469296, + "loss": 0.4315, + "step": 12201 + }, + { + "epoch": 0.6817331061262117, + "grad_norm": 0.3689860999584198, + "learning_rate": 0.0006613906320035858, + "loss": 0.5096, + "step": 12202 + }, + { + "epoch": 0.6817889767298936, + "grad_norm": 0.4090437889099121, + "learning_rate": 0.000661362617660242, + "loss": 0.4287, + "step": 12203 + }, + { + "epoch": 0.6818448473335754, + "grad_norm": 0.5090007781982422, + "learning_rate": 0.0006613346033168984, + "loss": 0.4568, + "step": 12204 + }, + { + "epoch": 0.6819007179372573, + "grad_norm": 0.41277015209198, + "learning_rate": 0.0006613065889735545, + "loss": 0.4315, + "step": 12205 + }, + { + "epoch": 0.6819565885409392, + "grad_norm": 0.32180869579315186, + "learning_rate": 0.0006612785746302107, + "loss": 0.3142, + "step": 12206 + }, + { + "epoch": 0.682012459144621, + "grad_norm": 0.49056267738342285, + "learning_rate": 0.0006612505602868669, + "loss": 0.4453, + "step": 12207 + }, + { + "epoch": 0.6820683297483029, + "grad_norm": 0.49976682662963867, + "learning_rate": 0.0006612225459435231, + "loss": 0.442, + "step": 12208 + }, + { + "epoch": 0.6821242003519848, + "grad_norm": 0.7014768719673157, + "learning_rate": 0.0006611945316001793, + "loss": 0.613, + "step": 12209 + }, + { + "epoch": 0.6821800709556667, + "grad_norm": 3.41719913482666, + "learning_rate": 0.0006611665172568355, + "loss": 0.4365, + "step": 12210 + }, + { + "epoch": 0.6822359415593485, + "grad_norm": 0.5215200185775757, + "learning_rate": 0.0006611385029134918, + "loss": 0.3951, + "step": 12211 + }, + { + "epoch": 0.6822918121630305, + "grad_norm": 1.229231834411621, + "learning_rate": 0.0006611104885701479, + "loss": 0.5865, + "step": 12212 + }, + { + "epoch": 0.6823476827667123, + "grad_norm": 0.6007933020591736, + "learning_rate": 0.0006610824742268041, + "loss": 0.4895, + "step": 12213 + }, + { + "epoch": 0.6824035533703942, + "grad_norm": 0.468875914812088, + "learning_rate": 0.0006610544598834603, + "loss": 0.4617, + "step": 12214 + }, + { + "epoch": 0.682459423974076, + "grad_norm": 0.6677796840667725, + "learning_rate": 0.0006610264455401166, + "loss": 0.4358, + "step": 12215 + }, + { + "epoch": 0.682515294577758, + "grad_norm": 0.42745816707611084, + "learning_rate": 0.0006609984311967727, + "loss": 0.4352, + "step": 12216 + }, + { + "epoch": 0.6825711651814398, + "grad_norm": 0.46790143847465515, + "learning_rate": 0.0006609704168534289, + "loss": 0.4392, + "step": 12217 + }, + { + "epoch": 0.6826270357851216, + "grad_norm": 0.35816842317581177, + "learning_rate": 0.0006609424025100852, + "loss": 0.3822, + "step": 12218 + }, + { + "epoch": 0.6826829063888036, + "grad_norm": 0.4370289444923401, + "learning_rate": 0.0006609143881667414, + "loss": 0.4296, + "step": 12219 + }, + { + "epoch": 0.6827387769924854, + "grad_norm": 0.4969431161880493, + "learning_rate": 0.0006608863738233975, + "loss": 0.3825, + "step": 12220 + }, + { + "epoch": 0.6827946475961673, + "grad_norm": 3.278639793395996, + "learning_rate": 0.0006608583594800538, + "loss": 0.4883, + "step": 12221 + }, + { + "epoch": 0.6828505181998491, + "grad_norm": 0.9645598530769348, + "learning_rate": 0.0006608303451367101, + "loss": 0.4439, + "step": 12222 + }, + { + "epoch": 0.682906388803531, + "grad_norm": 0.48043692111968994, + "learning_rate": 0.0006608023307933663, + "loss": 0.4863, + "step": 12223 + }, + { + "epoch": 0.6829622594072129, + "grad_norm": 0.4655659794807434, + "learning_rate": 0.0006607743164500224, + "loss": 0.3995, + "step": 12224 + }, + { + "epoch": 0.6830181300108947, + "grad_norm": 0.36219051480293274, + "learning_rate": 0.0006607463021066787, + "loss": 0.3955, + "step": 12225 + }, + { + "epoch": 0.6830740006145767, + "grad_norm": 0.6013150811195374, + "learning_rate": 0.0006607182877633349, + "loss": 0.4609, + "step": 12226 + }, + { + "epoch": 0.6831298712182585, + "grad_norm": 0.4733344614505768, + "learning_rate": 0.000660690273419991, + "loss": 0.3856, + "step": 12227 + }, + { + "epoch": 0.6831857418219404, + "grad_norm": 0.497595876455307, + "learning_rate": 0.0006606622590766472, + "loss": 0.4439, + "step": 12228 + }, + { + "epoch": 0.6832416124256223, + "grad_norm": 0.4180092513561249, + "learning_rate": 0.0006606342447333035, + "loss": 0.5122, + "step": 12229 + }, + { + "epoch": 0.6832974830293042, + "grad_norm": 0.43218445777893066, + "learning_rate": 0.0006606062303899597, + "loss": 0.3704, + "step": 12230 + }, + { + "epoch": 0.683353353632986, + "grad_norm": 1.1748855113983154, + "learning_rate": 0.0006605782160466158, + "loss": 0.382, + "step": 12231 + }, + { + "epoch": 0.6834092242366678, + "grad_norm": 0.6933474540710449, + "learning_rate": 0.0006605502017032721, + "loss": 0.4325, + "step": 12232 + }, + { + "epoch": 0.6834650948403498, + "grad_norm": 0.686015784740448, + "learning_rate": 0.0006605221873599283, + "loss": 0.5866, + "step": 12233 + }, + { + "epoch": 0.6835209654440316, + "grad_norm": 0.3392067849636078, + "learning_rate": 0.0006604941730165845, + "loss": 0.4106, + "step": 12234 + }, + { + "epoch": 0.6835768360477135, + "grad_norm": 0.6250699758529663, + "learning_rate": 0.0006604661586732406, + "loss": 0.6077, + "step": 12235 + }, + { + "epoch": 0.6836327066513954, + "grad_norm": 0.5005284547805786, + "learning_rate": 0.000660438144329897, + "loss": 0.4798, + "step": 12236 + }, + { + "epoch": 0.6836885772550773, + "grad_norm": 0.5760022401809692, + "learning_rate": 0.0006604101299865532, + "loss": 0.4482, + "step": 12237 + }, + { + "epoch": 0.6837444478587591, + "grad_norm": 1.0232678651809692, + "learning_rate": 0.0006603821156432094, + "loss": 0.4769, + "step": 12238 + }, + { + "epoch": 0.6838003184624409, + "grad_norm": 0.47398966550827026, + "learning_rate": 0.0006603541012998655, + "loss": 0.4013, + "step": 12239 + }, + { + "epoch": 0.6838561890661229, + "grad_norm": 1.313449740409851, + "learning_rate": 0.0006603260869565218, + "loss": 0.4372, + "step": 12240 + }, + { + "epoch": 0.6839120596698047, + "grad_norm": 0.4444301128387451, + "learning_rate": 0.000660298072613178, + "loss": 0.385, + "step": 12241 + }, + { + "epoch": 0.6839679302734866, + "grad_norm": 0.34318655729293823, + "learning_rate": 0.0006602700582698342, + "loss": 0.3679, + "step": 12242 + }, + { + "epoch": 0.6840238008771685, + "grad_norm": 0.5190429091453552, + "learning_rate": 0.0006602420439264904, + "loss": 0.5178, + "step": 12243 + }, + { + "epoch": 0.6840796714808504, + "grad_norm": 0.5612664222717285, + "learning_rate": 0.0006602140295831466, + "loss": 0.4143, + "step": 12244 + }, + { + "epoch": 0.6841355420845322, + "grad_norm": 0.7213748693466187, + "learning_rate": 0.0006601860152398028, + "loss": 0.4061, + "step": 12245 + }, + { + "epoch": 0.6841914126882142, + "grad_norm": 0.3936840891838074, + "learning_rate": 0.000660158000896459, + "loss": 0.4856, + "step": 12246 + }, + { + "epoch": 0.684247283291896, + "grad_norm": 0.4452683627605438, + "learning_rate": 0.0006601299865531152, + "loss": 0.5644, + "step": 12247 + }, + { + "epoch": 0.6843031538955778, + "grad_norm": 0.4156206548213959, + "learning_rate": 0.0006601019722097714, + "loss": 0.4302, + "step": 12248 + }, + { + "epoch": 0.6843590244992597, + "grad_norm": 0.8180496096611023, + "learning_rate": 0.0006600739578664276, + "loss": 0.542, + "step": 12249 + }, + { + "epoch": 0.6844148951029416, + "grad_norm": 1.7032221555709839, + "learning_rate": 0.0006600459435230838, + "loss": 0.4873, + "step": 12250 + }, + { + "epoch": 0.6844707657066235, + "grad_norm": 0.6767202019691467, + "learning_rate": 0.00066001792917974, + "loss": 0.3108, + "step": 12251 + }, + { + "epoch": 0.6845266363103053, + "grad_norm": 0.5809878706932068, + "learning_rate": 0.0006599899148363962, + "loss": 0.4257, + "step": 12252 + }, + { + "epoch": 0.6845825069139873, + "grad_norm": 0.49949684739112854, + "learning_rate": 0.0006599619004930524, + "loss": 0.4801, + "step": 12253 + }, + { + "epoch": 0.6846383775176691, + "grad_norm": 0.4479297697544098, + "learning_rate": 0.0006599338861497087, + "loss": 0.3433, + "step": 12254 + }, + { + "epoch": 0.6846942481213509, + "grad_norm": 0.6398472189903259, + "learning_rate": 0.0006599058718063649, + "loss": 0.4735, + "step": 12255 + }, + { + "epoch": 0.6847501187250328, + "grad_norm": 0.590287446975708, + "learning_rate": 0.0006598778574630211, + "loss": 0.4524, + "step": 12256 + }, + { + "epoch": 0.6848059893287147, + "grad_norm": 0.767348051071167, + "learning_rate": 0.0006598498431196774, + "loss": 0.4736, + "step": 12257 + }, + { + "epoch": 0.6848618599323966, + "grad_norm": 0.4661799967288971, + "learning_rate": 0.0006598218287763335, + "loss": 0.4625, + "step": 12258 + }, + { + "epoch": 0.6849177305360784, + "grad_norm": 0.9197534918785095, + "learning_rate": 0.0006597938144329897, + "loss": 0.4348, + "step": 12259 + }, + { + "epoch": 0.6849736011397604, + "grad_norm": 1.2513831853866577, + "learning_rate": 0.0006597658000896459, + "loss": 0.4294, + "step": 12260 + }, + { + "epoch": 0.6850294717434422, + "grad_norm": 0.5932397246360779, + "learning_rate": 0.0006597377857463022, + "loss": 0.431, + "step": 12261 + }, + { + "epoch": 0.685085342347124, + "grad_norm": 0.537572979927063, + "learning_rate": 0.0006597097714029583, + "loss": 0.3263, + "step": 12262 + }, + { + "epoch": 0.685141212950806, + "grad_norm": 0.47972768545150757, + "learning_rate": 0.0006596817570596145, + "loss": 0.4204, + "step": 12263 + }, + { + "epoch": 0.6851970835544878, + "grad_norm": 0.5310962796211243, + "learning_rate": 0.0006596537427162708, + "loss": 0.4955, + "step": 12264 + }, + { + "epoch": 0.6852529541581697, + "grad_norm": 0.5120587944984436, + "learning_rate": 0.000659625728372927, + "loss": 0.4929, + "step": 12265 + }, + { + "epoch": 0.6853088247618515, + "grad_norm": 0.37050455808639526, + "learning_rate": 0.0006595977140295831, + "loss": 0.3354, + "step": 12266 + }, + { + "epoch": 0.6853646953655335, + "grad_norm": 1.8886771202087402, + "learning_rate": 0.0006595696996862393, + "loss": 0.486, + "step": 12267 + }, + { + "epoch": 0.6854205659692153, + "grad_norm": 0.6789273023605347, + "learning_rate": 0.0006595416853428956, + "loss": 0.423, + "step": 12268 + }, + { + "epoch": 0.6854764365728971, + "grad_norm": 3.8334436416625977, + "learning_rate": 0.0006595136709995517, + "loss": 0.4418, + "step": 12269 + }, + { + "epoch": 0.6855323071765791, + "grad_norm": 0.35843491554260254, + "learning_rate": 0.0006594856566562079, + "loss": 0.3462, + "step": 12270 + }, + { + "epoch": 0.6855881777802609, + "grad_norm": 0.8044034838676453, + "learning_rate": 0.0006594576423128641, + "loss": 0.4336, + "step": 12271 + }, + { + "epoch": 0.6856440483839428, + "grad_norm": 0.7263255715370178, + "learning_rate": 0.0006594296279695205, + "loss": 0.4128, + "step": 12272 + }, + { + "epoch": 0.6856999189876246, + "grad_norm": 0.44200772047042847, + "learning_rate": 0.0006594016136261766, + "loss": 0.4247, + "step": 12273 + }, + { + "epoch": 0.6857557895913066, + "grad_norm": 1.0229723453521729, + "learning_rate": 0.0006593735992828328, + "loss": 0.4072, + "step": 12274 + }, + { + "epoch": 0.6858116601949884, + "grad_norm": 5.606417655944824, + "learning_rate": 0.0006593455849394891, + "loss": 0.4229, + "step": 12275 + }, + { + "epoch": 0.6858675307986702, + "grad_norm": 0.5073003768920898, + "learning_rate": 0.0006593175705961453, + "loss": 0.4645, + "step": 12276 + }, + { + "epoch": 0.6859234014023522, + "grad_norm": 0.5121041536331177, + "learning_rate": 0.0006592895562528014, + "loss": 0.3923, + "step": 12277 + }, + { + "epoch": 0.685979272006034, + "grad_norm": 0.434953898191452, + "learning_rate": 0.0006592615419094576, + "loss": 0.4556, + "step": 12278 + }, + { + "epoch": 0.6860351426097159, + "grad_norm": 0.484667032957077, + "learning_rate": 0.0006592335275661139, + "loss": 0.3873, + "step": 12279 + }, + { + "epoch": 0.6860910132133977, + "grad_norm": 0.9289417862892151, + "learning_rate": 0.0006592055132227701, + "loss": 0.4836, + "step": 12280 + }, + { + "epoch": 0.6861468838170797, + "grad_norm": 0.9068688750267029, + "learning_rate": 0.0006591774988794262, + "loss": 0.5675, + "step": 12281 + }, + { + "epoch": 0.6862027544207615, + "grad_norm": 0.5912303924560547, + "learning_rate": 0.0006591494845360825, + "loss": 0.4224, + "step": 12282 + }, + { + "epoch": 0.6862586250244433, + "grad_norm": 0.578102707862854, + "learning_rate": 0.0006591214701927387, + "loss": 0.4499, + "step": 12283 + }, + { + "epoch": 0.6863144956281253, + "grad_norm": 0.537879228591919, + "learning_rate": 0.0006590934558493949, + "loss": 0.4175, + "step": 12284 + }, + { + "epoch": 0.6863703662318071, + "grad_norm": 0.46350422501564026, + "learning_rate": 0.000659065441506051, + "loss": 0.4844, + "step": 12285 + }, + { + "epoch": 0.686426236835489, + "grad_norm": 0.5014481544494629, + "learning_rate": 0.0006590374271627073, + "loss": 0.3509, + "step": 12286 + }, + { + "epoch": 0.6864821074391709, + "grad_norm": 0.509566605091095, + "learning_rate": 0.0006590094128193636, + "loss": 0.4709, + "step": 12287 + }, + { + "epoch": 0.6865379780428528, + "grad_norm": 0.41533759236335754, + "learning_rate": 0.0006589813984760198, + "loss": 0.3898, + "step": 12288 + }, + { + "epoch": 0.6865938486465346, + "grad_norm": 0.867058277130127, + "learning_rate": 0.000658953384132676, + "loss": 0.5178, + "step": 12289 + }, + { + "epoch": 0.6866497192502165, + "grad_norm": 0.6219642162322998, + "learning_rate": 0.0006589253697893322, + "loss": 0.5137, + "step": 12290 + }, + { + "epoch": 0.6867055898538984, + "grad_norm": 0.4951665997505188, + "learning_rate": 0.0006588973554459884, + "loss": 0.3909, + "step": 12291 + }, + { + "epoch": 0.6867614604575802, + "grad_norm": 1.862715721130371, + "learning_rate": 0.0006588693411026445, + "loss": 0.4053, + "step": 12292 + }, + { + "epoch": 0.6868173310612621, + "grad_norm": 0.40303245186805725, + "learning_rate": 0.0006588413267593008, + "loss": 0.2997, + "step": 12293 + }, + { + "epoch": 0.686873201664944, + "grad_norm": 0.7202922105789185, + "learning_rate": 0.000658813312415957, + "loss": 0.4776, + "step": 12294 + }, + { + "epoch": 0.6869290722686259, + "grad_norm": 0.5889778137207031, + "learning_rate": 0.0006587852980726132, + "loss": 0.3594, + "step": 12295 + }, + { + "epoch": 0.6869849428723077, + "grad_norm": 1.0148833990097046, + "learning_rate": 0.0006587572837292694, + "loss": 0.3943, + "step": 12296 + }, + { + "epoch": 0.6870408134759896, + "grad_norm": 0.5266472101211548, + "learning_rate": 0.0006587292693859256, + "loss": 0.4914, + "step": 12297 + }, + { + "epoch": 0.6870966840796715, + "grad_norm": 0.4154301583766937, + "learning_rate": 0.0006587012550425818, + "loss": 0.3876, + "step": 12298 + }, + { + "epoch": 0.6871525546833533, + "grad_norm": 0.42289412021636963, + "learning_rate": 0.000658673240699238, + "loss": 0.5117, + "step": 12299 + }, + { + "epoch": 0.6872084252870352, + "grad_norm": 0.5242668390274048, + "learning_rate": 0.0006586452263558942, + "loss": 0.4129, + "step": 12300 + }, + { + "epoch": 0.6872642958907171, + "grad_norm": 0.43477779626846313, + "learning_rate": 0.0006586172120125504, + "loss": 0.4228, + "step": 12301 + }, + { + "epoch": 0.687320166494399, + "grad_norm": 0.9364448189735413, + "learning_rate": 0.0006585891976692066, + "loss": 0.5539, + "step": 12302 + }, + { + "epoch": 0.6873760370980808, + "grad_norm": 0.4951951205730438, + "learning_rate": 0.000658561183325863, + "loss": 0.3488, + "step": 12303 + }, + { + "epoch": 0.6874319077017628, + "grad_norm": 1.013813853263855, + "learning_rate": 0.000658533168982519, + "loss": 0.5539, + "step": 12304 + }, + { + "epoch": 0.6874877783054446, + "grad_norm": 0.3308449685573578, + "learning_rate": 0.0006585051546391753, + "loss": 0.3541, + "step": 12305 + }, + { + "epoch": 0.6875436489091264, + "grad_norm": 0.4052242040634155, + "learning_rate": 0.0006584771402958315, + "loss": 0.3559, + "step": 12306 + }, + { + "epoch": 0.6875995195128083, + "grad_norm": 0.48133212327957153, + "learning_rate": 0.0006584491259524878, + "loss": 0.402, + "step": 12307 + }, + { + "epoch": 0.6876553901164902, + "grad_norm": 0.49849236011505127, + "learning_rate": 0.0006584211116091439, + "loss": 0.496, + "step": 12308 + }, + { + "epoch": 0.6877112607201721, + "grad_norm": 0.45169320702552795, + "learning_rate": 0.0006583930972658001, + "loss": 0.3146, + "step": 12309 + }, + { + "epoch": 0.6877671313238539, + "grad_norm": 0.36836040019989014, + "learning_rate": 0.0006583650829224563, + "loss": 0.4145, + "step": 12310 + }, + { + "epoch": 0.6878230019275359, + "grad_norm": 1.1646658182144165, + "learning_rate": 0.0006583370685791126, + "loss": 0.5031, + "step": 12311 + }, + { + "epoch": 0.6878788725312177, + "grad_norm": 1.182687759399414, + "learning_rate": 0.0006583090542357687, + "loss": 0.4953, + "step": 12312 + }, + { + "epoch": 0.6879347431348996, + "grad_norm": 0.5369318127632141, + "learning_rate": 0.0006582810398924249, + "loss": 0.4116, + "step": 12313 + }, + { + "epoch": 0.6879906137385814, + "grad_norm": 0.4306829571723938, + "learning_rate": 0.0006582530255490812, + "loss": 0.3786, + "step": 12314 + }, + { + "epoch": 0.6880464843422633, + "grad_norm": 0.46830815076828003, + "learning_rate": 0.0006582250112057373, + "loss": 0.4953, + "step": 12315 + }, + { + "epoch": 0.6881023549459452, + "grad_norm": 0.45176824927330017, + "learning_rate": 0.0006581969968623935, + "loss": 0.5249, + "step": 12316 + }, + { + "epoch": 0.688158225549627, + "grad_norm": 0.9330127835273743, + "learning_rate": 0.0006581689825190497, + "loss": 0.3455, + "step": 12317 + }, + { + "epoch": 0.688214096153309, + "grad_norm": 0.5211748480796814, + "learning_rate": 0.000658140968175706, + "loss": 0.5157, + "step": 12318 + }, + { + "epoch": 0.6882699667569908, + "grad_norm": 1.7055243253707886, + "learning_rate": 0.0006581129538323621, + "loss": 0.5158, + "step": 12319 + }, + { + "epoch": 0.6883258373606727, + "grad_norm": 0.4864569902420044, + "learning_rate": 0.0006580849394890183, + "loss": 0.4125, + "step": 12320 + }, + { + "epoch": 0.6883817079643546, + "grad_norm": 0.6865062713623047, + "learning_rate": 0.0006580569251456747, + "loss": 0.4145, + "step": 12321 + }, + { + "epoch": 0.6884375785680364, + "grad_norm": 0.40384891629219055, + "learning_rate": 0.0006580289108023309, + "loss": 0.3771, + "step": 12322 + }, + { + "epoch": 0.6884934491717183, + "grad_norm": 0.49348893761634827, + "learning_rate": 0.000658000896458987, + "loss": 0.4481, + "step": 12323 + }, + { + "epoch": 0.6885493197754001, + "grad_norm": 0.3904980421066284, + "learning_rate": 0.0006579728821156432, + "loss": 0.4222, + "step": 12324 + }, + { + "epoch": 0.6886051903790821, + "grad_norm": 0.4685159921646118, + "learning_rate": 0.0006579448677722995, + "loss": 0.4635, + "step": 12325 + }, + { + "epoch": 0.6886610609827639, + "grad_norm": 0.37529489398002625, + "learning_rate": 0.0006579168534289557, + "loss": 0.4522, + "step": 12326 + }, + { + "epoch": 0.6887169315864458, + "grad_norm": 0.45543304085731506, + "learning_rate": 0.0006578888390856118, + "loss": 0.517, + "step": 12327 + }, + { + "epoch": 0.6887728021901277, + "grad_norm": 0.5884066820144653, + "learning_rate": 0.0006578608247422681, + "loss": 0.5069, + "step": 12328 + }, + { + "epoch": 0.6888286727938095, + "grad_norm": 1.1185544729232788, + "learning_rate": 0.0006578328103989243, + "loss": 0.4682, + "step": 12329 + }, + { + "epoch": 0.6888845433974914, + "grad_norm": 0.5465458035469055, + "learning_rate": 0.0006578047960555805, + "loss": 0.4485, + "step": 12330 + }, + { + "epoch": 0.6889404140011732, + "grad_norm": 0.6286165714263916, + "learning_rate": 0.0006577767817122366, + "loss": 0.4324, + "step": 12331 + }, + { + "epoch": 0.6889962846048552, + "grad_norm": 0.4545525312423706, + "learning_rate": 0.0006577487673688929, + "loss": 0.3873, + "step": 12332 + }, + { + "epoch": 0.689052155208537, + "grad_norm": 0.39186814427375793, + "learning_rate": 0.0006577207530255491, + "loss": 0.3375, + "step": 12333 + }, + { + "epoch": 0.6891080258122189, + "grad_norm": 0.4881627857685089, + "learning_rate": 0.0006576927386822052, + "loss": 0.5036, + "step": 12334 + }, + { + "epoch": 0.6891638964159008, + "grad_norm": 0.4025878608226776, + "learning_rate": 0.0006576647243388615, + "loss": 0.4755, + "step": 12335 + }, + { + "epoch": 0.6892197670195827, + "grad_norm": 0.5554612874984741, + "learning_rate": 0.0006576367099955177, + "loss": 0.4243, + "step": 12336 + }, + { + "epoch": 0.6892756376232645, + "grad_norm": 0.603357195854187, + "learning_rate": 0.000657608695652174, + "loss": 0.5259, + "step": 12337 + }, + { + "epoch": 0.6893315082269464, + "grad_norm": 0.5850807428359985, + "learning_rate": 0.00065758068130883, + "loss": 0.349, + "step": 12338 + }, + { + "epoch": 0.6893873788306283, + "grad_norm": 0.4177708327770233, + "learning_rate": 0.0006575526669654864, + "loss": 0.373, + "step": 12339 + }, + { + "epoch": 0.6894432494343101, + "grad_norm": 0.6364362835884094, + "learning_rate": 0.0006575246526221426, + "loss": 0.5665, + "step": 12340 + }, + { + "epoch": 0.689499120037992, + "grad_norm": 0.677216112613678, + "learning_rate": 0.0006574966382787988, + "loss": 0.5415, + "step": 12341 + }, + { + "epoch": 0.6895549906416739, + "grad_norm": 0.6359837651252747, + "learning_rate": 0.0006574686239354549, + "loss": 0.4386, + "step": 12342 + }, + { + "epoch": 0.6896108612453558, + "grad_norm": 9.877021789550781, + "learning_rate": 0.0006574406095921112, + "loss": 0.421, + "step": 12343 + }, + { + "epoch": 0.6896667318490376, + "grad_norm": 0.556947648525238, + "learning_rate": 0.0006574125952487674, + "loss": 0.4364, + "step": 12344 + }, + { + "epoch": 0.6897226024527195, + "grad_norm": 0.7899487614631653, + "learning_rate": 0.0006573845809054236, + "loss": 0.4223, + "step": 12345 + }, + { + "epoch": 0.6897784730564014, + "grad_norm": 0.411349356174469, + "learning_rate": 0.0006573565665620798, + "loss": 0.3411, + "step": 12346 + }, + { + "epoch": 0.6898343436600832, + "grad_norm": 0.41235920786857605, + "learning_rate": 0.000657328552218736, + "loss": 0.3468, + "step": 12347 + }, + { + "epoch": 0.6898902142637651, + "grad_norm": 0.4869711995124817, + "learning_rate": 0.0006573005378753922, + "loss": 0.4332, + "step": 12348 + }, + { + "epoch": 0.689946084867447, + "grad_norm": 0.9464502930641174, + "learning_rate": 0.0006572725235320484, + "loss": 0.4232, + "step": 12349 + }, + { + "epoch": 0.6900019554711289, + "grad_norm": 0.9713358879089355, + "learning_rate": 0.0006572445091887046, + "loss": 0.5657, + "step": 12350 + }, + { + "epoch": 0.6900578260748107, + "grad_norm": 0.5017536282539368, + "learning_rate": 0.0006572164948453608, + "loss": 0.4512, + "step": 12351 + }, + { + "epoch": 0.6901136966784926, + "grad_norm": 0.47309356927871704, + "learning_rate": 0.000657188480502017, + "loss": 0.4077, + "step": 12352 + }, + { + "epoch": 0.6901695672821745, + "grad_norm": 0.5770159959793091, + "learning_rate": 0.0006571604661586734, + "loss": 0.5549, + "step": 12353 + }, + { + "epoch": 0.6902254378858563, + "grad_norm": 0.7703874111175537, + "learning_rate": 0.0006571324518153294, + "loss": 0.4126, + "step": 12354 + }, + { + "epoch": 0.6902813084895383, + "grad_norm": 0.4894838035106659, + "learning_rate": 0.0006571044374719857, + "loss": 0.4068, + "step": 12355 + }, + { + "epoch": 0.6903371790932201, + "grad_norm": 0.3795333802700043, + "learning_rate": 0.0006570764231286419, + "loss": 0.3319, + "step": 12356 + }, + { + "epoch": 0.690393049696902, + "grad_norm": 0.4445364773273468, + "learning_rate": 0.0006570484087852981, + "loss": 0.4656, + "step": 12357 + }, + { + "epoch": 0.6904489203005838, + "grad_norm": 0.4124196469783783, + "learning_rate": 0.0006570203944419543, + "loss": 0.4325, + "step": 12358 + }, + { + "epoch": 0.6905047909042658, + "grad_norm": 0.46724647283554077, + "learning_rate": 0.0006569923800986105, + "loss": 0.4109, + "step": 12359 + }, + { + "epoch": 0.6905606615079476, + "grad_norm": 1.5697063207626343, + "learning_rate": 0.0006569643657552668, + "loss": 0.4442, + "step": 12360 + }, + { + "epoch": 0.6906165321116294, + "grad_norm": 0.6147964596748352, + "learning_rate": 0.0006569363514119229, + "loss": 0.513, + "step": 12361 + }, + { + "epoch": 0.6906724027153114, + "grad_norm": 0.5172843337059021, + "learning_rate": 0.0006569083370685791, + "loss": 0.5164, + "step": 12362 + }, + { + "epoch": 0.6907282733189932, + "grad_norm": 0.7066328525543213, + "learning_rate": 0.0006568803227252353, + "loss": 0.4132, + "step": 12363 + }, + { + "epoch": 0.6907841439226751, + "grad_norm": 3.6594626903533936, + "learning_rate": 0.0006568523083818916, + "loss": 0.5669, + "step": 12364 + }, + { + "epoch": 0.6908400145263569, + "grad_norm": 0.5304833650588989, + "learning_rate": 0.0006568242940385477, + "loss": 0.3995, + "step": 12365 + }, + { + "epoch": 0.6908958851300389, + "grad_norm": 0.690711498260498, + "learning_rate": 0.0006567962796952039, + "loss": 0.4413, + "step": 12366 + }, + { + "epoch": 0.6909517557337207, + "grad_norm": 0.5752975344657898, + "learning_rate": 0.0006567682653518602, + "loss": 0.4957, + "step": 12367 + }, + { + "epoch": 0.6910076263374025, + "grad_norm": 0.38540956377983093, + "learning_rate": 0.0006567402510085164, + "loss": 0.4582, + "step": 12368 + }, + { + "epoch": 0.6910634969410845, + "grad_norm": 0.6412515640258789, + "learning_rate": 0.0006567122366651725, + "loss": 0.4398, + "step": 12369 + }, + { + "epoch": 0.6911193675447663, + "grad_norm": 0.612745463848114, + "learning_rate": 0.0006566842223218287, + "loss": 0.4297, + "step": 12370 + }, + { + "epoch": 0.6911752381484482, + "grad_norm": 1.0125750303268433, + "learning_rate": 0.0006566562079784851, + "loss": 0.4423, + "step": 12371 + }, + { + "epoch": 0.6912311087521301, + "grad_norm": 0.3688487410545349, + "learning_rate": 0.0006566281936351413, + "loss": 0.4335, + "step": 12372 + }, + { + "epoch": 0.691286979355812, + "grad_norm": 0.48100632429122925, + "learning_rate": 0.0006566001792917974, + "loss": 0.4758, + "step": 12373 + }, + { + "epoch": 0.6913428499594938, + "grad_norm": 0.9707431793212891, + "learning_rate": 0.0006565721649484537, + "loss": 0.4332, + "step": 12374 + }, + { + "epoch": 0.6913987205631756, + "grad_norm": 1.1318894624710083, + "learning_rate": 0.0006565441506051099, + "loss": 0.4871, + "step": 12375 + }, + { + "epoch": 0.6914545911668576, + "grad_norm": 0.4296087920665741, + "learning_rate": 0.000656516136261766, + "loss": 0.4838, + "step": 12376 + }, + { + "epoch": 0.6915104617705394, + "grad_norm": 0.7212554812431335, + "learning_rate": 0.0006564881219184222, + "loss": 0.3773, + "step": 12377 + }, + { + "epoch": 0.6915663323742213, + "grad_norm": 0.5918447375297546, + "learning_rate": 0.0006564601075750785, + "loss": 0.4452, + "step": 12378 + }, + { + "epoch": 0.6916222029779032, + "grad_norm": 0.8290608525276184, + "learning_rate": 0.0006564320932317347, + "loss": 0.5497, + "step": 12379 + }, + { + "epoch": 0.6916780735815851, + "grad_norm": 0.5804015398025513, + "learning_rate": 0.0006564040788883908, + "loss": 0.5458, + "step": 12380 + }, + { + "epoch": 0.6917339441852669, + "grad_norm": 0.38470515608787537, + "learning_rate": 0.000656376064545047, + "loss": 0.3655, + "step": 12381 + }, + { + "epoch": 0.6917898147889487, + "grad_norm": 0.7739316821098328, + "learning_rate": 0.0006563480502017033, + "loss": 0.3946, + "step": 12382 + }, + { + "epoch": 0.6918456853926307, + "grad_norm": 0.3603934943675995, + "learning_rate": 0.0006563200358583595, + "loss": 0.494, + "step": 12383 + }, + { + "epoch": 0.6919015559963125, + "grad_norm": 0.4539802074432373, + "learning_rate": 0.0006562920215150156, + "loss": 0.4018, + "step": 12384 + }, + { + "epoch": 0.6919574265999944, + "grad_norm": 0.7388176918029785, + "learning_rate": 0.0006562640071716719, + "loss": 0.4729, + "step": 12385 + }, + { + "epoch": 0.6920132972036763, + "grad_norm": 0.5586922764778137, + "learning_rate": 0.0006562359928283281, + "loss": 0.4852, + "step": 12386 + }, + { + "epoch": 0.6920691678073582, + "grad_norm": 3.383394718170166, + "learning_rate": 0.0006562079784849843, + "loss": 0.4583, + "step": 12387 + }, + { + "epoch": 0.69212503841104, + "grad_norm": 0.5720994472503662, + "learning_rate": 0.0006561799641416404, + "loss": 0.4076, + "step": 12388 + }, + { + "epoch": 0.692180909014722, + "grad_norm": 0.8974199295043945, + "learning_rate": 0.0006561519497982968, + "loss": 0.3483, + "step": 12389 + }, + { + "epoch": 0.6922367796184038, + "grad_norm": 1.301135778427124, + "learning_rate": 0.000656123935454953, + "loss": 0.3909, + "step": 12390 + }, + { + "epoch": 0.6922926502220856, + "grad_norm": 0.40411683917045593, + "learning_rate": 0.0006560959211116092, + "loss": 0.4414, + "step": 12391 + }, + { + "epoch": 0.6923485208257675, + "grad_norm": 0.5068641304969788, + "learning_rate": 0.0006560679067682654, + "loss": 0.4303, + "step": 12392 + }, + { + "epoch": 0.6924043914294494, + "grad_norm": 0.6017261743545532, + "learning_rate": 0.0006560398924249216, + "loss": 0.6569, + "step": 12393 + }, + { + "epoch": 0.6924602620331313, + "grad_norm": 0.4588146507740021, + "learning_rate": 0.0006560118780815778, + "loss": 0.4253, + "step": 12394 + }, + { + "epoch": 0.6925161326368131, + "grad_norm": 0.6872242093086243, + "learning_rate": 0.000655983863738234, + "loss": 0.3881, + "step": 12395 + }, + { + "epoch": 0.6925720032404951, + "grad_norm": 2.478390693664551, + "learning_rate": 0.0006559558493948902, + "loss": 0.4496, + "step": 12396 + }, + { + "epoch": 0.6926278738441769, + "grad_norm": 0.6556323170661926, + "learning_rate": 0.0006559278350515464, + "loss": 0.3795, + "step": 12397 + }, + { + "epoch": 0.6926837444478587, + "grad_norm": 0.3856714963912964, + "learning_rate": 0.0006558998207082026, + "loss": 0.4112, + "step": 12398 + }, + { + "epoch": 0.6927396150515406, + "grad_norm": 0.47239378094673157, + "learning_rate": 0.0006558718063648588, + "loss": 0.3915, + "step": 12399 + }, + { + "epoch": 0.6927954856552225, + "grad_norm": 0.4735134243965149, + "learning_rate": 0.000655843792021515, + "loss": 0.2997, + "step": 12400 + }, + { + "epoch": 0.6928513562589044, + "grad_norm": 0.7392222881317139, + "learning_rate": 0.0006558157776781712, + "loss": 0.8262, + "step": 12401 + }, + { + "epoch": 0.6929072268625862, + "grad_norm": 0.6297357678413391, + "learning_rate": 0.0006557877633348274, + "loss": 0.4088, + "step": 12402 + }, + { + "epoch": 0.6929630974662682, + "grad_norm": 0.6884269714355469, + "learning_rate": 0.0006557597489914836, + "loss": 0.4291, + "step": 12403 + }, + { + "epoch": 0.69301896806995, + "grad_norm": 0.46951836347579956, + "learning_rate": 0.0006557317346481398, + "loss": 0.4223, + "step": 12404 + }, + { + "epoch": 0.6930748386736318, + "grad_norm": 0.43010950088500977, + "learning_rate": 0.000655703720304796, + "loss": 0.549, + "step": 12405 + }, + { + "epoch": 0.6931307092773138, + "grad_norm": 0.8549187183380127, + "learning_rate": 0.0006556757059614524, + "loss": 0.4602, + "step": 12406 + }, + { + "epoch": 0.6931865798809956, + "grad_norm": 0.42140525579452515, + "learning_rate": 0.0006556476916181085, + "loss": 0.4811, + "step": 12407 + }, + { + "epoch": 0.6932424504846775, + "grad_norm": 0.5342273116111755, + "learning_rate": 0.0006556196772747647, + "loss": 0.4492, + "step": 12408 + }, + { + "epoch": 0.6932983210883593, + "grad_norm": 0.6010997295379639, + "learning_rate": 0.0006555916629314209, + "loss": 0.3339, + "step": 12409 + }, + { + "epoch": 0.6933541916920413, + "grad_norm": 0.4859614372253418, + "learning_rate": 0.0006555636485880772, + "loss": 0.5011, + "step": 12410 + }, + { + "epoch": 0.6934100622957231, + "grad_norm": 0.40541020035743713, + "learning_rate": 0.0006555356342447333, + "loss": 0.4023, + "step": 12411 + }, + { + "epoch": 0.693465932899405, + "grad_norm": 0.5137203931808472, + "learning_rate": 0.0006555076199013895, + "loss": 0.3662, + "step": 12412 + }, + { + "epoch": 0.6935218035030869, + "grad_norm": 0.5841973423957825, + "learning_rate": 0.0006554796055580458, + "loss": 0.384, + "step": 12413 + }, + { + "epoch": 0.6935776741067687, + "grad_norm": 0.5078099966049194, + "learning_rate": 0.000655451591214702, + "loss": 0.4421, + "step": 12414 + }, + { + "epoch": 0.6936335447104506, + "grad_norm": 0.4626191258430481, + "learning_rate": 0.0006554235768713581, + "loss": 0.4963, + "step": 12415 + }, + { + "epoch": 0.6936894153141324, + "grad_norm": 0.42011168599128723, + "learning_rate": 0.0006553955625280143, + "loss": 0.4051, + "step": 12416 + }, + { + "epoch": 0.6937452859178144, + "grad_norm": 1.0899134874343872, + "learning_rate": 0.0006553675481846706, + "loss": 0.4197, + "step": 12417 + }, + { + "epoch": 0.6938011565214962, + "grad_norm": 0.41830822825431824, + "learning_rate": 0.0006553395338413267, + "loss": 0.4553, + "step": 12418 + }, + { + "epoch": 0.693857027125178, + "grad_norm": 0.7658846378326416, + "learning_rate": 0.0006553115194979829, + "loss": 0.4626, + "step": 12419 + }, + { + "epoch": 0.69391289772886, + "grad_norm": 0.6991179585456848, + "learning_rate": 0.0006552835051546391, + "loss": 0.5369, + "step": 12420 + }, + { + "epoch": 0.6939687683325418, + "grad_norm": 0.45468834042549133, + "learning_rate": 0.0006552554908112955, + "loss": 0.5479, + "step": 12421 + }, + { + "epoch": 0.6940246389362237, + "grad_norm": 0.5024108290672302, + "learning_rate": 0.0006552274764679516, + "loss": 0.4918, + "step": 12422 + }, + { + "epoch": 0.6940805095399056, + "grad_norm": 0.5538709163665771, + "learning_rate": 0.0006551994621246078, + "loss": 0.2924, + "step": 12423 + }, + { + "epoch": 0.6941363801435875, + "grad_norm": 0.6818821430206299, + "learning_rate": 0.0006551714477812641, + "loss": 0.4204, + "step": 12424 + }, + { + "epoch": 0.6941922507472693, + "grad_norm": 0.5659722685813904, + "learning_rate": 0.0006551434334379203, + "loss": 0.5019, + "step": 12425 + }, + { + "epoch": 0.6942481213509512, + "grad_norm": 0.7703965306282043, + "learning_rate": 0.0006551154190945764, + "loss": 0.4767, + "step": 12426 + }, + { + "epoch": 0.6943039919546331, + "grad_norm": 0.3505210280418396, + "learning_rate": 0.0006550874047512326, + "loss": 0.4318, + "step": 12427 + }, + { + "epoch": 0.6943598625583149, + "grad_norm": 0.36368346214294434, + "learning_rate": 0.0006550593904078889, + "loss": 0.3438, + "step": 12428 + }, + { + "epoch": 0.6944157331619968, + "grad_norm": 0.9703373312950134, + "learning_rate": 0.0006550313760645451, + "loss": 0.562, + "step": 12429 + }, + { + "epoch": 0.6944716037656787, + "grad_norm": 0.44626861810684204, + "learning_rate": 0.0006550033617212012, + "loss": 0.4124, + "step": 12430 + }, + { + "epoch": 0.6945274743693606, + "grad_norm": 0.5429732799530029, + "learning_rate": 0.0006549753473778575, + "loss": 0.4083, + "step": 12431 + }, + { + "epoch": 0.6945833449730424, + "grad_norm": 0.3813696801662445, + "learning_rate": 0.0006549473330345137, + "loss": 0.3168, + "step": 12432 + }, + { + "epoch": 0.6946392155767243, + "grad_norm": 0.38053417205810547, + "learning_rate": 0.0006549193186911699, + "loss": 0.4333, + "step": 12433 + }, + { + "epoch": 0.6946950861804062, + "grad_norm": 0.39903193712234497, + "learning_rate": 0.000654891304347826, + "loss": 0.394, + "step": 12434 + }, + { + "epoch": 0.694750956784088, + "grad_norm": 0.4331786632537842, + "learning_rate": 0.0006548632900044823, + "loss": 0.3937, + "step": 12435 + }, + { + "epoch": 0.6948068273877699, + "grad_norm": 0.39532893896102905, + "learning_rate": 0.0006548352756611385, + "loss": 0.3857, + "step": 12436 + }, + { + "epoch": 0.6948626979914518, + "grad_norm": 0.6744235157966614, + "learning_rate": 0.0006548072613177947, + "loss": 0.5046, + "step": 12437 + }, + { + "epoch": 0.6949185685951337, + "grad_norm": 0.6104267239570618, + "learning_rate": 0.000654779246974451, + "loss": 0.559, + "step": 12438 + }, + { + "epoch": 0.6949744391988155, + "grad_norm": 0.6823941469192505, + "learning_rate": 0.0006547512326311072, + "loss": 0.4283, + "step": 12439 + }, + { + "epoch": 0.6950303098024975, + "grad_norm": 0.6512045860290527, + "learning_rate": 0.0006547232182877634, + "loss": 0.4523, + "step": 12440 + }, + { + "epoch": 0.6950861804061793, + "grad_norm": 0.6811204552650452, + "learning_rate": 0.0006546952039444195, + "loss": 0.4878, + "step": 12441 + }, + { + "epoch": 0.6951420510098612, + "grad_norm": 0.6498299837112427, + "learning_rate": 0.0006546671896010758, + "loss": 0.6407, + "step": 12442 + }, + { + "epoch": 0.695197921613543, + "grad_norm": 1.2822033166885376, + "learning_rate": 0.000654639175257732, + "loss": 0.5768, + "step": 12443 + }, + { + "epoch": 0.6952537922172249, + "grad_norm": 1.2572790384292603, + "learning_rate": 0.0006546111609143882, + "loss": 0.3536, + "step": 12444 + }, + { + "epoch": 0.6953096628209068, + "grad_norm": 0.3888227343559265, + "learning_rate": 0.0006545831465710444, + "loss": 0.4048, + "step": 12445 + }, + { + "epoch": 0.6953655334245886, + "grad_norm": 0.4005681872367859, + "learning_rate": 0.0006545551322277006, + "loss": 0.3163, + "step": 12446 + }, + { + "epoch": 0.6954214040282706, + "grad_norm": 0.7777976989746094, + "learning_rate": 0.0006545271178843568, + "loss": 0.5779, + "step": 12447 + }, + { + "epoch": 0.6954772746319524, + "grad_norm": 0.5194401144981384, + "learning_rate": 0.000654499103541013, + "loss": 0.4865, + "step": 12448 + }, + { + "epoch": 0.6955331452356343, + "grad_norm": 0.409913033246994, + "learning_rate": 0.0006544710891976692, + "loss": 0.4304, + "step": 12449 + }, + { + "epoch": 0.6955890158393161, + "grad_norm": 0.3042415380477905, + "learning_rate": 0.0006544430748543254, + "loss": 0.334, + "step": 12450 + }, + { + "epoch": 0.695644886442998, + "grad_norm": 0.3752405047416687, + "learning_rate": 0.0006544150605109816, + "loss": 0.3479, + "step": 12451 + }, + { + "epoch": 0.6957007570466799, + "grad_norm": 0.4787050783634186, + "learning_rate": 0.0006543870461676378, + "loss": 0.3695, + "step": 12452 + }, + { + "epoch": 0.6957566276503617, + "grad_norm": 0.5405473709106445, + "learning_rate": 0.000654359031824294, + "loss": 0.4575, + "step": 12453 + }, + { + "epoch": 0.6958124982540437, + "grad_norm": 0.3838367462158203, + "learning_rate": 0.0006543310174809502, + "loss": 0.4668, + "step": 12454 + }, + { + "epoch": 0.6958683688577255, + "grad_norm": 0.6981995701789856, + "learning_rate": 0.0006543030031376065, + "loss": 0.4759, + "step": 12455 + }, + { + "epoch": 0.6959242394614074, + "grad_norm": 0.6496655941009521, + "learning_rate": 0.0006542749887942628, + "loss": 0.3703, + "step": 12456 + }, + { + "epoch": 0.6959801100650893, + "grad_norm": 0.5856876969337463, + "learning_rate": 0.0006542469744509189, + "loss": 0.3725, + "step": 12457 + }, + { + "epoch": 0.6960359806687711, + "grad_norm": 0.39370566606521606, + "learning_rate": 0.0006542189601075751, + "loss": 0.4034, + "step": 12458 + }, + { + "epoch": 0.696091851272453, + "grad_norm": 0.8162345886230469, + "learning_rate": 0.0006541909457642313, + "loss": 0.3481, + "step": 12459 + }, + { + "epoch": 0.6961477218761348, + "grad_norm": 0.6657698750495911, + "learning_rate": 0.0006541629314208875, + "loss": 0.4415, + "step": 12460 + }, + { + "epoch": 0.6962035924798168, + "grad_norm": 0.6437267661094666, + "learning_rate": 0.0006541349170775437, + "loss": 0.4346, + "step": 12461 + }, + { + "epoch": 0.6962594630834986, + "grad_norm": 0.9207749962806702, + "learning_rate": 0.0006541069027341999, + "loss": 0.514, + "step": 12462 + }, + { + "epoch": 0.6963153336871805, + "grad_norm": 0.5875070095062256, + "learning_rate": 0.0006540788883908562, + "loss": 0.4411, + "step": 12463 + }, + { + "epoch": 0.6963712042908624, + "grad_norm": 0.5361031889915466, + "learning_rate": 0.0006540508740475123, + "loss": 0.4494, + "step": 12464 + }, + { + "epoch": 0.6964270748945443, + "grad_norm": 0.43649065494537354, + "learning_rate": 0.0006540228597041685, + "loss": 0.5598, + "step": 12465 + }, + { + "epoch": 0.6964829454982261, + "grad_norm": 0.9541770219802856, + "learning_rate": 0.0006539948453608247, + "loss": 0.4363, + "step": 12466 + }, + { + "epoch": 0.6965388161019079, + "grad_norm": 0.8853284120559692, + "learning_rate": 0.000653966831017481, + "loss": 0.6047, + "step": 12467 + }, + { + "epoch": 0.6965946867055899, + "grad_norm": 1.0284041166305542, + "learning_rate": 0.0006539388166741371, + "loss": 0.4555, + "step": 12468 + }, + { + "epoch": 0.6966505573092717, + "grad_norm": 0.46419769525527954, + "learning_rate": 0.0006539108023307933, + "loss": 0.5164, + "step": 12469 + }, + { + "epoch": 0.6967064279129536, + "grad_norm": 0.5107366442680359, + "learning_rate": 0.0006538827879874496, + "loss": 0.4777, + "step": 12470 + }, + { + "epoch": 0.6967622985166355, + "grad_norm": 0.5688986778259277, + "learning_rate": 0.0006538547736441059, + "loss": 0.4089, + "step": 12471 + }, + { + "epoch": 0.6968181691203174, + "grad_norm": 0.3640749454498291, + "learning_rate": 0.000653826759300762, + "loss": 0.304, + "step": 12472 + }, + { + "epoch": 0.6968740397239992, + "grad_norm": 0.9997974038124084, + "learning_rate": 0.0006537987449574182, + "loss": 0.432, + "step": 12473 + }, + { + "epoch": 0.6969299103276811, + "grad_norm": 0.43687593936920166, + "learning_rate": 0.0006537707306140745, + "loss": 0.4713, + "step": 12474 + }, + { + "epoch": 0.696985780931363, + "grad_norm": 0.4060729742050171, + "learning_rate": 0.0006537427162707307, + "loss": 0.3797, + "step": 12475 + }, + { + "epoch": 0.6970416515350448, + "grad_norm": 0.6375582218170166, + "learning_rate": 0.0006537147019273868, + "loss": 0.3979, + "step": 12476 + }, + { + "epoch": 0.6970975221387267, + "grad_norm": 0.8401585221290588, + "learning_rate": 0.0006536866875840431, + "loss": 0.3794, + "step": 12477 + }, + { + "epoch": 0.6971533927424086, + "grad_norm": 0.7673923373222351, + "learning_rate": 0.0006536586732406993, + "loss": 0.529, + "step": 12478 + }, + { + "epoch": 0.6972092633460905, + "grad_norm": 0.4452807903289795, + "learning_rate": 0.0006536306588973555, + "loss": 0.4273, + "step": 12479 + }, + { + "epoch": 0.6972651339497723, + "grad_norm": 0.44349536299705505, + "learning_rate": 0.0006536026445540116, + "loss": 0.4193, + "step": 12480 + }, + { + "epoch": 0.6973210045534542, + "grad_norm": 1.0493369102478027, + "learning_rate": 0.0006535746302106679, + "loss": 0.3204, + "step": 12481 + }, + { + "epoch": 0.6973768751571361, + "grad_norm": 0.4893932640552521, + "learning_rate": 0.0006535466158673241, + "loss": 0.5665, + "step": 12482 + }, + { + "epoch": 0.6974327457608179, + "grad_norm": 0.5665848851203918, + "learning_rate": 0.0006535186015239802, + "loss": 0.4723, + "step": 12483 + }, + { + "epoch": 0.6974886163644998, + "grad_norm": 0.4613405168056488, + "learning_rate": 0.0006534905871806364, + "loss": 0.4333, + "step": 12484 + }, + { + "epoch": 0.6975444869681817, + "grad_norm": 0.4029140770435333, + "learning_rate": 0.0006534625728372927, + "loss": 0.3327, + "step": 12485 + }, + { + "epoch": 0.6976003575718636, + "grad_norm": 0.4724639356136322, + "learning_rate": 0.0006534345584939489, + "loss": 0.3364, + "step": 12486 + }, + { + "epoch": 0.6976562281755454, + "grad_norm": 0.4138375222682953, + "learning_rate": 0.000653406544150605, + "loss": 0.468, + "step": 12487 + }, + { + "epoch": 0.6977120987792274, + "grad_norm": 0.4055687189102173, + "learning_rate": 0.0006533785298072614, + "loss": 0.4314, + "step": 12488 + }, + { + "epoch": 0.6977679693829092, + "grad_norm": 1.2123990058898926, + "learning_rate": 0.0006533505154639176, + "loss": 0.4683, + "step": 12489 + }, + { + "epoch": 0.697823839986591, + "grad_norm": 0.3857737183570862, + "learning_rate": 0.0006533225011205738, + "loss": 0.4453, + "step": 12490 + }, + { + "epoch": 0.6978797105902729, + "grad_norm": 0.5594933032989502, + "learning_rate": 0.0006532944867772299, + "loss": 0.4284, + "step": 12491 + }, + { + "epoch": 0.6979355811939548, + "grad_norm": 0.4878815710544586, + "learning_rate": 0.0006532664724338862, + "loss": 0.4093, + "step": 12492 + }, + { + "epoch": 0.6979914517976367, + "grad_norm": 0.5961474180221558, + "learning_rate": 0.0006532384580905424, + "loss": 0.5267, + "step": 12493 + }, + { + "epoch": 0.6980473224013185, + "grad_norm": 1.1010987758636475, + "learning_rate": 0.0006532104437471986, + "loss": 0.51, + "step": 12494 + }, + { + "epoch": 0.6981031930050005, + "grad_norm": 0.5467108488082886, + "learning_rate": 0.0006531824294038548, + "loss": 0.5227, + "step": 12495 + }, + { + "epoch": 0.6981590636086823, + "grad_norm": 1.1433537006378174, + "learning_rate": 0.000653154415060511, + "loss": 0.4249, + "step": 12496 + }, + { + "epoch": 0.6982149342123641, + "grad_norm": 0.5675192475318909, + "learning_rate": 0.0006531264007171672, + "loss": 0.5442, + "step": 12497 + }, + { + "epoch": 0.6982708048160461, + "grad_norm": 1.317265510559082, + "learning_rate": 0.0006530983863738234, + "loss": 0.4473, + "step": 12498 + }, + { + "epoch": 0.6983266754197279, + "grad_norm": 0.4764583110809326, + "learning_rate": 0.0006530703720304796, + "loss": 0.5527, + "step": 12499 + }, + { + "epoch": 0.6983825460234098, + "grad_norm": 0.5621331334114075, + "learning_rate": 0.0006530423576871358, + "loss": 0.4541, + "step": 12500 + }, + { + "epoch": 0.6983825460234098, + "eval_cer": 0.09010070592342859, + "eval_loss": 0.3375411629676819, + "eval_runtime": 56.0502, + "eval_samples_per_second": 80.963, + "eval_steps_per_second": 5.067, + "eval_wer": 0.3573643194324181, + "step": 12500 + }, + { + "epoch": 0.6984384166270916, + "grad_norm": 0.7914776802062988, + "learning_rate": 0.000653014343343792, + "loss": 0.4549, + "step": 12501 + }, + { + "epoch": 0.6984942872307736, + "grad_norm": 0.581595242023468, + "learning_rate": 0.0006529863290004483, + "loss": 0.5253, + "step": 12502 + }, + { + "epoch": 0.6985501578344554, + "grad_norm": 0.4306609332561493, + "learning_rate": 0.0006529583146571044, + "loss": 0.4545, + "step": 12503 + }, + { + "epoch": 0.6986060284381372, + "grad_norm": 2.005943775177002, + "learning_rate": 0.0006529303003137606, + "loss": 0.4172, + "step": 12504 + }, + { + "epoch": 0.6986618990418192, + "grad_norm": 0.7283526062965393, + "learning_rate": 0.0006529022859704169, + "loss": 0.4646, + "step": 12505 + }, + { + "epoch": 0.698717769645501, + "grad_norm": 0.6237015128135681, + "learning_rate": 0.0006528742716270731, + "loss": 0.3991, + "step": 12506 + }, + { + "epoch": 0.6987736402491829, + "grad_norm": 0.5957909822463989, + "learning_rate": 0.0006528462572837293, + "loss": 0.4485, + "step": 12507 + }, + { + "epoch": 0.6988295108528647, + "grad_norm": 0.42107459902763367, + "learning_rate": 0.0006528182429403855, + "loss": 0.3868, + "step": 12508 + }, + { + "epoch": 0.6988853814565467, + "grad_norm": 0.6580045223236084, + "learning_rate": 0.0006527902285970418, + "loss": 0.4943, + "step": 12509 + }, + { + "epoch": 0.6989412520602285, + "grad_norm": 0.6459949016571045, + "learning_rate": 0.0006527622142536979, + "loss": 0.4028, + "step": 12510 + }, + { + "epoch": 0.6989971226639103, + "grad_norm": 1.4383331537246704, + "learning_rate": 0.0006527341999103541, + "loss": 0.4827, + "step": 12511 + }, + { + "epoch": 0.6990529932675923, + "grad_norm": 0.3945348262786865, + "learning_rate": 0.0006527061855670103, + "loss": 0.3413, + "step": 12512 + }, + { + "epoch": 0.6991088638712741, + "grad_norm": 0.39214691519737244, + "learning_rate": 0.0006526781712236666, + "loss": 0.3583, + "step": 12513 + }, + { + "epoch": 0.699164734474956, + "grad_norm": 1.101930856704712, + "learning_rate": 0.0006526501568803227, + "loss": 0.3808, + "step": 12514 + }, + { + "epoch": 0.6992206050786379, + "grad_norm": 0.5334333777427673, + "learning_rate": 0.0006526221425369789, + "loss": 0.5289, + "step": 12515 + }, + { + "epoch": 0.6992764756823198, + "grad_norm": 0.4199686348438263, + "learning_rate": 0.0006525941281936352, + "loss": 0.4554, + "step": 12516 + }, + { + "epoch": 0.6993323462860016, + "grad_norm": 0.40466412901878357, + "learning_rate": 0.0006525661138502914, + "loss": 0.3776, + "step": 12517 + }, + { + "epoch": 0.6993882168896834, + "grad_norm": 0.45769593119621277, + "learning_rate": 0.0006525380995069475, + "loss": 0.4111, + "step": 12518 + }, + { + "epoch": 0.6994440874933654, + "grad_norm": 0.4770458936691284, + "learning_rate": 0.0006525100851636037, + "loss": 0.4794, + "step": 12519 + }, + { + "epoch": 0.6994999580970472, + "grad_norm": 0.7833080291748047, + "learning_rate": 0.00065248207082026, + "loss": 0.406, + "step": 12520 + }, + { + "epoch": 0.6995558287007291, + "grad_norm": 0.3854668438434601, + "learning_rate": 0.0006524540564769163, + "loss": 0.4619, + "step": 12521 + }, + { + "epoch": 0.699611699304411, + "grad_norm": 0.40721720457077026, + "learning_rate": 0.0006524260421335724, + "loss": 0.344, + "step": 12522 + }, + { + "epoch": 0.6996675699080929, + "grad_norm": 0.5208889842033386, + "learning_rate": 0.0006523980277902286, + "loss": 0.4174, + "step": 12523 + }, + { + "epoch": 0.6997234405117747, + "grad_norm": 0.38133880496025085, + "learning_rate": 0.0006523700134468849, + "loss": 0.4909, + "step": 12524 + }, + { + "epoch": 0.6997793111154565, + "grad_norm": 0.42943957448005676, + "learning_rate": 0.000652341999103541, + "loss": 0.4344, + "step": 12525 + }, + { + "epoch": 0.6998351817191385, + "grad_norm": 0.5154833197593689, + "learning_rate": 0.0006523139847601972, + "loss": 0.5494, + "step": 12526 + }, + { + "epoch": 0.6998910523228203, + "grad_norm": 0.48272261023521423, + "learning_rate": 0.0006522859704168535, + "loss": 0.4218, + "step": 12527 + }, + { + "epoch": 0.6999469229265022, + "grad_norm": 1.7476847171783447, + "learning_rate": 0.0006522579560735097, + "loss": 0.4986, + "step": 12528 + }, + { + "epoch": 0.7000027935301841, + "grad_norm": 0.45337554812431335, + "learning_rate": 0.0006522299417301658, + "loss": 0.4437, + "step": 12529 + }, + { + "epoch": 0.700058664133866, + "grad_norm": 0.465181440114975, + "learning_rate": 0.000652201927386822, + "loss": 0.4595, + "step": 12530 + }, + { + "epoch": 0.7001145347375478, + "grad_norm": 1.9752265214920044, + "learning_rate": 0.0006521739130434783, + "loss": 0.6092, + "step": 12531 + }, + { + "epoch": 0.7001704053412298, + "grad_norm": 0.5274678468704224, + "learning_rate": 0.0006521458987001345, + "loss": 0.3736, + "step": 12532 + }, + { + "epoch": 0.7002262759449116, + "grad_norm": 0.6046198010444641, + "learning_rate": 0.0006521178843567906, + "loss": 0.4889, + "step": 12533 + }, + { + "epoch": 0.7002821465485934, + "grad_norm": 0.5951001644134521, + "learning_rate": 0.0006520898700134469, + "loss": 0.4707, + "step": 12534 + }, + { + "epoch": 0.7003380171522753, + "grad_norm": 0.8073389530181885, + "learning_rate": 0.0006520618556701031, + "loss": 0.7175, + "step": 12535 + }, + { + "epoch": 0.7003938877559572, + "grad_norm": 0.553976833820343, + "learning_rate": 0.0006520338413267593, + "loss": 0.3697, + "step": 12536 + }, + { + "epoch": 0.7004497583596391, + "grad_norm": 2.246562957763672, + "learning_rate": 0.0006520058269834154, + "loss": 0.4102, + "step": 12537 + }, + { + "epoch": 0.7005056289633209, + "grad_norm": 1.339470386505127, + "learning_rate": 0.0006519778126400718, + "loss": 0.4378, + "step": 12538 + }, + { + "epoch": 0.7005614995670029, + "grad_norm": 0.6469879746437073, + "learning_rate": 0.000651949798296728, + "loss": 0.3938, + "step": 12539 + }, + { + "epoch": 0.7006173701706847, + "grad_norm": 0.7258716821670532, + "learning_rate": 0.0006519217839533842, + "loss": 0.6378, + "step": 12540 + }, + { + "epoch": 0.7006732407743665, + "grad_norm": 5.494493007659912, + "learning_rate": 0.0006518937696100404, + "loss": 0.4577, + "step": 12541 + }, + { + "epoch": 0.7007291113780484, + "grad_norm": 1.158990740776062, + "learning_rate": 0.0006518657552666966, + "loss": 0.3642, + "step": 12542 + }, + { + "epoch": 0.7007849819817303, + "grad_norm": 0.5490526556968689, + "learning_rate": 0.0006518377409233528, + "loss": 0.4659, + "step": 12543 + }, + { + "epoch": 0.7008408525854122, + "grad_norm": 2.491509199142456, + "learning_rate": 0.000651809726580009, + "loss": 0.387, + "step": 12544 + }, + { + "epoch": 0.700896723189094, + "grad_norm": 0.6455645561218262, + "learning_rate": 0.0006517817122366652, + "loss": 0.457, + "step": 12545 + }, + { + "epoch": 0.700952593792776, + "grad_norm": 0.5082055330276489, + "learning_rate": 0.0006517536978933214, + "loss": 0.456, + "step": 12546 + }, + { + "epoch": 0.7010084643964578, + "grad_norm": 1.4105757474899292, + "learning_rate": 0.0006517256835499776, + "loss": 0.5998, + "step": 12547 + }, + { + "epoch": 0.7010643350001396, + "grad_norm": 0.4370853006839752, + "learning_rate": 0.0006516976692066338, + "loss": 0.451, + "step": 12548 + }, + { + "epoch": 0.7011202056038216, + "grad_norm": 2.0545575618743896, + "learning_rate": 0.00065166965486329, + "loss": 0.3677, + "step": 12549 + }, + { + "epoch": 0.7011760762075034, + "grad_norm": 0.4983333349227905, + "learning_rate": 0.0006516416405199462, + "loss": 0.4996, + "step": 12550 + }, + { + "epoch": 0.7012319468111853, + "grad_norm": 0.46462953090667725, + "learning_rate": 0.0006516136261766024, + "loss": 0.4844, + "step": 12551 + }, + { + "epoch": 0.7012878174148671, + "grad_norm": 0.5652801394462585, + "learning_rate": 0.0006515856118332586, + "loss": 0.6143, + "step": 12552 + }, + { + "epoch": 0.7013436880185491, + "grad_norm": 0.6559848785400391, + "learning_rate": 0.0006515575974899148, + "loss": 0.4489, + "step": 12553 + }, + { + "epoch": 0.7013995586222309, + "grad_norm": 0.38574153184890747, + "learning_rate": 0.000651529583146571, + "loss": 0.4791, + "step": 12554 + }, + { + "epoch": 0.7014554292259128, + "grad_norm": 0.40665388107299805, + "learning_rate": 0.0006515015688032274, + "loss": 0.3702, + "step": 12555 + }, + { + "epoch": 0.7015112998295947, + "grad_norm": 0.5149067640304565, + "learning_rate": 0.0006514735544598835, + "loss": 0.523, + "step": 12556 + }, + { + "epoch": 0.7015671704332765, + "grad_norm": 0.5217661261558533, + "learning_rate": 0.0006514455401165397, + "loss": 0.4288, + "step": 12557 + }, + { + "epoch": 0.7016230410369584, + "grad_norm": 0.41919568181037903, + "learning_rate": 0.0006514175257731959, + "loss": 0.4559, + "step": 12558 + }, + { + "epoch": 0.7016789116406402, + "grad_norm": 0.6591819524765015, + "learning_rate": 0.0006513895114298522, + "loss": 0.4878, + "step": 12559 + }, + { + "epoch": 0.7017347822443222, + "grad_norm": 2.773747444152832, + "learning_rate": 0.0006513614970865083, + "loss": 0.5164, + "step": 12560 + }, + { + "epoch": 0.701790652848004, + "grad_norm": 0.37052884697914124, + "learning_rate": 0.0006513334827431645, + "loss": 0.4084, + "step": 12561 + }, + { + "epoch": 0.7018465234516859, + "grad_norm": 0.37692418694496155, + "learning_rate": 0.0006513054683998207, + "loss": 0.4002, + "step": 12562 + }, + { + "epoch": 0.7019023940553678, + "grad_norm": 0.6584445238113403, + "learning_rate": 0.000651277454056477, + "loss": 0.5647, + "step": 12563 + }, + { + "epoch": 0.7019582646590496, + "grad_norm": 0.4484846889972687, + "learning_rate": 0.0006512494397131331, + "loss": 0.4712, + "step": 12564 + }, + { + "epoch": 0.7020141352627315, + "grad_norm": 0.762679934501648, + "learning_rate": 0.0006512214253697893, + "loss": 0.4769, + "step": 12565 + }, + { + "epoch": 0.7020700058664134, + "grad_norm": 0.4847736358642578, + "learning_rate": 0.0006511934110264456, + "loss": 0.4845, + "step": 12566 + }, + { + "epoch": 0.7021258764700953, + "grad_norm": 0.4513517916202545, + "learning_rate": 0.0006511653966831017, + "loss": 0.5437, + "step": 12567 + }, + { + "epoch": 0.7021817470737771, + "grad_norm": 0.4986242651939392, + "learning_rate": 0.0006511373823397579, + "loss": 0.3939, + "step": 12568 + }, + { + "epoch": 0.702237617677459, + "grad_norm": 0.738579273223877, + "learning_rate": 0.0006511093679964141, + "loss": 0.3853, + "step": 12569 + }, + { + "epoch": 0.7022934882811409, + "grad_norm": 1.8326221704483032, + "learning_rate": 0.0006510813536530704, + "loss": 0.4741, + "step": 12570 + }, + { + "epoch": 0.7023493588848227, + "grad_norm": 0.5769907832145691, + "learning_rate": 0.0006510533393097265, + "loss": 0.4076, + "step": 12571 + }, + { + "epoch": 0.7024052294885046, + "grad_norm": 2.1051831245422363, + "learning_rate": 0.0006510253249663828, + "loss": 0.4766, + "step": 12572 + }, + { + "epoch": 0.7024611000921865, + "grad_norm": 1.081831693649292, + "learning_rate": 0.0006509973106230391, + "loss": 0.5202, + "step": 12573 + }, + { + "epoch": 0.7025169706958684, + "grad_norm": 0.4195900857448578, + "learning_rate": 0.0006509692962796953, + "loss": 0.4201, + "step": 12574 + }, + { + "epoch": 0.7025728412995502, + "grad_norm": 0.6443397402763367, + "learning_rate": 0.0006509412819363514, + "loss": 0.56, + "step": 12575 + }, + { + "epoch": 0.7026287119032321, + "grad_norm": 0.5387607216835022, + "learning_rate": 0.0006509132675930076, + "loss": 0.6874, + "step": 12576 + }, + { + "epoch": 0.702684582506914, + "grad_norm": 0.5546450614929199, + "learning_rate": 0.0006508852532496639, + "loss": 0.4818, + "step": 12577 + }, + { + "epoch": 0.7027404531105959, + "grad_norm": 0.3970503509044647, + "learning_rate": 0.0006508572389063201, + "loss": 0.3711, + "step": 12578 + }, + { + "epoch": 0.7027963237142777, + "grad_norm": 2.104602575302124, + "learning_rate": 0.0006508292245629762, + "loss": 0.4944, + "step": 12579 + }, + { + "epoch": 0.7028521943179596, + "grad_norm": 0.38552325963974, + "learning_rate": 0.0006508012102196325, + "loss": 0.4398, + "step": 12580 + }, + { + "epoch": 0.7029080649216415, + "grad_norm": 0.4165954887866974, + "learning_rate": 0.0006507731958762887, + "loss": 0.4443, + "step": 12581 + }, + { + "epoch": 0.7029639355253233, + "grad_norm": 1.144707202911377, + "learning_rate": 0.0006507451815329449, + "loss": 0.405, + "step": 12582 + }, + { + "epoch": 0.7030198061290053, + "grad_norm": 0.4248158037662506, + "learning_rate": 0.000650717167189601, + "loss": 0.473, + "step": 12583 + }, + { + "epoch": 0.7030756767326871, + "grad_norm": 0.539726734161377, + "learning_rate": 0.0006506891528462573, + "loss": 0.4662, + "step": 12584 + }, + { + "epoch": 0.703131547336369, + "grad_norm": 2.4753708839416504, + "learning_rate": 0.0006506611385029135, + "loss": 0.4608, + "step": 12585 + }, + { + "epoch": 0.7031874179400508, + "grad_norm": 0.45003741979599, + "learning_rate": 0.0006506331241595697, + "loss": 0.3989, + "step": 12586 + }, + { + "epoch": 0.7032432885437327, + "grad_norm": 0.4516444504261017, + "learning_rate": 0.000650605109816226, + "loss": 0.497, + "step": 12587 + }, + { + "epoch": 0.7032991591474146, + "grad_norm": 0.44072243571281433, + "learning_rate": 0.0006505770954728822, + "loss": 0.5155, + "step": 12588 + }, + { + "epoch": 0.7033550297510964, + "grad_norm": 0.44851386547088623, + "learning_rate": 0.0006505490811295384, + "loss": 0.4651, + "step": 12589 + }, + { + "epoch": 0.7034109003547784, + "grad_norm": 0.4593474566936493, + "learning_rate": 0.0006505210667861945, + "loss": 0.4586, + "step": 12590 + }, + { + "epoch": 0.7034667709584602, + "grad_norm": 0.7007286548614502, + "learning_rate": 0.0006504930524428508, + "loss": 0.5439, + "step": 12591 + }, + { + "epoch": 0.7035226415621421, + "grad_norm": 1.9001071453094482, + "learning_rate": 0.000650465038099507, + "loss": 0.4456, + "step": 12592 + }, + { + "epoch": 0.7035785121658239, + "grad_norm": 0.4169517457485199, + "learning_rate": 0.0006504370237561632, + "loss": 0.4255, + "step": 12593 + }, + { + "epoch": 0.7036343827695059, + "grad_norm": 0.580703854560852, + "learning_rate": 0.0006504090094128193, + "loss": 0.5224, + "step": 12594 + }, + { + "epoch": 0.7036902533731877, + "grad_norm": 0.636456310749054, + "learning_rate": 0.0006503809950694756, + "loss": 0.3498, + "step": 12595 + }, + { + "epoch": 0.7037461239768695, + "grad_norm": 0.6758177280426025, + "learning_rate": 0.0006503529807261318, + "loss": 0.4089, + "step": 12596 + }, + { + "epoch": 0.7038019945805515, + "grad_norm": 0.7089087963104248, + "learning_rate": 0.000650324966382788, + "loss": 0.5086, + "step": 12597 + }, + { + "epoch": 0.7038578651842333, + "grad_norm": 0.4436291754245758, + "learning_rate": 0.0006502969520394442, + "loss": 0.4188, + "step": 12598 + }, + { + "epoch": 0.7039137357879152, + "grad_norm": 0.48724624514579773, + "learning_rate": 0.0006502689376961004, + "loss": 0.4259, + "step": 12599 + }, + { + "epoch": 0.7039696063915971, + "grad_norm": 0.9004136323928833, + "learning_rate": 0.0006502409233527566, + "loss": 0.708, + "step": 12600 + }, + { + "epoch": 0.704025476995279, + "grad_norm": 0.47161194682121277, + "learning_rate": 0.0006502129090094128, + "loss": 0.4455, + "step": 12601 + }, + { + "epoch": 0.7040813475989608, + "grad_norm": 0.805168092250824, + "learning_rate": 0.000650184894666069, + "loss": 0.5831, + "step": 12602 + }, + { + "epoch": 0.7041372182026426, + "grad_norm": 0.4293014705181122, + "learning_rate": 0.0006501568803227252, + "loss": 0.3439, + "step": 12603 + }, + { + "epoch": 0.7041930888063246, + "grad_norm": 0.41715195775032043, + "learning_rate": 0.0006501288659793814, + "loss": 0.4272, + "step": 12604 + }, + { + "epoch": 0.7042489594100064, + "grad_norm": 0.8402354121208191, + "learning_rate": 0.0006501008516360378, + "loss": 0.5902, + "step": 12605 + }, + { + "epoch": 0.7043048300136883, + "grad_norm": 0.3774196207523346, + "learning_rate": 0.0006500728372926939, + "loss": 0.4077, + "step": 12606 + }, + { + "epoch": 0.7043607006173702, + "grad_norm": 0.5531536936759949, + "learning_rate": 0.0006500448229493501, + "loss": 0.5421, + "step": 12607 + }, + { + "epoch": 0.7044165712210521, + "grad_norm": 0.4133262634277344, + "learning_rate": 0.0006500168086060063, + "loss": 0.3851, + "step": 12608 + }, + { + "epoch": 0.7044724418247339, + "grad_norm": 0.45653221011161804, + "learning_rate": 0.0006499887942626625, + "loss": 0.3893, + "step": 12609 + }, + { + "epoch": 0.7045283124284157, + "grad_norm": 1.9661084413528442, + "learning_rate": 0.0006499607799193187, + "loss": 0.5603, + "step": 12610 + }, + { + "epoch": 0.7045841830320977, + "grad_norm": 0.8259562253952026, + "learning_rate": 0.0006499327655759749, + "loss": 0.5315, + "step": 12611 + }, + { + "epoch": 0.7046400536357795, + "grad_norm": 0.4405726194381714, + "learning_rate": 0.0006499047512326312, + "loss": 0.3279, + "step": 12612 + }, + { + "epoch": 0.7046959242394614, + "grad_norm": 0.6673963665962219, + "learning_rate": 0.0006498767368892873, + "loss": 0.4555, + "step": 12613 + }, + { + "epoch": 0.7047517948431433, + "grad_norm": 0.7759480476379395, + "learning_rate": 0.0006498487225459435, + "loss": 0.4656, + "step": 12614 + }, + { + "epoch": 0.7048076654468252, + "grad_norm": 1.3405112028121948, + "learning_rate": 0.0006498207082025997, + "loss": 0.3984, + "step": 12615 + }, + { + "epoch": 0.704863536050507, + "grad_norm": 0.5985912084579468, + "learning_rate": 0.000649792693859256, + "loss": 0.383, + "step": 12616 + }, + { + "epoch": 0.704919406654189, + "grad_norm": 0.5204741954803467, + "learning_rate": 0.0006497646795159121, + "loss": 0.4351, + "step": 12617 + }, + { + "epoch": 0.7049752772578708, + "grad_norm": 0.6487236022949219, + "learning_rate": 0.0006497366651725683, + "loss": 0.3763, + "step": 12618 + }, + { + "epoch": 0.7050311478615526, + "grad_norm": 0.7357839941978455, + "learning_rate": 0.0006497086508292246, + "loss": 0.3823, + "step": 12619 + }, + { + "epoch": 0.7050870184652345, + "grad_norm": 0.5230579376220703, + "learning_rate": 0.0006496806364858808, + "loss": 0.5066, + "step": 12620 + }, + { + "epoch": 0.7051428890689164, + "grad_norm": 0.46423420310020447, + "learning_rate": 0.0006496526221425369, + "loss": 0.5208, + "step": 12621 + }, + { + "epoch": 0.7051987596725983, + "grad_norm": 0.4125761091709137, + "learning_rate": 0.0006496246077991931, + "loss": 0.4454, + "step": 12622 + }, + { + "epoch": 0.7052546302762801, + "grad_norm": 0.4252782166004181, + "learning_rate": 0.0006495965934558495, + "loss": 0.449, + "step": 12623 + }, + { + "epoch": 0.705310500879962, + "grad_norm": 0.3941310942173004, + "learning_rate": 0.0006495685791125057, + "loss": 0.3427, + "step": 12624 + }, + { + "epoch": 0.7053663714836439, + "grad_norm": 0.48103949427604675, + "learning_rate": 0.0006495405647691618, + "loss": 0.5442, + "step": 12625 + }, + { + "epoch": 0.7054222420873257, + "grad_norm": 0.4071162939071655, + "learning_rate": 0.0006495125504258181, + "loss": 0.4232, + "step": 12626 + }, + { + "epoch": 0.7054781126910076, + "grad_norm": 0.5169139504432678, + "learning_rate": 0.0006494845360824743, + "loss": 0.3432, + "step": 12627 + }, + { + "epoch": 0.7055339832946895, + "grad_norm": 0.6993414759635925, + "learning_rate": 0.0006494565217391305, + "loss": 0.4005, + "step": 12628 + }, + { + "epoch": 0.7055898538983714, + "grad_norm": 0.3642900288105011, + "learning_rate": 0.0006494285073957866, + "loss": 0.4101, + "step": 12629 + }, + { + "epoch": 0.7056457245020532, + "grad_norm": 0.4855663776397705, + "learning_rate": 0.0006494004930524429, + "loss": 0.5306, + "step": 12630 + }, + { + "epoch": 0.7057015951057352, + "grad_norm": 1.297998309135437, + "learning_rate": 0.0006493724787090991, + "loss": 0.3342, + "step": 12631 + }, + { + "epoch": 0.705757465709417, + "grad_norm": 0.5391308665275574, + "learning_rate": 0.0006493444643657552, + "loss": 0.512, + "step": 12632 + }, + { + "epoch": 0.7058133363130988, + "grad_norm": 0.414387047290802, + "learning_rate": 0.0006493164500224114, + "loss": 0.4554, + "step": 12633 + }, + { + "epoch": 0.7058692069167808, + "grad_norm": 0.44324424862861633, + "learning_rate": 0.0006492884356790677, + "loss": 0.2987, + "step": 12634 + }, + { + "epoch": 0.7059250775204626, + "grad_norm": 0.7372171878814697, + "learning_rate": 0.0006492604213357239, + "loss": 0.536, + "step": 12635 + }, + { + "epoch": 0.7059809481241445, + "grad_norm": 0.5872828364372253, + "learning_rate": 0.00064923240699238, + "loss": 0.5009, + "step": 12636 + }, + { + "epoch": 0.7060368187278263, + "grad_norm": 0.5148334503173828, + "learning_rate": 0.0006492043926490363, + "loss": 0.4101, + "step": 12637 + }, + { + "epoch": 0.7060926893315083, + "grad_norm": 0.43819186091423035, + "learning_rate": 0.0006491763783056926, + "loss": 0.5895, + "step": 12638 + }, + { + "epoch": 0.7061485599351901, + "grad_norm": 0.3660556674003601, + "learning_rate": 0.0006491483639623488, + "loss": 0.3754, + "step": 12639 + }, + { + "epoch": 0.7062044305388719, + "grad_norm": 0.56191486120224, + "learning_rate": 0.0006491203496190049, + "loss": 0.4462, + "step": 12640 + }, + { + "epoch": 0.7062603011425539, + "grad_norm": 0.36233896017074585, + "learning_rate": 0.0006490923352756612, + "loss": 0.411, + "step": 12641 + }, + { + "epoch": 0.7063161717462357, + "grad_norm": 1.3004298210144043, + "learning_rate": 0.0006490643209323174, + "loss": 0.5053, + "step": 12642 + }, + { + "epoch": 0.7063720423499176, + "grad_norm": 0.6685208082199097, + "learning_rate": 0.0006490363065889736, + "loss": 0.4987, + "step": 12643 + }, + { + "epoch": 0.7064279129535994, + "grad_norm": 0.3527875542640686, + "learning_rate": 0.0006490082922456298, + "loss": 0.4081, + "step": 12644 + }, + { + "epoch": 0.7064837835572814, + "grad_norm": 0.49977365136146545, + "learning_rate": 0.000648980277902286, + "loss": 0.4903, + "step": 12645 + }, + { + "epoch": 0.7065396541609632, + "grad_norm": 0.5331385135650635, + "learning_rate": 0.0006489522635589422, + "loss": 0.5397, + "step": 12646 + }, + { + "epoch": 0.706595524764645, + "grad_norm": 1.5992512702941895, + "learning_rate": 0.0006489242492155984, + "loss": 0.4089, + "step": 12647 + }, + { + "epoch": 0.706651395368327, + "grad_norm": 0.48003697395324707, + "learning_rate": 0.0006488962348722546, + "loss": 0.3727, + "step": 12648 + }, + { + "epoch": 0.7067072659720088, + "grad_norm": 0.7794275879859924, + "learning_rate": 0.0006488682205289108, + "loss": 0.4764, + "step": 12649 + }, + { + "epoch": 0.7067631365756907, + "grad_norm": 1.4320181608200073, + "learning_rate": 0.000648840206185567, + "loss": 0.5708, + "step": 12650 + }, + { + "epoch": 0.7068190071793726, + "grad_norm": 0.4379298686981201, + "learning_rate": 0.0006488121918422233, + "loss": 0.3909, + "step": 12651 + }, + { + "epoch": 0.7068748777830545, + "grad_norm": 0.4213232100009918, + "learning_rate": 0.0006487841774988794, + "loss": 0.5177, + "step": 12652 + }, + { + "epoch": 0.7069307483867363, + "grad_norm": 0.5488656759262085, + "learning_rate": 0.0006487561631555356, + "loss": 0.5233, + "step": 12653 + }, + { + "epoch": 0.7069866189904181, + "grad_norm": 1.0136057138442993, + "learning_rate": 0.0006487281488121918, + "loss": 0.406, + "step": 12654 + }, + { + "epoch": 0.7070424895941001, + "grad_norm": 0.4805604815483093, + "learning_rate": 0.000648700134468848, + "loss": 0.5489, + "step": 12655 + }, + { + "epoch": 0.7070983601977819, + "grad_norm": 0.41339918971061707, + "learning_rate": 0.0006486721201255043, + "loss": 0.3738, + "step": 12656 + }, + { + "epoch": 0.7071542308014638, + "grad_norm": 0.657025933265686, + "learning_rate": 0.0006486441057821605, + "loss": 0.7018, + "step": 12657 + }, + { + "epoch": 0.7072101014051457, + "grad_norm": 0.5366183519363403, + "learning_rate": 0.0006486160914388168, + "loss": 0.4922, + "step": 12658 + }, + { + "epoch": 0.7072659720088276, + "grad_norm": 5.80585241317749, + "learning_rate": 0.0006485880770954729, + "loss": 0.4387, + "step": 12659 + }, + { + "epoch": 0.7073218426125094, + "grad_norm": 0.591661274433136, + "learning_rate": 0.0006485600627521291, + "loss": 0.4515, + "step": 12660 + }, + { + "epoch": 0.7073777132161913, + "grad_norm": 0.7086001634597778, + "learning_rate": 0.0006485320484087853, + "loss": 0.5165, + "step": 12661 + }, + { + "epoch": 0.7074335838198732, + "grad_norm": 0.49627920985221863, + "learning_rate": 0.0006485040340654416, + "loss": 0.5047, + "step": 12662 + }, + { + "epoch": 0.707489454423555, + "grad_norm": 0.3379269540309906, + "learning_rate": 0.0006484760197220977, + "loss": 0.3818, + "step": 12663 + }, + { + "epoch": 0.7075453250272369, + "grad_norm": 1.102283000946045, + "learning_rate": 0.0006484480053787539, + "loss": 0.5096, + "step": 12664 + }, + { + "epoch": 0.7076011956309188, + "grad_norm": 0.6348732709884644, + "learning_rate": 0.0006484199910354101, + "loss": 0.4413, + "step": 12665 + }, + { + "epoch": 0.7076570662346007, + "grad_norm": 0.38413920998573303, + "learning_rate": 0.0006483919766920664, + "loss": 0.3469, + "step": 12666 + }, + { + "epoch": 0.7077129368382825, + "grad_norm": 0.5833083391189575, + "learning_rate": 0.0006483639623487225, + "loss": 0.5618, + "step": 12667 + }, + { + "epoch": 0.7077688074419645, + "grad_norm": 0.9728066325187683, + "learning_rate": 0.0006483359480053787, + "loss": 0.4628, + "step": 12668 + }, + { + "epoch": 0.7078246780456463, + "grad_norm": 0.5080624222755432, + "learning_rate": 0.000648307933662035, + "loss": 0.4804, + "step": 12669 + }, + { + "epoch": 0.7078805486493281, + "grad_norm": 0.5631571412086487, + "learning_rate": 0.0006482799193186912, + "loss": 0.4164, + "step": 12670 + }, + { + "epoch": 0.70793641925301, + "grad_norm": 0.43729498982429504, + "learning_rate": 0.0006482519049753473, + "loss": 0.4313, + "step": 12671 + }, + { + "epoch": 0.7079922898566919, + "grad_norm": 0.48494890332221985, + "learning_rate": 0.0006482238906320035, + "loss": 0.3311, + "step": 12672 + }, + { + "epoch": 0.7080481604603738, + "grad_norm": 0.492427796125412, + "learning_rate": 0.0006481958762886599, + "loss": 0.4531, + "step": 12673 + }, + { + "epoch": 0.7081040310640556, + "grad_norm": 0.4376203715801239, + "learning_rate": 0.000648167861945316, + "loss": 0.4466, + "step": 12674 + }, + { + "epoch": 0.7081599016677376, + "grad_norm": 0.5825027227401733, + "learning_rate": 0.0006481398476019722, + "loss": 0.4449, + "step": 12675 + }, + { + "epoch": 0.7082157722714194, + "grad_norm": 0.5889295339584351, + "learning_rate": 0.0006481118332586285, + "loss": 0.4249, + "step": 12676 + }, + { + "epoch": 0.7082716428751012, + "grad_norm": 0.40741175413131714, + "learning_rate": 0.0006480838189152847, + "loss": 0.4276, + "step": 12677 + }, + { + "epoch": 0.7083275134787831, + "grad_norm": 0.6613361239433289, + "learning_rate": 0.0006480558045719408, + "loss": 0.4565, + "step": 12678 + }, + { + "epoch": 0.708383384082465, + "grad_norm": 0.45276761054992676, + "learning_rate": 0.000648027790228597, + "loss": 0.5762, + "step": 12679 + }, + { + "epoch": 0.7084392546861469, + "grad_norm": 0.984363317489624, + "learning_rate": 0.0006479997758852533, + "loss": 0.4506, + "step": 12680 + }, + { + "epoch": 0.7084951252898287, + "grad_norm": 0.46131208539009094, + "learning_rate": 0.0006479717615419095, + "loss": 0.4549, + "step": 12681 + }, + { + "epoch": 0.7085509958935107, + "grad_norm": 0.566908061504364, + "learning_rate": 0.0006479437471985656, + "loss": 0.5106, + "step": 12682 + }, + { + "epoch": 0.7086068664971925, + "grad_norm": 0.36039188504219055, + "learning_rate": 0.0006479157328552219, + "loss": 0.3312, + "step": 12683 + }, + { + "epoch": 0.7086627371008744, + "grad_norm": 0.735946774482727, + "learning_rate": 0.0006478877185118781, + "loss": 0.4357, + "step": 12684 + }, + { + "epoch": 0.7087186077045563, + "grad_norm": 7.860744476318359, + "learning_rate": 0.0006478597041685343, + "loss": 0.4735, + "step": 12685 + }, + { + "epoch": 0.7087744783082381, + "grad_norm": 0.36996179819107056, + "learning_rate": 0.0006478316898251904, + "loss": 0.3982, + "step": 12686 + }, + { + "epoch": 0.70883034891192, + "grad_norm": 0.4047217071056366, + "learning_rate": 0.0006478036754818467, + "loss": 0.4037, + "step": 12687 + }, + { + "epoch": 0.7088862195156018, + "grad_norm": 0.9435495734214783, + "learning_rate": 0.000647775661138503, + "loss": 0.3665, + "step": 12688 + }, + { + "epoch": 0.7089420901192838, + "grad_norm": 0.41970014572143555, + "learning_rate": 0.0006477476467951592, + "loss": 0.4178, + "step": 12689 + }, + { + "epoch": 0.7089979607229656, + "grad_norm": 0.42563244700431824, + "learning_rate": 0.0006477196324518154, + "loss": 0.3952, + "step": 12690 + }, + { + "epoch": 0.7090538313266475, + "grad_norm": 0.5193626880645752, + "learning_rate": 0.0006476916181084716, + "loss": 0.45, + "step": 12691 + }, + { + "epoch": 0.7091097019303294, + "grad_norm": 0.4911293089389801, + "learning_rate": 0.0006476636037651278, + "loss": 0.3667, + "step": 12692 + }, + { + "epoch": 0.7091655725340112, + "grad_norm": 0.5935400128364563, + "learning_rate": 0.000647635589421784, + "loss": 0.5129, + "step": 12693 + }, + { + "epoch": 0.7092214431376931, + "grad_norm": 2.922717809677124, + "learning_rate": 0.0006476075750784402, + "loss": 0.4244, + "step": 12694 + }, + { + "epoch": 0.7092773137413749, + "grad_norm": 0.5871341228485107, + "learning_rate": 0.0006475795607350964, + "loss": 0.4318, + "step": 12695 + }, + { + "epoch": 0.7093331843450569, + "grad_norm": 3.9363930225372314, + "learning_rate": 0.0006475515463917526, + "loss": 0.4214, + "step": 12696 + }, + { + "epoch": 0.7093890549487387, + "grad_norm": 0.45426416397094727, + "learning_rate": 0.0006475235320484087, + "loss": 0.4033, + "step": 12697 + }, + { + "epoch": 0.7094449255524206, + "grad_norm": 0.7094062566757202, + "learning_rate": 0.000647495517705065, + "loss": 0.4826, + "step": 12698 + }, + { + "epoch": 0.7095007961561025, + "grad_norm": 0.7710968852043152, + "learning_rate": 0.0006474675033617212, + "loss": 0.4489, + "step": 12699 + }, + { + "epoch": 0.7095566667597843, + "grad_norm": 0.3787088692188263, + "learning_rate": 0.0006474394890183774, + "loss": 0.3671, + "step": 12700 + }, + { + "epoch": 0.7096125373634662, + "grad_norm": 0.3975520730018616, + "learning_rate": 0.0006474114746750336, + "loss": 0.4741, + "step": 12701 + }, + { + "epoch": 0.7096684079671481, + "grad_norm": 0.5002513527870178, + "learning_rate": 0.0006473834603316898, + "loss": 0.5212, + "step": 12702 + }, + { + "epoch": 0.70972427857083, + "grad_norm": 1.1043591499328613, + "learning_rate": 0.000647355445988346, + "loss": 0.4549, + "step": 12703 + }, + { + "epoch": 0.7097801491745118, + "grad_norm": 1.0098515748977661, + "learning_rate": 0.0006473274316450022, + "loss": 0.4971, + "step": 12704 + }, + { + "epoch": 0.7098360197781937, + "grad_norm": 0.35214024782180786, + "learning_rate": 0.0006472994173016584, + "loss": 0.426, + "step": 12705 + }, + { + "epoch": 0.7098918903818756, + "grad_norm": 0.551694929599762, + "learning_rate": 0.0006472714029583147, + "loss": 0.4502, + "step": 12706 + }, + { + "epoch": 0.7099477609855575, + "grad_norm": 0.5424116849899292, + "learning_rate": 0.0006472433886149709, + "loss": 0.5379, + "step": 12707 + }, + { + "epoch": 0.7100036315892393, + "grad_norm": 0.5020922422409058, + "learning_rate": 0.0006472153742716272, + "loss": 0.4394, + "step": 12708 + }, + { + "epoch": 0.7100595021929212, + "grad_norm": 0.4111941456794739, + "learning_rate": 0.0006471873599282833, + "loss": 0.5125, + "step": 12709 + }, + { + "epoch": 0.7101153727966031, + "grad_norm": 0.4151844382286072, + "learning_rate": 0.0006471593455849395, + "loss": 0.3898, + "step": 12710 + }, + { + "epoch": 0.7101712434002849, + "grad_norm": 0.41348397731781006, + "learning_rate": 0.0006471313312415957, + "loss": 0.4502, + "step": 12711 + }, + { + "epoch": 0.7102271140039668, + "grad_norm": 0.5175016522407532, + "learning_rate": 0.000647103316898252, + "loss": 0.3327, + "step": 12712 + }, + { + "epoch": 0.7102829846076487, + "grad_norm": 2.8703372478485107, + "learning_rate": 0.0006470753025549081, + "loss": 0.4554, + "step": 12713 + }, + { + "epoch": 0.7103388552113306, + "grad_norm": 0.4845414161682129, + "learning_rate": 0.0006470472882115643, + "loss": 0.4047, + "step": 12714 + }, + { + "epoch": 0.7103947258150124, + "grad_norm": 0.37285342812538147, + "learning_rate": 0.0006470192738682206, + "loss": 0.3783, + "step": 12715 + }, + { + "epoch": 0.7104505964186943, + "grad_norm": 0.573107123374939, + "learning_rate": 0.0006469912595248767, + "loss": 0.5592, + "step": 12716 + }, + { + "epoch": 0.7105064670223762, + "grad_norm": 0.6044049263000488, + "learning_rate": 0.0006469632451815329, + "loss": 0.5005, + "step": 12717 + }, + { + "epoch": 0.710562337626058, + "grad_norm": 0.4750550389289856, + "learning_rate": 0.0006469352308381891, + "loss": 0.5013, + "step": 12718 + }, + { + "epoch": 0.7106182082297399, + "grad_norm": 1.2174243927001953, + "learning_rate": 0.0006469072164948454, + "loss": 0.4833, + "step": 12719 + }, + { + "epoch": 0.7106740788334218, + "grad_norm": 0.6389219164848328, + "learning_rate": 0.0006468792021515015, + "loss": 0.4359, + "step": 12720 + }, + { + "epoch": 0.7107299494371037, + "grad_norm": 2.1323952674865723, + "learning_rate": 0.0006468511878081577, + "loss": 0.3986, + "step": 12721 + }, + { + "epoch": 0.7107858200407855, + "grad_norm": 0.4194241166114807, + "learning_rate": 0.0006468231734648141, + "loss": 0.4042, + "step": 12722 + }, + { + "epoch": 0.7108416906444674, + "grad_norm": 0.5750505924224854, + "learning_rate": 0.0006467951591214703, + "loss": 0.482, + "step": 12723 + }, + { + "epoch": 0.7108975612481493, + "grad_norm": 0.6212525963783264, + "learning_rate": 0.0006467671447781264, + "loss": 0.4057, + "step": 12724 + }, + { + "epoch": 0.7109534318518311, + "grad_norm": 0.4057518541812897, + "learning_rate": 0.0006467391304347826, + "loss": 0.414, + "step": 12725 + }, + { + "epoch": 0.7110093024555131, + "grad_norm": 0.5510073900222778, + "learning_rate": 0.0006467111160914389, + "loss": 0.416, + "step": 12726 + }, + { + "epoch": 0.7110651730591949, + "grad_norm": 0.38405588269233704, + "learning_rate": 0.0006466831017480951, + "loss": 0.327, + "step": 12727 + }, + { + "epoch": 0.7111210436628768, + "grad_norm": 0.40203917026519775, + "learning_rate": 0.0006466550874047512, + "loss": 0.4305, + "step": 12728 + }, + { + "epoch": 0.7111769142665586, + "grad_norm": 0.5213836431503296, + "learning_rate": 0.0006466270730614075, + "loss": 0.4961, + "step": 12729 + }, + { + "epoch": 0.7112327848702406, + "grad_norm": 0.7176899313926697, + "learning_rate": 0.0006465990587180637, + "loss": 0.4522, + "step": 12730 + }, + { + "epoch": 0.7112886554739224, + "grad_norm": 0.4406530559062958, + "learning_rate": 0.0006465710443747199, + "loss": 0.4204, + "step": 12731 + }, + { + "epoch": 0.7113445260776042, + "grad_norm": 0.44001373648643494, + "learning_rate": 0.000646543030031376, + "loss": 0.3889, + "step": 12732 + }, + { + "epoch": 0.7114003966812862, + "grad_norm": 1.0633901357650757, + "learning_rate": 0.0006465150156880323, + "loss": 0.5032, + "step": 12733 + }, + { + "epoch": 0.711456267284968, + "grad_norm": 0.6718016862869263, + "learning_rate": 0.0006464870013446885, + "loss": 0.4956, + "step": 12734 + }, + { + "epoch": 0.7115121378886499, + "grad_norm": 0.5228654742240906, + "learning_rate": 0.0006464589870013447, + "loss": 0.4117, + "step": 12735 + }, + { + "epoch": 0.7115680084923317, + "grad_norm": 0.8704761862754822, + "learning_rate": 0.0006464309726580008, + "loss": 0.4089, + "step": 12736 + }, + { + "epoch": 0.7116238790960137, + "grad_norm": 0.39860302209854126, + "learning_rate": 0.0006464029583146571, + "loss": 0.4489, + "step": 12737 + }, + { + "epoch": 0.7116797496996955, + "grad_norm": 0.5231443643569946, + "learning_rate": 0.0006463749439713133, + "loss": 0.3947, + "step": 12738 + }, + { + "epoch": 0.7117356203033773, + "grad_norm": 0.46453526616096497, + "learning_rate": 0.0006463469296279694, + "loss": 0.4116, + "step": 12739 + }, + { + "epoch": 0.7117914909070593, + "grad_norm": 0.46918049454689026, + "learning_rate": 0.0006463189152846258, + "loss": 0.425, + "step": 12740 + }, + { + "epoch": 0.7118473615107411, + "grad_norm": 0.4185551702976227, + "learning_rate": 0.000646290900941282, + "loss": 0.4549, + "step": 12741 + }, + { + "epoch": 0.711903232114423, + "grad_norm": 1.3084030151367188, + "learning_rate": 0.0006462628865979382, + "loss": 0.4593, + "step": 12742 + }, + { + "epoch": 0.7119591027181049, + "grad_norm": 0.7099047899246216, + "learning_rate": 0.0006462348722545943, + "loss": 0.4062, + "step": 12743 + }, + { + "epoch": 0.7120149733217868, + "grad_norm": 0.4754036068916321, + "learning_rate": 0.0006462068579112506, + "loss": 0.496, + "step": 12744 + }, + { + "epoch": 0.7120708439254686, + "grad_norm": 0.42506977915763855, + "learning_rate": 0.0006461788435679068, + "loss": 0.466, + "step": 12745 + }, + { + "epoch": 0.7121267145291504, + "grad_norm": 0.6062296032905579, + "learning_rate": 0.000646150829224563, + "loss": 0.4856, + "step": 12746 + }, + { + "epoch": 0.7121825851328324, + "grad_norm": 1.0281435251235962, + "learning_rate": 0.0006461228148812192, + "loss": 0.5751, + "step": 12747 + }, + { + "epoch": 0.7122384557365142, + "grad_norm": 0.6654332280158997, + "learning_rate": 0.0006460948005378754, + "loss": 0.3932, + "step": 12748 + }, + { + "epoch": 0.7122943263401961, + "grad_norm": 0.5775181651115417, + "learning_rate": 0.0006460667861945316, + "loss": 0.5653, + "step": 12749 + }, + { + "epoch": 0.712350196943878, + "grad_norm": 0.7709896564483643, + "learning_rate": 0.0006460387718511878, + "loss": 0.4815, + "step": 12750 + }, + { + "epoch": 0.7124060675475599, + "grad_norm": 0.5890455842018127, + "learning_rate": 0.000646010757507844, + "loss": 0.4711, + "step": 12751 + }, + { + "epoch": 0.7124619381512417, + "grad_norm": 0.42524826526641846, + "learning_rate": 0.0006459827431645002, + "loss": 0.3964, + "step": 12752 + }, + { + "epoch": 0.7125178087549235, + "grad_norm": 0.6964998245239258, + "learning_rate": 0.0006459547288211564, + "loss": 0.4245, + "step": 12753 + }, + { + "epoch": 0.7125736793586055, + "grad_norm": 5.262060165405273, + "learning_rate": 0.0006459267144778128, + "loss": 0.4669, + "step": 12754 + }, + { + "epoch": 0.7126295499622873, + "grad_norm": 1.5230352878570557, + "learning_rate": 0.0006458987001344688, + "loss": 0.4332, + "step": 12755 + }, + { + "epoch": 0.7126854205659692, + "grad_norm": 0.4491397738456726, + "learning_rate": 0.0006458706857911251, + "loss": 0.4939, + "step": 12756 + }, + { + "epoch": 0.7127412911696511, + "grad_norm": 0.5471433401107788, + "learning_rate": 0.0006458426714477813, + "loss": 0.6437, + "step": 12757 + }, + { + "epoch": 0.712797161773333, + "grad_norm": 0.46557602286338806, + "learning_rate": 0.0006458146571044375, + "loss": 0.3728, + "step": 12758 + }, + { + "epoch": 0.7128530323770148, + "grad_norm": 0.8584389090538025, + "learning_rate": 0.0006457866427610937, + "loss": 0.3112, + "step": 12759 + }, + { + "epoch": 0.7129089029806968, + "grad_norm": 1.48554527759552, + "learning_rate": 0.0006457586284177499, + "loss": 0.4, + "step": 12760 + }, + { + "epoch": 0.7129647735843786, + "grad_norm": 0.42930033802986145, + "learning_rate": 0.0006457306140744062, + "loss": 0.4792, + "step": 12761 + }, + { + "epoch": 0.7130206441880604, + "grad_norm": 0.5415902137756348, + "learning_rate": 0.0006457025997310623, + "loss": 0.459, + "step": 12762 + }, + { + "epoch": 0.7130765147917423, + "grad_norm": 0.5728740692138672, + "learning_rate": 0.0006456745853877185, + "loss": 0.6325, + "step": 12763 + }, + { + "epoch": 0.7131323853954242, + "grad_norm": 0.5203841924667358, + "learning_rate": 0.0006456465710443747, + "loss": 0.5743, + "step": 12764 + }, + { + "epoch": 0.7131882559991061, + "grad_norm": 0.42383748292922974, + "learning_rate": 0.000645618556701031, + "loss": 0.4223, + "step": 12765 + }, + { + "epoch": 0.7132441266027879, + "grad_norm": 0.49071547389030457, + "learning_rate": 0.0006455905423576871, + "loss": 0.5933, + "step": 12766 + }, + { + "epoch": 0.7132999972064699, + "grad_norm": 0.5173240900039673, + "learning_rate": 0.0006455625280143433, + "loss": 0.4863, + "step": 12767 + }, + { + "epoch": 0.7133558678101517, + "grad_norm": 0.5147274732589722, + "learning_rate": 0.0006455345136709996, + "loss": 0.3944, + "step": 12768 + }, + { + "epoch": 0.7134117384138335, + "grad_norm": 0.5425196886062622, + "learning_rate": 0.0006455064993276558, + "loss": 0.3618, + "step": 12769 + }, + { + "epoch": 0.7134676090175154, + "grad_norm": 1.8585262298583984, + "learning_rate": 0.0006454784849843119, + "loss": 0.5622, + "step": 12770 + }, + { + "epoch": 0.7135234796211973, + "grad_norm": 0.5909469127655029, + "learning_rate": 0.0006454504706409681, + "loss": 0.4046, + "step": 12771 + }, + { + "epoch": 0.7135793502248792, + "grad_norm": 0.6171122193336487, + "learning_rate": 0.0006454224562976245, + "loss": 0.4462, + "step": 12772 + }, + { + "epoch": 0.713635220828561, + "grad_norm": 1.0028129816055298, + "learning_rate": 0.0006453944419542807, + "loss": 0.5156, + "step": 12773 + }, + { + "epoch": 0.713691091432243, + "grad_norm": 0.3962969183921814, + "learning_rate": 0.0006453664276109368, + "loss": 0.4246, + "step": 12774 + }, + { + "epoch": 0.7137469620359248, + "grad_norm": 0.33679407835006714, + "learning_rate": 0.000645338413267593, + "loss": 0.3425, + "step": 12775 + }, + { + "epoch": 0.7138028326396066, + "grad_norm": 0.597343921661377, + "learning_rate": 0.0006453103989242493, + "loss": 0.5506, + "step": 12776 + }, + { + "epoch": 0.7138587032432886, + "grad_norm": 0.7353470921516418, + "learning_rate": 0.0006452823845809055, + "loss": 0.3746, + "step": 12777 + }, + { + "epoch": 0.7139145738469704, + "grad_norm": 0.7856991291046143, + "learning_rate": 0.0006452543702375616, + "loss": 0.485, + "step": 12778 + }, + { + "epoch": 0.7139704444506523, + "grad_norm": 0.450420618057251, + "learning_rate": 0.0006452263558942179, + "loss": 0.4024, + "step": 12779 + }, + { + "epoch": 0.7140263150543341, + "grad_norm": 1.8229130506515503, + "learning_rate": 0.0006451983415508741, + "loss": 0.3907, + "step": 12780 + }, + { + "epoch": 0.7140821856580161, + "grad_norm": 0.5820293426513672, + "learning_rate": 0.0006451703272075302, + "loss": 0.4479, + "step": 12781 + }, + { + "epoch": 0.7141380562616979, + "grad_norm": 0.4386962950229645, + "learning_rate": 0.0006451423128641864, + "loss": 0.3974, + "step": 12782 + }, + { + "epoch": 0.7141939268653797, + "grad_norm": 0.9503467679023743, + "learning_rate": 0.0006451142985208427, + "loss": 0.3247, + "step": 12783 + }, + { + "epoch": 0.7142497974690617, + "grad_norm": 0.4625675678253174, + "learning_rate": 0.0006450862841774989, + "loss": 0.4419, + "step": 12784 + }, + { + "epoch": 0.7143056680727435, + "grad_norm": 0.43302080035209656, + "learning_rate": 0.000645058269834155, + "loss": 0.3552, + "step": 12785 + }, + { + "epoch": 0.7143615386764254, + "grad_norm": 0.37406429648399353, + "learning_rate": 0.0006450302554908113, + "loss": 0.409, + "step": 12786 + }, + { + "epoch": 0.7144174092801072, + "grad_norm": 0.5006958842277527, + "learning_rate": 0.0006450022411474675, + "loss": 0.404, + "step": 12787 + }, + { + "epoch": 0.7144732798837892, + "grad_norm": 7.024693489074707, + "learning_rate": 0.0006449742268041237, + "loss": 0.4136, + "step": 12788 + }, + { + "epoch": 0.714529150487471, + "grad_norm": 0.5154906511306763, + "learning_rate": 0.0006449462124607798, + "loss": 0.4615, + "step": 12789 + }, + { + "epoch": 0.7145850210911528, + "grad_norm": 0.505351722240448, + "learning_rate": 0.0006449181981174362, + "loss": 0.3964, + "step": 12790 + }, + { + "epoch": 0.7146408916948348, + "grad_norm": 0.426897257566452, + "learning_rate": 0.0006448901837740924, + "loss": 0.3729, + "step": 12791 + }, + { + "epoch": 0.7146967622985166, + "grad_norm": 4.76810359954834, + "learning_rate": 0.0006448621694307486, + "loss": 0.4131, + "step": 12792 + }, + { + "epoch": 0.7147526329021985, + "grad_norm": 0.8412461876869202, + "learning_rate": 0.0006448341550874048, + "loss": 0.3824, + "step": 12793 + }, + { + "epoch": 0.7148085035058804, + "grad_norm": 1.4258438348770142, + "learning_rate": 0.000644806140744061, + "loss": 0.3125, + "step": 12794 + }, + { + "epoch": 0.7148643741095623, + "grad_norm": 0.5234453082084656, + "learning_rate": 0.0006447781264007172, + "loss": 0.4315, + "step": 12795 + }, + { + "epoch": 0.7149202447132441, + "grad_norm": 0.4359224736690521, + "learning_rate": 0.0006447501120573734, + "loss": 0.3444, + "step": 12796 + }, + { + "epoch": 0.714976115316926, + "grad_norm": 1.007910132408142, + "learning_rate": 0.0006447220977140296, + "loss": 0.4598, + "step": 12797 + }, + { + "epoch": 0.7150319859206079, + "grad_norm": 0.47854241728782654, + "learning_rate": 0.0006446940833706858, + "loss": 0.4748, + "step": 12798 + }, + { + "epoch": 0.7150878565242897, + "grad_norm": 3.453758478164673, + "learning_rate": 0.000644666069027342, + "loss": 0.3812, + "step": 12799 + }, + { + "epoch": 0.7151437271279716, + "grad_norm": 3.238358736038208, + "learning_rate": 0.0006446380546839982, + "loss": 0.3572, + "step": 12800 + }, + { + "epoch": 0.7151995977316535, + "grad_norm": 0.5310251712799072, + "learning_rate": 0.0006446100403406544, + "loss": 0.379, + "step": 12801 + }, + { + "epoch": 0.7152554683353354, + "grad_norm": 0.8712336421012878, + "learning_rate": 0.0006445820259973106, + "loss": 0.5877, + "step": 12802 + }, + { + "epoch": 0.7153113389390172, + "grad_norm": 0.7183954119682312, + "learning_rate": 0.0006445540116539668, + "loss": 0.4977, + "step": 12803 + }, + { + "epoch": 0.7153672095426991, + "grad_norm": 0.4659424126148224, + "learning_rate": 0.000644525997310623, + "loss": 0.5027, + "step": 12804 + }, + { + "epoch": 0.715423080146381, + "grad_norm": 0.4375970959663391, + "learning_rate": 0.0006444979829672792, + "loss": 0.4503, + "step": 12805 + }, + { + "epoch": 0.7154789507500628, + "grad_norm": 0.44976508617401123, + "learning_rate": 0.0006444699686239355, + "loss": 0.5625, + "step": 12806 + }, + { + "epoch": 0.7155348213537447, + "grad_norm": 6.47481107711792, + "learning_rate": 0.0006444419542805917, + "loss": 0.4286, + "step": 12807 + }, + { + "epoch": 0.7155906919574266, + "grad_norm": 0.5175030827522278, + "learning_rate": 0.0006444139399372479, + "loss": 0.405, + "step": 12808 + }, + { + "epoch": 0.7156465625611085, + "grad_norm": 0.3775107264518738, + "learning_rate": 0.0006443859255939041, + "loss": 0.405, + "step": 12809 + }, + { + "epoch": 0.7157024331647903, + "grad_norm": 0.4839082956314087, + "learning_rate": 0.0006443579112505603, + "loss": 0.4146, + "step": 12810 + }, + { + "epoch": 0.7157583037684723, + "grad_norm": 0.41997361183166504, + "learning_rate": 0.0006443298969072166, + "loss": 0.5054, + "step": 12811 + }, + { + "epoch": 0.7158141743721541, + "grad_norm": 0.5047481656074524, + "learning_rate": 0.0006443018825638727, + "loss": 0.4644, + "step": 12812 + }, + { + "epoch": 0.715870044975836, + "grad_norm": 1.6705337762832642, + "learning_rate": 0.0006442738682205289, + "loss": 0.5906, + "step": 12813 + }, + { + "epoch": 0.7159259155795178, + "grad_norm": 0.3982970118522644, + "learning_rate": 0.0006442458538771851, + "loss": 0.4569, + "step": 12814 + }, + { + "epoch": 0.7159817861831997, + "grad_norm": 1.1782114505767822, + "learning_rate": 0.0006442178395338414, + "loss": 0.3863, + "step": 12815 + }, + { + "epoch": 0.7160376567868816, + "grad_norm": 0.5812097191810608, + "learning_rate": 0.0006441898251904975, + "loss": 0.4452, + "step": 12816 + }, + { + "epoch": 0.7160935273905634, + "grad_norm": 0.6065754294395447, + "learning_rate": 0.0006441618108471537, + "loss": 0.4599, + "step": 12817 + }, + { + "epoch": 0.7161493979942454, + "grad_norm": 0.46953535079956055, + "learning_rate": 0.00064413379650381, + "loss": 0.4805, + "step": 12818 + }, + { + "epoch": 0.7162052685979272, + "grad_norm": 0.5913798809051514, + "learning_rate": 0.0006441057821604662, + "loss": 0.4838, + "step": 12819 + }, + { + "epoch": 0.716261139201609, + "grad_norm": 0.5376279950141907, + "learning_rate": 0.0006440777678171223, + "loss": 0.3553, + "step": 12820 + }, + { + "epoch": 0.7163170098052909, + "grad_norm": 1.624619960784912, + "learning_rate": 0.0006440497534737785, + "loss": 0.4803, + "step": 12821 + }, + { + "epoch": 0.7163728804089728, + "grad_norm": 0.6881061792373657, + "learning_rate": 0.0006440217391304349, + "loss": 0.5108, + "step": 12822 + }, + { + "epoch": 0.7164287510126547, + "grad_norm": 0.5574030876159668, + "learning_rate": 0.000643993724787091, + "loss": 0.3787, + "step": 12823 + }, + { + "epoch": 0.7164846216163365, + "grad_norm": 0.47949638962745667, + "learning_rate": 0.0006439657104437472, + "loss": 0.4332, + "step": 12824 + }, + { + "epoch": 0.7165404922200185, + "grad_norm": 0.6040207147598267, + "learning_rate": 0.0006439376961004035, + "loss": 0.53, + "step": 12825 + }, + { + "epoch": 0.7165963628237003, + "grad_norm": 0.49500221014022827, + "learning_rate": 0.0006439096817570597, + "loss": 0.4801, + "step": 12826 + }, + { + "epoch": 0.7166522334273822, + "grad_norm": 0.41759371757507324, + "learning_rate": 0.0006438816674137158, + "loss": 0.5164, + "step": 12827 + }, + { + "epoch": 0.7167081040310641, + "grad_norm": 0.5543555021286011, + "learning_rate": 0.000643853653070372, + "loss": 0.4842, + "step": 12828 + }, + { + "epoch": 0.716763974634746, + "grad_norm": 0.8573716878890991, + "learning_rate": 0.0006438256387270283, + "loss": 0.6575, + "step": 12829 + }, + { + "epoch": 0.7168198452384278, + "grad_norm": 0.42482444643974304, + "learning_rate": 0.0006437976243836845, + "loss": 0.45, + "step": 12830 + }, + { + "epoch": 0.7168757158421096, + "grad_norm": 0.5925284028053284, + "learning_rate": 0.0006437696100403406, + "loss": 0.4717, + "step": 12831 + }, + { + "epoch": 0.7169315864457916, + "grad_norm": 0.36851856112480164, + "learning_rate": 0.0006437415956969969, + "loss": 0.3456, + "step": 12832 + }, + { + "epoch": 0.7169874570494734, + "grad_norm": 0.4673929512500763, + "learning_rate": 0.0006437135813536531, + "loss": 0.5037, + "step": 12833 + }, + { + "epoch": 0.7170433276531553, + "grad_norm": 2.342897891998291, + "learning_rate": 0.0006436855670103093, + "loss": 0.428, + "step": 12834 + }, + { + "epoch": 0.7170991982568372, + "grad_norm": 0.5797029733657837, + "learning_rate": 0.0006436575526669654, + "loss": 0.4793, + "step": 12835 + }, + { + "epoch": 0.717155068860519, + "grad_norm": 0.5523281693458557, + "learning_rate": 0.0006436295383236217, + "loss": 0.4754, + "step": 12836 + }, + { + "epoch": 0.7172109394642009, + "grad_norm": 2.5703673362731934, + "learning_rate": 0.0006436015239802779, + "loss": 0.393, + "step": 12837 + }, + { + "epoch": 0.7172668100678827, + "grad_norm": 0.58282071352005, + "learning_rate": 0.0006435735096369341, + "loss": 0.4245, + "step": 12838 + }, + { + "epoch": 0.7173226806715647, + "grad_norm": 0.7467092871665955, + "learning_rate": 0.0006435454952935902, + "loss": 0.7187, + "step": 12839 + }, + { + "epoch": 0.7173785512752465, + "grad_norm": 0.3128630518913269, + "learning_rate": 0.0006435174809502466, + "loss": 0.4449, + "step": 12840 + }, + { + "epoch": 0.7174344218789284, + "grad_norm": 0.36210504174232483, + "learning_rate": 0.0006434894666069028, + "loss": 0.3582, + "step": 12841 + }, + { + "epoch": 0.7174902924826103, + "grad_norm": 0.4496181607246399, + "learning_rate": 0.000643461452263559, + "loss": 0.3992, + "step": 12842 + }, + { + "epoch": 0.7175461630862922, + "grad_norm": 0.5741847157478333, + "learning_rate": 0.0006434334379202152, + "loss": 0.3332, + "step": 12843 + }, + { + "epoch": 0.717602033689974, + "grad_norm": 0.7603506445884705, + "learning_rate": 0.0006434054235768714, + "loss": 0.5877, + "step": 12844 + }, + { + "epoch": 0.717657904293656, + "grad_norm": 0.7613046765327454, + "learning_rate": 0.0006433774092335276, + "loss": 0.4593, + "step": 12845 + }, + { + "epoch": 0.7177137748973378, + "grad_norm": 0.43357738852500916, + "learning_rate": 0.0006433493948901837, + "loss": 0.4452, + "step": 12846 + }, + { + "epoch": 0.7177696455010196, + "grad_norm": 0.45261117815971375, + "learning_rate": 0.00064332138054684, + "loss": 0.4401, + "step": 12847 + }, + { + "epoch": 0.7178255161047015, + "grad_norm": 0.551061749458313, + "learning_rate": 0.0006432933662034962, + "loss": 0.5222, + "step": 12848 + }, + { + "epoch": 0.7178813867083834, + "grad_norm": 0.5052083134651184, + "learning_rate": 0.0006432653518601524, + "loss": 0.5721, + "step": 12849 + }, + { + "epoch": 0.7179372573120653, + "grad_norm": 0.40800604224205017, + "learning_rate": 0.0006432373375168086, + "loss": 0.4506, + "step": 12850 + }, + { + "epoch": 0.7179931279157471, + "grad_norm": 0.9120592474937439, + "learning_rate": 0.0006432093231734648, + "loss": 0.3428, + "step": 12851 + }, + { + "epoch": 0.718048998519429, + "grad_norm": 0.4410371482372284, + "learning_rate": 0.000643181308830121, + "loss": 0.4413, + "step": 12852 + }, + { + "epoch": 0.7181048691231109, + "grad_norm": 0.3988586664199829, + "learning_rate": 0.0006431532944867772, + "loss": 0.3891, + "step": 12853 + }, + { + "epoch": 0.7181607397267927, + "grad_norm": 0.5389788746833801, + "learning_rate": 0.0006431252801434334, + "loss": 0.4777, + "step": 12854 + }, + { + "epoch": 0.7182166103304746, + "grad_norm": 0.4489149749279022, + "learning_rate": 0.0006430972658000896, + "loss": 0.3308, + "step": 12855 + }, + { + "epoch": 0.7182724809341565, + "grad_norm": 0.5170374512672424, + "learning_rate": 0.0006430692514567459, + "loss": 0.3702, + "step": 12856 + }, + { + "epoch": 0.7183283515378384, + "grad_norm": 0.4030223786830902, + "learning_rate": 0.0006430412371134022, + "loss": 0.372, + "step": 12857 + }, + { + "epoch": 0.7183842221415202, + "grad_norm": 0.7302391529083252, + "learning_rate": 0.0006430132227700583, + "loss": 0.4141, + "step": 12858 + }, + { + "epoch": 0.7184400927452022, + "grad_norm": 0.4951738119125366, + "learning_rate": 0.0006429852084267145, + "loss": 0.4324, + "step": 12859 + }, + { + "epoch": 0.718495963348884, + "grad_norm": 0.39017385244369507, + "learning_rate": 0.0006429571940833707, + "loss": 0.4462, + "step": 12860 + }, + { + "epoch": 0.7185518339525658, + "grad_norm": 0.4498579800128937, + "learning_rate": 0.000642929179740027, + "loss": 0.4423, + "step": 12861 + }, + { + "epoch": 0.7186077045562478, + "grad_norm": 0.47517311573028564, + "learning_rate": 0.0006429011653966831, + "loss": 0.4953, + "step": 12862 + }, + { + "epoch": 0.7186635751599296, + "grad_norm": 0.3745068907737732, + "learning_rate": 0.0006428731510533393, + "loss": 0.4652, + "step": 12863 + }, + { + "epoch": 0.7187194457636115, + "grad_norm": 0.7669312357902527, + "learning_rate": 0.0006428451367099956, + "loss": 0.7803, + "step": 12864 + }, + { + "epoch": 0.7187753163672933, + "grad_norm": 0.39866387844085693, + "learning_rate": 0.0006428171223666517, + "loss": 0.4146, + "step": 12865 + }, + { + "epoch": 0.7188311869709753, + "grad_norm": 0.7580296993255615, + "learning_rate": 0.0006427891080233079, + "loss": 0.4806, + "step": 12866 + }, + { + "epoch": 0.7188870575746571, + "grad_norm": 0.4651464521884918, + "learning_rate": 0.0006427610936799641, + "loss": 0.3779, + "step": 12867 + }, + { + "epoch": 0.7189429281783389, + "grad_norm": 0.42842426896095276, + "learning_rate": 0.0006427330793366204, + "loss": 0.4074, + "step": 12868 + }, + { + "epoch": 0.7189987987820209, + "grad_norm": 0.7584421634674072, + "learning_rate": 0.0006427050649932765, + "loss": 0.5088, + "step": 12869 + }, + { + "epoch": 0.7190546693857027, + "grad_norm": 0.3668591380119324, + "learning_rate": 0.0006426770506499327, + "loss": 0.4609, + "step": 12870 + }, + { + "epoch": 0.7191105399893846, + "grad_norm": 0.880664587020874, + "learning_rate": 0.000642649036306589, + "loss": 0.469, + "step": 12871 + }, + { + "epoch": 0.7191664105930664, + "grad_norm": 0.7287766337394714, + "learning_rate": 0.0006426210219632453, + "loss": 0.4552, + "step": 12872 + }, + { + "epoch": 0.7192222811967484, + "grad_norm": 0.4967601001262665, + "learning_rate": 0.0006425930076199014, + "loss": 0.4099, + "step": 12873 + }, + { + "epoch": 0.7192781518004302, + "grad_norm": 0.7067447304725647, + "learning_rate": 0.0006425649932765576, + "loss": 0.4938, + "step": 12874 + }, + { + "epoch": 0.719334022404112, + "grad_norm": 0.3627125024795532, + "learning_rate": 0.0006425369789332139, + "loss": 0.3653, + "step": 12875 + }, + { + "epoch": 0.719389893007794, + "grad_norm": 0.3776651918888092, + "learning_rate": 0.0006425089645898701, + "loss": 0.467, + "step": 12876 + }, + { + "epoch": 0.7194457636114758, + "grad_norm": 0.5178673267364502, + "learning_rate": 0.0006424809502465262, + "loss": 0.3621, + "step": 12877 + }, + { + "epoch": 0.7195016342151577, + "grad_norm": 1.6788049936294556, + "learning_rate": 0.0006424529359031824, + "loss": 0.4689, + "step": 12878 + }, + { + "epoch": 0.7195575048188396, + "grad_norm": 0.3956100344657898, + "learning_rate": 0.0006424249215598387, + "loss": 0.4435, + "step": 12879 + }, + { + "epoch": 0.7196133754225215, + "grad_norm": 0.3587244153022766, + "learning_rate": 0.0006423969072164949, + "loss": 0.3586, + "step": 12880 + }, + { + "epoch": 0.7196692460262033, + "grad_norm": 4.005620956420898, + "learning_rate": 0.000642368892873151, + "loss": 0.4262, + "step": 12881 + }, + { + "epoch": 0.7197251166298851, + "grad_norm": 0.4172760546207428, + "learning_rate": 0.0006423408785298073, + "loss": 0.4014, + "step": 12882 + }, + { + "epoch": 0.7197809872335671, + "grad_norm": 0.4992939531803131, + "learning_rate": 0.0006423128641864635, + "loss": 0.532, + "step": 12883 + }, + { + "epoch": 0.7198368578372489, + "grad_norm": 0.3785145580768585, + "learning_rate": 0.0006422848498431197, + "loss": 0.4685, + "step": 12884 + }, + { + "epoch": 0.7198927284409308, + "grad_norm": 0.45876529812812805, + "learning_rate": 0.0006422568354997758, + "loss": 0.2419, + "step": 12885 + }, + { + "epoch": 0.7199485990446127, + "grad_norm": 0.6058355569839478, + "learning_rate": 0.0006422288211564321, + "loss": 0.481, + "step": 12886 + }, + { + "epoch": 0.7200044696482946, + "grad_norm": 1.7238153219223022, + "learning_rate": 0.0006422008068130883, + "loss": 0.4587, + "step": 12887 + }, + { + "epoch": 0.7200603402519764, + "grad_norm": 0.5429101586341858, + "learning_rate": 0.0006421727924697444, + "loss": 0.4886, + "step": 12888 + }, + { + "epoch": 0.7201162108556582, + "grad_norm": 0.5299680829048157, + "learning_rate": 0.0006421447781264008, + "loss": 0.3894, + "step": 12889 + }, + { + "epoch": 0.7201720814593402, + "grad_norm": 0.49605318903923035, + "learning_rate": 0.000642116763783057, + "loss": 0.3726, + "step": 12890 + }, + { + "epoch": 0.720227952063022, + "grad_norm": 0.34402996301651, + "learning_rate": 0.0006420887494397132, + "loss": 0.4472, + "step": 12891 + }, + { + "epoch": 0.7202838226667039, + "grad_norm": 0.4127860367298126, + "learning_rate": 0.0006420607350963693, + "loss": 0.4595, + "step": 12892 + }, + { + "epoch": 0.7203396932703858, + "grad_norm": 0.37618330121040344, + "learning_rate": 0.0006420327207530256, + "loss": 0.3721, + "step": 12893 + }, + { + "epoch": 0.7203955638740677, + "grad_norm": 0.8608587980270386, + "learning_rate": 0.0006420047064096818, + "loss": 0.5019, + "step": 12894 + }, + { + "epoch": 0.7204514344777495, + "grad_norm": 0.5843793153762817, + "learning_rate": 0.000641976692066338, + "loss": 0.4695, + "step": 12895 + }, + { + "epoch": 0.7205073050814315, + "grad_norm": 0.48401957750320435, + "learning_rate": 0.0006419486777229942, + "loss": 0.5047, + "step": 12896 + }, + { + "epoch": 0.7205631756851133, + "grad_norm": 0.4059199094772339, + "learning_rate": 0.0006419206633796504, + "loss": 0.5222, + "step": 12897 + }, + { + "epoch": 0.7206190462887951, + "grad_norm": 0.4448688328266144, + "learning_rate": 0.0006418926490363066, + "loss": 0.4475, + "step": 12898 + }, + { + "epoch": 0.720674916892477, + "grad_norm": 0.5047444701194763, + "learning_rate": 0.0006418646346929628, + "loss": 0.4494, + "step": 12899 + }, + { + "epoch": 0.7207307874961589, + "grad_norm": 0.5241328477859497, + "learning_rate": 0.000641836620349619, + "loss": 0.5829, + "step": 12900 + }, + { + "epoch": 0.7207866580998408, + "grad_norm": 0.4265141487121582, + "learning_rate": 0.0006418086060062752, + "loss": 0.3179, + "step": 12901 + }, + { + "epoch": 0.7208425287035226, + "grad_norm": 0.38838356733322144, + "learning_rate": 0.0006417805916629314, + "loss": 0.4013, + "step": 12902 + }, + { + "epoch": 0.7208983993072046, + "grad_norm": 0.5480681657791138, + "learning_rate": 0.0006417525773195877, + "loss": 0.493, + "step": 12903 + }, + { + "epoch": 0.7209542699108864, + "grad_norm": 0.540864884853363, + "learning_rate": 0.0006417245629762438, + "loss": 0.4258, + "step": 12904 + }, + { + "epoch": 0.7210101405145682, + "grad_norm": 0.4705210030078888, + "learning_rate": 0.0006416965486329, + "loss": 0.5382, + "step": 12905 + }, + { + "epoch": 0.7210660111182501, + "grad_norm": 0.632996141910553, + "learning_rate": 0.0006416685342895563, + "loss": 0.4732, + "step": 12906 + }, + { + "epoch": 0.721121881721932, + "grad_norm": 0.37339267134666443, + "learning_rate": 0.0006416405199462125, + "loss": 0.4028, + "step": 12907 + }, + { + "epoch": 0.7211777523256139, + "grad_norm": 0.35705432295799255, + "learning_rate": 0.0006416125056028687, + "loss": 0.3232, + "step": 12908 + }, + { + "epoch": 0.7212336229292957, + "grad_norm": 1.5038275718688965, + "learning_rate": 0.0006415844912595249, + "loss": 0.4969, + "step": 12909 + }, + { + "epoch": 0.7212894935329777, + "grad_norm": 0.7750093936920166, + "learning_rate": 0.0006415564769161812, + "loss": 0.4846, + "step": 12910 + }, + { + "epoch": 0.7213453641366595, + "grad_norm": 0.5140912532806396, + "learning_rate": 0.0006415284625728373, + "loss": 0.4418, + "step": 12911 + }, + { + "epoch": 0.7214012347403413, + "grad_norm": 0.44574958086013794, + "learning_rate": 0.0006415004482294935, + "loss": 0.6036, + "step": 12912 + }, + { + "epoch": 0.7214571053440233, + "grad_norm": 0.578026533126831, + "learning_rate": 0.0006414724338861497, + "loss": 0.4186, + "step": 12913 + }, + { + "epoch": 0.7215129759477051, + "grad_norm": 0.5250725746154785, + "learning_rate": 0.000641444419542806, + "loss": 0.4122, + "step": 12914 + }, + { + "epoch": 0.721568846551387, + "grad_norm": 0.40875622630119324, + "learning_rate": 0.0006414164051994621, + "loss": 0.5027, + "step": 12915 + }, + { + "epoch": 0.7216247171550688, + "grad_norm": 0.58430415391922, + "learning_rate": 0.0006413883908561183, + "loss": 0.7234, + "step": 12916 + }, + { + "epoch": 0.7216805877587508, + "grad_norm": 0.971065878868103, + "learning_rate": 0.0006413603765127745, + "loss": 0.5122, + "step": 12917 + }, + { + "epoch": 0.7217364583624326, + "grad_norm": 0.8583533763885498, + "learning_rate": 0.0006413323621694308, + "loss": 0.4992, + "step": 12918 + }, + { + "epoch": 0.7217923289661144, + "grad_norm": 0.5172237157821655, + "learning_rate": 0.0006413043478260869, + "loss": 0.4026, + "step": 12919 + }, + { + "epoch": 0.7218481995697964, + "grad_norm": 0.4711349904537201, + "learning_rate": 0.0006412763334827431, + "loss": 0.3424, + "step": 12920 + }, + { + "epoch": 0.7219040701734782, + "grad_norm": 0.46587976813316345, + "learning_rate": 0.0006412483191393994, + "loss": 0.4376, + "step": 12921 + }, + { + "epoch": 0.7219599407771601, + "grad_norm": 0.8967381119728088, + "learning_rate": 0.0006412203047960557, + "loss": 0.3944, + "step": 12922 + }, + { + "epoch": 0.7220158113808419, + "grad_norm": 0.6537917256355286, + "learning_rate": 0.0006411922904527118, + "loss": 0.3793, + "step": 12923 + }, + { + "epoch": 0.7220716819845239, + "grad_norm": 0.4528687298297882, + "learning_rate": 0.000641164276109368, + "loss": 0.5002, + "step": 12924 + }, + { + "epoch": 0.7221275525882057, + "grad_norm": 0.3774677813053131, + "learning_rate": 0.0006411362617660243, + "loss": 0.453, + "step": 12925 + }, + { + "epoch": 0.7221834231918876, + "grad_norm": 0.6748971939086914, + "learning_rate": 0.0006411082474226805, + "loss": 0.3341, + "step": 12926 + }, + { + "epoch": 0.7222392937955695, + "grad_norm": 0.5715247988700867, + "learning_rate": 0.0006410802330793366, + "loss": 0.3962, + "step": 12927 + }, + { + "epoch": 0.7222951643992513, + "grad_norm": 0.9146735668182373, + "learning_rate": 0.0006410522187359929, + "loss": 0.6168, + "step": 12928 + }, + { + "epoch": 0.7223510350029332, + "grad_norm": 0.6795123815536499, + "learning_rate": 0.0006410242043926491, + "loss": 0.6897, + "step": 12929 + }, + { + "epoch": 0.722406905606615, + "grad_norm": 0.45049697160720825, + "learning_rate": 0.0006409961900493052, + "loss": 0.5255, + "step": 12930 + }, + { + "epoch": 0.722462776210297, + "grad_norm": 0.5404261946678162, + "learning_rate": 0.0006409681757059614, + "loss": 0.4295, + "step": 12931 + }, + { + "epoch": 0.7225186468139788, + "grad_norm": 0.4454360902309418, + "learning_rate": 0.0006409401613626177, + "loss": 0.4391, + "step": 12932 + }, + { + "epoch": 0.7225745174176607, + "grad_norm": 0.6281161308288574, + "learning_rate": 0.0006409121470192739, + "loss": 0.4169, + "step": 12933 + }, + { + "epoch": 0.7226303880213426, + "grad_norm": 0.5790579915046692, + "learning_rate": 0.00064088413267593, + "loss": 0.3826, + "step": 12934 + }, + { + "epoch": 0.7226862586250244, + "grad_norm": 0.651255190372467, + "learning_rate": 0.0006408561183325863, + "loss": 0.446, + "step": 12935 + }, + { + "epoch": 0.7227421292287063, + "grad_norm": 0.4070129990577698, + "learning_rate": 0.0006408281039892425, + "loss": 0.5246, + "step": 12936 + }, + { + "epoch": 0.7227979998323882, + "grad_norm": 1.3627382516860962, + "learning_rate": 0.0006408000896458987, + "loss": 0.4555, + "step": 12937 + }, + { + "epoch": 0.7228538704360701, + "grad_norm": 1.004325032234192, + "learning_rate": 0.0006407720753025548, + "loss": 0.5372, + "step": 12938 + }, + { + "epoch": 0.7229097410397519, + "grad_norm": 0.7766132950782776, + "learning_rate": 0.0006407440609592112, + "loss": 0.4419, + "step": 12939 + }, + { + "epoch": 0.7229656116434338, + "grad_norm": 0.6190351247787476, + "learning_rate": 0.0006407160466158674, + "loss": 0.2904, + "step": 12940 + }, + { + "epoch": 0.7230214822471157, + "grad_norm": 0.7537804245948792, + "learning_rate": 0.0006406880322725236, + "loss": 0.458, + "step": 12941 + }, + { + "epoch": 0.7230773528507976, + "grad_norm": 0.4590234160423279, + "learning_rate": 0.0006406600179291798, + "loss": 0.3576, + "step": 12942 + }, + { + "epoch": 0.7231332234544794, + "grad_norm": 0.575129508972168, + "learning_rate": 0.000640632003585836, + "loss": 0.4669, + "step": 12943 + }, + { + "epoch": 0.7231890940581613, + "grad_norm": 2.1075596809387207, + "learning_rate": 0.0006406039892424922, + "loss": 0.5193, + "step": 12944 + }, + { + "epoch": 0.7232449646618432, + "grad_norm": 0.5811126232147217, + "learning_rate": 0.0006405759748991484, + "loss": 0.3506, + "step": 12945 + }, + { + "epoch": 0.723300835265525, + "grad_norm": 0.6953825950622559, + "learning_rate": 0.0006405479605558046, + "loss": 0.4547, + "step": 12946 + }, + { + "epoch": 0.7233567058692069, + "grad_norm": 0.9603988528251648, + "learning_rate": 0.0006405199462124608, + "loss": 0.3895, + "step": 12947 + }, + { + "epoch": 0.7234125764728888, + "grad_norm": 0.4372292459011078, + "learning_rate": 0.000640491931869117, + "loss": 0.3913, + "step": 12948 + }, + { + "epoch": 0.7234684470765707, + "grad_norm": 0.48979735374450684, + "learning_rate": 0.0006404639175257731, + "loss": 0.3197, + "step": 12949 + }, + { + "epoch": 0.7235243176802525, + "grad_norm": 0.313239723443985, + "learning_rate": 0.0006404359031824294, + "loss": 0.3883, + "step": 12950 + }, + { + "epoch": 0.7235801882839344, + "grad_norm": 1.2724120616912842, + "learning_rate": 0.0006404078888390856, + "loss": 0.505, + "step": 12951 + }, + { + "epoch": 0.7236360588876163, + "grad_norm": 2.1170907020568848, + "learning_rate": 0.0006403798744957418, + "loss": 0.4654, + "step": 12952 + }, + { + "epoch": 0.7236919294912981, + "grad_norm": 0.5221952795982361, + "learning_rate": 0.000640351860152398, + "loss": 0.6207, + "step": 12953 + }, + { + "epoch": 0.7237478000949801, + "grad_norm": 0.5462852716445923, + "learning_rate": 0.0006403238458090542, + "loss": 0.4597, + "step": 12954 + }, + { + "epoch": 0.7238036706986619, + "grad_norm": 0.4365018904209137, + "learning_rate": 0.0006402958314657104, + "loss": 0.5113, + "step": 12955 + }, + { + "epoch": 0.7238595413023438, + "grad_norm": 0.4275396764278412, + "learning_rate": 0.0006402678171223667, + "loss": 0.4274, + "step": 12956 + }, + { + "epoch": 0.7239154119060256, + "grad_norm": 0.4603109359741211, + "learning_rate": 0.0006402398027790229, + "loss": 0.3691, + "step": 12957 + }, + { + "epoch": 0.7239712825097075, + "grad_norm": 0.6211150288581848, + "learning_rate": 0.0006402117884356791, + "loss": 0.4712, + "step": 12958 + }, + { + "epoch": 0.7240271531133894, + "grad_norm": 0.7449069619178772, + "learning_rate": 0.0006401837740923353, + "loss": 0.3816, + "step": 12959 + }, + { + "epoch": 0.7240830237170712, + "grad_norm": 0.5770741701126099, + "learning_rate": 0.0006401557597489916, + "loss": 0.4911, + "step": 12960 + }, + { + "epoch": 0.7241388943207532, + "grad_norm": 0.990352988243103, + "learning_rate": 0.0006401277454056477, + "loss": 0.4274, + "step": 12961 + }, + { + "epoch": 0.724194764924435, + "grad_norm": 2.2809369564056396, + "learning_rate": 0.0006400997310623039, + "loss": 0.564, + "step": 12962 + }, + { + "epoch": 0.7242506355281169, + "grad_norm": 0.9519371390342712, + "learning_rate": 0.0006400717167189601, + "loss": 0.3169, + "step": 12963 + }, + { + "epoch": 0.7243065061317987, + "grad_norm": 0.3670308291912079, + "learning_rate": 0.0006400437023756164, + "loss": 0.4621, + "step": 12964 + }, + { + "epoch": 0.7243623767354807, + "grad_norm": 0.5347092747688293, + "learning_rate": 0.0006400156880322725, + "loss": 0.4436, + "step": 12965 + }, + { + "epoch": 0.7244182473391625, + "grad_norm": 0.516142725944519, + "learning_rate": 0.0006399876736889287, + "loss": 0.4263, + "step": 12966 + }, + { + "epoch": 0.7244741179428443, + "grad_norm": 0.416640043258667, + "learning_rate": 0.000639959659345585, + "loss": 0.3777, + "step": 12967 + }, + { + "epoch": 0.7245299885465263, + "grad_norm": 0.517587423324585, + "learning_rate": 0.0006399316450022412, + "loss": 0.3942, + "step": 12968 + }, + { + "epoch": 0.7245858591502081, + "grad_norm": 0.4706045091152191, + "learning_rate": 0.0006399036306588973, + "loss": 0.4624, + "step": 12969 + }, + { + "epoch": 0.72464172975389, + "grad_norm": 0.830947756767273, + "learning_rate": 0.0006398756163155535, + "loss": 0.4159, + "step": 12970 + }, + { + "epoch": 0.7246976003575719, + "grad_norm": 0.4032474756240845, + "learning_rate": 0.0006398476019722098, + "loss": 0.4091, + "step": 12971 + }, + { + "epoch": 0.7247534709612538, + "grad_norm": 0.3746197819709778, + "learning_rate": 0.0006398195876288659, + "loss": 0.3961, + "step": 12972 + }, + { + "epoch": 0.7248093415649356, + "grad_norm": 0.4262131452560425, + "learning_rate": 0.0006397915732855221, + "loss": 0.4743, + "step": 12973 + }, + { + "epoch": 0.7248652121686174, + "grad_norm": 0.30471450090408325, + "learning_rate": 0.0006397635589421785, + "loss": 0.3789, + "step": 12974 + }, + { + "epoch": 0.7249210827722994, + "grad_norm": 0.4405056834220886, + "learning_rate": 0.0006397355445988347, + "loss": 0.4536, + "step": 12975 + }, + { + "epoch": 0.7249769533759812, + "grad_norm": 0.4690376818180084, + "learning_rate": 0.0006397075302554908, + "loss": 0.4024, + "step": 12976 + }, + { + "epoch": 0.7250328239796631, + "grad_norm": 0.3411528766155243, + "learning_rate": 0.000639679515912147, + "loss": 0.4034, + "step": 12977 + }, + { + "epoch": 0.725088694583345, + "grad_norm": 1.1752749681472778, + "learning_rate": 0.0006396515015688033, + "loss": 0.5863, + "step": 12978 + }, + { + "epoch": 0.7251445651870269, + "grad_norm": 0.4428904056549072, + "learning_rate": 0.0006396234872254595, + "loss": 0.5994, + "step": 12979 + }, + { + "epoch": 0.7252004357907087, + "grad_norm": 0.5251572728157043, + "learning_rate": 0.0006395954728821156, + "loss": 0.4777, + "step": 12980 + }, + { + "epoch": 0.7252563063943905, + "grad_norm": 0.8751762509346008, + "learning_rate": 0.0006395674585387719, + "loss": 0.4897, + "step": 12981 + }, + { + "epoch": 0.7253121769980725, + "grad_norm": 0.558754563331604, + "learning_rate": 0.0006395394441954281, + "loss": 0.5086, + "step": 12982 + }, + { + "epoch": 0.7253680476017543, + "grad_norm": 0.4428402781486511, + "learning_rate": 0.0006395114298520843, + "loss": 0.3901, + "step": 12983 + }, + { + "epoch": 0.7254239182054362, + "grad_norm": 0.45529359579086304, + "learning_rate": 0.0006394834155087404, + "loss": 0.4856, + "step": 12984 + }, + { + "epoch": 0.7254797888091181, + "grad_norm": 0.40204712748527527, + "learning_rate": 0.0006394554011653967, + "loss": 0.4259, + "step": 12985 + }, + { + "epoch": 0.7255356594128, + "grad_norm": 0.3416041135787964, + "learning_rate": 0.0006394273868220529, + "loss": 0.3783, + "step": 12986 + }, + { + "epoch": 0.7255915300164818, + "grad_norm": 0.38579466938972473, + "learning_rate": 0.0006393993724787091, + "loss": 0.3181, + "step": 12987 + }, + { + "epoch": 0.7256474006201638, + "grad_norm": 0.6245083808898926, + "learning_rate": 0.0006393713581353652, + "loss": 0.4349, + "step": 12988 + }, + { + "epoch": 0.7257032712238456, + "grad_norm": Infinity, + "learning_rate": 0.0006393713581353652, + "loss": 0.5343, + "step": 12989 + }, + { + "epoch": 0.7257591418275274, + "grad_norm": 0.41393500566482544, + "learning_rate": 0.0006393433437920216, + "loss": 0.4681, + "step": 12990 + }, + { + "epoch": 0.7258150124312093, + "grad_norm": 0.5237320065498352, + "learning_rate": 0.0006393153294486778, + "loss": 0.4438, + "step": 12991 + }, + { + "epoch": 0.7258708830348912, + "grad_norm": 0.3919839560985565, + "learning_rate": 0.000639287315105334, + "loss": 0.4273, + "step": 12992 + }, + { + "epoch": 0.7259267536385731, + "grad_norm": 0.4028712809085846, + "learning_rate": 0.0006392593007619902, + "loss": 0.4414, + "step": 12993 + }, + { + "epoch": 0.7259826242422549, + "grad_norm": 0.48024609684944153, + "learning_rate": 0.0006392312864186464, + "loss": 0.5518, + "step": 12994 + }, + { + "epoch": 0.7260384948459369, + "grad_norm": 0.463168203830719, + "learning_rate": 0.0006392032720753026, + "loss": 0.4534, + "step": 12995 + }, + { + "epoch": 0.7260943654496187, + "grad_norm": 0.4268948435783386, + "learning_rate": 0.0006391752577319587, + "loss": 0.4499, + "step": 12996 + }, + { + "epoch": 0.7261502360533005, + "grad_norm": 0.36533212661743164, + "learning_rate": 0.000639147243388615, + "loss": 0.3448, + "step": 12997 + }, + { + "epoch": 0.7262061066569824, + "grad_norm": 1.7267265319824219, + "learning_rate": 0.0006391192290452712, + "loss": 0.5667, + "step": 12998 + }, + { + "epoch": 0.7262619772606643, + "grad_norm": 0.3568171262741089, + "learning_rate": 0.0006390912147019274, + "loss": 0.3808, + "step": 12999 + }, + { + "epoch": 0.7263178478643462, + "grad_norm": 0.4948100745677948, + "learning_rate": 0.0006390632003585836, + "loss": 0.3769, + "step": 13000 + }, + { + "epoch": 0.7263178478643462, + "eval_cer": 0.09014434879382018, + "eval_loss": 0.3381013572216034, + "eval_runtime": 56.007, + "eval_samples_per_second": 81.026, + "eval_steps_per_second": 5.071, + "eval_wer": 0.3604647914862713, + "step": 13000 + }, + { + "epoch": 0.726373718468028, + "grad_norm": 0.38485386967658997, + "learning_rate": 0.0006390351860152398, + "loss": 0.3737, + "step": 13001 + }, + { + "epoch": 0.72642958907171, + "grad_norm": 0.8232678771018982, + "learning_rate": 0.000639007171671896, + "loss": 0.5822, + "step": 13002 + }, + { + "epoch": 0.7264854596753918, + "grad_norm": 0.5017839670181274, + "learning_rate": 0.0006389791573285522, + "loss": 0.5273, + "step": 13003 + }, + { + "epoch": 0.7265413302790736, + "grad_norm": 0.429004430770874, + "learning_rate": 0.0006389511429852084, + "loss": 0.5079, + "step": 13004 + }, + { + "epoch": 0.7265972008827556, + "grad_norm": 0.7629493474960327, + "learning_rate": 0.0006389231286418646, + "loss": 0.367, + "step": 13005 + }, + { + "epoch": 0.7266530714864374, + "grad_norm": 0.49411529302597046, + "learning_rate": 0.0006388951142985208, + "loss": 0.5789, + "step": 13006 + }, + { + "epoch": 0.7267089420901193, + "grad_norm": 0.6843600869178772, + "learning_rate": 0.0006388670999551772, + "loss": 0.3862, + "step": 13007 + }, + { + "epoch": 0.7267648126938011, + "grad_norm": 0.7294999361038208, + "learning_rate": 0.0006388390856118333, + "loss": 0.4629, + "step": 13008 + }, + { + "epoch": 0.7268206832974831, + "grad_norm": 0.6744655966758728, + "learning_rate": 0.0006388110712684895, + "loss": 0.4144, + "step": 13009 + }, + { + "epoch": 0.7268765539011649, + "grad_norm": 0.4803061783313751, + "learning_rate": 0.0006387830569251457, + "loss": 0.5301, + "step": 13010 + }, + { + "epoch": 0.7269324245048467, + "grad_norm": 2.389462947845459, + "learning_rate": 0.000638755042581802, + "loss": 0.3933, + "step": 13011 + }, + { + "epoch": 0.7269882951085287, + "grad_norm": 0.5878506898880005, + "learning_rate": 0.0006387270282384581, + "loss": 0.4618, + "step": 13012 + }, + { + "epoch": 0.7270441657122105, + "grad_norm": 0.6559337973594666, + "learning_rate": 0.0006386990138951143, + "loss": 0.4253, + "step": 13013 + }, + { + "epoch": 0.7271000363158924, + "grad_norm": 0.5608011484146118, + "learning_rate": 0.0006386709995517706, + "loss": 0.5528, + "step": 13014 + }, + { + "epoch": 0.7271559069195742, + "grad_norm": 0.452178955078125, + "learning_rate": 0.0006386429852084267, + "loss": 0.4559, + "step": 13015 + }, + { + "epoch": 0.7272117775232562, + "grad_norm": 0.5018419623374939, + "learning_rate": 0.0006386149708650829, + "loss": 0.5286, + "step": 13016 + }, + { + "epoch": 0.727267648126938, + "grad_norm": 0.5327479839324951, + "learning_rate": 0.0006385869565217391, + "loss": 0.5223, + "step": 13017 + }, + { + "epoch": 0.7273235187306198, + "grad_norm": 0.8455742001533508, + "learning_rate": 0.0006385589421783954, + "loss": 0.3628, + "step": 13018 + }, + { + "epoch": 0.7273793893343018, + "grad_norm": 0.6562570929527283, + "learning_rate": 0.0006385309278350515, + "loss": 0.5836, + "step": 13019 + }, + { + "epoch": 0.7274352599379836, + "grad_norm": 0.75368332862854, + "learning_rate": 0.0006385029134917077, + "loss": 0.4514, + "step": 13020 + }, + { + "epoch": 0.7274911305416655, + "grad_norm": 0.4258670210838318, + "learning_rate": 0.0006384748991483639, + "loss": 0.3244, + "step": 13021 + }, + { + "epoch": 0.7275470011453474, + "grad_norm": 9.372187614440918, + "learning_rate": 0.0006384468848050202, + "loss": 0.5242, + "step": 13022 + }, + { + "epoch": 0.7276028717490293, + "grad_norm": 0.5745556950569153, + "learning_rate": 0.0006384188704616763, + "loss": 0.4826, + "step": 13023 + }, + { + "epoch": 0.7276587423527111, + "grad_norm": 0.5806251764297485, + "learning_rate": 0.0006383908561183325, + "loss": 0.3489, + "step": 13024 + }, + { + "epoch": 0.727714612956393, + "grad_norm": 0.9372202754020691, + "learning_rate": 0.0006383628417749889, + "loss": 0.6112, + "step": 13025 + }, + { + "epoch": 0.7277704835600749, + "grad_norm": 0.4992195665836334, + "learning_rate": 0.0006383348274316451, + "loss": 0.4741, + "step": 13026 + }, + { + "epoch": 0.7278263541637567, + "grad_norm": 1.9475172758102417, + "learning_rate": 0.0006383068130883012, + "loss": 0.3575, + "step": 13027 + }, + { + "epoch": 0.7278822247674386, + "grad_norm": 1.0362733602523804, + "learning_rate": 0.0006382787987449574, + "loss": 0.3693, + "step": 13028 + }, + { + "epoch": 0.7279380953711205, + "grad_norm": 0.9764519929885864, + "learning_rate": 0.0006382507844016137, + "loss": 0.5212, + "step": 13029 + }, + { + "epoch": 0.7279939659748024, + "grad_norm": 4.646176815032959, + "learning_rate": 0.0006382227700582699, + "loss": 0.4103, + "step": 13030 + }, + { + "epoch": 0.7280498365784842, + "grad_norm": 0.8066773414611816, + "learning_rate": 0.000638194755714926, + "loss": 0.3991, + "step": 13031 + }, + { + "epoch": 0.728105707182166, + "grad_norm": 0.5831990838050842, + "learning_rate": 0.0006381667413715823, + "loss": 0.4249, + "step": 13032 + }, + { + "epoch": 0.728161577785848, + "grad_norm": 0.5040490031242371, + "learning_rate": 0.0006381387270282385, + "loss": 0.4958, + "step": 13033 + }, + { + "epoch": 0.7282174483895298, + "grad_norm": 0.998226523399353, + "learning_rate": 0.0006381107126848947, + "loss": 0.4388, + "step": 13034 + }, + { + "epoch": 0.7282733189932117, + "grad_norm": 1.5395350456237793, + "learning_rate": 0.0006380826983415508, + "loss": 0.4945, + "step": 13035 + }, + { + "epoch": 0.7283291895968936, + "grad_norm": 0.9256266355514526, + "learning_rate": 0.0006380546839982071, + "loss": 0.4168, + "step": 13036 + }, + { + "epoch": 0.7283850602005755, + "grad_norm": 0.5136159658432007, + "learning_rate": 0.0006380266696548633, + "loss": 0.4575, + "step": 13037 + }, + { + "epoch": 0.7284409308042573, + "grad_norm": 0.3429466187953949, + "learning_rate": 0.0006379986553115194, + "loss": 0.383, + "step": 13038 + }, + { + "epoch": 0.7284968014079393, + "grad_norm": 0.45766469836235046, + "learning_rate": 0.0006379706409681757, + "loss": 0.4427, + "step": 13039 + }, + { + "epoch": 0.7285526720116211, + "grad_norm": 1.1046693325042725, + "learning_rate": 0.000637942626624832, + "loss": 0.6045, + "step": 13040 + }, + { + "epoch": 0.728608542615303, + "grad_norm": 0.48894304037094116, + "learning_rate": 0.0006379146122814882, + "loss": 0.3562, + "step": 13041 + }, + { + "epoch": 0.7286644132189848, + "grad_norm": 0.40274736285209656, + "learning_rate": 0.0006378865979381443, + "loss": 0.4374, + "step": 13042 + }, + { + "epoch": 0.7287202838226667, + "grad_norm": 0.8680577278137207, + "learning_rate": 0.0006378585835948006, + "loss": 0.4181, + "step": 13043 + }, + { + "epoch": 0.7287761544263486, + "grad_norm": 0.48635387420654297, + "learning_rate": 0.0006378305692514568, + "loss": 0.4399, + "step": 13044 + }, + { + "epoch": 0.7288320250300304, + "grad_norm": 0.45793578028678894, + "learning_rate": 0.000637802554908113, + "loss": 0.5136, + "step": 13045 + }, + { + "epoch": 0.7288878956337124, + "grad_norm": 0.48538005352020264, + "learning_rate": 0.0006377745405647692, + "loss": 0.4929, + "step": 13046 + }, + { + "epoch": 0.7289437662373942, + "grad_norm": 0.4175281226634979, + "learning_rate": 0.0006377465262214254, + "loss": 0.506, + "step": 13047 + }, + { + "epoch": 0.728999636841076, + "grad_norm": 0.4253140091896057, + "learning_rate": 0.0006377185118780816, + "loss": 0.4221, + "step": 13048 + }, + { + "epoch": 0.7290555074447579, + "grad_norm": 0.5144241452217102, + "learning_rate": 0.0006376904975347378, + "loss": 0.4553, + "step": 13049 + }, + { + "epoch": 0.7291113780484398, + "grad_norm": 0.5408889651298523, + "learning_rate": 0.000637662483191394, + "loss": 0.5096, + "step": 13050 + }, + { + "epoch": 0.7291672486521217, + "grad_norm": 0.6370031833648682, + "learning_rate": 0.0006376344688480502, + "loss": 0.3893, + "step": 13051 + }, + { + "epoch": 0.7292231192558035, + "grad_norm": 0.4244908392429352, + "learning_rate": 0.0006376064545047064, + "loss": 0.443, + "step": 13052 + }, + { + "epoch": 0.7292789898594855, + "grad_norm": 0.4570315182209015, + "learning_rate": 0.0006375784401613627, + "loss": 0.4009, + "step": 13053 + }, + { + "epoch": 0.7293348604631673, + "grad_norm": 0.5359293222427368, + "learning_rate": 0.0006375504258180188, + "loss": 0.5417, + "step": 13054 + }, + { + "epoch": 0.7293907310668492, + "grad_norm": 0.5834537744522095, + "learning_rate": 0.000637522411474675, + "loss": 0.4046, + "step": 13055 + }, + { + "epoch": 0.7294466016705311, + "grad_norm": 0.414359986782074, + "learning_rate": 0.0006374943971313312, + "loss": 0.4771, + "step": 13056 + }, + { + "epoch": 0.7295024722742129, + "grad_norm": 0.6032629609107971, + "learning_rate": 0.0006374663827879874, + "loss": 0.3571, + "step": 13057 + }, + { + "epoch": 0.7295583428778948, + "grad_norm": 0.44690173864364624, + "learning_rate": 0.0006374383684446437, + "loss": 0.4337, + "step": 13058 + }, + { + "epoch": 0.7296142134815766, + "grad_norm": 0.9025493264198303, + "learning_rate": 0.0006374103541012999, + "loss": 0.5572, + "step": 13059 + }, + { + "epoch": 0.7296700840852586, + "grad_norm": 0.5187279582023621, + "learning_rate": 0.0006373823397579561, + "loss": 0.5064, + "step": 13060 + }, + { + "epoch": 0.7297259546889404, + "grad_norm": 0.7764899134635925, + "learning_rate": 0.0006373543254146123, + "loss": 0.4292, + "step": 13061 + }, + { + "epoch": 0.7297818252926223, + "grad_norm": 3.256715774536133, + "learning_rate": 0.0006373263110712685, + "loss": 0.4161, + "step": 13062 + }, + { + "epoch": 0.7298376958963042, + "grad_norm": 0.37384262681007385, + "learning_rate": 0.0006372982967279247, + "loss": 0.4169, + "step": 13063 + }, + { + "epoch": 0.729893566499986, + "grad_norm": 0.8492703437805176, + "learning_rate": 0.000637270282384581, + "loss": 0.4944, + "step": 13064 + }, + { + "epoch": 0.7299494371036679, + "grad_norm": 1.4354650974273682, + "learning_rate": 0.0006372422680412371, + "loss": 0.3804, + "step": 13065 + }, + { + "epoch": 0.7300053077073497, + "grad_norm": 0.4203684628009796, + "learning_rate": 0.0006372142536978933, + "loss": 0.4027, + "step": 13066 + }, + { + "epoch": 0.7300611783110317, + "grad_norm": 0.8187611103057861, + "learning_rate": 0.0006371862393545495, + "loss": 0.4158, + "step": 13067 + }, + { + "epoch": 0.7301170489147135, + "grad_norm": 0.6471171975135803, + "learning_rate": 0.0006371582250112058, + "loss": 0.4501, + "step": 13068 + }, + { + "epoch": 0.7301729195183954, + "grad_norm": 0.526681125164032, + "learning_rate": 0.0006371302106678619, + "loss": 0.5248, + "step": 13069 + }, + { + "epoch": 0.7302287901220773, + "grad_norm": 0.460177481174469, + "learning_rate": 0.0006371021963245181, + "loss": 0.3426, + "step": 13070 + }, + { + "epoch": 0.7302846607257591, + "grad_norm": 0.3749963343143463, + "learning_rate": 0.0006370741819811744, + "loss": 0.3611, + "step": 13071 + }, + { + "epoch": 0.730340531329441, + "grad_norm": 0.4386865794658661, + "learning_rate": 0.0006370461676378306, + "loss": 0.547, + "step": 13072 + }, + { + "epoch": 0.7303964019331229, + "grad_norm": 0.4132656157016754, + "learning_rate": 0.0006370181532944867, + "loss": 0.4068, + "step": 13073 + }, + { + "epoch": 0.7304522725368048, + "grad_norm": 0.5105851292610168, + "learning_rate": 0.000636990138951143, + "loss": 0.3849, + "step": 13074 + }, + { + "epoch": 0.7305081431404866, + "grad_norm": 0.6427749395370483, + "learning_rate": 0.0006369621246077993, + "loss": 0.5818, + "step": 13075 + }, + { + "epoch": 0.7305640137441685, + "grad_norm": 0.4654095768928528, + "learning_rate": 0.0006369341102644555, + "loss": 0.5963, + "step": 13076 + }, + { + "epoch": 0.7306198843478504, + "grad_norm": 0.5494029521942139, + "learning_rate": 0.0006369060959211116, + "loss": 0.4653, + "step": 13077 + }, + { + "epoch": 0.7306757549515323, + "grad_norm": 0.984795331954956, + "learning_rate": 0.0006368780815777679, + "loss": 0.4337, + "step": 13078 + }, + { + "epoch": 0.7307316255552141, + "grad_norm": 0.615123987197876, + "learning_rate": 0.0006368500672344241, + "loss": 0.4235, + "step": 13079 + }, + { + "epoch": 0.730787496158896, + "grad_norm": 0.6260597109794617, + "learning_rate": 0.0006368220528910802, + "loss": 0.528, + "step": 13080 + }, + { + "epoch": 0.7308433667625779, + "grad_norm": 0.5882424712181091, + "learning_rate": 0.0006367940385477364, + "loss": 0.5727, + "step": 13081 + }, + { + "epoch": 0.7308992373662597, + "grad_norm": 0.570377767086029, + "learning_rate": 0.0006367660242043927, + "loss": 0.5628, + "step": 13082 + }, + { + "epoch": 0.7309551079699416, + "grad_norm": 0.6066653728485107, + "learning_rate": 0.0006367380098610489, + "loss": 0.3906, + "step": 13083 + }, + { + "epoch": 0.7310109785736235, + "grad_norm": 0.4541183114051819, + "learning_rate": 0.000636709995517705, + "loss": 0.3636, + "step": 13084 + }, + { + "epoch": 0.7310668491773054, + "grad_norm": 0.5773184895515442, + "learning_rate": 0.0006366819811743613, + "loss": 0.5087, + "step": 13085 + }, + { + "epoch": 0.7311227197809872, + "grad_norm": 0.4667440354824066, + "learning_rate": 0.0006366539668310175, + "loss": 0.4233, + "step": 13086 + }, + { + "epoch": 0.7311785903846691, + "grad_norm": 0.5499706268310547, + "learning_rate": 0.0006366259524876737, + "loss": 0.5068, + "step": 13087 + }, + { + "epoch": 0.731234460988351, + "grad_norm": 0.511615514755249, + "learning_rate": 0.0006365979381443298, + "loss": 0.4216, + "step": 13088 + }, + { + "epoch": 0.7312903315920328, + "grad_norm": 0.4243183135986328, + "learning_rate": 0.0006365699238009861, + "loss": 0.4915, + "step": 13089 + }, + { + "epoch": 0.7313462021957148, + "grad_norm": 0.6055556535720825, + "learning_rate": 0.0006365419094576423, + "loss": 0.4703, + "step": 13090 + }, + { + "epoch": 0.7314020727993966, + "grad_norm": 0.41728752851486206, + "learning_rate": 0.0006365138951142986, + "loss": 0.428, + "step": 13091 + }, + { + "epoch": 0.7314579434030785, + "grad_norm": 0.584884762763977, + "learning_rate": 0.0006364858807709547, + "loss": 0.5812, + "step": 13092 + }, + { + "epoch": 0.7315138140067603, + "grad_norm": 0.8736400604248047, + "learning_rate": 0.000636457866427611, + "loss": 0.4465, + "step": 13093 + }, + { + "epoch": 0.7315696846104423, + "grad_norm": 1.5271164178848267, + "learning_rate": 0.0006364298520842672, + "loss": 0.3727, + "step": 13094 + }, + { + "epoch": 0.7316255552141241, + "grad_norm": 0.696060299873352, + "learning_rate": 0.0006364018377409234, + "loss": 0.4762, + "step": 13095 + }, + { + "epoch": 0.7316814258178059, + "grad_norm": 0.3716892600059509, + "learning_rate": 0.0006363738233975796, + "loss": 0.4708, + "step": 13096 + }, + { + "epoch": 0.7317372964214879, + "grad_norm": 0.7073664665222168, + "learning_rate": 0.0006363458090542358, + "loss": 0.5127, + "step": 13097 + }, + { + "epoch": 0.7317931670251697, + "grad_norm": 0.5641611218452454, + "learning_rate": 0.000636317794710892, + "loss": 0.3977, + "step": 13098 + }, + { + "epoch": 0.7318490376288516, + "grad_norm": 0.7270122170448303, + "learning_rate": 0.0006362897803675481, + "loss": 0.4028, + "step": 13099 + }, + { + "epoch": 0.7319049082325334, + "grad_norm": 1.0035496950149536, + "learning_rate": 0.0006362617660242044, + "loss": 0.3795, + "step": 13100 + }, + { + "epoch": 0.7319607788362154, + "grad_norm": 0.5459606647491455, + "learning_rate": 0.0006362337516808606, + "loss": 0.5876, + "step": 13101 + }, + { + "epoch": 0.7320166494398972, + "grad_norm": 0.5263988375663757, + "learning_rate": 0.0006362057373375168, + "loss": 0.5045, + "step": 13102 + }, + { + "epoch": 0.732072520043579, + "grad_norm": 0.6064448952674866, + "learning_rate": 0.000636177722994173, + "loss": 0.4617, + "step": 13103 + }, + { + "epoch": 0.732128390647261, + "grad_norm": 0.6902440786361694, + "learning_rate": 0.0006361497086508292, + "loss": 0.4963, + "step": 13104 + }, + { + "epoch": 0.7321842612509428, + "grad_norm": 1.3932130336761475, + "learning_rate": 0.0006361216943074854, + "loss": 0.4634, + "step": 13105 + }, + { + "epoch": 0.7322401318546247, + "grad_norm": 0.5276194214820862, + "learning_rate": 0.0006360936799641416, + "loss": 0.4517, + "step": 13106 + }, + { + "epoch": 0.7322960024583066, + "grad_norm": 0.46560075879096985, + "learning_rate": 0.0006360656656207978, + "loss": 0.4058, + "step": 13107 + }, + { + "epoch": 0.7323518730619885, + "grad_norm": 2.9792280197143555, + "learning_rate": 0.0006360376512774541, + "loss": 0.3838, + "step": 13108 + }, + { + "epoch": 0.7324077436656703, + "grad_norm": 0.8569866418838501, + "learning_rate": 0.0006360096369341103, + "loss": 0.5657, + "step": 13109 + }, + { + "epoch": 0.7324636142693521, + "grad_norm": 0.6691102385520935, + "learning_rate": 0.0006359816225907666, + "loss": 0.4921, + "step": 13110 + }, + { + "epoch": 0.7325194848730341, + "grad_norm": 1.7982752323150635, + "learning_rate": 0.0006359536082474227, + "loss": 0.4912, + "step": 13111 + }, + { + "epoch": 0.7325753554767159, + "grad_norm": 2.5395925045013428, + "learning_rate": 0.0006359255939040789, + "loss": 0.31, + "step": 13112 + }, + { + "epoch": 0.7326312260803978, + "grad_norm": 0.41238588094711304, + "learning_rate": 0.0006358975795607351, + "loss": 0.4064, + "step": 13113 + }, + { + "epoch": 0.7326870966840797, + "grad_norm": 0.5644327998161316, + "learning_rate": 0.0006358695652173914, + "loss": 0.5847, + "step": 13114 + }, + { + "epoch": 0.7327429672877616, + "grad_norm": 1.0689817667007446, + "learning_rate": 0.0006358415508740475, + "loss": 0.4108, + "step": 13115 + }, + { + "epoch": 0.7327988378914434, + "grad_norm": 1.5788346529006958, + "learning_rate": 0.0006358135365307037, + "loss": 0.3674, + "step": 13116 + }, + { + "epoch": 0.7328547084951252, + "grad_norm": 0.47355467081069946, + "learning_rate": 0.00063578552218736, + "loss": 0.475, + "step": 13117 + }, + { + "epoch": 0.7329105790988072, + "grad_norm": 0.4194372296333313, + "learning_rate": 0.0006357575078440162, + "loss": 0.4312, + "step": 13118 + }, + { + "epoch": 0.732966449702489, + "grad_norm": 0.9581978917121887, + "learning_rate": 0.0006357294935006723, + "loss": 0.4272, + "step": 13119 + }, + { + "epoch": 0.7330223203061709, + "grad_norm": 0.39042818546295166, + "learning_rate": 0.0006357014791573285, + "loss": 0.4919, + "step": 13120 + }, + { + "epoch": 0.7330781909098528, + "grad_norm": 0.6058931946754456, + "learning_rate": 0.0006356734648139848, + "loss": 0.4428, + "step": 13121 + }, + { + "epoch": 0.7331340615135347, + "grad_norm": 0.4681475758552551, + "learning_rate": 0.0006356454504706409, + "loss": 0.5046, + "step": 13122 + }, + { + "epoch": 0.7331899321172165, + "grad_norm": 0.528934895992279, + "learning_rate": 0.0006356174361272971, + "loss": 0.3616, + "step": 13123 + }, + { + "epoch": 0.7332458027208985, + "grad_norm": 0.5850439071655273, + "learning_rate": 0.0006355894217839535, + "loss": 0.3237, + "step": 13124 + }, + { + "epoch": 0.7333016733245803, + "grad_norm": 0.4370817244052887, + "learning_rate": 0.0006355614074406097, + "loss": 0.4773, + "step": 13125 + }, + { + "epoch": 0.7333575439282621, + "grad_norm": 2.814953565597534, + "learning_rate": 0.0006355333930972658, + "loss": 0.4874, + "step": 13126 + }, + { + "epoch": 0.733413414531944, + "grad_norm": 0.5575742125511169, + "learning_rate": 0.000635505378753922, + "loss": 0.4257, + "step": 13127 + }, + { + "epoch": 0.7334692851356259, + "grad_norm": 0.3344840109348297, + "learning_rate": 0.0006354773644105783, + "loss": 0.3864, + "step": 13128 + }, + { + "epoch": 0.7335251557393078, + "grad_norm": 0.7553679347038269, + "learning_rate": 0.0006354493500672345, + "loss": 0.4704, + "step": 13129 + }, + { + "epoch": 0.7335810263429896, + "grad_norm": 0.33257079124450684, + "learning_rate": 0.0006354213357238906, + "loss": 0.324, + "step": 13130 + }, + { + "epoch": 0.7336368969466716, + "grad_norm": 0.4289097487926483, + "learning_rate": 0.0006353933213805468, + "loss": 0.4292, + "step": 13131 + }, + { + "epoch": 0.7336927675503534, + "grad_norm": 0.47230449318885803, + "learning_rate": 0.0006353653070372031, + "loss": 0.4232, + "step": 13132 + }, + { + "epoch": 0.7337486381540352, + "grad_norm": 0.6013162732124329, + "learning_rate": 0.0006353372926938593, + "loss": 0.536, + "step": 13133 + }, + { + "epoch": 0.7338045087577171, + "grad_norm": 0.6086875796318054, + "learning_rate": 0.0006353092783505154, + "loss": 0.406, + "step": 13134 + }, + { + "epoch": 0.733860379361399, + "grad_norm": 0.8035362958908081, + "learning_rate": 0.0006352812640071717, + "loss": 0.3436, + "step": 13135 + }, + { + "epoch": 0.7339162499650809, + "grad_norm": 0.8885512948036194, + "learning_rate": 0.0006352532496638279, + "loss": 0.4718, + "step": 13136 + }, + { + "epoch": 0.7339721205687627, + "grad_norm": 0.79127436876297, + "learning_rate": 0.0006352252353204841, + "loss": 0.4535, + "step": 13137 + }, + { + "epoch": 0.7340279911724447, + "grad_norm": 0.9559204578399658, + "learning_rate": 0.0006351972209771402, + "loss": 0.7728, + "step": 13138 + }, + { + "epoch": 0.7340838617761265, + "grad_norm": 0.48470982909202576, + "learning_rate": 0.0006351692066337965, + "loss": 0.5208, + "step": 13139 + }, + { + "epoch": 0.7341397323798083, + "grad_norm": 1.4323457479476929, + "learning_rate": 0.0006351411922904527, + "loss": 0.4483, + "step": 13140 + }, + { + "epoch": 0.7341956029834903, + "grad_norm": 0.40012314915657043, + "learning_rate": 0.0006351131779471088, + "loss": 0.4308, + "step": 13141 + }, + { + "epoch": 0.7342514735871721, + "grad_norm": 1.5837302207946777, + "learning_rate": 0.0006350851636037652, + "loss": 0.4462, + "step": 13142 + }, + { + "epoch": 0.734307344190854, + "grad_norm": 0.7673121094703674, + "learning_rate": 0.0006350571492604214, + "loss": 0.4172, + "step": 13143 + }, + { + "epoch": 0.7343632147945358, + "grad_norm": 0.5104077458381653, + "learning_rate": 0.0006350291349170776, + "loss": 0.503, + "step": 13144 + }, + { + "epoch": 0.7344190853982178, + "grad_norm": 0.406713604927063, + "learning_rate": 0.0006350011205737337, + "loss": 0.381, + "step": 13145 + }, + { + "epoch": 0.7344749560018996, + "grad_norm": 0.3823848068714142, + "learning_rate": 0.00063497310623039, + "loss": 0.4231, + "step": 13146 + }, + { + "epoch": 0.7345308266055814, + "grad_norm": 0.43696966767311096, + "learning_rate": 0.0006349450918870462, + "loss": 0.4683, + "step": 13147 + }, + { + "epoch": 0.7345866972092634, + "grad_norm": 0.41811710596084595, + "learning_rate": 0.0006349170775437024, + "loss": 0.4988, + "step": 13148 + }, + { + "epoch": 0.7346425678129452, + "grad_norm": 1.4076104164123535, + "learning_rate": 0.0006348890632003586, + "loss": 0.4837, + "step": 13149 + }, + { + "epoch": 0.7346984384166271, + "grad_norm": 0.8368995785713196, + "learning_rate": 0.0006348610488570148, + "loss": 0.4552, + "step": 13150 + }, + { + "epoch": 0.7347543090203089, + "grad_norm": 0.3935741186141968, + "learning_rate": 0.000634833034513671, + "loss": 0.3297, + "step": 13151 + }, + { + "epoch": 0.7348101796239909, + "grad_norm": 2.6301348209381104, + "learning_rate": 0.0006348050201703272, + "loss": 0.4723, + "step": 13152 + }, + { + "epoch": 0.7348660502276727, + "grad_norm": 0.4053865075111389, + "learning_rate": 0.0006347770058269834, + "loss": 0.3475, + "step": 13153 + }, + { + "epoch": 0.7349219208313545, + "grad_norm": 0.44636791944503784, + "learning_rate": 0.0006347489914836396, + "loss": 0.4037, + "step": 13154 + }, + { + "epoch": 0.7349777914350365, + "grad_norm": 5.737041473388672, + "learning_rate": 0.0006347209771402958, + "loss": 0.3596, + "step": 13155 + }, + { + "epoch": 0.7350336620387183, + "grad_norm": 0.5227437615394592, + "learning_rate": 0.0006346929627969522, + "loss": 0.4477, + "step": 13156 + }, + { + "epoch": 0.7350895326424002, + "grad_norm": 0.459338515996933, + "learning_rate": 0.0006346649484536082, + "loss": 0.4451, + "step": 13157 + }, + { + "epoch": 0.735145403246082, + "grad_norm": 0.3733355402946472, + "learning_rate": 0.0006346369341102645, + "loss": 0.4329, + "step": 13158 + }, + { + "epoch": 0.735201273849764, + "grad_norm": 0.5184904932975769, + "learning_rate": 0.0006346089197669207, + "loss": 0.4628, + "step": 13159 + }, + { + "epoch": 0.7352571444534458, + "grad_norm": 0.8705956935882568, + "learning_rate": 0.000634580905423577, + "loss": 0.4687, + "step": 13160 + }, + { + "epoch": 0.7353130150571277, + "grad_norm": 0.4111528992652893, + "learning_rate": 0.0006345528910802331, + "loss": 0.4387, + "step": 13161 + }, + { + "epoch": 0.7353688856608096, + "grad_norm": 0.529769241809845, + "learning_rate": 0.0006345248767368893, + "loss": 0.4926, + "step": 13162 + }, + { + "epoch": 0.7354247562644914, + "grad_norm": 0.6913014054298401, + "learning_rate": 0.0006344968623935456, + "loss": 0.4106, + "step": 13163 + }, + { + "epoch": 0.7354806268681733, + "grad_norm": 0.6503148078918457, + "learning_rate": 0.0006344688480502017, + "loss": 0.3475, + "step": 13164 + }, + { + "epoch": 0.7355364974718552, + "grad_norm": 0.6124517917633057, + "learning_rate": 0.0006344408337068579, + "loss": 0.3779, + "step": 13165 + }, + { + "epoch": 0.7355923680755371, + "grad_norm": 0.5031937956809998, + "learning_rate": 0.0006344128193635141, + "loss": 0.4516, + "step": 13166 + }, + { + "epoch": 0.7356482386792189, + "grad_norm": 0.5035142302513123, + "learning_rate": 0.0006343848050201704, + "loss": 0.3449, + "step": 13167 + }, + { + "epoch": 0.7357041092829008, + "grad_norm": 0.682859480381012, + "learning_rate": 0.0006343567906768265, + "loss": 0.5679, + "step": 13168 + }, + { + "epoch": 0.7357599798865827, + "grad_norm": 0.6005233526229858, + "learning_rate": 0.0006343287763334827, + "loss": 0.4704, + "step": 13169 + }, + { + "epoch": 0.7358158504902645, + "grad_norm": 0.9721318483352661, + "learning_rate": 0.0006343007619901389, + "loss": 0.4408, + "step": 13170 + }, + { + "epoch": 0.7358717210939464, + "grad_norm": 0.7188464999198914, + "learning_rate": 0.0006342727476467952, + "loss": 0.3722, + "step": 13171 + }, + { + "epoch": 0.7359275916976283, + "grad_norm": 0.5276073217391968, + "learning_rate": 0.0006342447333034513, + "loss": 0.4633, + "step": 13172 + }, + { + "epoch": 0.7359834623013102, + "grad_norm": 0.4198119640350342, + "learning_rate": 0.0006342167189601075, + "loss": 0.4831, + "step": 13173 + }, + { + "epoch": 0.736039332904992, + "grad_norm": 0.6907301545143127, + "learning_rate": 0.0006341887046167639, + "loss": 0.4165, + "step": 13174 + }, + { + "epoch": 0.7360952035086739, + "grad_norm": 0.36328378319740295, + "learning_rate": 0.0006341606902734201, + "loss": 0.396, + "step": 13175 + }, + { + "epoch": 0.7361510741123558, + "grad_norm": 0.5017676949501038, + "learning_rate": 0.0006341326759300762, + "loss": 0.5292, + "step": 13176 + }, + { + "epoch": 0.7362069447160376, + "grad_norm": 0.720326840877533, + "learning_rate": 0.0006341046615867324, + "loss": 0.418, + "step": 13177 + }, + { + "epoch": 0.7362628153197195, + "grad_norm": 0.5478630661964417, + "learning_rate": 0.0006340766472433887, + "loss": 0.4218, + "step": 13178 + }, + { + "epoch": 0.7363186859234014, + "grad_norm": 0.5431253910064697, + "learning_rate": 0.0006340486329000449, + "loss": 0.3933, + "step": 13179 + }, + { + "epoch": 0.7363745565270833, + "grad_norm": 0.5987685918807983, + "learning_rate": 0.000634020618556701, + "loss": 0.4783, + "step": 13180 + }, + { + "epoch": 0.7364304271307651, + "grad_norm": 0.4891214370727539, + "learning_rate": 0.0006339926042133573, + "loss": 0.4458, + "step": 13181 + }, + { + "epoch": 0.7364862977344471, + "grad_norm": 0.5047645568847656, + "learning_rate": 0.0006339645898700135, + "loss": 0.4031, + "step": 13182 + }, + { + "epoch": 0.7365421683381289, + "grad_norm": 0.6516581177711487, + "learning_rate": 0.0006339365755266697, + "loss": 0.4411, + "step": 13183 + }, + { + "epoch": 0.7365980389418108, + "grad_norm": 0.8130183219909668, + "learning_rate": 0.0006339085611833258, + "loss": 0.3821, + "step": 13184 + }, + { + "epoch": 0.7366539095454926, + "grad_norm": 0.4008040726184845, + "learning_rate": 0.0006338805468399821, + "loss": 0.4258, + "step": 13185 + }, + { + "epoch": 0.7367097801491745, + "grad_norm": 0.39202946424484253, + "learning_rate": 0.0006338525324966383, + "loss": 0.4276, + "step": 13186 + }, + { + "epoch": 0.7367656507528564, + "grad_norm": 1.0365314483642578, + "learning_rate": 0.0006338245181532944, + "loss": 0.5014, + "step": 13187 + }, + { + "epoch": 0.7368215213565382, + "grad_norm": 0.4346409738063812, + "learning_rate": 0.0006337965038099507, + "loss": 0.3952, + "step": 13188 + }, + { + "epoch": 0.7368773919602202, + "grad_norm": 0.3548353910446167, + "learning_rate": 0.0006337684894666069, + "loss": 0.3597, + "step": 13189 + }, + { + "epoch": 0.736933262563902, + "grad_norm": 1.0201395750045776, + "learning_rate": 0.0006337404751232631, + "loss": 0.4531, + "step": 13190 + }, + { + "epoch": 0.7369891331675839, + "grad_norm": 0.41817784309387207, + "learning_rate": 0.0006337124607799192, + "loss": 0.4169, + "step": 13191 + }, + { + "epoch": 0.7370450037712657, + "grad_norm": 2.0850470066070557, + "learning_rate": 0.0006336844464365756, + "loss": 0.3808, + "step": 13192 + }, + { + "epoch": 0.7371008743749476, + "grad_norm": 0.5554848909378052, + "learning_rate": 0.0006336564320932318, + "loss": 0.3558, + "step": 13193 + }, + { + "epoch": 0.7371567449786295, + "grad_norm": 0.7243155837059021, + "learning_rate": 0.000633628417749888, + "loss": 0.4576, + "step": 13194 + }, + { + "epoch": 0.7372126155823113, + "grad_norm": 0.4506779611110687, + "learning_rate": 0.0006336004034065442, + "loss": 0.406, + "step": 13195 + }, + { + "epoch": 0.7372684861859933, + "grad_norm": 0.7841095328330994, + "learning_rate": 0.0006335723890632004, + "loss": 0.4323, + "step": 13196 + }, + { + "epoch": 0.7373243567896751, + "grad_norm": 0.532844603061676, + "learning_rate": 0.0006335443747198566, + "loss": 0.4794, + "step": 13197 + }, + { + "epoch": 0.737380227393357, + "grad_norm": 0.3992776572704315, + "learning_rate": 0.0006335163603765128, + "loss": 0.5267, + "step": 13198 + }, + { + "epoch": 0.7374360979970389, + "grad_norm": 0.6555945873260498, + "learning_rate": 0.000633488346033169, + "loss": 0.3366, + "step": 13199 + }, + { + "epoch": 0.7374919686007207, + "grad_norm": 0.6519894599914551, + "learning_rate": 0.0006334603316898252, + "loss": 0.3255, + "step": 13200 + }, + { + "epoch": 0.7375478392044026, + "grad_norm": 0.7823646068572998, + "learning_rate": 0.0006334323173464814, + "loss": 0.4242, + "step": 13201 + }, + { + "epoch": 0.7376037098080844, + "grad_norm": 0.6012877821922302, + "learning_rate": 0.0006334043030031376, + "loss": 0.4662, + "step": 13202 + }, + { + "epoch": 0.7376595804117664, + "grad_norm": 2.2964749336242676, + "learning_rate": 0.0006333762886597938, + "loss": 0.4047, + "step": 13203 + }, + { + "epoch": 0.7377154510154482, + "grad_norm": 0.5129268169403076, + "learning_rate": 0.00063334827431645, + "loss": 0.4309, + "step": 13204 + }, + { + "epoch": 0.7377713216191301, + "grad_norm": 0.46364158391952515, + "learning_rate": 0.0006333202599731062, + "loss": 0.4354, + "step": 13205 + }, + { + "epoch": 0.737827192222812, + "grad_norm": 0.580616295337677, + "learning_rate": 0.0006332922456297624, + "loss": 0.381, + "step": 13206 + }, + { + "epoch": 0.7378830628264939, + "grad_norm": 0.4945921301841736, + "learning_rate": 0.0006332642312864186, + "loss": 0.4525, + "step": 13207 + }, + { + "epoch": 0.7379389334301757, + "grad_norm": 0.4340871572494507, + "learning_rate": 0.0006332362169430749, + "loss": 0.4346, + "step": 13208 + }, + { + "epoch": 0.7379948040338575, + "grad_norm": 0.5681777596473694, + "learning_rate": 0.0006332082025997311, + "loss": 0.6244, + "step": 13209 + }, + { + "epoch": 0.7380506746375395, + "grad_norm": 0.5616172552108765, + "learning_rate": 0.0006331801882563873, + "loss": 0.5225, + "step": 13210 + }, + { + "epoch": 0.7381065452412213, + "grad_norm": 0.5344250202178955, + "learning_rate": 0.0006331521739130435, + "loss": 0.4958, + "step": 13211 + }, + { + "epoch": 0.7381624158449032, + "grad_norm": 0.5675156116485596, + "learning_rate": 0.0006331241595696997, + "loss": 0.4536, + "step": 13212 + }, + { + "epoch": 0.7382182864485851, + "grad_norm": 0.49438169598579407, + "learning_rate": 0.000633096145226356, + "loss": 0.6138, + "step": 13213 + }, + { + "epoch": 0.738274157052267, + "grad_norm": 0.6790849566459656, + "learning_rate": 0.0006330681308830121, + "loss": 0.3881, + "step": 13214 + }, + { + "epoch": 0.7383300276559488, + "grad_norm": 0.6387701630592346, + "learning_rate": 0.0006330401165396683, + "loss": 0.44, + "step": 13215 + }, + { + "epoch": 0.7383858982596307, + "grad_norm": 1.0874027013778687, + "learning_rate": 0.0006330121021963245, + "loss": 0.556, + "step": 13216 + }, + { + "epoch": 0.7384417688633126, + "grad_norm": 0.41816574335098267, + "learning_rate": 0.0006329840878529808, + "loss": 0.3796, + "step": 13217 + }, + { + "epoch": 0.7384976394669944, + "grad_norm": 0.454549640417099, + "learning_rate": 0.0006329560735096369, + "loss": 0.3496, + "step": 13218 + }, + { + "epoch": 0.7385535100706763, + "grad_norm": 0.370614230632782, + "learning_rate": 0.0006329280591662931, + "loss": 0.3454, + "step": 13219 + }, + { + "epoch": 0.7386093806743582, + "grad_norm": 0.4411345422267914, + "learning_rate": 0.0006329000448229494, + "loss": 0.486, + "step": 13220 + }, + { + "epoch": 0.7386652512780401, + "grad_norm": 0.4899822473526001, + "learning_rate": 0.0006328720304796056, + "loss": 0.4808, + "step": 13221 + }, + { + "epoch": 0.7387211218817219, + "grad_norm": 0.6213960647583008, + "learning_rate": 0.0006328440161362617, + "loss": 0.5706, + "step": 13222 + }, + { + "epoch": 0.7387769924854038, + "grad_norm": 0.7081360220909119, + "learning_rate": 0.0006328160017929179, + "loss": 0.3871, + "step": 13223 + }, + { + "epoch": 0.7388328630890857, + "grad_norm": 0.654776930809021, + "learning_rate": 0.0006327879874495743, + "loss": 0.5575, + "step": 13224 + }, + { + "epoch": 0.7388887336927675, + "grad_norm": 1.0194714069366455, + "learning_rate": 0.0006327599731062305, + "loss": 0.3808, + "step": 13225 + }, + { + "epoch": 0.7389446042964494, + "grad_norm": 0.43404439091682434, + "learning_rate": 0.0006327319587628866, + "loss": 0.4748, + "step": 13226 + }, + { + "epoch": 0.7390004749001313, + "grad_norm": 0.4197913706302643, + "learning_rate": 0.0006327039444195429, + "loss": 0.4161, + "step": 13227 + }, + { + "epoch": 0.7390563455038132, + "grad_norm": 0.5548290014266968, + "learning_rate": 0.0006326759300761991, + "loss": 0.6071, + "step": 13228 + }, + { + "epoch": 0.739112216107495, + "grad_norm": 0.41753697395324707, + "learning_rate": 0.0006326479157328552, + "loss": 0.3674, + "step": 13229 + }, + { + "epoch": 0.739168086711177, + "grad_norm": 1.4053220748901367, + "learning_rate": 0.0006326199013895114, + "loss": 0.7036, + "step": 13230 + }, + { + "epoch": 0.7392239573148588, + "grad_norm": 0.6249507069587708, + "learning_rate": 0.0006325918870461677, + "loss": 0.4896, + "step": 13231 + }, + { + "epoch": 0.7392798279185406, + "grad_norm": 0.5418022871017456, + "learning_rate": 0.0006325638727028239, + "loss": 0.4266, + "step": 13232 + }, + { + "epoch": 0.7393356985222226, + "grad_norm": 0.4587683379650116, + "learning_rate": 0.00063253585835948, + "loss": 0.5188, + "step": 13233 + }, + { + "epoch": 0.7393915691259044, + "grad_norm": 1.3566430807113647, + "learning_rate": 0.0006325078440161362, + "loss": 0.4123, + "step": 13234 + }, + { + "epoch": 0.7394474397295863, + "grad_norm": 0.7144213914871216, + "learning_rate": 0.0006324798296727925, + "loss": 0.4426, + "step": 13235 + }, + { + "epoch": 0.7395033103332681, + "grad_norm": 0.603209376335144, + "learning_rate": 0.0006324518153294487, + "loss": 0.4197, + "step": 13236 + }, + { + "epoch": 0.7395591809369501, + "grad_norm": 0.6586359739303589, + "learning_rate": 0.0006324238009861048, + "loss": 0.8366, + "step": 13237 + }, + { + "epoch": 0.7396150515406319, + "grad_norm": 4.614570617675781, + "learning_rate": 0.0006323957866427611, + "loss": 0.4667, + "step": 13238 + }, + { + "epoch": 0.7396709221443137, + "grad_norm": 0.4566130042076111, + "learning_rate": 0.0006323677722994173, + "loss": 0.3775, + "step": 13239 + }, + { + "epoch": 0.7397267927479957, + "grad_norm": 0.44623857736587524, + "learning_rate": 0.0006323397579560735, + "loss": 0.3623, + "step": 13240 + }, + { + "epoch": 0.7397826633516775, + "grad_norm": 0.5101231336593628, + "learning_rate": 0.0006323117436127296, + "loss": 0.4947, + "step": 13241 + }, + { + "epoch": 0.7398385339553594, + "grad_norm": 0.4178868234157562, + "learning_rate": 0.000632283729269386, + "loss": 0.4487, + "step": 13242 + }, + { + "epoch": 0.7398944045590412, + "grad_norm": 1.6827003955841064, + "learning_rate": 0.0006322557149260422, + "loss": 0.5351, + "step": 13243 + }, + { + "epoch": 0.7399502751627232, + "grad_norm": 0.5120741724967957, + "learning_rate": 0.0006322277005826984, + "loss": 0.5353, + "step": 13244 + }, + { + "epoch": 0.740006145766405, + "grad_norm": 0.5344571471214294, + "learning_rate": 0.0006321996862393546, + "loss": 0.5227, + "step": 13245 + }, + { + "epoch": 0.7400620163700868, + "grad_norm": 0.666038990020752, + "learning_rate": 0.0006321716718960108, + "loss": 0.4998, + "step": 13246 + }, + { + "epoch": 0.7401178869737688, + "grad_norm": 0.5354704856872559, + "learning_rate": 0.000632143657552667, + "loss": 0.4991, + "step": 13247 + }, + { + "epoch": 0.7401737575774506, + "grad_norm": 0.3844585716724396, + "learning_rate": 0.0006321156432093231, + "loss": 0.4516, + "step": 13248 + }, + { + "epoch": 0.7402296281811325, + "grad_norm": 0.45989352464675903, + "learning_rate": 0.0006320876288659794, + "loss": 0.3708, + "step": 13249 + }, + { + "epoch": 0.7402854987848144, + "grad_norm": 0.43160709738731384, + "learning_rate": 0.0006320596145226356, + "loss": 0.436, + "step": 13250 + }, + { + "epoch": 0.7403413693884963, + "grad_norm": 0.4339282214641571, + "learning_rate": 0.0006320316001792918, + "loss": 0.3987, + "step": 13251 + }, + { + "epoch": 0.7403972399921781, + "grad_norm": 0.5428483486175537, + "learning_rate": 0.000632003585835948, + "loss": 0.4813, + "step": 13252 + }, + { + "epoch": 0.7404531105958599, + "grad_norm": 0.4556979537010193, + "learning_rate": 0.0006319755714926042, + "loss": 0.4069, + "step": 13253 + }, + { + "epoch": 0.7405089811995419, + "grad_norm": 0.5507538914680481, + "learning_rate": 0.0006319475571492604, + "loss": 0.4064, + "step": 13254 + }, + { + "epoch": 0.7405648518032237, + "grad_norm": 0.388295441865921, + "learning_rate": 0.0006319195428059166, + "loss": 0.4031, + "step": 13255 + }, + { + "epoch": 0.7406207224069056, + "grad_norm": 0.3906558156013489, + "learning_rate": 0.0006318915284625728, + "loss": 0.3566, + "step": 13256 + }, + { + "epoch": 0.7406765930105875, + "grad_norm": 0.7019330859184265, + "learning_rate": 0.000631863514119229, + "loss": 0.5595, + "step": 13257 + }, + { + "epoch": 0.7407324636142694, + "grad_norm": 0.47992604970932007, + "learning_rate": 0.0006318354997758853, + "loss": 0.4564, + "step": 13258 + }, + { + "epoch": 0.7407883342179512, + "grad_norm": 0.4903627932071686, + "learning_rate": 0.0006318074854325416, + "loss": 0.4886, + "step": 13259 + }, + { + "epoch": 0.740844204821633, + "grad_norm": 0.5603652000427246, + "learning_rate": 0.0006317794710891977, + "loss": 0.4427, + "step": 13260 + }, + { + "epoch": 0.740900075425315, + "grad_norm": 0.4653174579143524, + "learning_rate": 0.0006317514567458539, + "loss": 0.5364, + "step": 13261 + }, + { + "epoch": 0.7409559460289968, + "grad_norm": 0.30961108207702637, + "learning_rate": 0.0006317234424025101, + "loss": 0.3492, + "step": 13262 + }, + { + "epoch": 0.7410118166326787, + "grad_norm": 0.3760753870010376, + "learning_rate": 0.0006316954280591664, + "loss": 0.3782, + "step": 13263 + }, + { + "epoch": 0.7410676872363606, + "grad_norm": 1.6391338109970093, + "learning_rate": 0.0006316674137158225, + "loss": 0.591, + "step": 13264 + }, + { + "epoch": 0.7411235578400425, + "grad_norm": 2.5959277153015137, + "learning_rate": 0.0006316393993724787, + "loss": 0.5094, + "step": 13265 + }, + { + "epoch": 0.7411794284437243, + "grad_norm": 0.8130508661270142, + "learning_rate": 0.000631611385029135, + "loss": 0.4255, + "step": 13266 + }, + { + "epoch": 0.7412352990474063, + "grad_norm": 0.8260494470596313, + "learning_rate": 0.0006315833706857912, + "loss": 0.5127, + "step": 13267 + }, + { + "epoch": 0.7412911696510881, + "grad_norm": 0.5084616541862488, + "learning_rate": 0.0006315553563424473, + "loss": 0.4397, + "step": 13268 + }, + { + "epoch": 0.7413470402547699, + "grad_norm": 1.2674381732940674, + "learning_rate": 0.0006315273419991035, + "loss": 0.5024, + "step": 13269 + }, + { + "epoch": 0.7414029108584518, + "grad_norm": 0.4924171566963196, + "learning_rate": 0.0006314993276557598, + "loss": 0.4459, + "step": 13270 + }, + { + "epoch": 0.7414587814621337, + "grad_norm": 0.44559261202812195, + "learning_rate": 0.0006314713133124159, + "loss": 0.4621, + "step": 13271 + }, + { + "epoch": 0.7415146520658156, + "grad_norm": 0.41469433903694153, + "learning_rate": 0.0006314432989690721, + "loss": 0.4993, + "step": 13272 + }, + { + "epoch": 0.7415705226694974, + "grad_norm": 0.501128077507019, + "learning_rate": 0.0006314152846257283, + "loss": 0.5014, + "step": 13273 + }, + { + "epoch": 0.7416263932731794, + "grad_norm": 0.5118246078491211, + "learning_rate": 0.0006313872702823847, + "loss": 0.3387, + "step": 13274 + }, + { + "epoch": 0.7416822638768612, + "grad_norm": 0.9050430655479431, + "learning_rate": 0.0006313592559390408, + "loss": 0.4323, + "step": 13275 + }, + { + "epoch": 0.741738134480543, + "grad_norm": 0.4797267019748688, + "learning_rate": 0.000631331241595697, + "loss": 0.3772, + "step": 13276 + }, + { + "epoch": 0.7417940050842249, + "grad_norm": 0.6175113916397095, + "learning_rate": 0.0006313032272523533, + "loss": 0.3765, + "step": 13277 + }, + { + "epoch": 0.7418498756879068, + "grad_norm": 0.49298977851867676, + "learning_rate": 0.0006312752129090095, + "loss": 0.4781, + "step": 13278 + }, + { + "epoch": 0.7419057462915887, + "grad_norm": 4.885138988494873, + "learning_rate": 0.0006312471985656656, + "loss": 0.4532, + "step": 13279 + }, + { + "epoch": 0.7419616168952705, + "grad_norm": 0.5069345831871033, + "learning_rate": 0.0006312191842223218, + "loss": 0.542, + "step": 13280 + }, + { + "epoch": 0.7420174874989525, + "grad_norm": 0.4130367636680603, + "learning_rate": 0.0006311911698789781, + "loss": 0.3791, + "step": 13281 + }, + { + "epoch": 0.7420733581026343, + "grad_norm": 0.48047319054603577, + "learning_rate": 0.0006311631555356343, + "loss": 0.5791, + "step": 13282 + }, + { + "epoch": 0.7421292287063161, + "grad_norm": 0.5390379428863525, + "learning_rate": 0.0006311351411922904, + "loss": 0.3939, + "step": 13283 + }, + { + "epoch": 0.7421850993099981, + "grad_norm": 0.6175254583358765, + "learning_rate": 0.0006311071268489467, + "loss": 0.4243, + "step": 13284 + }, + { + "epoch": 0.7422409699136799, + "grad_norm": 0.46131187677383423, + "learning_rate": 0.0006310791125056029, + "loss": 0.4459, + "step": 13285 + }, + { + "epoch": 0.7422968405173618, + "grad_norm": 0.41458970308303833, + "learning_rate": 0.0006310510981622591, + "loss": 0.4705, + "step": 13286 + }, + { + "epoch": 0.7423527111210436, + "grad_norm": 1.6675711870193481, + "learning_rate": 0.0006310230838189152, + "loss": 0.5491, + "step": 13287 + }, + { + "epoch": 0.7424085817247256, + "grad_norm": 0.6484756469726562, + "learning_rate": 0.0006309950694755715, + "loss": 0.4143, + "step": 13288 + }, + { + "epoch": 0.7424644523284074, + "grad_norm": 1.1285018920898438, + "learning_rate": 0.0006309670551322277, + "loss": 0.4219, + "step": 13289 + }, + { + "epoch": 0.7425203229320892, + "grad_norm": 0.5204700231552124, + "learning_rate": 0.0006309390407888838, + "loss": 0.5211, + "step": 13290 + }, + { + "epoch": 0.7425761935357712, + "grad_norm": 0.6188123226165771, + "learning_rate": 0.0006309110264455402, + "loss": 0.3804, + "step": 13291 + }, + { + "epoch": 0.742632064139453, + "grad_norm": 0.45598018169403076, + "learning_rate": 0.0006308830121021964, + "loss": 0.378, + "step": 13292 + }, + { + "epoch": 0.7426879347431349, + "grad_norm": 4.439958095550537, + "learning_rate": 0.0006308549977588526, + "loss": 0.3968, + "step": 13293 + }, + { + "epoch": 0.7427438053468167, + "grad_norm": 0.6411669850349426, + "learning_rate": 0.0006308269834155087, + "loss": 0.4665, + "step": 13294 + }, + { + "epoch": 0.7427996759504987, + "grad_norm": 0.5799289345741272, + "learning_rate": 0.000630798969072165, + "loss": 0.5382, + "step": 13295 + }, + { + "epoch": 0.7428555465541805, + "grad_norm": 0.39914536476135254, + "learning_rate": 0.0006307709547288212, + "loss": 0.3061, + "step": 13296 + }, + { + "epoch": 0.7429114171578624, + "grad_norm": 1.563480257987976, + "learning_rate": 0.0006307429403854774, + "loss": 0.4567, + "step": 13297 + }, + { + "epoch": 0.7429672877615443, + "grad_norm": 0.42845913767814636, + "learning_rate": 0.0006307149260421336, + "loss": 0.526, + "step": 13298 + }, + { + "epoch": 0.7430231583652261, + "grad_norm": 0.5416348576545715, + "learning_rate": 0.0006306869116987898, + "loss": 0.403, + "step": 13299 + }, + { + "epoch": 0.743079028968908, + "grad_norm": 4.770346164703369, + "learning_rate": 0.000630658897355446, + "loss": 0.4391, + "step": 13300 + }, + { + "epoch": 0.7431348995725899, + "grad_norm": 0.48590293526649475, + "learning_rate": 0.0006306308830121022, + "loss": 0.4767, + "step": 13301 + }, + { + "epoch": 0.7431907701762718, + "grad_norm": 2.4837803840637207, + "learning_rate": 0.0006306028686687584, + "loss": 0.5094, + "step": 13302 + }, + { + "epoch": 0.7432466407799536, + "grad_norm": 0.3937985599040985, + "learning_rate": 0.0006305748543254146, + "loss": 0.4227, + "step": 13303 + }, + { + "epoch": 0.7433025113836355, + "grad_norm": 0.6820511817932129, + "learning_rate": 0.0006305468399820708, + "loss": 0.4442, + "step": 13304 + }, + { + "epoch": 0.7433583819873174, + "grad_norm": 0.3319840133190155, + "learning_rate": 0.0006305188256387271, + "loss": 0.3898, + "step": 13305 + }, + { + "epoch": 0.7434142525909992, + "grad_norm": 1.7986401319503784, + "learning_rate": 0.0006304908112953832, + "loss": 0.4764, + "step": 13306 + }, + { + "epoch": 0.7434701231946811, + "grad_norm": 1.679598093032837, + "learning_rate": 0.0006304627969520394, + "loss": 0.422, + "step": 13307 + }, + { + "epoch": 0.743525993798363, + "grad_norm": 0.4615201950073242, + "learning_rate": 0.0006304347826086957, + "loss": 0.4441, + "step": 13308 + }, + { + "epoch": 0.7435818644020449, + "grad_norm": 0.5334694981575012, + "learning_rate": 0.000630406768265352, + "loss": 0.7014, + "step": 13309 + }, + { + "epoch": 0.7436377350057267, + "grad_norm": 0.5213630199432373, + "learning_rate": 0.0006303787539220081, + "loss": 0.4861, + "step": 13310 + }, + { + "epoch": 0.7436936056094086, + "grad_norm": 0.4448128342628479, + "learning_rate": 0.0006303507395786643, + "loss": 0.3681, + "step": 13311 + }, + { + "epoch": 0.7437494762130905, + "grad_norm": 0.5819923281669617, + "learning_rate": 0.0006303227252353205, + "loss": 0.4684, + "step": 13312 + }, + { + "epoch": 0.7438053468167724, + "grad_norm": 0.44244131445884705, + "learning_rate": 0.0006302947108919767, + "loss": 0.4651, + "step": 13313 + }, + { + "epoch": 0.7438612174204542, + "grad_norm": 0.4410144090652466, + "learning_rate": 0.0006302666965486329, + "loss": 0.4486, + "step": 13314 + }, + { + "epoch": 0.7439170880241361, + "grad_norm": 0.5069446563720703, + "learning_rate": 0.0006302386822052891, + "loss": 0.4859, + "step": 13315 + }, + { + "epoch": 0.743972958627818, + "grad_norm": 0.5461729168891907, + "learning_rate": 0.0006302106678619454, + "loss": 0.516, + "step": 13316 + }, + { + "epoch": 0.7440288292314998, + "grad_norm": 0.9523918628692627, + "learning_rate": 0.0006301826535186015, + "loss": 0.4096, + "step": 13317 + }, + { + "epoch": 0.7440846998351818, + "grad_norm": 0.5091478228569031, + "learning_rate": 0.0006301546391752577, + "loss": 0.4398, + "step": 13318 + }, + { + "epoch": 0.7441405704388636, + "grad_norm": 0.6333234906196594, + "learning_rate": 0.0006301266248319139, + "loss": 0.4275, + "step": 13319 + }, + { + "epoch": 0.7441964410425455, + "grad_norm": 0.4504019320011139, + "learning_rate": 0.0006300986104885702, + "loss": 0.4389, + "step": 13320 + }, + { + "epoch": 0.7442523116462273, + "grad_norm": 0.4236977994441986, + "learning_rate": 0.0006300705961452263, + "loss": 0.5497, + "step": 13321 + }, + { + "epoch": 0.7443081822499092, + "grad_norm": 0.7635577321052551, + "learning_rate": 0.0006300425818018825, + "loss": 0.4191, + "step": 13322 + }, + { + "epoch": 0.7443640528535911, + "grad_norm": 0.40436068177223206, + "learning_rate": 0.0006300145674585388, + "loss": 0.4122, + "step": 13323 + }, + { + "epoch": 0.7444199234572729, + "grad_norm": 0.3395594358444214, + "learning_rate": 0.000629986553115195, + "loss": 0.3489, + "step": 13324 + }, + { + "epoch": 0.7444757940609549, + "grad_norm": 0.43191656470298767, + "learning_rate": 0.0006299585387718512, + "loss": 0.4274, + "step": 13325 + }, + { + "epoch": 0.7445316646646367, + "grad_norm": 0.6294686794281006, + "learning_rate": 0.0006299305244285074, + "loss": 0.4508, + "step": 13326 + }, + { + "epoch": 0.7445875352683186, + "grad_norm": 0.4615964889526367, + "learning_rate": 0.0006299025100851637, + "loss": 0.5018, + "step": 13327 + }, + { + "epoch": 0.7446434058720004, + "grad_norm": 0.49758821725845337, + "learning_rate": 0.0006298744957418199, + "loss": 0.3264, + "step": 13328 + }, + { + "epoch": 0.7446992764756823, + "grad_norm": 0.5054938793182373, + "learning_rate": 0.000629846481398476, + "loss": 0.4736, + "step": 13329 + }, + { + "epoch": 0.7447551470793642, + "grad_norm": 4.722684383392334, + "learning_rate": 0.0006298184670551323, + "loss": 0.4804, + "step": 13330 + }, + { + "epoch": 0.744811017683046, + "grad_norm": 0.4039250612258911, + "learning_rate": 0.0006297904527117885, + "loss": 0.4211, + "step": 13331 + }, + { + "epoch": 0.744866888286728, + "grad_norm": 0.6158595681190491, + "learning_rate": 0.0006297624383684447, + "loss": 0.5243, + "step": 13332 + }, + { + "epoch": 0.7449227588904098, + "grad_norm": 1.6458386182785034, + "learning_rate": 0.0006297344240251008, + "loss": 0.4462, + "step": 13333 + }, + { + "epoch": 0.7449786294940917, + "grad_norm": 0.4700935482978821, + "learning_rate": 0.0006297064096817571, + "loss": 0.4947, + "step": 13334 + }, + { + "epoch": 0.7450345000977736, + "grad_norm": 0.4438633322715759, + "learning_rate": 0.0006296783953384133, + "loss": 0.3491, + "step": 13335 + }, + { + "epoch": 0.7450903707014555, + "grad_norm": 0.5092584490776062, + "learning_rate": 0.0006296503809950694, + "loss": 0.602, + "step": 13336 + }, + { + "epoch": 0.7451462413051373, + "grad_norm": 0.5443742871284485, + "learning_rate": 0.0006296223666517257, + "loss": 0.3718, + "step": 13337 + }, + { + "epoch": 0.7452021119088191, + "grad_norm": 1.5289788246154785, + "learning_rate": 0.0006295943523083819, + "loss": 0.4516, + "step": 13338 + }, + { + "epoch": 0.7452579825125011, + "grad_norm": 0.3984651565551758, + "learning_rate": 0.0006295663379650381, + "loss": 0.4624, + "step": 13339 + }, + { + "epoch": 0.7453138531161829, + "grad_norm": 0.7666395306587219, + "learning_rate": 0.0006295383236216942, + "loss": 0.4659, + "step": 13340 + }, + { + "epoch": 0.7453697237198648, + "grad_norm": 0.7225918173789978, + "learning_rate": 0.0006295103092783506, + "loss": 0.4479, + "step": 13341 + }, + { + "epoch": 0.7454255943235467, + "grad_norm": 1.17424476146698, + "learning_rate": 0.0006294822949350068, + "loss": 0.3974, + "step": 13342 + }, + { + "epoch": 0.7454814649272286, + "grad_norm": 0.5248534083366394, + "learning_rate": 0.000629454280591663, + "loss": 0.4418, + "step": 13343 + }, + { + "epoch": 0.7455373355309104, + "grad_norm": 1.8060722351074219, + "learning_rate": 0.0006294262662483191, + "loss": 0.4099, + "step": 13344 + }, + { + "epoch": 0.7455932061345922, + "grad_norm": 1.9985966682434082, + "learning_rate": 0.0006293982519049754, + "loss": 0.5992, + "step": 13345 + }, + { + "epoch": 0.7456490767382742, + "grad_norm": 0.6111483573913574, + "learning_rate": 0.0006293702375616316, + "loss": 0.4234, + "step": 13346 + }, + { + "epoch": 0.745704947341956, + "grad_norm": 0.8491196036338806, + "learning_rate": 0.0006293422232182878, + "loss": 0.4277, + "step": 13347 + }, + { + "epoch": 0.7457608179456379, + "grad_norm": 0.3442361652851105, + "learning_rate": 0.000629314208874944, + "loss": 0.3656, + "step": 13348 + }, + { + "epoch": 0.7458166885493198, + "grad_norm": 0.5199165344238281, + "learning_rate": 0.0006292861945316002, + "loss": 0.4567, + "step": 13349 + }, + { + "epoch": 0.7458725591530017, + "grad_norm": 0.3850245177745819, + "learning_rate": 0.0006292581801882564, + "loss": 0.4386, + "step": 13350 + }, + { + "epoch": 0.7459284297566835, + "grad_norm": 0.5529813766479492, + "learning_rate": 0.0006292301658449126, + "loss": 0.4685, + "step": 13351 + }, + { + "epoch": 0.7459843003603654, + "grad_norm": 0.38068604469299316, + "learning_rate": 0.0006292021515015688, + "loss": 0.3852, + "step": 13352 + }, + { + "epoch": 0.7460401709640473, + "grad_norm": 1.1642863750457764, + "learning_rate": 0.000629174137158225, + "loss": 0.4155, + "step": 13353 + }, + { + "epoch": 0.7460960415677291, + "grad_norm": 0.505713701248169, + "learning_rate": 0.0006291461228148812, + "loss": 0.303, + "step": 13354 + }, + { + "epoch": 0.746151912171411, + "grad_norm": 0.40578892827033997, + "learning_rate": 0.0006291181084715374, + "loss": 0.4668, + "step": 13355 + }, + { + "epoch": 0.7462077827750929, + "grad_norm": 0.4539047181606293, + "learning_rate": 0.0006290900941281936, + "loss": 0.3971, + "step": 13356 + }, + { + "epoch": 0.7462636533787748, + "grad_norm": 0.43949514627456665, + "learning_rate": 0.0006290620797848498, + "loss": 0.4961, + "step": 13357 + }, + { + "epoch": 0.7463195239824566, + "grad_norm": 0.5134631395339966, + "learning_rate": 0.000629034065441506, + "loss": 0.3894, + "step": 13358 + }, + { + "epoch": 0.7463753945861386, + "grad_norm": 1.6555739641189575, + "learning_rate": 0.0006290060510981623, + "loss": 0.5165, + "step": 13359 + }, + { + "epoch": 0.7464312651898204, + "grad_norm": 0.5523767471313477, + "learning_rate": 0.0006289780367548185, + "loss": 0.4794, + "step": 13360 + }, + { + "epoch": 0.7464871357935022, + "grad_norm": 0.4588357210159302, + "learning_rate": 0.0006289500224114747, + "loss": 0.4065, + "step": 13361 + }, + { + "epoch": 0.7465430063971841, + "grad_norm": 0.4990856945514679, + "learning_rate": 0.000628922008068131, + "loss": 0.3854, + "step": 13362 + }, + { + "epoch": 0.746598877000866, + "grad_norm": 0.4529532790184021, + "learning_rate": 0.0006288939937247871, + "loss": 0.3685, + "step": 13363 + }, + { + "epoch": 0.7466547476045479, + "grad_norm": 3.8709230422973633, + "learning_rate": 0.0006288659793814433, + "loss": 0.4999, + "step": 13364 + }, + { + "epoch": 0.7467106182082297, + "grad_norm": 0.3490064740180969, + "learning_rate": 0.0006288379650380995, + "loss": 0.375, + "step": 13365 + }, + { + "epoch": 0.7467664888119117, + "grad_norm": 0.5476385354995728, + "learning_rate": 0.0006288099506947558, + "loss": 0.3718, + "step": 13366 + }, + { + "epoch": 0.7468223594155935, + "grad_norm": 1.4578619003295898, + "learning_rate": 0.0006287819363514119, + "loss": 0.4654, + "step": 13367 + }, + { + "epoch": 0.7468782300192753, + "grad_norm": 0.5275477170944214, + "learning_rate": 0.0006287539220080681, + "loss": 0.5584, + "step": 13368 + }, + { + "epoch": 0.7469341006229572, + "grad_norm": 0.44399651885032654, + "learning_rate": 0.0006287259076647244, + "loss": 0.5098, + "step": 13369 + }, + { + "epoch": 0.7469899712266391, + "grad_norm": 0.4890243113040924, + "learning_rate": 0.0006286978933213806, + "loss": 0.5997, + "step": 13370 + }, + { + "epoch": 0.747045841830321, + "grad_norm": 0.9660753011703491, + "learning_rate": 0.0006286698789780367, + "loss": 0.4342, + "step": 13371 + }, + { + "epoch": 0.7471017124340028, + "grad_norm": 0.4609638750553131, + "learning_rate": 0.0006286418646346929, + "loss": 0.4206, + "step": 13372 + }, + { + "epoch": 0.7471575830376848, + "grad_norm": 0.6291660666465759, + "learning_rate": 0.0006286138502913492, + "loss": 0.4727, + "step": 13373 + }, + { + "epoch": 0.7472134536413666, + "grad_norm": 7.0433502197265625, + "learning_rate": 0.0006285858359480055, + "loss": 0.4323, + "step": 13374 + }, + { + "epoch": 0.7472693242450484, + "grad_norm": 0.4218199551105499, + "learning_rate": 0.0006285578216046615, + "loss": 0.3874, + "step": 13375 + }, + { + "epoch": 0.7473251948487304, + "grad_norm": 0.5652777552604675, + "learning_rate": 0.0006285298072613179, + "loss": 0.6064, + "step": 13376 + }, + { + "epoch": 0.7473810654524122, + "grad_norm": 0.41820666193962097, + "learning_rate": 0.0006285017929179741, + "loss": 0.3787, + "step": 13377 + }, + { + "epoch": 0.7474369360560941, + "grad_norm": 0.5290716290473938, + "learning_rate": 0.0006284737785746302, + "loss": 0.5651, + "step": 13378 + }, + { + "epoch": 0.7474928066597759, + "grad_norm": 0.387228399515152, + "learning_rate": 0.0006284457642312864, + "loss": 0.3288, + "step": 13379 + }, + { + "epoch": 0.7475486772634579, + "grad_norm": 0.9385089874267578, + "learning_rate": 0.0006284177498879427, + "loss": 0.4278, + "step": 13380 + }, + { + "epoch": 0.7476045478671397, + "grad_norm": 1.155389666557312, + "learning_rate": 0.0006283897355445989, + "loss": 0.3564, + "step": 13381 + }, + { + "epoch": 0.7476604184708215, + "grad_norm": 0.42893552780151367, + "learning_rate": 0.000628361721201255, + "loss": 0.4597, + "step": 13382 + }, + { + "epoch": 0.7477162890745035, + "grad_norm": 0.3841746151447296, + "learning_rate": 0.0006283337068579112, + "loss": 0.4656, + "step": 13383 + }, + { + "epoch": 0.7477721596781853, + "grad_norm": 0.4032016694545746, + "learning_rate": 0.0006283056925145675, + "loss": 0.345, + "step": 13384 + }, + { + "epoch": 0.7478280302818672, + "grad_norm": 2.2065465450286865, + "learning_rate": 0.0006282776781712237, + "loss": 0.4033, + "step": 13385 + }, + { + "epoch": 0.747883900885549, + "grad_norm": 1.503813624382019, + "learning_rate": 0.0006282496638278798, + "loss": 0.4237, + "step": 13386 + }, + { + "epoch": 0.747939771489231, + "grad_norm": 0.4030596911907196, + "learning_rate": 0.0006282216494845361, + "loss": 0.4286, + "step": 13387 + }, + { + "epoch": 0.7479956420929128, + "grad_norm": 0.38072219491004944, + "learning_rate": 0.0006281936351411923, + "loss": 0.4332, + "step": 13388 + }, + { + "epoch": 0.7480515126965946, + "grad_norm": 0.4279056489467621, + "learning_rate": 0.0006281656207978485, + "loss": 0.5291, + "step": 13389 + }, + { + "epoch": 0.7481073833002766, + "grad_norm": 0.5513595342636108, + "learning_rate": 0.0006281376064545046, + "loss": 0.4473, + "step": 13390 + }, + { + "epoch": 0.7481632539039584, + "grad_norm": 2.499844551086426, + "learning_rate": 0.000628109592111161, + "loss": 0.4641, + "step": 13391 + }, + { + "epoch": 0.7482191245076403, + "grad_norm": 0.7124417424201965, + "learning_rate": 0.0006280815777678172, + "loss": 0.4952, + "step": 13392 + }, + { + "epoch": 0.7482749951113222, + "grad_norm": 0.5735433101654053, + "learning_rate": 0.0006280535634244734, + "loss": 0.5758, + "step": 13393 + }, + { + "epoch": 0.7483308657150041, + "grad_norm": 0.4740225672721863, + "learning_rate": 0.0006280255490811296, + "loss": 0.3938, + "step": 13394 + }, + { + "epoch": 0.7483867363186859, + "grad_norm": 0.48352429270744324, + "learning_rate": 0.0006279975347377858, + "loss": 0.4139, + "step": 13395 + }, + { + "epoch": 0.7484426069223677, + "grad_norm": 0.4375140070915222, + "learning_rate": 0.000627969520394442, + "loss": 0.4125, + "step": 13396 + }, + { + "epoch": 0.7484984775260497, + "grad_norm": 0.5228789448738098, + "learning_rate": 0.0006279415060510981, + "loss": 0.468, + "step": 13397 + }, + { + "epoch": 0.7485543481297315, + "grad_norm": 0.738344132900238, + "learning_rate": 0.0006279134917077544, + "loss": 0.7105, + "step": 13398 + }, + { + "epoch": 0.7486102187334134, + "grad_norm": 0.41923993825912476, + "learning_rate": 0.0006278854773644106, + "loss": 0.3677, + "step": 13399 + }, + { + "epoch": 0.7486660893370953, + "grad_norm": 0.47236156463623047, + "learning_rate": 0.0006278574630210668, + "loss": 0.4978, + "step": 13400 + }, + { + "epoch": 0.7487219599407772, + "grad_norm": 0.4082318842411041, + "learning_rate": 0.000627829448677723, + "loss": 0.4011, + "step": 13401 + }, + { + "epoch": 0.748777830544459, + "grad_norm": 0.51844722032547, + "learning_rate": 0.0006278014343343792, + "loss": 0.5339, + "step": 13402 + }, + { + "epoch": 0.7488337011481409, + "grad_norm": 0.36014071106910706, + "learning_rate": 0.0006277734199910354, + "loss": 0.3801, + "step": 13403 + }, + { + "epoch": 0.7488895717518228, + "grad_norm": 0.45396196842193604, + "learning_rate": 0.0006277454056476916, + "loss": 0.4376, + "step": 13404 + }, + { + "epoch": 0.7489454423555046, + "grad_norm": 0.5240358114242554, + "learning_rate": 0.0006277173913043478, + "loss": 0.4035, + "step": 13405 + }, + { + "epoch": 0.7490013129591865, + "grad_norm": 0.46362411975860596, + "learning_rate": 0.000627689376961004, + "loss": 0.4521, + "step": 13406 + }, + { + "epoch": 0.7490571835628684, + "grad_norm": 0.4722874164581299, + "learning_rate": 0.0006276613626176602, + "loss": 0.4001, + "step": 13407 + }, + { + "epoch": 0.7491130541665503, + "grad_norm": 0.4018875062465668, + "learning_rate": 0.0006276333482743166, + "loss": 0.3015, + "step": 13408 + }, + { + "epoch": 0.7491689247702321, + "grad_norm": 0.4613550007343292, + "learning_rate": 0.0006276053339309727, + "loss": 0.3249, + "step": 13409 + }, + { + "epoch": 0.7492247953739141, + "grad_norm": 0.38273221254348755, + "learning_rate": 0.0006275773195876289, + "loss": 0.3522, + "step": 13410 + }, + { + "epoch": 0.7492806659775959, + "grad_norm": 0.39930489659309387, + "learning_rate": 0.0006275493052442851, + "loss": 0.3551, + "step": 13411 + }, + { + "epoch": 0.7493365365812777, + "grad_norm": 0.4791734218597412, + "learning_rate": 0.0006275212909009414, + "loss": 0.4612, + "step": 13412 + }, + { + "epoch": 0.7493924071849596, + "grad_norm": 0.47667452692985535, + "learning_rate": 0.0006274932765575975, + "loss": 0.4767, + "step": 13413 + }, + { + "epoch": 0.7494482777886415, + "grad_norm": 0.5383352637290955, + "learning_rate": 0.0006274652622142537, + "loss": 0.3816, + "step": 13414 + }, + { + "epoch": 0.7495041483923234, + "grad_norm": 0.5089861750602722, + "learning_rate": 0.0006274372478709099, + "loss": 0.3836, + "step": 13415 + }, + { + "epoch": 0.7495600189960052, + "grad_norm": 0.4007509648799896, + "learning_rate": 0.0006274092335275662, + "loss": 0.3965, + "step": 13416 + }, + { + "epoch": 0.7496158895996872, + "grad_norm": 0.39363938570022583, + "learning_rate": 0.0006273812191842223, + "loss": 0.5207, + "step": 13417 + }, + { + "epoch": 0.749671760203369, + "grad_norm": 0.5094574689865112, + "learning_rate": 0.0006273532048408785, + "loss": 0.3649, + "step": 13418 + }, + { + "epoch": 0.7497276308070508, + "grad_norm": 0.5621894001960754, + "learning_rate": 0.0006273251904975348, + "loss": 0.4566, + "step": 13419 + }, + { + "epoch": 0.7497835014107327, + "grad_norm": 1.1557666063308716, + "learning_rate": 0.0006272971761541909, + "loss": 0.7455, + "step": 13420 + }, + { + "epoch": 0.7498393720144146, + "grad_norm": 1.3081743717193604, + "learning_rate": 0.0006272691618108471, + "loss": 0.4313, + "step": 13421 + }, + { + "epoch": 0.7498952426180965, + "grad_norm": 0.5412591099739075, + "learning_rate": 0.0006272411474675033, + "loss": 0.423, + "step": 13422 + }, + { + "epoch": 0.7499511132217783, + "grad_norm": 1.006557583808899, + "learning_rate": 0.0006272131331241596, + "loss": 0.5939, + "step": 13423 + }, + { + "epoch": 0.7500069838254603, + "grad_norm": 0.7721410989761353, + "learning_rate": 0.0006271851187808157, + "loss": 0.4989, + "step": 13424 + }, + { + "epoch": 0.7500628544291421, + "grad_norm": 0.3267631530761719, + "learning_rate": 0.000627157104437472, + "loss": 0.4154, + "step": 13425 + }, + { + "epoch": 0.750118725032824, + "grad_norm": 0.8137548565864563, + "learning_rate": 0.0006271290900941283, + "loss": 0.6726, + "step": 13426 + }, + { + "epoch": 0.7501745956365059, + "grad_norm": 0.4777452051639557, + "learning_rate": 0.0006271010757507845, + "loss": 0.5162, + "step": 13427 + }, + { + "epoch": 0.7502304662401877, + "grad_norm": 0.48777851462364197, + "learning_rate": 0.0006270730614074406, + "loss": 0.4613, + "step": 13428 + }, + { + "epoch": 0.7502863368438696, + "grad_norm": 0.4694766700267792, + "learning_rate": 0.0006270450470640968, + "loss": 0.3806, + "step": 13429 + }, + { + "epoch": 0.7503422074475514, + "grad_norm": 0.5693162083625793, + "learning_rate": 0.0006270170327207531, + "loss": 0.3966, + "step": 13430 + }, + { + "epoch": 0.7503980780512334, + "grad_norm": 0.4867722988128662, + "learning_rate": 0.0006269890183774093, + "loss": 0.3946, + "step": 13431 + }, + { + "epoch": 0.7504539486549152, + "grad_norm": 0.5500354170799255, + "learning_rate": 0.0006269610040340654, + "loss": 0.4906, + "step": 13432 + }, + { + "epoch": 0.750509819258597, + "grad_norm": 0.42842790484428406, + "learning_rate": 0.0006269329896907217, + "loss": 0.4755, + "step": 13433 + }, + { + "epoch": 0.750565689862279, + "grad_norm": 0.6911308169364929, + "learning_rate": 0.0006269049753473779, + "loss": 0.5383, + "step": 13434 + }, + { + "epoch": 0.7506215604659608, + "grad_norm": 0.5792227387428284, + "learning_rate": 0.0006268769610040341, + "loss": 0.5016, + "step": 13435 + }, + { + "epoch": 0.7506774310696427, + "grad_norm": 0.35939303040504456, + "learning_rate": 0.0006268489466606902, + "loss": 0.3791, + "step": 13436 + }, + { + "epoch": 0.7507333016733245, + "grad_norm": 0.4780380427837372, + "learning_rate": 0.0006268209323173465, + "loss": 0.4157, + "step": 13437 + }, + { + "epoch": 0.7507891722770065, + "grad_norm": 0.6193230748176575, + "learning_rate": 0.0006267929179740027, + "loss": 0.4487, + "step": 13438 + }, + { + "epoch": 0.7508450428806883, + "grad_norm": 0.4579247236251831, + "learning_rate": 0.0006267649036306588, + "loss": 0.4367, + "step": 13439 + }, + { + "epoch": 0.7509009134843702, + "grad_norm": 0.4677896499633789, + "learning_rate": 0.0006267368892873151, + "loss": 0.3953, + "step": 13440 + }, + { + "epoch": 0.7509567840880521, + "grad_norm": 0.434489369392395, + "learning_rate": 0.0006267088749439713, + "loss": 0.469, + "step": 13441 + }, + { + "epoch": 0.751012654691734, + "grad_norm": 1.343895673751831, + "learning_rate": 0.0006266808606006276, + "loss": 0.4121, + "step": 13442 + }, + { + "epoch": 0.7510685252954158, + "grad_norm": 0.5919949412345886, + "learning_rate": 0.0006266528462572837, + "loss": 0.4224, + "step": 13443 + }, + { + "epoch": 0.7511243958990977, + "grad_norm": 0.5093335509300232, + "learning_rate": 0.00062662483191394, + "loss": 0.371, + "step": 13444 + }, + { + "epoch": 0.7511802665027796, + "grad_norm": 0.4703071415424347, + "learning_rate": 0.0006265968175705962, + "loss": 0.5162, + "step": 13445 + }, + { + "epoch": 0.7512361371064614, + "grad_norm": 0.46090421080589294, + "learning_rate": 0.0006265688032272524, + "loss": 0.4799, + "step": 13446 + }, + { + "epoch": 0.7512920077101433, + "grad_norm": 0.6824076175689697, + "learning_rate": 0.0006265407888839085, + "loss": 0.4024, + "step": 13447 + }, + { + "epoch": 0.7513478783138252, + "grad_norm": 0.524598240852356, + "learning_rate": 0.0006265127745405648, + "loss": 0.4224, + "step": 13448 + }, + { + "epoch": 0.751403748917507, + "grad_norm": 1.497862696647644, + "learning_rate": 0.000626484760197221, + "loss": 0.4456, + "step": 13449 + }, + { + "epoch": 0.7514596195211889, + "grad_norm": 0.4912664294242859, + "learning_rate": 0.0006264567458538772, + "loss": 0.5281, + "step": 13450 + }, + { + "epoch": 0.7515154901248708, + "grad_norm": 0.5028389096260071, + "learning_rate": 0.0006264287315105334, + "loss": 0.3889, + "step": 13451 + }, + { + "epoch": 0.7515713607285527, + "grad_norm": 0.6284744739532471, + "learning_rate": 0.0006264007171671896, + "loss": 0.5811, + "step": 13452 + }, + { + "epoch": 0.7516272313322345, + "grad_norm": 0.654765784740448, + "learning_rate": 0.0006263727028238458, + "loss": 0.4282, + "step": 13453 + }, + { + "epoch": 0.7516831019359164, + "grad_norm": 0.5373620986938477, + "learning_rate": 0.000626344688480502, + "loss": 0.3853, + "step": 13454 + }, + { + "epoch": 0.7517389725395983, + "grad_norm": 0.43407905101776123, + "learning_rate": 0.0006263166741371582, + "loss": 0.3614, + "step": 13455 + }, + { + "epoch": 0.7517948431432802, + "grad_norm": 0.44167953729629517, + "learning_rate": 0.0006262886597938144, + "loss": 0.4227, + "step": 13456 + }, + { + "epoch": 0.751850713746962, + "grad_norm": 0.7024281024932861, + "learning_rate": 0.0006262606454504706, + "loss": 0.5021, + "step": 13457 + }, + { + "epoch": 0.751906584350644, + "grad_norm": 0.3794041872024536, + "learning_rate": 0.000626232631107127, + "loss": 0.4324, + "step": 13458 + }, + { + "epoch": 0.7519624549543258, + "grad_norm": 0.5322843194007874, + "learning_rate": 0.0006262046167637831, + "loss": 0.5449, + "step": 13459 + }, + { + "epoch": 0.7520183255580076, + "grad_norm": 0.4911141097545624, + "learning_rate": 0.0006261766024204393, + "loss": 0.3651, + "step": 13460 + }, + { + "epoch": 0.7520741961616896, + "grad_norm": 1.217954397201538, + "learning_rate": 0.0006261485880770955, + "loss": 0.6434, + "step": 13461 + }, + { + "epoch": 0.7521300667653714, + "grad_norm": 0.6401230692863464, + "learning_rate": 0.0006261205737337517, + "loss": 0.4757, + "step": 13462 + }, + { + "epoch": 0.7521859373690533, + "grad_norm": 0.42391064763069153, + "learning_rate": 0.0006260925593904079, + "loss": 0.4495, + "step": 13463 + }, + { + "epoch": 0.7522418079727351, + "grad_norm": 0.5131352543830872, + "learning_rate": 0.0006260645450470641, + "loss": 0.3837, + "step": 13464 + }, + { + "epoch": 0.752297678576417, + "grad_norm": 0.43195635080337524, + "learning_rate": 0.0006260365307037204, + "loss": 0.5794, + "step": 13465 + }, + { + "epoch": 0.7523535491800989, + "grad_norm": 0.3968721330165863, + "learning_rate": 0.0006260085163603765, + "loss": 0.4451, + "step": 13466 + }, + { + "epoch": 0.7524094197837807, + "grad_norm": 0.4541391432285309, + "learning_rate": 0.0006259805020170327, + "loss": 0.4545, + "step": 13467 + }, + { + "epoch": 0.7524652903874627, + "grad_norm": 4.411153793334961, + "learning_rate": 0.0006259524876736889, + "loss": 0.5016, + "step": 13468 + }, + { + "epoch": 0.7525211609911445, + "grad_norm": 1.0929205417633057, + "learning_rate": 0.0006259244733303452, + "loss": 0.6199, + "step": 13469 + }, + { + "epoch": 0.7525770315948264, + "grad_norm": 0.5058714747428894, + "learning_rate": 0.0006258964589870013, + "loss": 0.3608, + "step": 13470 + }, + { + "epoch": 0.7526329021985082, + "grad_norm": 0.4812338948249817, + "learning_rate": 0.0006258684446436575, + "loss": 0.424, + "step": 13471 + }, + { + "epoch": 0.7526887728021902, + "grad_norm": 0.5262874364852905, + "learning_rate": 0.0006258404303003138, + "loss": 0.3469, + "step": 13472 + }, + { + "epoch": 0.752744643405872, + "grad_norm": 0.43449467420578003, + "learning_rate": 0.00062581241595697, + "loss": 0.4239, + "step": 13473 + }, + { + "epoch": 0.7528005140095538, + "grad_norm": 0.4951923191547394, + "learning_rate": 0.0006257844016136261, + "loss": 0.4379, + "step": 13474 + }, + { + "epoch": 0.7528563846132358, + "grad_norm": 0.4451347887516022, + "learning_rate": 0.0006257563872702823, + "loss": 0.4496, + "step": 13475 + }, + { + "epoch": 0.7529122552169176, + "grad_norm": 0.43187251687049866, + "learning_rate": 0.0006257283729269387, + "loss": 0.4769, + "step": 13476 + }, + { + "epoch": 0.7529681258205995, + "grad_norm": 0.30071544647216797, + "learning_rate": 0.0006257003585835949, + "loss": 0.2762, + "step": 13477 + }, + { + "epoch": 0.7530239964242814, + "grad_norm": 0.3156162202358246, + "learning_rate": 0.000625672344240251, + "loss": 0.322, + "step": 13478 + }, + { + "epoch": 0.7530798670279633, + "grad_norm": 0.5908116698265076, + "learning_rate": 0.0006256443298969073, + "loss": 0.4073, + "step": 13479 + }, + { + "epoch": 0.7531357376316451, + "grad_norm": 0.4463503956794739, + "learning_rate": 0.0006256163155535635, + "loss": 0.4964, + "step": 13480 + }, + { + "epoch": 0.7531916082353269, + "grad_norm": 0.7842648029327393, + "learning_rate": 0.0006255883012102196, + "loss": 0.4133, + "step": 13481 + }, + { + "epoch": 0.7532474788390089, + "grad_norm": 0.3478870093822479, + "learning_rate": 0.0006255602868668758, + "loss": 0.4, + "step": 13482 + }, + { + "epoch": 0.7533033494426907, + "grad_norm": 0.47592613101005554, + "learning_rate": 0.0006255322725235321, + "loss": 0.3997, + "step": 13483 + }, + { + "epoch": 0.7533592200463726, + "grad_norm": 0.34629398584365845, + "learning_rate": 0.0006255042581801883, + "loss": 0.3342, + "step": 13484 + }, + { + "epoch": 0.7534150906500545, + "grad_norm": 0.45139265060424805, + "learning_rate": 0.0006254762438368444, + "loss": 0.4134, + "step": 13485 + }, + { + "epoch": 0.7534709612537364, + "grad_norm": 0.6952793598175049, + "learning_rate": 0.0006254482294935006, + "loss": 0.551, + "step": 13486 + }, + { + "epoch": 0.7535268318574182, + "grad_norm": 0.6015214920043945, + "learning_rate": 0.0006254202151501569, + "loss": 0.456, + "step": 13487 + }, + { + "epoch": 0.7535827024611, + "grad_norm": 0.5027416348457336, + "learning_rate": 0.0006253922008068131, + "loss": 0.5039, + "step": 13488 + }, + { + "epoch": 0.753638573064782, + "grad_norm": 0.44052600860595703, + "learning_rate": 0.0006253641864634692, + "loss": 0.4374, + "step": 13489 + }, + { + "epoch": 0.7536944436684638, + "grad_norm": 0.512904942035675, + "learning_rate": 0.0006253361721201255, + "loss": 0.4982, + "step": 13490 + }, + { + "epoch": 0.7537503142721457, + "grad_norm": 1.4416441917419434, + "learning_rate": 0.0006253081577767817, + "loss": 0.4581, + "step": 13491 + }, + { + "epoch": 0.7538061848758276, + "grad_norm": 0.5275624990463257, + "learning_rate": 0.000625280143433438, + "loss": 0.4413, + "step": 13492 + }, + { + "epoch": 0.7538620554795095, + "grad_norm": 4.719262599945068, + "learning_rate": 0.000625252129090094, + "loss": 0.5623, + "step": 13493 + }, + { + "epoch": 0.7539179260831913, + "grad_norm": 0.4179215133190155, + "learning_rate": 0.0006252241147467504, + "loss": 0.4171, + "step": 13494 + }, + { + "epoch": 0.7539737966868733, + "grad_norm": 0.5142239928245544, + "learning_rate": 0.0006251961004034066, + "loss": 0.4578, + "step": 13495 + }, + { + "epoch": 0.7540296672905551, + "grad_norm": 1.257567048072815, + "learning_rate": 0.0006251680860600628, + "loss": 0.4663, + "step": 13496 + }, + { + "epoch": 0.7540855378942369, + "grad_norm": 0.9452446699142456, + "learning_rate": 0.000625140071716719, + "loss": 0.6378, + "step": 13497 + }, + { + "epoch": 0.7541414084979188, + "grad_norm": 0.5383964776992798, + "learning_rate": 0.0006251120573733752, + "loss": 0.3984, + "step": 13498 + }, + { + "epoch": 0.7541972791016007, + "grad_norm": 0.3810681700706482, + "learning_rate": 0.0006250840430300314, + "loss": 0.3908, + "step": 13499 + }, + { + "epoch": 0.7542531497052826, + "grad_norm": 0.4446561336517334, + "learning_rate": 0.0006250560286866876, + "loss": 0.3915, + "step": 13500 + }, + { + "epoch": 0.7542531497052826, + "eval_cer": 0.08915692885121054, + "eval_loss": 0.33570900559425354, + "eval_runtime": 56.2781, + "eval_samples_per_second": 80.635, + "eval_steps_per_second": 5.046, + "eval_wer": 0.35381693248791934, + "step": 13500 + }, + { + "epoch": 0.7543090203089644, + "grad_norm": 0.9660778641700745, + "learning_rate": 0.0006250280143433438, + "loss": 0.5589, + "step": 13501 + }, + { + "epoch": 0.7543648909126464, + "grad_norm": 0.399791955947876, + "learning_rate": 0.000625, + "loss": 0.3778, + "step": 13502 + }, + { + "epoch": 0.7544207615163282, + "grad_norm": 0.9530764222145081, + "learning_rate": 0.0006249719856566562, + "loss": 0.2983, + "step": 13503 + }, + { + "epoch": 0.75447663212001, + "grad_norm": 0.45742085576057434, + "learning_rate": 0.0006249439713133124, + "loss": 0.4323, + "step": 13504 + }, + { + "epoch": 0.7545325027236919, + "grad_norm": 0.3384399116039276, + "learning_rate": 0.0006249159569699686, + "loss": 0.3697, + "step": 13505 + }, + { + "epoch": 0.7545883733273738, + "grad_norm": 0.4627189040184021, + "learning_rate": 0.0006248879426266248, + "loss": 0.4667, + "step": 13506 + }, + { + "epoch": 0.7546442439310557, + "grad_norm": 0.500180184841156, + "learning_rate": 0.000624859928283281, + "loss": 0.4536, + "step": 13507 + }, + { + "epoch": 0.7547001145347375, + "grad_norm": 0.40457531809806824, + "learning_rate": 0.0006248319139399372, + "loss": 0.4204, + "step": 13508 + }, + { + "epoch": 0.7547559851384195, + "grad_norm": 0.5841420888900757, + "learning_rate": 0.0006248038995965935, + "loss": 0.5126, + "step": 13509 + }, + { + "epoch": 0.7548118557421013, + "grad_norm": 1.9289007186889648, + "learning_rate": 0.0006247758852532497, + "loss": 0.3629, + "step": 13510 + }, + { + "epoch": 0.7548677263457831, + "grad_norm": 0.5593510270118713, + "learning_rate": 0.000624747870909906, + "loss": 0.4688, + "step": 13511 + }, + { + "epoch": 0.7549235969494651, + "grad_norm": 0.3029554486274719, + "learning_rate": 0.0006247198565665621, + "loss": 0.3578, + "step": 13512 + }, + { + "epoch": 0.7549794675531469, + "grad_norm": 0.5162441730499268, + "learning_rate": 0.0006246918422232183, + "loss": 0.4774, + "step": 13513 + }, + { + "epoch": 0.7550353381568288, + "grad_norm": 0.4092152416706085, + "learning_rate": 0.0006246638278798745, + "loss": 0.3963, + "step": 13514 + }, + { + "epoch": 0.7550912087605106, + "grad_norm": 0.46890953183174133, + "learning_rate": 0.0006246358135365308, + "loss": 0.5044, + "step": 13515 + }, + { + "epoch": 0.7551470793641926, + "grad_norm": 0.5315017700195312, + "learning_rate": 0.0006246077991931869, + "loss": 0.4797, + "step": 13516 + }, + { + "epoch": 0.7552029499678744, + "grad_norm": 0.34394362568855286, + "learning_rate": 0.0006245797848498431, + "loss": 0.4742, + "step": 13517 + }, + { + "epoch": 0.7552588205715562, + "grad_norm": 0.4420415461063385, + "learning_rate": 0.0006245517705064994, + "loss": 0.4943, + "step": 13518 + }, + { + "epoch": 0.7553146911752382, + "grad_norm": 0.403260201215744, + "learning_rate": 0.0006245237561631556, + "loss": 0.3957, + "step": 13519 + }, + { + "epoch": 0.75537056177892, + "grad_norm": 3.412578821182251, + "learning_rate": 0.0006244957418198117, + "loss": 0.5629, + "step": 13520 + }, + { + "epoch": 0.7554264323826019, + "grad_norm": 0.48920756578445435, + "learning_rate": 0.0006244677274764679, + "loss": 0.5269, + "step": 13521 + }, + { + "epoch": 0.7554823029862837, + "grad_norm": 0.5408328771591187, + "learning_rate": 0.0006244397131331242, + "loss": 0.4228, + "step": 13522 + }, + { + "epoch": 0.7555381735899657, + "grad_norm": 0.6103297472000122, + "learning_rate": 0.0006244116987897804, + "loss": 0.3924, + "step": 13523 + }, + { + "epoch": 0.7555940441936475, + "grad_norm": 0.4869939386844635, + "learning_rate": 0.0006243836844464365, + "loss": 0.3948, + "step": 13524 + }, + { + "epoch": 0.7556499147973293, + "grad_norm": 1.5316511392593384, + "learning_rate": 0.0006243556701030927, + "loss": 0.7777, + "step": 13525 + }, + { + "epoch": 0.7557057854010113, + "grad_norm": 0.5362390875816345, + "learning_rate": 0.0006243276557597491, + "loss": 0.406, + "step": 13526 + }, + { + "epoch": 0.7557616560046931, + "grad_norm": 0.9890533685684204, + "learning_rate": 0.0006242996414164052, + "loss": 0.3891, + "step": 13527 + }, + { + "epoch": 0.755817526608375, + "grad_norm": 0.6059514284133911, + "learning_rate": 0.0006242716270730614, + "loss": 0.462, + "step": 13528 + }, + { + "epoch": 0.7558733972120569, + "grad_norm": 0.3723866641521454, + "learning_rate": 0.0006242436127297177, + "loss": 0.4474, + "step": 13529 + }, + { + "epoch": 0.7559292678157388, + "grad_norm": 4.78813362121582, + "learning_rate": 0.0006242155983863739, + "loss": 0.4578, + "step": 13530 + }, + { + "epoch": 0.7559851384194206, + "grad_norm": 0.45488429069519043, + "learning_rate": 0.00062418758404303, + "loss": 0.3555, + "step": 13531 + }, + { + "epoch": 0.7560410090231025, + "grad_norm": 0.6911581754684448, + "learning_rate": 0.0006241595696996862, + "loss": 0.7032, + "step": 13532 + }, + { + "epoch": 0.7560968796267844, + "grad_norm": 1.620263695716858, + "learning_rate": 0.0006241315553563425, + "loss": 0.4359, + "step": 13533 + }, + { + "epoch": 0.7561527502304662, + "grad_norm": 0.45222049951553345, + "learning_rate": 0.0006241035410129987, + "loss": 0.4342, + "step": 13534 + }, + { + "epoch": 0.7562086208341481, + "grad_norm": 0.7071042060852051, + "learning_rate": 0.0006240755266696548, + "loss": 0.4985, + "step": 13535 + }, + { + "epoch": 0.75626449143783, + "grad_norm": 0.40437260270118713, + "learning_rate": 0.0006240475123263111, + "loss": 0.3886, + "step": 13536 + }, + { + "epoch": 0.7563203620415119, + "grad_norm": 1.0192621946334839, + "learning_rate": 0.0006240194979829673, + "loss": 0.4884, + "step": 13537 + }, + { + "epoch": 0.7563762326451937, + "grad_norm": 0.5238025188446045, + "learning_rate": 0.0006239914836396235, + "loss": 0.5132, + "step": 13538 + }, + { + "epoch": 0.7564321032488756, + "grad_norm": 0.4588794410228729, + "learning_rate": 0.0006239634692962796, + "loss": 0.3624, + "step": 13539 + }, + { + "epoch": 0.7564879738525575, + "grad_norm": 0.44645044207572937, + "learning_rate": 0.0006239354549529359, + "loss": 0.4909, + "step": 13540 + }, + { + "epoch": 0.7565438444562393, + "grad_norm": 0.7891680002212524, + "learning_rate": 0.0006239074406095921, + "loss": 0.358, + "step": 13541 + }, + { + "epoch": 0.7565997150599212, + "grad_norm": 0.5396749377250671, + "learning_rate": 0.0006238794262662484, + "loss": 0.5852, + "step": 13542 + }, + { + "epoch": 0.7566555856636031, + "grad_norm": 0.5277355909347534, + "learning_rate": 0.0006238514119229046, + "loss": 0.5163, + "step": 13543 + }, + { + "epoch": 0.756711456267285, + "grad_norm": 4.61224365234375, + "learning_rate": 0.0006238233975795608, + "loss": 0.3763, + "step": 13544 + }, + { + "epoch": 0.7567673268709668, + "grad_norm": 0.4026711881160736, + "learning_rate": 0.000623795383236217, + "loss": 0.3148, + "step": 13545 + }, + { + "epoch": 0.7568231974746488, + "grad_norm": 0.4206242859363556, + "learning_rate": 0.0006237673688928731, + "loss": 0.4823, + "step": 13546 + }, + { + "epoch": 0.7568790680783306, + "grad_norm": 0.5849800705909729, + "learning_rate": 0.0006237393545495294, + "loss": 0.4691, + "step": 13547 + }, + { + "epoch": 0.7569349386820124, + "grad_norm": 0.43612366914749146, + "learning_rate": 0.0006237113402061856, + "loss": 0.4185, + "step": 13548 + }, + { + "epoch": 0.7569908092856943, + "grad_norm": 0.41645675897598267, + "learning_rate": 0.0006236833258628418, + "loss": 0.5336, + "step": 13549 + }, + { + "epoch": 0.7570466798893762, + "grad_norm": 0.5326940417289734, + "learning_rate": 0.000623655311519498, + "loss": 0.3088, + "step": 13550 + }, + { + "epoch": 0.7571025504930581, + "grad_norm": 0.4570246934890747, + "learning_rate": 0.0006236272971761542, + "loss": 0.5128, + "step": 13551 + }, + { + "epoch": 0.7571584210967399, + "grad_norm": 0.48519167304039, + "learning_rate": 0.0006235992828328104, + "loss": 0.4496, + "step": 13552 + }, + { + "epoch": 0.7572142917004219, + "grad_norm": 0.5578323006629944, + "learning_rate": 0.0006235712684894666, + "loss": 0.5393, + "step": 13553 + }, + { + "epoch": 0.7572701623041037, + "grad_norm": 0.5963954329490662, + "learning_rate": 0.0006235432541461228, + "loss": 0.3414, + "step": 13554 + }, + { + "epoch": 0.7573260329077856, + "grad_norm": 1.4339220523834229, + "learning_rate": 0.000623515239802779, + "loss": 0.7136, + "step": 13555 + }, + { + "epoch": 0.7573819035114674, + "grad_norm": 0.33755961060523987, + "learning_rate": 0.0006234872254594352, + "loss": 0.3765, + "step": 13556 + }, + { + "epoch": 0.7574377741151493, + "grad_norm": 0.6429595351219177, + "learning_rate": 0.0006234592111160915, + "loss": 0.4039, + "step": 13557 + }, + { + "epoch": 0.7574936447188312, + "grad_norm": 0.40535250306129456, + "learning_rate": 0.0006234311967727476, + "loss": 0.4205, + "step": 13558 + }, + { + "epoch": 0.757549515322513, + "grad_norm": 0.3470677137374878, + "learning_rate": 0.0006234031824294039, + "loss": 0.4472, + "step": 13559 + }, + { + "epoch": 0.757605385926195, + "grad_norm": 0.5131312608718872, + "learning_rate": 0.0006233751680860601, + "loss": 0.4263, + "step": 13560 + }, + { + "epoch": 0.7576612565298768, + "grad_norm": 0.8655538558959961, + "learning_rate": 0.0006233471537427164, + "loss": 0.5543, + "step": 13561 + }, + { + "epoch": 0.7577171271335587, + "grad_norm": 0.3741232752799988, + "learning_rate": 0.0006233191393993725, + "loss": 0.4706, + "step": 13562 + }, + { + "epoch": 0.7577729977372406, + "grad_norm": 6.083619594573975, + "learning_rate": 0.0006232911250560287, + "loss": 0.4668, + "step": 13563 + }, + { + "epoch": 0.7578288683409224, + "grad_norm": 0.5429067611694336, + "learning_rate": 0.0006232631107126849, + "loss": 0.4425, + "step": 13564 + }, + { + "epoch": 0.7578847389446043, + "grad_norm": 0.6934283375740051, + "learning_rate": 0.0006232350963693412, + "loss": 0.4836, + "step": 13565 + }, + { + "epoch": 0.7579406095482861, + "grad_norm": 0.5906968116760254, + "learning_rate": 0.0006232070820259973, + "loss": 0.6146, + "step": 13566 + }, + { + "epoch": 0.7579964801519681, + "grad_norm": 0.4569457769393921, + "learning_rate": 0.0006231790676826535, + "loss": 0.7377, + "step": 13567 + }, + { + "epoch": 0.7580523507556499, + "grad_norm": 0.4023842513561249, + "learning_rate": 0.0006231510533393098, + "loss": 0.456, + "step": 13568 + }, + { + "epoch": 0.7581082213593318, + "grad_norm": 0.5379443168640137, + "learning_rate": 0.0006231230389959659, + "loss": 0.4229, + "step": 13569 + }, + { + "epoch": 0.7581640919630137, + "grad_norm": 0.5712339878082275, + "learning_rate": 0.0006230950246526221, + "loss": 0.4796, + "step": 13570 + }, + { + "epoch": 0.7582199625666955, + "grad_norm": 0.47981202602386475, + "learning_rate": 0.0006230670103092783, + "loss": 0.4645, + "step": 13571 + }, + { + "epoch": 0.7582758331703774, + "grad_norm": 0.4389262795448303, + "learning_rate": 0.0006230389959659346, + "loss": 0.4065, + "step": 13572 + }, + { + "epoch": 0.7583317037740592, + "grad_norm": 1.5648125410079956, + "learning_rate": 0.0006230109816225907, + "loss": 0.3489, + "step": 13573 + }, + { + "epoch": 0.7583875743777412, + "grad_norm": 0.6157087683677673, + "learning_rate": 0.0006229829672792469, + "loss": 0.4199, + "step": 13574 + }, + { + "epoch": 0.758443444981423, + "grad_norm": 0.6159941554069519, + "learning_rate": 0.0006229549529359033, + "loss": 0.4748, + "step": 13575 + }, + { + "epoch": 0.7584993155851049, + "grad_norm": 0.5593623518943787, + "learning_rate": 0.0006229269385925595, + "loss": 0.4719, + "step": 13576 + }, + { + "epoch": 0.7585551861887868, + "grad_norm": 0.5208501219749451, + "learning_rate": 0.0006228989242492156, + "loss": 0.5206, + "step": 13577 + }, + { + "epoch": 0.7586110567924687, + "grad_norm": 0.5383267998695374, + "learning_rate": 0.0006228709099058718, + "loss": 0.5777, + "step": 13578 + }, + { + "epoch": 0.7586669273961505, + "grad_norm": 0.5637678503990173, + "learning_rate": 0.0006228428955625281, + "loss": 0.4576, + "step": 13579 + }, + { + "epoch": 0.7587227979998324, + "grad_norm": 0.6789091229438782, + "learning_rate": 0.0006228148812191843, + "loss": 0.4385, + "step": 13580 + }, + { + "epoch": 0.7587786686035143, + "grad_norm": 0.9839410185813904, + "learning_rate": 0.0006227868668758404, + "loss": 0.4234, + "step": 13581 + }, + { + "epoch": 0.7588345392071961, + "grad_norm": 0.6336453557014465, + "learning_rate": 0.0006227588525324967, + "loss": 0.5259, + "step": 13582 + }, + { + "epoch": 0.758890409810878, + "grad_norm": 0.4440670311450958, + "learning_rate": 0.0006227308381891529, + "loss": 0.3566, + "step": 13583 + }, + { + "epoch": 0.7589462804145599, + "grad_norm": 0.5145084857940674, + "learning_rate": 0.0006227028238458091, + "loss": 0.4579, + "step": 13584 + }, + { + "epoch": 0.7590021510182418, + "grad_norm": 0.4438595473766327, + "learning_rate": 0.0006226748095024652, + "loss": 0.518, + "step": 13585 + }, + { + "epoch": 0.7590580216219236, + "grad_norm": 0.6171905398368835, + "learning_rate": 0.0006226467951591215, + "loss": 0.5327, + "step": 13586 + }, + { + "epoch": 0.7591138922256055, + "grad_norm": 1.9182665348052979, + "learning_rate": 0.0006226187808157777, + "loss": 0.3977, + "step": 13587 + }, + { + "epoch": 0.7591697628292874, + "grad_norm": 0.9055777788162231, + "learning_rate": 0.0006225907664724338, + "loss": 0.4982, + "step": 13588 + }, + { + "epoch": 0.7592256334329692, + "grad_norm": 0.5947664380073547, + "learning_rate": 0.0006225627521290901, + "loss": 0.5377, + "step": 13589 + }, + { + "epoch": 0.7592815040366511, + "grad_norm": 0.4751300811767578, + "learning_rate": 0.0006225347377857463, + "loss": 0.4134, + "step": 13590 + }, + { + "epoch": 0.759337374640333, + "grad_norm": 0.4518619477748871, + "learning_rate": 0.0006225067234424025, + "loss": 0.4448, + "step": 13591 + }, + { + "epoch": 0.7593932452440149, + "grad_norm": 5.75033712387085, + "learning_rate": 0.0006224787090990586, + "loss": 0.4007, + "step": 13592 + }, + { + "epoch": 0.7594491158476967, + "grad_norm": 0.5849319100379944, + "learning_rate": 0.000622450694755715, + "loss": 0.3771, + "step": 13593 + }, + { + "epoch": 0.7595049864513786, + "grad_norm": 0.5550528764724731, + "learning_rate": 0.0006224226804123712, + "loss": 0.3961, + "step": 13594 + }, + { + "epoch": 0.7595608570550605, + "grad_norm": 0.6857936382293701, + "learning_rate": 0.0006223946660690274, + "loss": 0.5512, + "step": 13595 + }, + { + "epoch": 0.7596167276587423, + "grad_norm": 0.48756536841392517, + "learning_rate": 0.0006223666517256835, + "loss": 0.4372, + "step": 13596 + }, + { + "epoch": 0.7596725982624242, + "grad_norm": 0.3595375716686249, + "learning_rate": 0.0006223386373823398, + "loss": 0.4078, + "step": 13597 + }, + { + "epoch": 0.7597284688661061, + "grad_norm": 0.4951455891132355, + "learning_rate": 0.000622310623038996, + "loss": 0.3466, + "step": 13598 + }, + { + "epoch": 0.759784339469788, + "grad_norm": 0.43644043803215027, + "learning_rate": 0.0006222826086956522, + "loss": 0.4506, + "step": 13599 + }, + { + "epoch": 0.7598402100734698, + "grad_norm": 0.6850645542144775, + "learning_rate": 0.0006222545943523084, + "loss": 0.414, + "step": 13600 + }, + { + "epoch": 0.7598960806771518, + "grad_norm": 0.45781657099723816, + "learning_rate": 0.0006222265800089646, + "loss": 0.4722, + "step": 13601 + }, + { + "epoch": 0.7599519512808336, + "grad_norm": 0.4359513819217682, + "learning_rate": 0.0006221985656656208, + "loss": 0.4056, + "step": 13602 + }, + { + "epoch": 0.7600078218845154, + "grad_norm": 0.36981043219566345, + "learning_rate": 0.000622170551322277, + "loss": 0.4128, + "step": 13603 + }, + { + "epoch": 0.7600636924881974, + "grad_norm": 0.40673384070396423, + "learning_rate": 0.0006221425369789332, + "loss": 0.3911, + "step": 13604 + }, + { + "epoch": 0.7601195630918792, + "grad_norm": 0.6699557304382324, + "learning_rate": 0.0006221145226355894, + "loss": 0.4572, + "step": 13605 + }, + { + "epoch": 0.7601754336955611, + "grad_norm": 0.4153314530849457, + "learning_rate": 0.0006220865082922456, + "loss": 0.4361, + "step": 13606 + }, + { + "epoch": 0.7602313042992429, + "grad_norm": 0.42069968581199646, + "learning_rate": 0.000622058493948902, + "loss": 0.3977, + "step": 13607 + }, + { + "epoch": 0.7602871749029249, + "grad_norm": 0.45540615916252136, + "learning_rate": 0.000622030479605558, + "loss": 0.4932, + "step": 13608 + }, + { + "epoch": 0.7603430455066067, + "grad_norm": 0.7538439035415649, + "learning_rate": 0.0006220024652622143, + "loss": 0.5588, + "step": 13609 + }, + { + "epoch": 0.7603989161102885, + "grad_norm": 0.6010199785232544, + "learning_rate": 0.0006219744509188705, + "loss": 0.4387, + "step": 13610 + }, + { + "epoch": 0.7604547867139705, + "grad_norm": 0.45126956701278687, + "learning_rate": 0.0006219464365755267, + "loss": 0.3665, + "step": 13611 + }, + { + "epoch": 0.7605106573176523, + "grad_norm": 0.4069528877735138, + "learning_rate": 0.0006219184222321829, + "loss": 0.4736, + "step": 13612 + }, + { + "epoch": 0.7605665279213342, + "grad_norm": 0.6317915320396423, + "learning_rate": 0.0006218904078888391, + "loss": 0.4847, + "step": 13613 + }, + { + "epoch": 0.760622398525016, + "grad_norm": 0.3173142671585083, + "learning_rate": 0.0006218623935454954, + "loss": 0.3987, + "step": 13614 + }, + { + "epoch": 0.760678269128698, + "grad_norm": 0.6854267120361328, + "learning_rate": 0.0006218343792021515, + "loss": 0.4178, + "step": 13615 + }, + { + "epoch": 0.7607341397323798, + "grad_norm": 0.4075023829936981, + "learning_rate": 0.0006218063648588077, + "loss": 0.4321, + "step": 13616 + }, + { + "epoch": 0.7607900103360616, + "grad_norm": 0.44793838262557983, + "learning_rate": 0.0006217783505154639, + "loss": 0.3154, + "step": 13617 + }, + { + "epoch": 0.7608458809397436, + "grad_norm": 6.839022636413574, + "learning_rate": 0.0006217503361721202, + "loss": 0.4265, + "step": 13618 + }, + { + "epoch": 0.7609017515434254, + "grad_norm": 0.39629340171813965, + "learning_rate": 0.0006217223218287763, + "loss": 0.4962, + "step": 13619 + }, + { + "epoch": 0.7609576221471073, + "grad_norm": 0.4590931236743927, + "learning_rate": 0.0006216943074854325, + "loss": 0.4426, + "step": 13620 + }, + { + "epoch": 0.7610134927507892, + "grad_norm": 0.3694497048854828, + "learning_rate": 0.0006216662931420888, + "loss": 0.4236, + "step": 13621 + }, + { + "epoch": 0.7610693633544711, + "grad_norm": 0.3411204218864441, + "learning_rate": 0.000621638278798745, + "loss": 0.4056, + "step": 13622 + }, + { + "epoch": 0.7611252339581529, + "grad_norm": 0.48798611760139465, + "learning_rate": 0.0006216102644554011, + "loss": 0.5079, + "step": 13623 + }, + { + "epoch": 0.7611811045618347, + "grad_norm": 0.4604344367980957, + "learning_rate": 0.0006215822501120573, + "loss": 0.3698, + "step": 13624 + }, + { + "epoch": 0.7612369751655167, + "grad_norm": 0.5628504157066345, + "learning_rate": 0.0006215542357687137, + "loss": 0.6382, + "step": 13625 + }, + { + "epoch": 0.7612928457691985, + "grad_norm": 0.4186537563800812, + "learning_rate": 0.0006215262214253699, + "loss": 0.445, + "step": 13626 + }, + { + "epoch": 0.7613487163728804, + "grad_norm": 0.706211268901825, + "learning_rate": 0.000621498207082026, + "loss": 0.6657, + "step": 13627 + }, + { + "epoch": 0.7614045869765623, + "grad_norm": 0.5199497938156128, + "learning_rate": 0.0006214701927386822, + "loss": 0.452, + "step": 13628 + }, + { + "epoch": 0.7614604575802442, + "grad_norm": 0.44888395071029663, + "learning_rate": 0.0006214421783953385, + "loss": 0.4452, + "step": 13629 + }, + { + "epoch": 0.761516328183926, + "grad_norm": 0.5468723177909851, + "learning_rate": 0.0006214141640519946, + "loss": 0.435, + "step": 13630 + }, + { + "epoch": 0.7615721987876078, + "grad_norm": 0.4244318902492523, + "learning_rate": 0.0006213861497086508, + "loss": 0.3951, + "step": 13631 + }, + { + "epoch": 0.7616280693912898, + "grad_norm": 0.6280615329742432, + "learning_rate": 0.0006213581353653071, + "loss": 0.4412, + "step": 13632 + }, + { + "epoch": 0.7616839399949716, + "grad_norm": 0.6840070486068726, + "learning_rate": 0.0006213301210219633, + "loss": 0.4092, + "step": 13633 + }, + { + "epoch": 0.7617398105986535, + "grad_norm": 2.976569652557373, + "learning_rate": 0.0006213021066786194, + "loss": 0.4888, + "step": 13634 + }, + { + "epoch": 0.7617956812023354, + "grad_norm": 0.49249324202537537, + "learning_rate": 0.0006212740923352756, + "loss": 0.5313, + "step": 13635 + }, + { + "epoch": 0.7618515518060173, + "grad_norm": 0.4598885178565979, + "learning_rate": 0.0006212460779919319, + "loss": 0.3691, + "step": 13636 + }, + { + "epoch": 0.7619074224096991, + "grad_norm": 0.48066794872283936, + "learning_rate": 0.0006212180636485881, + "loss": 0.4811, + "step": 13637 + }, + { + "epoch": 0.7619632930133811, + "grad_norm": 0.5236368775367737, + "learning_rate": 0.0006211900493052442, + "loss": 0.3693, + "step": 13638 + }, + { + "epoch": 0.7620191636170629, + "grad_norm": 0.541206955909729, + "learning_rate": 0.0006211620349619005, + "loss": 0.4417, + "step": 13639 + }, + { + "epoch": 0.7620750342207447, + "grad_norm": 1.1403980255126953, + "learning_rate": 0.0006211340206185567, + "loss": 0.4813, + "step": 13640 + }, + { + "epoch": 0.7621309048244266, + "grad_norm": 0.9293106198310852, + "learning_rate": 0.000621106006275213, + "loss": 0.6384, + "step": 13641 + }, + { + "epoch": 0.7621867754281085, + "grad_norm": 0.6660730838775635, + "learning_rate": 0.000621077991931869, + "loss": 0.5436, + "step": 13642 + }, + { + "epoch": 0.7622426460317904, + "grad_norm": 0.45282185077667236, + "learning_rate": 0.0006210499775885254, + "loss": 0.5956, + "step": 13643 + }, + { + "epoch": 0.7622985166354722, + "grad_norm": 2.647803783416748, + "learning_rate": 0.0006210219632451816, + "loss": 0.4473, + "step": 13644 + }, + { + "epoch": 0.7623543872391542, + "grad_norm": 0.42648452520370483, + "learning_rate": 0.0006209939489018378, + "loss": 0.4169, + "step": 13645 + }, + { + "epoch": 0.762410257842836, + "grad_norm": 0.3613019287586212, + "learning_rate": 0.000620965934558494, + "loss": 0.4701, + "step": 13646 + }, + { + "epoch": 0.7624661284465178, + "grad_norm": 0.6015993356704712, + "learning_rate": 0.0006209379202151502, + "loss": 0.5043, + "step": 13647 + }, + { + "epoch": 0.7625219990501997, + "grad_norm": 0.5902588963508606, + "learning_rate": 0.0006209099058718064, + "loss": 0.54, + "step": 13648 + }, + { + "epoch": 0.7625778696538816, + "grad_norm": 0.4790465235710144, + "learning_rate": 0.0006208818915284626, + "loss": 0.4436, + "step": 13649 + }, + { + "epoch": 0.7626337402575635, + "grad_norm": 0.43290042877197266, + "learning_rate": 0.0006208538771851188, + "loss": 0.4784, + "step": 13650 + }, + { + "epoch": 0.7626896108612453, + "grad_norm": 0.4316630959510803, + "learning_rate": 0.000620825862841775, + "loss": 0.3588, + "step": 13651 + }, + { + "epoch": 0.7627454814649273, + "grad_norm": 0.5512781143188477, + "learning_rate": 0.0006207978484984312, + "loss": 0.4688, + "step": 13652 + }, + { + "epoch": 0.7628013520686091, + "grad_norm": 0.43146857619285583, + "learning_rate": 0.0006207698341550874, + "loss": 0.5119, + "step": 13653 + }, + { + "epoch": 0.762857222672291, + "grad_norm": 0.8228240013122559, + "learning_rate": 0.0006207418198117436, + "loss": 0.4511, + "step": 13654 + }, + { + "epoch": 0.7629130932759729, + "grad_norm": 1.1140505075454712, + "learning_rate": 0.0006207138054683998, + "loss": 0.354, + "step": 13655 + }, + { + "epoch": 0.7629689638796547, + "grad_norm": 0.7858426570892334, + "learning_rate": 0.000620685791125056, + "loss": 0.5431, + "step": 13656 + }, + { + "epoch": 0.7630248344833366, + "grad_norm": 0.43534258008003235, + "learning_rate": 0.0006206577767817122, + "loss": 0.4361, + "step": 13657 + }, + { + "epoch": 0.7630807050870184, + "grad_norm": 1.3967347145080566, + "learning_rate": 0.0006206297624383684, + "loss": 0.4492, + "step": 13658 + }, + { + "epoch": 0.7631365756907004, + "grad_norm": 0.4733524024486542, + "learning_rate": 0.0006206017480950247, + "loss": 0.5083, + "step": 13659 + }, + { + "epoch": 0.7631924462943822, + "grad_norm": 0.43910321593284607, + "learning_rate": 0.000620573733751681, + "loss": 0.5088, + "step": 13660 + }, + { + "epoch": 0.763248316898064, + "grad_norm": 0.3690272569656372, + "learning_rate": 0.0006205457194083371, + "loss": 0.4489, + "step": 13661 + }, + { + "epoch": 0.763304187501746, + "grad_norm": 0.4265716075897217, + "learning_rate": 0.0006205177050649933, + "loss": 0.3132, + "step": 13662 + }, + { + "epoch": 0.7633600581054278, + "grad_norm": 0.4416729807853699, + "learning_rate": 0.0006204896907216495, + "loss": 0.3542, + "step": 13663 + }, + { + "epoch": 0.7634159287091097, + "grad_norm": 0.4634413421154022, + "learning_rate": 0.0006204616763783058, + "loss": 0.451, + "step": 13664 + }, + { + "epoch": 0.7634717993127915, + "grad_norm": 0.4868599474430084, + "learning_rate": 0.0006204336620349619, + "loss": 0.5384, + "step": 13665 + }, + { + "epoch": 0.7635276699164735, + "grad_norm": 0.5502709150314331, + "learning_rate": 0.0006204056476916181, + "loss": 0.4574, + "step": 13666 + }, + { + "epoch": 0.7635835405201553, + "grad_norm": 0.4444175660610199, + "learning_rate": 0.0006203776333482743, + "loss": 0.5164, + "step": 13667 + }, + { + "epoch": 0.7636394111238372, + "grad_norm": 0.3411906659603119, + "learning_rate": 0.0006203496190049306, + "loss": 0.3728, + "step": 13668 + }, + { + "epoch": 0.7636952817275191, + "grad_norm": 0.5689351558685303, + "learning_rate": 0.0006203216046615867, + "loss": 0.4964, + "step": 13669 + }, + { + "epoch": 0.7637511523312009, + "grad_norm": 0.4233696758747101, + "learning_rate": 0.0006202935903182429, + "loss": 0.4161, + "step": 13670 + }, + { + "epoch": 0.7638070229348828, + "grad_norm": 0.729720950126648, + "learning_rate": 0.0006202655759748992, + "loss": 0.5964, + "step": 13671 + }, + { + "epoch": 0.7638628935385647, + "grad_norm": 0.44464820623397827, + "learning_rate": 0.0006202375616315553, + "loss": 0.3597, + "step": 13672 + }, + { + "epoch": 0.7639187641422466, + "grad_norm": 2.2276554107666016, + "learning_rate": 0.0006202095472882115, + "loss": 0.4112, + "step": 13673 + }, + { + "epoch": 0.7639746347459284, + "grad_norm": 0.6149027943611145, + "learning_rate": 0.0006201815329448677, + "loss": 0.504, + "step": 13674 + }, + { + "epoch": 0.7640305053496103, + "grad_norm": 0.4815657138824463, + "learning_rate": 0.000620153518601524, + "loss": 0.4287, + "step": 13675 + }, + { + "epoch": 0.7640863759532922, + "grad_norm": 0.4623292088508606, + "learning_rate": 0.0006201255042581802, + "loss": 0.4777, + "step": 13676 + }, + { + "epoch": 0.764142246556974, + "grad_norm": 0.46486446261405945, + "learning_rate": 0.0006200974899148364, + "loss": 0.5005, + "step": 13677 + }, + { + "epoch": 0.7641981171606559, + "grad_norm": 0.5729174613952637, + "learning_rate": 0.0006200694755714927, + "loss": 0.3275, + "step": 13678 + }, + { + "epoch": 0.7642539877643378, + "grad_norm": 0.5294298529624939, + "learning_rate": 0.0006200414612281489, + "loss": 0.4477, + "step": 13679 + }, + { + "epoch": 0.7643098583680197, + "grad_norm": 0.4707203209400177, + "learning_rate": 0.000620013446884805, + "loss": 0.4835, + "step": 13680 + }, + { + "epoch": 0.7643657289717015, + "grad_norm": 0.6474797129631042, + "learning_rate": 0.0006199854325414612, + "loss": 0.4537, + "step": 13681 + }, + { + "epoch": 0.7644215995753834, + "grad_norm": 0.451957106590271, + "learning_rate": 0.0006199574181981175, + "loss": 0.3142, + "step": 13682 + }, + { + "epoch": 0.7644774701790653, + "grad_norm": 0.4979381263256073, + "learning_rate": 0.0006199294038547737, + "loss": 0.4294, + "step": 13683 + }, + { + "epoch": 0.7645333407827472, + "grad_norm": 0.5723061561584473, + "learning_rate": 0.0006199013895114298, + "loss": 0.5067, + "step": 13684 + }, + { + "epoch": 0.764589211386429, + "grad_norm": 0.5633653998374939, + "learning_rate": 0.0006198733751680861, + "loss": 0.4734, + "step": 13685 + }, + { + "epoch": 0.7646450819901109, + "grad_norm": 0.5323101282119751, + "learning_rate": 0.0006198453608247423, + "loss": 0.4229, + "step": 13686 + }, + { + "epoch": 0.7647009525937928, + "grad_norm": 1.7737184762954712, + "learning_rate": 0.0006198173464813985, + "loss": 0.4652, + "step": 13687 + }, + { + "epoch": 0.7647568231974746, + "grad_norm": 1.0872939825057983, + "learning_rate": 0.0006197893321380546, + "loss": 0.4188, + "step": 13688 + }, + { + "epoch": 0.7648126938011566, + "grad_norm": 0.46440789103507996, + "learning_rate": 0.0006197613177947109, + "loss": 0.3664, + "step": 13689 + }, + { + "epoch": 0.7648685644048384, + "grad_norm": 0.4306761920452118, + "learning_rate": 0.0006197333034513671, + "loss": 0.3681, + "step": 13690 + }, + { + "epoch": 0.7649244350085203, + "grad_norm": 0.33172252774238586, + "learning_rate": 0.0006197052891080233, + "loss": 0.3101, + "step": 13691 + }, + { + "epoch": 0.7649803056122021, + "grad_norm": 1.1203075647354126, + "learning_rate": 0.0006196772747646796, + "loss": 0.474, + "step": 13692 + }, + { + "epoch": 0.765036176215884, + "grad_norm": 0.8442895412445068, + "learning_rate": 0.0006196492604213358, + "loss": 0.3564, + "step": 13693 + }, + { + "epoch": 0.7650920468195659, + "grad_norm": 0.5466575622558594, + "learning_rate": 0.000619621246077992, + "loss": 0.4258, + "step": 13694 + }, + { + "epoch": 0.7651479174232477, + "grad_norm": 0.43785369396209717, + "learning_rate": 0.0006195932317346481, + "loss": 0.4377, + "step": 13695 + }, + { + "epoch": 0.7652037880269297, + "grad_norm": 0.6755947470664978, + "learning_rate": 0.0006195652173913044, + "loss": 0.3846, + "step": 13696 + }, + { + "epoch": 0.7652596586306115, + "grad_norm": 0.6963968873023987, + "learning_rate": 0.0006195372030479606, + "loss": 0.4751, + "step": 13697 + }, + { + "epoch": 0.7653155292342934, + "grad_norm": 0.6316248178482056, + "learning_rate": 0.0006195091887046168, + "loss": 0.4779, + "step": 13698 + }, + { + "epoch": 0.7653713998379752, + "grad_norm": 0.7907214164733887, + "learning_rate": 0.0006194811743612729, + "loss": 0.4327, + "step": 13699 + }, + { + "epoch": 0.7654272704416571, + "grad_norm": 0.491260826587677, + "learning_rate": 0.0006194531600179292, + "loss": 0.4619, + "step": 13700 + }, + { + "epoch": 0.765483141045339, + "grad_norm": 0.6578371524810791, + "learning_rate": 0.0006194251456745854, + "loss": 0.3859, + "step": 13701 + }, + { + "epoch": 0.7655390116490208, + "grad_norm": 0.44552192091941833, + "learning_rate": 0.0006193971313312416, + "loss": 0.3934, + "step": 13702 + }, + { + "epoch": 0.7655948822527028, + "grad_norm": 0.46700504422187805, + "learning_rate": 0.0006193691169878978, + "loss": 0.4287, + "step": 13703 + }, + { + "epoch": 0.7656507528563846, + "grad_norm": 0.7383966445922852, + "learning_rate": 0.000619341102644554, + "loss": 0.474, + "step": 13704 + }, + { + "epoch": 0.7657066234600665, + "grad_norm": 0.43930739164352417, + "learning_rate": 0.0006193130883012102, + "loss": 0.4782, + "step": 13705 + }, + { + "epoch": 0.7657624940637484, + "grad_norm": 1.292265772819519, + "learning_rate": 0.0006192850739578664, + "loss": 0.4316, + "step": 13706 + }, + { + "epoch": 0.7658183646674303, + "grad_norm": 1.1409741640090942, + "learning_rate": 0.0006192570596145226, + "loss": 0.3834, + "step": 13707 + }, + { + "epoch": 0.7658742352711121, + "grad_norm": 0.48576563596725464, + "learning_rate": 0.0006192290452711788, + "loss": 0.5863, + "step": 13708 + }, + { + "epoch": 0.7659301058747939, + "grad_norm": 0.5722666382789612, + "learning_rate": 0.000619201030927835, + "loss": 0.652, + "step": 13709 + }, + { + "epoch": 0.7659859764784759, + "grad_norm": 0.46896448731422424, + "learning_rate": 0.0006191730165844914, + "loss": 0.3744, + "step": 13710 + }, + { + "epoch": 0.7660418470821577, + "grad_norm": 0.4816475808620453, + "learning_rate": 0.0006191450022411475, + "loss": 0.3279, + "step": 13711 + }, + { + "epoch": 0.7660977176858396, + "grad_norm": 0.42319512367248535, + "learning_rate": 0.0006191169878978037, + "loss": 0.4211, + "step": 13712 + }, + { + "epoch": 0.7661535882895215, + "grad_norm": 0.39940008521080017, + "learning_rate": 0.0006190889735544599, + "loss": 0.3827, + "step": 13713 + }, + { + "epoch": 0.7662094588932034, + "grad_norm": 0.4834328889846802, + "learning_rate": 0.0006190609592111162, + "loss": 0.4432, + "step": 13714 + }, + { + "epoch": 0.7662653294968852, + "grad_norm": 0.4876575171947479, + "learning_rate": 0.0006190329448677723, + "loss": 0.3532, + "step": 13715 + }, + { + "epoch": 0.766321200100567, + "grad_norm": 0.5117378830909729, + "learning_rate": 0.0006190049305244285, + "loss": 0.4989, + "step": 13716 + }, + { + "epoch": 0.766377070704249, + "grad_norm": 0.39643150568008423, + "learning_rate": 0.0006189769161810848, + "loss": 0.3573, + "step": 13717 + }, + { + "epoch": 0.7664329413079308, + "grad_norm": 0.9292975664138794, + "learning_rate": 0.0006189489018377409, + "loss": 0.4227, + "step": 13718 + }, + { + "epoch": 0.7664888119116127, + "grad_norm": 1.5017409324645996, + "learning_rate": 0.0006189208874943971, + "loss": 0.4691, + "step": 13719 + }, + { + "epoch": 0.7665446825152946, + "grad_norm": 1.0191526412963867, + "learning_rate": 0.0006188928731510533, + "loss": 0.5038, + "step": 13720 + }, + { + "epoch": 0.7666005531189765, + "grad_norm": 0.4351777732372284, + "learning_rate": 0.0006188648588077096, + "loss": 0.4153, + "step": 13721 + }, + { + "epoch": 0.7666564237226583, + "grad_norm": 0.39949050545692444, + "learning_rate": 0.0006188368444643657, + "loss": 0.4675, + "step": 13722 + }, + { + "epoch": 0.7667122943263402, + "grad_norm": 1.053642988204956, + "learning_rate": 0.0006188088301210219, + "loss": 0.4796, + "step": 13723 + }, + { + "epoch": 0.7667681649300221, + "grad_norm": 0.4099563956260681, + "learning_rate": 0.0006187808157776782, + "loss": 0.4381, + "step": 13724 + }, + { + "epoch": 0.7668240355337039, + "grad_norm": 0.43815165758132935, + "learning_rate": 0.0006187528014343345, + "loss": 0.4616, + "step": 13725 + }, + { + "epoch": 0.7668799061373858, + "grad_norm": 0.5605183243751526, + "learning_rate": 0.0006187247870909905, + "loss": 0.4209, + "step": 13726 + }, + { + "epoch": 0.7669357767410677, + "grad_norm": 1.0458377599716187, + "learning_rate": 0.0006186967727476468, + "loss": 0.5053, + "step": 13727 + }, + { + "epoch": 0.7669916473447496, + "grad_norm": 0.3832378089427948, + "learning_rate": 0.0006186687584043031, + "loss": 0.4116, + "step": 13728 + }, + { + "epoch": 0.7670475179484314, + "grad_norm": 0.45284709334373474, + "learning_rate": 0.0006186407440609593, + "loss": 0.5206, + "step": 13729 + }, + { + "epoch": 0.7671033885521134, + "grad_norm": 0.40273481607437134, + "learning_rate": 0.0006186127297176154, + "loss": 0.4226, + "step": 13730 + }, + { + "epoch": 0.7671592591557952, + "grad_norm": 0.6553599834442139, + "learning_rate": 0.0006185847153742717, + "loss": 0.47, + "step": 13731 + }, + { + "epoch": 0.767215129759477, + "grad_norm": 0.48325565457344055, + "learning_rate": 0.0006185567010309279, + "loss": 0.5207, + "step": 13732 + }, + { + "epoch": 0.7672710003631589, + "grad_norm": 0.434982568025589, + "learning_rate": 0.0006185286866875841, + "loss": 0.5086, + "step": 13733 + }, + { + "epoch": 0.7673268709668408, + "grad_norm": 0.6789858937263489, + "learning_rate": 0.0006185006723442402, + "loss": 0.4799, + "step": 13734 + }, + { + "epoch": 0.7673827415705227, + "grad_norm": 0.4370322525501251, + "learning_rate": 0.0006184726580008965, + "loss": 0.4947, + "step": 13735 + }, + { + "epoch": 0.7674386121742045, + "grad_norm": 0.5720311999320984, + "learning_rate": 0.0006184446436575527, + "loss": 0.4152, + "step": 13736 + }, + { + "epoch": 0.7674944827778865, + "grad_norm": 2.453423261642456, + "learning_rate": 0.0006184166293142088, + "loss": 0.4455, + "step": 13737 + }, + { + "epoch": 0.7675503533815683, + "grad_norm": 0.4247485101222992, + "learning_rate": 0.000618388614970865, + "loss": 0.459, + "step": 13738 + }, + { + "epoch": 0.7676062239852501, + "grad_norm": 0.35720816254615784, + "learning_rate": 0.0006183606006275213, + "loss": 0.4192, + "step": 13739 + }, + { + "epoch": 0.7676620945889321, + "grad_norm": 2.0336995124816895, + "learning_rate": 0.0006183325862841775, + "loss": 0.3307, + "step": 13740 + }, + { + "epoch": 0.7677179651926139, + "grad_norm": 0.4266883432865143, + "learning_rate": 0.0006183045719408336, + "loss": 0.407, + "step": 13741 + }, + { + "epoch": 0.7677738357962958, + "grad_norm": 0.3382103145122528, + "learning_rate": 0.00061827655759749, + "loss": 0.3558, + "step": 13742 + }, + { + "epoch": 0.7678297063999776, + "grad_norm": 0.46616655588150024, + "learning_rate": 0.0006182485432541462, + "loss": 0.3555, + "step": 13743 + }, + { + "epoch": 0.7678855770036596, + "grad_norm": 0.5501828789710999, + "learning_rate": 0.0006182205289108024, + "loss": 0.6643, + "step": 13744 + }, + { + "epoch": 0.7679414476073414, + "grad_norm": 0.3650079667568207, + "learning_rate": 0.0006181925145674585, + "loss": 0.3801, + "step": 13745 + }, + { + "epoch": 0.7679973182110232, + "grad_norm": 0.4903980493545532, + "learning_rate": 0.0006181645002241148, + "loss": 0.4488, + "step": 13746 + }, + { + "epoch": 0.7680531888147052, + "grad_norm": 0.7259668707847595, + "learning_rate": 0.000618136485880771, + "loss": 0.4404, + "step": 13747 + }, + { + "epoch": 0.768109059418387, + "grad_norm": 2.433959722518921, + "learning_rate": 0.0006181084715374272, + "loss": 0.5846, + "step": 13748 + }, + { + "epoch": 0.7681649300220689, + "grad_norm": 0.37062332034111023, + "learning_rate": 0.0006180804571940834, + "loss": 0.4315, + "step": 13749 + }, + { + "epoch": 0.7682208006257507, + "grad_norm": 0.5414018630981445, + "learning_rate": 0.0006180524428507396, + "loss": 0.4599, + "step": 13750 + }, + { + "epoch": 0.7682766712294327, + "grad_norm": 0.5277739763259888, + "learning_rate": 0.0006180244285073958, + "loss": 0.4886, + "step": 13751 + }, + { + "epoch": 0.7683325418331145, + "grad_norm": 0.6202769875526428, + "learning_rate": 0.000617996414164052, + "loss": 0.5621, + "step": 13752 + }, + { + "epoch": 0.7683884124367963, + "grad_norm": 0.5181243419647217, + "learning_rate": 0.0006179683998207082, + "loss": 0.3403, + "step": 13753 + }, + { + "epoch": 0.7684442830404783, + "grad_norm": 1.3661659955978394, + "learning_rate": 0.0006179403854773644, + "loss": 0.4774, + "step": 13754 + }, + { + "epoch": 0.7685001536441601, + "grad_norm": 0.6714478135108948, + "learning_rate": 0.0006179123711340206, + "loss": 0.5425, + "step": 13755 + }, + { + "epoch": 0.768556024247842, + "grad_norm": 0.7271973490715027, + "learning_rate": 0.0006178843567906769, + "loss": 0.4536, + "step": 13756 + }, + { + "epoch": 0.7686118948515239, + "grad_norm": 0.5053291320800781, + "learning_rate": 0.000617856342447333, + "loss": 0.4621, + "step": 13757 + }, + { + "epoch": 0.7686677654552058, + "grad_norm": 0.4523674547672272, + "learning_rate": 0.0006178283281039892, + "loss": 0.3992, + "step": 13758 + }, + { + "epoch": 0.7687236360588876, + "grad_norm": 0.6782209873199463, + "learning_rate": 0.0006178003137606454, + "loss": 0.4881, + "step": 13759 + }, + { + "epoch": 0.7687795066625694, + "grad_norm": 0.7626388669013977, + "learning_rate": 0.0006177722994173017, + "loss": 0.4253, + "step": 13760 + }, + { + "epoch": 0.7688353772662514, + "grad_norm": 0.569532036781311, + "learning_rate": 0.0006177442850739579, + "loss": 0.5738, + "step": 13761 + }, + { + "epoch": 0.7688912478699332, + "grad_norm": 1.7104371786117554, + "learning_rate": 0.0006177162707306141, + "loss": 0.5932, + "step": 13762 + }, + { + "epoch": 0.7689471184736151, + "grad_norm": 0.4157220423221588, + "learning_rate": 0.0006176882563872704, + "loss": 0.3967, + "step": 13763 + }, + { + "epoch": 0.769002989077297, + "grad_norm": 0.8119778633117676, + "learning_rate": 0.0006176602420439265, + "loss": 0.4394, + "step": 13764 + }, + { + "epoch": 0.7690588596809789, + "grad_norm": 0.4390336871147156, + "learning_rate": 0.0006176322277005827, + "loss": 0.5034, + "step": 13765 + }, + { + "epoch": 0.7691147302846607, + "grad_norm": 0.5242427587509155, + "learning_rate": 0.0006176042133572389, + "loss": 0.5254, + "step": 13766 + }, + { + "epoch": 0.7691706008883425, + "grad_norm": 0.4531094431877136, + "learning_rate": 0.0006175761990138952, + "loss": 0.4871, + "step": 13767 + }, + { + "epoch": 0.7692264714920245, + "grad_norm": 0.4686170220375061, + "learning_rate": 0.0006175481846705513, + "loss": 0.5428, + "step": 13768 + }, + { + "epoch": 0.7692823420957063, + "grad_norm": 0.42379310727119446, + "learning_rate": 0.0006175201703272075, + "loss": 0.4257, + "step": 13769 + }, + { + "epoch": 0.7693382126993882, + "grad_norm": 0.41410496830940247, + "learning_rate": 0.0006174921559838638, + "loss": 0.3691, + "step": 13770 + }, + { + "epoch": 0.7693940833030701, + "grad_norm": 0.409134179353714, + "learning_rate": 0.00061746414164052, + "loss": 0.4651, + "step": 13771 + }, + { + "epoch": 0.769449953906752, + "grad_norm": 0.8311570882797241, + "learning_rate": 0.0006174361272971761, + "loss": 0.4288, + "step": 13772 + }, + { + "epoch": 0.7695058245104338, + "grad_norm": 0.4404568672180176, + "learning_rate": 0.0006174081129538323, + "loss": 0.3975, + "step": 13773 + }, + { + "epoch": 0.7695616951141158, + "grad_norm": 0.48529765009880066, + "learning_rate": 0.0006173800986104886, + "loss": 0.5571, + "step": 13774 + }, + { + "epoch": 0.7696175657177976, + "grad_norm": 0.41991275548934937, + "learning_rate": 0.0006173520842671449, + "loss": 0.4058, + "step": 13775 + }, + { + "epoch": 0.7696734363214794, + "grad_norm": 0.3925504982471466, + "learning_rate": 0.000617324069923801, + "loss": 0.3968, + "step": 13776 + }, + { + "epoch": 0.7697293069251613, + "grad_norm": 0.4103698432445526, + "learning_rate": 0.0006172960555804572, + "loss": 0.541, + "step": 13777 + }, + { + "epoch": 0.7697851775288432, + "grad_norm": 0.366923451423645, + "learning_rate": 0.0006172680412371135, + "loss": 0.3772, + "step": 13778 + }, + { + "epoch": 0.7698410481325251, + "grad_norm": 0.3422240912914276, + "learning_rate": 0.0006172400268937696, + "loss": 0.3559, + "step": 13779 + }, + { + "epoch": 0.7698969187362069, + "grad_norm": 0.5306734442710876, + "learning_rate": 0.0006172120125504258, + "loss": 0.4194, + "step": 13780 + }, + { + "epoch": 0.7699527893398889, + "grad_norm": 0.47817885875701904, + "learning_rate": 0.0006171839982070821, + "loss": 0.4486, + "step": 13781 + }, + { + "epoch": 0.7700086599435707, + "grad_norm": 0.39212536811828613, + "learning_rate": 0.0006171559838637383, + "loss": 0.3605, + "step": 13782 + }, + { + "epoch": 0.7700645305472525, + "grad_norm": 0.9430112838745117, + "learning_rate": 0.0006171279695203944, + "loss": 0.4428, + "step": 13783 + }, + { + "epoch": 0.7701204011509344, + "grad_norm": 0.6084392070770264, + "learning_rate": 0.0006170999551770506, + "loss": 0.4805, + "step": 13784 + }, + { + "epoch": 0.7701762717546163, + "grad_norm": 1.5682194232940674, + "learning_rate": 0.0006170719408337069, + "loss": 0.4222, + "step": 13785 + }, + { + "epoch": 0.7702321423582982, + "grad_norm": 0.37974536418914795, + "learning_rate": 0.0006170439264903631, + "loss": 0.4063, + "step": 13786 + }, + { + "epoch": 0.77028801296198, + "grad_norm": 0.4018388092517853, + "learning_rate": 0.0006170159121470192, + "loss": 0.4381, + "step": 13787 + }, + { + "epoch": 0.770343883565662, + "grad_norm": 0.6101363301277161, + "learning_rate": 0.0006169878978036755, + "loss": 0.3971, + "step": 13788 + }, + { + "epoch": 0.7703997541693438, + "grad_norm": 1.1781911849975586, + "learning_rate": 0.0006169598834603317, + "loss": 0.3731, + "step": 13789 + }, + { + "epoch": 0.7704556247730256, + "grad_norm": 0.8007503747940063, + "learning_rate": 0.0006169318691169879, + "loss": 0.3674, + "step": 13790 + }, + { + "epoch": 0.7705114953767076, + "grad_norm": 0.4086199998855591, + "learning_rate": 0.000616903854773644, + "loss": 0.4183, + "step": 13791 + }, + { + "epoch": 0.7705673659803894, + "grad_norm": 0.40586909651756287, + "learning_rate": 0.0006168758404303003, + "loss": 0.5278, + "step": 13792 + }, + { + "epoch": 0.7706232365840713, + "grad_norm": 0.5846885442733765, + "learning_rate": 0.0006168478260869566, + "loss": 0.4142, + "step": 13793 + }, + { + "epoch": 0.7706791071877531, + "grad_norm": 0.3958839178085327, + "learning_rate": 0.0006168198117436128, + "loss": 0.3424, + "step": 13794 + }, + { + "epoch": 0.7707349777914351, + "grad_norm": 0.5748491287231445, + "learning_rate": 0.000616791797400269, + "loss": 0.383, + "step": 13795 + }, + { + "epoch": 0.7707908483951169, + "grad_norm": 0.7566101551055908, + "learning_rate": 0.0006167637830569252, + "loss": 0.3883, + "step": 13796 + }, + { + "epoch": 0.7708467189987988, + "grad_norm": 0.4107487201690674, + "learning_rate": 0.0006167357687135814, + "loss": 0.3582, + "step": 13797 + }, + { + "epoch": 0.7709025896024807, + "grad_norm": 0.524856686592102, + "learning_rate": 0.0006167077543702376, + "loss": 0.4739, + "step": 13798 + }, + { + "epoch": 0.7709584602061625, + "grad_norm": 0.45643529295921326, + "learning_rate": 0.0006166797400268938, + "loss": 0.339, + "step": 13799 + }, + { + "epoch": 0.7710143308098444, + "grad_norm": 0.5885153412818909, + "learning_rate": 0.00061665172568355, + "loss": 0.3565, + "step": 13800 + }, + { + "epoch": 0.7710702014135262, + "grad_norm": 0.7097814083099365, + "learning_rate": 0.0006166237113402062, + "loss": 0.4262, + "step": 13801 + }, + { + "epoch": 0.7711260720172082, + "grad_norm": 0.606116771697998, + "learning_rate": 0.0006165956969968624, + "loss": 0.4075, + "step": 13802 + }, + { + "epoch": 0.77118194262089, + "grad_norm": 3.9477429389953613, + "learning_rate": 0.0006165676826535186, + "loss": 0.4451, + "step": 13803 + }, + { + "epoch": 0.7712378132245719, + "grad_norm": 0.4600481688976288, + "learning_rate": 0.0006165396683101748, + "loss": 0.4236, + "step": 13804 + }, + { + "epoch": 0.7712936838282538, + "grad_norm": 0.4312556982040405, + "learning_rate": 0.000616511653966831, + "loss": 0.4147, + "step": 13805 + }, + { + "epoch": 0.7713495544319356, + "grad_norm": 0.585964024066925, + "learning_rate": 0.0006164836396234872, + "loss": 0.4527, + "step": 13806 + }, + { + "epoch": 0.7714054250356175, + "grad_norm": 0.4923916161060333, + "learning_rate": 0.0006164556252801434, + "loss": 0.5528, + "step": 13807 + }, + { + "epoch": 0.7714612956392994, + "grad_norm": 0.4098256230354309, + "learning_rate": 0.0006164276109367996, + "loss": 0.4936, + "step": 13808 + }, + { + "epoch": 0.7715171662429813, + "grad_norm": 0.6140130162239075, + "learning_rate": 0.0006163995965934558, + "loss": 0.3764, + "step": 13809 + }, + { + "epoch": 0.7715730368466631, + "grad_norm": 1.363246202468872, + "learning_rate": 0.0006163715822501121, + "loss": 0.4897, + "step": 13810 + }, + { + "epoch": 0.771628907450345, + "grad_norm": 0.6083342432975769, + "learning_rate": 0.0006163435679067683, + "loss": 0.4896, + "step": 13811 + }, + { + "epoch": 0.7716847780540269, + "grad_norm": 0.6494155526161194, + "learning_rate": 0.0006163155535634245, + "loss": 0.5826, + "step": 13812 + }, + { + "epoch": 0.7717406486577087, + "grad_norm": 0.8161655068397522, + "learning_rate": 0.0006162875392200808, + "loss": 0.6286, + "step": 13813 + }, + { + "epoch": 0.7717965192613906, + "grad_norm": 0.9480617642402649, + "learning_rate": 0.0006162595248767369, + "loss": 0.4825, + "step": 13814 + }, + { + "epoch": 0.7718523898650725, + "grad_norm": 0.5088974833488464, + "learning_rate": 0.0006162315105333931, + "loss": 0.4219, + "step": 13815 + }, + { + "epoch": 0.7719082604687544, + "grad_norm": 0.3384435474872589, + "learning_rate": 0.0006162034961900493, + "loss": 0.3873, + "step": 13816 + }, + { + "epoch": 0.7719641310724362, + "grad_norm": 0.6154324412345886, + "learning_rate": 0.0006161754818467056, + "loss": 0.4306, + "step": 13817 + }, + { + "epoch": 0.7720200016761181, + "grad_norm": 0.46951624751091003, + "learning_rate": 0.0006161474675033617, + "loss": 0.3875, + "step": 13818 + }, + { + "epoch": 0.7720758722798, + "grad_norm": 0.468429833650589, + "learning_rate": 0.0006161194531600179, + "loss": 0.4541, + "step": 13819 + }, + { + "epoch": 0.7721317428834819, + "grad_norm": 0.6512152552604675, + "learning_rate": 0.0006160914388166742, + "loss": 0.4183, + "step": 13820 + }, + { + "epoch": 0.7721876134871637, + "grad_norm": 0.344382107257843, + "learning_rate": 0.0006160634244733303, + "loss": 0.387, + "step": 13821 + }, + { + "epoch": 0.7722434840908456, + "grad_norm": 0.4151298999786377, + "learning_rate": 0.0006160354101299865, + "loss": 0.4664, + "step": 13822 + }, + { + "epoch": 0.7722993546945275, + "grad_norm": 0.4635462164878845, + "learning_rate": 0.0006160073957866427, + "loss": 0.3863, + "step": 13823 + }, + { + "epoch": 0.7723552252982093, + "grad_norm": 0.5584434270858765, + "learning_rate": 0.000615979381443299, + "loss": 0.3989, + "step": 13824 + }, + { + "epoch": 0.7724110959018912, + "grad_norm": 0.5499560236930847, + "learning_rate": 0.0006159513670999551, + "loss": 0.4118, + "step": 13825 + }, + { + "epoch": 0.7724669665055731, + "grad_norm": 2.565215826034546, + "learning_rate": 0.0006159233527566113, + "loss": 0.4987, + "step": 13826 + }, + { + "epoch": 0.772522837109255, + "grad_norm": 0.6134344339370728, + "learning_rate": 0.0006158953384132677, + "loss": 0.3779, + "step": 13827 + }, + { + "epoch": 0.7725787077129368, + "grad_norm": 0.5549224019050598, + "learning_rate": 0.0006158673240699239, + "loss": 0.4569, + "step": 13828 + }, + { + "epoch": 0.7726345783166187, + "grad_norm": 0.7294390797615051, + "learning_rate": 0.00061583930972658, + "loss": 0.5581, + "step": 13829 + }, + { + "epoch": 0.7726904489203006, + "grad_norm": 0.4016130566596985, + "learning_rate": 0.0006158112953832362, + "loss": 0.4151, + "step": 13830 + }, + { + "epoch": 0.7727463195239824, + "grad_norm": 0.6072956919670105, + "learning_rate": 0.0006157832810398925, + "loss": 0.412, + "step": 13831 + }, + { + "epoch": 0.7728021901276644, + "grad_norm": 0.6477127075195312, + "learning_rate": 0.0006157552666965487, + "loss": 0.284, + "step": 13832 + }, + { + "epoch": 0.7728580607313462, + "grad_norm": 0.48194485902786255, + "learning_rate": 0.0006157272523532048, + "loss": 0.4483, + "step": 13833 + }, + { + "epoch": 0.7729139313350281, + "grad_norm": 1.1335233449935913, + "learning_rate": 0.0006156992380098611, + "loss": 0.3806, + "step": 13834 + }, + { + "epoch": 0.7729698019387099, + "grad_norm": 0.3520949184894562, + "learning_rate": 0.0006156712236665173, + "loss": 0.4702, + "step": 13835 + }, + { + "epoch": 0.7730256725423919, + "grad_norm": 0.7685045599937439, + "learning_rate": 0.0006156432093231735, + "loss": 0.5788, + "step": 13836 + }, + { + "epoch": 0.7730815431460737, + "grad_norm": 0.6570239067077637, + "learning_rate": 0.0006156151949798296, + "loss": 0.6186, + "step": 13837 + }, + { + "epoch": 0.7731374137497555, + "grad_norm": 0.5104888081550598, + "learning_rate": 0.0006155871806364859, + "loss": 0.5015, + "step": 13838 + }, + { + "epoch": 0.7731932843534375, + "grad_norm": 0.350246399641037, + "learning_rate": 0.0006155591662931421, + "loss": 0.3146, + "step": 13839 + }, + { + "epoch": 0.7732491549571193, + "grad_norm": 0.438010573387146, + "learning_rate": 0.0006155311519497983, + "loss": 0.4013, + "step": 13840 + }, + { + "epoch": 0.7733050255608012, + "grad_norm": 0.5238475799560547, + "learning_rate": 0.0006155031376064544, + "loss": 0.3723, + "step": 13841 + }, + { + "epoch": 0.773360896164483, + "grad_norm": 1.1460880041122437, + "learning_rate": 0.0006154751232631107, + "loss": 0.5286, + "step": 13842 + }, + { + "epoch": 0.773416766768165, + "grad_norm": 0.38824573159217834, + "learning_rate": 0.000615447108919767, + "loss": 0.3349, + "step": 13843 + }, + { + "epoch": 0.7734726373718468, + "grad_norm": 0.43952807784080505, + "learning_rate": 0.000615419094576423, + "loss": 0.3932, + "step": 13844 + }, + { + "epoch": 0.7735285079755286, + "grad_norm": 1.0843236446380615, + "learning_rate": 0.0006153910802330794, + "loss": 0.5442, + "step": 13845 + }, + { + "epoch": 0.7735843785792106, + "grad_norm": 0.7805629968643188, + "learning_rate": 0.0006153630658897356, + "loss": 0.557, + "step": 13846 + }, + { + "epoch": 0.7736402491828924, + "grad_norm": 0.5975000858306885, + "learning_rate": 0.0006153350515463918, + "loss": 0.4468, + "step": 13847 + }, + { + "epoch": 0.7736961197865743, + "grad_norm": 1.225149154663086, + "learning_rate": 0.0006153070372030479, + "loss": 0.7588, + "step": 13848 + }, + { + "epoch": 0.7737519903902562, + "grad_norm": 0.486422598361969, + "learning_rate": 0.0006152790228597042, + "loss": 0.3867, + "step": 13849 + }, + { + "epoch": 0.7738078609939381, + "grad_norm": 0.4967392683029175, + "learning_rate": 0.0006152510085163604, + "loss": 0.4283, + "step": 13850 + }, + { + "epoch": 0.7738637315976199, + "grad_norm": 0.41810840368270874, + "learning_rate": 0.0006152229941730166, + "loss": 0.4075, + "step": 13851 + }, + { + "epoch": 0.7739196022013017, + "grad_norm": 0.7147414684295654, + "learning_rate": 0.0006151949798296728, + "loss": 0.6264, + "step": 13852 + }, + { + "epoch": 0.7739754728049837, + "grad_norm": 0.6068586111068726, + "learning_rate": 0.000615166965486329, + "loss": 0.4448, + "step": 13853 + }, + { + "epoch": 0.7740313434086655, + "grad_norm": 0.4803554117679596, + "learning_rate": 0.0006151389511429852, + "loss": 0.4503, + "step": 13854 + }, + { + "epoch": 0.7740872140123474, + "grad_norm": 1.5236860513687134, + "learning_rate": 0.0006151109367996414, + "loss": 0.5091, + "step": 13855 + }, + { + "epoch": 0.7741430846160293, + "grad_norm": 0.38155877590179443, + "learning_rate": 0.0006150829224562976, + "loss": 0.4099, + "step": 13856 + }, + { + "epoch": 0.7741989552197112, + "grad_norm": 0.514411449432373, + "learning_rate": 0.0006150549081129538, + "loss": 0.5434, + "step": 13857 + }, + { + "epoch": 0.774254825823393, + "grad_norm": 0.5105291604995728, + "learning_rate": 0.00061502689376961, + "loss": 0.4119, + "step": 13858 + }, + { + "epoch": 0.7743106964270748, + "grad_norm": 0.4898536801338196, + "learning_rate": 0.0006149988794262664, + "loss": 0.3571, + "step": 13859 + }, + { + "epoch": 0.7743665670307568, + "grad_norm": 0.4796188771724701, + "learning_rate": 0.0006149708650829225, + "loss": 0.383, + "step": 13860 + }, + { + "epoch": 0.7744224376344386, + "grad_norm": 0.664761483669281, + "learning_rate": 0.0006149428507395787, + "loss": 0.3868, + "step": 13861 + }, + { + "epoch": 0.7744783082381205, + "grad_norm": 0.44272491335868835, + "learning_rate": 0.0006149148363962349, + "loss": 0.4925, + "step": 13862 + }, + { + "epoch": 0.7745341788418024, + "grad_norm": 0.3791176378726959, + "learning_rate": 0.0006148868220528912, + "loss": 0.4006, + "step": 13863 + }, + { + "epoch": 0.7745900494454843, + "grad_norm": 0.48545214533805847, + "learning_rate": 0.0006148588077095473, + "loss": 0.4384, + "step": 13864 + }, + { + "epoch": 0.7746459200491661, + "grad_norm": 0.389407217502594, + "learning_rate": 0.0006148307933662035, + "loss": 0.5058, + "step": 13865 + }, + { + "epoch": 0.774701790652848, + "grad_norm": 0.8046534657478333, + "learning_rate": 0.0006148027790228598, + "loss": 0.3395, + "step": 13866 + }, + { + "epoch": 0.7747576612565299, + "grad_norm": 0.4507969319820404, + "learning_rate": 0.0006147747646795159, + "loss": 0.4266, + "step": 13867 + }, + { + "epoch": 0.7748135318602117, + "grad_norm": 0.36194008588790894, + "learning_rate": 0.0006147467503361721, + "loss": 0.4599, + "step": 13868 + }, + { + "epoch": 0.7748694024638936, + "grad_norm": 0.4733697175979614, + "learning_rate": 0.0006147187359928283, + "loss": 0.5738, + "step": 13869 + }, + { + "epoch": 0.7749252730675755, + "grad_norm": 0.4646453857421875, + "learning_rate": 0.0006146907216494846, + "loss": 0.4558, + "step": 13870 + }, + { + "epoch": 0.7749811436712574, + "grad_norm": 3.1038312911987305, + "learning_rate": 0.0006146627073061407, + "loss": 0.7248, + "step": 13871 + }, + { + "epoch": 0.7750370142749392, + "grad_norm": 0.3914531171321869, + "learning_rate": 0.0006146346929627969, + "loss": 0.4115, + "step": 13872 + }, + { + "epoch": 0.7750928848786212, + "grad_norm": 1.435227632522583, + "learning_rate": 0.0006146066786194532, + "loss": 0.3974, + "step": 13873 + }, + { + "epoch": 0.775148755482303, + "grad_norm": 0.4587695300579071, + "learning_rate": 0.0006145786642761094, + "loss": 0.3506, + "step": 13874 + }, + { + "epoch": 0.7752046260859848, + "grad_norm": 2.1001546382904053, + "learning_rate": 0.0006145506499327655, + "loss": 0.4938, + "step": 13875 + }, + { + "epoch": 0.7752604966896667, + "grad_norm": 0.8445022702217102, + "learning_rate": 0.0006145226355894217, + "loss": 0.3788, + "step": 13876 + }, + { + "epoch": 0.7753163672933486, + "grad_norm": 0.33768561482429504, + "learning_rate": 0.0006144946212460781, + "loss": 0.3573, + "step": 13877 + }, + { + "epoch": 0.7753722378970305, + "grad_norm": 1.013732671737671, + "learning_rate": 0.0006144666069027343, + "loss": 0.4696, + "step": 13878 + }, + { + "epoch": 0.7754281085007123, + "grad_norm": 0.5863798260688782, + "learning_rate": 0.0006144385925593904, + "loss": 0.5155, + "step": 13879 + }, + { + "epoch": 0.7754839791043943, + "grad_norm": 0.8960651159286499, + "learning_rate": 0.0006144105782160466, + "loss": 0.4415, + "step": 13880 + }, + { + "epoch": 0.7755398497080761, + "grad_norm": 0.36648160219192505, + "learning_rate": 0.0006143825638727029, + "loss": 0.4244, + "step": 13881 + }, + { + "epoch": 0.7755957203117579, + "grad_norm": 0.39367836713790894, + "learning_rate": 0.0006143545495293591, + "loss": 0.4009, + "step": 13882 + }, + { + "epoch": 0.7756515909154399, + "grad_norm": 0.7609037160873413, + "learning_rate": 0.0006143265351860152, + "loss": 0.3703, + "step": 13883 + }, + { + "epoch": 0.7757074615191217, + "grad_norm": 1.3078489303588867, + "learning_rate": 0.0006142985208426715, + "loss": 0.4871, + "step": 13884 + }, + { + "epoch": 0.7757633321228036, + "grad_norm": 0.45456695556640625, + "learning_rate": 0.0006142705064993277, + "loss": 0.3861, + "step": 13885 + }, + { + "epoch": 0.7758192027264854, + "grad_norm": 0.6172959208488464, + "learning_rate": 0.0006142424921559838, + "loss": 0.4922, + "step": 13886 + }, + { + "epoch": 0.7758750733301674, + "grad_norm": 0.435154527425766, + "learning_rate": 0.00061421447781264, + "loss": 0.302, + "step": 13887 + }, + { + "epoch": 0.7759309439338492, + "grad_norm": 0.8391597270965576, + "learning_rate": 0.0006141864634692963, + "loss": 0.5227, + "step": 13888 + }, + { + "epoch": 0.775986814537531, + "grad_norm": 3.717573642730713, + "learning_rate": 0.0006141584491259525, + "loss": 0.6535, + "step": 13889 + }, + { + "epoch": 0.776042685141213, + "grad_norm": 0.4167819917201996, + "learning_rate": 0.0006141304347826086, + "loss": 0.4034, + "step": 13890 + }, + { + "epoch": 0.7760985557448948, + "grad_norm": 7.199531078338623, + "learning_rate": 0.0006141024204392649, + "loss": 0.5645, + "step": 13891 + }, + { + "epoch": 0.7761544263485767, + "grad_norm": 0.67750084400177, + "learning_rate": 0.0006140744060959211, + "loss": 0.5198, + "step": 13892 + }, + { + "epoch": 0.7762102969522585, + "grad_norm": 0.7211425304412842, + "learning_rate": 0.0006140463917525774, + "loss": 0.4474, + "step": 13893 + }, + { + "epoch": 0.7762661675559405, + "grad_norm": 0.740273118019104, + "learning_rate": 0.0006140183774092335, + "loss": 0.4771, + "step": 13894 + }, + { + "epoch": 0.7763220381596223, + "grad_norm": 5.6421709060668945, + "learning_rate": 0.0006139903630658898, + "loss": 0.4038, + "step": 13895 + }, + { + "epoch": 0.7763779087633041, + "grad_norm": 1.108949065208435, + "learning_rate": 0.000613962348722546, + "loss": 0.4699, + "step": 13896 + }, + { + "epoch": 0.7764337793669861, + "grad_norm": 0.5528332591056824, + "learning_rate": 0.0006139343343792022, + "loss": 0.5789, + "step": 13897 + }, + { + "epoch": 0.7764896499706679, + "grad_norm": 0.6197617650032043, + "learning_rate": 0.0006139063200358584, + "loss": 0.4648, + "step": 13898 + }, + { + "epoch": 0.7765455205743498, + "grad_norm": 0.6033675670623779, + "learning_rate": 0.0006138783056925146, + "loss": 0.4364, + "step": 13899 + }, + { + "epoch": 0.7766013911780317, + "grad_norm": 0.4961022436618805, + "learning_rate": 0.0006138502913491708, + "loss": 0.4239, + "step": 13900 + }, + { + "epoch": 0.7766572617817136, + "grad_norm": 0.6176660060882568, + "learning_rate": 0.000613822277005827, + "loss": 0.4375, + "step": 13901 + }, + { + "epoch": 0.7767131323853954, + "grad_norm": 0.5567438006401062, + "learning_rate": 0.0006137942626624832, + "loss": 0.5138, + "step": 13902 + }, + { + "epoch": 0.7767690029890773, + "grad_norm": 0.46664389967918396, + "learning_rate": 0.0006137662483191394, + "loss": 0.3344, + "step": 13903 + }, + { + "epoch": 0.7768248735927592, + "grad_norm": 0.3939303755760193, + "learning_rate": 0.0006137382339757956, + "loss": 0.4593, + "step": 13904 + }, + { + "epoch": 0.776880744196441, + "grad_norm": 0.9356236457824707, + "learning_rate": 0.0006137102196324519, + "loss": 0.4533, + "step": 13905 + }, + { + "epoch": 0.7769366148001229, + "grad_norm": 0.5402829051017761, + "learning_rate": 0.000613682205289108, + "loss": 0.472, + "step": 13906 + }, + { + "epoch": 0.7769924854038048, + "grad_norm": 3.3545875549316406, + "learning_rate": 0.0006136541909457642, + "loss": 0.3533, + "step": 13907 + }, + { + "epoch": 0.7770483560074867, + "grad_norm": 0.6035195589065552, + "learning_rate": 0.0006136261766024204, + "loss": 0.6272, + "step": 13908 + }, + { + "epoch": 0.7771042266111685, + "grad_norm": 0.6171263456344604, + "learning_rate": 0.0006135981622590766, + "loss": 0.4413, + "step": 13909 + }, + { + "epoch": 0.7771600972148504, + "grad_norm": 0.45323804020881653, + "learning_rate": 0.0006135701479157329, + "loss": 0.4445, + "step": 13910 + }, + { + "epoch": 0.7772159678185323, + "grad_norm": 0.5028830170631409, + "learning_rate": 0.0006135421335723891, + "loss": 0.4487, + "step": 13911 + }, + { + "epoch": 0.7772718384222141, + "grad_norm": 0.4040389060974121, + "learning_rate": 0.0006135141192290454, + "loss": 0.3821, + "step": 13912 + }, + { + "epoch": 0.777327709025896, + "grad_norm": 1.14820396900177, + "learning_rate": 0.0006134861048857015, + "loss": 0.4142, + "step": 13913 + }, + { + "epoch": 0.7773835796295779, + "grad_norm": 1.860353946685791, + "learning_rate": 0.0006134580905423577, + "loss": 0.4402, + "step": 13914 + }, + { + "epoch": 0.7774394502332598, + "grad_norm": 0.3918580412864685, + "learning_rate": 0.0006134300761990139, + "loss": 0.3908, + "step": 13915 + }, + { + "epoch": 0.7774953208369416, + "grad_norm": 0.5685648322105408, + "learning_rate": 0.0006134020618556702, + "loss": 0.425, + "step": 13916 + }, + { + "epoch": 0.7775511914406236, + "grad_norm": 0.6462746262550354, + "learning_rate": 0.0006133740475123263, + "loss": 0.4527, + "step": 13917 + }, + { + "epoch": 0.7776070620443054, + "grad_norm": 0.6797937750816345, + "learning_rate": 0.0006133460331689825, + "loss": 0.61, + "step": 13918 + }, + { + "epoch": 0.7776629326479872, + "grad_norm": 0.46049702167510986, + "learning_rate": 0.0006133180188256387, + "loss": 0.4403, + "step": 13919 + }, + { + "epoch": 0.7777188032516691, + "grad_norm": 0.38274839520454407, + "learning_rate": 0.000613290004482295, + "loss": 0.3908, + "step": 13920 + }, + { + "epoch": 0.777774673855351, + "grad_norm": 0.472062349319458, + "learning_rate": 0.0006132619901389511, + "loss": 0.4405, + "step": 13921 + }, + { + "epoch": 0.7778305444590329, + "grad_norm": 0.5152544975280762, + "learning_rate": 0.0006132339757956073, + "loss": 0.3781, + "step": 13922 + }, + { + "epoch": 0.7778864150627147, + "grad_norm": 7.128126621246338, + "learning_rate": 0.0006132059614522636, + "loss": 0.4493, + "step": 13923 + }, + { + "epoch": 0.7779422856663967, + "grad_norm": 0.5339787602424622, + "learning_rate": 0.0006131779471089198, + "loss": 0.5307, + "step": 13924 + }, + { + "epoch": 0.7779981562700785, + "grad_norm": 0.5221497416496277, + "learning_rate": 0.0006131499327655759, + "loss": 0.5892, + "step": 13925 + }, + { + "epoch": 0.7780540268737604, + "grad_norm": 1.5358433723449707, + "learning_rate": 0.0006131219184222321, + "loss": 0.4557, + "step": 13926 + }, + { + "epoch": 0.7781098974774422, + "grad_norm": 0.5025541186332703, + "learning_rate": 0.0006130939040788885, + "loss": 0.3908, + "step": 13927 + }, + { + "epoch": 0.7781657680811241, + "grad_norm": 0.4310501217842102, + "learning_rate": 0.0006130658897355446, + "loss": 0.4188, + "step": 13928 + }, + { + "epoch": 0.778221638684806, + "grad_norm": 0.47798284888267517, + "learning_rate": 0.0006130378753922008, + "loss": 0.4825, + "step": 13929 + }, + { + "epoch": 0.7782775092884878, + "grad_norm": 1.3164056539535522, + "learning_rate": 0.0006130098610488571, + "loss": 0.4066, + "step": 13930 + }, + { + "epoch": 0.7783333798921698, + "grad_norm": 0.9037346243858337, + "learning_rate": 0.0006129818467055133, + "loss": 0.3624, + "step": 13931 + }, + { + "epoch": 0.7783892504958516, + "grad_norm": 0.5430248975753784, + "learning_rate": 0.0006129538323621694, + "loss": 0.3701, + "step": 13932 + }, + { + "epoch": 0.7784451210995335, + "grad_norm": 0.43240320682525635, + "learning_rate": 0.0006129258180188256, + "loss": 0.3782, + "step": 13933 + }, + { + "epoch": 0.7785009917032154, + "grad_norm": 0.49823302030563354, + "learning_rate": 0.0006128978036754819, + "loss": 0.4596, + "step": 13934 + }, + { + "epoch": 0.7785568623068972, + "grad_norm": 0.4914328455924988, + "learning_rate": 0.0006128697893321381, + "loss": 0.317, + "step": 13935 + }, + { + "epoch": 0.7786127329105791, + "grad_norm": 0.6914281845092773, + "learning_rate": 0.0006128417749887942, + "loss": 0.4082, + "step": 13936 + }, + { + "epoch": 0.7786686035142609, + "grad_norm": 0.4167875647544861, + "learning_rate": 0.0006128137606454505, + "loss": 0.5057, + "step": 13937 + }, + { + "epoch": 0.7787244741179429, + "grad_norm": 0.7185355424880981, + "learning_rate": 0.0006127857463021067, + "loss": 0.4943, + "step": 13938 + }, + { + "epoch": 0.7787803447216247, + "grad_norm": 0.45757466554641724, + "learning_rate": 0.0006127577319587629, + "loss": 0.4785, + "step": 13939 + }, + { + "epoch": 0.7788362153253066, + "grad_norm": 0.4608824849128723, + "learning_rate": 0.000612729717615419, + "loss": 0.4141, + "step": 13940 + }, + { + "epoch": 0.7788920859289885, + "grad_norm": 0.42361757159233093, + "learning_rate": 0.0006127017032720753, + "loss": 0.4665, + "step": 13941 + }, + { + "epoch": 0.7789479565326703, + "grad_norm": 1.220296025276184, + "learning_rate": 0.0006126736889287315, + "loss": 0.5487, + "step": 13942 + }, + { + "epoch": 0.7790038271363522, + "grad_norm": 0.7193284034729004, + "learning_rate": 0.0006126456745853878, + "loss": 0.4461, + "step": 13943 + }, + { + "epoch": 0.779059697740034, + "grad_norm": 0.41412195563316345, + "learning_rate": 0.000612617660242044, + "loss": 0.459, + "step": 13944 + }, + { + "epoch": 0.779115568343716, + "grad_norm": 0.4420609474182129, + "learning_rate": 0.0006125896458987002, + "loss": 0.4367, + "step": 13945 + }, + { + "epoch": 0.7791714389473978, + "grad_norm": 0.4767250418663025, + "learning_rate": 0.0006125616315553564, + "loss": 0.4787, + "step": 13946 + }, + { + "epoch": 0.7792273095510797, + "grad_norm": 0.9959251880645752, + "learning_rate": 0.0006125336172120126, + "loss": 0.5085, + "step": 13947 + }, + { + "epoch": 0.7792831801547616, + "grad_norm": 0.7586161494255066, + "learning_rate": 0.0006125056028686688, + "loss": 0.516, + "step": 13948 + }, + { + "epoch": 0.7793390507584435, + "grad_norm": 0.6478872299194336, + "learning_rate": 0.000612477588525325, + "loss": 0.4583, + "step": 13949 + }, + { + "epoch": 0.7793949213621253, + "grad_norm": 0.44629916548728943, + "learning_rate": 0.0006124495741819812, + "loss": 0.4097, + "step": 13950 + }, + { + "epoch": 0.7794507919658072, + "grad_norm": 0.5712570548057556, + "learning_rate": 0.0006124215598386373, + "loss": 0.3823, + "step": 13951 + }, + { + "epoch": 0.7795066625694891, + "grad_norm": 0.38568487763404846, + "learning_rate": 0.0006123935454952936, + "loss": 0.3691, + "step": 13952 + }, + { + "epoch": 0.7795625331731709, + "grad_norm": 0.47608673572540283, + "learning_rate": 0.0006123655311519498, + "loss": 0.4125, + "step": 13953 + }, + { + "epoch": 0.7796184037768528, + "grad_norm": 0.5506026744842529, + "learning_rate": 0.000612337516808606, + "loss": 0.4119, + "step": 13954 + }, + { + "epoch": 0.7796742743805347, + "grad_norm": 1.6653090715408325, + "learning_rate": 0.0006123095024652622, + "loss": 0.4518, + "step": 13955 + }, + { + "epoch": 0.7797301449842166, + "grad_norm": 0.5523743033409119, + "learning_rate": 0.0006122814881219184, + "loss": 0.501, + "step": 13956 + }, + { + "epoch": 0.7797860155878984, + "grad_norm": 0.5521410703659058, + "learning_rate": 0.0006122534737785746, + "loss": 0.4737, + "step": 13957 + }, + { + "epoch": 0.7798418861915803, + "grad_norm": 0.7654346227645874, + "learning_rate": 0.0006122254594352308, + "loss": 0.3753, + "step": 13958 + }, + { + "epoch": 0.7798977567952622, + "grad_norm": 0.34542983770370483, + "learning_rate": 0.000612197445091887, + "loss": 0.4451, + "step": 13959 + }, + { + "epoch": 0.779953627398944, + "grad_norm": 0.4153323173522949, + "learning_rate": 0.0006121694307485433, + "loss": 0.3716, + "step": 13960 + }, + { + "epoch": 0.7800094980026259, + "grad_norm": 0.6382514238357544, + "learning_rate": 0.0006121414164051995, + "loss": 0.4097, + "step": 13961 + }, + { + "epoch": 0.7800653686063078, + "grad_norm": 0.4304259121417999, + "learning_rate": 0.0006121134020618558, + "loss": 0.36, + "step": 13962 + }, + { + "epoch": 0.7801212392099897, + "grad_norm": 0.6184456944465637, + "learning_rate": 0.0006120853877185119, + "loss": 0.4393, + "step": 13963 + }, + { + "epoch": 0.7801771098136715, + "grad_norm": 0.5436460971832275, + "learning_rate": 0.0006120573733751681, + "loss": 0.4632, + "step": 13964 + }, + { + "epoch": 0.7802329804173534, + "grad_norm": 0.6049615740776062, + "learning_rate": 0.0006120293590318243, + "loss": 0.4864, + "step": 13965 + }, + { + "epoch": 0.7802888510210353, + "grad_norm": 0.7908092141151428, + "learning_rate": 0.0006120013446884806, + "loss": 0.6058, + "step": 13966 + }, + { + "epoch": 0.7803447216247171, + "grad_norm": 0.45694971084594727, + "learning_rate": 0.0006119733303451367, + "loss": 0.5223, + "step": 13967 + }, + { + "epoch": 0.7804005922283991, + "grad_norm": 0.41688019037246704, + "learning_rate": 0.0006119453160017929, + "loss": 0.4636, + "step": 13968 + }, + { + "epoch": 0.7804564628320809, + "grad_norm": 0.45415231585502625, + "learning_rate": 0.0006119173016584492, + "loss": 0.465, + "step": 13969 + }, + { + "epoch": 0.7805123334357628, + "grad_norm": 2.692054510116577, + "learning_rate": 0.0006118892873151053, + "loss": 0.4912, + "step": 13970 + }, + { + "epoch": 0.7805682040394446, + "grad_norm": 0.5852211713790894, + "learning_rate": 0.0006118612729717615, + "loss": 0.6166, + "step": 13971 + }, + { + "epoch": 0.7806240746431266, + "grad_norm": 0.657265841960907, + "learning_rate": 0.0006118332586284177, + "loss": 0.5069, + "step": 13972 + }, + { + "epoch": 0.7806799452468084, + "grad_norm": 0.4689260423183441, + "learning_rate": 0.000611805244285074, + "loss": 0.5013, + "step": 13973 + }, + { + "epoch": 0.7807358158504902, + "grad_norm": 0.502858579158783, + "learning_rate": 0.0006117772299417301, + "loss": 0.3875, + "step": 13974 + }, + { + "epoch": 0.7807916864541722, + "grad_norm": 0.5287750959396362, + "learning_rate": 0.0006117492155983863, + "loss": 0.4807, + "step": 13975 + }, + { + "epoch": 0.780847557057854, + "grad_norm": 0.3894875645637512, + "learning_rate": 0.0006117212012550427, + "loss": 0.3477, + "step": 13976 + }, + { + "epoch": 0.7809034276615359, + "grad_norm": 1.8044291734695435, + "learning_rate": 0.0006116931869116989, + "loss": 0.5091, + "step": 13977 + }, + { + "epoch": 0.7809592982652177, + "grad_norm": 0.6624112129211426, + "learning_rate": 0.000611665172568355, + "loss": 0.4904, + "step": 13978 + }, + { + "epoch": 0.7810151688688997, + "grad_norm": 0.4585859477519989, + "learning_rate": 0.0006116371582250112, + "loss": 0.5106, + "step": 13979 + }, + { + "epoch": 0.7810710394725815, + "grad_norm": 0.6200411915779114, + "learning_rate": 0.0006116091438816675, + "loss": 0.5622, + "step": 13980 + }, + { + "epoch": 0.7811269100762633, + "grad_norm": 0.33737313747406006, + "learning_rate": 0.0006115811295383237, + "loss": 0.3901, + "step": 13981 + }, + { + "epoch": 0.7811827806799453, + "grad_norm": 0.4465620219707489, + "learning_rate": 0.0006115531151949798, + "loss": 0.4272, + "step": 13982 + }, + { + "epoch": 0.7812386512836271, + "grad_norm": 0.5171107649803162, + "learning_rate": 0.0006115251008516361, + "loss": 0.5592, + "step": 13983 + }, + { + "epoch": 0.781294521887309, + "grad_norm": 0.3815889358520508, + "learning_rate": 0.0006114970865082923, + "loss": 0.3834, + "step": 13984 + }, + { + "epoch": 0.7813503924909909, + "grad_norm": 0.5594921112060547, + "learning_rate": 0.0006114690721649485, + "loss": 0.4397, + "step": 13985 + }, + { + "epoch": 0.7814062630946728, + "grad_norm": 0.5994755625724792, + "learning_rate": 0.0006114410578216046, + "loss": 0.4316, + "step": 13986 + }, + { + "epoch": 0.7814621336983546, + "grad_norm": 0.5266635417938232, + "learning_rate": 0.0006114130434782609, + "loss": 0.4704, + "step": 13987 + }, + { + "epoch": 0.7815180043020364, + "grad_norm": 0.40535449981689453, + "learning_rate": 0.0006113850291349171, + "loss": 0.4514, + "step": 13988 + }, + { + "epoch": 0.7815738749057184, + "grad_norm": 0.49640530347824097, + "learning_rate": 0.0006113570147915733, + "loss": 0.4711, + "step": 13989 + }, + { + "epoch": 0.7816297455094002, + "grad_norm": 0.8138172626495361, + "learning_rate": 0.0006113290004482294, + "loss": 0.439, + "step": 13990 + }, + { + "epoch": 0.7816856161130821, + "grad_norm": 0.9353523254394531, + "learning_rate": 0.0006113009861048857, + "loss": 0.453, + "step": 13991 + }, + { + "epoch": 0.781741486716764, + "grad_norm": 0.47526901960372925, + "learning_rate": 0.000611272971761542, + "loss": 0.4462, + "step": 13992 + }, + { + "epoch": 0.7817973573204459, + "grad_norm": 0.7562959790229797, + "learning_rate": 0.000611244957418198, + "loss": 0.4498, + "step": 13993 + }, + { + "epoch": 0.7818532279241277, + "grad_norm": 0.4222371280193329, + "learning_rate": 0.0006112169430748544, + "loss": 0.3386, + "step": 13994 + }, + { + "epoch": 0.7819090985278095, + "grad_norm": 0.5054277777671814, + "learning_rate": 0.0006111889287315106, + "loss": 0.4627, + "step": 13995 + }, + { + "epoch": 0.7819649691314915, + "grad_norm": 0.4616905450820923, + "learning_rate": 0.0006111609143881668, + "loss": 0.5081, + "step": 13996 + }, + { + "epoch": 0.7820208397351733, + "grad_norm": 3.1584157943725586, + "learning_rate": 0.0006111329000448229, + "loss": 0.4648, + "step": 13997 + }, + { + "epoch": 0.7820767103388552, + "grad_norm": 0.5458407998085022, + "learning_rate": 0.0006111048857014792, + "loss": 0.4682, + "step": 13998 + }, + { + "epoch": 0.7821325809425371, + "grad_norm": 0.427781879901886, + "learning_rate": 0.0006110768713581354, + "loss": 0.3712, + "step": 13999 + }, + { + "epoch": 0.782188451546219, + "grad_norm": 0.6624724864959717, + "learning_rate": 0.0006110488570147916, + "loss": 0.5068, + "step": 14000 + }, + { + "epoch": 0.782188451546219, + "eval_cer": 0.0891623842100095, + "eval_loss": 0.3372998535633087, + "eval_runtime": 56.0728, + "eval_samples_per_second": 80.93, + "eval_steps_per_second": 5.065, + "eval_wer": 0.3518337476606799, + "step": 14000 + }, + { + "epoch": 0.7822443221499008, + "grad_norm": 1.3786218166351318, + "learning_rate": 0.0006110208426714478, + "loss": 0.3934, + "step": 14001 + }, + { + "epoch": 0.7823001927535828, + "grad_norm": 1.111899495124817, + "learning_rate": 0.000610992828328104, + "loss": 0.4746, + "step": 14002 + }, + { + "epoch": 0.7823560633572646, + "grad_norm": 0.8088793158531189, + "learning_rate": 0.0006109648139847602, + "loss": 0.5401, + "step": 14003 + }, + { + "epoch": 0.7824119339609464, + "grad_norm": 0.4357903003692627, + "learning_rate": 0.0006109367996414164, + "loss": 0.3472, + "step": 14004 + }, + { + "epoch": 0.7824678045646283, + "grad_norm": 0.7322601675987244, + "learning_rate": 0.0006109087852980726, + "loss": 0.4515, + "step": 14005 + }, + { + "epoch": 0.7825236751683102, + "grad_norm": 0.717999279499054, + "learning_rate": 0.0006108807709547288, + "loss": 0.474, + "step": 14006 + }, + { + "epoch": 0.7825795457719921, + "grad_norm": 0.4278312921524048, + "learning_rate": 0.000610852756611385, + "loss": 0.3628, + "step": 14007 + }, + { + "epoch": 0.7826354163756739, + "grad_norm": 0.8685829043388367, + "learning_rate": 0.0006108247422680413, + "loss": 0.5441, + "step": 14008 + }, + { + "epoch": 0.7826912869793559, + "grad_norm": 0.6622011661529541, + "learning_rate": 0.0006107967279246974, + "loss": 0.3485, + "step": 14009 + }, + { + "epoch": 0.7827471575830377, + "grad_norm": 0.5639208555221558, + "learning_rate": 0.0006107687135813537, + "loss": 0.4074, + "step": 14010 + }, + { + "epoch": 0.7828030281867195, + "grad_norm": 1.158021330833435, + "learning_rate": 0.0006107406992380099, + "loss": 0.8034, + "step": 14011 + }, + { + "epoch": 0.7828588987904014, + "grad_norm": 0.4006234407424927, + "learning_rate": 0.0006107126848946661, + "loss": 0.518, + "step": 14012 + }, + { + "epoch": 0.7829147693940833, + "grad_norm": 0.3972032964229584, + "learning_rate": 0.0006106846705513223, + "loss": 0.4928, + "step": 14013 + }, + { + "epoch": 0.7829706399977652, + "grad_norm": 0.49122461676597595, + "learning_rate": 0.0006106566562079785, + "loss": 0.4127, + "step": 14014 + }, + { + "epoch": 0.783026510601447, + "grad_norm": 0.6052459478378296, + "learning_rate": 0.0006106286418646348, + "loss": 0.355, + "step": 14015 + }, + { + "epoch": 0.783082381205129, + "grad_norm": 0.5672774910926819, + "learning_rate": 0.0006106006275212909, + "loss": 0.452, + "step": 14016 + }, + { + "epoch": 0.7831382518088108, + "grad_norm": 0.5662577748298645, + "learning_rate": 0.0006105726131779471, + "loss": 0.4011, + "step": 14017 + }, + { + "epoch": 0.7831941224124926, + "grad_norm": 0.5181400775909424, + "learning_rate": 0.0006105445988346033, + "loss": 0.3755, + "step": 14018 + }, + { + "epoch": 0.7832499930161746, + "grad_norm": 0.48926013708114624, + "learning_rate": 0.0006105165844912596, + "loss": 0.416, + "step": 14019 + }, + { + "epoch": 0.7833058636198564, + "grad_norm": 0.5265619158744812, + "learning_rate": 0.0006104885701479157, + "loss": 0.4907, + "step": 14020 + }, + { + "epoch": 0.7833617342235383, + "grad_norm": 0.46691426634788513, + "learning_rate": 0.0006104605558045719, + "loss": 0.5026, + "step": 14021 + }, + { + "epoch": 0.7834176048272201, + "grad_norm": 0.3610619902610779, + "learning_rate": 0.0006104325414612281, + "loss": 0.4496, + "step": 14022 + }, + { + "epoch": 0.7834734754309021, + "grad_norm": 0.3900066912174225, + "learning_rate": 0.0006104045271178844, + "loss": 0.4212, + "step": 14023 + }, + { + "epoch": 0.7835293460345839, + "grad_norm": 0.3446091115474701, + "learning_rate": 0.0006103765127745405, + "loss": 0.4284, + "step": 14024 + }, + { + "epoch": 0.7835852166382657, + "grad_norm": 1.0608267784118652, + "learning_rate": 0.0006103484984311967, + "loss": 0.4703, + "step": 14025 + }, + { + "epoch": 0.7836410872419477, + "grad_norm": 0.5812452435493469, + "learning_rate": 0.000610320484087853, + "loss": 0.3757, + "step": 14026 + }, + { + "epoch": 0.7836969578456295, + "grad_norm": 0.5822714567184448, + "learning_rate": 0.0006102924697445093, + "loss": 0.3987, + "step": 14027 + }, + { + "epoch": 0.7837528284493114, + "grad_norm": 0.7281882762908936, + "learning_rate": 0.0006102644554011654, + "loss": 0.3921, + "step": 14028 + }, + { + "epoch": 0.7838086990529932, + "grad_norm": 0.5080806016921997, + "learning_rate": 0.0006102364410578216, + "loss": 0.4172, + "step": 14029 + }, + { + "epoch": 0.7838645696566752, + "grad_norm": 0.6439521312713623, + "learning_rate": 0.0006102084267144779, + "loss": 0.4852, + "step": 14030 + }, + { + "epoch": 0.783920440260357, + "grad_norm": 0.9081251621246338, + "learning_rate": 0.0006101804123711341, + "loss": 0.4741, + "step": 14031 + }, + { + "epoch": 0.7839763108640389, + "grad_norm": 0.4548361003398895, + "learning_rate": 0.0006101523980277902, + "loss": 0.4599, + "step": 14032 + }, + { + "epoch": 0.7840321814677208, + "grad_norm": 0.6361120939254761, + "learning_rate": 0.0006101243836844465, + "loss": 0.5829, + "step": 14033 + }, + { + "epoch": 0.7840880520714026, + "grad_norm": 0.6921722888946533, + "learning_rate": 0.0006100963693411027, + "loss": 0.5123, + "step": 14034 + }, + { + "epoch": 0.7841439226750845, + "grad_norm": 0.39056381583213806, + "learning_rate": 0.0006100683549977588, + "loss": 0.354, + "step": 14035 + }, + { + "epoch": 0.7841997932787663, + "grad_norm": 0.5046605467796326, + "learning_rate": 0.000610040340654415, + "loss": 0.4141, + "step": 14036 + }, + { + "epoch": 0.7842556638824483, + "grad_norm": 0.49393874406814575, + "learning_rate": 0.0006100123263110713, + "loss": 0.4338, + "step": 14037 + }, + { + "epoch": 0.7843115344861301, + "grad_norm": 0.4790928363800049, + "learning_rate": 0.0006099843119677275, + "loss": 0.3415, + "step": 14038 + }, + { + "epoch": 0.784367405089812, + "grad_norm": 0.5965774655342102, + "learning_rate": 0.0006099562976243836, + "loss": 0.4994, + "step": 14039 + }, + { + "epoch": 0.7844232756934939, + "grad_norm": 0.628102719783783, + "learning_rate": 0.0006099282832810399, + "loss": 0.4088, + "step": 14040 + }, + { + "epoch": 0.7844791462971757, + "grad_norm": 0.6165995001792908, + "learning_rate": 0.0006099002689376961, + "loss": 0.427, + "step": 14041 + }, + { + "epoch": 0.7845350169008576, + "grad_norm": 0.5740131139755249, + "learning_rate": 0.0006098722545943523, + "loss": 0.3617, + "step": 14042 + }, + { + "epoch": 0.7845908875045395, + "grad_norm": 0.3879907429218292, + "learning_rate": 0.0006098442402510084, + "loss": 0.4437, + "step": 14043 + }, + { + "epoch": 0.7846467581082214, + "grad_norm": 2.2630503177642822, + "learning_rate": 0.0006098162259076648, + "loss": 0.3879, + "step": 14044 + }, + { + "epoch": 0.7847026287119032, + "grad_norm": 0.5991496443748474, + "learning_rate": 0.000609788211564321, + "loss": 0.4481, + "step": 14045 + }, + { + "epoch": 0.7847584993155851, + "grad_norm": 0.4962519705295563, + "learning_rate": 0.0006097601972209772, + "loss": 0.3218, + "step": 14046 + }, + { + "epoch": 0.784814369919267, + "grad_norm": 0.45602405071258545, + "learning_rate": 0.0006097321828776334, + "loss": 0.4283, + "step": 14047 + }, + { + "epoch": 0.7848702405229488, + "grad_norm": 0.6172324419021606, + "learning_rate": 0.0006097041685342896, + "loss": 0.6385, + "step": 14048 + }, + { + "epoch": 0.7849261111266307, + "grad_norm": 0.4858020544052124, + "learning_rate": 0.0006096761541909458, + "loss": 0.4083, + "step": 14049 + }, + { + "epoch": 0.7849819817303126, + "grad_norm": 0.5142977237701416, + "learning_rate": 0.000609648139847602, + "loss": 0.5252, + "step": 14050 + }, + { + "epoch": 0.7850378523339945, + "grad_norm": 0.4616599678993225, + "learning_rate": 0.0006096201255042582, + "loss": 0.396, + "step": 14051 + }, + { + "epoch": 0.7850937229376763, + "grad_norm": 0.40672749280929565, + "learning_rate": 0.0006095921111609144, + "loss": 0.477, + "step": 14052 + }, + { + "epoch": 0.7851495935413582, + "grad_norm": 0.4252544939517975, + "learning_rate": 0.0006095640968175706, + "loss": 0.4766, + "step": 14053 + }, + { + "epoch": 0.7852054641450401, + "grad_norm": 0.47561633586883545, + "learning_rate": 0.0006095360824742269, + "loss": 0.3715, + "step": 14054 + }, + { + "epoch": 0.785261334748722, + "grad_norm": 0.357504665851593, + "learning_rate": 0.000609508068130883, + "loss": 0.3609, + "step": 14055 + }, + { + "epoch": 0.7853172053524038, + "grad_norm": 0.4081772565841675, + "learning_rate": 0.0006094800537875392, + "loss": 0.5485, + "step": 14056 + }, + { + "epoch": 0.7853730759560857, + "grad_norm": 0.42895087599754333, + "learning_rate": 0.0006094520394441954, + "loss": 0.3429, + "step": 14057 + }, + { + "epoch": 0.7854289465597676, + "grad_norm": 0.4609881639480591, + "learning_rate": 0.0006094240251008516, + "loss": 0.4918, + "step": 14058 + }, + { + "epoch": 0.7854848171634494, + "grad_norm": 0.6033467650413513, + "learning_rate": 0.0006093960107575078, + "loss": 0.418, + "step": 14059 + }, + { + "epoch": 0.7855406877671314, + "grad_norm": 0.5170384645462036, + "learning_rate": 0.000609367996414164, + "loss": 0.434, + "step": 14060 + }, + { + "epoch": 0.7855965583708132, + "grad_norm": 0.9726393222808838, + "learning_rate": 0.0006093399820708203, + "loss": 0.4553, + "step": 14061 + }, + { + "epoch": 0.785652428974495, + "grad_norm": 0.3288683593273163, + "learning_rate": 0.0006093119677274765, + "loss": 0.4157, + "step": 14062 + }, + { + "epoch": 0.7857082995781769, + "grad_norm": 0.37786865234375, + "learning_rate": 0.0006092839533841327, + "loss": 0.4205, + "step": 14063 + }, + { + "epoch": 0.7857641701818588, + "grad_norm": 3.7560248374938965, + "learning_rate": 0.0006092559390407889, + "loss": 0.5352, + "step": 14064 + }, + { + "epoch": 0.7858200407855407, + "grad_norm": 0.3882884979248047, + "learning_rate": 0.0006092279246974452, + "loss": 0.4253, + "step": 14065 + }, + { + "epoch": 0.7858759113892225, + "grad_norm": 0.9994754791259766, + "learning_rate": 0.0006091999103541013, + "loss": 0.4564, + "step": 14066 + }, + { + "epoch": 0.7859317819929045, + "grad_norm": 0.9700228571891785, + "learning_rate": 0.0006091718960107575, + "loss": 0.5515, + "step": 14067 + }, + { + "epoch": 0.7859876525965863, + "grad_norm": 0.5927050113677979, + "learning_rate": 0.0006091438816674137, + "loss": 0.447, + "step": 14068 + }, + { + "epoch": 0.7860435232002682, + "grad_norm": 0.6875207424163818, + "learning_rate": 0.00060911586732407, + "loss": 0.5129, + "step": 14069 + }, + { + "epoch": 0.78609939380395, + "grad_norm": 0.6222660541534424, + "learning_rate": 0.0006090878529807261, + "loss": 0.4315, + "step": 14070 + }, + { + "epoch": 0.786155264407632, + "grad_norm": 0.3984541594982147, + "learning_rate": 0.0006090598386373823, + "loss": 0.4605, + "step": 14071 + }, + { + "epoch": 0.7862111350113138, + "grad_norm": 0.5592263340950012, + "learning_rate": 0.0006090318242940386, + "loss": 0.5658, + "step": 14072 + }, + { + "epoch": 0.7862670056149956, + "grad_norm": 0.6096998453140259, + "learning_rate": 0.0006090038099506948, + "loss": 0.4351, + "step": 14073 + }, + { + "epoch": 0.7863228762186776, + "grad_norm": 0.593959629535675, + "learning_rate": 0.0006089757956073509, + "loss": 0.6654, + "step": 14074 + }, + { + "epoch": 0.7863787468223594, + "grad_norm": 0.3257344961166382, + "learning_rate": 0.0006089477812640071, + "loss": 0.3154, + "step": 14075 + }, + { + "epoch": 0.7864346174260413, + "grad_norm": 1.3043227195739746, + "learning_rate": 0.0006089197669206635, + "loss": 0.47, + "step": 14076 + }, + { + "epoch": 0.7864904880297232, + "grad_norm": 0.35866081714630127, + "learning_rate": 0.0006088917525773195, + "loss": 0.3479, + "step": 14077 + }, + { + "epoch": 0.786546358633405, + "grad_norm": 0.6905713081359863, + "learning_rate": 0.0006088637382339758, + "loss": 0.3767, + "step": 14078 + }, + { + "epoch": 0.7866022292370869, + "grad_norm": 0.4566895663738251, + "learning_rate": 0.0006088357238906321, + "loss": 0.5672, + "step": 14079 + }, + { + "epoch": 0.7866580998407687, + "grad_norm": 0.7584131956100464, + "learning_rate": 0.0006088077095472883, + "loss": 0.4994, + "step": 14080 + }, + { + "epoch": 0.7867139704444507, + "grad_norm": 0.4960346817970276, + "learning_rate": 0.0006087796952039444, + "loss": 0.4418, + "step": 14081 + }, + { + "epoch": 0.7867698410481325, + "grad_norm": 0.35571786761283875, + "learning_rate": 0.0006087516808606006, + "loss": 0.3831, + "step": 14082 + }, + { + "epoch": 0.7868257116518144, + "grad_norm": 0.7240315675735474, + "learning_rate": 0.0006087236665172569, + "loss": 0.5273, + "step": 14083 + }, + { + "epoch": 0.7868815822554963, + "grad_norm": 0.5080440044403076, + "learning_rate": 0.0006086956521739131, + "loss": 0.4632, + "step": 14084 + }, + { + "epoch": 0.7869374528591782, + "grad_norm": 0.44193437695503235, + "learning_rate": 0.0006086676378305692, + "loss": 0.4324, + "step": 14085 + }, + { + "epoch": 0.78699332346286, + "grad_norm": 0.40668484568595886, + "learning_rate": 0.0006086396234872255, + "loss": 0.3975, + "step": 14086 + }, + { + "epoch": 0.7870491940665418, + "grad_norm": 0.5552467703819275, + "learning_rate": 0.0006086116091438817, + "loss": 0.4589, + "step": 14087 + }, + { + "epoch": 0.7871050646702238, + "grad_norm": 0.4255148470401764, + "learning_rate": 0.0006085835948005379, + "loss": 0.4481, + "step": 14088 + }, + { + "epoch": 0.7871609352739056, + "grad_norm": 0.3803299367427826, + "learning_rate": 0.000608555580457194, + "loss": 0.4518, + "step": 14089 + }, + { + "epoch": 0.7872168058775875, + "grad_norm": 0.48680269718170166, + "learning_rate": 0.0006085275661138503, + "loss": 0.4771, + "step": 14090 + }, + { + "epoch": 0.7872726764812694, + "grad_norm": 0.46074178814888, + "learning_rate": 0.0006084995517705065, + "loss": 0.5691, + "step": 14091 + }, + { + "epoch": 0.7873285470849513, + "grad_norm": 0.4701012372970581, + "learning_rate": 0.0006084715374271627, + "loss": 0.4345, + "step": 14092 + }, + { + "epoch": 0.7873844176886331, + "grad_norm": 0.6031306982040405, + "learning_rate": 0.0006084435230838188, + "loss": 0.4495, + "step": 14093 + }, + { + "epoch": 0.787440288292315, + "grad_norm": 0.39219948649406433, + "learning_rate": 0.0006084155087404752, + "loss": 0.3879, + "step": 14094 + }, + { + "epoch": 0.7874961588959969, + "grad_norm": 0.5127809643745422, + "learning_rate": 0.0006083874943971314, + "loss": 0.539, + "step": 14095 + }, + { + "epoch": 0.7875520294996787, + "grad_norm": 0.3679586946964264, + "learning_rate": 0.0006083594800537876, + "loss": 0.3665, + "step": 14096 + }, + { + "epoch": 0.7876079001033606, + "grad_norm": 0.5041347146034241, + "learning_rate": 0.0006083314657104438, + "loss": 0.4298, + "step": 14097 + }, + { + "epoch": 0.7876637707070425, + "grad_norm": 0.47278815507888794, + "learning_rate": 0.0006083034513671, + "loss": 0.4169, + "step": 14098 + }, + { + "epoch": 0.7877196413107244, + "grad_norm": 0.5627928376197815, + "learning_rate": 0.0006082754370237562, + "loss": 0.3871, + "step": 14099 + }, + { + "epoch": 0.7877755119144062, + "grad_norm": 0.4217686951160431, + "learning_rate": 0.0006082474226804123, + "loss": 0.3887, + "step": 14100 + }, + { + "epoch": 0.7878313825180882, + "grad_norm": 1.1777300834655762, + "learning_rate": 0.0006082194083370686, + "loss": 0.3974, + "step": 14101 + }, + { + "epoch": 0.78788725312177, + "grad_norm": 0.43014663457870483, + "learning_rate": 0.0006081913939937248, + "loss": 0.3527, + "step": 14102 + }, + { + "epoch": 0.7879431237254518, + "grad_norm": 1.9842658042907715, + "learning_rate": 0.000608163379650381, + "loss": 0.389, + "step": 14103 + }, + { + "epoch": 0.7879989943291337, + "grad_norm": 0.4915000796318054, + "learning_rate": 0.0006081353653070372, + "loss": 0.502, + "step": 14104 + }, + { + "epoch": 0.7880548649328156, + "grad_norm": 0.5770915150642395, + "learning_rate": 0.0006081073509636934, + "loss": 0.4323, + "step": 14105 + }, + { + "epoch": 0.7881107355364975, + "grad_norm": 2.2114713191986084, + "learning_rate": 0.0006080793366203496, + "loss": 0.4956, + "step": 14106 + }, + { + "epoch": 0.7881666061401793, + "grad_norm": 0.5673012733459473, + "learning_rate": 0.0006080513222770058, + "loss": 0.474, + "step": 14107 + }, + { + "epoch": 0.7882224767438613, + "grad_norm": 0.4351632297039032, + "learning_rate": 0.000608023307933662, + "loss": 0.382, + "step": 14108 + }, + { + "epoch": 0.7882783473475431, + "grad_norm": 0.7737995982170105, + "learning_rate": 0.0006079952935903182, + "loss": 0.6671, + "step": 14109 + }, + { + "epoch": 0.7883342179512249, + "grad_norm": 5.937624454498291, + "learning_rate": 0.0006079672792469744, + "loss": 0.417, + "step": 14110 + }, + { + "epoch": 0.7883900885549069, + "grad_norm": 0.630242109298706, + "learning_rate": 0.0006079392649036308, + "loss": 0.5494, + "step": 14111 + }, + { + "epoch": 0.7884459591585887, + "grad_norm": 0.40778419375419617, + "learning_rate": 0.0006079112505602869, + "loss": 0.3663, + "step": 14112 + }, + { + "epoch": 0.7885018297622706, + "grad_norm": 0.4879855513572693, + "learning_rate": 0.0006078832362169431, + "loss": 0.5393, + "step": 14113 + }, + { + "epoch": 0.7885577003659524, + "grad_norm": 0.5817978382110596, + "learning_rate": 0.0006078552218735993, + "loss": 0.4631, + "step": 14114 + }, + { + "epoch": 0.7886135709696344, + "grad_norm": 1.3693115711212158, + "learning_rate": 0.0006078272075302556, + "loss": 0.3657, + "step": 14115 + }, + { + "epoch": 0.7886694415733162, + "grad_norm": 6.336624622344971, + "learning_rate": 0.0006077991931869117, + "loss": 0.332, + "step": 14116 + }, + { + "epoch": 0.788725312176998, + "grad_norm": 5.348274230957031, + "learning_rate": 0.0006077711788435679, + "loss": 0.5088, + "step": 14117 + }, + { + "epoch": 0.78878118278068, + "grad_norm": 0.5691168308258057, + "learning_rate": 0.0006077431645002242, + "loss": 0.4258, + "step": 14118 + }, + { + "epoch": 0.7888370533843618, + "grad_norm": 0.33297592401504517, + "learning_rate": 0.0006077151501568803, + "loss": 0.3955, + "step": 14119 + }, + { + "epoch": 0.7888929239880437, + "grad_norm": 0.454280823469162, + "learning_rate": 0.0006076871358135365, + "loss": 0.5051, + "step": 14120 + }, + { + "epoch": 0.7889487945917255, + "grad_norm": 0.5271598696708679, + "learning_rate": 0.0006076591214701927, + "loss": 0.4096, + "step": 14121 + }, + { + "epoch": 0.7890046651954075, + "grad_norm": 0.5940399765968323, + "learning_rate": 0.000607631107126849, + "loss": 0.4331, + "step": 14122 + }, + { + "epoch": 0.7890605357990893, + "grad_norm": 0.48216718435287476, + "learning_rate": 0.0006076030927835051, + "loss": 0.421, + "step": 14123 + }, + { + "epoch": 0.7891164064027711, + "grad_norm": 0.49709048867225647, + "learning_rate": 0.0006075750784401613, + "loss": 0.5101, + "step": 14124 + }, + { + "epoch": 0.7891722770064531, + "grad_norm": 1.312953233718872, + "learning_rate": 0.0006075470640968176, + "loss": 0.4523, + "step": 14125 + }, + { + "epoch": 0.7892281476101349, + "grad_norm": 1.696645736694336, + "learning_rate": 0.0006075190497534739, + "loss": 0.5783, + "step": 14126 + }, + { + "epoch": 0.7892840182138168, + "grad_norm": 0.41933557391166687, + "learning_rate": 0.00060749103541013, + "loss": 0.4045, + "step": 14127 + }, + { + "epoch": 0.7893398888174987, + "grad_norm": 0.37353917956352234, + "learning_rate": 0.0006074630210667862, + "loss": 0.4414, + "step": 14128 + }, + { + "epoch": 0.7893957594211806, + "grad_norm": 0.46945932507514954, + "learning_rate": 0.0006074350067234425, + "loss": 0.451, + "step": 14129 + }, + { + "epoch": 0.7894516300248624, + "grad_norm": 0.3844485580921173, + "learning_rate": 0.0006074069923800987, + "loss": 0.4509, + "step": 14130 + }, + { + "epoch": 0.7895075006285442, + "grad_norm": 0.4352136552333832, + "learning_rate": 0.0006073789780367548, + "loss": 0.2923, + "step": 14131 + }, + { + "epoch": 0.7895633712322262, + "grad_norm": 0.5711892247200012, + "learning_rate": 0.000607350963693411, + "loss": 0.3717, + "step": 14132 + }, + { + "epoch": 0.789619241835908, + "grad_norm": 0.8045820593833923, + "learning_rate": 0.0006073229493500673, + "loss": 0.4973, + "step": 14133 + }, + { + "epoch": 0.7896751124395899, + "grad_norm": 0.47133395075798035, + "learning_rate": 0.0006072949350067235, + "loss": 0.4157, + "step": 14134 + }, + { + "epoch": 0.7897309830432718, + "grad_norm": 0.40595120191574097, + "learning_rate": 0.0006072669206633796, + "loss": 0.4462, + "step": 14135 + }, + { + "epoch": 0.7897868536469537, + "grad_norm": 1.7646424770355225, + "learning_rate": 0.0006072389063200359, + "loss": 0.3678, + "step": 14136 + }, + { + "epoch": 0.7898427242506355, + "grad_norm": 0.35995393991470337, + "learning_rate": 0.0006072108919766921, + "loss": 0.3354, + "step": 14137 + }, + { + "epoch": 0.7898985948543173, + "grad_norm": 0.40332233905792236, + "learning_rate": 0.0006071828776333483, + "loss": 0.4258, + "step": 14138 + }, + { + "epoch": 0.7899544654579993, + "grad_norm": 0.7899670600891113, + "learning_rate": 0.0006071548632900044, + "loss": 0.5328, + "step": 14139 + }, + { + "epoch": 0.7900103360616811, + "grad_norm": 0.34228503704071045, + "learning_rate": 0.0006071268489466607, + "loss": 0.45, + "step": 14140 + }, + { + "epoch": 0.790066206665363, + "grad_norm": 0.6811981797218323, + "learning_rate": 0.0006070988346033169, + "loss": 0.535, + "step": 14141 + }, + { + "epoch": 0.7901220772690449, + "grad_norm": 0.5869759321212769, + "learning_rate": 0.000607070820259973, + "loss": 0.4024, + "step": 14142 + }, + { + "epoch": 0.7901779478727268, + "grad_norm": 0.5932109355926514, + "learning_rate": 0.0006070428059166294, + "loss": 0.5069, + "step": 14143 + }, + { + "epoch": 0.7902338184764086, + "grad_norm": 0.8369225859642029, + "learning_rate": 0.0006070147915732856, + "loss": 0.6032, + "step": 14144 + }, + { + "epoch": 0.7902896890800906, + "grad_norm": 0.466627836227417, + "learning_rate": 0.0006069867772299418, + "loss": 0.5216, + "step": 14145 + }, + { + "epoch": 0.7903455596837724, + "grad_norm": 0.5076679587364197, + "learning_rate": 0.0006069587628865979, + "loss": 0.4455, + "step": 14146 + }, + { + "epoch": 0.7904014302874542, + "grad_norm": 0.47467395663261414, + "learning_rate": 0.0006069307485432542, + "loss": 0.4254, + "step": 14147 + }, + { + "epoch": 0.7904573008911361, + "grad_norm": 0.6567107439041138, + "learning_rate": 0.0006069027341999104, + "loss": 0.5601, + "step": 14148 + }, + { + "epoch": 0.790513171494818, + "grad_norm": 0.6372073888778687, + "learning_rate": 0.0006068747198565666, + "loss": 0.5439, + "step": 14149 + }, + { + "epoch": 0.7905690420984999, + "grad_norm": 0.6415491104125977, + "learning_rate": 0.0006068467055132228, + "loss": 0.5464, + "step": 14150 + }, + { + "epoch": 0.7906249127021817, + "grad_norm": 2.454070568084717, + "learning_rate": 0.000606818691169879, + "loss": 0.4521, + "step": 14151 + }, + { + "epoch": 0.7906807833058637, + "grad_norm": 1.5228549242019653, + "learning_rate": 0.0006067906768265352, + "loss": 0.4241, + "step": 14152 + }, + { + "epoch": 0.7907366539095455, + "grad_norm": 0.5501197576522827, + "learning_rate": 0.0006067626624831914, + "loss": 0.3928, + "step": 14153 + }, + { + "epoch": 0.7907925245132273, + "grad_norm": 0.3440093398094177, + "learning_rate": 0.0006067346481398476, + "loss": 0.4036, + "step": 14154 + }, + { + "epoch": 0.7908483951169092, + "grad_norm": 0.39159095287323, + "learning_rate": 0.0006067066337965038, + "loss": 0.4617, + "step": 14155 + }, + { + "epoch": 0.7909042657205911, + "grad_norm": 0.9768584370613098, + "learning_rate": 0.00060667861945316, + "loss": 0.4015, + "step": 14156 + }, + { + "epoch": 0.790960136324273, + "grad_norm": 0.4739052951335907, + "learning_rate": 0.0006066506051098163, + "loss": 0.3958, + "step": 14157 + }, + { + "epoch": 0.7910160069279548, + "grad_norm": 0.6268057227134705, + "learning_rate": 0.0006066225907664724, + "loss": 0.5391, + "step": 14158 + }, + { + "epoch": 0.7910718775316368, + "grad_norm": 0.4079590141773224, + "learning_rate": 0.0006065945764231286, + "loss": 0.356, + "step": 14159 + }, + { + "epoch": 0.7911277481353186, + "grad_norm": 0.39615824818611145, + "learning_rate": 0.0006065665620797848, + "loss": 0.3813, + "step": 14160 + }, + { + "epoch": 0.7911836187390004, + "grad_norm": 0.537329375743866, + "learning_rate": 0.0006065385477364411, + "loss": 0.3951, + "step": 14161 + }, + { + "epoch": 0.7912394893426824, + "grad_norm": 0.4694070518016815, + "learning_rate": 0.0006065105333930973, + "loss": 0.4684, + "step": 14162 + }, + { + "epoch": 0.7912953599463642, + "grad_norm": 0.3749573230743408, + "learning_rate": 0.0006064825190497535, + "loss": 0.2991, + "step": 14163 + }, + { + "epoch": 0.7913512305500461, + "grad_norm": 0.6500875949859619, + "learning_rate": 0.0006064545047064098, + "loss": 0.5822, + "step": 14164 + }, + { + "epoch": 0.7914071011537279, + "grad_norm": 0.429640531539917, + "learning_rate": 0.0006064264903630659, + "loss": 0.3668, + "step": 14165 + }, + { + "epoch": 0.7914629717574099, + "grad_norm": 0.5350689888000488, + "learning_rate": 0.0006063984760197221, + "loss": 0.4292, + "step": 14166 + }, + { + "epoch": 0.7915188423610917, + "grad_norm": 0.5179438591003418, + "learning_rate": 0.0006063704616763783, + "loss": 0.5435, + "step": 14167 + }, + { + "epoch": 0.7915747129647736, + "grad_norm": 0.5390998125076294, + "learning_rate": 0.0006063424473330346, + "loss": 0.4209, + "step": 14168 + }, + { + "epoch": 0.7916305835684555, + "grad_norm": 0.4465797543525696, + "learning_rate": 0.0006063144329896907, + "loss": 0.4721, + "step": 14169 + }, + { + "epoch": 0.7916864541721373, + "grad_norm": 0.3735269010066986, + "learning_rate": 0.0006062864186463469, + "loss": 0.406, + "step": 14170 + }, + { + "epoch": 0.7917423247758192, + "grad_norm": 0.7673088312149048, + "learning_rate": 0.0006062584043030031, + "loss": 0.4355, + "step": 14171 + }, + { + "epoch": 0.791798195379501, + "grad_norm": 0.5755178928375244, + "learning_rate": 0.0006062303899596594, + "loss": 0.4427, + "step": 14172 + }, + { + "epoch": 0.791854065983183, + "grad_norm": 0.4520193338394165, + "learning_rate": 0.0006062023756163155, + "loss": 0.4792, + "step": 14173 + }, + { + "epoch": 0.7919099365868648, + "grad_norm": 4.66713285446167, + "learning_rate": 0.0006061743612729717, + "loss": 0.4704, + "step": 14174 + }, + { + "epoch": 0.7919658071905467, + "grad_norm": 0.400450736284256, + "learning_rate": 0.000606146346929628, + "loss": 0.4514, + "step": 14175 + }, + { + "epoch": 0.7920216777942286, + "grad_norm": 0.36648261547088623, + "learning_rate": 0.0006061183325862843, + "loss": 0.3977, + "step": 14176 + }, + { + "epoch": 0.7920775483979104, + "grad_norm": 0.9633039832115173, + "learning_rate": 0.0006060903182429403, + "loss": 0.3426, + "step": 14177 + }, + { + "epoch": 0.7921334190015923, + "grad_norm": 0.6330431699752808, + "learning_rate": 0.0006060623038995966, + "loss": 0.3899, + "step": 14178 + }, + { + "epoch": 0.7921892896052742, + "grad_norm": 0.37327149510383606, + "learning_rate": 0.0006060342895562529, + "loss": 0.47, + "step": 14179 + }, + { + "epoch": 0.7922451602089561, + "grad_norm": 0.45820656418800354, + "learning_rate": 0.0006060062752129091, + "loss": 0.4119, + "step": 14180 + }, + { + "epoch": 0.7923010308126379, + "grad_norm": 0.6864424347877502, + "learning_rate": 0.0006059782608695652, + "loss": 0.6095, + "step": 14181 + }, + { + "epoch": 0.7923569014163198, + "grad_norm": 0.6118533611297607, + "learning_rate": 0.0006059502465262215, + "loss": 0.436, + "step": 14182 + }, + { + "epoch": 0.7924127720200017, + "grad_norm": 0.3686070442199707, + "learning_rate": 0.0006059222321828777, + "loss": 0.4126, + "step": 14183 + }, + { + "epoch": 0.7924686426236836, + "grad_norm": 0.46791622042655945, + "learning_rate": 0.0006058942178395338, + "loss": 0.3717, + "step": 14184 + }, + { + "epoch": 0.7925245132273654, + "grad_norm": 0.5385743379592896, + "learning_rate": 0.00060586620349619, + "loss": 0.593, + "step": 14185 + }, + { + "epoch": 0.7925803838310473, + "grad_norm": 0.4428853988647461, + "learning_rate": 0.0006058381891528463, + "loss": 0.4279, + "step": 14186 + }, + { + "epoch": 0.7926362544347292, + "grad_norm": 0.526531457901001, + "learning_rate": 0.0006058101748095025, + "loss": 0.5262, + "step": 14187 + }, + { + "epoch": 0.792692125038411, + "grad_norm": 1.089352011680603, + "learning_rate": 0.0006057821604661586, + "loss": 0.5472, + "step": 14188 + }, + { + "epoch": 0.7927479956420929, + "grad_norm": 1.9165740013122559, + "learning_rate": 0.0006057541461228149, + "loss": 0.4333, + "step": 14189 + }, + { + "epoch": 0.7928038662457748, + "grad_norm": 0.49396824836730957, + "learning_rate": 0.0006057261317794711, + "loss": 0.6405, + "step": 14190 + }, + { + "epoch": 0.7928597368494567, + "grad_norm": 0.44131940603256226, + "learning_rate": 0.0006056981174361273, + "loss": 0.4582, + "step": 14191 + }, + { + "epoch": 0.7929156074531385, + "grad_norm": 0.9089382290840149, + "learning_rate": 0.0006056701030927834, + "loss": 0.4377, + "step": 14192 + }, + { + "epoch": 0.7929714780568204, + "grad_norm": 0.47419872879981995, + "learning_rate": 0.0006056420887494397, + "loss": 0.4224, + "step": 14193 + }, + { + "epoch": 0.7930273486605023, + "grad_norm": 0.4307591915130615, + "learning_rate": 0.000605614074406096, + "loss": 0.4687, + "step": 14194 + }, + { + "epoch": 0.7930832192641841, + "grad_norm": 0.43245887756347656, + "learning_rate": 0.0006055860600627522, + "loss": 0.4299, + "step": 14195 + }, + { + "epoch": 0.7931390898678661, + "grad_norm": 0.46589285135269165, + "learning_rate": 0.0006055580457194084, + "loss": 0.4391, + "step": 14196 + }, + { + "epoch": 0.7931949604715479, + "grad_norm": 0.4567476809024811, + "learning_rate": 0.0006055300313760646, + "loss": 0.5039, + "step": 14197 + }, + { + "epoch": 0.7932508310752298, + "grad_norm": 0.5976382493972778, + "learning_rate": 0.0006055020170327208, + "loss": 0.4238, + "step": 14198 + }, + { + "epoch": 0.7933067016789116, + "grad_norm": 0.6263655424118042, + "learning_rate": 0.000605474002689377, + "loss": 0.4763, + "step": 14199 + }, + { + "epoch": 0.7933625722825935, + "grad_norm": 0.49854135513305664, + "learning_rate": 0.0006054459883460332, + "loss": 0.3343, + "step": 14200 + }, + { + "epoch": 0.7934184428862754, + "grad_norm": 0.41031014919281006, + "learning_rate": 0.0006054179740026894, + "loss": 0.4363, + "step": 14201 + }, + { + "epoch": 0.7934743134899572, + "grad_norm": 0.3731561601161957, + "learning_rate": 0.0006053899596593456, + "loss": 0.4743, + "step": 14202 + }, + { + "epoch": 0.7935301840936392, + "grad_norm": 0.35983526706695557, + "learning_rate": 0.0006053619453160018, + "loss": 0.4267, + "step": 14203 + }, + { + "epoch": 0.793586054697321, + "grad_norm": 1.0928095579147339, + "learning_rate": 0.000605333930972658, + "loss": 0.4974, + "step": 14204 + }, + { + "epoch": 0.7936419253010029, + "grad_norm": 0.45083215832710266, + "learning_rate": 0.0006053059166293142, + "loss": 0.5169, + "step": 14205 + }, + { + "epoch": 0.7936977959046847, + "grad_norm": 0.37486809492111206, + "learning_rate": 0.0006052779022859704, + "loss": 0.4201, + "step": 14206 + }, + { + "epoch": 0.7937536665083667, + "grad_norm": 1.1276204586029053, + "learning_rate": 0.0006052498879426266, + "loss": 0.4302, + "step": 14207 + }, + { + "epoch": 0.7938095371120485, + "grad_norm": 2.8414838314056396, + "learning_rate": 0.0006052218735992828, + "loss": 0.3867, + "step": 14208 + }, + { + "epoch": 0.7938654077157303, + "grad_norm": 0.43569281697273254, + "learning_rate": 0.000605193859255939, + "loss": 0.3994, + "step": 14209 + }, + { + "epoch": 0.7939212783194123, + "grad_norm": 0.7788124084472656, + "learning_rate": 0.0006051658449125952, + "loss": 0.4676, + "step": 14210 + }, + { + "epoch": 0.7939771489230941, + "grad_norm": 1.24380624294281, + "learning_rate": 0.0006051378305692515, + "loss": 0.4382, + "step": 14211 + }, + { + "epoch": 0.794033019526776, + "grad_norm": 0.6285657286643982, + "learning_rate": 0.0006051098162259077, + "loss": 0.5247, + "step": 14212 + }, + { + "epoch": 0.7940888901304579, + "grad_norm": 0.599338948726654, + "learning_rate": 0.0006050818018825639, + "loss": 0.4193, + "step": 14213 + }, + { + "epoch": 0.7941447607341398, + "grad_norm": 1.0311521291732788, + "learning_rate": 0.0006050537875392202, + "loss": 0.3461, + "step": 14214 + }, + { + "epoch": 0.7942006313378216, + "grad_norm": 0.5102466940879822, + "learning_rate": 0.0006050257731958763, + "loss": 0.4788, + "step": 14215 + }, + { + "epoch": 0.7942565019415034, + "grad_norm": 0.6010972857475281, + "learning_rate": 0.0006049977588525325, + "loss": 0.6064, + "step": 14216 + }, + { + "epoch": 0.7943123725451854, + "grad_norm": 0.5190659761428833, + "learning_rate": 0.0006049697445091887, + "loss": 0.4355, + "step": 14217 + }, + { + "epoch": 0.7943682431488672, + "grad_norm": 0.5172007083892822, + "learning_rate": 0.000604941730165845, + "loss": 0.3765, + "step": 14218 + }, + { + "epoch": 0.7944241137525491, + "grad_norm": 0.5050421953201294, + "learning_rate": 0.0006049137158225011, + "loss": 0.4534, + "step": 14219 + }, + { + "epoch": 0.794479984356231, + "grad_norm": 2.14225697517395, + "learning_rate": 0.0006048857014791573, + "loss": 0.4448, + "step": 14220 + }, + { + "epoch": 0.7945358549599129, + "grad_norm": 0.6063327193260193, + "learning_rate": 0.0006048576871358136, + "loss": 0.4944, + "step": 14221 + }, + { + "epoch": 0.7945917255635947, + "grad_norm": 0.43762287497520447, + "learning_rate": 0.0006048296727924698, + "loss": 0.3754, + "step": 14222 + }, + { + "epoch": 0.7946475961672765, + "grad_norm": 0.49397358298301697, + "learning_rate": 0.0006048016584491259, + "loss": 0.4299, + "step": 14223 + }, + { + "epoch": 0.7947034667709585, + "grad_norm": 2.2023189067840576, + "learning_rate": 0.0006047736441057821, + "loss": 0.4522, + "step": 14224 + }, + { + "epoch": 0.7947593373746403, + "grad_norm": 0.352195680141449, + "learning_rate": 0.0006047456297624384, + "loss": 0.4344, + "step": 14225 + }, + { + "epoch": 0.7948152079783222, + "grad_norm": 1.3073042631149292, + "learning_rate": 0.0006047176154190945, + "loss": 0.5662, + "step": 14226 + }, + { + "epoch": 0.7948710785820041, + "grad_norm": 0.9064087867736816, + "learning_rate": 0.0006046896010757507, + "loss": 0.5041, + "step": 14227 + }, + { + "epoch": 0.794926949185686, + "grad_norm": 0.9618933796882629, + "learning_rate": 0.0006046615867324071, + "loss": 0.5394, + "step": 14228 + }, + { + "epoch": 0.7949828197893678, + "grad_norm": 0.494211345911026, + "learning_rate": 0.0006046335723890633, + "loss": 0.4857, + "step": 14229 + }, + { + "epoch": 0.7950386903930498, + "grad_norm": 0.6401739120483398, + "learning_rate": 0.0006046055580457194, + "loss": 0.4484, + "step": 14230 + }, + { + "epoch": 0.7950945609967316, + "grad_norm": 0.42550283670425415, + "learning_rate": 0.0006045775437023756, + "loss": 0.3038, + "step": 14231 + }, + { + "epoch": 0.7951504316004134, + "grad_norm": 0.5636771321296692, + "learning_rate": 0.0006045495293590319, + "loss": 0.4304, + "step": 14232 + }, + { + "epoch": 0.7952063022040953, + "grad_norm": 0.56814044713974, + "learning_rate": 0.0006045215150156881, + "loss": 0.5119, + "step": 14233 + }, + { + "epoch": 0.7952621728077772, + "grad_norm": 0.808536171913147, + "learning_rate": 0.0006044935006723442, + "loss": 0.6043, + "step": 14234 + }, + { + "epoch": 0.7953180434114591, + "grad_norm": 1.0086287260055542, + "learning_rate": 0.0006044654863290004, + "loss": 0.4137, + "step": 14235 + }, + { + "epoch": 0.7953739140151409, + "grad_norm": 0.5745257139205933, + "learning_rate": 0.0006044374719856567, + "loss": 0.4481, + "step": 14236 + }, + { + "epoch": 0.7954297846188229, + "grad_norm": 0.5450400710105896, + "learning_rate": 0.0006044094576423129, + "loss": 0.4902, + "step": 14237 + }, + { + "epoch": 0.7954856552225047, + "grad_norm": 0.7397739887237549, + "learning_rate": 0.000604381443298969, + "loss": 0.535, + "step": 14238 + }, + { + "epoch": 0.7955415258261865, + "grad_norm": 0.4094224274158478, + "learning_rate": 0.0006043534289556253, + "loss": 0.3749, + "step": 14239 + }, + { + "epoch": 0.7955973964298684, + "grad_norm": 0.43468838930130005, + "learning_rate": 0.0006043254146122815, + "loss": 0.4103, + "step": 14240 + }, + { + "epoch": 0.7956532670335503, + "grad_norm": 0.8325844407081604, + "learning_rate": 0.0006042974002689377, + "loss": 0.4473, + "step": 14241 + }, + { + "epoch": 0.7957091376372322, + "grad_norm": 0.4020022749900818, + "learning_rate": 0.0006042693859255938, + "loss": 0.4276, + "step": 14242 + }, + { + "epoch": 0.795765008240914, + "grad_norm": 0.5418186187744141, + "learning_rate": 0.0006042413715822501, + "loss": 0.5116, + "step": 14243 + }, + { + "epoch": 0.795820878844596, + "grad_norm": 0.650407075881958, + "learning_rate": 0.0006042133572389064, + "loss": 0.4384, + "step": 14244 + }, + { + "epoch": 0.7958767494482778, + "grad_norm": 0.505726158618927, + "learning_rate": 0.0006041853428955626, + "loss": 0.4006, + "step": 14245 + }, + { + "epoch": 0.7959326200519596, + "grad_norm": 0.49521297216415405, + "learning_rate": 0.0006041573285522188, + "loss": 0.4545, + "step": 14246 + }, + { + "epoch": 0.7959884906556416, + "grad_norm": 0.49289610981941223, + "learning_rate": 0.000604129314208875, + "loss": 0.4055, + "step": 14247 + }, + { + "epoch": 0.7960443612593234, + "grad_norm": 0.9135069847106934, + "learning_rate": 0.0006041012998655312, + "loss": 0.5513, + "step": 14248 + }, + { + "epoch": 0.7961002318630053, + "grad_norm": 0.5536870956420898, + "learning_rate": 0.0006040732855221873, + "loss": 0.3803, + "step": 14249 + }, + { + "epoch": 0.7961561024666871, + "grad_norm": 0.5520544648170471, + "learning_rate": 0.0006040452711788436, + "loss": 0.4488, + "step": 14250 + }, + { + "epoch": 0.7962119730703691, + "grad_norm": 0.46383801102638245, + "learning_rate": 0.0006040172568354998, + "loss": 0.4597, + "step": 14251 + }, + { + "epoch": 0.7962678436740509, + "grad_norm": 0.3435573875904083, + "learning_rate": 0.000603989242492156, + "loss": 0.4335, + "step": 14252 + }, + { + "epoch": 0.7963237142777327, + "grad_norm": 0.41778698563575745, + "learning_rate": 0.0006039612281488122, + "loss": 0.4623, + "step": 14253 + }, + { + "epoch": 0.7963795848814147, + "grad_norm": 0.5490212440490723, + "learning_rate": 0.0006039332138054684, + "loss": 0.4156, + "step": 14254 + }, + { + "epoch": 0.7964354554850965, + "grad_norm": 0.44405221939086914, + "learning_rate": 0.0006039051994621246, + "loss": 0.515, + "step": 14255 + }, + { + "epoch": 0.7964913260887784, + "grad_norm": 0.6299079060554504, + "learning_rate": 0.0006038771851187808, + "loss": 0.4432, + "step": 14256 + }, + { + "epoch": 0.7965471966924602, + "grad_norm": 1.1195448637008667, + "learning_rate": 0.000603849170775437, + "loss": 0.4108, + "step": 14257 + }, + { + "epoch": 0.7966030672961422, + "grad_norm": 0.4629983603954315, + "learning_rate": 0.0006038211564320932, + "loss": 0.3419, + "step": 14258 + }, + { + "epoch": 0.796658937899824, + "grad_norm": 0.3534344732761383, + "learning_rate": 0.0006037931420887494, + "loss": 0.3545, + "step": 14259 + }, + { + "epoch": 0.7967148085035058, + "grad_norm": 0.45007169246673584, + "learning_rate": 0.0006037651277454058, + "loss": 0.373, + "step": 14260 + }, + { + "epoch": 0.7967706791071878, + "grad_norm": 0.5019956827163696, + "learning_rate": 0.0006037371134020619, + "loss": 0.4271, + "step": 14261 + }, + { + "epoch": 0.7968265497108696, + "grad_norm": 0.4126591682434082, + "learning_rate": 0.0006037090990587181, + "loss": 0.3567, + "step": 14262 + }, + { + "epoch": 0.7968824203145515, + "grad_norm": 0.7321553230285645, + "learning_rate": 0.0006036810847153743, + "loss": 0.4915, + "step": 14263 + }, + { + "epoch": 0.7969382909182333, + "grad_norm": 0.693905234336853, + "learning_rate": 0.0006036530703720306, + "loss": 0.439, + "step": 14264 + }, + { + "epoch": 0.7969941615219153, + "grad_norm": 0.4907208979129791, + "learning_rate": 0.0006036250560286867, + "loss": 0.4989, + "step": 14265 + }, + { + "epoch": 0.7970500321255971, + "grad_norm": 0.41151976585388184, + "learning_rate": 0.0006035970416853429, + "loss": 0.442, + "step": 14266 + }, + { + "epoch": 0.797105902729279, + "grad_norm": 0.39752379059791565, + "learning_rate": 0.0006035690273419992, + "loss": 0.4678, + "step": 14267 + }, + { + "epoch": 0.7971617733329609, + "grad_norm": 0.539659321308136, + "learning_rate": 0.0006035410129986553, + "loss": 0.4083, + "step": 14268 + }, + { + "epoch": 0.7972176439366427, + "grad_norm": 0.7537956833839417, + "learning_rate": 0.0006035129986553115, + "loss": 0.5023, + "step": 14269 + }, + { + "epoch": 0.7972735145403246, + "grad_norm": 0.7522028088569641, + "learning_rate": 0.0006034849843119677, + "loss": 0.4126, + "step": 14270 + }, + { + "epoch": 0.7973293851440065, + "grad_norm": 0.9616773724555969, + "learning_rate": 0.000603456969968624, + "loss": 0.3545, + "step": 14271 + }, + { + "epoch": 0.7973852557476884, + "grad_norm": 0.542099118232727, + "learning_rate": 0.0006034289556252801, + "loss": 0.4396, + "step": 14272 + }, + { + "epoch": 0.7974411263513702, + "grad_norm": 0.4309632182121277, + "learning_rate": 0.0006034009412819363, + "loss": 0.3815, + "step": 14273 + }, + { + "epoch": 0.797496996955052, + "grad_norm": 0.6237158179283142, + "learning_rate": 0.0006033729269385925, + "loss": 0.4785, + "step": 14274 + }, + { + "epoch": 0.797552867558734, + "grad_norm": 0.4139920175075531, + "learning_rate": 0.0006033449125952488, + "loss": 0.4125, + "step": 14275 + }, + { + "epoch": 0.7976087381624158, + "grad_norm": 0.42430832982063293, + "learning_rate": 0.0006033168982519049, + "loss": 0.5052, + "step": 14276 + }, + { + "epoch": 0.7976646087660977, + "grad_norm": 0.37687772512435913, + "learning_rate": 0.0006032888839085611, + "loss": 0.447, + "step": 14277 + }, + { + "epoch": 0.7977204793697796, + "grad_norm": 0.41586410999298096, + "learning_rate": 0.0006032608695652175, + "loss": 0.4095, + "step": 14278 + }, + { + "epoch": 0.7977763499734615, + "grad_norm": 0.45413070917129517, + "learning_rate": 0.0006032328552218737, + "loss": 0.378, + "step": 14279 + }, + { + "epoch": 0.7978322205771433, + "grad_norm": 0.4401607811450958, + "learning_rate": 0.0006032048408785298, + "loss": 0.4563, + "step": 14280 + }, + { + "epoch": 0.7978880911808252, + "grad_norm": 0.31400495767593384, + "learning_rate": 0.000603176826535186, + "loss": 0.3956, + "step": 14281 + }, + { + "epoch": 0.7979439617845071, + "grad_norm": 0.9002634882926941, + "learning_rate": 0.0006031488121918423, + "loss": 0.3802, + "step": 14282 + }, + { + "epoch": 0.797999832388189, + "grad_norm": 1.460768461227417, + "learning_rate": 0.0006031207978484985, + "loss": 0.3738, + "step": 14283 + }, + { + "epoch": 0.7980557029918708, + "grad_norm": 0.5005025267601013, + "learning_rate": 0.0006030927835051546, + "loss": 0.5041, + "step": 14284 + }, + { + "epoch": 0.7981115735955527, + "grad_norm": 0.45310792326927185, + "learning_rate": 0.0006030647691618109, + "loss": 0.4475, + "step": 14285 + }, + { + "epoch": 0.7981674441992346, + "grad_norm": 0.6895287036895752, + "learning_rate": 0.0006030367548184671, + "loss": 0.4076, + "step": 14286 + }, + { + "epoch": 0.7982233148029164, + "grad_norm": 0.6701183319091797, + "learning_rate": 0.0006030087404751233, + "loss": 0.434, + "step": 14287 + }, + { + "epoch": 0.7982791854065984, + "grad_norm": 0.4004763066768646, + "learning_rate": 0.0006029807261317794, + "loss": 0.3807, + "step": 14288 + }, + { + "epoch": 0.7983350560102802, + "grad_norm": 0.7773367166519165, + "learning_rate": 0.0006029527117884357, + "loss": 0.4348, + "step": 14289 + }, + { + "epoch": 0.798390926613962, + "grad_norm": 0.42177459597587585, + "learning_rate": 0.0006029246974450919, + "loss": 0.46, + "step": 14290 + }, + { + "epoch": 0.7984467972176439, + "grad_norm": 0.45709362626075745, + "learning_rate": 0.000602896683101748, + "loss": 0.4173, + "step": 14291 + }, + { + "epoch": 0.7985026678213258, + "grad_norm": 0.5874821543693542, + "learning_rate": 0.0006028686687584043, + "loss": 0.3552, + "step": 14292 + }, + { + "epoch": 0.7985585384250077, + "grad_norm": 0.48274821043014526, + "learning_rate": 0.0006028406544150605, + "loss": 0.4498, + "step": 14293 + }, + { + "epoch": 0.7986144090286895, + "grad_norm": 0.37187835574150085, + "learning_rate": 0.0006028126400717168, + "loss": 0.3615, + "step": 14294 + }, + { + "epoch": 0.7986702796323715, + "grad_norm": 0.6189165115356445, + "learning_rate": 0.0006027846257283729, + "loss": 0.6194, + "step": 14295 + }, + { + "epoch": 0.7987261502360533, + "grad_norm": 1.7074635028839111, + "learning_rate": 0.0006027566113850292, + "loss": 0.4686, + "step": 14296 + }, + { + "epoch": 0.7987820208397352, + "grad_norm": 0.3784674108028412, + "learning_rate": 0.0006027285970416854, + "loss": 0.4675, + "step": 14297 + }, + { + "epoch": 0.798837891443417, + "grad_norm": 0.6024179458618164, + "learning_rate": 0.0006027005826983416, + "loss": 0.4388, + "step": 14298 + }, + { + "epoch": 0.7988937620470989, + "grad_norm": 0.5115712285041809, + "learning_rate": 0.0006026725683549978, + "loss": 0.4826, + "step": 14299 + }, + { + "epoch": 0.7989496326507808, + "grad_norm": 0.5198671221733093, + "learning_rate": 0.000602644554011654, + "loss": 0.4672, + "step": 14300 + }, + { + "epoch": 0.7990055032544626, + "grad_norm": 0.6027021408081055, + "learning_rate": 0.0006026165396683102, + "loss": 0.4158, + "step": 14301 + }, + { + "epoch": 0.7990613738581446, + "grad_norm": 0.627221405506134, + "learning_rate": 0.0006025885253249664, + "loss": 0.3947, + "step": 14302 + }, + { + "epoch": 0.7991172444618264, + "grad_norm": 0.5583813786506653, + "learning_rate": 0.0006025605109816226, + "loss": 0.3053, + "step": 14303 + }, + { + "epoch": 0.7991731150655083, + "grad_norm": 1.2131731510162354, + "learning_rate": 0.0006025324966382788, + "loss": 0.5994, + "step": 14304 + }, + { + "epoch": 0.7992289856691902, + "grad_norm": 0.4211658537387848, + "learning_rate": 0.000602504482294935, + "loss": 0.3744, + "step": 14305 + }, + { + "epoch": 0.799284856272872, + "grad_norm": 1.7770990133285522, + "learning_rate": 0.0006024764679515913, + "loss": 0.3799, + "step": 14306 + }, + { + "epoch": 0.7993407268765539, + "grad_norm": 0.6938738822937012, + "learning_rate": 0.0006024484536082474, + "loss": 0.4232, + "step": 14307 + }, + { + "epoch": 0.7993965974802357, + "grad_norm": 0.5907264947891235, + "learning_rate": 0.0006024204392649036, + "loss": 0.4693, + "step": 14308 + }, + { + "epoch": 0.7994524680839177, + "grad_norm": 0.565606415271759, + "learning_rate": 0.0006023924249215598, + "loss": 0.4583, + "step": 14309 + }, + { + "epoch": 0.7995083386875995, + "grad_norm": 0.40532201528549194, + "learning_rate": 0.000602364410578216, + "loss": 0.5325, + "step": 14310 + }, + { + "epoch": 0.7995642092912814, + "grad_norm": 0.5176790356636047, + "learning_rate": 0.0006023363962348723, + "loss": 0.435, + "step": 14311 + }, + { + "epoch": 0.7996200798949633, + "grad_norm": 0.7050508260726929, + "learning_rate": 0.0006023083818915285, + "loss": 0.5204, + "step": 14312 + }, + { + "epoch": 0.7996759504986451, + "grad_norm": 0.5082020163536072, + "learning_rate": 0.0006022803675481847, + "loss": 0.4775, + "step": 14313 + }, + { + "epoch": 0.799731821102327, + "grad_norm": 0.6015868782997131, + "learning_rate": 0.0006022523532048409, + "loss": 0.4248, + "step": 14314 + }, + { + "epoch": 0.7997876917060088, + "grad_norm": 0.37193918228149414, + "learning_rate": 0.0006022243388614971, + "loss": 0.3837, + "step": 14315 + }, + { + "epoch": 0.7998435623096908, + "grad_norm": 0.5029812455177307, + "learning_rate": 0.0006021963245181533, + "loss": 0.7433, + "step": 14316 + }, + { + "epoch": 0.7998994329133726, + "grad_norm": 1.9355747699737549, + "learning_rate": 0.0006021683101748096, + "loss": 0.3883, + "step": 14317 + }, + { + "epoch": 0.7999553035170545, + "grad_norm": 0.669674813747406, + "learning_rate": 0.0006021402958314657, + "loss": 0.442, + "step": 14318 + }, + { + "epoch": 0.8000111741207364, + "grad_norm": 0.6001459360122681, + "learning_rate": 0.0006021122814881219, + "loss": 0.4739, + "step": 14319 + }, + { + "epoch": 0.8000670447244183, + "grad_norm": 0.44072040915489197, + "learning_rate": 0.0006020842671447781, + "loss": 0.4258, + "step": 14320 + }, + { + "epoch": 0.8001229153281001, + "grad_norm": 3.668551206588745, + "learning_rate": 0.0006020562528014344, + "loss": 0.3383, + "step": 14321 + }, + { + "epoch": 0.800178785931782, + "grad_norm": 0.44906577467918396, + "learning_rate": 0.0006020282384580905, + "loss": 0.4104, + "step": 14322 + }, + { + "epoch": 0.8002346565354639, + "grad_norm": 0.48193925619125366, + "learning_rate": 0.0006020002241147467, + "loss": 0.394, + "step": 14323 + }, + { + "epoch": 0.8002905271391457, + "grad_norm": 1.0879452228546143, + "learning_rate": 0.000601972209771403, + "loss": 0.4516, + "step": 14324 + }, + { + "epoch": 0.8003463977428276, + "grad_norm": 0.48540040850639343, + "learning_rate": 0.0006019441954280592, + "loss": 0.3897, + "step": 14325 + }, + { + "epoch": 0.8004022683465095, + "grad_norm": 0.5986642241477966, + "learning_rate": 0.0006019161810847153, + "loss": 0.3865, + "step": 14326 + }, + { + "epoch": 0.8004581389501914, + "grad_norm": 1.160934329032898, + "learning_rate": 0.0006018881667413715, + "loss": 0.614, + "step": 14327 + }, + { + "epoch": 0.8005140095538732, + "grad_norm": 0.573562741279602, + "learning_rate": 0.0006018601523980279, + "loss": 0.3547, + "step": 14328 + }, + { + "epoch": 0.8005698801575551, + "grad_norm": 0.478268563747406, + "learning_rate": 0.0006018321380546841, + "loss": 0.408, + "step": 14329 + }, + { + "epoch": 0.800625750761237, + "grad_norm": 0.7219024300575256, + "learning_rate": 0.0006018041237113402, + "loss": 0.5844, + "step": 14330 + }, + { + "epoch": 0.8006816213649188, + "grad_norm": 0.5865113139152527, + "learning_rate": 0.0006017761093679965, + "loss": 0.4649, + "step": 14331 + }, + { + "epoch": 0.8007374919686007, + "grad_norm": 0.3761843144893646, + "learning_rate": 0.0006017480950246527, + "loss": 0.4238, + "step": 14332 + }, + { + "epoch": 0.8007933625722826, + "grad_norm": 0.35936203598976135, + "learning_rate": 0.0006017200806813088, + "loss": 0.3357, + "step": 14333 + }, + { + "epoch": 0.8008492331759645, + "grad_norm": 1.6409215927124023, + "learning_rate": 0.000601692066337965, + "loss": 0.4496, + "step": 14334 + }, + { + "epoch": 0.8009051037796463, + "grad_norm": 0.508169412612915, + "learning_rate": 0.0006016640519946213, + "loss": 0.432, + "step": 14335 + }, + { + "epoch": 0.8009609743833283, + "grad_norm": 0.4148443639278412, + "learning_rate": 0.0006016360376512775, + "loss": 0.3484, + "step": 14336 + }, + { + "epoch": 0.8010168449870101, + "grad_norm": 0.5572779178619385, + "learning_rate": 0.0006016080233079336, + "loss": 0.4177, + "step": 14337 + }, + { + "epoch": 0.8010727155906919, + "grad_norm": 0.7898359298706055, + "learning_rate": 0.0006015800089645899, + "loss": 0.4574, + "step": 14338 + }, + { + "epoch": 0.8011285861943739, + "grad_norm": 2.3313217163085938, + "learning_rate": 0.0006015519946212461, + "loss": 0.4807, + "step": 14339 + }, + { + "epoch": 0.8011844567980557, + "grad_norm": 1.0399423837661743, + "learning_rate": 0.0006015239802779023, + "loss": 0.4422, + "step": 14340 + }, + { + "epoch": 0.8012403274017376, + "grad_norm": 0.4995505213737488, + "learning_rate": 0.0006014959659345584, + "loss": 0.455, + "step": 14341 + }, + { + "epoch": 0.8012961980054194, + "grad_norm": 2.666106939315796, + "learning_rate": 0.0006014679515912147, + "loss": 0.4509, + "step": 14342 + }, + { + "epoch": 0.8013520686091014, + "grad_norm": 0.7991487383842468, + "learning_rate": 0.000601439937247871, + "loss": 0.4263, + "step": 14343 + }, + { + "epoch": 0.8014079392127832, + "grad_norm": 1.7705636024475098, + "learning_rate": 0.0006014119229045272, + "loss": 0.4609, + "step": 14344 + }, + { + "epoch": 0.801463809816465, + "grad_norm": 0.43014785647392273, + "learning_rate": 0.0006013839085611833, + "loss": 0.4458, + "step": 14345 + }, + { + "epoch": 0.801519680420147, + "grad_norm": 0.7202755808830261, + "learning_rate": 0.0006013558942178396, + "loss": 0.4195, + "step": 14346 + }, + { + "epoch": 0.8015755510238288, + "grad_norm": 0.4124045968055725, + "learning_rate": 0.0006013278798744958, + "loss": 0.4817, + "step": 14347 + }, + { + "epoch": 0.8016314216275107, + "grad_norm": 0.8255464434623718, + "learning_rate": 0.000601299865531152, + "loss": 0.5609, + "step": 14348 + }, + { + "epoch": 0.8016872922311925, + "grad_norm": 2.04376220703125, + "learning_rate": 0.0006012718511878082, + "loss": 0.4598, + "step": 14349 + }, + { + "epoch": 0.8017431628348745, + "grad_norm": 0.40146514773368835, + "learning_rate": 0.0006012438368444644, + "loss": 0.5087, + "step": 14350 + }, + { + "epoch": 0.8017990334385563, + "grad_norm": 0.6607266664505005, + "learning_rate": 0.0006012158225011206, + "loss": 0.4332, + "step": 14351 + }, + { + "epoch": 0.8018549040422381, + "grad_norm": 0.8543229699134827, + "learning_rate": 0.0006011878081577767, + "loss": 0.608, + "step": 14352 + }, + { + "epoch": 0.8019107746459201, + "grad_norm": 0.4086782932281494, + "learning_rate": 0.000601159793814433, + "loss": 0.4211, + "step": 14353 + }, + { + "epoch": 0.8019666452496019, + "grad_norm": 0.5647338628768921, + "learning_rate": 0.0006011317794710892, + "loss": 0.4644, + "step": 14354 + }, + { + "epoch": 0.8020225158532838, + "grad_norm": 0.41400569677352905, + "learning_rate": 0.0006011037651277454, + "loss": 0.497, + "step": 14355 + }, + { + "epoch": 0.8020783864569657, + "grad_norm": 0.5169005393981934, + "learning_rate": 0.0006010757507844016, + "loss": 0.5007, + "step": 14356 + }, + { + "epoch": 0.8021342570606476, + "grad_norm": 0.5213384032249451, + "learning_rate": 0.0006010477364410578, + "loss": 0.4946, + "step": 14357 + }, + { + "epoch": 0.8021901276643294, + "grad_norm": 1.1150003671646118, + "learning_rate": 0.000601019722097714, + "loss": 0.5352, + "step": 14358 + }, + { + "epoch": 0.8022459982680112, + "grad_norm": 0.5426008701324463, + "learning_rate": 0.0006009917077543702, + "loss": 0.5494, + "step": 14359 + }, + { + "epoch": 0.8023018688716932, + "grad_norm": 0.9401735067367554, + "learning_rate": 0.0006009636934110264, + "loss": 0.4386, + "step": 14360 + }, + { + "epoch": 0.802357739475375, + "grad_norm": 0.43458566069602966, + "learning_rate": 0.0006009356790676827, + "loss": 0.3563, + "step": 14361 + }, + { + "epoch": 0.8024136100790569, + "grad_norm": 0.38755956292152405, + "learning_rate": 0.0006009076647243389, + "loss": 0.3528, + "step": 14362 + }, + { + "epoch": 0.8024694806827388, + "grad_norm": 0.4425658583641052, + "learning_rate": 0.0006008796503809952, + "loss": 0.4489, + "step": 14363 + }, + { + "epoch": 0.8025253512864207, + "grad_norm": 0.9833215475082397, + "learning_rate": 0.0006008516360376513, + "loss": 0.4719, + "step": 14364 + }, + { + "epoch": 0.8025812218901025, + "grad_norm": 0.49974316358566284, + "learning_rate": 0.0006008236216943075, + "loss": 0.506, + "step": 14365 + }, + { + "epoch": 0.8026370924937843, + "grad_norm": 0.8458643555641174, + "learning_rate": 0.0006007956073509637, + "loss": 0.4604, + "step": 14366 + }, + { + "epoch": 0.8026929630974663, + "grad_norm": 0.38542312383651733, + "learning_rate": 0.00060076759300762, + "loss": 0.4404, + "step": 14367 + }, + { + "epoch": 0.8027488337011481, + "grad_norm": 1.0454856157302856, + "learning_rate": 0.0006007395786642761, + "loss": 0.4639, + "step": 14368 + }, + { + "epoch": 0.80280470430483, + "grad_norm": 0.521971583366394, + "learning_rate": 0.0006007115643209323, + "loss": 0.4722, + "step": 14369 + }, + { + "epoch": 0.8028605749085119, + "grad_norm": 1.6614502668380737, + "learning_rate": 0.0006006835499775886, + "loss": 0.4265, + "step": 14370 + }, + { + "epoch": 0.8029164455121938, + "grad_norm": 0.4563922584056854, + "learning_rate": 0.0006006555356342448, + "loss": 0.4616, + "step": 14371 + }, + { + "epoch": 0.8029723161158756, + "grad_norm": 0.6055052280426025, + "learning_rate": 0.0006006275212909009, + "loss": 0.4441, + "step": 14372 + }, + { + "epoch": 0.8030281867195576, + "grad_norm": 1.38217294216156, + "learning_rate": 0.0006005995069475571, + "loss": 0.4159, + "step": 14373 + }, + { + "epoch": 0.8030840573232394, + "grad_norm": 0.5870234966278076, + "learning_rate": 0.0006005714926042134, + "loss": 0.3438, + "step": 14374 + }, + { + "epoch": 0.8031399279269212, + "grad_norm": 0.6687636375427246, + "learning_rate": 0.0006005434782608695, + "loss": 0.459, + "step": 14375 + }, + { + "epoch": 0.8031957985306031, + "grad_norm": 0.7308162450790405, + "learning_rate": 0.0006005154639175257, + "loss": 0.539, + "step": 14376 + }, + { + "epoch": 0.803251669134285, + "grad_norm": 0.6683363914489746, + "learning_rate": 0.0006004874495741819, + "loss": 0.5389, + "step": 14377 + }, + { + "epoch": 0.8033075397379669, + "grad_norm": 0.4786112904548645, + "learning_rate": 0.0006004594352308383, + "loss": 0.4783, + "step": 14378 + }, + { + "epoch": 0.8033634103416487, + "grad_norm": 0.40479567646980286, + "learning_rate": 0.0006004314208874944, + "loss": 0.448, + "step": 14379 + }, + { + "epoch": 0.8034192809453307, + "grad_norm": 0.5012235045433044, + "learning_rate": 0.0006004034065441506, + "loss": 0.555, + "step": 14380 + }, + { + "epoch": 0.8034751515490125, + "grad_norm": 0.6483058333396912, + "learning_rate": 0.0006003753922008069, + "loss": 0.5398, + "step": 14381 + }, + { + "epoch": 0.8035310221526943, + "grad_norm": 0.6137022972106934, + "learning_rate": 0.0006003473778574631, + "loss": 0.4383, + "step": 14382 + }, + { + "epoch": 0.8035868927563762, + "grad_norm": 0.4393083155155182, + "learning_rate": 0.0006003193635141192, + "loss": 0.4932, + "step": 14383 + }, + { + "epoch": 0.8036427633600581, + "grad_norm": 0.48147454857826233, + "learning_rate": 0.0006002913491707754, + "loss": 0.4237, + "step": 14384 + }, + { + "epoch": 0.80369863396374, + "grad_norm": 0.8195571899414062, + "learning_rate": 0.0006002633348274317, + "loss": 0.393, + "step": 14385 + }, + { + "epoch": 0.8037545045674218, + "grad_norm": 0.5223113894462585, + "learning_rate": 0.0006002353204840879, + "loss": 0.3597, + "step": 14386 + }, + { + "epoch": 0.8038103751711038, + "grad_norm": 0.46745628118515015, + "learning_rate": 0.000600207306140744, + "loss": 0.4324, + "step": 14387 + }, + { + "epoch": 0.8038662457747856, + "grad_norm": 0.5996381044387817, + "learning_rate": 0.0006001792917974003, + "loss": 0.384, + "step": 14388 + }, + { + "epoch": 0.8039221163784674, + "grad_norm": 0.5614415407180786, + "learning_rate": 0.0006001512774540565, + "loss": 0.5871, + "step": 14389 + }, + { + "epoch": 0.8039779869821494, + "grad_norm": 0.39475390315055847, + "learning_rate": 0.0006001232631107127, + "loss": 0.4082, + "step": 14390 + }, + { + "epoch": 0.8040338575858312, + "grad_norm": 0.47284331917762756, + "learning_rate": 0.0006000952487673688, + "loss": 0.4438, + "step": 14391 + }, + { + "epoch": 0.8040897281895131, + "grad_norm": 0.9276883006095886, + "learning_rate": 0.0006000672344240251, + "loss": 0.5271, + "step": 14392 + }, + { + "epoch": 0.8041455987931949, + "grad_norm": 0.8103336095809937, + "learning_rate": 0.0006000392200806813, + "loss": 0.4755, + "step": 14393 + }, + { + "epoch": 0.8042014693968769, + "grad_norm": 0.576869547367096, + "learning_rate": 0.0006000112057373376, + "loss": 0.4704, + "step": 14394 + }, + { + "epoch": 0.8042573400005587, + "grad_norm": 0.5160900950431824, + "learning_rate": 0.0005999831913939938, + "loss": 0.3672, + "step": 14395 + }, + { + "epoch": 0.8043132106042405, + "grad_norm": 25.0606689453125, + "learning_rate": 0.00059995517705065, + "loss": 0.4968, + "step": 14396 + }, + { + "epoch": 0.8043690812079225, + "grad_norm": 0.4565785825252533, + "learning_rate": 0.0005999271627073062, + "loss": 0.4622, + "step": 14397 + }, + { + "epoch": 0.8044249518116043, + "grad_norm": 0.5580068826675415, + "learning_rate": 0.0005998991483639623, + "loss": 0.3442, + "step": 14398 + }, + { + "epoch": 0.8044808224152862, + "grad_norm": 1.0893858671188354, + "learning_rate": 0.0005998711340206186, + "loss": 0.5986, + "step": 14399 + }, + { + "epoch": 0.804536693018968, + "grad_norm": 0.5648353099822998, + "learning_rate": 0.0005998431196772748, + "loss": 0.429, + "step": 14400 + }, + { + "epoch": 0.80459256362265, + "grad_norm": 0.5408662557601929, + "learning_rate": 0.000599815105333931, + "loss": 0.4834, + "step": 14401 + }, + { + "epoch": 0.8046484342263318, + "grad_norm": 0.35870298743247986, + "learning_rate": 0.0005997870909905872, + "loss": 0.379, + "step": 14402 + }, + { + "epoch": 0.8047043048300137, + "grad_norm": 0.4495464265346527, + "learning_rate": 0.0005997590766472434, + "loss": 0.3557, + "step": 14403 + }, + { + "epoch": 0.8047601754336956, + "grad_norm": 0.6588979959487915, + "learning_rate": 0.0005997310623038996, + "loss": 0.5012, + "step": 14404 + }, + { + "epoch": 0.8048160460373774, + "grad_norm": 0.4167587161064148, + "learning_rate": 0.0005997030479605558, + "loss": 0.504, + "step": 14405 + }, + { + "epoch": 0.8048719166410593, + "grad_norm": 0.5871827602386475, + "learning_rate": 0.000599675033617212, + "loss": 0.4123, + "step": 14406 + }, + { + "epoch": 0.8049277872447412, + "grad_norm": 0.4439360797405243, + "learning_rate": 0.0005996470192738682, + "loss": 0.4065, + "step": 14407 + }, + { + "epoch": 0.8049836578484231, + "grad_norm": 0.3892970085144043, + "learning_rate": 0.0005996190049305244, + "loss": 0.3252, + "step": 14408 + }, + { + "epoch": 0.8050395284521049, + "grad_norm": 0.54803866147995, + "learning_rate": 0.0005995909905871807, + "loss": 0.3592, + "step": 14409 + }, + { + "epoch": 0.8050953990557868, + "grad_norm": 0.38755080103874207, + "learning_rate": 0.0005995629762438368, + "loss": 0.3168, + "step": 14410 + }, + { + "epoch": 0.8051512696594687, + "grad_norm": 0.373807430267334, + "learning_rate": 0.000599534961900493, + "loss": 0.4121, + "step": 14411 + }, + { + "epoch": 0.8052071402631505, + "grad_norm": 0.5018885731697083, + "learning_rate": 0.0005995069475571493, + "loss": 0.4572, + "step": 14412 + }, + { + "epoch": 0.8052630108668324, + "grad_norm": 0.3999671936035156, + "learning_rate": 0.0005994789332138056, + "loss": 0.4267, + "step": 14413 + }, + { + "epoch": 0.8053188814705143, + "grad_norm": 0.556525707244873, + "learning_rate": 0.0005994509188704617, + "loss": 0.4186, + "step": 14414 + }, + { + "epoch": 0.8053747520741962, + "grad_norm": 0.8708495497703552, + "learning_rate": 0.0005994229045271179, + "loss": 0.609, + "step": 14415 + }, + { + "epoch": 0.805430622677878, + "grad_norm": 0.8844956159591675, + "learning_rate": 0.0005993948901837741, + "loss": 0.4187, + "step": 14416 + }, + { + "epoch": 0.8054864932815599, + "grad_norm": 0.694723904132843, + "learning_rate": 0.0005993668758404303, + "loss": 0.5546, + "step": 14417 + }, + { + "epoch": 0.8055423638852418, + "grad_norm": 0.5065619945526123, + "learning_rate": 0.0005993388614970865, + "loss": 0.5103, + "step": 14418 + }, + { + "epoch": 0.8055982344889236, + "grad_norm": 0.5505942702293396, + "learning_rate": 0.0005993108471537427, + "loss": 0.4235, + "step": 14419 + }, + { + "epoch": 0.8056541050926055, + "grad_norm": 0.5296710729598999, + "learning_rate": 0.000599282832810399, + "loss": 0.4342, + "step": 14420 + }, + { + "epoch": 0.8057099756962874, + "grad_norm": 0.8569499254226685, + "learning_rate": 0.0005992548184670551, + "loss": 0.4561, + "step": 14421 + }, + { + "epoch": 0.8057658462999693, + "grad_norm": 0.6119349002838135, + "learning_rate": 0.0005992268041237113, + "loss": 0.5757, + "step": 14422 + }, + { + "epoch": 0.8058217169036511, + "grad_norm": 0.5866276025772095, + "learning_rate": 0.0005991987897803675, + "loss": 0.5039, + "step": 14423 + }, + { + "epoch": 0.8058775875073331, + "grad_norm": 0.490354984998703, + "learning_rate": 0.0005991707754370238, + "loss": 0.5129, + "step": 14424 + }, + { + "epoch": 0.8059334581110149, + "grad_norm": 0.4625978171825409, + "learning_rate": 0.0005991427610936799, + "loss": 0.3781, + "step": 14425 + }, + { + "epoch": 0.8059893287146968, + "grad_norm": 13.543953895568848, + "learning_rate": 0.0005991147467503361, + "loss": 0.4502, + "step": 14426 + }, + { + "epoch": 0.8060451993183786, + "grad_norm": 0.5315485596656799, + "learning_rate": 0.0005990867324069925, + "loss": 0.4377, + "step": 14427 + }, + { + "epoch": 0.8061010699220605, + "grad_norm": 0.9049780368804932, + "learning_rate": 0.0005990587180636487, + "loss": 0.3549, + "step": 14428 + }, + { + "epoch": 0.8061569405257424, + "grad_norm": 0.6597656011581421, + "learning_rate": 0.0005990307037203048, + "loss": 0.5802, + "step": 14429 + }, + { + "epoch": 0.8062128111294242, + "grad_norm": 0.43280965089797974, + "learning_rate": 0.000599002689376961, + "loss": 0.458, + "step": 14430 + }, + { + "epoch": 0.8062686817331062, + "grad_norm": 0.4221387207508087, + "learning_rate": 0.0005989746750336173, + "loss": 0.543, + "step": 14431 + }, + { + "epoch": 0.806324552336788, + "grad_norm": 0.4778335690498352, + "learning_rate": 0.0005989466606902735, + "loss": 0.4803, + "step": 14432 + }, + { + "epoch": 0.8063804229404699, + "grad_norm": 0.4785165786743164, + "learning_rate": 0.0005989186463469296, + "loss": 0.6011, + "step": 14433 + }, + { + "epoch": 0.8064362935441517, + "grad_norm": 0.5255254507064819, + "learning_rate": 0.0005988906320035859, + "loss": 0.6146, + "step": 14434 + }, + { + "epoch": 0.8064921641478336, + "grad_norm": 0.7326489686965942, + "learning_rate": 0.0005988626176602421, + "loss": 0.5447, + "step": 14435 + }, + { + "epoch": 0.8065480347515155, + "grad_norm": 0.7772290110588074, + "learning_rate": 0.0005988346033168983, + "loss": 0.3602, + "step": 14436 + }, + { + "epoch": 0.8066039053551973, + "grad_norm": 0.6663302779197693, + "learning_rate": 0.0005988065889735544, + "loss": 0.3953, + "step": 14437 + }, + { + "epoch": 0.8066597759588793, + "grad_norm": 0.45718348026275635, + "learning_rate": 0.0005987785746302107, + "loss": 0.4989, + "step": 14438 + }, + { + "epoch": 0.8067156465625611, + "grad_norm": 3.7716691493988037, + "learning_rate": 0.0005987505602868669, + "loss": 0.3413, + "step": 14439 + }, + { + "epoch": 0.806771517166243, + "grad_norm": 0.461769700050354, + "learning_rate": 0.000598722545943523, + "loss": 0.4672, + "step": 14440 + }, + { + "epoch": 0.8068273877699249, + "grad_norm": 0.8097370266914368, + "learning_rate": 0.0005986945316001793, + "loss": 0.5055, + "step": 14441 + }, + { + "epoch": 0.8068832583736067, + "grad_norm": 0.49743103981018066, + "learning_rate": 0.0005986665172568355, + "loss": 0.4101, + "step": 14442 + }, + { + "epoch": 0.8069391289772886, + "grad_norm": 1.0691065788269043, + "learning_rate": 0.0005986385029134917, + "loss": 0.4578, + "step": 14443 + }, + { + "epoch": 0.8069949995809704, + "grad_norm": 0.6816514134407043, + "learning_rate": 0.0005986104885701478, + "loss": 0.4989, + "step": 14444 + }, + { + "epoch": 0.8070508701846524, + "grad_norm": 0.5720085501670837, + "learning_rate": 0.0005985824742268042, + "loss": 0.4643, + "step": 14445 + }, + { + "epoch": 0.8071067407883342, + "grad_norm": 0.6279597282409668, + "learning_rate": 0.0005985544598834604, + "loss": 0.4407, + "step": 14446 + }, + { + "epoch": 0.8071626113920161, + "grad_norm": 0.5180456638336182, + "learning_rate": 0.0005985264455401166, + "loss": 0.4618, + "step": 14447 + }, + { + "epoch": 0.807218481995698, + "grad_norm": 0.8800845146179199, + "learning_rate": 0.0005984984311967727, + "loss": 0.4738, + "step": 14448 + }, + { + "epoch": 0.8072743525993799, + "grad_norm": 0.4969417452812195, + "learning_rate": 0.000598470416853429, + "loss": 0.4676, + "step": 14449 + }, + { + "epoch": 0.8073302232030617, + "grad_norm": 0.7281315326690674, + "learning_rate": 0.0005984424025100852, + "loss": 0.6242, + "step": 14450 + }, + { + "epoch": 0.8073860938067435, + "grad_norm": 0.385092556476593, + "learning_rate": 0.0005984143881667414, + "loss": 0.4856, + "step": 14451 + }, + { + "epoch": 0.8074419644104255, + "grad_norm": 0.5236669182777405, + "learning_rate": 0.0005983863738233976, + "loss": 0.3903, + "step": 14452 + }, + { + "epoch": 0.8074978350141073, + "grad_norm": 0.5887762308120728, + "learning_rate": 0.0005983583594800538, + "loss": 0.5456, + "step": 14453 + }, + { + "epoch": 0.8075537056177892, + "grad_norm": 0.8273738622665405, + "learning_rate": 0.00059833034513671, + "loss": 0.3698, + "step": 14454 + }, + { + "epoch": 0.8076095762214711, + "grad_norm": 1.0151456594467163, + "learning_rate": 0.0005983023307933662, + "loss": 0.5762, + "step": 14455 + }, + { + "epoch": 0.807665446825153, + "grad_norm": 0.6862366199493408, + "learning_rate": 0.0005982743164500224, + "loss": 0.4893, + "step": 14456 + }, + { + "epoch": 0.8077213174288348, + "grad_norm": 0.5636329650878906, + "learning_rate": 0.0005982463021066786, + "loss": 0.4877, + "step": 14457 + }, + { + "epoch": 0.8077771880325167, + "grad_norm": 0.4148663282394409, + "learning_rate": 0.0005982182877633348, + "loss": 0.3658, + "step": 14458 + }, + { + "epoch": 0.8078330586361986, + "grad_norm": 0.5205398201942444, + "learning_rate": 0.000598190273419991, + "loss": 0.4396, + "step": 14459 + }, + { + "epoch": 0.8078889292398804, + "grad_norm": 0.2947588860988617, + "learning_rate": 0.0005981622590766472, + "loss": 0.3934, + "step": 14460 + }, + { + "epoch": 0.8079447998435623, + "grad_norm": 0.7378969192504883, + "learning_rate": 0.0005981342447333034, + "loss": 0.5439, + "step": 14461 + }, + { + "epoch": 0.8080006704472442, + "grad_norm": 0.5909087061882019, + "learning_rate": 0.0005981062303899597, + "loss": 0.4985, + "step": 14462 + }, + { + "epoch": 0.8080565410509261, + "grad_norm": 0.4655504524707794, + "learning_rate": 0.0005980782160466159, + "loss": 0.4756, + "step": 14463 + }, + { + "epoch": 0.8081124116546079, + "grad_norm": 6.450228214263916, + "learning_rate": 0.0005980502017032721, + "loss": 0.4789, + "step": 14464 + }, + { + "epoch": 0.8081682822582898, + "grad_norm": 3.12542986869812, + "learning_rate": 0.0005980221873599283, + "loss": 0.6143, + "step": 14465 + }, + { + "epoch": 0.8082241528619717, + "grad_norm": 0.5703584551811218, + "learning_rate": 0.0005979941730165846, + "loss": 0.4145, + "step": 14466 + }, + { + "epoch": 0.8082800234656535, + "grad_norm": 0.6562813520431519, + "learning_rate": 0.0005979661586732407, + "loss": 0.3901, + "step": 14467 + }, + { + "epoch": 0.8083358940693354, + "grad_norm": 0.7508659362792969, + "learning_rate": 0.0005979381443298969, + "loss": 0.4576, + "step": 14468 + }, + { + "epoch": 0.8083917646730173, + "grad_norm": 0.33519476652145386, + "learning_rate": 0.0005979101299865531, + "loss": 0.3166, + "step": 14469 + }, + { + "epoch": 0.8084476352766992, + "grad_norm": 0.40878719091415405, + "learning_rate": 0.0005978821156432094, + "loss": 0.5629, + "step": 14470 + }, + { + "epoch": 0.808503505880381, + "grad_norm": 0.44766196608543396, + "learning_rate": 0.0005978541012998655, + "loss": 0.4147, + "step": 14471 + }, + { + "epoch": 0.808559376484063, + "grad_norm": 0.4756188988685608, + "learning_rate": 0.0005978260869565217, + "loss": 0.4778, + "step": 14472 + }, + { + "epoch": 0.8086152470877448, + "grad_norm": 0.4110649526119232, + "learning_rate": 0.000597798072613178, + "loss": 0.4399, + "step": 14473 + }, + { + "epoch": 0.8086711176914266, + "grad_norm": 0.48343342542648315, + "learning_rate": 0.0005977700582698342, + "loss": 0.4164, + "step": 14474 + }, + { + "epoch": 0.8087269882951085, + "grad_norm": 1.0843844413757324, + "learning_rate": 0.0005977420439264903, + "loss": 0.4945, + "step": 14475 + }, + { + "epoch": 0.8087828588987904, + "grad_norm": 0.6729439496994019, + "learning_rate": 0.0005977140295831465, + "loss": 0.4134, + "step": 14476 + }, + { + "epoch": 0.8088387295024723, + "grad_norm": 0.7704885005950928, + "learning_rate": 0.0005976860152398029, + "loss": 0.7017, + "step": 14477 + }, + { + "epoch": 0.8088946001061541, + "grad_norm": 0.3407328426837921, + "learning_rate": 0.0005976580008964591, + "loss": 0.4067, + "step": 14478 + }, + { + "epoch": 0.8089504707098361, + "grad_norm": 0.48451781272888184, + "learning_rate": 0.0005976299865531152, + "loss": 0.4555, + "step": 14479 + }, + { + "epoch": 0.8090063413135179, + "grad_norm": 0.6332307457923889, + "learning_rate": 0.0005976019722097715, + "loss": 0.4528, + "step": 14480 + }, + { + "epoch": 0.8090622119171997, + "grad_norm": 0.41055095195770264, + "learning_rate": 0.0005975739578664277, + "loss": 0.4587, + "step": 14481 + }, + { + "epoch": 0.8091180825208817, + "grad_norm": 0.3633083403110504, + "learning_rate": 0.0005975459435230838, + "loss": 0.5069, + "step": 14482 + }, + { + "epoch": 0.8091739531245635, + "grad_norm": 2.0332190990448, + "learning_rate": 0.00059751792917974, + "loss": 0.4892, + "step": 14483 + }, + { + "epoch": 0.8092298237282454, + "grad_norm": 0.4407368004322052, + "learning_rate": 0.0005974899148363963, + "loss": 0.5297, + "step": 14484 + }, + { + "epoch": 0.8092856943319272, + "grad_norm": 0.5266661643981934, + "learning_rate": 0.0005974619004930525, + "loss": 0.4516, + "step": 14485 + }, + { + "epoch": 0.8093415649356092, + "grad_norm": 0.4388616383075714, + "learning_rate": 0.0005974338861497086, + "loss": 0.4477, + "step": 14486 + }, + { + "epoch": 0.809397435539291, + "grad_norm": 0.7078152298927307, + "learning_rate": 0.0005974058718063648, + "loss": 0.4958, + "step": 14487 + }, + { + "epoch": 0.8094533061429728, + "grad_norm": 13.081334114074707, + "learning_rate": 0.0005973778574630211, + "loss": 0.5269, + "step": 14488 + }, + { + "epoch": 0.8095091767466548, + "grad_norm": 0.6103470921516418, + "learning_rate": 0.0005973498431196773, + "loss": 0.4385, + "step": 14489 + }, + { + "epoch": 0.8095650473503366, + "grad_norm": 0.6887192130088806, + "learning_rate": 0.0005973218287763334, + "loss": 0.5708, + "step": 14490 + }, + { + "epoch": 0.8096209179540185, + "grad_norm": 0.5180385112762451, + "learning_rate": 0.0005972938144329897, + "loss": 0.4609, + "step": 14491 + }, + { + "epoch": 0.8096767885577003, + "grad_norm": 0.4850904643535614, + "learning_rate": 0.0005972658000896459, + "loss": 0.4614, + "step": 14492 + }, + { + "epoch": 0.8097326591613823, + "grad_norm": 0.37198033928871155, + "learning_rate": 0.0005972377857463021, + "loss": 0.4874, + "step": 14493 + }, + { + "epoch": 0.8097885297650641, + "grad_norm": 0.624351441860199, + "learning_rate": 0.0005972097714029582, + "loss": 0.4287, + "step": 14494 + }, + { + "epoch": 0.8098444003687459, + "grad_norm": 1.0560650825500488, + "learning_rate": 0.0005971817570596146, + "loss": 0.4939, + "step": 14495 + }, + { + "epoch": 0.8099002709724279, + "grad_norm": 0.575063169002533, + "learning_rate": 0.0005971537427162708, + "loss": 0.4951, + "step": 14496 + }, + { + "epoch": 0.8099561415761097, + "grad_norm": 0.4741264283657074, + "learning_rate": 0.000597125728372927, + "loss": 0.3344, + "step": 14497 + }, + { + "epoch": 0.8100120121797916, + "grad_norm": 2.45522141456604, + "learning_rate": 0.0005970977140295832, + "loss": 0.4646, + "step": 14498 + }, + { + "epoch": 0.8100678827834735, + "grad_norm": 0.4571155309677124, + "learning_rate": 0.0005970696996862394, + "loss": 0.6211, + "step": 14499 + }, + { + "epoch": 0.8101237533871554, + "grad_norm": 0.4708005487918854, + "learning_rate": 0.0005970416853428956, + "loss": 0.3922, + "step": 14500 + }, + { + "epoch": 0.8101237533871554, + "eval_cer": 0.08953334860833798, + "eval_loss": 0.33621639013290405, + "eval_runtime": 56.0765, + "eval_samples_per_second": 80.925, + "eval_steps_per_second": 5.065, + "eval_wer": 0.35557665986983605, + "step": 14500 + }, + { + "epoch": 0.8101796239908372, + "grad_norm": 0.3240039646625519, + "learning_rate": 0.0005970136709995517, + "loss": 0.3471, + "step": 14501 + }, + { + "epoch": 0.810235494594519, + "grad_norm": 2.1206531524658203, + "learning_rate": 0.000596985656656208, + "loss": 0.4037, + "step": 14502 + }, + { + "epoch": 0.810291365198201, + "grad_norm": 0.5933811068534851, + "learning_rate": 0.0005969576423128642, + "loss": 0.4205, + "step": 14503 + }, + { + "epoch": 0.8103472358018828, + "grad_norm": 1.8521788120269775, + "learning_rate": 0.0005969296279695204, + "loss": 0.4167, + "step": 14504 + }, + { + "epoch": 0.8104031064055647, + "grad_norm": 1.1249024868011475, + "learning_rate": 0.0005969016136261766, + "loss": 0.4783, + "step": 14505 + }, + { + "epoch": 0.8104589770092466, + "grad_norm": 1.2872071266174316, + "learning_rate": 0.0005968735992828328, + "loss": 0.5353, + "step": 14506 + }, + { + "epoch": 0.8105148476129285, + "grad_norm": 0.46065330505371094, + "learning_rate": 0.000596845584939489, + "loss": 0.4579, + "step": 14507 + }, + { + "epoch": 0.8105707182166103, + "grad_norm": 4.597671031951904, + "learning_rate": 0.0005968175705961452, + "loss": 0.475, + "step": 14508 + }, + { + "epoch": 0.8106265888202921, + "grad_norm": 0.4780883193016052, + "learning_rate": 0.0005967895562528014, + "loss": 0.4443, + "step": 14509 + }, + { + "epoch": 0.8106824594239741, + "grad_norm": 0.6015709638595581, + "learning_rate": 0.0005967615419094576, + "loss": 0.3721, + "step": 14510 + }, + { + "epoch": 0.8107383300276559, + "grad_norm": 0.4920697808265686, + "learning_rate": 0.0005967335275661138, + "loss": 0.5726, + "step": 14511 + }, + { + "epoch": 0.8107942006313378, + "grad_norm": 0.407473623752594, + "learning_rate": 0.0005967055132227702, + "loss": 0.3627, + "step": 14512 + }, + { + "epoch": 0.8108500712350197, + "grad_norm": 0.3724028468132019, + "learning_rate": 0.0005966774988794263, + "loss": 0.3815, + "step": 14513 + }, + { + "epoch": 0.8109059418387016, + "grad_norm": 0.3963870406150818, + "learning_rate": 0.0005966494845360825, + "loss": 0.4193, + "step": 14514 + }, + { + "epoch": 0.8109618124423834, + "grad_norm": 0.4951949715614319, + "learning_rate": 0.0005966214701927387, + "loss": 0.5352, + "step": 14515 + }, + { + "epoch": 0.8110176830460654, + "grad_norm": 0.49916157126426697, + "learning_rate": 0.000596593455849395, + "loss": 0.515, + "step": 14516 + }, + { + "epoch": 0.8110735536497472, + "grad_norm": 0.31842997670173645, + "learning_rate": 0.0005965654415060511, + "loss": 0.4256, + "step": 14517 + }, + { + "epoch": 0.811129424253429, + "grad_norm": 0.37065038084983826, + "learning_rate": 0.0005965374271627073, + "loss": 0.3444, + "step": 14518 + }, + { + "epoch": 0.8111852948571109, + "grad_norm": 0.7230265736579895, + "learning_rate": 0.0005965094128193636, + "loss": 0.4032, + "step": 14519 + }, + { + "epoch": 0.8112411654607928, + "grad_norm": 0.4636567533016205, + "learning_rate": 0.0005964813984760198, + "loss": 0.4301, + "step": 14520 + }, + { + "epoch": 0.8112970360644747, + "grad_norm": 0.4529419541358948, + "learning_rate": 0.0005964533841326759, + "loss": 0.4219, + "step": 14521 + }, + { + "epoch": 0.8113529066681565, + "grad_norm": 1.081578254699707, + "learning_rate": 0.0005964253697893321, + "loss": 0.4404, + "step": 14522 + }, + { + "epoch": 0.8114087772718385, + "grad_norm": 0.3915141522884369, + "learning_rate": 0.0005963973554459884, + "loss": 0.384, + "step": 14523 + }, + { + "epoch": 0.8114646478755203, + "grad_norm": 0.5197691917419434, + "learning_rate": 0.0005963693411026445, + "loss": 0.3627, + "step": 14524 + }, + { + "epoch": 0.8115205184792021, + "grad_norm": 0.7998167872428894, + "learning_rate": 0.0005963413267593007, + "loss": 0.4813, + "step": 14525 + }, + { + "epoch": 0.811576389082884, + "grad_norm": 19.762100219726562, + "learning_rate": 0.0005963133124159569, + "loss": 0.5033, + "step": 14526 + }, + { + "epoch": 0.8116322596865659, + "grad_norm": 0.44800734519958496, + "learning_rate": 0.0005962852980726133, + "loss": 0.3731, + "step": 14527 + }, + { + "epoch": 0.8116881302902478, + "grad_norm": 0.5171303749084473, + "learning_rate": 0.0005962572837292693, + "loss": 0.4459, + "step": 14528 + }, + { + "epoch": 0.8117440008939296, + "grad_norm": 0.4245359003543854, + "learning_rate": 0.0005962292693859256, + "loss": 0.4521, + "step": 14529 + }, + { + "epoch": 0.8117998714976116, + "grad_norm": 0.5299193859100342, + "learning_rate": 0.0005962012550425819, + "loss": 0.4514, + "step": 14530 + }, + { + "epoch": 0.8118557421012934, + "grad_norm": 0.43041786551475525, + "learning_rate": 0.0005961732406992381, + "loss": 0.4981, + "step": 14531 + }, + { + "epoch": 0.8119116127049752, + "grad_norm": 0.8274111747741699, + "learning_rate": 0.0005961452263558942, + "loss": 0.3802, + "step": 14532 + }, + { + "epoch": 0.8119674833086572, + "grad_norm": 6.941343784332275, + "learning_rate": 0.0005961172120125504, + "loss": 0.368, + "step": 14533 + }, + { + "epoch": 0.812023353912339, + "grad_norm": 1.2757103443145752, + "learning_rate": 0.0005960891976692067, + "loss": 0.3789, + "step": 14534 + }, + { + "epoch": 0.8120792245160209, + "grad_norm": 0.6592363119125366, + "learning_rate": 0.0005960611833258629, + "loss": 0.3469, + "step": 14535 + }, + { + "epoch": 0.8121350951197027, + "grad_norm": 0.48031774163246155, + "learning_rate": 0.000596033168982519, + "loss": 0.472, + "step": 14536 + }, + { + "epoch": 0.8121909657233847, + "grad_norm": 0.3662095069885254, + "learning_rate": 0.0005960051546391753, + "loss": 0.3865, + "step": 14537 + }, + { + "epoch": 0.8122468363270665, + "grad_norm": 0.5852396488189697, + "learning_rate": 0.0005959771402958315, + "loss": 0.44, + "step": 14538 + }, + { + "epoch": 0.8123027069307484, + "grad_norm": 0.4378916621208191, + "learning_rate": 0.0005959491259524877, + "loss": 0.3672, + "step": 14539 + }, + { + "epoch": 0.8123585775344303, + "grad_norm": 0.39777305722236633, + "learning_rate": 0.0005959211116091438, + "loss": 0.3957, + "step": 14540 + }, + { + "epoch": 0.8124144481381121, + "grad_norm": 0.4404458999633789, + "learning_rate": 0.0005958930972658001, + "loss": 0.5138, + "step": 14541 + }, + { + "epoch": 0.812470318741794, + "grad_norm": 0.4953605532646179, + "learning_rate": 0.0005958650829224563, + "loss": 0.381, + "step": 14542 + }, + { + "epoch": 0.8125261893454758, + "grad_norm": 0.5294392704963684, + "learning_rate": 0.0005958370685791125, + "loss": 0.4764, + "step": 14543 + }, + { + "epoch": 0.8125820599491578, + "grad_norm": 0.38293761014938354, + "learning_rate": 0.0005958090542357687, + "loss": 0.3732, + "step": 14544 + }, + { + "epoch": 0.8126379305528396, + "grad_norm": 0.6140350103378296, + "learning_rate": 0.000595781039892425, + "loss": 0.4749, + "step": 14545 + }, + { + "epoch": 0.8126938011565215, + "grad_norm": 0.5323939919471741, + "learning_rate": 0.0005957530255490812, + "loss": 0.4651, + "step": 14546 + }, + { + "epoch": 0.8127496717602034, + "grad_norm": 0.35419243574142456, + "learning_rate": 0.0005957250112057373, + "loss": 0.3324, + "step": 14547 + }, + { + "epoch": 0.8128055423638852, + "grad_norm": 0.3886476159095764, + "learning_rate": 0.0005956969968623936, + "loss": 0.4414, + "step": 14548 + }, + { + "epoch": 0.8128614129675671, + "grad_norm": 0.9681345224380493, + "learning_rate": 0.0005956689825190498, + "loss": 0.3126, + "step": 14549 + }, + { + "epoch": 0.812917283571249, + "grad_norm": 0.5247407555580139, + "learning_rate": 0.000595640968175706, + "loss": 0.3457, + "step": 14550 + }, + { + "epoch": 0.8129731541749309, + "grad_norm": 0.505911111831665, + "learning_rate": 0.0005956129538323622, + "loss": 0.506, + "step": 14551 + }, + { + "epoch": 0.8130290247786127, + "grad_norm": 0.5212088227272034, + "learning_rate": 0.0005955849394890184, + "loss": 0.5815, + "step": 14552 + }, + { + "epoch": 0.8130848953822946, + "grad_norm": 0.49986374378204346, + "learning_rate": 0.0005955569251456746, + "loss": 0.511, + "step": 14553 + }, + { + "epoch": 0.8131407659859765, + "grad_norm": 0.4330500066280365, + "learning_rate": 0.0005955289108023308, + "loss": 0.3343, + "step": 14554 + }, + { + "epoch": 0.8131966365896584, + "grad_norm": 0.6062993407249451, + "learning_rate": 0.000595500896458987, + "loss": 0.4383, + "step": 14555 + }, + { + "epoch": 0.8132525071933402, + "grad_norm": 0.5184759497642517, + "learning_rate": 0.0005954728821156432, + "loss": 0.4152, + "step": 14556 + }, + { + "epoch": 0.8133083777970221, + "grad_norm": 0.5403144955635071, + "learning_rate": 0.0005954448677722994, + "loss": 0.3808, + "step": 14557 + }, + { + "epoch": 0.813364248400704, + "grad_norm": 0.8069104552268982, + "learning_rate": 0.0005954168534289556, + "loss": 0.3796, + "step": 14558 + }, + { + "epoch": 0.8134201190043858, + "grad_norm": 0.6485400199890137, + "learning_rate": 0.0005953888390856118, + "loss": 0.5065, + "step": 14559 + }, + { + "epoch": 0.8134759896080677, + "grad_norm": 0.4083744287490845, + "learning_rate": 0.000595360824742268, + "loss": 0.4859, + "step": 14560 + }, + { + "epoch": 0.8135318602117496, + "grad_norm": 0.701560378074646, + "learning_rate": 0.0005953328103989242, + "loss": 0.5536, + "step": 14561 + }, + { + "epoch": 0.8135877308154315, + "grad_norm": 0.4325343072414398, + "learning_rate": 0.0005953047960555806, + "loss": 0.4923, + "step": 14562 + }, + { + "epoch": 0.8136436014191133, + "grad_norm": 0.8437586426734924, + "learning_rate": 0.0005952767817122367, + "loss": 0.4095, + "step": 14563 + }, + { + "epoch": 0.8136994720227952, + "grad_norm": 0.7071218490600586, + "learning_rate": 0.0005952487673688929, + "loss": 0.4293, + "step": 14564 + }, + { + "epoch": 0.8137553426264771, + "grad_norm": 0.5765233635902405, + "learning_rate": 0.0005952207530255491, + "loss": 0.562, + "step": 14565 + }, + { + "epoch": 0.8138112132301589, + "grad_norm": 0.41047582030296326, + "learning_rate": 0.0005951927386822053, + "loss": 0.484, + "step": 14566 + }, + { + "epoch": 0.8138670838338409, + "grad_norm": 0.3944631814956665, + "learning_rate": 0.0005951647243388615, + "loss": 0.425, + "step": 14567 + }, + { + "epoch": 0.8139229544375227, + "grad_norm": 0.7500340342521667, + "learning_rate": 0.0005951367099955177, + "loss": 0.4199, + "step": 14568 + }, + { + "epoch": 0.8139788250412046, + "grad_norm": 0.5037770867347717, + "learning_rate": 0.000595108695652174, + "loss": 0.5573, + "step": 14569 + }, + { + "epoch": 0.8140346956448864, + "grad_norm": 0.38967448472976685, + "learning_rate": 0.0005950806813088301, + "loss": 0.3894, + "step": 14570 + }, + { + "epoch": 0.8140905662485683, + "grad_norm": 0.3997982442378998, + "learning_rate": 0.0005950526669654863, + "loss": 0.36, + "step": 14571 + }, + { + "epoch": 0.8141464368522502, + "grad_norm": 0.4902600646018982, + "learning_rate": 0.0005950246526221425, + "loss": 0.4864, + "step": 14572 + }, + { + "epoch": 0.814202307455932, + "grad_norm": 0.4280129373073578, + "learning_rate": 0.0005949966382787988, + "loss": 0.4824, + "step": 14573 + }, + { + "epoch": 0.814258178059614, + "grad_norm": 0.5490202903747559, + "learning_rate": 0.0005949686239354549, + "loss": 0.4851, + "step": 14574 + }, + { + "epoch": 0.8143140486632958, + "grad_norm": 1.5203070640563965, + "learning_rate": 0.0005949406095921111, + "loss": 0.4655, + "step": 14575 + }, + { + "epoch": 0.8143699192669777, + "grad_norm": 0.5937241911888123, + "learning_rate": 0.0005949125952487674, + "loss": 0.3513, + "step": 14576 + }, + { + "epoch": 0.8144257898706595, + "grad_norm": 0.6565181612968445, + "learning_rate": 0.0005948845809054236, + "loss": 0.4742, + "step": 14577 + }, + { + "epoch": 0.8144816604743415, + "grad_norm": 0.505731463432312, + "learning_rate": 0.0005948565665620797, + "loss": 0.5031, + "step": 14578 + }, + { + "epoch": 0.8145375310780233, + "grad_norm": 0.49092283844947815, + "learning_rate": 0.000594828552218736, + "loss": 0.5153, + "step": 14579 + }, + { + "epoch": 0.8145934016817051, + "grad_norm": 0.36384040117263794, + "learning_rate": 0.0005948005378753923, + "loss": 0.4763, + "step": 14580 + }, + { + "epoch": 0.8146492722853871, + "grad_norm": 0.4305999279022217, + "learning_rate": 0.0005947725235320485, + "loss": 0.3732, + "step": 14581 + }, + { + "epoch": 0.8147051428890689, + "grad_norm": 0.5499919056892395, + "learning_rate": 0.0005947445091887046, + "loss": 0.3303, + "step": 14582 + }, + { + "epoch": 0.8147610134927508, + "grad_norm": 0.5560824871063232, + "learning_rate": 0.0005947164948453609, + "loss": 0.5287, + "step": 14583 + }, + { + "epoch": 0.8148168840964327, + "grad_norm": 3.855597972869873, + "learning_rate": 0.0005946884805020171, + "loss": 0.3966, + "step": 14584 + }, + { + "epoch": 0.8148727547001146, + "grad_norm": 0.435097873210907, + "learning_rate": 0.0005946604661586733, + "loss": 0.5759, + "step": 14585 + }, + { + "epoch": 0.8149286253037964, + "grad_norm": 2.232847213745117, + "learning_rate": 0.0005946324518153294, + "loss": 0.3307, + "step": 14586 + }, + { + "epoch": 0.8149844959074782, + "grad_norm": 0.47184571623802185, + "learning_rate": 0.0005946044374719857, + "loss": 0.4472, + "step": 14587 + }, + { + "epoch": 0.8150403665111602, + "grad_norm": 0.4529981315135956, + "learning_rate": 0.0005945764231286419, + "loss": 0.4539, + "step": 14588 + }, + { + "epoch": 0.815096237114842, + "grad_norm": 0.6511526107788086, + "learning_rate": 0.000594548408785298, + "loss": 0.4735, + "step": 14589 + }, + { + "epoch": 0.8151521077185239, + "grad_norm": 0.3900088965892792, + "learning_rate": 0.0005945203944419542, + "loss": 0.3852, + "step": 14590 + }, + { + "epoch": 0.8152079783222058, + "grad_norm": 0.5800966024398804, + "learning_rate": 0.0005944923800986105, + "loss": 0.4934, + "step": 14591 + }, + { + "epoch": 0.8152638489258877, + "grad_norm": 0.7066552639007568, + "learning_rate": 0.0005944643657552667, + "loss": 0.6976, + "step": 14592 + }, + { + "epoch": 0.8153197195295695, + "grad_norm": 0.609285295009613, + "learning_rate": 0.0005944363514119228, + "loss": 0.4599, + "step": 14593 + }, + { + "epoch": 0.8153755901332513, + "grad_norm": 0.6313188076019287, + "learning_rate": 0.0005944083370685791, + "loss": 0.5945, + "step": 14594 + }, + { + "epoch": 0.8154314607369333, + "grad_norm": 0.6619378328323364, + "learning_rate": 0.0005943803227252354, + "loss": 0.4043, + "step": 14595 + }, + { + "epoch": 0.8154873313406151, + "grad_norm": 0.6312944293022156, + "learning_rate": 0.0005943523083818916, + "loss": 0.4554, + "step": 14596 + }, + { + "epoch": 0.815543201944297, + "grad_norm": 0.45417413115501404, + "learning_rate": 0.0005943242940385477, + "loss": 0.5421, + "step": 14597 + }, + { + "epoch": 0.8155990725479789, + "grad_norm": 0.4214315712451935, + "learning_rate": 0.000594296279695204, + "loss": 0.3786, + "step": 14598 + }, + { + "epoch": 0.8156549431516608, + "grad_norm": 0.4296373724937439, + "learning_rate": 0.0005942682653518602, + "loss": 0.4813, + "step": 14599 + }, + { + "epoch": 0.8157108137553426, + "grad_norm": 0.5404746532440186, + "learning_rate": 0.0005942402510085164, + "loss": 0.4628, + "step": 14600 + }, + { + "epoch": 0.8157666843590246, + "grad_norm": 0.5518640875816345, + "learning_rate": 0.0005942122366651726, + "loss": 0.4125, + "step": 14601 + }, + { + "epoch": 0.8158225549627064, + "grad_norm": 0.5071491003036499, + "learning_rate": 0.0005941842223218288, + "loss": 0.4771, + "step": 14602 + }, + { + "epoch": 0.8158784255663882, + "grad_norm": 0.9693794846534729, + "learning_rate": 0.000594156207978485, + "loss": 0.5128, + "step": 14603 + }, + { + "epoch": 0.8159342961700701, + "grad_norm": 1.832018256187439, + "learning_rate": 0.0005941281936351412, + "loss": 0.3785, + "step": 14604 + }, + { + "epoch": 0.815990166773752, + "grad_norm": 0.4662865102291107, + "learning_rate": 0.0005941001792917974, + "loss": 0.5008, + "step": 14605 + }, + { + "epoch": 0.8160460373774339, + "grad_norm": 0.7825879454612732, + "learning_rate": 0.0005940721649484536, + "loss": 0.484, + "step": 14606 + }, + { + "epoch": 0.8161019079811157, + "grad_norm": 0.4147186577320099, + "learning_rate": 0.0005940441506051098, + "loss": 0.4168, + "step": 14607 + }, + { + "epoch": 0.8161577785847977, + "grad_norm": 0.4593331813812256, + "learning_rate": 0.000594016136261766, + "loss": 0.4802, + "step": 14608 + }, + { + "epoch": 0.8162136491884795, + "grad_norm": 0.8570152521133423, + "learning_rate": 0.0005939881219184222, + "loss": 0.5032, + "step": 14609 + }, + { + "epoch": 0.8162695197921613, + "grad_norm": 0.679020881652832, + "learning_rate": 0.0005939601075750784, + "loss": 0.4371, + "step": 14610 + }, + { + "epoch": 0.8163253903958432, + "grad_norm": 0.5632423162460327, + "learning_rate": 0.0005939320932317346, + "loss": 0.4178, + "step": 14611 + }, + { + "epoch": 0.8163812609995251, + "grad_norm": 0.6812543272972107, + "learning_rate": 0.0005939040788883909, + "loss": 0.4817, + "step": 14612 + }, + { + "epoch": 0.816437131603207, + "grad_norm": 0.4432829022407532, + "learning_rate": 0.0005938760645450471, + "loss": 0.3845, + "step": 14613 + }, + { + "epoch": 0.8164930022068888, + "grad_norm": 0.2985711097717285, + "learning_rate": 0.0005938480502017033, + "loss": 0.3899, + "step": 14614 + }, + { + "epoch": 0.8165488728105708, + "grad_norm": 0.5619593858718872, + "learning_rate": 0.0005938200358583596, + "loss": 0.4204, + "step": 14615 + }, + { + "epoch": 0.8166047434142526, + "grad_norm": 0.6847980618476868, + "learning_rate": 0.0005937920215150157, + "loss": 0.4609, + "step": 14616 + }, + { + "epoch": 0.8166606140179344, + "grad_norm": 1.1029422283172607, + "learning_rate": 0.0005937640071716719, + "loss": 0.4212, + "step": 14617 + }, + { + "epoch": 0.8167164846216164, + "grad_norm": 0.4136258363723755, + "learning_rate": 0.0005937359928283281, + "loss": 0.4225, + "step": 14618 + }, + { + "epoch": 0.8167723552252982, + "grad_norm": 0.4861225485801697, + "learning_rate": 0.0005937079784849844, + "loss": 0.5889, + "step": 14619 + }, + { + "epoch": 0.8168282258289801, + "grad_norm": 0.5168942809104919, + "learning_rate": 0.0005936799641416405, + "loss": 0.3837, + "step": 14620 + }, + { + "epoch": 0.8168840964326619, + "grad_norm": 0.996921181678772, + "learning_rate": 0.0005936519497982967, + "loss": 0.4391, + "step": 14621 + }, + { + "epoch": 0.8169399670363439, + "grad_norm": 0.3705909252166748, + "learning_rate": 0.000593623935454953, + "loss": 0.4018, + "step": 14622 + }, + { + "epoch": 0.8169958376400257, + "grad_norm": 2.26311993598938, + "learning_rate": 0.0005935959211116092, + "loss": 0.4095, + "step": 14623 + }, + { + "epoch": 0.8170517082437075, + "grad_norm": 0.4775497317314148, + "learning_rate": 0.0005935679067682653, + "loss": 0.4063, + "step": 14624 + }, + { + "epoch": 0.8171075788473895, + "grad_norm": 1.2979120016098022, + "learning_rate": 0.0005935398924249215, + "loss": 0.4253, + "step": 14625 + }, + { + "epoch": 0.8171634494510713, + "grad_norm": 0.4674529731273651, + "learning_rate": 0.0005935118780815778, + "loss": 0.399, + "step": 14626 + }, + { + "epoch": 0.8172193200547532, + "grad_norm": 3.3399994373321533, + "learning_rate": 0.000593483863738234, + "loss": 0.4219, + "step": 14627 + }, + { + "epoch": 0.817275190658435, + "grad_norm": 0.559684693813324, + "learning_rate": 0.0005934558493948901, + "loss": 0.4925, + "step": 14628 + }, + { + "epoch": 0.817331061262117, + "grad_norm": 0.6635794043540955, + "learning_rate": 0.0005934278350515464, + "loss": 0.5677, + "step": 14629 + }, + { + "epoch": 0.8173869318657988, + "grad_norm": 2.2276525497436523, + "learning_rate": 0.0005933998207082027, + "loss": 0.4042, + "step": 14630 + }, + { + "epoch": 0.8174428024694806, + "grad_norm": 0.5148231983184814, + "learning_rate": 0.0005933718063648588, + "loss": 0.4259, + "step": 14631 + }, + { + "epoch": 0.8174986730731626, + "grad_norm": 0.6032353043556213, + "learning_rate": 0.000593343792021515, + "loss": 0.3832, + "step": 14632 + }, + { + "epoch": 0.8175545436768444, + "grad_norm": 0.45490074157714844, + "learning_rate": 0.0005933157776781713, + "loss": 0.378, + "step": 14633 + }, + { + "epoch": 0.8176104142805263, + "grad_norm": 0.5690136551856995, + "learning_rate": 0.0005932877633348275, + "loss": 0.4496, + "step": 14634 + }, + { + "epoch": 0.8176662848842082, + "grad_norm": 0.2970782518386841, + "learning_rate": 0.0005932597489914836, + "loss": 0.3524, + "step": 14635 + }, + { + "epoch": 0.8177221554878901, + "grad_norm": 0.5259761810302734, + "learning_rate": 0.0005932317346481398, + "loss": 0.3381, + "step": 14636 + }, + { + "epoch": 0.8177780260915719, + "grad_norm": 0.8885822892189026, + "learning_rate": 0.0005932037203047961, + "loss": 0.4749, + "step": 14637 + }, + { + "epoch": 0.8178338966952537, + "grad_norm": 0.4375818371772766, + "learning_rate": 0.0005931757059614523, + "loss": 0.409, + "step": 14638 + }, + { + "epoch": 0.8178897672989357, + "grad_norm": 0.5786561965942383, + "learning_rate": 0.0005931476916181084, + "loss": 0.5339, + "step": 14639 + }, + { + "epoch": 0.8179456379026175, + "grad_norm": 0.4494718313217163, + "learning_rate": 0.0005931196772747647, + "loss": 0.4805, + "step": 14640 + }, + { + "epoch": 0.8180015085062994, + "grad_norm": 0.4517863392829895, + "learning_rate": 0.0005930916629314209, + "loss": 0.3966, + "step": 14641 + }, + { + "epoch": 0.8180573791099813, + "grad_norm": 1.306328535079956, + "learning_rate": 0.0005930636485880771, + "loss": 0.6359, + "step": 14642 + }, + { + "epoch": 0.8181132497136632, + "grad_norm": 0.5240480303764343, + "learning_rate": 0.0005930356342447332, + "loss": 0.4562, + "step": 14643 + }, + { + "epoch": 0.818169120317345, + "grad_norm": 2.183229923248291, + "learning_rate": 0.0005930076199013895, + "loss": 0.5008, + "step": 14644 + }, + { + "epoch": 0.8182249909210269, + "grad_norm": 0.6741382479667664, + "learning_rate": 0.0005929796055580458, + "loss": 0.4614, + "step": 14645 + }, + { + "epoch": 0.8182808615247088, + "grad_norm": 0.5587326288223267, + "learning_rate": 0.000592951591214702, + "loss": 0.5388, + "step": 14646 + }, + { + "epoch": 0.8183367321283906, + "grad_norm": 0.6870570778846741, + "learning_rate": 0.0005929235768713582, + "loss": 0.4874, + "step": 14647 + }, + { + "epoch": 0.8183926027320725, + "grad_norm": 0.5876993536949158, + "learning_rate": 0.0005928955625280144, + "loss": 0.6124, + "step": 14648 + }, + { + "epoch": 0.8184484733357544, + "grad_norm": 0.5123735666275024, + "learning_rate": 0.0005928675481846706, + "loss": 0.3598, + "step": 14649 + }, + { + "epoch": 0.8185043439394363, + "grad_norm": 4.796054840087891, + "learning_rate": 0.0005928395338413267, + "loss": 0.3289, + "step": 14650 + }, + { + "epoch": 0.8185602145431181, + "grad_norm": 1.3197240829467773, + "learning_rate": 0.000592811519497983, + "loss": 0.4604, + "step": 14651 + }, + { + "epoch": 0.8186160851468001, + "grad_norm": 2.0584349632263184, + "learning_rate": 0.0005927835051546392, + "loss": 0.4959, + "step": 14652 + }, + { + "epoch": 0.8186719557504819, + "grad_norm": 0.6058129668235779, + "learning_rate": 0.0005927554908112954, + "loss": 0.4433, + "step": 14653 + }, + { + "epoch": 0.8187278263541637, + "grad_norm": 0.5466804504394531, + "learning_rate": 0.0005927274764679516, + "loss": 0.5072, + "step": 14654 + }, + { + "epoch": 0.8187836969578456, + "grad_norm": 0.5261551141738892, + "learning_rate": 0.0005926994621246078, + "loss": 0.48, + "step": 14655 + }, + { + "epoch": 0.8188395675615275, + "grad_norm": 0.5265743732452393, + "learning_rate": 0.000592671447781264, + "loss": 0.3689, + "step": 14656 + }, + { + "epoch": 0.8188954381652094, + "grad_norm": 0.45533987879753113, + "learning_rate": 0.0005926434334379202, + "loss": 0.4842, + "step": 14657 + }, + { + "epoch": 0.8189513087688912, + "grad_norm": 1.0299392938613892, + "learning_rate": 0.0005926154190945764, + "loss": 0.3974, + "step": 14658 + }, + { + "epoch": 0.8190071793725732, + "grad_norm": 1.0651124715805054, + "learning_rate": 0.0005925874047512326, + "loss": 0.5739, + "step": 14659 + }, + { + "epoch": 0.819063049976255, + "grad_norm": 0.44463062286376953, + "learning_rate": 0.0005925593904078888, + "loss": 0.5013, + "step": 14660 + }, + { + "epoch": 0.8191189205799368, + "grad_norm": 0.581109881401062, + "learning_rate": 0.0005925313760645452, + "loss": 0.5616, + "step": 14661 + }, + { + "epoch": 0.8191747911836187, + "grad_norm": 0.4801687002182007, + "learning_rate": 0.0005925033617212013, + "loss": 0.4153, + "step": 14662 + }, + { + "epoch": 0.8192306617873006, + "grad_norm": 0.4213036000728607, + "learning_rate": 0.0005924753473778575, + "loss": 0.4867, + "step": 14663 + }, + { + "epoch": 0.8192865323909825, + "grad_norm": 0.4270215630531311, + "learning_rate": 0.0005924473330345137, + "loss": 0.3796, + "step": 14664 + }, + { + "epoch": 0.8193424029946643, + "grad_norm": 0.9531970024108887, + "learning_rate": 0.00059241931869117, + "loss": 0.4265, + "step": 14665 + }, + { + "epoch": 0.8193982735983463, + "grad_norm": 0.5900970101356506, + "learning_rate": 0.0005923913043478261, + "loss": 0.4593, + "step": 14666 + }, + { + "epoch": 0.8194541442020281, + "grad_norm": 0.7029227614402771, + "learning_rate": 0.0005923632900044823, + "loss": 0.5299, + "step": 14667 + }, + { + "epoch": 0.81951001480571, + "grad_norm": 0.7078645825386047, + "learning_rate": 0.0005923352756611385, + "loss": 0.5141, + "step": 14668 + }, + { + "epoch": 0.8195658854093919, + "grad_norm": 0.4843161702156067, + "learning_rate": 0.0005923072613177948, + "loss": 0.4967, + "step": 14669 + }, + { + "epoch": 0.8196217560130737, + "grad_norm": 1.1446616649627686, + "learning_rate": 0.0005922792469744509, + "loss": 0.5235, + "step": 14670 + }, + { + "epoch": 0.8196776266167556, + "grad_norm": 1.4342095851898193, + "learning_rate": 0.0005922512326311071, + "loss": 0.5525, + "step": 14671 + }, + { + "epoch": 0.8197334972204374, + "grad_norm": 0.6068954467773438, + "learning_rate": 0.0005922232182877634, + "loss": 0.4905, + "step": 14672 + }, + { + "epoch": 0.8197893678241194, + "grad_norm": 0.5865341424942017, + "learning_rate": 0.0005921952039444195, + "loss": 0.3124, + "step": 14673 + }, + { + "epoch": 0.8198452384278012, + "grad_norm": 3.841092109680176, + "learning_rate": 0.0005921671896010757, + "loss": 0.4734, + "step": 14674 + }, + { + "epoch": 0.819901109031483, + "grad_norm": 0.9350895881652832, + "learning_rate": 0.0005921391752577319, + "loss": 0.3616, + "step": 14675 + }, + { + "epoch": 0.819956979635165, + "grad_norm": 7.19405460357666, + "learning_rate": 0.0005921111609143882, + "loss": 0.3967, + "step": 14676 + }, + { + "epoch": 0.8200128502388468, + "grad_norm": 0.5283555388450623, + "learning_rate": 0.0005920831465710443, + "loss": 0.4276, + "step": 14677 + }, + { + "epoch": 0.8200687208425287, + "grad_norm": 0.5545494556427002, + "learning_rate": 0.0005920551322277005, + "loss": 0.4948, + "step": 14678 + }, + { + "epoch": 0.8201245914462105, + "grad_norm": 0.45709919929504395, + "learning_rate": 0.0005920271178843569, + "loss": 0.44, + "step": 14679 + }, + { + "epoch": 0.8201804620498925, + "grad_norm": 0.664887547492981, + "learning_rate": 0.0005919991035410131, + "loss": 0.4736, + "step": 14680 + }, + { + "epoch": 0.8202363326535743, + "grad_norm": 3.1192171573638916, + "learning_rate": 0.0005919710891976692, + "loss": 0.4331, + "step": 14681 + }, + { + "epoch": 0.8202922032572562, + "grad_norm": 0.6217215657234192, + "learning_rate": 0.0005919430748543254, + "loss": 0.5059, + "step": 14682 + }, + { + "epoch": 0.8203480738609381, + "grad_norm": 2.0192906856536865, + "learning_rate": 0.0005919150605109817, + "loss": 0.4581, + "step": 14683 + }, + { + "epoch": 0.82040394446462, + "grad_norm": 0.47037366032600403, + "learning_rate": 0.0005918870461676379, + "loss": 0.3951, + "step": 14684 + }, + { + "epoch": 0.8204598150683018, + "grad_norm": 0.3852527141571045, + "learning_rate": 0.000591859031824294, + "loss": 0.4012, + "step": 14685 + }, + { + "epoch": 0.8205156856719837, + "grad_norm": 0.5487754344940186, + "learning_rate": 0.0005918310174809503, + "loss": 0.4714, + "step": 14686 + }, + { + "epoch": 0.8205715562756656, + "grad_norm": 0.4515005648136139, + "learning_rate": 0.0005918030031376065, + "loss": 0.3767, + "step": 14687 + }, + { + "epoch": 0.8206274268793474, + "grad_norm": 0.384339302778244, + "learning_rate": 0.0005917749887942627, + "loss": 0.4684, + "step": 14688 + }, + { + "epoch": 0.8206832974830293, + "grad_norm": 0.4024884104728699, + "learning_rate": 0.0005917469744509188, + "loss": 0.3701, + "step": 14689 + }, + { + "epoch": 0.8207391680867112, + "grad_norm": 0.5399254560470581, + "learning_rate": 0.0005917189601075751, + "loss": 0.4188, + "step": 14690 + }, + { + "epoch": 0.820795038690393, + "grad_norm": 0.3376947343349457, + "learning_rate": 0.0005916909457642313, + "loss": 0.3079, + "step": 14691 + }, + { + "epoch": 0.8208509092940749, + "grad_norm": 0.6082528829574585, + "learning_rate": 0.0005916629314208874, + "loss": 0.5326, + "step": 14692 + }, + { + "epoch": 0.8209067798977568, + "grad_norm": 0.568178653717041, + "learning_rate": 0.0005916349170775437, + "loss": 0.6802, + "step": 14693 + }, + { + "epoch": 0.8209626505014387, + "grad_norm": 0.48825615644454956, + "learning_rate": 0.0005916069027342, + "loss": 0.4696, + "step": 14694 + }, + { + "epoch": 0.8210185211051205, + "grad_norm": 1.3598657846450806, + "learning_rate": 0.0005915788883908562, + "loss": 0.4182, + "step": 14695 + }, + { + "epoch": 0.8210743917088024, + "grad_norm": 0.5113895535469055, + "learning_rate": 0.0005915508740475123, + "loss": 0.4702, + "step": 14696 + }, + { + "epoch": 0.8211302623124843, + "grad_norm": 0.7306109070777893, + "learning_rate": 0.0005915228597041686, + "loss": 0.4805, + "step": 14697 + }, + { + "epoch": 0.8211861329161662, + "grad_norm": 0.43581274151802063, + "learning_rate": 0.0005914948453608248, + "loss": 0.5981, + "step": 14698 + }, + { + "epoch": 0.821242003519848, + "grad_norm": 0.5192616581916809, + "learning_rate": 0.000591466831017481, + "loss": 0.5117, + "step": 14699 + }, + { + "epoch": 0.82129787412353, + "grad_norm": 0.4786086976528168, + "learning_rate": 0.0005914388166741371, + "loss": 0.3369, + "step": 14700 + }, + { + "epoch": 0.8213537447272118, + "grad_norm": 1.0013450384140015, + "learning_rate": 0.0005914108023307934, + "loss": 0.4618, + "step": 14701 + }, + { + "epoch": 0.8214096153308936, + "grad_norm": 0.41074487566947937, + "learning_rate": 0.0005913827879874496, + "loss": 0.4197, + "step": 14702 + }, + { + "epoch": 0.8214654859345755, + "grad_norm": 0.7857810258865356, + "learning_rate": 0.0005913547736441058, + "loss": 0.5141, + "step": 14703 + }, + { + "epoch": 0.8215213565382574, + "grad_norm": 0.4371468424797058, + "learning_rate": 0.000591326759300762, + "loss": 0.3142, + "step": 14704 + }, + { + "epoch": 0.8215772271419393, + "grad_norm": 1.9844624996185303, + "learning_rate": 0.0005912987449574182, + "loss": 0.453, + "step": 14705 + }, + { + "epoch": 0.8216330977456211, + "grad_norm": 0.720429003238678, + "learning_rate": 0.0005912707306140744, + "loss": 0.5518, + "step": 14706 + }, + { + "epoch": 0.821688968349303, + "grad_norm": 3.105156660079956, + "learning_rate": 0.0005912427162707306, + "loss": 0.4956, + "step": 14707 + }, + { + "epoch": 0.8217448389529849, + "grad_norm": 0.4367329180240631, + "learning_rate": 0.0005912147019273868, + "loss": 0.3164, + "step": 14708 + }, + { + "epoch": 0.8218007095566667, + "grad_norm": 0.4717821478843689, + "learning_rate": 0.000591186687584043, + "loss": 0.4278, + "step": 14709 + }, + { + "epoch": 0.8218565801603487, + "grad_norm": 0.517623245716095, + "learning_rate": 0.0005911586732406992, + "loss": 0.4836, + "step": 14710 + }, + { + "epoch": 0.8219124507640305, + "grad_norm": 0.45945999026298523, + "learning_rate": 0.0005911306588973556, + "loss": 0.4715, + "step": 14711 + }, + { + "epoch": 0.8219683213677124, + "grad_norm": 0.6065245270729065, + "learning_rate": 0.0005911026445540117, + "loss": 0.3756, + "step": 14712 + }, + { + "epoch": 0.8220241919713942, + "grad_norm": 1.7805904150009155, + "learning_rate": 0.0005910746302106679, + "loss": 0.4429, + "step": 14713 + }, + { + "epoch": 0.8220800625750762, + "grad_norm": 0.41316312551498413, + "learning_rate": 0.0005910466158673241, + "loss": 0.5978, + "step": 14714 + }, + { + "epoch": 0.822135933178758, + "grad_norm": 2.5162997245788574, + "learning_rate": 0.0005910186015239803, + "loss": 0.328, + "step": 14715 + }, + { + "epoch": 0.8221918037824398, + "grad_norm": 0.42758163809776306, + "learning_rate": 0.0005909905871806365, + "loss": 0.4913, + "step": 14716 + }, + { + "epoch": 0.8222476743861218, + "grad_norm": 0.5143594741821289, + "learning_rate": 0.0005909625728372927, + "loss": 0.4318, + "step": 14717 + }, + { + "epoch": 0.8223035449898036, + "grad_norm": 0.4016572833061218, + "learning_rate": 0.000590934558493949, + "loss": 0.3524, + "step": 14718 + }, + { + "epoch": 0.8223594155934855, + "grad_norm": 0.37048354744911194, + "learning_rate": 0.0005909065441506051, + "loss": 0.4806, + "step": 14719 + }, + { + "epoch": 0.8224152861971673, + "grad_norm": 0.6306585669517517, + "learning_rate": 0.0005908785298072613, + "loss": 0.6877, + "step": 14720 + }, + { + "epoch": 0.8224711568008493, + "grad_norm": 0.3685491681098938, + "learning_rate": 0.0005908505154639175, + "loss": 0.4654, + "step": 14721 + }, + { + "epoch": 0.8225270274045311, + "grad_norm": 0.7585979104042053, + "learning_rate": 0.0005908225011205738, + "loss": 0.4782, + "step": 14722 + }, + { + "epoch": 0.8225828980082129, + "grad_norm": 0.492938756942749, + "learning_rate": 0.0005907944867772299, + "loss": 0.4155, + "step": 14723 + }, + { + "epoch": 0.8226387686118949, + "grad_norm": 0.46632570028305054, + "learning_rate": 0.0005907664724338861, + "loss": 0.4075, + "step": 14724 + }, + { + "epoch": 0.8226946392155767, + "grad_norm": 0.6653192639350891, + "learning_rate": 0.0005907384580905424, + "loss": 0.4523, + "step": 14725 + }, + { + "epoch": 0.8227505098192586, + "grad_norm": 0.6115115284919739, + "learning_rate": 0.0005907104437471986, + "loss": 0.4508, + "step": 14726 + }, + { + "epoch": 0.8228063804229405, + "grad_norm": 0.47011932730674744, + "learning_rate": 0.0005906824294038547, + "loss": 0.4092, + "step": 14727 + }, + { + "epoch": 0.8228622510266224, + "grad_norm": 0.6480539441108704, + "learning_rate": 0.0005906544150605109, + "loss": 0.5746, + "step": 14728 + }, + { + "epoch": 0.8229181216303042, + "grad_norm": 0.5257002115249634, + "learning_rate": 0.0005906264007171673, + "loss": 0.4995, + "step": 14729 + }, + { + "epoch": 0.822973992233986, + "grad_norm": 0.36006417870521545, + "learning_rate": 0.0005905983863738235, + "loss": 0.339, + "step": 14730 + }, + { + "epoch": 0.823029862837668, + "grad_norm": 0.5387992858886719, + "learning_rate": 0.0005905703720304796, + "loss": 0.4225, + "step": 14731 + }, + { + "epoch": 0.8230857334413498, + "grad_norm": 0.470643013715744, + "learning_rate": 0.0005905423576871359, + "loss": 0.5132, + "step": 14732 + }, + { + "epoch": 0.8231416040450317, + "grad_norm": 0.7173253893852234, + "learning_rate": 0.0005905143433437921, + "loss": 0.4993, + "step": 14733 + }, + { + "epoch": 0.8231974746487136, + "grad_norm": 0.4204113483428955, + "learning_rate": 0.0005904863290004483, + "loss": 0.4081, + "step": 14734 + }, + { + "epoch": 0.8232533452523955, + "grad_norm": 0.47951993346214294, + "learning_rate": 0.0005904583146571044, + "loss": 0.4196, + "step": 14735 + }, + { + "epoch": 0.8233092158560773, + "grad_norm": 0.3369995653629303, + "learning_rate": 0.0005904303003137607, + "loss": 0.3668, + "step": 14736 + }, + { + "epoch": 0.8233650864597591, + "grad_norm": 0.4173097610473633, + "learning_rate": 0.0005904022859704169, + "loss": 0.4052, + "step": 14737 + }, + { + "epoch": 0.8234209570634411, + "grad_norm": 0.5550920963287354, + "learning_rate": 0.000590374271627073, + "loss": 0.5948, + "step": 14738 + }, + { + "epoch": 0.8234768276671229, + "grad_norm": 0.34274211525917053, + "learning_rate": 0.0005903462572837292, + "loss": 0.3867, + "step": 14739 + }, + { + "epoch": 0.8235326982708048, + "grad_norm": 0.7116803526878357, + "learning_rate": 0.0005903182429403855, + "loss": 0.3787, + "step": 14740 + }, + { + "epoch": 0.8235885688744867, + "grad_norm": 0.8392075896263123, + "learning_rate": 0.0005902902285970417, + "loss": 0.4672, + "step": 14741 + }, + { + "epoch": 0.8236444394781686, + "grad_norm": 0.4131726026535034, + "learning_rate": 0.0005902622142536978, + "loss": 0.4059, + "step": 14742 + }, + { + "epoch": 0.8237003100818504, + "grad_norm": 0.41162097454071045, + "learning_rate": 0.0005902341999103541, + "loss": 0.3719, + "step": 14743 + }, + { + "epoch": 0.8237561806855324, + "grad_norm": 0.5730288028717041, + "learning_rate": 0.0005902061855670103, + "loss": 0.4425, + "step": 14744 + }, + { + "epoch": 0.8238120512892142, + "grad_norm": 0.49356403946876526, + "learning_rate": 0.0005901781712236666, + "loss": 0.5299, + "step": 14745 + }, + { + "epoch": 0.823867921892896, + "grad_norm": 0.5506628751754761, + "learning_rate": 0.0005901501568803226, + "loss": 0.4148, + "step": 14746 + }, + { + "epoch": 0.8239237924965779, + "grad_norm": 2.3833844661712646, + "learning_rate": 0.000590122142536979, + "loss": 0.4445, + "step": 14747 + }, + { + "epoch": 0.8239796631002598, + "grad_norm": 2.8890960216522217, + "learning_rate": 0.0005900941281936352, + "loss": 0.3462, + "step": 14748 + }, + { + "epoch": 0.8240355337039417, + "grad_norm": 0.4731200933456421, + "learning_rate": 0.0005900661138502914, + "loss": 0.4689, + "step": 14749 + }, + { + "epoch": 0.8240914043076235, + "grad_norm": 0.3628256022930145, + "learning_rate": 0.0005900380995069476, + "loss": 0.3511, + "step": 14750 + }, + { + "epoch": 0.8241472749113055, + "grad_norm": 0.4990544021129608, + "learning_rate": 0.0005900100851636038, + "loss": 0.5253, + "step": 14751 + }, + { + "epoch": 0.8242031455149873, + "grad_norm": 0.44128745794296265, + "learning_rate": 0.00058998207082026, + "loss": 0.374, + "step": 14752 + }, + { + "epoch": 0.8242590161186691, + "grad_norm": 0.3465210497379303, + "learning_rate": 0.0005899540564769162, + "loss": 0.4056, + "step": 14753 + }, + { + "epoch": 0.824314886722351, + "grad_norm": 2.408761978149414, + "learning_rate": 0.0005899260421335724, + "loss": 0.3609, + "step": 14754 + }, + { + "epoch": 0.8243707573260329, + "grad_norm": 0.4123425781726837, + "learning_rate": 0.0005898980277902286, + "loss": 0.4368, + "step": 14755 + }, + { + "epoch": 0.8244266279297148, + "grad_norm": 0.4471760392189026, + "learning_rate": 0.0005898700134468848, + "loss": 0.6239, + "step": 14756 + }, + { + "epoch": 0.8244824985333966, + "grad_norm": 0.33286330103874207, + "learning_rate": 0.000589841999103541, + "loss": 0.3227, + "step": 14757 + }, + { + "epoch": 0.8245383691370786, + "grad_norm": 0.4215811789035797, + "learning_rate": 0.0005898139847601972, + "loss": 0.4111, + "step": 14758 + }, + { + "epoch": 0.8245942397407604, + "grad_norm": 0.5933043956756592, + "learning_rate": 0.0005897859704168534, + "loss": 0.4583, + "step": 14759 + }, + { + "epoch": 0.8246501103444422, + "grad_norm": 0.4597631096839905, + "learning_rate": 0.0005897579560735096, + "loss": 0.3789, + "step": 14760 + }, + { + "epoch": 0.8247059809481242, + "grad_norm": 0.546693742275238, + "learning_rate": 0.0005897299417301658, + "loss": 0.4808, + "step": 14761 + }, + { + "epoch": 0.824761851551806, + "grad_norm": 0.3995975852012634, + "learning_rate": 0.000589701927386822, + "loss": 0.4176, + "step": 14762 + }, + { + "epoch": 0.8248177221554879, + "grad_norm": 0.4173557162284851, + "learning_rate": 0.0005896739130434783, + "loss": 0.4058, + "step": 14763 + }, + { + "epoch": 0.8248735927591697, + "grad_norm": 0.42167940735816956, + "learning_rate": 0.0005896458987001346, + "loss": 0.3637, + "step": 14764 + }, + { + "epoch": 0.8249294633628517, + "grad_norm": 0.4062604308128357, + "learning_rate": 0.0005896178843567907, + "loss": 0.4482, + "step": 14765 + }, + { + "epoch": 0.8249853339665335, + "grad_norm": 0.7031906247138977, + "learning_rate": 0.0005895898700134469, + "loss": 0.4819, + "step": 14766 + }, + { + "epoch": 0.8250412045702153, + "grad_norm": 0.4205365777015686, + "learning_rate": 0.0005895618556701031, + "loss": 0.438, + "step": 14767 + }, + { + "epoch": 0.8250970751738973, + "grad_norm": 1.570701241493225, + "learning_rate": 0.0005895338413267594, + "loss": 0.4959, + "step": 14768 + }, + { + "epoch": 0.8251529457775791, + "grad_norm": 0.3631305694580078, + "learning_rate": 0.0005895058269834155, + "loss": 0.4232, + "step": 14769 + }, + { + "epoch": 0.825208816381261, + "grad_norm": 0.4153154492378235, + "learning_rate": 0.0005894778126400717, + "loss": 0.4035, + "step": 14770 + }, + { + "epoch": 0.8252646869849428, + "grad_norm": 0.5415488481521606, + "learning_rate": 0.0005894497982967279, + "loss": 0.5114, + "step": 14771 + }, + { + "epoch": 0.8253205575886248, + "grad_norm": 0.43782442808151245, + "learning_rate": 0.0005894217839533842, + "loss": 0.5657, + "step": 14772 + }, + { + "epoch": 0.8253764281923066, + "grad_norm": 0.5667098164558411, + "learning_rate": 0.0005893937696100403, + "loss": 0.3685, + "step": 14773 + }, + { + "epoch": 0.8254322987959885, + "grad_norm": 0.38496145606040955, + "learning_rate": 0.0005893657552666965, + "loss": 0.3498, + "step": 14774 + }, + { + "epoch": 0.8254881693996704, + "grad_norm": 0.5882765650749207, + "learning_rate": 0.0005893377409233528, + "loss": 0.4354, + "step": 14775 + }, + { + "epoch": 0.8255440400033522, + "grad_norm": 0.7868071794509888, + "learning_rate": 0.000589309726580009, + "loss": 0.5198, + "step": 14776 + }, + { + "epoch": 0.8255999106070341, + "grad_norm": 0.41160595417022705, + "learning_rate": 0.0005892817122366651, + "loss": 0.5183, + "step": 14777 + }, + { + "epoch": 0.825655781210716, + "grad_norm": 1.0242929458618164, + "learning_rate": 0.0005892536978933213, + "loss": 0.5075, + "step": 14778 + }, + { + "epoch": 0.8257116518143979, + "grad_norm": 0.5963667035102844, + "learning_rate": 0.0005892256835499777, + "loss": 0.3981, + "step": 14779 + }, + { + "epoch": 0.8257675224180797, + "grad_norm": 1.0677555799484253, + "learning_rate": 0.0005891976692066338, + "loss": 0.5064, + "step": 14780 + }, + { + "epoch": 0.8258233930217616, + "grad_norm": 0.4356894791126251, + "learning_rate": 0.00058916965486329, + "loss": 0.4436, + "step": 14781 + }, + { + "epoch": 0.8258792636254435, + "grad_norm": 0.4771660566329956, + "learning_rate": 0.0005891416405199463, + "loss": 0.3759, + "step": 14782 + }, + { + "epoch": 0.8259351342291253, + "grad_norm": 0.35447096824645996, + "learning_rate": 0.0005891136261766025, + "loss": 0.3807, + "step": 14783 + }, + { + "epoch": 0.8259910048328072, + "grad_norm": 0.3714160621166229, + "learning_rate": 0.0005890856118332586, + "loss": 0.4432, + "step": 14784 + }, + { + "epoch": 0.8260468754364891, + "grad_norm": 0.4525142312049866, + "learning_rate": 0.0005890575974899148, + "loss": 0.4841, + "step": 14785 + }, + { + "epoch": 0.826102746040171, + "grad_norm": 0.742901086807251, + "learning_rate": 0.0005890295831465711, + "loss": 0.5686, + "step": 14786 + }, + { + "epoch": 0.8261586166438528, + "grad_norm": 0.4894007742404938, + "learning_rate": 0.0005890015688032273, + "loss": 0.6049, + "step": 14787 + }, + { + "epoch": 0.8262144872475347, + "grad_norm": 0.434557169675827, + "learning_rate": 0.0005889735544598834, + "loss": 0.4145, + "step": 14788 + }, + { + "epoch": 0.8262703578512166, + "grad_norm": 0.9923653602600098, + "learning_rate": 0.0005889455401165397, + "loss": 0.4137, + "step": 14789 + }, + { + "epoch": 0.8263262284548984, + "grad_norm": 0.5420666337013245, + "learning_rate": 0.0005889175257731959, + "loss": 0.4318, + "step": 14790 + }, + { + "epoch": 0.8263820990585803, + "grad_norm": 0.447165310382843, + "learning_rate": 0.0005888895114298521, + "loss": 0.3688, + "step": 14791 + }, + { + "epoch": 0.8264379696622622, + "grad_norm": 0.40860864520072937, + "learning_rate": 0.0005888614970865082, + "loss": 0.4848, + "step": 14792 + }, + { + "epoch": 0.8264938402659441, + "grad_norm": 3.1950833797454834, + "learning_rate": 0.0005888334827431645, + "loss": 0.3485, + "step": 14793 + }, + { + "epoch": 0.8265497108696259, + "grad_norm": 2.339820146560669, + "learning_rate": 0.0005888054683998207, + "loss": 0.4457, + "step": 14794 + }, + { + "epoch": 0.8266055814733079, + "grad_norm": 0.4362979233264923, + "learning_rate": 0.000588777454056477, + "loss": 0.4442, + "step": 14795 + }, + { + "epoch": 0.8266614520769897, + "grad_norm": 0.5099694728851318, + "learning_rate": 0.0005887494397131332, + "loss": 0.4408, + "step": 14796 + }, + { + "epoch": 0.8267173226806716, + "grad_norm": 0.3808230757713318, + "learning_rate": 0.0005887214253697894, + "loss": 0.3981, + "step": 14797 + }, + { + "epoch": 0.8267731932843534, + "grad_norm": 0.5416070818901062, + "learning_rate": 0.0005886934110264456, + "loss": 0.4419, + "step": 14798 + }, + { + "epoch": 0.8268290638880353, + "grad_norm": 0.3734534978866577, + "learning_rate": 0.0005886653966831017, + "loss": 0.3532, + "step": 14799 + }, + { + "epoch": 0.8268849344917172, + "grad_norm": 0.3399306535720825, + "learning_rate": 0.000588637382339758, + "loss": 0.3603, + "step": 14800 + }, + { + "epoch": 0.826940805095399, + "grad_norm": 0.3559456169605255, + "learning_rate": 0.0005886093679964142, + "loss": 0.3512, + "step": 14801 + }, + { + "epoch": 0.826996675699081, + "grad_norm": 0.7699991464614868, + "learning_rate": 0.0005885813536530704, + "loss": 0.5309, + "step": 14802 + }, + { + "epoch": 0.8270525463027628, + "grad_norm": 0.43886396288871765, + "learning_rate": 0.0005885533393097265, + "loss": 0.4178, + "step": 14803 + }, + { + "epoch": 0.8271084169064447, + "grad_norm": 0.611238420009613, + "learning_rate": 0.0005885253249663828, + "loss": 0.34, + "step": 14804 + }, + { + "epoch": 0.8271642875101265, + "grad_norm": 0.41927671432495117, + "learning_rate": 0.000588497310623039, + "loss": 0.3881, + "step": 14805 + }, + { + "epoch": 0.8272201581138084, + "grad_norm": 0.338059663772583, + "learning_rate": 0.0005884692962796952, + "loss": 0.375, + "step": 14806 + }, + { + "epoch": 0.8272760287174903, + "grad_norm": 0.36773887276649475, + "learning_rate": 0.0005884412819363514, + "loss": 0.419, + "step": 14807 + }, + { + "epoch": 0.8273318993211721, + "grad_norm": 0.5592237114906311, + "learning_rate": 0.0005884132675930076, + "loss": 0.4241, + "step": 14808 + }, + { + "epoch": 0.8273877699248541, + "grad_norm": 0.6524297595024109, + "learning_rate": 0.0005883852532496638, + "loss": 0.3911, + "step": 14809 + }, + { + "epoch": 0.8274436405285359, + "grad_norm": 1.7864493131637573, + "learning_rate": 0.00058835723890632, + "loss": 0.4735, + "step": 14810 + }, + { + "epoch": 0.8274995111322178, + "grad_norm": 0.8441469669342041, + "learning_rate": 0.0005883292245629762, + "loss": 0.4445, + "step": 14811 + }, + { + "epoch": 0.8275553817358997, + "grad_norm": 1.0586445331573486, + "learning_rate": 0.0005883012102196325, + "loss": 0.7701, + "step": 14812 + }, + { + "epoch": 0.8276112523395815, + "grad_norm": 0.4868951439857483, + "learning_rate": 0.0005882731958762887, + "loss": 0.4968, + "step": 14813 + }, + { + "epoch": 0.8276671229432634, + "grad_norm": 1.4952939748764038, + "learning_rate": 0.000588245181532945, + "loss": 0.4928, + "step": 14814 + }, + { + "epoch": 0.8277229935469452, + "grad_norm": 0.3909667134284973, + "learning_rate": 0.0005882171671896011, + "loss": 0.4428, + "step": 14815 + }, + { + "epoch": 0.8277788641506272, + "grad_norm": 0.479107528924942, + "learning_rate": 0.0005881891528462573, + "loss": 0.507, + "step": 14816 + }, + { + "epoch": 0.827834734754309, + "grad_norm": 0.7318060994148254, + "learning_rate": 0.0005881611385029135, + "loss": 0.4325, + "step": 14817 + }, + { + "epoch": 0.8278906053579909, + "grad_norm": 7.578203201293945, + "learning_rate": 0.0005881331241595698, + "loss": 0.534, + "step": 14818 + }, + { + "epoch": 0.8279464759616728, + "grad_norm": 0.7079473733901978, + "learning_rate": 0.0005881051098162259, + "loss": 0.4095, + "step": 14819 + }, + { + "epoch": 0.8280023465653547, + "grad_norm": 0.7048851847648621, + "learning_rate": 0.0005880770954728821, + "loss": 0.4114, + "step": 14820 + }, + { + "epoch": 0.8280582171690365, + "grad_norm": 0.5724986791610718, + "learning_rate": 0.0005880490811295384, + "loss": 0.4214, + "step": 14821 + }, + { + "epoch": 0.8281140877727183, + "grad_norm": 0.4024474024772644, + "learning_rate": 0.0005880210667861945, + "loss": 0.3791, + "step": 14822 + }, + { + "epoch": 0.8281699583764003, + "grad_norm": 0.7500112652778625, + "learning_rate": 0.0005879930524428507, + "loss": 0.5102, + "step": 14823 + }, + { + "epoch": 0.8282258289800821, + "grad_norm": 1.8536382913589478, + "learning_rate": 0.0005879650380995069, + "loss": 0.3723, + "step": 14824 + }, + { + "epoch": 0.828281699583764, + "grad_norm": 0.4901805818080902, + "learning_rate": 0.0005879370237561632, + "loss": 0.4453, + "step": 14825 + }, + { + "epoch": 0.8283375701874459, + "grad_norm": 0.7653189301490784, + "learning_rate": 0.0005879090094128193, + "loss": 0.3871, + "step": 14826 + }, + { + "epoch": 0.8283934407911278, + "grad_norm": 0.6858178973197937, + "learning_rate": 0.0005878809950694755, + "loss": 0.4218, + "step": 14827 + }, + { + "epoch": 0.8284493113948096, + "grad_norm": 0.5893157720565796, + "learning_rate": 0.0005878529807261319, + "loss": 0.3908, + "step": 14828 + }, + { + "epoch": 0.8285051819984915, + "grad_norm": 0.9362946152687073, + "learning_rate": 0.0005878249663827881, + "loss": 0.488, + "step": 14829 + }, + { + "epoch": 0.8285610526021734, + "grad_norm": 0.5760895013809204, + "learning_rate": 0.0005877969520394442, + "loss": 0.4537, + "step": 14830 + }, + { + "epoch": 0.8286169232058552, + "grad_norm": 4.224344730377197, + "learning_rate": 0.0005877689376961004, + "loss": 0.4753, + "step": 14831 + }, + { + "epoch": 0.8286727938095371, + "grad_norm": 0.6936929225921631, + "learning_rate": 0.0005877409233527567, + "loss": 0.4354, + "step": 14832 + }, + { + "epoch": 0.828728664413219, + "grad_norm": 0.492016077041626, + "learning_rate": 0.0005877129090094129, + "loss": 0.321, + "step": 14833 + }, + { + "epoch": 0.8287845350169009, + "grad_norm": 0.4433666169643402, + "learning_rate": 0.000587684894666069, + "loss": 0.4843, + "step": 14834 + }, + { + "epoch": 0.8288404056205827, + "grad_norm": 0.4254678189754486, + "learning_rate": 0.0005876568803227253, + "loss": 0.3915, + "step": 14835 + }, + { + "epoch": 0.8288962762242646, + "grad_norm": 0.4803553819656372, + "learning_rate": 0.0005876288659793815, + "loss": 0.4257, + "step": 14836 + }, + { + "epoch": 0.8289521468279465, + "grad_norm": 1.4052300453186035, + "learning_rate": 0.0005876008516360377, + "loss": 0.3298, + "step": 14837 + }, + { + "epoch": 0.8290080174316283, + "grad_norm": 0.3753562271595001, + "learning_rate": 0.0005875728372926938, + "loss": 0.4656, + "step": 14838 + }, + { + "epoch": 0.8290638880353102, + "grad_norm": 0.44970980286598206, + "learning_rate": 0.0005875448229493501, + "loss": 0.4796, + "step": 14839 + }, + { + "epoch": 0.8291197586389921, + "grad_norm": 0.7757684588432312, + "learning_rate": 0.0005875168086060063, + "loss": 0.3935, + "step": 14840 + }, + { + "epoch": 0.829175629242674, + "grad_norm": 0.308284193277359, + "learning_rate": 0.0005874887942626624, + "loss": 0.3872, + "step": 14841 + }, + { + "epoch": 0.8292314998463558, + "grad_norm": 0.4406222999095917, + "learning_rate": 0.0005874607799193186, + "loss": 0.4193, + "step": 14842 + }, + { + "epoch": 0.8292873704500378, + "grad_norm": 0.4289402663707733, + "learning_rate": 0.0005874327655759749, + "loss": 0.4093, + "step": 14843 + }, + { + "epoch": 0.8293432410537196, + "grad_norm": 0.7220064401626587, + "learning_rate": 0.0005874047512326311, + "loss": 0.6436, + "step": 14844 + }, + { + "epoch": 0.8293991116574014, + "grad_norm": 0.5225938558578491, + "learning_rate": 0.0005873767368892872, + "loss": 0.3908, + "step": 14845 + }, + { + "epoch": 0.8294549822610834, + "grad_norm": 0.5184041857719421, + "learning_rate": 0.0005873487225459436, + "loss": 0.4315, + "step": 14846 + }, + { + "epoch": 0.8295108528647652, + "grad_norm": 0.46213841438293457, + "learning_rate": 0.0005873207082025998, + "loss": 0.3816, + "step": 14847 + }, + { + "epoch": 0.8295667234684471, + "grad_norm": 4.135205268859863, + "learning_rate": 0.000587292693859256, + "loss": 0.4502, + "step": 14848 + }, + { + "epoch": 0.8296225940721289, + "grad_norm": 0.47956395149230957, + "learning_rate": 0.0005872646795159121, + "loss": 0.3748, + "step": 14849 + }, + { + "epoch": 0.8296784646758109, + "grad_norm": 0.5600651502609253, + "learning_rate": 0.0005872366651725684, + "loss": 0.4974, + "step": 14850 + }, + { + "epoch": 0.8297343352794927, + "grad_norm": 0.6617973446846008, + "learning_rate": 0.0005872086508292246, + "loss": 0.4055, + "step": 14851 + }, + { + "epoch": 0.8297902058831745, + "grad_norm": 0.5596491098403931, + "learning_rate": 0.0005871806364858808, + "loss": 0.505, + "step": 14852 + }, + { + "epoch": 0.8298460764868565, + "grad_norm": 0.32690396904945374, + "learning_rate": 0.000587152622142537, + "loss": 0.4212, + "step": 14853 + }, + { + "epoch": 0.8299019470905383, + "grad_norm": 0.4003518223762512, + "learning_rate": 0.0005871246077991932, + "loss": 0.3288, + "step": 14854 + }, + { + "epoch": 0.8299578176942202, + "grad_norm": 0.46469414234161377, + "learning_rate": 0.0005870965934558494, + "loss": 0.369, + "step": 14855 + }, + { + "epoch": 0.830013688297902, + "grad_norm": 8.156113624572754, + "learning_rate": 0.0005870685791125056, + "loss": 0.4332, + "step": 14856 + }, + { + "epoch": 0.830069558901584, + "grad_norm": 0.45452311635017395, + "learning_rate": 0.0005870405647691618, + "loss": 0.3946, + "step": 14857 + }, + { + "epoch": 0.8301254295052658, + "grad_norm": 1.905517578125, + "learning_rate": 0.000587012550425818, + "loss": 0.3676, + "step": 14858 + }, + { + "epoch": 0.8301813001089476, + "grad_norm": 1.6461156606674194, + "learning_rate": 0.0005869845360824742, + "loss": 0.4787, + "step": 14859 + }, + { + "epoch": 0.8302371707126296, + "grad_norm": 0.47930604219436646, + "learning_rate": 0.0005869565217391305, + "loss": 0.4289, + "step": 14860 + }, + { + "epoch": 0.8302930413163114, + "grad_norm": 0.5701023936271667, + "learning_rate": 0.0005869285073957866, + "loss": 0.3942, + "step": 14861 + }, + { + "epoch": 0.8303489119199933, + "grad_norm": 2.8071415424346924, + "learning_rate": 0.0005869004930524428, + "loss": 0.4053, + "step": 14862 + }, + { + "epoch": 0.8304047825236752, + "grad_norm": 8.634937286376953, + "learning_rate": 0.0005868724787090991, + "loss": 0.6516, + "step": 14863 + }, + { + "epoch": 0.8304606531273571, + "grad_norm": 0.4251364767551422, + "learning_rate": 0.0005868444643657553, + "loss": 0.452, + "step": 14864 + }, + { + "epoch": 0.8305165237310389, + "grad_norm": 0.49660876393318176, + "learning_rate": 0.0005868164500224115, + "loss": 0.4169, + "step": 14865 + }, + { + "epoch": 0.8305723943347207, + "grad_norm": 1.0801740884780884, + "learning_rate": 0.0005867884356790677, + "loss": 0.4286, + "step": 14866 + }, + { + "epoch": 0.8306282649384027, + "grad_norm": 0.8054161071777344, + "learning_rate": 0.000586760421335724, + "loss": 0.5116, + "step": 14867 + }, + { + "epoch": 0.8306841355420845, + "grad_norm": 0.3343879282474518, + "learning_rate": 0.0005867324069923801, + "loss": 0.4375, + "step": 14868 + }, + { + "epoch": 0.8307400061457664, + "grad_norm": 0.4895366132259369, + "learning_rate": 0.0005867043926490363, + "loss": 0.5144, + "step": 14869 + }, + { + "epoch": 0.8307958767494483, + "grad_norm": 0.5474954843521118, + "learning_rate": 0.0005866763783056925, + "loss": 0.5234, + "step": 14870 + }, + { + "epoch": 0.8308517473531302, + "grad_norm": 0.4197629690170288, + "learning_rate": 0.0005866483639623488, + "loss": 0.4147, + "step": 14871 + }, + { + "epoch": 0.830907617956812, + "grad_norm": 0.6081820130348206, + "learning_rate": 0.0005866203496190049, + "loss": 0.5586, + "step": 14872 + }, + { + "epoch": 0.8309634885604938, + "grad_norm": 0.398892343044281, + "learning_rate": 0.0005865923352756611, + "loss": 0.4586, + "step": 14873 + }, + { + "epoch": 0.8310193591641758, + "grad_norm": 0.6287072896957397, + "learning_rate": 0.0005865643209323174, + "loss": 0.5318, + "step": 14874 + }, + { + "epoch": 0.8310752297678576, + "grad_norm": 0.6885213255882263, + "learning_rate": 0.0005865363065889736, + "loss": 0.3967, + "step": 14875 + }, + { + "epoch": 0.8311311003715395, + "grad_norm": 0.4574962556362152, + "learning_rate": 0.0005865082922456297, + "loss": 0.4588, + "step": 14876 + }, + { + "epoch": 0.8311869709752214, + "grad_norm": 0.7830962538719177, + "learning_rate": 0.0005864802779022859, + "loss": 0.597, + "step": 14877 + }, + { + "epoch": 0.8312428415789033, + "grad_norm": 0.39953356981277466, + "learning_rate": 0.0005864522635589423, + "loss": 0.3571, + "step": 14878 + }, + { + "epoch": 0.8312987121825851, + "grad_norm": 0.5808364748954773, + "learning_rate": 0.0005864242492155985, + "loss": 0.3911, + "step": 14879 + }, + { + "epoch": 0.8313545827862671, + "grad_norm": 2.124690532684326, + "learning_rate": 0.0005863962348722546, + "loss": 0.436, + "step": 14880 + }, + { + "epoch": 0.8314104533899489, + "grad_norm": 0.9418643712997437, + "learning_rate": 0.0005863682205289108, + "loss": 0.4228, + "step": 14881 + }, + { + "epoch": 0.8314663239936307, + "grad_norm": 0.501953125, + "learning_rate": 0.0005863402061855671, + "loss": 0.4176, + "step": 14882 + }, + { + "epoch": 0.8315221945973126, + "grad_norm": 0.46201395988464355, + "learning_rate": 0.0005863121918422233, + "loss": 0.3732, + "step": 14883 + }, + { + "epoch": 0.8315780652009945, + "grad_norm": 0.40242674946784973, + "learning_rate": 0.0005862841774988794, + "loss": 0.3525, + "step": 14884 + }, + { + "epoch": 0.8316339358046764, + "grad_norm": 3.15429425239563, + "learning_rate": 0.0005862561631555357, + "loss": 0.4228, + "step": 14885 + }, + { + "epoch": 0.8316898064083582, + "grad_norm": 1.2312870025634766, + "learning_rate": 0.0005862281488121919, + "loss": 0.4637, + "step": 14886 + }, + { + "epoch": 0.8317456770120402, + "grad_norm": 0.4678957462310791, + "learning_rate": 0.000586200134468848, + "loss": 0.38, + "step": 14887 + }, + { + "epoch": 0.831801547615722, + "grad_norm": 0.45952457189559937, + "learning_rate": 0.0005861721201255042, + "loss": 0.4183, + "step": 14888 + }, + { + "epoch": 0.8318574182194038, + "grad_norm": 0.47408193349838257, + "learning_rate": 0.0005861441057821605, + "loss": 0.5648, + "step": 14889 + }, + { + "epoch": 0.8319132888230857, + "grad_norm": 0.39054733514785767, + "learning_rate": 0.0005861160914388167, + "loss": 0.3888, + "step": 14890 + }, + { + "epoch": 0.8319691594267676, + "grad_norm": 0.5531524419784546, + "learning_rate": 0.0005860880770954728, + "loss": 0.4404, + "step": 14891 + }, + { + "epoch": 0.8320250300304495, + "grad_norm": 0.4005955755710602, + "learning_rate": 0.0005860600627521291, + "loss": 0.4778, + "step": 14892 + }, + { + "epoch": 0.8320809006341313, + "grad_norm": 0.6166980862617493, + "learning_rate": 0.0005860320484087853, + "loss": 0.4683, + "step": 14893 + }, + { + "epoch": 0.8321367712378133, + "grad_norm": 0.5292292237281799, + "learning_rate": 0.0005860040340654415, + "loss": 0.4272, + "step": 14894 + }, + { + "epoch": 0.8321926418414951, + "grad_norm": 0.4246763586997986, + "learning_rate": 0.0005859760197220976, + "loss": 0.4545, + "step": 14895 + }, + { + "epoch": 0.832248512445177, + "grad_norm": 0.411670058965683, + "learning_rate": 0.000585948005378754, + "loss": 0.4153, + "step": 14896 + }, + { + "epoch": 0.8323043830488589, + "grad_norm": 0.47462233901023865, + "learning_rate": 0.0005859199910354102, + "loss": 0.4575, + "step": 14897 + }, + { + "epoch": 0.8323602536525407, + "grad_norm": 1.4858464002609253, + "learning_rate": 0.0005858919766920664, + "loss": 0.404, + "step": 14898 + }, + { + "epoch": 0.8324161242562226, + "grad_norm": 0.7387720942497253, + "learning_rate": 0.0005858639623487226, + "loss": 0.5276, + "step": 14899 + }, + { + "epoch": 0.8324719948599044, + "grad_norm": 0.6455279588699341, + "learning_rate": 0.0005858359480053788, + "loss": 0.5029, + "step": 14900 + }, + { + "epoch": 0.8325278654635864, + "grad_norm": 0.4530632793903351, + "learning_rate": 0.000585807933662035, + "loss": 0.3837, + "step": 14901 + }, + { + "epoch": 0.8325837360672682, + "grad_norm": 0.43752044439315796, + "learning_rate": 0.0005857799193186912, + "loss": 0.4212, + "step": 14902 + }, + { + "epoch": 0.83263960667095, + "grad_norm": 0.5020350813865662, + "learning_rate": 0.0005857519049753474, + "loss": 0.4153, + "step": 14903 + }, + { + "epoch": 0.832695477274632, + "grad_norm": 0.4073018431663513, + "learning_rate": 0.0005857238906320036, + "loss": 0.4885, + "step": 14904 + }, + { + "epoch": 0.8327513478783138, + "grad_norm": 0.8712196350097656, + "learning_rate": 0.0005856958762886598, + "loss": 0.37, + "step": 14905 + }, + { + "epoch": 0.8328072184819957, + "grad_norm": 2.0875768661499023, + "learning_rate": 0.000585667861945316, + "loss": 0.391, + "step": 14906 + }, + { + "epoch": 0.8328630890856775, + "grad_norm": 0.4901418387889862, + "learning_rate": 0.0005856398476019722, + "loss": 0.474, + "step": 14907 + }, + { + "epoch": 0.8329189596893595, + "grad_norm": 0.42290619015693665, + "learning_rate": 0.0005856118332586284, + "loss": 0.3495, + "step": 14908 + }, + { + "epoch": 0.8329748302930413, + "grad_norm": 0.5297216176986694, + "learning_rate": 0.0005855838189152846, + "loss": 0.4977, + "step": 14909 + }, + { + "epoch": 0.8330307008967232, + "grad_norm": 0.3563157021999359, + "learning_rate": 0.0005855558045719408, + "loss": 0.6325, + "step": 14910 + }, + { + "epoch": 0.8330865715004051, + "grad_norm": 0.3535958528518677, + "learning_rate": 0.000585527790228597, + "loss": 0.4752, + "step": 14911 + }, + { + "epoch": 0.8331424421040869, + "grad_norm": 0.5089762806892395, + "learning_rate": 0.0005854997758852532, + "loss": 0.6236, + "step": 14912 + }, + { + "epoch": 0.8331983127077688, + "grad_norm": 0.6250039339065552, + "learning_rate": 0.0005854717615419096, + "loss": 0.3967, + "step": 14913 + }, + { + "epoch": 0.8332541833114507, + "grad_norm": 1.1347746849060059, + "learning_rate": 0.0005854437471985657, + "loss": 0.4311, + "step": 14914 + }, + { + "epoch": 0.8333100539151326, + "grad_norm": 0.49475300312042236, + "learning_rate": 0.0005854157328552219, + "loss": 0.356, + "step": 14915 + }, + { + "epoch": 0.8333659245188144, + "grad_norm": 0.4998260736465454, + "learning_rate": 0.0005853877185118781, + "loss": 0.4364, + "step": 14916 + }, + { + "epoch": 0.8334217951224963, + "grad_norm": 0.4004567861557007, + "learning_rate": 0.0005853597041685344, + "loss": 0.3947, + "step": 14917 + }, + { + "epoch": 0.8334776657261782, + "grad_norm": 0.4050907790660858, + "learning_rate": 0.0005853316898251905, + "loss": 0.4607, + "step": 14918 + }, + { + "epoch": 0.83353353632986, + "grad_norm": 0.496646523475647, + "learning_rate": 0.0005853036754818467, + "loss": 0.3447, + "step": 14919 + }, + { + "epoch": 0.8335894069335419, + "grad_norm": 0.3993310332298279, + "learning_rate": 0.0005852756611385029, + "loss": 0.5625, + "step": 14920 + }, + { + "epoch": 0.8336452775372238, + "grad_norm": 0.4469067454338074, + "learning_rate": 0.0005852476467951592, + "loss": 0.4389, + "step": 14921 + }, + { + "epoch": 0.8337011481409057, + "grad_norm": 0.5026975274085999, + "learning_rate": 0.0005852196324518153, + "loss": 0.4782, + "step": 14922 + }, + { + "epoch": 0.8337570187445875, + "grad_norm": 0.5034511685371399, + "learning_rate": 0.0005851916181084715, + "loss": 0.451, + "step": 14923 + }, + { + "epoch": 0.8338128893482694, + "grad_norm": 0.5574461221694946, + "learning_rate": 0.0005851636037651278, + "loss": 0.4672, + "step": 14924 + }, + { + "epoch": 0.8338687599519513, + "grad_norm": 0.5843635201454163, + "learning_rate": 0.000585135589421784, + "loss": 0.4029, + "step": 14925 + }, + { + "epoch": 0.8339246305556332, + "grad_norm": 1.3601716756820679, + "learning_rate": 0.0005851075750784401, + "loss": 0.4548, + "step": 14926 + }, + { + "epoch": 0.833980501159315, + "grad_norm": 0.7469363212585449, + "learning_rate": 0.0005850795607350963, + "loss": 0.4372, + "step": 14927 + }, + { + "epoch": 0.8340363717629969, + "grad_norm": 0.8636696934700012, + "learning_rate": 0.0005850515463917526, + "loss": 0.5864, + "step": 14928 + }, + { + "epoch": 0.8340922423666788, + "grad_norm": 0.59710693359375, + "learning_rate": 0.0005850235320484087, + "loss": 0.5167, + "step": 14929 + }, + { + "epoch": 0.8341481129703606, + "grad_norm": 0.5251189470291138, + "learning_rate": 0.000584995517705065, + "loss": 0.3583, + "step": 14930 + }, + { + "epoch": 0.8342039835740425, + "grad_norm": 0.43068012595176697, + "learning_rate": 0.0005849675033617213, + "loss": 0.467, + "step": 14931 + }, + { + "epoch": 0.8342598541777244, + "grad_norm": 0.4794663190841675, + "learning_rate": 0.0005849394890183775, + "loss": 0.399, + "step": 14932 + }, + { + "epoch": 0.8343157247814063, + "grad_norm": 0.4686000347137451, + "learning_rate": 0.0005849114746750336, + "loss": 0.5485, + "step": 14933 + }, + { + "epoch": 0.8343715953850881, + "grad_norm": 1.3285489082336426, + "learning_rate": 0.0005848834603316898, + "loss": 0.4046, + "step": 14934 + }, + { + "epoch": 0.83442746598877, + "grad_norm": 0.43957868218421936, + "learning_rate": 0.0005848554459883461, + "loss": 0.4328, + "step": 14935 + }, + { + "epoch": 0.8344833365924519, + "grad_norm": 1.903619408607483, + "learning_rate": 0.0005848274316450023, + "loss": 0.4037, + "step": 14936 + }, + { + "epoch": 0.8345392071961337, + "grad_norm": 0.6172997951507568, + "learning_rate": 0.0005847994173016584, + "loss": 0.4181, + "step": 14937 + }, + { + "epoch": 0.8345950777998157, + "grad_norm": 0.5720274448394775, + "learning_rate": 0.0005847714029583147, + "loss": 0.4478, + "step": 14938 + }, + { + "epoch": 0.8346509484034975, + "grad_norm": 0.3903510570526123, + "learning_rate": 0.0005847433886149709, + "loss": 0.352, + "step": 14939 + }, + { + "epoch": 0.8347068190071794, + "grad_norm": 2.2747087478637695, + "learning_rate": 0.0005847153742716271, + "loss": 0.4771, + "step": 14940 + }, + { + "epoch": 0.8347626896108612, + "grad_norm": 0.7364723682403564, + "learning_rate": 0.0005846873599282832, + "loss": 0.3557, + "step": 14941 + }, + { + "epoch": 0.8348185602145431, + "grad_norm": 0.3513237535953522, + "learning_rate": 0.0005846593455849395, + "loss": 0.4264, + "step": 14942 + }, + { + "epoch": 0.834874430818225, + "grad_norm": 0.5717436671257019, + "learning_rate": 0.0005846313312415957, + "loss": 0.5425, + "step": 14943 + }, + { + "epoch": 0.8349303014219068, + "grad_norm": 1.1996631622314453, + "learning_rate": 0.0005846033168982519, + "loss": 0.3748, + "step": 14944 + }, + { + "epoch": 0.8349861720255888, + "grad_norm": 0.3992021083831787, + "learning_rate": 0.0005845753025549081, + "loss": 0.4446, + "step": 14945 + }, + { + "epoch": 0.8350420426292706, + "grad_norm": 0.43658187985420227, + "learning_rate": 0.0005845472882115644, + "loss": 0.4091, + "step": 14946 + }, + { + "epoch": 0.8350979132329525, + "grad_norm": 0.415792852640152, + "learning_rate": 0.0005845192738682206, + "loss": 0.4625, + "step": 14947 + }, + { + "epoch": 0.8351537838366343, + "grad_norm": 0.4393376410007477, + "learning_rate": 0.0005844912595248767, + "loss": 0.3982, + "step": 14948 + }, + { + "epoch": 0.8352096544403163, + "grad_norm": 0.3986091613769531, + "learning_rate": 0.000584463245181533, + "loss": 0.3455, + "step": 14949 + }, + { + "epoch": 0.8352655250439981, + "grad_norm": 0.45517686009407043, + "learning_rate": 0.0005844352308381892, + "loss": 0.4318, + "step": 14950 + }, + { + "epoch": 0.8353213956476799, + "grad_norm": 1.3875819444656372, + "learning_rate": 0.0005844072164948454, + "loss": 0.5061, + "step": 14951 + }, + { + "epoch": 0.8353772662513619, + "grad_norm": 0.4352225661277771, + "learning_rate": 0.0005843792021515015, + "loss": 0.4563, + "step": 14952 + }, + { + "epoch": 0.8354331368550437, + "grad_norm": 0.39697545766830444, + "learning_rate": 0.0005843511878081578, + "loss": 0.4805, + "step": 14953 + }, + { + "epoch": 0.8354890074587256, + "grad_norm": 0.5876631140708923, + "learning_rate": 0.000584323173464814, + "loss": 0.4532, + "step": 14954 + }, + { + "epoch": 0.8355448780624075, + "grad_norm": 0.4586201012134552, + "learning_rate": 0.0005842951591214702, + "loss": 0.4754, + "step": 14955 + }, + { + "epoch": 0.8356007486660894, + "grad_norm": 0.9270184636116028, + "learning_rate": 0.0005842671447781264, + "loss": 0.5447, + "step": 14956 + }, + { + "epoch": 0.8356566192697712, + "grad_norm": 0.4664750099182129, + "learning_rate": 0.0005842391304347826, + "loss": 0.4723, + "step": 14957 + }, + { + "epoch": 0.835712489873453, + "grad_norm": 0.3886289596557617, + "learning_rate": 0.0005842111160914388, + "loss": 0.4186, + "step": 14958 + }, + { + "epoch": 0.835768360477135, + "grad_norm": 0.44799530506134033, + "learning_rate": 0.000584183101748095, + "loss": 0.3843, + "step": 14959 + }, + { + "epoch": 0.8358242310808168, + "grad_norm": 0.9238483309745789, + "learning_rate": 0.0005841550874047512, + "loss": 0.4701, + "step": 14960 + }, + { + "epoch": 0.8358801016844987, + "grad_norm": 0.36842846870422363, + "learning_rate": 0.0005841270730614074, + "loss": 0.4492, + "step": 14961 + }, + { + "epoch": 0.8359359722881806, + "grad_norm": 0.4914831221103668, + "learning_rate": 0.0005840990587180636, + "loss": 0.544, + "step": 14962 + }, + { + "epoch": 0.8359918428918625, + "grad_norm": 0.4123392105102539, + "learning_rate": 0.00058407104437472, + "loss": 0.4328, + "step": 14963 + }, + { + "epoch": 0.8360477134955443, + "grad_norm": 0.4238136112689972, + "learning_rate": 0.0005840430300313761, + "loss": 0.3493, + "step": 14964 + }, + { + "epoch": 0.8361035840992261, + "grad_norm": 0.5038748979568481, + "learning_rate": 0.0005840150156880323, + "loss": 0.4549, + "step": 14965 + }, + { + "epoch": 0.8361594547029081, + "grad_norm": 0.45802050828933716, + "learning_rate": 0.0005839870013446885, + "loss": 0.3602, + "step": 14966 + }, + { + "epoch": 0.8362153253065899, + "grad_norm": 0.4460276961326599, + "learning_rate": 0.0005839589870013448, + "loss": 0.4917, + "step": 14967 + }, + { + "epoch": 0.8362711959102718, + "grad_norm": 0.7873501777648926, + "learning_rate": 0.0005839309726580009, + "loss": 0.3781, + "step": 14968 + }, + { + "epoch": 0.8363270665139537, + "grad_norm": 0.502784788608551, + "learning_rate": 0.0005839029583146571, + "loss": 0.3375, + "step": 14969 + }, + { + "epoch": 0.8363829371176356, + "grad_norm": 0.6931352615356445, + "learning_rate": 0.0005838749439713134, + "loss": 0.4955, + "step": 14970 + }, + { + "epoch": 0.8364388077213174, + "grad_norm": 1.3126068115234375, + "learning_rate": 0.0005838469296279695, + "loss": 0.3434, + "step": 14971 + }, + { + "epoch": 0.8364946783249994, + "grad_norm": 0.5440928339958191, + "learning_rate": 0.0005838189152846257, + "loss": 0.4942, + "step": 14972 + }, + { + "epoch": 0.8365505489286812, + "grad_norm": 0.5840023159980774, + "learning_rate": 0.0005837909009412819, + "loss": 0.5646, + "step": 14973 + }, + { + "epoch": 0.836606419532363, + "grad_norm": 0.5818125605583191, + "learning_rate": 0.0005837628865979382, + "loss": 0.8002, + "step": 14974 + }, + { + "epoch": 0.8366622901360449, + "grad_norm": 0.38564106822013855, + "learning_rate": 0.0005837348722545943, + "loss": 0.4568, + "step": 14975 + }, + { + "epoch": 0.8367181607397268, + "grad_norm": 0.4677940905094147, + "learning_rate": 0.0005837068579112505, + "loss": 0.3177, + "step": 14976 + }, + { + "epoch": 0.8367740313434087, + "grad_norm": 0.4389524459838867, + "learning_rate": 0.0005836788435679068, + "loss": 0.5113, + "step": 14977 + }, + { + "epoch": 0.8368299019470905, + "grad_norm": 0.37930211424827576, + "learning_rate": 0.000583650829224563, + "loss": 0.3498, + "step": 14978 + }, + { + "epoch": 0.8368857725507725, + "grad_norm": 0.5645079016685486, + "learning_rate": 0.0005836228148812191, + "loss": 0.3542, + "step": 14979 + }, + { + "epoch": 0.8369416431544543, + "grad_norm": 1.3135865926742554, + "learning_rate": 0.0005835948005378754, + "loss": 0.4984, + "step": 14980 + }, + { + "epoch": 0.8369975137581361, + "grad_norm": 0.3299943804740906, + "learning_rate": 0.0005835667861945317, + "loss": 0.3344, + "step": 14981 + }, + { + "epoch": 0.837053384361818, + "grad_norm": 0.8126201629638672, + "learning_rate": 0.0005835387718511879, + "loss": 0.5877, + "step": 14982 + }, + { + "epoch": 0.8371092549654999, + "grad_norm": 0.4021969437599182, + "learning_rate": 0.000583510757507844, + "loss": 0.4558, + "step": 14983 + }, + { + "epoch": 0.8371651255691818, + "grad_norm": 0.3913135826587677, + "learning_rate": 0.0005834827431645002, + "loss": 0.3839, + "step": 14984 + }, + { + "epoch": 0.8372209961728636, + "grad_norm": 0.4719770550727844, + "learning_rate": 0.0005834547288211565, + "loss": 0.4796, + "step": 14985 + }, + { + "epoch": 0.8372768667765456, + "grad_norm": 0.35099321603775024, + "learning_rate": 0.0005834267144778127, + "loss": 0.3828, + "step": 14986 + }, + { + "epoch": 0.8373327373802274, + "grad_norm": 0.39432477951049805, + "learning_rate": 0.0005833987001344688, + "loss": 0.4786, + "step": 14987 + }, + { + "epoch": 0.8373886079839092, + "grad_norm": 0.4623650312423706, + "learning_rate": 0.0005833706857911251, + "loss": 0.4178, + "step": 14988 + }, + { + "epoch": 0.8374444785875912, + "grad_norm": 0.479951411485672, + "learning_rate": 0.0005833426714477813, + "loss": 0.3565, + "step": 14989 + }, + { + "epoch": 0.837500349191273, + "grad_norm": 0.7222636342048645, + "learning_rate": 0.0005833146571044374, + "loss": 0.4865, + "step": 14990 + }, + { + "epoch": 0.8375562197949549, + "grad_norm": 0.6043869256973267, + "learning_rate": 0.0005832866427610936, + "loss": 0.3303, + "step": 14991 + }, + { + "epoch": 0.8376120903986367, + "grad_norm": 0.43930840492248535, + "learning_rate": 0.0005832586284177499, + "loss": 0.3749, + "step": 14992 + }, + { + "epoch": 0.8376679610023187, + "grad_norm": 0.5090981125831604, + "learning_rate": 0.0005832306140744061, + "loss": 0.3893, + "step": 14993 + }, + { + "epoch": 0.8377238316060005, + "grad_norm": 0.50479656457901, + "learning_rate": 0.0005832025997310622, + "loss": 0.4602, + "step": 14994 + }, + { + "epoch": 0.8377797022096823, + "grad_norm": 0.3970533013343811, + "learning_rate": 0.0005831745853877185, + "loss": 0.3952, + "step": 14995 + }, + { + "epoch": 0.8378355728133643, + "grad_norm": 0.44813641905784607, + "learning_rate": 0.0005831465710443748, + "loss": 0.4903, + "step": 14996 + }, + { + "epoch": 0.8378914434170461, + "grad_norm": Infinity, + "learning_rate": 0.0005831465710443748, + "loss": 0.4702, + "step": 14997 + }, + { + "epoch": 0.837947314020728, + "grad_norm": 0.42392411828041077, + "learning_rate": 0.000583118556701031, + "loss": 0.457, + "step": 14998 + }, + { + "epoch": 0.8380031846244098, + "grad_norm": 1.3480775356292725, + "learning_rate": 0.0005830905423576871, + "loss": 0.3714, + "step": 14999 + }, + { + "epoch": 0.8380590552280918, + "grad_norm": 0.9962452054023743, + "learning_rate": 0.0005830625280143434, + "loss": 0.3928, + "step": 15000 + }, + { + "epoch": 0.8380590552280918, + "eval_cer": 0.0886986787120989, + "eval_loss": 0.3336978554725647, + "eval_runtime": 56.0092, + "eval_samples_per_second": 81.022, + "eval_steps_per_second": 5.071, + "eval_wer": 0.34892880087148404, + "step": 15000 + }, + { + "epoch": 0.8381149258317736, + "grad_norm": 0.9815027117729187, + "learning_rate": 0.0005830345136709996, + "loss": 0.4186, + "step": 15001 + }, + { + "epoch": 0.8381707964354554, + "grad_norm": 0.44439542293548584, + "learning_rate": 0.0005830064993276558, + "loss": 0.4404, + "step": 15002 + }, + { + "epoch": 0.8382266670391374, + "grad_norm": 0.5624732375144958, + "learning_rate": 0.000582978484984312, + "loss": 0.4378, + "step": 15003 + }, + { + "epoch": 0.8382825376428192, + "grad_norm": 0.5731766819953918, + "learning_rate": 0.0005829504706409682, + "loss": 0.5239, + "step": 15004 + }, + { + "epoch": 0.8383384082465011, + "grad_norm": 0.8919277787208557, + "learning_rate": 0.0005829224562976244, + "loss": 0.4549, + "step": 15005 + }, + { + "epoch": 0.838394278850183, + "grad_norm": 0.36471977829933167, + "learning_rate": 0.0005828944419542806, + "loss": 0.3607, + "step": 15006 + }, + { + "epoch": 0.8384501494538649, + "grad_norm": 0.433077871799469, + "learning_rate": 0.0005828664276109368, + "loss": 0.4041, + "step": 15007 + }, + { + "epoch": 0.8385060200575467, + "grad_norm": 0.3599492013454437, + "learning_rate": 0.000582838413267593, + "loss": 0.2955, + "step": 15008 + }, + { + "epoch": 0.8385618906612285, + "grad_norm": 0.4911680221557617, + "learning_rate": 0.0005828103989242492, + "loss": 0.4832, + "step": 15009 + }, + { + "epoch": 0.8386177612649105, + "grad_norm": 0.6699753999710083, + "learning_rate": 0.0005827823845809055, + "loss": 0.3182, + "step": 15010 + }, + { + "epoch": 0.8386736318685923, + "grad_norm": 0.35436052083969116, + "learning_rate": 0.0005827543702375616, + "loss": 0.3827, + "step": 15011 + }, + { + "epoch": 0.8387295024722742, + "grad_norm": 0.5152038335800171, + "learning_rate": 0.0005827263558942178, + "loss": 0.4327, + "step": 15012 + }, + { + "epoch": 0.8387853730759561, + "grad_norm": 0.6827511191368103, + "learning_rate": 0.000582698341550874, + "loss": 0.5098, + "step": 15013 + }, + { + "epoch": 0.838841243679638, + "grad_norm": 0.48247480392456055, + "learning_rate": 0.0005826703272075303, + "loss": 0.3917, + "step": 15014 + }, + { + "epoch": 0.8388971142833198, + "grad_norm": 0.6690773963928223, + "learning_rate": 0.0005826423128641865, + "loss": 0.5077, + "step": 15015 + }, + { + "epoch": 0.8389529848870017, + "grad_norm": 0.5357568860054016, + "learning_rate": 0.0005826142985208427, + "loss": 0.5524, + "step": 15016 + }, + { + "epoch": 0.8390088554906836, + "grad_norm": 0.4460785984992981, + "learning_rate": 0.000582586284177499, + "loss": 0.4951, + "step": 15017 + }, + { + "epoch": 0.8390647260943654, + "grad_norm": 0.8157033920288086, + "learning_rate": 0.0005825582698341551, + "loss": 0.4257, + "step": 15018 + }, + { + "epoch": 0.8391205966980473, + "grad_norm": 0.5287913084030151, + "learning_rate": 0.0005825302554908113, + "loss": 0.5189, + "step": 15019 + }, + { + "epoch": 0.8391764673017292, + "grad_norm": 0.3843144178390503, + "learning_rate": 0.0005825022411474675, + "loss": 0.3725, + "step": 15020 + }, + { + "epoch": 0.8392323379054111, + "grad_norm": 0.495021790266037, + "learning_rate": 0.0005824742268041238, + "loss": 0.5309, + "step": 15021 + }, + { + "epoch": 0.8392882085090929, + "grad_norm": 0.5957879424095154, + "learning_rate": 0.0005824462124607799, + "loss": 0.6078, + "step": 15022 + }, + { + "epoch": 0.8393440791127749, + "grad_norm": 0.5911241769790649, + "learning_rate": 0.0005824181981174361, + "loss": 0.3479, + "step": 15023 + }, + { + "epoch": 0.8393999497164567, + "grad_norm": 0.3649478256702423, + "learning_rate": 0.0005823901837740923, + "loss": 0.3744, + "step": 15024 + }, + { + "epoch": 0.8394558203201385, + "grad_norm": 0.5660577416419983, + "learning_rate": 0.0005823621694307486, + "loss": 0.4149, + "step": 15025 + }, + { + "epoch": 0.8395116909238204, + "grad_norm": 0.5862229466438293, + "learning_rate": 0.0005823341550874047, + "loss": 0.4682, + "step": 15026 + }, + { + "epoch": 0.8395675615275023, + "grad_norm": 0.5154262781143188, + "learning_rate": 0.0005823061407440609, + "loss": 0.4683, + "step": 15027 + }, + { + "epoch": 0.8396234321311842, + "grad_norm": 0.5355793833732605, + "learning_rate": 0.0005822781264007172, + "loss": 0.4113, + "step": 15028 + }, + { + "epoch": 0.839679302734866, + "grad_norm": 0.4427765905857086, + "learning_rate": 0.0005822501120573734, + "loss": 0.3497, + "step": 15029 + }, + { + "epoch": 0.839735173338548, + "grad_norm": 0.3978462517261505, + "learning_rate": 0.0005822220977140295, + "loss": 0.4639, + "step": 15030 + }, + { + "epoch": 0.8397910439422298, + "grad_norm": 0.3940868079662323, + "learning_rate": 0.0005821940833706858, + "loss": 0.3889, + "step": 15031 + }, + { + "epoch": 0.8398469145459116, + "grad_norm": 0.6467211842536926, + "learning_rate": 0.0005821660690273421, + "loss": 0.4524, + "step": 15032 + }, + { + "epoch": 0.8399027851495935, + "grad_norm": 0.37111034989356995, + "learning_rate": 0.0005821380546839982, + "loss": 0.434, + "step": 15033 + }, + { + "epoch": 0.8399586557532754, + "grad_norm": 0.3533463478088379, + "learning_rate": 0.0005821100403406544, + "loss": 0.4317, + "step": 15034 + }, + { + "epoch": 0.8400145263569573, + "grad_norm": 0.7967902421951294, + "learning_rate": 0.0005820820259973107, + "loss": 0.3698, + "step": 15035 + }, + { + "epoch": 0.8400703969606391, + "grad_norm": 0.4312323331832886, + "learning_rate": 0.0005820540116539669, + "loss": 0.4455, + "step": 15036 + }, + { + "epoch": 0.8401262675643211, + "grad_norm": 0.4914570748806, + "learning_rate": 0.000582025997310623, + "loss": 0.4407, + "step": 15037 + }, + { + "epoch": 0.8401821381680029, + "grad_norm": 0.43962571024894714, + "learning_rate": 0.0005819979829672792, + "loss": 0.5616, + "step": 15038 + }, + { + "epoch": 0.8402380087716848, + "grad_norm": 0.5201588273048401, + "learning_rate": 0.0005819699686239355, + "loss": 0.3582, + "step": 15039 + }, + { + "epoch": 0.8402938793753667, + "grad_norm": 0.5597681403160095, + "learning_rate": 0.0005819419542805917, + "loss": 0.4566, + "step": 15040 + }, + { + "epoch": 0.8403497499790485, + "grad_norm": 0.4189048409461975, + "learning_rate": 0.0005819139399372478, + "loss": 0.3537, + "step": 15041 + }, + { + "epoch": 0.8404056205827304, + "grad_norm": 1.7070579528808594, + "learning_rate": 0.0005818859255939041, + "loss": 0.3162, + "step": 15042 + }, + { + "epoch": 0.8404614911864122, + "grad_norm": 0.4139566719532013, + "learning_rate": 0.0005818579112505603, + "loss": 0.3908, + "step": 15043 + }, + { + "epoch": 0.8405173617900942, + "grad_norm": 0.5718859434127808, + "learning_rate": 0.0005818298969072165, + "loss": 0.4689, + "step": 15044 + }, + { + "epoch": 0.840573232393776, + "grad_norm": 0.5090233683586121, + "learning_rate": 0.0005818018825638726, + "loss": 0.5736, + "step": 15045 + }, + { + "epoch": 0.8406291029974579, + "grad_norm": 0.43490004539489746, + "learning_rate": 0.000581773868220529, + "loss": 0.4062, + "step": 15046 + }, + { + "epoch": 0.8406849736011398, + "grad_norm": 0.32079604268074036, + "learning_rate": 0.0005817458538771852, + "loss": 0.431, + "step": 15047 + }, + { + "epoch": 0.8407408442048216, + "grad_norm": 0.4077328145503998, + "learning_rate": 0.0005817178395338414, + "loss": 0.4658, + "step": 15048 + }, + { + "epoch": 0.8407967148085035, + "grad_norm": 0.6749886274337769, + "learning_rate": 0.0005816898251904976, + "loss": 0.3846, + "step": 15049 + }, + { + "epoch": 0.8408525854121853, + "grad_norm": 0.7128700017929077, + "learning_rate": 0.0005816618108471538, + "loss": 0.4607, + "step": 15050 + }, + { + "epoch": 0.8409084560158673, + "grad_norm": 1.1581944227218628, + "learning_rate": 0.00058163379650381, + "loss": 0.4468, + "step": 15051 + }, + { + "epoch": 0.8409643266195491, + "grad_norm": 0.4506048858165741, + "learning_rate": 0.0005816057821604662, + "loss": 0.4429, + "step": 15052 + }, + { + "epoch": 0.841020197223231, + "grad_norm": 1.1571632623672485, + "learning_rate": 0.0005815777678171224, + "loss": 0.4367, + "step": 15053 + }, + { + "epoch": 0.8410760678269129, + "grad_norm": 0.3846244215965271, + "learning_rate": 0.0005815497534737786, + "loss": 0.4245, + "step": 15054 + }, + { + "epoch": 0.8411319384305947, + "grad_norm": 0.45383962988853455, + "learning_rate": 0.0005815217391304348, + "loss": 0.4784, + "step": 15055 + }, + { + "epoch": 0.8411878090342766, + "grad_norm": 0.5239003300666809, + "learning_rate": 0.0005814937247870909, + "loss": 0.5528, + "step": 15056 + }, + { + "epoch": 0.8412436796379585, + "grad_norm": 0.6229205131530762, + "learning_rate": 0.0005814657104437472, + "loss": 0.7067, + "step": 15057 + }, + { + "epoch": 0.8412995502416404, + "grad_norm": 1.0328432321548462, + "learning_rate": 0.0005814376961004034, + "loss": 0.5525, + "step": 15058 + }, + { + "epoch": 0.8413554208453222, + "grad_norm": 0.4060608446598053, + "learning_rate": 0.0005814096817570596, + "loss": 0.4954, + "step": 15059 + }, + { + "epoch": 0.8414112914490041, + "grad_norm": 0.5572602152824402, + "learning_rate": 0.0005813816674137158, + "loss": 0.3867, + "step": 15060 + }, + { + "epoch": 0.841467162052686, + "grad_norm": 0.8521888852119446, + "learning_rate": 0.000581353653070372, + "loss": 0.4001, + "step": 15061 + }, + { + "epoch": 0.8415230326563679, + "grad_norm": 0.3787970542907715, + "learning_rate": 0.0005813256387270282, + "loss": 0.3761, + "step": 15062 + }, + { + "epoch": 0.8415789032600497, + "grad_norm": 0.5883147716522217, + "learning_rate": 0.0005812976243836844, + "loss": 0.5155, + "step": 15063 + }, + { + "epoch": 0.8416347738637316, + "grad_norm": 0.5559608340263367, + "learning_rate": 0.0005812696100403407, + "loss": 0.3988, + "step": 15064 + }, + { + "epoch": 0.8416906444674135, + "grad_norm": 0.7778342366218567, + "learning_rate": 0.0005812415956969969, + "loss": 0.4777, + "step": 15065 + }, + { + "epoch": 0.8417465150710953, + "grad_norm": 0.5511471629142761, + "learning_rate": 0.0005812135813536531, + "loss": 0.4208, + "step": 15066 + }, + { + "epoch": 0.8418023856747772, + "grad_norm": 6.471615791320801, + "learning_rate": 0.0005811855670103094, + "loss": 0.3531, + "step": 15067 + }, + { + "epoch": 0.8418582562784591, + "grad_norm": 1.06554114818573, + "learning_rate": 0.0005811575526669655, + "loss": 0.3763, + "step": 15068 + }, + { + "epoch": 0.841914126882141, + "grad_norm": 0.4352147579193115, + "learning_rate": 0.0005811295383236217, + "loss": 0.4248, + "step": 15069 + }, + { + "epoch": 0.8419699974858228, + "grad_norm": 1.632891297340393, + "learning_rate": 0.0005811015239802779, + "loss": 0.3764, + "step": 15070 + }, + { + "epoch": 0.8420258680895047, + "grad_norm": 1.4002124071121216, + "learning_rate": 0.0005810735096369342, + "loss": 0.4575, + "step": 15071 + }, + { + "epoch": 0.8420817386931866, + "grad_norm": 0.46669283509254456, + "learning_rate": 0.0005810454952935903, + "loss": 0.439, + "step": 15072 + }, + { + "epoch": 0.8421376092968684, + "grad_norm": 0.5133165717124939, + "learning_rate": 0.0005810174809502465, + "loss": 0.4538, + "step": 15073 + }, + { + "epoch": 0.8421934799005504, + "grad_norm": 1.312511682510376, + "learning_rate": 0.0005809894666069028, + "loss": 0.4197, + "step": 15074 + }, + { + "epoch": 0.8422493505042322, + "grad_norm": 0.49738752841949463, + "learning_rate": 0.000580961452263559, + "loss": 0.4887, + "step": 15075 + }, + { + "epoch": 0.8423052211079141, + "grad_norm": 0.49836045503616333, + "learning_rate": 0.0005809334379202151, + "loss": 0.4456, + "step": 15076 + }, + { + "epoch": 0.8423610917115959, + "grad_norm": 0.5526162385940552, + "learning_rate": 0.0005809054235768713, + "loss": 0.4144, + "step": 15077 + }, + { + "epoch": 0.8424169623152779, + "grad_norm": 1.043245553970337, + "learning_rate": 0.0005808774092335276, + "loss": 0.4742, + "step": 15078 + }, + { + "epoch": 0.8424728329189597, + "grad_norm": 0.5071491003036499, + "learning_rate": 0.0005808493948901837, + "loss": 0.3809, + "step": 15079 + }, + { + "epoch": 0.8425287035226415, + "grad_norm": 0.3873586654663086, + "learning_rate": 0.0005808213805468399, + "loss": 0.4456, + "step": 15080 + }, + { + "epoch": 0.8425845741263235, + "grad_norm": 1.3335319757461548, + "learning_rate": 0.0005807933662034963, + "loss": 0.5053, + "step": 15081 + }, + { + "epoch": 0.8426404447300053, + "grad_norm": 0.5985936522483826, + "learning_rate": 0.0005807653518601525, + "loss": 0.424, + "step": 15082 + }, + { + "epoch": 0.8426963153336872, + "grad_norm": 0.35307031869888306, + "learning_rate": 0.0005807373375168086, + "loss": 0.366, + "step": 15083 + }, + { + "epoch": 0.842752185937369, + "grad_norm": 0.5393702387809753, + "learning_rate": 0.0005807093231734648, + "loss": 0.4616, + "step": 15084 + }, + { + "epoch": 0.842808056541051, + "grad_norm": 0.5981107950210571, + "learning_rate": 0.0005806813088301211, + "loss": 0.3864, + "step": 15085 + }, + { + "epoch": 0.8428639271447328, + "grad_norm": 6.319785118103027, + "learning_rate": 0.0005806532944867773, + "loss": 0.4601, + "step": 15086 + }, + { + "epoch": 0.8429197977484146, + "grad_norm": 0.5819458365440369, + "learning_rate": 0.0005806252801434334, + "loss": 0.4187, + "step": 15087 + }, + { + "epoch": 0.8429756683520966, + "grad_norm": 0.9158802032470703, + "learning_rate": 0.0005805972658000897, + "loss": 0.5406, + "step": 15088 + }, + { + "epoch": 0.8430315389557784, + "grad_norm": 0.38979196548461914, + "learning_rate": 0.0005805692514567459, + "loss": 0.5245, + "step": 15089 + }, + { + "epoch": 0.8430874095594603, + "grad_norm": 0.4979667365550995, + "learning_rate": 0.0005805412371134021, + "loss": 0.4226, + "step": 15090 + }, + { + "epoch": 0.8431432801631422, + "grad_norm": 0.4457850754261017, + "learning_rate": 0.0005805132227700582, + "loss": 0.405, + "step": 15091 + }, + { + "epoch": 0.8431991507668241, + "grad_norm": 0.5032418370246887, + "learning_rate": 0.0005804852084267145, + "loss": 0.4179, + "step": 15092 + }, + { + "epoch": 0.8432550213705059, + "grad_norm": 0.3584097623825073, + "learning_rate": 0.0005804571940833707, + "loss": 0.3593, + "step": 15093 + }, + { + "epoch": 0.8433108919741877, + "grad_norm": 0.6177075505256653, + "learning_rate": 0.0005804291797400269, + "loss": 0.4799, + "step": 15094 + }, + { + "epoch": 0.8433667625778697, + "grad_norm": 1.285349726676941, + "learning_rate": 0.000580401165396683, + "loss": 0.4343, + "step": 15095 + }, + { + "epoch": 0.8434226331815515, + "grad_norm": 1.122065544128418, + "learning_rate": 0.0005803731510533393, + "loss": 0.446, + "step": 15096 + }, + { + "epoch": 0.8434785037852334, + "grad_norm": 0.6170968413352966, + "learning_rate": 0.0005803451367099956, + "loss": 0.453, + "step": 15097 + }, + { + "epoch": 0.8435343743889153, + "grad_norm": 0.7411781549453735, + "learning_rate": 0.0005803171223666516, + "loss": 0.6158, + "step": 15098 + }, + { + "epoch": 0.8435902449925972, + "grad_norm": 0.5827665328979492, + "learning_rate": 0.000580289108023308, + "loss": 0.3955, + "step": 15099 + }, + { + "epoch": 0.843646115596279, + "grad_norm": 0.48470333218574524, + "learning_rate": 0.0005802610936799642, + "loss": 0.4396, + "step": 15100 + }, + { + "epoch": 0.8437019861999608, + "grad_norm": 0.4223266541957855, + "learning_rate": 0.0005802330793366204, + "loss": 0.3939, + "step": 15101 + }, + { + "epoch": 0.8437578568036428, + "grad_norm": 0.6328983306884766, + "learning_rate": 0.0005802050649932765, + "loss": 0.4695, + "step": 15102 + }, + { + "epoch": 0.8438137274073246, + "grad_norm": 0.568468451499939, + "learning_rate": 0.0005801770506499328, + "loss": 0.5497, + "step": 15103 + }, + { + "epoch": 0.8438695980110065, + "grad_norm": 0.4492528438568115, + "learning_rate": 0.000580149036306589, + "loss": 0.4839, + "step": 15104 + }, + { + "epoch": 0.8439254686146884, + "grad_norm": 0.3742114007472992, + "learning_rate": 0.0005801210219632452, + "loss": 0.3905, + "step": 15105 + }, + { + "epoch": 0.8439813392183703, + "grad_norm": 0.5725758671760559, + "learning_rate": 0.0005800930076199014, + "loss": 0.5468, + "step": 15106 + }, + { + "epoch": 0.8440372098220521, + "grad_norm": 0.5436546206474304, + "learning_rate": 0.0005800649932765576, + "loss": 0.4478, + "step": 15107 + }, + { + "epoch": 0.844093080425734, + "grad_norm": 2.5672714710235596, + "learning_rate": 0.0005800369789332138, + "loss": 0.5397, + "step": 15108 + }, + { + "epoch": 0.8441489510294159, + "grad_norm": 1.3945249319076538, + "learning_rate": 0.00058000896458987, + "loss": 0.4408, + "step": 15109 + }, + { + "epoch": 0.8442048216330977, + "grad_norm": 0.6180221438407898, + "learning_rate": 0.0005799809502465262, + "loss": 0.4583, + "step": 15110 + }, + { + "epoch": 0.8442606922367796, + "grad_norm": 0.7271673679351807, + "learning_rate": 0.0005799529359031824, + "loss": 0.4242, + "step": 15111 + }, + { + "epoch": 0.8443165628404615, + "grad_norm": 0.5434330105781555, + "learning_rate": 0.0005799249215598386, + "loss": 0.5748, + "step": 15112 + }, + { + "epoch": 0.8443724334441434, + "grad_norm": 0.5204839706420898, + "learning_rate": 0.000579896907216495, + "loss": 0.5472, + "step": 15113 + }, + { + "epoch": 0.8444283040478252, + "grad_norm": 0.6124056577682495, + "learning_rate": 0.000579868892873151, + "loss": 0.42, + "step": 15114 + }, + { + "epoch": 0.8444841746515072, + "grad_norm": 0.5149624347686768, + "learning_rate": 0.0005798408785298073, + "loss": 0.4949, + "step": 15115 + }, + { + "epoch": 0.844540045255189, + "grad_norm": 0.396028995513916, + "learning_rate": 0.0005798128641864635, + "loss": 0.3782, + "step": 15116 + }, + { + "epoch": 0.8445959158588708, + "grad_norm": 0.5165128111839294, + "learning_rate": 0.0005797848498431198, + "loss": 0.4795, + "step": 15117 + }, + { + "epoch": 0.8446517864625527, + "grad_norm": 0.34048712253570557, + "learning_rate": 0.0005797568354997759, + "loss": 0.4133, + "step": 15118 + }, + { + "epoch": 0.8447076570662346, + "grad_norm": 0.8641983866691589, + "learning_rate": 0.0005797288211564321, + "loss": 0.4135, + "step": 15119 + }, + { + "epoch": 0.8447635276699165, + "grad_norm": 0.569739043712616, + "learning_rate": 0.0005797008068130884, + "loss": 0.4482, + "step": 15120 + }, + { + "epoch": 0.8448193982735983, + "grad_norm": 0.9858090877532959, + "learning_rate": 0.0005796727924697445, + "loss": 0.414, + "step": 15121 + }, + { + "epoch": 0.8448752688772803, + "grad_norm": 0.5736381411552429, + "learning_rate": 0.0005796447781264007, + "loss": 0.5171, + "step": 15122 + }, + { + "epoch": 0.8449311394809621, + "grad_norm": 0.4731729030609131, + "learning_rate": 0.0005796167637830569, + "loss": 0.3459, + "step": 15123 + }, + { + "epoch": 0.8449870100846439, + "grad_norm": 0.8031646609306335, + "learning_rate": 0.0005795887494397132, + "loss": 0.4523, + "step": 15124 + }, + { + "epoch": 0.8450428806883259, + "grad_norm": 0.35747313499450684, + "learning_rate": 0.0005795607350963693, + "loss": 0.3626, + "step": 15125 + }, + { + "epoch": 0.8450987512920077, + "grad_norm": 0.5262337923049927, + "learning_rate": 0.0005795327207530255, + "loss": 0.4678, + "step": 15126 + }, + { + "epoch": 0.8451546218956896, + "grad_norm": 0.37621167302131653, + "learning_rate": 0.0005795047064096818, + "loss": 0.3616, + "step": 15127 + }, + { + "epoch": 0.8452104924993714, + "grad_norm": 0.4049895107746124, + "learning_rate": 0.000579476692066338, + "loss": 0.5555, + "step": 15128 + }, + { + "epoch": 0.8452663631030534, + "grad_norm": 0.4097277522087097, + "learning_rate": 0.0005794486777229941, + "loss": 0.4072, + "step": 15129 + }, + { + "epoch": 0.8453222337067352, + "grad_norm": 0.42606455087661743, + "learning_rate": 0.0005794206633796503, + "loss": 0.3783, + "step": 15130 + }, + { + "epoch": 0.845378104310417, + "grad_norm": 0.6451665163040161, + "learning_rate": 0.0005793926490363067, + "loss": 0.4917, + "step": 15131 + }, + { + "epoch": 0.845433974914099, + "grad_norm": 0.4678861200809479, + "learning_rate": 0.0005793646346929629, + "loss": 0.375, + "step": 15132 + }, + { + "epoch": 0.8454898455177808, + "grad_norm": 0.4203849136829376, + "learning_rate": 0.000579336620349619, + "loss": 0.4065, + "step": 15133 + }, + { + "epoch": 0.8455457161214627, + "grad_norm": 0.4906863868236542, + "learning_rate": 0.0005793086060062752, + "loss": 0.3841, + "step": 15134 + }, + { + "epoch": 0.8456015867251445, + "grad_norm": 2.6910033226013184, + "learning_rate": 0.0005792805916629315, + "loss": 0.392, + "step": 15135 + }, + { + "epoch": 0.8456574573288265, + "grad_norm": 0.6538512706756592, + "learning_rate": 0.0005792525773195877, + "loss": 0.3703, + "step": 15136 + }, + { + "epoch": 0.8457133279325083, + "grad_norm": 0.7447658777236938, + "learning_rate": 0.0005792245629762438, + "loss": 0.3884, + "step": 15137 + }, + { + "epoch": 0.8457691985361901, + "grad_norm": 1.015478491783142, + "learning_rate": 0.0005791965486329001, + "loss": 0.3851, + "step": 15138 + }, + { + "epoch": 0.8458250691398721, + "grad_norm": 0.46300429105758667, + "learning_rate": 0.0005791685342895563, + "loss": 0.3624, + "step": 15139 + }, + { + "epoch": 0.8458809397435539, + "grad_norm": 0.5395675897598267, + "learning_rate": 0.0005791405199462124, + "loss": 0.4376, + "step": 15140 + }, + { + "epoch": 0.8459368103472358, + "grad_norm": 0.952738881111145, + "learning_rate": 0.0005791125056028686, + "loss": 0.7217, + "step": 15141 + }, + { + "epoch": 0.8459926809509176, + "grad_norm": 0.41212838888168335, + "learning_rate": 0.0005790844912595249, + "loss": 0.3209, + "step": 15142 + }, + { + "epoch": 0.8460485515545996, + "grad_norm": 0.48823681473731995, + "learning_rate": 0.0005790564769161811, + "loss": 0.5191, + "step": 15143 + }, + { + "epoch": 0.8461044221582814, + "grad_norm": 0.40681615471839905, + "learning_rate": 0.0005790284625728372, + "loss": 0.4495, + "step": 15144 + }, + { + "epoch": 0.8461602927619633, + "grad_norm": 0.4633169174194336, + "learning_rate": 0.0005790004482294935, + "loss": 0.4598, + "step": 15145 + }, + { + "epoch": 0.8462161633656452, + "grad_norm": 0.5090947151184082, + "learning_rate": 0.0005789724338861497, + "loss": 0.5679, + "step": 15146 + }, + { + "epoch": 0.846272033969327, + "grad_norm": 0.5290205478668213, + "learning_rate": 0.000578944419542806, + "loss": 0.4484, + "step": 15147 + }, + { + "epoch": 0.8463279045730089, + "grad_norm": 0.4995565414428711, + "learning_rate": 0.000578916405199462, + "loss": 0.3841, + "step": 15148 + }, + { + "epoch": 0.8463837751766908, + "grad_norm": 5.997913360595703, + "learning_rate": 0.0005788883908561184, + "loss": 0.4156, + "step": 15149 + }, + { + "epoch": 0.8464396457803727, + "grad_norm": 0.5214009881019592, + "learning_rate": 0.0005788603765127746, + "loss": 0.5003, + "step": 15150 + }, + { + "epoch": 0.8464955163840545, + "grad_norm": 0.47460824251174927, + "learning_rate": 0.0005788323621694308, + "loss": 0.4896, + "step": 15151 + }, + { + "epoch": 0.8465513869877364, + "grad_norm": 0.4407256543636322, + "learning_rate": 0.000578804347826087, + "loss": 0.4064, + "step": 15152 + }, + { + "epoch": 0.8466072575914183, + "grad_norm": 1.8230875730514526, + "learning_rate": 0.0005787763334827432, + "loss": 0.4314, + "step": 15153 + }, + { + "epoch": 0.8466631281951001, + "grad_norm": 1.442104458808899, + "learning_rate": 0.0005787483191393994, + "loss": 0.476, + "step": 15154 + }, + { + "epoch": 0.846718998798782, + "grad_norm": 0.438617467880249, + "learning_rate": 0.0005787203047960556, + "loss": 0.3845, + "step": 15155 + }, + { + "epoch": 0.8467748694024639, + "grad_norm": 0.5659894943237305, + "learning_rate": 0.0005786922904527118, + "loss": 0.4299, + "step": 15156 + }, + { + "epoch": 0.8468307400061458, + "grad_norm": 0.5777814984321594, + "learning_rate": 0.000578664276109368, + "loss": 0.5052, + "step": 15157 + }, + { + "epoch": 0.8468866106098276, + "grad_norm": 0.3744737505912781, + "learning_rate": 0.0005786362617660242, + "loss": 0.4269, + "step": 15158 + }, + { + "epoch": 0.8469424812135095, + "grad_norm": 0.49723726511001587, + "learning_rate": 0.0005786082474226805, + "loss": 0.4071, + "step": 15159 + }, + { + "epoch": 0.8469983518171914, + "grad_norm": 0.4691278040409088, + "learning_rate": 0.0005785802330793366, + "loss": 0.3843, + "step": 15160 + }, + { + "epoch": 0.8470542224208732, + "grad_norm": 0.3268478214740753, + "learning_rate": 0.0005785522187359928, + "loss": 0.3371, + "step": 15161 + }, + { + "epoch": 0.8471100930245551, + "grad_norm": 0.753089964389801, + "learning_rate": 0.000578524204392649, + "loss": 0.4714, + "step": 15162 + }, + { + "epoch": 0.847165963628237, + "grad_norm": 0.5763358473777771, + "learning_rate": 0.0005784961900493052, + "loss": 0.5633, + "step": 15163 + }, + { + "epoch": 0.8472218342319189, + "grad_norm": 0.4324769079685211, + "learning_rate": 0.0005784681757059615, + "loss": 0.4006, + "step": 15164 + }, + { + "epoch": 0.8472777048356007, + "grad_norm": 2.722118616104126, + "learning_rate": 0.0005784401613626177, + "loss": 0.4359, + "step": 15165 + }, + { + "epoch": 0.8473335754392827, + "grad_norm": 0.5163480043411255, + "learning_rate": 0.0005784121470192739, + "loss": 0.4649, + "step": 15166 + }, + { + "epoch": 0.8473894460429645, + "grad_norm": 2.2998766899108887, + "learning_rate": 0.0005783841326759301, + "loss": 0.3986, + "step": 15167 + }, + { + "epoch": 0.8474453166466464, + "grad_norm": 0.7606044411659241, + "learning_rate": 0.0005783561183325863, + "loss": 0.398, + "step": 15168 + }, + { + "epoch": 0.8475011872503282, + "grad_norm": 0.492575466632843, + "learning_rate": 0.0005783281039892425, + "loss": 0.3635, + "step": 15169 + }, + { + "epoch": 0.8475570578540101, + "grad_norm": 1.0617002248764038, + "learning_rate": 0.0005783000896458988, + "loss": 0.6817, + "step": 15170 + }, + { + "epoch": 0.847612928457692, + "grad_norm": 0.3440079092979431, + "learning_rate": 0.0005782720753025549, + "loss": 0.449, + "step": 15171 + }, + { + "epoch": 0.8476687990613738, + "grad_norm": 0.47734788060188293, + "learning_rate": 0.0005782440609592111, + "loss": 0.3821, + "step": 15172 + }, + { + "epoch": 0.8477246696650558, + "grad_norm": 0.3852957785129547, + "learning_rate": 0.0005782160466158673, + "loss": 0.4557, + "step": 15173 + }, + { + "epoch": 0.8477805402687376, + "grad_norm": 0.6616795659065247, + "learning_rate": 0.0005781880322725236, + "loss": 0.3633, + "step": 15174 + }, + { + "epoch": 0.8478364108724195, + "grad_norm": 0.6865038871765137, + "learning_rate": 0.0005781600179291797, + "loss": 0.4987, + "step": 15175 + }, + { + "epoch": 0.8478922814761013, + "grad_norm": 0.40377500653266907, + "learning_rate": 0.0005781320035858359, + "loss": 0.3209, + "step": 15176 + }, + { + "epoch": 0.8479481520797832, + "grad_norm": 0.35087108612060547, + "learning_rate": 0.0005781039892424922, + "loss": 0.2867, + "step": 15177 + }, + { + "epoch": 0.8480040226834651, + "grad_norm": 0.49747297167778015, + "learning_rate": 0.0005780759748991484, + "loss": 0.5278, + "step": 15178 + }, + { + "epoch": 0.8480598932871469, + "grad_norm": 0.5273075103759766, + "learning_rate": 0.0005780479605558045, + "loss": 0.4855, + "step": 15179 + }, + { + "epoch": 0.8481157638908289, + "grad_norm": 0.5617144703865051, + "learning_rate": 0.0005780199462124607, + "loss": 0.6241, + "step": 15180 + }, + { + "epoch": 0.8481716344945107, + "grad_norm": 0.4916396737098694, + "learning_rate": 0.0005779919318691171, + "loss": 0.3771, + "step": 15181 + }, + { + "epoch": 0.8482275050981926, + "grad_norm": 0.6417993307113647, + "learning_rate": 0.0005779639175257732, + "loss": 0.4482, + "step": 15182 + }, + { + "epoch": 0.8482833757018745, + "grad_norm": 0.35986751317977905, + "learning_rate": 0.0005779359031824294, + "loss": 0.4339, + "step": 15183 + }, + { + "epoch": 0.8483392463055563, + "grad_norm": 0.8579294085502625, + "learning_rate": 0.0005779078888390857, + "loss": 0.5417, + "step": 15184 + }, + { + "epoch": 0.8483951169092382, + "grad_norm": 0.42219334840774536, + "learning_rate": 0.0005778798744957419, + "loss": 0.3768, + "step": 15185 + }, + { + "epoch": 0.84845098751292, + "grad_norm": 0.5599267482757568, + "learning_rate": 0.000577851860152398, + "loss": 0.5632, + "step": 15186 + }, + { + "epoch": 0.848506858116602, + "grad_norm": 0.3659818470478058, + "learning_rate": 0.0005778238458090542, + "loss": 0.4096, + "step": 15187 + }, + { + "epoch": 0.8485627287202838, + "grad_norm": 0.6000370383262634, + "learning_rate": 0.0005777958314657105, + "loss": 0.4928, + "step": 15188 + }, + { + "epoch": 0.8486185993239657, + "grad_norm": 0.702105700969696, + "learning_rate": 0.0005777678171223667, + "loss": 0.487, + "step": 15189 + }, + { + "epoch": 0.8486744699276476, + "grad_norm": 0.49233174324035645, + "learning_rate": 0.0005777398027790228, + "loss": 0.4123, + "step": 15190 + }, + { + "epoch": 0.8487303405313295, + "grad_norm": 0.7289968729019165, + "learning_rate": 0.0005777117884356791, + "loss": 0.6157, + "step": 15191 + }, + { + "epoch": 0.8487862111350113, + "grad_norm": 0.6613783240318298, + "learning_rate": 0.0005776837740923353, + "loss": 0.4859, + "step": 15192 + }, + { + "epoch": 0.8488420817386931, + "grad_norm": 0.3322209119796753, + "learning_rate": 0.0005776557597489915, + "loss": 0.3243, + "step": 15193 + }, + { + "epoch": 0.8488979523423751, + "grad_norm": 0.5234158039093018, + "learning_rate": 0.0005776277454056476, + "loss": 0.4096, + "step": 15194 + }, + { + "epoch": 0.8489538229460569, + "grad_norm": 0.38436248898506165, + "learning_rate": 0.0005775997310623039, + "loss": 0.3924, + "step": 15195 + }, + { + "epoch": 0.8490096935497388, + "grad_norm": 0.35384389758110046, + "learning_rate": 0.0005775717167189601, + "loss": 0.3691, + "step": 15196 + }, + { + "epoch": 0.8490655641534207, + "grad_norm": 0.6691365838050842, + "learning_rate": 0.0005775437023756164, + "loss": 0.5019, + "step": 15197 + }, + { + "epoch": 0.8491214347571026, + "grad_norm": 0.5314581394195557, + "learning_rate": 0.0005775156880322724, + "loss": 0.513, + "step": 15198 + }, + { + "epoch": 0.8491773053607844, + "grad_norm": 0.7930400967597961, + "learning_rate": 0.0005774876736889288, + "loss": 0.4098, + "step": 15199 + }, + { + "epoch": 0.8492331759644663, + "grad_norm": 0.36648401618003845, + "learning_rate": 0.000577459659345585, + "loss": 0.3931, + "step": 15200 + }, + { + "epoch": 0.8492890465681482, + "grad_norm": 0.39929357171058655, + "learning_rate": 0.0005774316450022412, + "loss": 0.3813, + "step": 15201 + }, + { + "epoch": 0.84934491717183, + "grad_norm": 0.43765586614608765, + "learning_rate": 0.0005774036306588974, + "loss": 0.4486, + "step": 15202 + }, + { + "epoch": 0.8494007877755119, + "grad_norm": 3.9585890769958496, + "learning_rate": 0.0005773756163155536, + "loss": 0.3739, + "step": 15203 + }, + { + "epoch": 0.8494566583791938, + "grad_norm": 0.6577576398849487, + "learning_rate": 0.0005773476019722098, + "loss": 0.4961, + "step": 15204 + }, + { + "epoch": 0.8495125289828757, + "grad_norm": 0.7150046825408936, + "learning_rate": 0.0005773195876288659, + "loss": 0.416, + "step": 15205 + }, + { + "epoch": 0.8495683995865575, + "grad_norm": 2.318979501724243, + "learning_rate": 0.0005772915732855222, + "loss": 0.403, + "step": 15206 + }, + { + "epoch": 0.8496242701902395, + "grad_norm": 0.34829166531562805, + "learning_rate": 0.0005772635589421784, + "loss": 0.3615, + "step": 15207 + }, + { + "epoch": 0.8496801407939213, + "grad_norm": 0.339344322681427, + "learning_rate": 0.0005772355445988346, + "loss": 0.3703, + "step": 15208 + }, + { + "epoch": 0.8497360113976031, + "grad_norm": 0.3386315107345581, + "learning_rate": 0.0005772075302554908, + "loss": 0.403, + "step": 15209 + }, + { + "epoch": 0.849791882001285, + "grad_norm": 0.44034552574157715, + "learning_rate": 0.000577179515912147, + "loss": 0.3893, + "step": 15210 + }, + { + "epoch": 0.8498477526049669, + "grad_norm": 0.519141435623169, + "learning_rate": 0.0005771515015688032, + "loss": 0.4019, + "step": 15211 + }, + { + "epoch": 0.8499036232086488, + "grad_norm": 0.44443774223327637, + "learning_rate": 0.0005771234872254594, + "loss": 0.4469, + "step": 15212 + }, + { + "epoch": 0.8499594938123306, + "grad_norm": 0.5289946794509888, + "learning_rate": 0.0005770954728821156, + "loss": 0.4486, + "step": 15213 + }, + { + "epoch": 0.8500153644160126, + "grad_norm": 0.4469730854034424, + "learning_rate": 0.0005770674585387718, + "loss": 0.5469, + "step": 15214 + }, + { + "epoch": 0.8500712350196944, + "grad_norm": 0.3996174931526184, + "learning_rate": 0.0005770394441954281, + "loss": 0.3916, + "step": 15215 + }, + { + "epoch": 0.8501271056233762, + "grad_norm": 0.4481546878814697, + "learning_rate": 0.0005770114298520844, + "loss": 0.4267, + "step": 15216 + }, + { + "epoch": 0.8501829762270582, + "grad_norm": 0.38839149475097656, + "learning_rate": 0.0005769834155087405, + "loss": 0.5166, + "step": 15217 + }, + { + "epoch": 0.85023884683074, + "grad_norm": 1.4960763454437256, + "learning_rate": 0.0005769554011653967, + "loss": 0.5303, + "step": 15218 + }, + { + "epoch": 0.8502947174344219, + "grad_norm": 0.39706096053123474, + "learning_rate": 0.0005769273868220529, + "loss": 0.4653, + "step": 15219 + }, + { + "epoch": 0.8503505880381037, + "grad_norm": 0.43443870544433594, + "learning_rate": 0.0005768993724787092, + "loss": 0.3488, + "step": 15220 + }, + { + "epoch": 0.8504064586417857, + "grad_norm": 0.39041730761528015, + "learning_rate": 0.0005768713581353653, + "loss": 0.4032, + "step": 15221 + }, + { + "epoch": 0.8504623292454675, + "grad_norm": 0.35675913095474243, + "learning_rate": 0.0005768433437920215, + "loss": 0.3405, + "step": 15222 + }, + { + "epoch": 0.8505181998491493, + "grad_norm": 0.5241184234619141, + "learning_rate": 0.0005768153294486778, + "loss": 0.4978, + "step": 15223 + }, + { + "epoch": 0.8505740704528313, + "grad_norm": 0.4203619658946991, + "learning_rate": 0.000576787315105334, + "loss": 0.4179, + "step": 15224 + }, + { + "epoch": 0.8506299410565131, + "grad_norm": 0.5330315232276917, + "learning_rate": 0.0005767593007619901, + "loss": 0.4924, + "step": 15225 + }, + { + "epoch": 0.850685811660195, + "grad_norm": 0.3402199447154999, + "learning_rate": 0.0005767312864186463, + "loss": 0.4664, + "step": 15226 + }, + { + "epoch": 0.8507416822638768, + "grad_norm": 0.5232788920402527, + "learning_rate": 0.0005767032720753026, + "loss": 0.4381, + "step": 15227 + }, + { + "epoch": 0.8507975528675588, + "grad_norm": 0.6165463924407959, + "learning_rate": 0.0005766752577319587, + "loss": 0.4464, + "step": 15228 + }, + { + "epoch": 0.8508534234712406, + "grad_norm": 0.8282992243766785, + "learning_rate": 0.0005766472433886149, + "loss": 0.5076, + "step": 15229 + }, + { + "epoch": 0.8509092940749224, + "grad_norm": 0.8339760303497314, + "learning_rate": 0.0005766192290452713, + "loss": 0.4195, + "step": 15230 + }, + { + "epoch": 0.8509651646786044, + "grad_norm": 0.4239168167114258, + "learning_rate": 0.0005765912147019275, + "loss": 0.4274, + "step": 15231 + }, + { + "epoch": 0.8510210352822862, + "grad_norm": 0.4132409393787384, + "learning_rate": 0.0005765632003585836, + "loss": 0.4931, + "step": 15232 + }, + { + "epoch": 0.8510769058859681, + "grad_norm": 0.5095260739326477, + "learning_rate": 0.0005765351860152398, + "loss": 0.434, + "step": 15233 + }, + { + "epoch": 0.85113277648965, + "grad_norm": 2.3199384212493896, + "learning_rate": 0.0005765071716718961, + "loss": 0.4583, + "step": 15234 + }, + { + "epoch": 0.8511886470933319, + "grad_norm": 0.5074347853660583, + "learning_rate": 0.0005764791573285523, + "loss": 0.4804, + "step": 15235 + }, + { + "epoch": 0.8512445176970137, + "grad_norm": 0.8785603046417236, + "learning_rate": 0.0005764511429852084, + "loss": 0.3778, + "step": 15236 + }, + { + "epoch": 0.8513003883006955, + "grad_norm": 0.6038690805435181, + "learning_rate": 0.0005764231286418646, + "loss": 0.618, + "step": 15237 + }, + { + "epoch": 0.8513562589043775, + "grad_norm": 0.8723292350769043, + "learning_rate": 0.0005763951142985209, + "loss": 0.5057, + "step": 15238 + }, + { + "epoch": 0.8514121295080593, + "grad_norm": 1.0981069803237915, + "learning_rate": 0.0005763670999551771, + "loss": 0.5287, + "step": 15239 + }, + { + "epoch": 0.8514680001117412, + "grad_norm": 0.40260541439056396, + "learning_rate": 0.0005763390856118332, + "loss": 0.411, + "step": 15240 + }, + { + "epoch": 0.8515238707154231, + "grad_norm": 0.5512939691543579, + "learning_rate": 0.0005763110712684895, + "loss": 0.3872, + "step": 15241 + }, + { + "epoch": 0.851579741319105, + "grad_norm": 0.5319110155105591, + "learning_rate": 0.0005762830569251457, + "loss": 0.3194, + "step": 15242 + }, + { + "epoch": 0.8516356119227868, + "grad_norm": 0.3418392539024353, + "learning_rate": 0.0005762550425818019, + "loss": 0.3508, + "step": 15243 + }, + { + "epoch": 0.8516914825264686, + "grad_norm": 0.6912153363227844, + "learning_rate": 0.000576227028238458, + "loss": 0.383, + "step": 15244 + }, + { + "epoch": 0.8517473531301506, + "grad_norm": 0.37081554532051086, + "learning_rate": 0.0005761990138951143, + "loss": 0.4324, + "step": 15245 + }, + { + "epoch": 0.8518032237338324, + "grad_norm": 0.7316744327545166, + "learning_rate": 0.0005761709995517705, + "loss": 0.5801, + "step": 15246 + }, + { + "epoch": 0.8518590943375143, + "grad_norm": 0.7758939862251282, + "learning_rate": 0.0005761429852084266, + "loss": 0.4622, + "step": 15247 + }, + { + "epoch": 0.8519149649411962, + "grad_norm": 0.5177994966506958, + "learning_rate": 0.000576114970865083, + "loss": 0.4318, + "step": 15248 + }, + { + "epoch": 0.8519708355448781, + "grad_norm": 1.7095259428024292, + "learning_rate": 0.0005760869565217392, + "loss": 0.4269, + "step": 15249 + }, + { + "epoch": 0.8520267061485599, + "grad_norm": 0.9957188963890076, + "learning_rate": 0.0005760589421783954, + "loss": 0.4725, + "step": 15250 + }, + { + "epoch": 0.8520825767522419, + "grad_norm": 0.5018069744110107, + "learning_rate": 0.0005760309278350515, + "loss": 0.5844, + "step": 15251 + }, + { + "epoch": 0.8521384473559237, + "grad_norm": 0.4428724944591522, + "learning_rate": 0.0005760029134917078, + "loss": 0.4353, + "step": 15252 + }, + { + "epoch": 0.8521943179596055, + "grad_norm": 0.4136973023414612, + "learning_rate": 0.000575974899148364, + "loss": 0.4074, + "step": 15253 + }, + { + "epoch": 0.8522501885632874, + "grad_norm": 0.9140169620513916, + "learning_rate": 0.0005759468848050202, + "loss": 0.4927, + "step": 15254 + }, + { + "epoch": 0.8523060591669693, + "grad_norm": 1.0212284326553345, + "learning_rate": 0.0005759188704616764, + "loss": 0.4565, + "step": 15255 + }, + { + "epoch": 0.8523619297706512, + "grad_norm": 0.6234208345413208, + "learning_rate": 0.0005758908561183326, + "loss": 0.4244, + "step": 15256 + }, + { + "epoch": 0.852417800374333, + "grad_norm": 0.33623629808425903, + "learning_rate": 0.0005758628417749888, + "loss": 0.3131, + "step": 15257 + }, + { + "epoch": 0.852473670978015, + "grad_norm": 0.3599293529987335, + "learning_rate": 0.000575834827431645, + "loss": 0.4633, + "step": 15258 + }, + { + "epoch": 0.8525295415816968, + "grad_norm": 0.5084460377693176, + "learning_rate": 0.0005758068130883012, + "loss": 0.4279, + "step": 15259 + }, + { + "epoch": 0.8525854121853786, + "grad_norm": 0.4287366271018982, + "learning_rate": 0.0005757787987449574, + "loss": 0.2971, + "step": 15260 + }, + { + "epoch": 0.8526412827890605, + "grad_norm": 1.38046395778656, + "learning_rate": 0.0005757507844016136, + "loss": 0.453, + "step": 15261 + }, + { + "epoch": 0.8526971533927424, + "grad_norm": 0.4331337809562683, + "learning_rate": 0.0005757227700582699, + "loss": 0.3827, + "step": 15262 + }, + { + "epoch": 0.8527530239964243, + "grad_norm": 0.5129404664039612, + "learning_rate": 0.000575694755714926, + "loss": 0.5631, + "step": 15263 + }, + { + "epoch": 0.8528088946001061, + "grad_norm": 0.5004639625549316, + "learning_rate": 0.0005756667413715822, + "loss": 0.3519, + "step": 15264 + }, + { + "epoch": 0.8528647652037881, + "grad_norm": 0.4388630986213684, + "learning_rate": 0.0005756387270282385, + "loss": 0.5046, + "step": 15265 + }, + { + "epoch": 0.8529206358074699, + "grad_norm": 0.5614573359489441, + "learning_rate": 0.0005756107126848948, + "loss": 0.3857, + "step": 15266 + }, + { + "epoch": 0.8529765064111517, + "grad_norm": 0.5045706629753113, + "learning_rate": 0.0005755826983415509, + "loss": 0.4936, + "step": 15267 + }, + { + "epoch": 0.8530323770148337, + "grad_norm": 0.49913179874420166, + "learning_rate": 0.0005755546839982071, + "loss": 0.477, + "step": 15268 + }, + { + "epoch": 0.8530882476185155, + "grad_norm": 0.8226732611656189, + "learning_rate": 0.0005755266696548634, + "loss": 0.4212, + "step": 15269 + }, + { + "epoch": 0.8531441182221974, + "grad_norm": 0.46405449509620667, + "learning_rate": 0.0005754986553115195, + "loss": 0.4636, + "step": 15270 + }, + { + "epoch": 0.8531999888258792, + "grad_norm": 0.45342960953712463, + "learning_rate": 0.0005754706409681757, + "loss": 0.3075, + "step": 15271 + }, + { + "epoch": 0.8532558594295612, + "grad_norm": 0.9978341460227966, + "learning_rate": 0.0005754426266248319, + "loss": 0.4437, + "step": 15272 + }, + { + "epoch": 0.853311730033243, + "grad_norm": 0.282858669757843, + "learning_rate": 0.0005754146122814882, + "loss": 0.3055, + "step": 15273 + }, + { + "epoch": 0.8533676006369249, + "grad_norm": 0.3889446556568146, + "learning_rate": 0.0005753865979381443, + "loss": 0.4072, + "step": 15274 + }, + { + "epoch": 0.8534234712406068, + "grad_norm": 0.5094013214111328, + "learning_rate": 0.0005753585835948005, + "loss": 0.5375, + "step": 15275 + }, + { + "epoch": 0.8534793418442886, + "grad_norm": 0.5551899075508118, + "learning_rate": 0.0005753305692514567, + "loss": 0.3999, + "step": 15276 + }, + { + "epoch": 0.8535352124479705, + "grad_norm": 0.47146981954574585, + "learning_rate": 0.000575302554908113, + "loss": 0.454, + "step": 15277 + }, + { + "epoch": 0.8535910830516523, + "grad_norm": 0.38793501257896423, + "learning_rate": 0.0005752745405647691, + "loss": 0.4502, + "step": 15278 + }, + { + "epoch": 0.8536469536553343, + "grad_norm": 1.2778829336166382, + "learning_rate": 0.0005752465262214253, + "loss": 0.4887, + "step": 15279 + }, + { + "epoch": 0.8537028242590161, + "grad_norm": 0.4838353991508484, + "learning_rate": 0.0005752185118780816, + "loss": 0.3905, + "step": 15280 + }, + { + "epoch": 0.853758694862698, + "grad_norm": 3.120619297027588, + "learning_rate": 0.0005751904975347379, + "loss": 0.592, + "step": 15281 + }, + { + "epoch": 0.8538145654663799, + "grad_norm": 0.5351167917251587, + "learning_rate": 0.000575162483191394, + "loss": 0.4243, + "step": 15282 + }, + { + "epoch": 0.8538704360700617, + "grad_norm": 0.48654428124427795, + "learning_rate": 0.0005751344688480502, + "loss": 0.3856, + "step": 15283 + }, + { + "epoch": 0.8539263066737436, + "grad_norm": 0.3940756022930145, + "learning_rate": 0.0005751064545047065, + "loss": 0.34, + "step": 15284 + }, + { + "epoch": 0.8539821772774255, + "grad_norm": 0.5167561173439026, + "learning_rate": 0.0005750784401613627, + "loss": 0.4487, + "step": 15285 + }, + { + "epoch": 0.8540380478811074, + "grad_norm": 0.40598127245903015, + "learning_rate": 0.0005750504258180188, + "loss": 0.4317, + "step": 15286 + }, + { + "epoch": 0.8540939184847892, + "grad_norm": 0.6319817304611206, + "learning_rate": 0.0005750224114746751, + "loss": 0.4455, + "step": 15287 + }, + { + "epoch": 0.8541497890884711, + "grad_norm": 1.059013843536377, + "learning_rate": 0.0005749943971313313, + "loss": 0.4947, + "step": 15288 + }, + { + "epoch": 0.854205659692153, + "grad_norm": 0.5813087821006775, + "learning_rate": 0.0005749663827879874, + "loss": 0.4422, + "step": 15289 + }, + { + "epoch": 0.8542615302958348, + "grad_norm": 0.35916024446487427, + "learning_rate": 0.0005749383684446436, + "loss": 0.4541, + "step": 15290 + }, + { + "epoch": 0.8543174008995167, + "grad_norm": 1.130506157875061, + "learning_rate": 0.0005749103541012999, + "loss": 0.4211, + "step": 15291 + }, + { + "epoch": 0.8543732715031986, + "grad_norm": 0.32441747188568115, + "learning_rate": 0.0005748823397579561, + "loss": 0.4565, + "step": 15292 + }, + { + "epoch": 0.8544291421068805, + "grad_norm": 0.5821933746337891, + "learning_rate": 0.0005748543254146122, + "loss": 0.5518, + "step": 15293 + }, + { + "epoch": 0.8544850127105623, + "grad_norm": 0.4099457263946533, + "learning_rate": 0.0005748263110712685, + "loss": 0.4914, + "step": 15294 + }, + { + "epoch": 0.8545408833142442, + "grad_norm": 0.6264086365699768, + "learning_rate": 0.0005747982967279247, + "loss": 0.4441, + "step": 15295 + }, + { + "epoch": 0.8545967539179261, + "grad_norm": 7.523664951324463, + "learning_rate": 0.0005747702823845809, + "loss": 0.3984, + "step": 15296 + }, + { + "epoch": 0.854652624521608, + "grad_norm": 0.3435061275959015, + "learning_rate": 0.000574742268041237, + "loss": 0.3869, + "step": 15297 + }, + { + "epoch": 0.8547084951252898, + "grad_norm": 0.6535807847976685, + "learning_rate": 0.0005747142536978934, + "loss": 0.442, + "step": 15298 + }, + { + "epoch": 0.8547643657289717, + "grad_norm": 0.9499971270561218, + "learning_rate": 0.0005746862393545496, + "loss": 0.4973, + "step": 15299 + }, + { + "epoch": 0.8548202363326536, + "grad_norm": 0.5038163065910339, + "learning_rate": 0.0005746582250112058, + "loss": 0.4398, + "step": 15300 + }, + { + "epoch": 0.8548761069363354, + "grad_norm": 0.49750038981437683, + "learning_rate": 0.000574630210667862, + "loss": 0.3684, + "step": 15301 + }, + { + "epoch": 0.8549319775400174, + "grad_norm": 0.5342135429382324, + "learning_rate": 0.0005746021963245182, + "loss": 0.4019, + "step": 15302 + }, + { + "epoch": 0.8549878481436992, + "grad_norm": 0.6449026465415955, + "learning_rate": 0.0005745741819811744, + "loss": 0.4654, + "step": 15303 + }, + { + "epoch": 0.855043718747381, + "grad_norm": 0.5922929048538208, + "learning_rate": 0.0005745461676378306, + "loss": 0.3255, + "step": 15304 + }, + { + "epoch": 0.8550995893510629, + "grad_norm": 0.4799043834209442, + "learning_rate": 0.0005745181532944868, + "loss": 0.4799, + "step": 15305 + }, + { + "epoch": 0.8551554599547448, + "grad_norm": 4.299559116363525, + "learning_rate": 0.000574490138951143, + "loss": 0.5954, + "step": 15306 + }, + { + "epoch": 0.8552113305584267, + "grad_norm": 0.7288792133331299, + "learning_rate": 0.0005744621246077992, + "loss": 0.4381, + "step": 15307 + }, + { + "epoch": 0.8552672011621085, + "grad_norm": 0.42280519008636475, + "learning_rate": 0.0005744341102644555, + "loss": 0.4453, + "step": 15308 + }, + { + "epoch": 0.8553230717657905, + "grad_norm": 0.4212965667247772, + "learning_rate": 0.0005744060959211116, + "loss": 0.4799, + "step": 15309 + }, + { + "epoch": 0.8553789423694723, + "grad_norm": 0.7992929816246033, + "learning_rate": 0.0005743780815777678, + "loss": 0.4443, + "step": 15310 + }, + { + "epoch": 0.8554348129731542, + "grad_norm": 0.35906022787094116, + "learning_rate": 0.000574350067234424, + "loss": 0.4296, + "step": 15311 + }, + { + "epoch": 0.855490683576836, + "grad_norm": 0.5169544816017151, + "learning_rate": 0.0005743220528910802, + "loss": 0.5137, + "step": 15312 + }, + { + "epoch": 0.855546554180518, + "grad_norm": 0.3971608877182007, + "learning_rate": 0.0005742940385477364, + "loss": 0.4433, + "step": 15313 + }, + { + "epoch": 0.8556024247841998, + "grad_norm": 0.6079486012458801, + "learning_rate": 0.0005742660242043926, + "loss": 0.3469, + "step": 15314 + }, + { + "epoch": 0.8556582953878816, + "grad_norm": 0.5801418423652649, + "learning_rate": 0.0005742380098610489, + "loss": 0.5194, + "step": 15315 + }, + { + "epoch": 0.8557141659915636, + "grad_norm": 0.5435479879379272, + "learning_rate": 0.0005742099955177051, + "loss": 0.3914, + "step": 15316 + }, + { + "epoch": 0.8557700365952454, + "grad_norm": 0.42790696024894714, + "learning_rate": 0.0005741819811743613, + "loss": 0.3714, + "step": 15317 + }, + { + "epoch": 0.8558259071989273, + "grad_norm": 0.4587008059024811, + "learning_rate": 0.0005741539668310175, + "loss": 0.3517, + "step": 15318 + }, + { + "epoch": 0.8558817778026092, + "grad_norm": 1.444393515586853, + "learning_rate": 0.0005741259524876738, + "loss": 0.557, + "step": 15319 + }, + { + "epoch": 0.855937648406291, + "grad_norm": 0.4097956120967865, + "learning_rate": 0.0005740979381443299, + "loss": 0.503, + "step": 15320 + }, + { + "epoch": 0.8559935190099729, + "grad_norm": 0.5221260786056519, + "learning_rate": 0.0005740699238009861, + "loss": 0.5654, + "step": 15321 + }, + { + "epoch": 0.8560493896136547, + "grad_norm": 1.8472442626953125, + "learning_rate": 0.0005740419094576423, + "loss": 0.445, + "step": 15322 + }, + { + "epoch": 0.8561052602173367, + "grad_norm": 0.6461547017097473, + "learning_rate": 0.0005740138951142986, + "loss": 0.4833, + "step": 15323 + }, + { + "epoch": 0.8561611308210185, + "grad_norm": 2.874835968017578, + "learning_rate": 0.0005739858807709547, + "loss": 0.592, + "step": 15324 + }, + { + "epoch": 0.8562170014247004, + "grad_norm": 2.101097822189331, + "learning_rate": 0.0005739578664276109, + "loss": 0.7648, + "step": 15325 + }, + { + "epoch": 0.8562728720283823, + "grad_norm": 0.478605717420578, + "learning_rate": 0.0005739298520842672, + "loss": 0.471, + "step": 15326 + }, + { + "epoch": 0.8563287426320642, + "grad_norm": 0.40589088201522827, + "learning_rate": 0.0005739018377409234, + "loss": 0.432, + "step": 15327 + }, + { + "epoch": 0.856384613235746, + "grad_norm": 0.3580128848552704, + "learning_rate": 0.0005738738233975795, + "loss": 0.3805, + "step": 15328 + }, + { + "epoch": 0.8564404838394278, + "grad_norm": 0.5793143510818481, + "learning_rate": 0.0005738458090542357, + "loss": 0.4309, + "step": 15329 + }, + { + "epoch": 0.8564963544431098, + "grad_norm": 0.6283808350563049, + "learning_rate": 0.000573817794710892, + "loss": 0.4021, + "step": 15330 + }, + { + "epoch": 0.8565522250467916, + "grad_norm": 0.8534294366836548, + "learning_rate": 0.0005737897803675481, + "loss": 0.5185, + "step": 15331 + }, + { + "epoch": 0.8566080956504735, + "grad_norm": 0.49442511796951294, + "learning_rate": 0.0005737617660242044, + "loss": 0.4474, + "step": 15332 + }, + { + "epoch": 0.8566639662541554, + "grad_norm": 0.5709825754165649, + "learning_rate": 0.0005737337516808607, + "loss": 0.5558, + "step": 15333 + }, + { + "epoch": 0.8567198368578373, + "grad_norm": 0.4283294081687927, + "learning_rate": 0.0005737057373375169, + "loss": 0.4313, + "step": 15334 + }, + { + "epoch": 0.8567757074615191, + "grad_norm": 0.46788477897644043, + "learning_rate": 0.000573677722994173, + "loss": 0.3889, + "step": 15335 + }, + { + "epoch": 0.856831578065201, + "grad_norm": 0.48583894968032837, + "learning_rate": 0.0005736497086508292, + "loss": 0.4364, + "step": 15336 + }, + { + "epoch": 0.8568874486688829, + "grad_norm": 0.4500178098678589, + "learning_rate": 0.0005736216943074855, + "loss": 0.4151, + "step": 15337 + }, + { + "epoch": 0.8569433192725647, + "grad_norm": 0.40389344096183777, + "learning_rate": 0.0005735936799641417, + "loss": 0.4466, + "step": 15338 + }, + { + "epoch": 0.8569991898762466, + "grad_norm": 1.3746812343597412, + "learning_rate": 0.0005735656656207978, + "loss": 0.6772, + "step": 15339 + }, + { + "epoch": 0.8570550604799285, + "grad_norm": 0.6320574283599854, + "learning_rate": 0.0005735376512774541, + "loss": 0.4681, + "step": 15340 + }, + { + "epoch": 0.8571109310836104, + "grad_norm": 2.8778984546661377, + "learning_rate": 0.0005735096369341103, + "loss": 0.6329, + "step": 15341 + }, + { + "epoch": 0.8571668016872922, + "grad_norm": 0.7436590790748596, + "learning_rate": 0.0005734816225907665, + "loss": 0.4703, + "step": 15342 + }, + { + "epoch": 0.8572226722909742, + "grad_norm": 0.6069959402084351, + "learning_rate": 0.0005734536082474226, + "loss": 0.4549, + "step": 15343 + }, + { + "epoch": 0.857278542894656, + "grad_norm": 0.4183422923088074, + "learning_rate": 0.0005734255939040789, + "loss": 0.4982, + "step": 15344 + }, + { + "epoch": 0.8573344134983378, + "grad_norm": 0.6070115566253662, + "learning_rate": 0.0005733975795607351, + "loss": 0.5206, + "step": 15345 + }, + { + "epoch": 0.8573902841020197, + "grad_norm": 0.48223739862442017, + "learning_rate": 0.0005733695652173913, + "loss": 0.5073, + "step": 15346 + }, + { + "epoch": 0.8574461547057016, + "grad_norm": 0.5069206357002258, + "learning_rate": 0.0005733415508740474, + "loss": 0.4647, + "step": 15347 + }, + { + "epoch": 0.8575020253093835, + "grad_norm": 0.4291059374809265, + "learning_rate": 0.0005733135365307038, + "loss": 0.3589, + "step": 15348 + }, + { + "epoch": 0.8575578959130653, + "grad_norm": 0.4462362825870514, + "learning_rate": 0.00057328552218736, + "loss": 0.4886, + "step": 15349 + }, + { + "epoch": 0.8576137665167473, + "grad_norm": 0.7498332262039185, + "learning_rate": 0.0005732575078440162, + "loss": 0.4563, + "step": 15350 + }, + { + "epoch": 0.8576696371204291, + "grad_norm": 0.4908643662929535, + "learning_rate": 0.0005732294935006724, + "loss": 0.5272, + "step": 15351 + }, + { + "epoch": 0.8577255077241109, + "grad_norm": 1.3664236068725586, + "learning_rate": 0.0005732014791573286, + "loss": 0.436, + "step": 15352 + }, + { + "epoch": 0.8577813783277929, + "grad_norm": 1.1869257688522339, + "learning_rate": 0.0005731734648139848, + "loss": 0.4762, + "step": 15353 + }, + { + "epoch": 0.8578372489314747, + "grad_norm": 0.5021815299987793, + "learning_rate": 0.0005731454504706409, + "loss": 0.3623, + "step": 15354 + }, + { + "epoch": 0.8578931195351566, + "grad_norm": 0.4177873432636261, + "learning_rate": 0.0005731174361272972, + "loss": 0.4246, + "step": 15355 + }, + { + "epoch": 0.8579489901388384, + "grad_norm": 0.5851925015449524, + "learning_rate": 0.0005730894217839534, + "loss": 0.4575, + "step": 15356 + }, + { + "epoch": 0.8580048607425204, + "grad_norm": 0.37027472257614136, + "learning_rate": 0.0005730614074406096, + "loss": 0.4517, + "step": 15357 + }, + { + "epoch": 0.8580607313462022, + "grad_norm": 0.3970695436000824, + "learning_rate": 0.0005730333930972658, + "loss": 0.447, + "step": 15358 + }, + { + "epoch": 0.858116601949884, + "grad_norm": 1.041153073310852, + "learning_rate": 0.000573005378753922, + "loss": 0.4589, + "step": 15359 + }, + { + "epoch": 0.858172472553566, + "grad_norm": 0.5062230825424194, + "learning_rate": 0.0005729773644105782, + "loss": 0.4105, + "step": 15360 + }, + { + "epoch": 0.8582283431572478, + "grad_norm": 0.34087565541267395, + "learning_rate": 0.0005729493500672344, + "loss": 0.4131, + "step": 15361 + }, + { + "epoch": 0.8582842137609297, + "grad_norm": 0.552529513835907, + "learning_rate": 0.0005729213357238906, + "loss": 0.6663, + "step": 15362 + }, + { + "epoch": 0.8583400843646115, + "grad_norm": 0.40995749831199646, + "learning_rate": 0.0005728933213805468, + "loss": 0.518, + "step": 15363 + }, + { + "epoch": 0.8583959549682935, + "grad_norm": 0.423371285200119, + "learning_rate": 0.000572865307037203, + "loss": 0.3668, + "step": 15364 + }, + { + "epoch": 0.8584518255719753, + "grad_norm": 0.47994765639305115, + "learning_rate": 0.0005728372926938594, + "loss": 0.4531, + "step": 15365 + }, + { + "epoch": 0.8585076961756571, + "grad_norm": 2.291760206222534, + "learning_rate": 0.0005728092783505155, + "loss": 0.3774, + "step": 15366 + }, + { + "epoch": 0.8585635667793391, + "grad_norm": 0.5191513895988464, + "learning_rate": 0.0005727812640071717, + "loss": 0.4823, + "step": 15367 + }, + { + "epoch": 0.8586194373830209, + "grad_norm": 10.602723121643066, + "learning_rate": 0.0005727532496638279, + "loss": 0.465, + "step": 15368 + }, + { + "epoch": 0.8586753079867028, + "grad_norm": 0.3860258460044861, + "learning_rate": 0.0005727252353204842, + "loss": 0.4776, + "step": 15369 + }, + { + "epoch": 0.8587311785903846, + "grad_norm": 0.4844697415828705, + "learning_rate": 0.0005726972209771403, + "loss": 0.5005, + "step": 15370 + }, + { + "epoch": 0.8587870491940666, + "grad_norm": 0.4536154568195343, + "learning_rate": 0.0005726692066337965, + "loss": 0.5689, + "step": 15371 + }, + { + "epoch": 0.8588429197977484, + "grad_norm": 0.37212610244750977, + "learning_rate": 0.0005726411922904528, + "loss": 0.4124, + "step": 15372 + }, + { + "epoch": 0.8588987904014302, + "grad_norm": 0.7798500061035156, + "learning_rate": 0.0005726131779471089, + "loss": 0.349, + "step": 15373 + }, + { + "epoch": 0.8589546610051122, + "grad_norm": 0.4349924623966217, + "learning_rate": 0.0005725851636037651, + "loss": 0.3527, + "step": 15374 + }, + { + "epoch": 0.859010531608794, + "grad_norm": 0.3477824628353119, + "learning_rate": 0.0005725571492604213, + "loss": 0.3958, + "step": 15375 + }, + { + "epoch": 0.8590664022124759, + "grad_norm": 0.44681069254875183, + "learning_rate": 0.0005725291349170776, + "loss": 0.5793, + "step": 15376 + }, + { + "epoch": 0.8591222728161578, + "grad_norm": 0.44077569246292114, + "learning_rate": 0.0005725011205737337, + "loss": 0.5508, + "step": 15377 + }, + { + "epoch": 0.8591781434198397, + "grad_norm": 0.956451952457428, + "learning_rate": 0.0005724731062303899, + "loss": 0.4461, + "step": 15378 + }, + { + "epoch": 0.8592340140235215, + "grad_norm": 0.3935163915157318, + "learning_rate": 0.0005724450918870461, + "loss": 0.2809, + "step": 15379 + }, + { + "epoch": 0.8592898846272033, + "grad_norm": 0.602623462677002, + "learning_rate": 0.0005724170775437024, + "loss": 0.5048, + "step": 15380 + }, + { + "epoch": 0.8593457552308853, + "grad_norm": 2.7756996154785156, + "learning_rate": 0.0005723890632003585, + "loss": 0.4913, + "step": 15381 + }, + { + "epoch": 0.8594016258345671, + "grad_norm": 0.47414636611938477, + "learning_rate": 0.0005723610488570148, + "loss": 0.4812, + "step": 15382 + }, + { + "epoch": 0.859457496438249, + "grad_norm": 0.37748655676841736, + "learning_rate": 0.0005723330345136711, + "loss": 0.474, + "step": 15383 + }, + { + "epoch": 0.8595133670419309, + "grad_norm": 0.7322424054145813, + "learning_rate": 0.0005723050201703273, + "loss": 0.5342, + "step": 15384 + }, + { + "epoch": 0.8595692376456128, + "grad_norm": 0.7105609774589539, + "learning_rate": 0.0005722770058269834, + "loss": 0.3857, + "step": 15385 + }, + { + "epoch": 0.8596251082492946, + "grad_norm": 0.702343761920929, + "learning_rate": 0.0005722489914836396, + "loss": 0.3845, + "step": 15386 + }, + { + "epoch": 0.8596809788529765, + "grad_norm": 0.3759746849536896, + "learning_rate": 0.0005722209771402959, + "loss": 0.4038, + "step": 15387 + }, + { + "epoch": 0.8597368494566584, + "grad_norm": 0.4729771018028259, + "learning_rate": 0.0005721929627969521, + "loss": 0.4217, + "step": 15388 + }, + { + "epoch": 0.8597927200603402, + "grad_norm": 0.4510622024536133, + "learning_rate": 0.0005721649484536082, + "loss": 0.4041, + "step": 15389 + }, + { + "epoch": 0.8598485906640221, + "grad_norm": 0.4846169948577881, + "learning_rate": 0.0005721369341102645, + "loss": 0.4803, + "step": 15390 + }, + { + "epoch": 0.859904461267704, + "grad_norm": 1.6877529621124268, + "learning_rate": 0.0005721089197669207, + "loss": 0.3786, + "step": 15391 + }, + { + "epoch": 0.8599603318713859, + "grad_norm": 0.44502148032188416, + "learning_rate": 0.0005720809054235769, + "loss": 0.4035, + "step": 15392 + }, + { + "epoch": 0.8600162024750677, + "grad_norm": 2.558959722518921, + "learning_rate": 0.000572052891080233, + "loss": 0.5145, + "step": 15393 + }, + { + "epoch": 0.8600720730787497, + "grad_norm": 0.5541415214538574, + "learning_rate": 0.0005720248767368893, + "loss": 0.4871, + "step": 15394 + }, + { + "epoch": 0.8601279436824315, + "grad_norm": 0.5128536820411682, + "learning_rate": 0.0005719968623935455, + "loss": 0.4336, + "step": 15395 + }, + { + "epoch": 0.8601838142861133, + "grad_norm": 0.39193156361579895, + "learning_rate": 0.0005719688480502016, + "loss": 0.5169, + "step": 15396 + }, + { + "epoch": 0.8602396848897952, + "grad_norm": 1.3162139654159546, + "learning_rate": 0.000571940833706858, + "loss": 0.4814, + "step": 15397 + }, + { + "epoch": 0.8602955554934771, + "grad_norm": 0.5030894875526428, + "learning_rate": 0.0005719128193635142, + "loss": 0.3714, + "step": 15398 + }, + { + "epoch": 0.860351426097159, + "grad_norm": 1.009965181350708, + "learning_rate": 0.0005718848050201704, + "loss": 0.3698, + "step": 15399 + }, + { + "epoch": 0.8604072967008408, + "grad_norm": 0.5209161639213562, + "learning_rate": 0.0005718567906768265, + "loss": 0.4697, + "step": 15400 + }, + { + "epoch": 0.8604631673045228, + "grad_norm": 0.3677946925163269, + "learning_rate": 0.0005718287763334828, + "loss": 0.3825, + "step": 15401 + }, + { + "epoch": 0.8605190379082046, + "grad_norm": 0.37187331914901733, + "learning_rate": 0.000571800761990139, + "loss": 0.343, + "step": 15402 + }, + { + "epoch": 0.8605749085118864, + "grad_norm": 0.6815358996391296, + "learning_rate": 0.0005717727476467952, + "loss": 0.5022, + "step": 15403 + }, + { + "epoch": 0.8606307791155683, + "grad_norm": 1.6876106262207031, + "learning_rate": 0.0005717447333034514, + "loss": 0.4885, + "step": 15404 + }, + { + "epoch": 0.8606866497192502, + "grad_norm": 0.43052202463150024, + "learning_rate": 0.0005717167189601076, + "loss": 0.4124, + "step": 15405 + }, + { + "epoch": 0.8607425203229321, + "grad_norm": 0.38612666726112366, + "learning_rate": 0.0005716887046167638, + "loss": 0.3823, + "step": 15406 + }, + { + "epoch": 0.8607983909266139, + "grad_norm": 0.4049907624721527, + "learning_rate": 0.00057166069027342, + "loss": 0.4246, + "step": 15407 + }, + { + "epoch": 0.8608542615302959, + "grad_norm": 0.6055341958999634, + "learning_rate": 0.0005716326759300762, + "loss": 0.6362, + "step": 15408 + }, + { + "epoch": 0.8609101321339777, + "grad_norm": 0.3396017253398895, + "learning_rate": 0.0005716046615867324, + "loss": 0.3363, + "step": 15409 + }, + { + "epoch": 0.8609660027376596, + "grad_norm": 0.6201258897781372, + "learning_rate": 0.0005715766472433886, + "loss": 0.4322, + "step": 15410 + }, + { + "epoch": 0.8610218733413415, + "grad_norm": 0.4579618275165558, + "learning_rate": 0.0005715486329000449, + "loss": 0.269, + "step": 15411 + }, + { + "epoch": 0.8610777439450233, + "grad_norm": 0.3782704174518585, + "learning_rate": 0.000571520618556701, + "loss": 0.4019, + "step": 15412 + }, + { + "epoch": 0.8611336145487052, + "grad_norm": 0.8226367831230164, + "learning_rate": 0.0005714926042133572, + "loss": 0.4012, + "step": 15413 + }, + { + "epoch": 0.861189485152387, + "grad_norm": 0.5446321964263916, + "learning_rate": 0.0005714645898700134, + "loss": 0.4141, + "step": 15414 + }, + { + "epoch": 0.861245355756069, + "grad_norm": 0.5961794853210449, + "learning_rate": 0.0005714365755266698, + "loss": 0.4714, + "step": 15415 + }, + { + "epoch": 0.8613012263597508, + "grad_norm": 0.7101730108261108, + "learning_rate": 0.0005714085611833259, + "loss": 0.5638, + "step": 15416 + }, + { + "epoch": 0.8613570969634327, + "grad_norm": 0.6919733881950378, + "learning_rate": 0.0005713805468399821, + "loss": 0.4637, + "step": 15417 + }, + { + "epoch": 0.8614129675671146, + "grad_norm": 0.4412024915218353, + "learning_rate": 0.0005713525324966383, + "loss": 0.3493, + "step": 15418 + }, + { + "epoch": 0.8614688381707964, + "grad_norm": 0.6393027305603027, + "learning_rate": 0.0005713245181532945, + "loss": 0.3919, + "step": 15419 + }, + { + "epoch": 0.8615247087744783, + "grad_norm": 3.111780881881714, + "learning_rate": 0.0005712965038099507, + "loss": 0.4347, + "step": 15420 + }, + { + "epoch": 0.8615805793781601, + "grad_norm": 0.40580350160598755, + "learning_rate": 0.0005712684894666069, + "loss": 0.4597, + "step": 15421 + }, + { + "epoch": 0.8616364499818421, + "grad_norm": 0.7262750864028931, + "learning_rate": 0.0005712404751232632, + "loss": 0.497, + "step": 15422 + }, + { + "epoch": 0.8616923205855239, + "grad_norm": 0.6351814866065979, + "learning_rate": 0.0005712124607799193, + "loss": 0.3772, + "step": 15423 + }, + { + "epoch": 0.8617481911892058, + "grad_norm": 1.9809865951538086, + "learning_rate": 0.0005711844464365755, + "loss": 0.4253, + "step": 15424 + }, + { + "epoch": 0.8618040617928877, + "grad_norm": 2.5654969215393066, + "learning_rate": 0.0005711564320932317, + "loss": 0.4171, + "step": 15425 + }, + { + "epoch": 0.8618599323965696, + "grad_norm": 0.3620027005672455, + "learning_rate": 0.000571128417749888, + "loss": 0.4249, + "step": 15426 + }, + { + "epoch": 0.8619158030002514, + "grad_norm": 0.668380618095398, + "learning_rate": 0.0005711004034065441, + "loss": 0.3387, + "step": 15427 + }, + { + "epoch": 0.8619716736039333, + "grad_norm": 0.484364777803421, + "learning_rate": 0.0005710723890632003, + "loss": 0.4747, + "step": 15428 + }, + { + "epoch": 0.8620275442076152, + "grad_norm": 0.4083690643310547, + "learning_rate": 0.0005710443747198566, + "loss": 0.3985, + "step": 15429 + }, + { + "epoch": 0.862083414811297, + "grad_norm": 0.523146390914917, + "learning_rate": 0.0005710163603765128, + "loss": 0.3928, + "step": 15430 + }, + { + "epoch": 0.8621392854149789, + "grad_norm": 1.3838988542556763, + "learning_rate": 0.0005709883460331689, + "loss": 0.5271, + "step": 15431 + }, + { + "epoch": 0.8621951560186608, + "grad_norm": 0.37045028805732727, + "learning_rate": 0.0005709603316898252, + "loss": 0.393, + "step": 15432 + }, + { + "epoch": 0.8622510266223427, + "grad_norm": 0.3855557441711426, + "learning_rate": 0.0005709323173464815, + "loss": 0.4246, + "step": 15433 + }, + { + "epoch": 0.8623068972260245, + "grad_norm": 0.3974785804748535, + "learning_rate": 0.0005709043030031377, + "loss": 0.3444, + "step": 15434 + }, + { + "epoch": 0.8623627678297064, + "grad_norm": 0.4101777970790863, + "learning_rate": 0.0005708762886597938, + "loss": 0.5452, + "step": 15435 + }, + { + "epoch": 0.8624186384333883, + "grad_norm": 0.4243045449256897, + "learning_rate": 0.0005708482743164501, + "loss": 0.4159, + "step": 15436 + }, + { + "epoch": 0.8624745090370701, + "grad_norm": 0.70592200756073, + "learning_rate": 0.0005708202599731063, + "loss": 0.4881, + "step": 15437 + }, + { + "epoch": 0.862530379640752, + "grad_norm": 0.8170251846313477, + "learning_rate": 0.0005707922456297624, + "loss": 0.3708, + "step": 15438 + }, + { + "epoch": 0.8625862502444339, + "grad_norm": 0.5262956023216248, + "learning_rate": 0.0005707642312864186, + "loss": 0.388, + "step": 15439 + }, + { + "epoch": 0.8626421208481158, + "grad_norm": 0.4808351397514343, + "learning_rate": 0.0005707362169430749, + "loss": 0.35, + "step": 15440 + }, + { + "epoch": 0.8626979914517976, + "grad_norm": 0.4123280644416809, + "learning_rate": 0.0005707082025997311, + "loss": 0.4284, + "step": 15441 + }, + { + "epoch": 0.8627538620554795, + "grad_norm": 0.3574284613132477, + "learning_rate": 0.0005706801882563872, + "loss": 0.3933, + "step": 15442 + }, + { + "epoch": 0.8628097326591614, + "grad_norm": 3.310447931289673, + "learning_rate": 0.0005706521739130435, + "loss": 0.4849, + "step": 15443 + }, + { + "epoch": 0.8628656032628432, + "grad_norm": 1.9137107133865356, + "learning_rate": 0.0005706241595696997, + "loss": 0.3366, + "step": 15444 + }, + { + "epoch": 0.8629214738665252, + "grad_norm": 0.5317107439041138, + "learning_rate": 0.0005705961452263559, + "loss": 0.4286, + "step": 15445 + }, + { + "epoch": 0.862977344470207, + "grad_norm": 0.3237667977809906, + "learning_rate": 0.000570568130883012, + "loss": 0.386, + "step": 15446 + }, + { + "epoch": 0.8630332150738889, + "grad_norm": 0.32049763202667236, + "learning_rate": 0.0005705401165396683, + "loss": 0.3623, + "step": 15447 + }, + { + "epoch": 0.8630890856775707, + "grad_norm": 0.43236857652664185, + "learning_rate": 0.0005705121021963246, + "loss": 0.4199, + "step": 15448 + }, + { + "epoch": 0.8631449562812527, + "grad_norm": 1.5554472208023071, + "learning_rate": 0.0005704840878529808, + "loss": 0.4392, + "step": 15449 + }, + { + "epoch": 0.8632008268849345, + "grad_norm": 2.0164148807525635, + "learning_rate": 0.0005704560735096369, + "loss": 0.4731, + "step": 15450 + }, + { + "epoch": 0.8632566974886163, + "grad_norm": 0.5925301909446716, + "learning_rate": 0.0005704280591662932, + "loss": 0.4471, + "step": 15451 + }, + { + "epoch": 0.8633125680922983, + "grad_norm": 7.397457599639893, + "learning_rate": 0.0005704000448229494, + "loss": 0.4213, + "step": 15452 + }, + { + "epoch": 0.8633684386959801, + "grad_norm": 0.5516096353530884, + "learning_rate": 0.0005703720304796056, + "loss": 0.3836, + "step": 15453 + }, + { + "epoch": 0.863424309299662, + "grad_norm": 0.4891369342803955, + "learning_rate": 0.0005703440161362618, + "loss": 0.4367, + "step": 15454 + }, + { + "epoch": 0.8634801799033438, + "grad_norm": 0.7032449841499329, + "learning_rate": 0.000570316001792918, + "loss": 0.4081, + "step": 15455 + }, + { + "epoch": 0.8635360505070258, + "grad_norm": 0.4239242374897003, + "learning_rate": 0.0005702879874495742, + "loss": 0.3525, + "step": 15456 + }, + { + "epoch": 0.8635919211107076, + "grad_norm": 0.3239501118659973, + "learning_rate": 0.0005702599731062304, + "loss": 0.3445, + "step": 15457 + }, + { + "epoch": 0.8636477917143894, + "grad_norm": 3.529973030090332, + "learning_rate": 0.0005702319587628866, + "loss": 0.4305, + "step": 15458 + }, + { + "epoch": 0.8637036623180714, + "grad_norm": 0.42659610509872437, + "learning_rate": 0.0005702039444195428, + "loss": 0.5632, + "step": 15459 + }, + { + "epoch": 0.8637595329217532, + "grad_norm": 0.3595368266105652, + "learning_rate": 0.000570175930076199, + "loss": 0.4085, + "step": 15460 + }, + { + "epoch": 0.8638154035254351, + "grad_norm": 0.48112934827804565, + "learning_rate": 0.0005701479157328552, + "loss": 0.3113, + "step": 15461 + }, + { + "epoch": 0.863871274129117, + "grad_norm": 0.8132485747337341, + "learning_rate": 0.0005701199013895114, + "loss": 0.5686, + "step": 15462 + }, + { + "epoch": 0.8639271447327989, + "grad_norm": 1.7429637908935547, + "learning_rate": 0.0005700918870461676, + "loss": 0.3204, + "step": 15463 + }, + { + "epoch": 0.8639830153364807, + "grad_norm": 0.7975624799728394, + "learning_rate": 0.0005700638727028238, + "loss": 0.3704, + "step": 15464 + }, + { + "epoch": 0.8640388859401625, + "grad_norm": 0.5061265230178833, + "learning_rate": 0.00057003585835948, + "loss": 0.5182, + "step": 15465 + }, + { + "epoch": 0.8640947565438445, + "grad_norm": 0.5712950825691223, + "learning_rate": 0.0005700078440161363, + "loss": 0.4588, + "step": 15466 + }, + { + "epoch": 0.8641506271475263, + "grad_norm": 0.4287416934967041, + "learning_rate": 0.0005699798296727925, + "loss": 0.4761, + "step": 15467 + }, + { + "epoch": 0.8642064977512082, + "grad_norm": 1.1039386987686157, + "learning_rate": 0.0005699518153294488, + "loss": 0.4651, + "step": 15468 + }, + { + "epoch": 0.8642623683548901, + "grad_norm": 0.6128109097480774, + "learning_rate": 0.0005699238009861049, + "loss": 0.4755, + "step": 15469 + }, + { + "epoch": 0.864318238958572, + "grad_norm": 0.584194004535675, + "learning_rate": 0.0005698957866427611, + "loss": 0.3598, + "step": 15470 + }, + { + "epoch": 0.8643741095622538, + "grad_norm": 0.47922056913375854, + "learning_rate": 0.0005698677722994173, + "loss": 0.4503, + "step": 15471 + }, + { + "epoch": 0.8644299801659356, + "grad_norm": 0.9774891138076782, + "learning_rate": 0.0005698397579560736, + "loss": 0.4892, + "step": 15472 + }, + { + "epoch": 0.8644858507696176, + "grad_norm": 0.4462221562862396, + "learning_rate": 0.0005698117436127297, + "loss": 0.4646, + "step": 15473 + }, + { + "epoch": 0.8645417213732994, + "grad_norm": 0.48591914772987366, + "learning_rate": 0.0005697837292693859, + "loss": 0.4072, + "step": 15474 + }, + { + "epoch": 0.8645975919769813, + "grad_norm": 0.577107846736908, + "learning_rate": 0.0005697557149260422, + "loss": 0.4946, + "step": 15475 + }, + { + "epoch": 0.8646534625806632, + "grad_norm": 0.6095107793807983, + "learning_rate": 0.0005697277005826984, + "loss": 0.4898, + "step": 15476 + }, + { + "epoch": 0.8647093331843451, + "grad_norm": 0.5460664629936218, + "learning_rate": 0.0005696996862393545, + "loss": 0.4516, + "step": 15477 + }, + { + "epoch": 0.8647652037880269, + "grad_norm": 0.46723636984825134, + "learning_rate": 0.0005696716718960107, + "loss": 0.5623, + "step": 15478 + }, + { + "epoch": 0.8648210743917089, + "grad_norm": 0.44155529141426086, + "learning_rate": 0.000569643657552667, + "loss": 0.3761, + "step": 15479 + }, + { + "epoch": 0.8648769449953907, + "grad_norm": 0.5775860548019409, + "learning_rate": 0.0005696156432093231, + "loss": 0.454, + "step": 15480 + }, + { + "epoch": 0.8649328155990725, + "grad_norm": 0.5015895962715149, + "learning_rate": 0.0005695876288659793, + "loss": 0.4121, + "step": 15481 + }, + { + "epoch": 0.8649886862027544, + "grad_norm": 1.0878278017044067, + "learning_rate": 0.0005695596145226357, + "loss": 0.3857, + "step": 15482 + }, + { + "epoch": 0.8650445568064363, + "grad_norm": 0.5444540977478027, + "learning_rate": 0.0005695316001792919, + "loss": 0.4977, + "step": 15483 + }, + { + "epoch": 0.8651004274101182, + "grad_norm": 0.6149139404296875, + "learning_rate": 0.000569503585835948, + "loss": 0.4147, + "step": 15484 + }, + { + "epoch": 0.8651562980138, + "grad_norm": 0.5803570747375488, + "learning_rate": 0.0005694755714926042, + "loss": 0.452, + "step": 15485 + }, + { + "epoch": 0.865212168617482, + "grad_norm": 2.8610007762908936, + "learning_rate": 0.0005694475571492605, + "loss": 0.4625, + "step": 15486 + }, + { + "epoch": 0.8652680392211638, + "grad_norm": 0.45050716400146484, + "learning_rate": 0.0005694195428059167, + "loss": 0.4304, + "step": 15487 + }, + { + "epoch": 0.8653239098248456, + "grad_norm": 1.185347557067871, + "learning_rate": 0.0005693915284625728, + "loss": 0.4777, + "step": 15488 + }, + { + "epoch": 0.8653797804285275, + "grad_norm": 0.657272458076477, + "learning_rate": 0.000569363514119229, + "loss": 0.5438, + "step": 15489 + }, + { + "epoch": 0.8654356510322094, + "grad_norm": 0.6599783301353455, + "learning_rate": 0.0005693354997758853, + "loss": 0.4826, + "step": 15490 + }, + { + "epoch": 0.8654915216358913, + "grad_norm": 0.640741229057312, + "learning_rate": 0.0005693074854325415, + "loss": 0.5409, + "step": 15491 + }, + { + "epoch": 0.8655473922395731, + "grad_norm": 0.7081790566444397, + "learning_rate": 0.0005692794710891976, + "loss": 0.5766, + "step": 15492 + }, + { + "epoch": 0.8656032628432551, + "grad_norm": 0.36559757590293884, + "learning_rate": 0.0005692514567458539, + "loss": 0.4326, + "step": 15493 + }, + { + "epoch": 0.8656591334469369, + "grad_norm": 0.7256369590759277, + "learning_rate": 0.0005692234424025101, + "loss": 0.3364, + "step": 15494 + }, + { + "epoch": 0.8657150040506187, + "grad_norm": 0.4505833089351654, + "learning_rate": 0.0005691954280591663, + "loss": 0.3828, + "step": 15495 + }, + { + "epoch": 0.8657708746543007, + "grad_norm": 0.5016536712646484, + "learning_rate": 0.0005691674137158224, + "loss": 0.3761, + "step": 15496 + }, + { + "epoch": 0.8658267452579825, + "grad_norm": 0.8432008028030396, + "learning_rate": 0.0005691393993724787, + "loss": 0.5389, + "step": 15497 + }, + { + "epoch": 0.8658826158616644, + "grad_norm": 0.3302515149116516, + "learning_rate": 0.000569111385029135, + "loss": 0.3086, + "step": 15498 + }, + { + "epoch": 0.8659384864653462, + "grad_norm": 0.43161702156066895, + "learning_rate": 0.0005690833706857912, + "loss": 0.5473, + "step": 15499 + }, + { + "epoch": 0.8659943570690282, + "grad_norm": 0.4181564450263977, + "learning_rate": 0.0005690553563424474, + "loss": 0.487, + "step": 15500 + }, + { + "epoch": 0.8659943570690282, + "eval_cer": 0.0897297415251001, + "eval_loss": 0.3350467085838318, + "eval_runtime": 56.1753, + "eval_samples_per_second": 80.783, + "eval_steps_per_second": 5.056, + "eval_wer": 0.358006759587721, + "step": 15500 + }, + { + "epoch": 0.86605022767271, + "grad_norm": 0.3217324912548065, + "learning_rate": 0.0005690273419991036, + "loss": 0.4057, + "step": 15501 + }, + { + "epoch": 0.8661060982763918, + "grad_norm": 0.4222371578216553, + "learning_rate": 0.0005689993276557598, + "loss": 0.4763, + "step": 15502 + }, + { + "epoch": 0.8661619688800738, + "grad_norm": 0.8424919247627258, + "learning_rate": 0.0005689713133124159, + "loss": 0.4381, + "step": 15503 + }, + { + "epoch": 0.8662178394837556, + "grad_norm": 0.7755890488624573, + "learning_rate": 0.0005689432989690722, + "loss": 0.5231, + "step": 15504 + }, + { + "epoch": 0.8662737100874375, + "grad_norm": 0.6010269522666931, + "learning_rate": 0.0005689152846257284, + "loss": 0.4335, + "step": 15505 + }, + { + "epoch": 0.8663295806911193, + "grad_norm": 1.1871554851531982, + "learning_rate": 0.0005688872702823846, + "loss": 0.4287, + "step": 15506 + }, + { + "epoch": 0.8663854512948013, + "grad_norm": 0.6348364949226379, + "learning_rate": 0.0005688592559390408, + "loss": 0.4755, + "step": 15507 + }, + { + "epoch": 0.8664413218984831, + "grad_norm": 0.4026443362236023, + "learning_rate": 0.000568831241595697, + "loss": 0.359, + "step": 15508 + }, + { + "epoch": 0.866497192502165, + "grad_norm": 0.3900725543498993, + "learning_rate": 0.0005688032272523532, + "loss": 0.3755, + "step": 15509 + }, + { + "epoch": 0.8665530631058469, + "grad_norm": 0.41263726353645325, + "learning_rate": 0.0005687752129090094, + "loss": 0.5119, + "step": 15510 + }, + { + "epoch": 0.8666089337095287, + "grad_norm": 6.055846214294434, + "learning_rate": 0.0005687471985656656, + "loss": 0.4143, + "step": 15511 + }, + { + "epoch": 0.8666648043132106, + "grad_norm": 0.49169448018074036, + "learning_rate": 0.0005687191842223218, + "loss": 0.412, + "step": 15512 + }, + { + "epoch": 0.8667206749168925, + "grad_norm": 0.452382892370224, + "learning_rate": 0.000568691169878978, + "loss": 0.3739, + "step": 15513 + }, + { + "epoch": 0.8667765455205744, + "grad_norm": 0.4041098654270172, + "learning_rate": 0.0005686631555356344, + "loss": 0.3615, + "step": 15514 + }, + { + "epoch": 0.8668324161242562, + "grad_norm": 0.5063281655311584, + "learning_rate": 0.0005686351411922905, + "loss": 0.532, + "step": 15515 + }, + { + "epoch": 0.866888286727938, + "grad_norm": 2.303149938583374, + "learning_rate": 0.0005686071268489467, + "loss": 0.397, + "step": 15516 + }, + { + "epoch": 0.86694415733162, + "grad_norm": 0.384382426738739, + "learning_rate": 0.0005685791125056029, + "loss": 0.3033, + "step": 15517 + }, + { + "epoch": 0.8670000279353018, + "grad_norm": 0.5568134188652039, + "learning_rate": 0.0005685510981622592, + "loss": 0.4203, + "step": 15518 + }, + { + "epoch": 0.8670558985389837, + "grad_norm": 0.6913617253303528, + "learning_rate": 0.0005685230838189153, + "loss": 0.4914, + "step": 15519 + }, + { + "epoch": 0.8671117691426656, + "grad_norm": 0.40669846534729004, + "learning_rate": 0.0005684950694755715, + "loss": 0.3523, + "step": 15520 + }, + { + "epoch": 0.8671676397463475, + "grad_norm": 0.45391175150871277, + "learning_rate": 0.0005684670551322278, + "loss": 0.4065, + "step": 15521 + }, + { + "epoch": 0.8672235103500293, + "grad_norm": 0.8699300289154053, + "learning_rate": 0.0005684390407888839, + "loss": 0.6173, + "step": 15522 + }, + { + "epoch": 0.8672793809537112, + "grad_norm": 0.44101205468177795, + "learning_rate": 0.0005684110264455401, + "loss": 0.4178, + "step": 15523 + }, + { + "epoch": 0.8673352515573931, + "grad_norm": 0.7072166204452515, + "learning_rate": 0.0005683830121021963, + "loss": 0.3744, + "step": 15524 + }, + { + "epoch": 0.867391122161075, + "grad_norm": 0.7462141513824463, + "learning_rate": 0.0005683549977588526, + "loss": 0.6832, + "step": 15525 + }, + { + "epoch": 0.8674469927647568, + "grad_norm": 0.6891704797744751, + "learning_rate": 0.0005683269834155087, + "loss": 0.5262, + "step": 15526 + }, + { + "epoch": 0.8675028633684387, + "grad_norm": 1.0756359100341797, + "learning_rate": 0.0005682989690721649, + "loss": 0.4496, + "step": 15527 + }, + { + "epoch": 0.8675587339721206, + "grad_norm": 0.43415069580078125, + "learning_rate": 0.0005682709547288211, + "loss": 0.4254, + "step": 15528 + }, + { + "epoch": 0.8676146045758024, + "grad_norm": 0.6608605980873108, + "learning_rate": 0.0005682429403854774, + "loss": 0.3936, + "step": 15529 + }, + { + "epoch": 0.8676704751794844, + "grad_norm": 0.3828410804271698, + "learning_rate": 0.0005682149260421335, + "loss": 0.3824, + "step": 15530 + }, + { + "epoch": 0.8677263457831662, + "grad_norm": 0.9293229579925537, + "learning_rate": 0.0005681869116987897, + "loss": 0.5519, + "step": 15531 + }, + { + "epoch": 0.867782216386848, + "grad_norm": 0.3368742763996124, + "learning_rate": 0.0005681588973554461, + "loss": 0.3555, + "step": 15532 + }, + { + "epoch": 0.8678380869905299, + "grad_norm": 0.4046901762485504, + "learning_rate": 0.0005681308830121023, + "loss": 0.4866, + "step": 15533 + }, + { + "epoch": 0.8678939575942118, + "grad_norm": 0.5146494507789612, + "learning_rate": 0.0005681028686687584, + "loss": 0.4872, + "step": 15534 + }, + { + "epoch": 0.8679498281978937, + "grad_norm": 0.6942834258079529, + "learning_rate": 0.0005680748543254146, + "loss": 0.5578, + "step": 15535 + }, + { + "epoch": 0.8680056988015755, + "grad_norm": 0.5304747223854065, + "learning_rate": 0.0005680468399820709, + "loss": 0.4153, + "step": 15536 + }, + { + "epoch": 0.8680615694052575, + "grad_norm": 0.49851444363594055, + "learning_rate": 0.0005680188256387271, + "loss": 0.6091, + "step": 15537 + }, + { + "epoch": 0.8681174400089393, + "grad_norm": 0.799596905708313, + "learning_rate": 0.0005679908112953832, + "loss": 0.506, + "step": 15538 + }, + { + "epoch": 0.8681733106126212, + "grad_norm": 0.5407029390335083, + "learning_rate": 0.0005679627969520395, + "loss": 0.4193, + "step": 15539 + }, + { + "epoch": 0.868229181216303, + "grad_norm": 0.34662675857543945, + "learning_rate": 0.0005679347826086957, + "loss": 0.3861, + "step": 15540 + }, + { + "epoch": 0.8682850518199849, + "grad_norm": 0.3954973518848419, + "learning_rate": 0.0005679067682653519, + "loss": 0.4014, + "step": 15541 + }, + { + "epoch": 0.8683409224236668, + "grad_norm": 0.4545825123786926, + "learning_rate": 0.000567878753922008, + "loss": 0.459, + "step": 15542 + }, + { + "epoch": 0.8683967930273486, + "grad_norm": 0.3604993522167206, + "learning_rate": 0.0005678507395786643, + "loss": 0.4942, + "step": 15543 + }, + { + "epoch": 0.8684526636310306, + "grad_norm": 0.41741928458213806, + "learning_rate": 0.0005678227252353205, + "loss": 0.4798, + "step": 15544 + }, + { + "epoch": 0.8685085342347124, + "grad_norm": 0.36523929238319397, + "learning_rate": 0.0005677947108919766, + "loss": 0.4275, + "step": 15545 + }, + { + "epoch": 0.8685644048383943, + "grad_norm": 0.31770041584968567, + "learning_rate": 0.0005677666965486329, + "loss": 0.3129, + "step": 15546 + }, + { + "epoch": 0.8686202754420762, + "grad_norm": 0.7366705536842346, + "learning_rate": 0.0005677386822052891, + "loss": 0.5033, + "step": 15547 + }, + { + "epoch": 0.868676146045758, + "grad_norm": 0.4980665445327759, + "learning_rate": 0.0005677106678619454, + "loss": 0.4971, + "step": 15548 + }, + { + "epoch": 0.8687320166494399, + "grad_norm": 0.5067864656448364, + "learning_rate": 0.0005676826535186014, + "loss": 0.4454, + "step": 15549 + }, + { + "epoch": 0.8687878872531217, + "grad_norm": 0.5294049382209778, + "learning_rate": 0.0005676546391752578, + "loss": 0.4245, + "step": 15550 + }, + { + "epoch": 0.8688437578568037, + "grad_norm": 0.5976264476776123, + "learning_rate": 0.000567626624831914, + "loss": 0.606, + "step": 15551 + }, + { + "epoch": 0.8688996284604855, + "grad_norm": 0.6345810890197754, + "learning_rate": 0.0005675986104885702, + "loss": 0.4638, + "step": 15552 + }, + { + "epoch": 0.8689554990641674, + "grad_norm": 0.6303694844245911, + "learning_rate": 0.0005675705961452264, + "loss": 0.4774, + "step": 15553 + }, + { + "epoch": 0.8690113696678493, + "grad_norm": 0.5195894241333008, + "learning_rate": 0.0005675425818018826, + "loss": 0.4223, + "step": 15554 + }, + { + "epoch": 0.8690672402715311, + "grad_norm": 1.414796233177185, + "learning_rate": 0.0005675145674585388, + "loss": 0.4427, + "step": 15555 + }, + { + "epoch": 0.869123110875213, + "grad_norm": 0.40186262130737305, + "learning_rate": 0.000567486553115195, + "loss": 0.4661, + "step": 15556 + }, + { + "epoch": 0.8691789814788948, + "grad_norm": 0.35705137252807617, + "learning_rate": 0.0005674585387718512, + "loss": 0.3627, + "step": 15557 + }, + { + "epoch": 0.8692348520825768, + "grad_norm": 0.4687630832195282, + "learning_rate": 0.0005674305244285074, + "loss": 0.4716, + "step": 15558 + }, + { + "epoch": 0.8692907226862586, + "grad_norm": 0.44288280606269836, + "learning_rate": 0.0005674025100851636, + "loss": 0.4686, + "step": 15559 + }, + { + "epoch": 0.8693465932899405, + "grad_norm": 0.8087920546531677, + "learning_rate": 0.0005673744957418198, + "loss": 0.4771, + "step": 15560 + }, + { + "epoch": 0.8694024638936224, + "grad_norm": 0.3421419858932495, + "learning_rate": 0.000567346481398476, + "loss": 0.3247, + "step": 15561 + }, + { + "epoch": 0.8694583344973043, + "grad_norm": 0.7028408050537109, + "learning_rate": 0.0005673184670551322, + "loss": 0.4998, + "step": 15562 + }, + { + "epoch": 0.8695142051009861, + "grad_norm": 0.36452487111091614, + "learning_rate": 0.0005672904527117884, + "loss": 0.3572, + "step": 15563 + }, + { + "epoch": 0.869570075704668, + "grad_norm": 0.40399497747421265, + "learning_rate": 0.0005672624383684448, + "loss": 0.4893, + "step": 15564 + }, + { + "epoch": 0.8696259463083499, + "grad_norm": 0.4128575623035431, + "learning_rate": 0.0005672344240251008, + "loss": 0.3783, + "step": 15565 + }, + { + "epoch": 0.8696818169120317, + "grad_norm": 0.600670337677002, + "learning_rate": 0.0005672064096817571, + "loss": 0.4422, + "step": 15566 + }, + { + "epoch": 0.8697376875157136, + "grad_norm": 0.5535273551940918, + "learning_rate": 0.0005671783953384133, + "loss": 0.4698, + "step": 15567 + }, + { + "epoch": 0.8697935581193955, + "grad_norm": 0.5258636474609375, + "learning_rate": 0.0005671503809950695, + "loss": 0.3391, + "step": 15568 + }, + { + "epoch": 0.8698494287230774, + "grad_norm": 0.3509262800216675, + "learning_rate": 0.0005671223666517257, + "loss": 0.3717, + "step": 15569 + }, + { + "epoch": 0.8699052993267592, + "grad_norm": 4.0546088218688965, + "learning_rate": 0.0005670943523083819, + "loss": 0.5088, + "step": 15570 + }, + { + "epoch": 0.8699611699304411, + "grad_norm": 0.4153389036655426, + "learning_rate": 0.0005670663379650382, + "loss": 0.3747, + "step": 15571 + }, + { + "epoch": 0.870017040534123, + "grad_norm": 0.37170976400375366, + "learning_rate": 0.0005670383236216943, + "loss": 0.3895, + "step": 15572 + }, + { + "epoch": 0.8700729111378048, + "grad_norm": 0.434535413980484, + "learning_rate": 0.0005670103092783505, + "loss": 0.4133, + "step": 15573 + }, + { + "epoch": 0.8701287817414867, + "grad_norm": 0.38594892621040344, + "learning_rate": 0.0005669822949350067, + "loss": 0.2789, + "step": 15574 + }, + { + "epoch": 0.8701846523451686, + "grad_norm": 0.39838358759880066, + "learning_rate": 0.000566954280591663, + "loss": 0.4482, + "step": 15575 + }, + { + "epoch": 0.8702405229488505, + "grad_norm": 0.4734336733818054, + "learning_rate": 0.0005669262662483191, + "loss": 0.3979, + "step": 15576 + }, + { + "epoch": 0.8702963935525323, + "grad_norm": 0.44306278228759766, + "learning_rate": 0.0005668982519049753, + "loss": 0.5301, + "step": 15577 + }, + { + "epoch": 0.8703522641562143, + "grad_norm": 0.41628578305244446, + "learning_rate": 0.0005668702375616316, + "loss": 0.4644, + "step": 15578 + }, + { + "epoch": 0.8704081347598961, + "grad_norm": 0.7745109796524048, + "learning_rate": 0.0005668422232182878, + "loss": 0.6039, + "step": 15579 + }, + { + "epoch": 0.8704640053635779, + "grad_norm": 0.5829578638076782, + "learning_rate": 0.0005668142088749439, + "loss": 0.4011, + "step": 15580 + }, + { + "epoch": 0.8705198759672598, + "grad_norm": 0.4152379333972931, + "learning_rate": 0.0005667861945316001, + "loss": 0.3644, + "step": 15581 + }, + { + "epoch": 0.8705757465709417, + "grad_norm": 0.40739700198173523, + "learning_rate": 0.0005667581801882565, + "loss": 0.4324, + "step": 15582 + }, + { + "epoch": 0.8706316171746236, + "grad_norm": 0.8340111374855042, + "learning_rate": 0.0005667301658449127, + "loss": 0.4829, + "step": 15583 + }, + { + "epoch": 0.8706874877783054, + "grad_norm": 0.40554511547088623, + "learning_rate": 0.0005667021515015688, + "loss": 0.3942, + "step": 15584 + }, + { + "epoch": 0.8707433583819874, + "grad_norm": 0.8848171234130859, + "learning_rate": 0.0005666741371582251, + "loss": 0.4266, + "step": 15585 + }, + { + "epoch": 0.8707992289856692, + "grad_norm": 0.3429234027862549, + "learning_rate": 0.0005666461228148813, + "loss": 0.4409, + "step": 15586 + }, + { + "epoch": 0.870855099589351, + "grad_norm": 0.5744084119796753, + "learning_rate": 0.0005666181084715374, + "loss": 0.3763, + "step": 15587 + }, + { + "epoch": 0.870910970193033, + "grad_norm": 0.6483365893363953, + "learning_rate": 0.0005665900941281936, + "loss": 0.7651, + "step": 15588 + }, + { + "epoch": 0.8709668407967148, + "grad_norm": 2.3268089294433594, + "learning_rate": 0.0005665620797848499, + "loss": 0.4317, + "step": 15589 + }, + { + "epoch": 0.8710227114003967, + "grad_norm": 0.4210550785064697, + "learning_rate": 0.0005665340654415061, + "loss": 0.3751, + "step": 15590 + }, + { + "epoch": 0.8710785820040785, + "grad_norm": 0.3629619777202606, + "learning_rate": 0.0005665060510981622, + "loss": 0.4738, + "step": 15591 + }, + { + "epoch": 0.8711344526077605, + "grad_norm": 0.4166366159915924, + "learning_rate": 0.0005664780367548184, + "loss": 0.4476, + "step": 15592 + }, + { + "epoch": 0.8711903232114423, + "grad_norm": 0.6801640391349792, + "learning_rate": 0.0005664500224114747, + "loss": 0.4122, + "step": 15593 + }, + { + "epoch": 0.8712461938151241, + "grad_norm": 0.507371723651886, + "learning_rate": 0.0005664220080681309, + "loss": 0.5513, + "step": 15594 + }, + { + "epoch": 0.8713020644188061, + "grad_norm": 0.3605135381221771, + "learning_rate": 0.000566393993724787, + "loss": 0.4277, + "step": 15595 + }, + { + "epoch": 0.8713579350224879, + "grad_norm": 0.5699144601821899, + "learning_rate": 0.0005663659793814433, + "loss": 0.3643, + "step": 15596 + }, + { + "epoch": 0.8714138056261698, + "grad_norm": 0.5424829125404358, + "learning_rate": 0.0005663379650380995, + "loss": 0.4943, + "step": 15597 + }, + { + "epoch": 0.8714696762298516, + "grad_norm": 0.5088583827018738, + "learning_rate": 0.0005663099506947557, + "loss": 0.4448, + "step": 15598 + }, + { + "epoch": 0.8715255468335336, + "grad_norm": 0.3700294494628906, + "learning_rate": 0.0005662819363514118, + "loss": 0.3716, + "step": 15599 + }, + { + "epoch": 0.8715814174372154, + "grad_norm": 0.6158493757247925, + "learning_rate": 0.0005662539220080682, + "loss": 0.5379, + "step": 15600 + }, + { + "epoch": 0.8716372880408972, + "grad_norm": 0.6974049210548401, + "learning_rate": 0.0005662259076647244, + "loss": 0.3718, + "step": 15601 + }, + { + "epoch": 0.8716931586445792, + "grad_norm": 0.48210370540618896, + "learning_rate": 0.0005661978933213806, + "loss": 0.503, + "step": 15602 + }, + { + "epoch": 0.871749029248261, + "grad_norm": 0.43593278527259827, + "learning_rate": 0.0005661698789780368, + "loss": 0.4136, + "step": 15603 + }, + { + "epoch": 0.8718048998519429, + "grad_norm": 0.5292016863822937, + "learning_rate": 0.000566141864634693, + "loss": 0.4683, + "step": 15604 + }, + { + "epoch": 0.8718607704556248, + "grad_norm": 0.9269499182701111, + "learning_rate": 0.0005661138502913492, + "loss": 0.4661, + "step": 15605 + }, + { + "epoch": 0.8719166410593067, + "grad_norm": 0.36909008026123047, + "learning_rate": 0.0005660858359480054, + "loss": 0.4455, + "step": 15606 + }, + { + "epoch": 0.8719725116629885, + "grad_norm": 0.5120836496353149, + "learning_rate": 0.0005660578216046616, + "loss": 0.5182, + "step": 15607 + }, + { + "epoch": 0.8720283822666703, + "grad_norm": 1.5719554424285889, + "learning_rate": 0.0005660298072613178, + "loss": 0.5636, + "step": 15608 + }, + { + "epoch": 0.8720842528703523, + "grad_norm": 0.40238356590270996, + "learning_rate": 0.000566001792917974, + "loss": 0.4029, + "step": 15609 + }, + { + "epoch": 0.8721401234740341, + "grad_norm": 0.3156065046787262, + "learning_rate": 0.0005659737785746302, + "loss": 0.3616, + "step": 15610 + }, + { + "epoch": 0.872195994077716, + "grad_norm": 0.3819146156311035, + "learning_rate": 0.0005659457642312864, + "loss": 0.3404, + "step": 15611 + }, + { + "epoch": 0.8722518646813979, + "grad_norm": 0.6023192405700684, + "learning_rate": 0.0005659177498879426, + "loss": 0.5572, + "step": 15612 + }, + { + "epoch": 0.8723077352850798, + "grad_norm": 0.46306318044662476, + "learning_rate": 0.0005658897355445988, + "loss": 0.4321, + "step": 15613 + }, + { + "epoch": 0.8723636058887616, + "grad_norm": 0.585515558719635, + "learning_rate": 0.000565861721201255, + "loss": 0.4169, + "step": 15614 + }, + { + "epoch": 0.8724194764924434, + "grad_norm": 0.4183819591999054, + "learning_rate": 0.0005658337068579112, + "loss": 0.4549, + "step": 15615 + }, + { + "epoch": 0.8724753470961254, + "grad_norm": 1.1183807849884033, + "learning_rate": 0.0005658056925145675, + "loss": 0.4148, + "step": 15616 + }, + { + "epoch": 0.8725312176998072, + "grad_norm": 5.021268367767334, + "learning_rate": 0.0005657776781712238, + "loss": 0.3742, + "step": 15617 + }, + { + "epoch": 0.8725870883034891, + "grad_norm": 0.759063720703125, + "learning_rate": 0.0005657496638278799, + "loss": 0.3675, + "step": 15618 + }, + { + "epoch": 0.872642958907171, + "grad_norm": 1.1385579109191895, + "learning_rate": 0.0005657216494845361, + "loss": 0.384, + "step": 15619 + }, + { + "epoch": 0.8726988295108529, + "grad_norm": 0.45480936765670776, + "learning_rate": 0.0005656936351411923, + "loss": 0.3932, + "step": 15620 + }, + { + "epoch": 0.8727547001145347, + "grad_norm": 0.8526319265365601, + "learning_rate": 0.0005656656207978486, + "loss": 0.3639, + "step": 15621 + }, + { + "epoch": 0.8728105707182167, + "grad_norm": 0.4300589859485626, + "learning_rate": 0.0005656376064545047, + "loss": 0.4656, + "step": 15622 + }, + { + "epoch": 0.8728664413218985, + "grad_norm": 0.44630762934684753, + "learning_rate": 0.0005656095921111609, + "loss": 0.4613, + "step": 15623 + }, + { + "epoch": 0.8729223119255803, + "grad_norm": 15.02567195892334, + "learning_rate": 0.0005655815777678172, + "loss": 0.5353, + "step": 15624 + }, + { + "epoch": 0.8729781825292622, + "grad_norm": 0.5070521831512451, + "learning_rate": 0.0005655535634244734, + "loss": 0.4483, + "step": 15625 + }, + { + "epoch": 0.8730340531329441, + "grad_norm": 0.8941291570663452, + "learning_rate": 0.0005655255490811295, + "loss": 0.3199, + "step": 15626 + }, + { + "epoch": 0.873089923736626, + "grad_norm": 0.4559859037399292, + "learning_rate": 0.0005654975347377857, + "loss": 0.3443, + "step": 15627 + }, + { + "epoch": 0.8731457943403078, + "grad_norm": 0.7440001368522644, + "learning_rate": 0.000565469520394442, + "loss": 0.5119, + "step": 15628 + }, + { + "epoch": 0.8732016649439898, + "grad_norm": 0.9972832202911377, + "learning_rate": 0.0005654415060510981, + "loss": 0.5076, + "step": 15629 + }, + { + "epoch": 0.8732575355476716, + "grad_norm": 0.31577399373054504, + "learning_rate": 0.0005654134917077543, + "loss": 0.3543, + "step": 15630 + }, + { + "epoch": 0.8733134061513534, + "grad_norm": 0.3290747404098511, + "learning_rate": 0.0005653854773644105, + "loss": 0.3922, + "step": 15631 + }, + { + "epoch": 0.8733692767550353, + "grad_norm": 0.6234573125839233, + "learning_rate": 0.0005653574630210669, + "loss": 0.4243, + "step": 15632 + }, + { + "epoch": 0.8734251473587172, + "grad_norm": 0.47538477182388306, + "learning_rate": 0.000565329448677723, + "loss": 0.3692, + "step": 15633 + }, + { + "epoch": 0.8734810179623991, + "grad_norm": 0.386666476726532, + "learning_rate": 0.0005653014343343792, + "loss": 0.4952, + "step": 15634 + }, + { + "epoch": 0.8735368885660809, + "grad_norm": 0.49188706278800964, + "learning_rate": 0.0005652734199910355, + "loss": 0.5857, + "step": 15635 + }, + { + "epoch": 0.8735927591697629, + "grad_norm": 0.4234575927257538, + "learning_rate": 0.0005652454056476917, + "loss": 0.3978, + "step": 15636 + }, + { + "epoch": 0.8736486297734447, + "grad_norm": 0.5979828238487244, + "learning_rate": 0.0005652173913043478, + "loss": 0.5258, + "step": 15637 + }, + { + "epoch": 0.8737045003771265, + "grad_norm": 0.5010218024253845, + "learning_rate": 0.000565189376961004, + "loss": 0.6311, + "step": 15638 + }, + { + "epoch": 0.8737603709808085, + "grad_norm": 0.37362855672836304, + "learning_rate": 0.0005651613626176603, + "loss": 0.3337, + "step": 15639 + }, + { + "epoch": 0.8738162415844903, + "grad_norm": 0.4632851183414459, + "learning_rate": 0.0005651333482743165, + "loss": 0.4478, + "step": 15640 + }, + { + "epoch": 0.8738721121881722, + "grad_norm": 0.4855749309062958, + "learning_rate": 0.0005651053339309726, + "loss": 0.4783, + "step": 15641 + }, + { + "epoch": 0.873927982791854, + "grad_norm": 0.4815387427806854, + "learning_rate": 0.0005650773195876289, + "loss": 0.4253, + "step": 15642 + }, + { + "epoch": 0.873983853395536, + "grad_norm": 1.5447298288345337, + "learning_rate": 0.0005650493052442851, + "loss": 0.5572, + "step": 15643 + }, + { + "epoch": 0.8740397239992178, + "grad_norm": 0.42875176668167114, + "learning_rate": 0.0005650212909009413, + "loss": 0.4743, + "step": 15644 + }, + { + "epoch": 0.8740955946028997, + "grad_norm": 0.49657517671585083, + "learning_rate": 0.0005649932765575974, + "loss": 0.3735, + "step": 15645 + }, + { + "epoch": 0.8741514652065816, + "grad_norm": 0.8785173296928406, + "learning_rate": 0.0005649652622142537, + "loss": 0.3225, + "step": 15646 + }, + { + "epoch": 0.8742073358102634, + "grad_norm": 0.4296983480453491, + "learning_rate": 0.0005649372478709099, + "loss": 0.4087, + "step": 15647 + }, + { + "epoch": 0.8742632064139453, + "grad_norm": 0.5186469554901123, + "learning_rate": 0.0005649092335275661, + "loss": 0.5078, + "step": 15648 + }, + { + "epoch": 0.8743190770176271, + "grad_norm": 0.9322760701179504, + "learning_rate": 0.0005648812191842224, + "loss": 0.3953, + "step": 15649 + }, + { + "epoch": 0.8743749476213091, + "grad_norm": 3.8627078533172607, + "learning_rate": 0.0005648532048408786, + "loss": 0.6654, + "step": 15650 + }, + { + "epoch": 0.8744308182249909, + "grad_norm": 0.42584919929504395, + "learning_rate": 0.0005648251904975348, + "loss": 0.3528, + "step": 15651 + }, + { + "epoch": 0.8744866888286728, + "grad_norm": 0.858518660068512, + "learning_rate": 0.0005647971761541909, + "loss": 0.5681, + "step": 15652 + }, + { + "epoch": 0.8745425594323547, + "grad_norm": 0.4805578589439392, + "learning_rate": 0.0005647691618108472, + "loss": 0.5868, + "step": 15653 + }, + { + "epoch": 0.8745984300360365, + "grad_norm": 0.5563173890113831, + "learning_rate": 0.0005647411474675034, + "loss": 0.5978, + "step": 15654 + }, + { + "epoch": 0.8746543006397184, + "grad_norm": 0.47049590945243835, + "learning_rate": 0.0005647131331241596, + "loss": 0.4459, + "step": 15655 + }, + { + "epoch": 0.8747101712434003, + "grad_norm": 5.326593399047852, + "learning_rate": 0.0005646851187808158, + "loss": 0.5706, + "step": 15656 + }, + { + "epoch": 0.8747660418470822, + "grad_norm": 0.6046302318572998, + "learning_rate": 0.000564657104437472, + "loss": 0.4917, + "step": 15657 + }, + { + "epoch": 0.874821912450764, + "grad_norm": 1.1510456800460815, + "learning_rate": 0.0005646290900941282, + "loss": 0.4093, + "step": 15658 + }, + { + "epoch": 0.8748777830544459, + "grad_norm": 0.5254102945327759, + "learning_rate": 0.0005646010757507844, + "loss": 0.4975, + "step": 15659 + }, + { + "epoch": 0.8749336536581278, + "grad_norm": 0.5020347833633423, + "learning_rate": 0.0005645730614074406, + "loss": 0.4511, + "step": 15660 + }, + { + "epoch": 0.8749895242618096, + "grad_norm": 0.5132710933685303, + "learning_rate": 0.0005645450470640968, + "loss": 0.3798, + "step": 15661 + }, + { + "epoch": 0.8750453948654915, + "grad_norm": 0.3693525791168213, + "learning_rate": 0.000564517032720753, + "loss": 0.4602, + "step": 15662 + }, + { + "epoch": 0.8751012654691734, + "grad_norm": 1.5387340784072876, + "learning_rate": 0.0005644890183774093, + "loss": 0.3611, + "step": 15663 + }, + { + "epoch": 0.8751571360728553, + "grad_norm": 0.4025547504425049, + "learning_rate": 0.0005644610040340654, + "loss": 0.4183, + "step": 15664 + }, + { + "epoch": 0.8752130066765371, + "grad_norm": 2.7363016605377197, + "learning_rate": 0.0005644329896907216, + "loss": 0.4639, + "step": 15665 + }, + { + "epoch": 0.875268877280219, + "grad_norm": 0.573032557964325, + "learning_rate": 0.0005644049753473779, + "loss": 0.3954, + "step": 15666 + }, + { + "epoch": 0.8753247478839009, + "grad_norm": 0.732589840888977, + "learning_rate": 0.0005643769610040342, + "loss": 0.5319, + "step": 15667 + }, + { + "epoch": 0.8753806184875828, + "grad_norm": 3.0331692695617676, + "learning_rate": 0.0005643489466606903, + "loss": 0.3796, + "step": 15668 + }, + { + "epoch": 0.8754364890912646, + "grad_norm": 0.5315291285514832, + "learning_rate": 0.0005643209323173465, + "loss": 0.382, + "step": 15669 + }, + { + "epoch": 0.8754923596949465, + "grad_norm": 3.1562108993530273, + "learning_rate": 0.0005642929179740027, + "loss": 0.4426, + "step": 15670 + }, + { + "epoch": 0.8755482302986284, + "grad_norm": 0.41537222266197205, + "learning_rate": 0.0005642649036306589, + "loss": 0.4766, + "step": 15671 + }, + { + "epoch": 0.8756041009023102, + "grad_norm": 0.6032430529594421, + "learning_rate": 0.0005642368892873151, + "loss": 0.4363, + "step": 15672 + }, + { + "epoch": 0.8756599715059922, + "grad_norm": 0.7626582980155945, + "learning_rate": 0.0005642088749439713, + "loss": 0.459, + "step": 15673 + }, + { + "epoch": 0.875715842109674, + "grad_norm": 0.4906750023365021, + "learning_rate": 0.0005641808606006276, + "loss": 0.4535, + "step": 15674 + }, + { + "epoch": 0.8757717127133559, + "grad_norm": 0.47283732891082764, + "learning_rate": 0.0005641528462572837, + "loss": 0.4783, + "step": 15675 + }, + { + "epoch": 0.8758275833170377, + "grad_norm": 0.43194153904914856, + "learning_rate": 0.0005641248319139399, + "loss": 0.5245, + "step": 15676 + }, + { + "epoch": 0.8758834539207196, + "grad_norm": 7.022581577301025, + "learning_rate": 0.0005640968175705961, + "loss": 0.4684, + "step": 15677 + }, + { + "epoch": 0.8759393245244015, + "grad_norm": 0.41412556171417236, + "learning_rate": 0.0005640688032272524, + "loss": 0.4121, + "step": 15678 + }, + { + "epoch": 0.8759951951280833, + "grad_norm": 1.3615074157714844, + "learning_rate": 0.0005640407888839085, + "loss": 0.5485, + "step": 15679 + }, + { + "epoch": 0.8760510657317653, + "grad_norm": 0.7192856669425964, + "learning_rate": 0.0005640127745405647, + "loss": 0.4565, + "step": 15680 + }, + { + "epoch": 0.8761069363354471, + "grad_norm": 0.4712297022342682, + "learning_rate": 0.000563984760197221, + "loss": 0.4807, + "step": 15681 + }, + { + "epoch": 0.876162806939129, + "grad_norm": 0.382110059261322, + "learning_rate": 0.0005639567458538773, + "loss": 0.389, + "step": 15682 + }, + { + "epoch": 0.8762186775428108, + "grad_norm": 0.8522199392318726, + "learning_rate": 0.0005639287315105334, + "loss": 0.3702, + "step": 15683 + }, + { + "epoch": 0.8762745481464927, + "grad_norm": 0.5892477035522461, + "learning_rate": 0.0005639007171671896, + "loss": 0.3963, + "step": 15684 + }, + { + "epoch": 0.8763304187501746, + "grad_norm": 0.49911120533943176, + "learning_rate": 0.0005638727028238459, + "loss": 0.5553, + "step": 15685 + }, + { + "epoch": 0.8763862893538564, + "grad_norm": 0.5822484493255615, + "learning_rate": 0.0005638446884805021, + "loss": 0.5358, + "step": 15686 + }, + { + "epoch": 0.8764421599575384, + "grad_norm": 0.36750462651252747, + "learning_rate": 0.0005638166741371582, + "loss": 0.4509, + "step": 15687 + }, + { + "epoch": 0.8764980305612202, + "grad_norm": 0.420761376619339, + "learning_rate": 0.0005637886597938145, + "loss": 0.3825, + "step": 15688 + }, + { + "epoch": 0.8765539011649021, + "grad_norm": 0.429214745759964, + "learning_rate": 0.0005637606454504707, + "loss": 0.3506, + "step": 15689 + }, + { + "epoch": 0.876609771768584, + "grad_norm": 0.5788127183914185, + "learning_rate": 0.0005637326311071269, + "loss": 0.3764, + "step": 15690 + }, + { + "epoch": 0.8766656423722659, + "grad_norm": 0.48898056149482727, + "learning_rate": 0.000563704616763783, + "loss": 0.4146, + "step": 15691 + }, + { + "epoch": 0.8767215129759477, + "grad_norm": 2.2822680473327637, + "learning_rate": 0.0005636766024204393, + "loss": 0.4601, + "step": 15692 + }, + { + "epoch": 0.8767773835796295, + "grad_norm": 1.101352334022522, + "learning_rate": 0.0005636485880770955, + "loss": 0.4329, + "step": 15693 + }, + { + "epoch": 0.8768332541833115, + "grad_norm": 0.7381768822669983, + "learning_rate": 0.0005636205737337516, + "loss": 0.4666, + "step": 15694 + }, + { + "epoch": 0.8768891247869933, + "grad_norm": 0.4034422039985657, + "learning_rate": 0.0005635925593904079, + "loss": 0.332, + "step": 15695 + }, + { + "epoch": 0.8769449953906752, + "grad_norm": 0.43841031193733215, + "learning_rate": 0.0005635645450470641, + "loss": 0.4024, + "step": 15696 + }, + { + "epoch": 0.8770008659943571, + "grad_norm": 2.534879446029663, + "learning_rate": 0.0005635365307037203, + "loss": 0.3743, + "step": 15697 + }, + { + "epoch": 0.877056736598039, + "grad_norm": 0.5635911822319031, + "learning_rate": 0.0005635085163603764, + "loss": 0.3763, + "step": 15698 + }, + { + "epoch": 0.8771126072017208, + "grad_norm": 0.47352975606918335, + "learning_rate": 0.0005634805020170328, + "loss": 0.7128, + "step": 15699 + }, + { + "epoch": 0.8771684778054026, + "grad_norm": 1.1542483568191528, + "learning_rate": 0.000563452487673689, + "loss": 0.4185, + "step": 15700 + }, + { + "epoch": 0.8772243484090846, + "grad_norm": 0.4086190462112427, + "learning_rate": 0.0005634244733303452, + "loss": 0.4144, + "step": 15701 + }, + { + "epoch": 0.8772802190127664, + "grad_norm": 1.5467733144760132, + "learning_rate": 0.0005633964589870013, + "loss": 0.3991, + "step": 15702 + }, + { + "epoch": 0.8773360896164483, + "grad_norm": 14.590704917907715, + "learning_rate": 0.0005633684446436576, + "loss": 0.4706, + "step": 15703 + }, + { + "epoch": 0.8773919602201302, + "grad_norm": 0.7785755395889282, + "learning_rate": 0.0005633404303003138, + "loss": 0.5597, + "step": 15704 + }, + { + "epoch": 0.8774478308238121, + "grad_norm": 0.3645383417606354, + "learning_rate": 0.00056331241595697, + "loss": 0.5245, + "step": 15705 + }, + { + "epoch": 0.8775037014274939, + "grad_norm": 1.0634894371032715, + "learning_rate": 0.0005632844016136262, + "loss": 0.4177, + "step": 15706 + }, + { + "epoch": 0.8775595720311758, + "grad_norm": 1.0705634355545044, + "learning_rate": 0.0005632563872702824, + "loss": 0.5165, + "step": 15707 + }, + { + "epoch": 0.8776154426348577, + "grad_norm": 0.6703816056251526, + "learning_rate": 0.0005632283729269386, + "loss": 0.551, + "step": 15708 + }, + { + "epoch": 0.8776713132385395, + "grad_norm": 0.3830517530441284, + "learning_rate": 0.0005632003585835948, + "loss": 0.4425, + "step": 15709 + }, + { + "epoch": 0.8777271838422214, + "grad_norm": 0.49357619881629944, + "learning_rate": 0.000563172344240251, + "loss": 0.4238, + "step": 15710 + }, + { + "epoch": 0.8777830544459033, + "grad_norm": 0.7117871046066284, + "learning_rate": 0.0005631443298969072, + "loss": 0.4635, + "step": 15711 + }, + { + "epoch": 0.8778389250495852, + "grad_norm": 1.0245977640151978, + "learning_rate": 0.0005631163155535634, + "loss": 0.4109, + "step": 15712 + }, + { + "epoch": 0.877894795653267, + "grad_norm": 0.4246136248111725, + "learning_rate": 0.0005630883012102196, + "loss": 0.3877, + "step": 15713 + }, + { + "epoch": 0.877950666256949, + "grad_norm": 0.685916543006897, + "learning_rate": 0.0005630602868668758, + "loss": 0.458, + "step": 15714 + }, + { + "epoch": 0.8780065368606308, + "grad_norm": 0.5592381954193115, + "learning_rate": 0.000563032272523532, + "loss": 0.4939, + "step": 15715 + }, + { + "epoch": 0.8780624074643126, + "grad_norm": 0.5766861438751221, + "learning_rate": 0.0005630042581801883, + "loss": 0.4272, + "step": 15716 + }, + { + "epoch": 0.8781182780679945, + "grad_norm": 1.683650255203247, + "learning_rate": 0.0005629762438368445, + "loss": 0.362, + "step": 15717 + }, + { + "epoch": 0.8781741486716764, + "grad_norm": 0.5497289896011353, + "learning_rate": 0.0005629482294935007, + "loss": 0.4589, + "step": 15718 + }, + { + "epoch": 0.8782300192753583, + "grad_norm": 0.536236584186554, + "learning_rate": 0.0005629202151501569, + "loss": 0.3952, + "step": 15719 + }, + { + "epoch": 0.8782858898790401, + "grad_norm": 0.7623941898345947, + "learning_rate": 0.0005628922008068132, + "loss": 0.4231, + "step": 15720 + }, + { + "epoch": 0.8783417604827221, + "grad_norm": 0.45077306032180786, + "learning_rate": 0.0005628641864634693, + "loss": 0.4815, + "step": 15721 + }, + { + "epoch": 0.8783976310864039, + "grad_norm": 0.926594614982605, + "learning_rate": 0.0005628361721201255, + "loss": 0.4353, + "step": 15722 + }, + { + "epoch": 0.8784535016900857, + "grad_norm": 0.5758428573608398, + "learning_rate": 0.0005628081577767817, + "loss": 0.4389, + "step": 15723 + }, + { + "epoch": 0.8785093722937677, + "grad_norm": 0.40980255603790283, + "learning_rate": 0.000562780143433438, + "loss": 0.4072, + "step": 15724 + }, + { + "epoch": 0.8785652428974495, + "grad_norm": 1.8538466691970825, + "learning_rate": 0.0005627521290900941, + "loss": 0.4135, + "step": 15725 + }, + { + "epoch": 0.8786211135011314, + "grad_norm": 0.594677746295929, + "learning_rate": 0.0005627241147467503, + "loss": 0.4153, + "step": 15726 + }, + { + "epoch": 0.8786769841048132, + "grad_norm": 0.6507859230041504, + "learning_rate": 0.0005626961004034066, + "loss": 0.4379, + "step": 15727 + }, + { + "epoch": 0.8787328547084952, + "grad_norm": 0.5464143753051758, + "learning_rate": 0.0005626680860600628, + "loss": 0.4614, + "step": 15728 + }, + { + "epoch": 0.878788725312177, + "grad_norm": 1.3898305892944336, + "learning_rate": 0.0005626400717167189, + "loss": 0.4485, + "step": 15729 + }, + { + "epoch": 0.8788445959158588, + "grad_norm": 0.8124890923500061, + "learning_rate": 0.0005626120573733751, + "loss": 0.4755, + "step": 15730 + }, + { + "epoch": 0.8789004665195408, + "grad_norm": 0.8208053112030029, + "learning_rate": 0.0005625840430300314, + "loss": 0.4035, + "step": 15731 + }, + { + "epoch": 0.8789563371232226, + "grad_norm": 0.37136808037757874, + "learning_rate": 0.0005625560286866877, + "loss": 0.3662, + "step": 15732 + }, + { + "epoch": 0.8790122077269045, + "grad_norm": 0.4130733013153076, + "learning_rate": 0.0005625280143433438, + "loss": 0.4305, + "step": 15733 + }, + { + "epoch": 0.8790680783305863, + "grad_norm": 0.451446533203125, + "learning_rate": 0.0005625000000000001, + "loss": 0.5035, + "step": 15734 + }, + { + "epoch": 0.8791239489342683, + "grad_norm": 0.420561283826828, + "learning_rate": 0.0005624719856566563, + "loss": 0.3914, + "step": 15735 + }, + { + "epoch": 0.8791798195379501, + "grad_norm": 0.7355039715766907, + "learning_rate": 0.0005624439713133124, + "loss": 0.3718, + "step": 15736 + }, + { + "epoch": 0.8792356901416319, + "grad_norm": 0.6187682151794434, + "learning_rate": 0.0005624159569699686, + "loss": 0.5341, + "step": 15737 + }, + { + "epoch": 0.8792915607453139, + "grad_norm": 0.8441882729530334, + "learning_rate": 0.0005623879426266249, + "loss": 0.3944, + "step": 15738 + }, + { + "epoch": 0.8793474313489957, + "grad_norm": 0.4764167070388794, + "learning_rate": 0.0005623599282832811, + "loss": 0.4042, + "step": 15739 + }, + { + "epoch": 0.8794033019526776, + "grad_norm": 0.8448765873908997, + "learning_rate": 0.0005623319139399372, + "loss": 0.4288, + "step": 15740 + }, + { + "epoch": 0.8794591725563595, + "grad_norm": 1.0719871520996094, + "learning_rate": 0.0005623038995965934, + "loss": 0.4017, + "step": 15741 + }, + { + "epoch": 0.8795150431600414, + "grad_norm": 0.4391495883464813, + "learning_rate": 0.0005622758852532497, + "loss": 0.418, + "step": 15742 + }, + { + "epoch": 0.8795709137637232, + "grad_norm": 0.4316994249820709, + "learning_rate": 0.0005622478709099059, + "loss": 0.4102, + "step": 15743 + }, + { + "epoch": 0.879626784367405, + "grad_norm": 0.5204192996025085, + "learning_rate": 0.000562219856566562, + "loss": 0.4542, + "step": 15744 + }, + { + "epoch": 0.879682654971087, + "grad_norm": 0.4642358422279358, + "learning_rate": 0.0005621918422232183, + "loss": 0.4724, + "step": 15745 + }, + { + "epoch": 0.8797385255747688, + "grad_norm": 0.8664199709892273, + "learning_rate": 0.0005621638278798745, + "loss": 0.4841, + "step": 15746 + }, + { + "epoch": 0.8797943961784507, + "grad_norm": 0.8419342637062073, + "learning_rate": 0.0005621358135365307, + "loss": 0.4892, + "step": 15747 + }, + { + "epoch": 0.8798502667821326, + "grad_norm": 0.5130323171615601, + "learning_rate": 0.0005621077991931868, + "loss": 0.4999, + "step": 15748 + }, + { + "epoch": 0.8799061373858145, + "grad_norm": 0.3515557646751404, + "learning_rate": 0.0005620797848498432, + "loss": 0.3965, + "step": 15749 + }, + { + "epoch": 0.8799620079894963, + "grad_norm": 0.4702773094177246, + "learning_rate": 0.0005620517705064994, + "loss": 0.5141, + "step": 15750 + }, + { + "epoch": 0.8800178785931781, + "grad_norm": 0.7177945375442505, + "learning_rate": 0.0005620237561631556, + "loss": 0.422, + "step": 15751 + }, + { + "epoch": 0.8800737491968601, + "grad_norm": 0.9034556150436401, + "learning_rate": 0.0005619957418198118, + "loss": 0.4622, + "step": 15752 + }, + { + "epoch": 0.8801296198005419, + "grad_norm": 1.0782307386398315, + "learning_rate": 0.000561967727476468, + "loss": 0.4302, + "step": 15753 + }, + { + "epoch": 0.8801854904042238, + "grad_norm": 0.38013333082199097, + "learning_rate": 0.0005619397131331242, + "loss": 0.3584, + "step": 15754 + }, + { + "epoch": 0.8802413610079057, + "grad_norm": 0.663632869720459, + "learning_rate": 0.0005619116987897804, + "loss": 0.5096, + "step": 15755 + }, + { + "epoch": 0.8802972316115876, + "grad_norm": 0.5985325574874878, + "learning_rate": 0.0005618836844464366, + "loss": 0.4317, + "step": 15756 + }, + { + "epoch": 0.8803531022152694, + "grad_norm": 0.3906570374965668, + "learning_rate": 0.0005618556701030928, + "loss": 0.3604, + "step": 15757 + }, + { + "epoch": 0.8804089728189514, + "grad_norm": 0.6488112807273865, + "learning_rate": 0.000561827655759749, + "loss": 0.4344, + "step": 15758 + }, + { + "epoch": 0.8804648434226332, + "grad_norm": 0.7297333478927612, + "learning_rate": 0.0005617996414164052, + "loss": 0.4075, + "step": 15759 + }, + { + "epoch": 0.880520714026315, + "grad_norm": 0.5327625274658203, + "learning_rate": 0.0005617716270730614, + "loss": 0.5041, + "step": 15760 + }, + { + "epoch": 0.8805765846299969, + "grad_norm": 0.3808210790157318, + "learning_rate": 0.0005617436127297176, + "loss": 0.351, + "step": 15761 + }, + { + "epoch": 0.8806324552336788, + "grad_norm": 1.0965451002120972, + "learning_rate": 0.0005617155983863738, + "loss": 0.5247, + "step": 15762 + }, + { + "epoch": 0.8806883258373607, + "grad_norm": 0.6610701084136963, + "learning_rate": 0.00056168758404303, + "loss": 0.6361, + "step": 15763 + }, + { + "epoch": 0.8807441964410425, + "grad_norm": 0.7227989435195923, + "learning_rate": 0.0005616595696996862, + "loss": 0.3209, + "step": 15764 + }, + { + "epoch": 0.8808000670447245, + "grad_norm": 0.6478484272956848, + "learning_rate": 0.0005616315553563424, + "loss": 0.4815, + "step": 15765 + }, + { + "epoch": 0.8808559376484063, + "grad_norm": 0.4123193025588989, + "learning_rate": 0.0005616035410129988, + "loss": 0.4627, + "step": 15766 + }, + { + "epoch": 0.8809118082520881, + "grad_norm": 0.5504810214042664, + "learning_rate": 0.0005615755266696549, + "loss": 0.3399, + "step": 15767 + }, + { + "epoch": 0.88096767885577, + "grad_norm": 0.4050418734550476, + "learning_rate": 0.0005615475123263111, + "loss": 0.398, + "step": 15768 + }, + { + "epoch": 0.8810235494594519, + "grad_norm": 0.3998817205429077, + "learning_rate": 0.0005615194979829673, + "loss": 0.4736, + "step": 15769 + }, + { + "epoch": 0.8810794200631338, + "grad_norm": 1.6283667087554932, + "learning_rate": 0.0005614914836396236, + "loss": 0.4167, + "step": 15770 + }, + { + "epoch": 0.8811352906668156, + "grad_norm": 0.4423868656158447, + "learning_rate": 0.0005614634692962797, + "loss": 0.5092, + "step": 15771 + }, + { + "epoch": 0.8811911612704976, + "grad_norm": 0.3749750852584839, + "learning_rate": 0.0005614354549529359, + "loss": 0.3598, + "step": 15772 + }, + { + "epoch": 0.8812470318741794, + "grad_norm": 0.449134886264801, + "learning_rate": 0.0005614074406095921, + "loss": 0.4465, + "step": 15773 + }, + { + "epoch": 0.8813029024778612, + "grad_norm": 0.6128026247024536, + "learning_rate": 0.0005613794262662484, + "loss": 0.3881, + "step": 15774 + }, + { + "epoch": 0.8813587730815432, + "grad_norm": 5.229954719543457, + "learning_rate": 0.0005613514119229045, + "loss": 0.6039, + "step": 15775 + }, + { + "epoch": 0.881414643685225, + "grad_norm": 0.8812405467033386, + "learning_rate": 0.0005613233975795607, + "loss": 0.345, + "step": 15776 + }, + { + "epoch": 0.8814705142889069, + "grad_norm": 0.44368645548820496, + "learning_rate": 0.000561295383236217, + "loss": 0.4435, + "step": 15777 + }, + { + "epoch": 0.8815263848925887, + "grad_norm": 0.5021790862083435, + "learning_rate": 0.0005612673688928731, + "loss": 0.4246, + "step": 15778 + }, + { + "epoch": 0.8815822554962707, + "grad_norm": 0.46310800313949585, + "learning_rate": 0.0005612393545495293, + "loss": 0.4198, + "step": 15779 + }, + { + "epoch": 0.8816381260999525, + "grad_norm": 0.4456344544887543, + "learning_rate": 0.0005612113402061855, + "loss": 0.4018, + "step": 15780 + }, + { + "epoch": 0.8816939967036344, + "grad_norm": 1.319220781326294, + "learning_rate": 0.0005611833258628418, + "loss": 0.6561, + "step": 15781 + }, + { + "epoch": 0.8817498673073163, + "grad_norm": 0.4201894700527191, + "learning_rate": 0.0005611553115194979, + "loss": 0.4262, + "step": 15782 + }, + { + "epoch": 0.8818057379109981, + "grad_norm": 0.4602068066596985, + "learning_rate": 0.0005611272971761542, + "loss": 0.4126, + "step": 15783 + }, + { + "epoch": 0.88186160851468, + "grad_norm": 0.7959999442100525, + "learning_rate": 0.0005610992828328105, + "loss": 0.5482, + "step": 15784 + }, + { + "epoch": 0.8819174791183618, + "grad_norm": 0.386572927236557, + "learning_rate": 0.0005610712684894667, + "loss": 0.435, + "step": 15785 + }, + { + "epoch": 0.8819733497220438, + "grad_norm": 0.6325757503509521, + "learning_rate": 0.0005610432541461228, + "loss": 0.5572, + "step": 15786 + }, + { + "epoch": 0.8820292203257256, + "grad_norm": 0.5429150462150574, + "learning_rate": 0.000561015239802779, + "loss": 0.4408, + "step": 15787 + }, + { + "epoch": 0.8820850909294075, + "grad_norm": 0.623649537563324, + "learning_rate": 0.0005609872254594353, + "loss": 0.4604, + "step": 15788 + }, + { + "epoch": 0.8821409615330894, + "grad_norm": 1.0045404434204102, + "learning_rate": 0.0005609592111160915, + "loss": 0.4746, + "step": 15789 + }, + { + "epoch": 0.8821968321367712, + "grad_norm": 0.353819340467453, + "learning_rate": 0.0005609311967727476, + "loss": 0.4503, + "step": 15790 + }, + { + "epoch": 0.8822527027404531, + "grad_norm": 0.45374152064323425, + "learning_rate": 0.0005609031824294039, + "loss": 0.5417, + "step": 15791 + }, + { + "epoch": 0.882308573344135, + "grad_norm": 0.8123939037322998, + "learning_rate": 0.0005608751680860601, + "loss": 0.4353, + "step": 15792 + }, + { + "epoch": 0.8823644439478169, + "grad_norm": 0.9229682683944702, + "learning_rate": 0.0005608471537427163, + "loss": 0.5225, + "step": 15793 + }, + { + "epoch": 0.8824203145514987, + "grad_norm": 0.3993086516857147, + "learning_rate": 0.0005608191393993724, + "loss": 0.4559, + "step": 15794 + }, + { + "epoch": 0.8824761851551806, + "grad_norm": 0.4682874083518982, + "learning_rate": 0.0005607911250560287, + "loss": 0.4999, + "step": 15795 + }, + { + "epoch": 0.8825320557588625, + "grad_norm": 0.4480965733528137, + "learning_rate": 0.0005607631107126849, + "loss": 0.3931, + "step": 15796 + }, + { + "epoch": 0.8825879263625444, + "grad_norm": 0.4351092278957367, + "learning_rate": 0.0005607350963693411, + "loss": 0.3788, + "step": 15797 + }, + { + "epoch": 0.8826437969662262, + "grad_norm": 0.45054277777671814, + "learning_rate": 0.0005607070820259973, + "loss": 0.4462, + "step": 15798 + }, + { + "epoch": 0.8826996675699081, + "grad_norm": 0.38447731733322144, + "learning_rate": 0.0005606790676826536, + "loss": 0.4439, + "step": 15799 + }, + { + "epoch": 0.88275553817359, + "grad_norm": 0.5988203287124634, + "learning_rate": 0.0005606510533393098, + "loss": 0.4999, + "step": 15800 + }, + { + "epoch": 0.8828114087772718, + "grad_norm": 0.42915526032447815, + "learning_rate": 0.0005606230389959659, + "loss": 0.3628, + "step": 15801 + }, + { + "epoch": 0.8828672793809537, + "grad_norm": 0.6621850728988647, + "learning_rate": 0.0005605950246526222, + "loss": 0.3565, + "step": 15802 + }, + { + "epoch": 0.8829231499846356, + "grad_norm": 0.7220125198364258, + "learning_rate": 0.0005605670103092784, + "loss": 0.4501, + "step": 15803 + }, + { + "epoch": 0.8829790205883175, + "grad_norm": 0.35491499304771423, + "learning_rate": 0.0005605389959659346, + "loss": 0.4251, + "step": 15804 + }, + { + "epoch": 0.8830348911919993, + "grad_norm": 0.5271967649459839, + "learning_rate": 0.0005605109816225907, + "loss": 0.4965, + "step": 15805 + }, + { + "epoch": 0.8830907617956812, + "grad_norm": 1.471633791923523, + "learning_rate": 0.000560482967279247, + "loss": 0.4412, + "step": 15806 + }, + { + "epoch": 0.8831466323993631, + "grad_norm": 1.2486097812652588, + "learning_rate": 0.0005604549529359032, + "loss": 0.4587, + "step": 15807 + }, + { + "epoch": 0.8832025030030449, + "grad_norm": 0.541053056716919, + "learning_rate": 0.0005604269385925594, + "loss": 0.4661, + "step": 15808 + }, + { + "epoch": 0.8832583736067268, + "grad_norm": 9.21313190460205, + "learning_rate": 0.0005603989242492156, + "loss": 0.3495, + "step": 15809 + }, + { + "epoch": 0.8833142442104087, + "grad_norm": 1.0234848260879517, + "learning_rate": 0.0005603709099058718, + "loss": 0.4088, + "step": 15810 + }, + { + "epoch": 0.8833701148140906, + "grad_norm": 0.407552033662796, + "learning_rate": 0.000560342895562528, + "loss": 0.4161, + "step": 15811 + }, + { + "epoch": 0.8834259854177724, + "grad_norm": 0.6531028151512146, + "learning_rate": 0.0005603148812191842, + "loss": 0.4232, + "step": 15812 + }, + { + "epoch": 0.8834818560214543, + "grad_norm": 0.7799391746520996, + "learning_rate": 0.0005602868668758404, + "loss": 0.5048, + "step": 15813 + }, + { + "epoch": 0.8835377266251362, + "grad_norm": 1.5924577713012695, + "learning_rate": 0.0005602588525324966, + "loss": 0.4202, + "step": 15814 + }, + { + "epoch": 0.883593597228818, + "grad_norm": 0.3515469431877136, + "learning_rate": 0.0005602308381891528, + "loss": 0.4265, + "step": 15815 + }, + { + "epoch": 0.8836494678325, + "grad_norm": 0.4949039816856384, + "learning_rate": 0.0005602028238458092, + "loss": 0.4156, + "step": 15816 + }, + { + "epoch": 0.8837053384361818, + "grad_norm": 0.617041289806366, + "learning_rate": 0.0005601748095024653, + "loss": 0.4709, + "step": 15817 + }, + { + "epoch": 0.8837612090398637, + "grad_norm": 0.6689000725746155, + "learning_rate": 0.0005601467951591215, + "loss": 0.4184, + "step": 15818 + }, + { + "epoch": 0.8838170796435455, + "grad_norm": 0.4144006073474884, + "learning_rate": 0.0005601187808157777, + "loss": 0.3493, + "step": 15819 + }, + { + "epoch": 0.8838729502472275, + "grad_norm": 0.43044090270996094, + "learning_rate": 0.0005600907664724339, + "loss": 0.4725, + "step": 15820 + }, + { + "epoch": 0.8839288208509093, + "grad_norm": 0.3775036633014679, + "learning_rate": 0.0005600627521290901, + "loss": 0.4028, + "step": 15821 + }, + { + "epoch": 0.8839846914545911, + "grad_norm": 0.5350784659385681, + "learning_rate": 0.0005600347377857463, + "loss": 0.5095, + "step": 15822 + }, + { + "epoch": 0.8840405620582731, + "grad_norm": 0.7977792024612427, + "learning_rate": 0.0005600067234424026, + "loss": 0.4778, + "step": 15823 + }, + { + "epoch": 0.8840964326619549, + "grad_norm": 1.3997169733047485, + "learning_rate": 0.0005599787090990587, + "loss": 0.7472, + "step": 15824 + }, + { + "epoch": 0.8841523032656368, + "grad_norm": 0.5872367024421692, + "learning_rate": 0.0005599506947557149, + "loss": 0.4624, + "step": 15825 + }, + { + "epoch": 0.8842081738693186, + "grad_norm": 0.6319361329078674, + "learning_rate": 0.0005599226804123711, + "loss": 0.5088, + "step": 15826 + }, + { + "epoch": 0.8842640444730006, + "grad_norm": 0.4598875641822815, + "learning_rate": 0.0005598946660690274, + "loss": 0.4021, + "step": 15827 + }, + { + "epoch": 0.8843199150766824, + "grad_norm": 1.1631600856781006, + "learning_rate": 0.0005598666517256835, + "loss": 0.4959, + "step": 15828 + }, + { + "epoch": 0.8843757856803642, + "grad_norm": 0.6288260221481323, + "learning_rate": 0.0005598386373823397, + "loss": 0.4112, + "step": 15829 + }, + { + "epoch": 0.8844316562840462, + "grad_norm": 0.36311641335487366, + "learning_rate": 0.000559810623038996, + "loss": 0.3945, + "step": 15830 + }, + { + "epoch": 0.884487526887728, + "grad_norm": 0.5442600250244141, + "learning_rate": 0.0005597826086956522, + "loss": 0.4262, + "step": 15831 + }, + { + "epoch": 0.8845433974914099, + "grad_norm": 0.5525892972946167, + "learning_rate": 0.0005597545943523083, + "loss": 0.4622, + "step": 15832 + }, + { + "epoch": 0.8845992680950918, + "grad_norm": 0.5501536130905151, + "learning_rate": 0.0005597265800089646, + "loss": 0.4961, + "step": 15833 + }, + { + "epoch": 0.8846551386987737, + "grad_norm": 0.4826008975505829, + "learning_rate": 0.0005596985656656209, + "loss": 0.399, + "step": 15834 + }, + { + "epoch": 0.8847110093024555, + "grad_norm": 1.0044835805892944, + "learning_rate": 0.0005596705513222771, + "loss": 0.3566, + "step": 15835 + }, + { + "epoch": 0.8847668799061373, + "grad_norm": 0.5316320657730103, + "learning_rate": 0.0005596425369789332, + "loss": 0.5865, + "step": 15836 + }, + { + "epoch": 0.8848227505098193, + "grad_norm": 0.5190406441688538, + "learning_rate": 0.0005596145226355895, + "loss": 0.4664, + "step": 15837 + }, + { + "epoch": 0.8848786211135011, + "grad_norm": 0.4167915880680084, + "learning_rate": 0.0005595865082922457, + "loss": 0.3654, + "step": 15838 + }, + { + "epoch": 0.884934491717183, + "grad_norm": 0.45328474044799805, + "learning_rate": 0.0005595584939489019, + "loss": 0.4777, + "step": 15839 + }, + { + "epoch": 0.8849903623208649, + "grad_norm": 0.8611881732940674, + "learning_rate": 0.000559530479605558, + "loss": 0.4581, + "step": 15840 + }, + { + "epoch": 0.8850462329245468, + "grad_norm": 0.3742764890193939, + "learning_rate": 0.0005595024652622143, + "loss": 0.3987, + "step": 15841 + }, + { + "epoch": 0.8851021035282286, + "grad_norm": 0.3999614417552948, + "learning_rate": 0.0005594744509188705, + "loss": 0.3964, + "step": 15842 + }, + { + "epoch": 0.8851579741319104, + "grad_norm": 0.42785415053367615, + "learning_rate": 0.0005594464365755266, + "loss": 0.5146, + "step": 15843 + }, + { + "epoch": 0.8852138447355924, + "grad_norm": 0.6375609636306763, + "learning_rate": 0.0005594184222321828, + "loss": 0.3949, + "step": 15844 + }, + { + "epoch": 0.8852697153392742, + "grad_norm": 0.5588764548301697, + "learning_rate": 0.0005593904078888391, + "loss": 0.5385, + "step": 15845 + }, + { + "epoch": 0.8853255859429561, + "grad_norm": 0.7144248485565186, + "learning_rate": 0.0005593623935454953, + "loss": 0.3437, + "step": 15846 + }, + { + "epoch": 0.885381456546638, + "grad_norm": 0.40872931480407715, + "learning_rate": 0.0005593343792021514, + "loss": 0.365, + "step": 15847 + }, + { + "epoch": 0.8854373271503199, + "grad_norm": 0.8467289805412292, + "learning_rate": 0.0005593063648588077, + "loss": 0.405, + "step": 15848 + }, + { + "epoch": 0.8854931977540017, + "grad_norm": 0.41520071029663086, + "learning_rate": 0.000559278350515464, + "loss": 0.534, + "step": 15849 + }, + { + "epoch": 0.8855490683576837, + "grad_norm": 0.3426154553890228, + "learning_rate": 0.0005592503361721202, + "loss": 0.3676, + "step": 15850 + }, + { + "epoch": 0.8856049389613655, + "grad_norm": 0.4585864543914795, + "learning_rate": 0.0005592223218287763, + "loss": 0.5038, + "step": 15851 + }, + { + "epoch": 0.8856608095650473, + "grad_norm": 0.4620489478111267, + "learning_rate": 0.0005591943074854326, + "loss": 0.4361, + "step": 15852 + }, + { + "epoch": 0.8857166801687292, + "grad_norm": 0.3981071412563324, + "learning_rate": 0.0005591662931420888, + "loss": 0.6403, + "step": 15853 + }, + { + "epoch": 0.8857725507724111, + "grad_norm": 0.4231365919113159, + "learning_rate": 0.000559138278798745, + "loss": 0.4713, + "step": 15854 + }, + { + "epoch": 0.885828421376093, + "grad_norm": 0.8807493448257446, + "learning_rate": 0.0005591102644554012, + "loss": 0.5037, + "step": 15855 + }, + { + "epoch": 0.8858842919797748, + "grad_norm": 0.4725566506385803, + "learning_rate": 0.0005590822501120574, + "loss": 0.4329, + "step": 15856 + }, + { + "epoch": 0.8859401625834568, + "grad_norm": 0.37803059816360474, + "learning_rate": 0.0005590542357687136, + "loss": 0.4396, + "step": 15857 + }, + { + "epoch": 0.8859960331871386, + "grad_norm": 0.3818085193634033, + "learning_rate": 0.0005590262214253698, + "loss": 0.4713, + "step": 15858 + }, + { + "epoch": 0.8860519037908204, + "grad_norm": 0.4499889016151428, + "learning_rate": 0.000558998207082026, + "loss": 0.4663, + "step": 15859 + }, + { + "epoch": 0.8861077743945023, + "grad_norm": 0.41598284244537354, + "learning_rate": 0.0005589701927386822, + "loss": 0.3549, + "step": 15860 + }, + { + "epoch": 0.8861636449981842, + "grad_norm": 0.6574746966362, + "learning_rate": 0.0005589421783953384, + "loss": 0.4608, + "step": 15861 + }, + { + "epoch": 0.8862195156018661, + "grad_norm": 0.4261842966079712, + "learning_rate": 0.0005589141640519946, + "loss": 0.424, + "step": 15862 + }, + { + "epoch": 0.8862753862055479, + "grad_norm": 0.4157867431640625, + "learning_rate": 0.0005588861497086508, + "loss": 0.4328, + "step": 15863 + }, + { + "epoch": 0.8863312568092299, + "grad_norm": 0.34732869267463684, + "learning_rate": 0.000558858135365307, + "loss": 0.4745, + "step": 15864 + }, + { + "epoch": 0.8863871274129117, + "grad_norm": 0.4169768989086151, + "learning_rate": 0.0005588301210219632, + "loss": 0.4614, + "step": 15865 + }, + { + "epoch": 0.8864429980165935, + "grad_norm": 0.48797738552093506, + "learning_rate": 0.0005588021066786195, + "loss": 0.4391, + "step": 15866 + }, + { + "epoch": 0.8864988686202755, + "grad_norm": 0.8445791602134705, + "learning_rate": 0.0005587740923352757, + "loss": 0.4407, + "step": 15867 + }, + { + "epoch": 0.8865547392239573, + "grad_norm": 0.5347010493278503, + "learning_rate": 0.0005587460779919319, + "loss": 0.4875, + "step": 15868 + }, + { + "epoch": 0.8866106098276392, + "grad_norm": 0.7978647351264954, + "learning_rate": 0.0005587180636485882, + "loss": 0.6123, + "step": 15869 + }, + { + "epoch": 0.886666480431321, + "grad_norm": 0.5048972964286804, + "learning_rate": 0.0005586900493052443, + "loss": 0.4697, + "step": 15870 + }, + { + "epoch": 0.886722351035003, + "grad_norm": 0.5395758748054504, + "learning_rate": 0.0005586620349619005, + "loss": 0.4577, + "step": 15871 + }, + { + "epoch": 0.8867782216386848, + "grad_norm": 0.9362608194351196, + "learning_rate": 0.0005586340206185567, + "loss": 0.4717, + "step": 15872 + }, + { + "epoch": 0.8868340922423666, + "grad_norm": 0.7155297994613647, + "learning_rate": 0.000558606006275213, + "loss": 0.4774, + "step": 15873 + }, + { + "epoch": 0.8868899628460486, + "grad_norm": 0.7609071731567383, + "learning_rate": 0.0005585779919318691, + "loss": 0.4104, + "step": 15874 + }, + { + "epoch": 0.8869458334497304, + "grad_norm": 0.5101801156997681, + "learning_rate": 0.0005585499775885253, + "loss": 0.5159, + "step": 15875 + }, + { + "epoch": 0.8870017040534123, + "grad_norm": 0.4746586084365845, + "learning_rate": 0.0005585219632451816, + "loss": 0.3578, + "step": 15876 + }, + { + "epoch": 0.8870575746570941, + "grad_norm": 0.6845768690109253, + "learning_rate": 0.0005584939489018378, + "loss": 0.5057, + "step": 15877 + }, + { + "epoch": 0.8871134452607761, + "grad_norm": 1.1894069910049438, + "learning_rate": 0.0005584659345584939, + "loss": 0.4048, + "step": 15878 + }, + { + "epoch": 0.8871693158644579, + "grad_norm": 0.440525084733963, + "learning_rate": 0.0005584379202151501, + "loss": 0.3484, + "step": 15879 + }, + { + "epoch": 0.8872251864681397, + "grad_norm": 0.4959937334060669, + "learning_rate": 0.0005584099058718064, + "loss": 0.4388, + "step": 15880 + }, + { + "epoch": 0.8872810570718217, + "grad_norm": 0.7620554566383362, + "learning_rate": 0.0005583818915284626, + "loss": 0.4944, + "step": 15881 + }, + { + "epoch": 0.8873369276755035, + "grad_norm": 0.8696513772010803, + "learning_rate": 0.0005583538771851187, + "loss": 0.4733, + "step": 15882 + }, + { + "epoch": 0.8873927982791854, + "grad_norm": 0.40308380126953125, + "learning_rate": 0.000558325862841775, + "loss": 0.4207, + "step": 15883 + }, + { + "epoch": 0.8874486688828673, + "grad_norm": 0.5704174637794495, + "learning_rate": 0.0005582978484984313, + "loss": 0.4053, + "step": 15884 + }, + { + "epoch": 0.8875045394865492, + "grad_norm": 0.45627331733703613, + "learning_rate": 0.0005582698341550874, + "loss": 0.3207, + "step": 15885 + }, + { + "epoch": 0.887560410090231, + "grad_norm": 0.4022524654865265, + "learning_rate": 0.0005582418198117436, + "loss": 0.4567, + "step": 15886 + }, + { + "epoch": 0.8876162806939129, + "grad_norm": 0.4518255889415741, + "learning_rate": 0.0005582138054683999, + "loss": 0.405, + "step": 15887 + }, + { + "epoch": 0.8876721512975948, + "grad_norm": 1.591443419456482, + "learning_rate": 0.0005581857911250561, + "loss": 0.3799, + "step": 15888 + }, + { + "epoch": 0.8877280219012766, + "grad_norm": 0.541336178779602, + "learning_rate": 0.0005581577767817122, + "loss": 0.3861, + "step": 15889 + }, + { + "epoch": 0.8877838925049585, + "grad_norm": 0.4092230796813965, + "learning_rate": 0.0005581297624383684, + "loss": 0.3888, + "step": 15890 + }, + { + "epoch": 0.8878397631086404, + "grad_norm": 0.5519083738327026, + "learning_rate": 0.0005581017480950247, + "loss": 0.4707, + "step": 15891 + }, + { + "epoch": 0.8878956337123223, + "grad_norm": 0.3248414099216461, + "learning_rate": 0.0005580737337516809, + "loss": 0.4251, + "step": 15892 + }, + { + "epoch": 0.8879515043160041, + "grad_norm": 0.7667018175125122, + "learning_rate": 0.000558045719408337, + "loss": 0.3932, + "step": 15893 + }, + { + "epoch": 0.888007374919686, + "grad_norm": 0.634650707244873, + "learning_rate": 0.0005580177050649933, + "loss": 0.3745, + "step": 15894 + }, + { + "epoch": 0.8880632455233679, + "grad_norm": 0.37328311800956726, + "learning_rate": 0.0005579896907216495, + "loss": 0.3736, + "step": 15895 + }, + { + "epoch": 0.8881191161270497, + "grad_norm": 0.5672295689582825, + "learning_rate": 0.0005579616763783057, + "loss": 0.5943, + "step": 15896 + }, + { + "epoch": 0.8881749867307316, + "grad_norm": 0.5581481456756592, + "learning_rate": 0.0005579336620349618, + "loss": 0.3659, + "step": 15897 + }, + { + "epoch": 0.8882308573344135, + "grad_norm": 0.5046303272247314, + "learning_rate": 0.0005579056476916181, + "loss": 0.4779, + "step": 15898 + }, + { + "epoch": 0.8882867279380954, + "grad_norm": 0.7123607397079468, + "learning_rate": 0.0005578776333482744, + "loss": 0.3303, + "step": 15899 + }, + { + "epoch": 0.8883425985417772, + "grad_norm": 0.5555926561355591, + "learning_rate": 0.0005578496190049306, + "loss": 0.4248, + "step": 15900 + }, + { + "epoch": 0.8883984691454592, + "grad_norm": 0.9561766982078552, + "learning_rate": 0.0005578216046615868, + "loss": 0.4143, + "step": 15901 + }, + { + "epoch": 0.888454339749141, + "grad_norm": 1.765165090560913, + "learning_rate": 0.000557793590318243, + "loss": 0.5429, + "step": 15902 + }, + { + "epoch": 0.8885102103528228, + "grad_norm": 0.405038058757782, + "learning_rate": 0.0005577655759748992, + "loss": 0.3586, + "step": 15903 + }, + { + "epoch": 0.8885660809565047, + "grad_norm": 0.5786087512969971, + "learning_rate": 0.0005577375616315553, + "loss": 0.4367, + "step": 15904 + }, + { + "epoch": 0.8886219515601866, + "grad_norm": 0.36182621121406555, + "learning_rate": 0.0005577095472882116, + "loss": 0.3604, + "step": 15905 + }, + { + "epoch": 0.8886778221638685, + "grad_norm": 0.6286565661430359, + "learning_rate": 0.0005576815329448678, + "loss": 0.4692, + "step": 15906 + }, + { + "epoch": 0.8887336927675503, + "grad_norm": 0.4116819500923157, + "learning_rate": 0.000557653518601524, + "loss": 0.4048, + "step": 15907 + }, + { + "epoch": 0.8887895633712323, + "grad_norm": 0.47086167335510254, + "learning_rate": 0.0005576255042581802, + "loss": 0.3871, + "step": 15908 + }, + { + "epoch": 0.8888454339749141, + "grad_norm": 0.5287370085716248, + "learning_rate": 0.0005575974899148364, + "loss": 0.4735, + "step": 15909 + }, + { + "epoch": 0.888901304578596, + "grad_norm": 0.4539856016635895, + "learning_rate": 0.0005575694755714926, + "loss": 0.4234, + "step": 15910 + }, + { + "epoch": 0.8889571751822778, + "grad_norm": 0.6468028426170349, + "learning_rate": 0.0005575414612281488, + "loss": 0.5094, + "step": 15911 + }, + { + "epoch": 0.8890130457859597, + "grad_norm": 0.7729212045669556, + "learning_rate": 0.000557513446884805, + "loss": 0.5442, + "step": 15912 + }, + { + "epoch": 0.8890689163896416, + "grad_norm": 3.630685806274414, + "learning_rate": 0.0005574854325414612, + "loss": 0.4162, + "step": 15913 + }, + { + "epoch": 0.8891247869933234, + "grad_norm": 0.479492723941803, + "learning_rate": 0.0005574574181981174, + "loss": 0.3847, + "step": 15914 + }, + { + "epoch": 0.8891806575970054, + "grad_norm": 2.0701777935028076, + "learning_rate": 0.0005574294038547736, + "loss": 0.4437, + "step": 15915 + }, + { + "epoch": 0.8892365282006872, + "grad_norm": 0.706827700138092, + "learning_rate": 0.0005574013895114298, + "loss": 0.3342, + "step": 15916 + }, + { + "epoch": 0.889292398804369, + "grad_norm": 0.39747563004493713, + "learning_rate": 0.0005573733751680861, + "loss": 0.3427, + "step": 15917 + }, + { + "epoch": 0.889348269408051, + "grad_norm": 0.6735577583312988, + "learning_rate": 0.0005573453608247423, + "loss": 0.4561, + "step": 15918 + }, + { + "epoch": 0.8894041400117328, + "grad_norm": 0.5385830402374268, + "learning_rate": 0.0005573173464813986, + "loss": 0.447, + "step": 15919 + }, + { + "epoch": 0.8894600106154147, + "grad_norm": 2.9998953342437744, + "learning_rate": 0.0005572893321380547, + "loss": 0.4097, + "step": 15920 + }, + { + "epoch": 0.8895158812190965, + "grad_norm": 0.6136702299118042, + "learning_rate": 0.0005572613177947109, + "loss": 0.3692, + "step": 15921 + }, + { + "epoch": 0.8895717518227785, + "grad_norm": 0.4277360737323761, + "learning_rate": 0.0005572333034513671, + "loss": 0.4537, + "step": 15922 + }, + { + "epoch": 0.8896276224264603, + "grad_norm": 0.5654351711273193, + "learning_rate": 0.0005572052891080234, + "loss": 0.4009, + "step": 15923 + }, + { + "epoch": 0.8896834930301422, + "grad_norm": 0.3876180350780487, + "learning_rate": 0.0005571772747646795, + "loss": 0.4961, + "step": 15924 + }, + { + "epoch": 0.8897393636338241, + "grad_norm": 0.4792751967906952, + "learning_rate": 0.0005571492604213357, + "loss": 0.4588, + "step": 15925 + }, + { + "epoch": 0.889795234237506, + "grad_norm": 0.3404037356376648, + "learning_rate": 0.000557121246077992, + "loss": 0.3532, + "step": 15926 + }, + { + "epoch": 0.8898511048411878, + "grad_norm": 0.6377606987953186, + "learning_rate": 0.0005570932317346481, + "loss": 0.4787, + "step": 15927 + }, + { + "epoch": 0.8899069754448696, + "grad_norm": 0.5192388296127319, + "learning_rate": 0.0005570652173913043, + "loss": 0.4221, + "step": 15928 + }, + { + "epoch": 0.8899628460485516, + "grad_norm": 0.8972006440162659, + "learning_rate": 0.0005570372030479605, + "loss": 0.6456, + "step": 15929 + }, + { + "epoch": 0.8900187166522334, + "grad_norm": 0.9363482594490051, + "learning_rate": 0.0005570091887046168, + "loss": 0.4195, + "step": 15930 + }, + { + "epoch": 0.8900745872559153, + "grad_norm": 0.5092593431472778, + "learning_rate": 0.0005569811743612729, + "loss": 0.4509, + "step": 15931 + }, + { + "epoch": 0.8901304578595972, + "grad_norm": 0.5068438053131104, + "learning_rate": 0.0005569531600179291, + "loss": 0.4101, + "step": 15932 + }, + { + "epoch": 0.890186328463279, + "grad_norm": 0.63824862241745, + "learning_rate": 0.0005569251456745855, + "loss": 0.4457, + "step": 15933 + }, + { + "epoch": 0.8902421990669609, + "grad_norm": 0.6485647559165955, + "learning_rate": 0.0005568971313312417, + "loss": 0.487, + "step": 15934 + }, + { + "epoch": 0.8902980696706428, + "grad_norm": 0.5763607621192932, + "learning_rate": 0.0005568691169878978, + "loss": 0.4753, + "step": 15935 + }, + { + "epoch": 0.8903539402743247, + "grad_norm": 0.5541970729827881, + "learning_rate": 0.000556841102644554, + "loss": 0.4385, + "step": 15936 + }, + { + "epoch": 0.8904098108780065, + "grad_norm": 0.6730737686157227, + "learning_rate": 0.0005568130883012103, + "loss": 0.4019, + "step": 15937 + }, + { + "epoch": 0.8904656814816884, + "grad_norm": 0.5110547542572021, + "learning_rate": 0.0005567850739578665, + "loss": 0.4405, + "step": 15938 + }, + { + "epoch": 0.8905215520853703, + "grad_norm": 0.4100966155529022, + "learning_rate": 0.0005567570596145226, + "loss": 0.3792, + "step": 15939 + }, + { + "epoch": 0.8905774226890522, + "grad_norm": 1.8250812292099, + "learning_rate": 0.0005567290452711789, + "loss": 0.7083, + "step": 15940 + }, + { + "epoch": 0.890633293292734, + "grad_norm": 0.5934962034225464, + "learning_rate": 0.0005567010309278351, + "loss": 0.5287, + "step": 15941 + }, + { + "epoch": 0.890689163896416, + "grad_norm": 0.6651060581207275, + "learning_rate": 0.0005566730165844913, + "loss": 0.5269, + "step": 15942 + }, + { + "epoch": 0.8907450345000978, + "grad_norm": 0.6665958762168884, + "learning_rate": 0.0005566450022411474, + "loss": 0.5335, + "step": 15943 + }, + { + "epoch": 0.8908009051037796, + "grad_norm": 0.3267519772052765, + "learning_rate": 0.0005566169878978037, + "loss": 0.4062, + "step": 15944 + }, + { + "epoch": 0.8908567757074615, + "grad_norm": 0.5238247513771057, + "learning_rate": 0.0005565889735544599, + "loss": 0.496, + "step": 15945 + }, + { + "epoch": 0.8909126463111434, + "grad_norm": 0.630884051322937, + "learning_rate": 0.0005565609592111161, + "loss": 0.4605, + "step": 15946 + }, + { + "epoch": 0.8909685169148253, + "grad_norm": 0.8428598642349243, + "learning_rate": 0.0005565329448677722, + "loss": 0.343, + "step": 15947 + }, + { + "epoch": 0.8910243875185071, + "grad_norm": 0.6964418292045593, + "learning_rate": 0.0005565049305244285, + "loss": 0.3528, + "step": 15948 + }, + { + "epoch": 0.891080258122189, + "grad_norm": 0.5135846138000488, + "learning_rate": 0.0005564769161810847, + "loss": 0.4789, + "step": 15949 + }, + { + "epoch": 0.8911361287258709, + "grad_norm": 0.40763843059539795, + "learning_rate": 0.0005564489018377408, + "loss": 0.3118, + "step": 15950 + }, + { + "epoch": 0.8911919993295527, + "grad_norm": 1.2600289583206177, + "learning_rate": 0.0005564208874943972, + "loss": 0.6446, + "step": 15951 + }, + { + "epoch": 0.8912478699332347, + "grad_norm": 0.33124032616615295, + "learning_rate": 0.0005563928731510534, + "loss": 0.4276, + "step": 15952 + }, + { + "epoch": 0.8913037405369165, + "grad_norm": 0.5601537823677063, + "learning_rate": 0.0005563648588077096, + "loss": 0.4277, + "step": 15953 + }, + { + "epoch": 0.8913596111405984, + "grad_norm": 0.4205104112625122, + "learning_rate": 0.0005563368444643657, + "loss": 0.378, + "step": 15954 + }, + { + "epoch": 0.8914154817442802, + "grad_norm": 0.4327593445777893, + "learning_rate": 0.000556308830121022, + "loss": 0.3056, + "step": 15955 + }, + { + "epoch": 0.8914713523479622, + "grad_norm": 0.4373265206813812, + "learning_rate": 0.0005562808157776782, + "loss": 0.48, + "step": 15956 + }, + { + "epoch": 0.891527222951644, + "grad_norm": 0.4948159158229828, + "learning_rate": 0.0005562528014343344, + "loss": 0.4817, + "step": 15957 + }, + { + "epoch": 0.8915830935553258, + "grad_norm": 0.6465309858322144, + "learning_rate": 0.0005562247870909906, + "loss": 0.4469, + "step": 15958 + }, + { + "epoch": 0.8916389641590078, + "grad_norm": 0.4033644199371338, + "learning_rate": 0.0005561967727476468, + "loss": 0.3923, + "step": 15959 + }, + { + "epoch": 0.8916948347626896, + "grad_norm": 0.43542173504829407, + "learning_rate": 0.000556168758404303, + "loss": 0.3784, + "step": 15960 + }, + { + "epoch": 0.8917507053663715, + "grad_norm": 0.5116628408432007, + "learning_rate": 0.0005561407440609592, + "loss": 0.4993, + "step": 15961 + }, + { + "epoch": 0.8918065759700533, + "grad_norm": 0.4240174889564514, + "learning_rate": 0.0005561127297176154, + "loss": 0.6476, + "step": 15962 + }, + { + "epoch": 0.8918624465737353, + "grad_norm": 0.42289283871650696, + "learning_rate": 0.0005560847153742716, + "loss": 0.3752, + "step": 15963 + }, + { + "epoch": 0.8919183171774171, + "grad_norm": 0.4227098822593689, + "learning_rate": 0.0005560567010309278, + "loss": 0.4678, + "step": 15964 + }, + { + "epoch": 0.8919741877810989, + "grad_norm": 0.5922468900680542, + "learning_rate": 0.0005560286866875842, + "loss": 0.3984, + "step": 15965 + }, + { + "epoch": 0.8920300583847809, + "grad_norm": 1.040076732635498, + "learning_rate": 0.0005560006723442402, + "loss": 0.4084, + "step": 15966 + }, + { + "epoch": 0.8920859289884627, + "grad_norm": 0.5324835181236267, + "learning_rate": 0.0005559726580008965, + "loss": 0.5237, + "step": 15967 + }, + { + "epoch": 0.8921417995921446, + "grad_norm": 0.9070014357566833, + "learning_rate": 0.0005559446436575527, + "loss": 0.3478, + "step": 15968 + }, + { + "epoch": 0.8921976701958265, + "grad_norm": 0.5666398406028748, + "learning_rate": 0.0005559166293142089, + "loss": 0.4306, + "step": 15969 + }, + { + "epoch": 0.8922535407995084, + "grad_norm": 0.411516398191452, + "learning_rate": 0.0005558886149708651, + "loss": 0.4198, + "step": 15970 + }, + { + "epoch": 0.8923094114031902, + "grad_norm": 0.49901142716407776, + "learning_rate": 0.0005558606006275213, + "loss": 0.3989, + "step": 15971 + }, + { + "epoch": 0.892365282006872, + "grad_norm": 0.386125385761261, + "learning_rate": 0.0005558325862841776, + "loss": 0.3569, + "step": 15972 + }, + { + "epoch": 0.892421152610554, + "grad_norm": 0.4053994417190552, + "learning_rate": 0.0005558045719408337, + "loss": 0.4803, + "step": 15973 + }, + { + "epoch": 0.8924770232142358, + "grad_norm": 0.42667752504348755, + "learning_rate": 0.0005557765575974899, + "loss": 0.3792, + "step": 15974 + }, + { + "epoch": 0.8925328938179177, + "grad_norm": 0.39611661434173584, + "learning_rate": 0.0005557485432541461, + "loss": 0.5073, + "step": 15975 + }, + { + "epoch": 0.8925887644215996, + "grad_norm": 0.4864833950996399, + "learning_rate": 0.0005557205289108024, + "loss": 0.3719, + "step": 15976 + }, + { + "epoch": 0.8926446350252815, + "grad_norm": 0.83418208360672, + "learning_rate": 0.0005556925145674585, + "loss": 0.4524, + "step": 15977 + }, + { + "epoch": 0.8927005056289633, + "grad_norm": 0.40378817915916443, + "learning_rate": 0.0005556645002241147, + "loss": 0.4034, + "step": 15978 + }, + { + "epoch": 0.8927563762326451, + "grad_norm": 0.4115784764289856, + "learning_rate": 0.000555636485880771, + "loss": 0.3511, + "step": 15979 + }, + { + "epoch": 0.8928122468363271, + "grad_norm": 0.5940119624137878, + "learning_rate": 0.0005556084715374272, + "loss": 0.5257, + "step": 15980 + }, + { + "epoch": 0.8928681174400089, + "grad_norm": 0.5356873869895935, + "learning_rate": 0.0005555804571940833, + "loss": 0.5362, + "step": 15981 + }, + { + "epoch": 0.8929239880436908, + "grad_norm": 0.59042888879776, + "learning_rate": 0.0005555524428507395, + "loss": 0.4281, + "step": 15982 + }, + { + "epoch": 0.8929798586473727, + "grad_norm": 0.5547558069229126, + "learning_rate": 0.0005555244285073959, + "loss": 0.5181, + "step": 15983 + }, + { + "epoch": 0.8930357292510546, + "grad_norm": 1.1311691999435425, + "learning_rate": 0.0005554964141640521, + "loss": 0.3659, + "step": 15984 + }, + { + "epoch": 0.8930915998547364, + "grad_norm": 1.021254301071167, + "learning_rate": 0.0005554683998207082, + "loss": 0.5458, + "step": 15985 + }, + { + "epoch": 0.8931474704584184, + "grad_norm": 0.7002360224723816, + "learning_rate": 0.0005554403854773644, + "loss": 0.4115, + "step": 15986 + }, + { + "epoch": 0.8932033410621002, + "grad_norm": 1.945334792137146, + "learning_rate": 0.0005554123711340207, + "loss": 0.3496, + "step": 15987 + }, + { + "epoch": 0.893259211665782, + "grad_norm": 1.25892972946167, + "learning_rate": 0.0005553843567906769, + "loss": 0.4467, + "step": 15988 + }, + { + "epoch": 0.8933150822694639, + "grad_norm": 0.44322386384010315, + "learning_rate": 0.000555356342447333, + "loss": 0.4346, + "step": 15989 + }, + { + "epoch": 0.8933709528731458, + "grad_norm": 0.4617787003517151, + "learning_rate": 0.0005553283281039893, + "loss": 0.5286, + "step": 15990 + }, + { + "epoch": 0.8934268234768277, + "grad_norm": 0.3880421817302704, + "learning_rate": 0.0005553003137606455, + "loss": 0.4543, + "step": 15991 + }, + { + "epoch": 0.8934826940805095, + "grad_norm": 0.4398204982280731, + "learning_rate": 0.0005552722994173016, + "loss": 0.4273, + "step": 15992 + }, + { + "epoch": 0.8935385646841915, + "grad_norm": 0.745162308216095, + "learning_rate": 0.0005552442850739578, + "loss": 0.5309, + "step": 15993 + }, + { + "epoch": 0.8935944352878733, + "grad_norm": 0.525987982749939, + "learning_rate": 0.0005552162707306141, + "loss": 0.4403, + "step": 15994 + }, + { + "epoch": 0.8936503058915551, + "grad_norm": 0.8355324864387512, + "learning_rate": 0.0005551882563872703, + "loss": 0.3912, + "step": 15995 + }, + { + "epoch": 0.893706176495237, + "grad_norm": 0.48753052949905396, + "learning_rate": 0.0005551602420439264, + "loss": 0.4849, + "step": 15996 + }, + { + "epoch": 0.8937620470989189, + "grad_norm": 0.49426284432411194, + "learning_rate": 0.0005551322277005827, + "loss": 0.49, + "step": 15997 + }, + { + "epoch": 0.8938179177026008, + "grad_norm": 0.49013814330101013, + "learning_rate": 0.0005551042133572389, + "loss": 0.4116, + "step": 15998 + }, + { + "epoch": 0.8938737883062826, + "grad_norm": 0.4772787392139435, + "learning_rate": 0.0005550761990138951, + "loss": 0.4183, + "step": 15999 + }, + { + "epoch": 0.8939296589099646, + "grad_norm": 0.5681507587432861, + "learning_rate": 0.0005550481846705512, + "loss": 0.3974, + "step": 16000 + }, + { + "epoch": 0.8939296589099646, + "eval_cer": 0.0891623842100095, + "eval_loss": 0.33301815390586853, + "eval_runtime": 56.3151, + "eval_samples_per_second": 80.582, + "eval_steps_per_second": 5.043, + "eval_wer": 0.35451523700455295, + "step": 16000 + }, + { + "epoch": 0.8939855295136464, + "grad_norm": 0.4345226585865021, + "learning_rate": 0.0005550201703272076, + "loss": 0.4245, + "step": 16001 + }, + { + "epoch": 0.8940414001173282, + "grad_norm": 0.38833895325660706, + "learning_rate": 0.0005549921559838638, + "loss": 0.4423, + "step": 16002 + }, + { + "epoch": 0.8940972707210102, + "grad_norm": 0.47264644503593445, + "learning_rate": 0.00055496414164052, + "loss": 0.4151, + "step": 16003 + }, + { + "epoch": 0.894153141324692, + "grad_norm": 0.6787971258163452, + "learning_rate": 0.0005549361272971762, + "loss": 0.3255, + "step": 16004 + }, + { + "epoch": 0.8942090119283739, + "grad_norm": 0.6422802805900574, + "learning_rate": 0.0005549081129538324, + "loss": 0.593, + "step": 16005 + }, + { + "epoch": 0.8942648825320557, + "grad_norm": 0.44517356157302856, + "learning_rate": 0.0005548800986104886, + "loss": 0.3984, + "step": 16006 + }, + { + "epoch": 0.8943207531357377, + "grad_norm": 0.4397099018096924, + "learning_rate": 0.0005548520842671448, + "loss": 0.4349, + "step": 16007 + }, + { + "epoch": 0.8943766237394195, + "grad_norm": 1.9869226217269897, + "learning_rate": 0.000554824069923801, + "loss": 0.6805, + "step": 16008 + }, + { + "epoch": 0.8944324943431013, + "grad_norm": 0.3598397672176361, + "learning_rate": 0.0005547960555804572, + "loss": 0.4244, + "step": 16009 + }, + { + "epoch": 0.8944883649467833, + "grad_norm": 0.8175591826438904, + "learning_rate": 0.0005547680412371134, + "loss": 0.3804, + "step": 16010 + }, + { + "epoch": 0.8945442355504651, + "grad_norm": 0.4798179566860199, + "learning_rate": 0.0005547400268937696, + "loss": 0.4195, + "step": 16011 + }, + { + "epoch": 0.894600106154147, + "grad_norm": 0.9072717428207397, + "learning_rate": 0.0005547120125504258, + "loss": 0.5485, + "step": 16012 + }, + { + "epoch": 0.8946559767578288, + "grad_norm": 0.46193063259124756, + "learning_rate": 0.000554683998207082, + "loss": 0.2847, + "step": 16013 + }, + { + "epoch": 0.8947118473615108, + "grad_norm": 0.36403605341911316, + "learning_rate": 0.0005546559838637382, + "loss": 0.3376, + "step": 16014 + }, + { + "epoch": 0.8947677179651926, + "grad_norm": 0.6371790170669556, + "learning_rate": 0.0005546279695203944, + "loss": 0.4573, + "step": 16015 + }, + { + "epoch": 0.8948235885688745, + "grad_norm": 0.4804866313934326, + "learning_rate": 0.0005545999551770506, + "loss": 0.4958, + "step": 16016 + }, + { + "epoch": 0.8948794591725564, + "grad_norm": 0.5673686861991882, + "learning_rate": 0.0005545719408337069, + "loss": 0.3585, + "step": 16017 + }, + { + "epoch": 0.8949353297762382, + "grad_norm": 0.4398680627346039, + "learning_rate": 0.0005545439264903632, + "loss": 0.4605, + "step": 16018 + }, + { + "epoch": 0.8949912003799201, + "grad_norm": 0.37322598695755005, + "learning_rate": 0.0005545159121470193, + "loss": 0.389, + "step": 16019 + }, + { + "epoch": 0.895047070983602, + "grad_norm": 0.4629228115081787, + "learning_rate": 0.0005544878978036755, + "loss": 0.3398, + "step": 16020 + }, + { + "epoch": 0.8951029415872839, + "grad_norm": 0.5664471983909607, + "learning_rate": 0.0005544598834603317, + "loss": 0.3788, + "step": 16021 + }, + { + "epoch": 0.8951588121909657, + "grad_norm": 0.4005005359649658, + "learning_rate": 0.000554431869116988, + "loss": 0.3782, + "step": 16022 + }, + { + "epoch": 0.8952146827946476, + "grad_norm": 0.40311571955680847, + "learning_rate": 0.0005544038547736441, + "loss": 0.4987, + "step": 16023 + }, + { + "epoch": 0.8952705533983295, + "grad_norm": 0.404862642288208, + "learning_rate": 0.0005543758404303003, + "loss": 0.336, + "step": 16024 + }, + { + "epoch": 0.8953264240020113, + "grad_norm": 0.5880100131034851, + "learning_rate": 0.0005543478260869565, + "loss": 0.3799, + "step": 16025 + }, + { + "epoch": 0.8953822946056932, + "grad_norm": 0.6306424140930176, + "learning_rate": 0.0005543198117436128, + "loss": 0.5458, + "step": 16026 + }, + { + "epoch": 0.8954381652093751, + "grad_norm": 0.6363179683685303, + "learning_rate": 0.0005542917974002689, + "loss": 0.4474, + "step": 16027 + }, + { + "epoch": 0.895494035813057, + "grad_norm": 0.34831154346466064, + "learning_rate": 0.0005542637830569251, + "loss": 0.4064, + "step": 16028 + }, + { + "epoch": 0.8955499064167388, + "grad_norm": 0.6895803213119507, + "learning_rate": 0.0005542357687135814, + "loss": 0.4209, + "step": 16029 + }, + { + "epoch": 0.8956057770204207, + "grad_norm": 0.3487863838672638, + "learning_rate": 0.0005542077543702376, + "loss": 0.4686, + "step": 16030 + }, + { + "epoch": 0.8956616476241026, + "grad_norm": 0.6383845806121826, + "learning_rate": 0.0005541797400268937, + "loss": 0.6434, + "step": 16031 + }, + { + "epoch": 0.8957175182277844, + "grad_norm": 0.7170008420944214, + "learning_rate": 0.0005541517256835499, + "loss": 0.4028, + "step": 16032 + }, + { + "epoch": 0.8957733888314663, + "grad_norm": 0.5015184879302979, + "learning_rate": 0.0005541237113402063, + "loss": 0.4676, + "step": 16033 + }, + { + "epoch": 0.8958292594351482, + "grad_norm": 0.3632389307022095, + "learning_rate": 0.0005540956969968624, + "loss": 0.403, + "step": 16034 + }, + { + "epoch": 0.8958851300388301, + "grad_norm": 0.7040925025939941, + "learning_rate": 0.0005540676826535186, + "loss": 0.4925, + "step": 16035 + }, + { + "epoch": 0.8959410006425119, + "grad_norm": 1.0111677646636963, + "learning_rate": 0.0005540396683101749, + "loss": 0.4425, + "step": 16036 + }, + { + "epoch": 0.8959968712461938, + "grad_norm": 0.4213480055332184, + "learning_rate": 0.0005540116539668311, + "loss": 0.4588, + "step": 16037 + }, + { + "epoch": 0.8960527418498757, + "grad_norm": 0.479381263256073, + "learning_rate": 0.0005539836396234872, + "loss": 0.4309, + "step": 16038 + }, + { + "epoch": 0.8961086124535576, + "grad_norm": 0.6671374440193176, + "learning_rate": 0.0005539556252801434, + "loss": 0.3614, + "step": 16039 + }, + { + "epoch": 0.8961644830572394, + "grad_norm": 0.4740208685398102, + "learning_rate": 0.0005539276109367997, + "loss": 0.442, + "step": 16040 + }, + { + "epoch": 0.8962203536609213, + "grad_norm": 0.45326700806617737, + "learning_rate": 0.0005538995965934559, + "loss": 0.4405, + "step": 16041 + }, + { + "epoch": 0.8962762242646032, + "grad_norm": 0.47111448645591736, + "learning_rate": 0.000553871582250112, + "loss": 0.434, + "step": 16042 + }, + { + "epoch": 0.896332094868285, + "grad_norm": 0.44584140181541443, + "learning_rate": 0.0005538435679067683, + "loss": 0.4214, + "step": 16043 + }, + { + "epoch": 0.896387965471967, + "grad_norm": 0.409782350063324, + "learning_rate": 0.0005538155535634245, + "loss": 0.3416, + "step": 16044 + }, + { + "epoch": 0.8964438360756488, + "grad_norm": 0.6128621101379395, + "learning_rate": 0.0005537875392200807, + "loss": 0.5497, + "step": 16045 + }, + { + "epoch": 0.8964997066793307, + "grad_norm": 0.7723132371902466, + "learning_rate": 0.0005537595248767368, + "loss": 0.4237, + "step": 16046 + }, + { + "epoch": 0.8965555772830125, + "grad_norm": 1.176539659500122, + "learning_rate": 0.0005537315105333931, + "loss": 0.432, + "step": 16047 + }, + { + "epoch": 0.8966114478866944, + "grad_norm": 0.7906206846237183, + "learning_rate": 0.0005537034961900493, + "loss": 0.3549, + "step": 16048 + }, + { + "epoch": 0.8966673184903763, + "grad_norm": 0.5124810934066772, + "learning_rate": 0.0005536754818467055, + "loss": 0.4131, + "step": 16049 + }, + { + "epoch": 0.8967231890940581, + "grad_norm": 0.6526705622673035, + "learning_rate": 0.0005536474675033618, + "loss": 0.3667, + "step": 16050 + }, + { + "epoch": 0.8967790596977401, + "grad_norm": 8.774215698242188, + "learning_rate": 0.000553619453160018, + "loss": 0.4172, + "step": 16051 + }, + { + "epoch": 0.8968349303014219, + "grad_norm": 0.6174339056015015, + "learning_rate": 0.0005535914388166742, + "loss": 0.4447, + "step": 16052 + }, + { + "epoch": 0.8968908009051038, + "grad_norm": 0.4894152283668518, + "learning_rate": 0.0005535634244733303, + "loss": 0.4854, + "step": 16053 + }, + { + "epoch": 0.8969466715087856, + "grad_norm": 0.47129523754119873, + "learning_rate": 0.0005535354101299866, + "loss": 0.4473, + "step": 16054 + }, + { + "epoch": 0.8970025421124675, + "grad_norm": 0.4162922203540802, + "learning_rate": 0.0005535073957866428, + "loss": 0.3901, + "step": 16055 + }, + { + "epoch": 0.8970584127161494, + "grad_norm": 0.618709146976471, + "learning_rate": 0.000553479381443299, + "loss": 0.4843, + "step": 16056 + }, + { + "epoch": 0.8971142833198312, + "grad_norm": 0.7525842785835266, + "learning_rate": 0.0005534513670999551, + "loss": 0.4527, + "step": 16057 + }, + { + "epoch": 0.8971701539235132, + "grad_norm": 0.6800097227096558, + "learning_rate": 0.0005534233527566114, + "loss": 0.5491, + "step": 16058 + }, + { + "epoch": 0.897226024527195, + "grad_norm": 0.4999452233314514, + "learning_rate": 0.0005533953384132676, + "loss": 0.349, + "step": 16059 + }, + { + "epoch": 0.8972818951308769, + "grad_norm": 0.45355355739593506, + "learning_rate": 0.0005533673240699238, + "loss": 0.4819, + "step": 16060 + }, + { + "epoch": 0.8973377657345588, + "grad_norm": 0.5153102874755859, + "learning_rate": 0.00055333930972658, + "loss": 0.5302, + "step": 16061 + }, + { + "epoch": 0.8973936363382407, + "grad_norm": 0.7790489792823792, + "learning_rate": 0.0005533112953832362, + "loss": 0.5817, + "step": 16062 + }, + { + "epoch": 0.8974495069419225, + "grad_norm": 0.4886455535888672, + "learning_rate": 0.0005532832810398924, + "loss": 0.5061, + "step": 16063 + }, + { + "epoch": 0.8975053775456043, + "grad_norm": 1.2953146696090698, + "learning_rate": 0.0005532552666965486, + "loss": 0.4816, + "step": 16064 + }, + { + "epoch": 0.8975612481492863, + "grad_norm": 0.45061391592025757, + "learning_rate": 0.0005532272523532048, + "loss": 0.4257, + "step": 16065 + }, + { + "epoch": 0.8976171187529681, + "grad_norm": 0.9140583872795105, + "learning_rate": 0.000553199238009861, + "loss": 0.3822, + "step": 16066 + }, + { + "epoch": 0.89767298935665, + "grad_norm": 0.3869934380054474, + "learning_rate": 0.0005531712236665173, + "loss": 0.3249, + "step": 16067 + }, + { + "epoch": 0.8977288599603319, + "grad_norm": 0.5828748345375061, + "learning_rate": 0.0005531432093231736, + "loss": 0.5375, + "step": 16068 + }, + { + "epoch": 0.8977847305640138, + "grad_norm": 0.3185482621192932, + "learning_rate": 0.0005531151949798297, + "loss": 0.4287, + "step": 16069 + }, + { + "epoch": 0.8978406011676956, + "grad_norm": 0.7365001440048218, + "learning_rate": 0.0005530871806364859, + "loss": 0.4622, + "step": 16070 + }, + { + "epoch": 0.8978964717713774, + "grad_norm": 0.6330768465995789, + "learning_rate": 0.0005530591662931421, + "loss": 0.5156, + "step": 16071 + }, + { + "epoch": 0.8979523423750594, + "grad_norm": 0.37974584102630615, + "learning_rate": 0.0005530311519497984, + "loss": 0.3642, + "step": 16072 + }, + { + "epoch": 0.8980082129787412, + "grad_norm": 0.3453536331653595, + "learning_rate": 0.0005530031376064545, + "loss": 0.3567, + "step": 16073 + }, + { + "epoch": 0.8980640835824231, + "grad_norm": 3.8604114055633545, + "learning_rate": 0.0005529751232631107, + "loss": 0.4341, + "step": 16074 + }, + { + "epoch": 0.898119954186105, + "grad_norm": 3.1437699794769287, + "learning_rate": 0.000552947108919767, + "loss": 0.4234, + "step": 16075 + }, + { + "epoch": 0.8981758247897869, + "grad_norm": 0.7751939296722412, + "learning_rate": 0.0005529190945764231, + "loss": 0.3737, + "step": 16076 + }, + { + "epoch": 0.8982316953934687, + "grad_norm": 0.5779013633728027, + "learning_rate": 0.0005528910802330793, + "loss": 0.5094, + "step": 16077 + }, + { + "epoch": 0.8982875659971506, + "grad_norm": 0.6257683038711548, + "learning_rate": 0.0005528630658897355, + "loss": 0.4435, + "step": 16078 + }, + { + "epoch": 0.8983434366008325, + "grad_norm": 0.6337378025054932, + "learning_rate": 0.0005528350515463918, + "loss": 0.5224, + "step": 16079 + }, + { + "epoch": 0.8983993072045143, + "grad_norm": 0.5711128115653992, + "learning_rate": 0.0005528070372030479, + "loss": 0.5043, + "step": 16080 + }, + { + "epoch": 0.8984551778081962, + "grad_norm": 0.9932848215103149, + "learning_rate": 0.0005527790228597041, + "loss": 0.4121, + "step": 16081 + }, + { + "epoch": 0.8985110484118781, + "grad_norm": 0.4103388488292694, + "learning_rate": 0.0005527510085163604, + "loss": 0.5223, + "step": 16082 + }, + { + "epoch": 0.89856691901556, + "grad_norm": 0.658194363117218, + "learning_rate": 0.0005527229941730167, + "loss": 0.457, + "step": 16083 + }, + { + "epoch": 0.8986227896192418, + "grad_norm": 0.6118751764297485, + "learning_rate": 0.0005526949798296728, + "loss": 0.4001, + "step": 16084 + }, + { + "epoch": 0.8986786602229238, + "grad_norm": 0.7221964597702026, + "learning_rate": 0.000552666965486329, + "loss": 0.3892, + "step": 16085 + }, + { + "epoch": 0.8987345308266056, + "grad_norm": 0.8024883270263672, + "learning_rate": 0.0005526389511429853, + "loss": 0.502, + "step": 16086 + }, + { + "epoch": 0.8987904014302874, + "grad_norm": 0.5247349739074707, + "learning_rate": 0.0005526109367996415, + "loss": 0.4267, + "step": 16087 + }, + { + "epoch": 0.8988462720339693, + "grad_norm": 1.103219985961914, + "learning_rate": 0.0005525829224562976, + "loss": 0.5419, + "step": 16088 + }, + { + "epoch": 0.8989021426376512, + "grad_norm": 0.5103521943092346, + "learning_rate": 0.0005525549081129539, + "loss": 0.4844, + "step": 16089 + }, + { + "epoch": 0.8989580132413331, + "grad_norm": 0.5734323859214783, + "learning_rate": 0.0005525268937696101, + "loss": 0.4511, + "step": 16090 + }, + { + "epoch": 0.8990138838450149, + "grad_norm": 0.4384804964065552, + "learning_rate": 0.0005524988794262663, + "loss": 0.394, + "step": 16091 + }, + { + "epoch": 0.8990697544486969, + "grad_norm": 0.7842878103256226, + "learning_rate": 0.0005524708650829224, + "loss": 0.6133, + "step": 16092 + }, + { + "epoch": 0.8991256250523787, + "grad_norm": 0.3885398805141449, + "learning_rate": 0.0005524428507395787, + "loss": 0.3813, + "step": 16093 + }, + { + "epoch": 0.8991814956560605, + "grad_norm": 0.48152807354927063, + "learning_rate": 0.0005524148363962349, + "loss": 0.5125, + "step": 16094 + }, + { + "epoch": 0.8992373662597425, + "grad_norm": 0.4013349115848541, + "learning_rate": 0.0005523868220528911, + "loss": 0.4935, + "step": 16095 + }, + { + "epoch": 0.8992932368634243, + "grad_norm": 0.5783447623252869, + "learning_rate": 0.0005523588077095472, + "loss": 0.4123, + "step": 16096 + }, + { + "epoch": 0.8993491074671062, + "grad_norm": 0.49301469326019287, + "learning_rate": 0.0005523307933662035, + "loss": 0.4331, + "step": 16097 + }, + { + "epoch": 0.899404978070788, + "grad_norm": 1.402397632598877, + "learning_rate": 0.0005523027790228597, + "loss": 0.4306, + "step": 16098 + }, + { + "epoch": 0.89946084867447, + "grad_norm": 0.5085805058479309, + "learning_rate": 0.0005522747646795158, + "loss": 0.4133, + "step": 16099 + }, + { + "epoch": 0.8995167192781518, + "grad_norm": 0.6873524188995361, + "learning_rate": 0.0005522467503361722, + "loss": 0.4453, + "step": 16100 + }, + { + "epoch": 0.8995725898818336, + "grad_norm": 0.6222655177116394, + "learning_rate": 0.0005522187359928284, + "loss": 0.4918, + "step": 16101 + }, + { + "epoch": 0.8996284604855156, + "grad_norm": 0.3842146694660187, + "learning_rate": 0.0005521907216494846, + "loss": 0.3705, + "step": 16102 + }, + { + "epoch": 0.8996843310891974, + "grad_norm": 0.44285205006599426, + "learning_rate": 0.0005521627073061407, + "loss": 0.3433, + "step": 16103 + }, + { + "epoch": 0.8997402016928793, + "grad_norm": 0.46175000071525574, + "learning_rate": 0.000552134692962797, + "loss": 0.4833, + "step": 16104 + }, + { + "epoch": 0.8997960722965611, + "grad_norm": 0.704734742641449, + "learning_rate": 0.0005521066786194532, + "loss": 0.4879, + "step": 16105 + }, + { + "epoch": 0.8998519429002431, + "grad_norm": 0.6319592595100403, + "learning_rate": 0.0005520786642761094, + "loss": 0.4642, + "step": 16106 + }, + { + "epoch": 0.8999078135039249, + "grad_norm": 0.7551725506782532, + "learning_rate": 0.0005520506499327656, + "loss": 0.4654, + "step": 16107 + }, + { + "epoch": 0.8999636841076067, + "grad_norm": 0.48179879784584045, + "learning_rate": 0.0005520226355894218, + "loss": 0.4104, + "step": 16108 + }, + { + "epoch": 0.9000195547112887, + "grad_norm": 0.8384383916854858, + "learning_rate": 0.000551994621246078, + "loss": 0.5757, + "step": 16109 + }, + { + "epoch": 0.9000754253149705, + "grad_norm": 0.562502384185791, + "learning_rate": 0.0005519666069027342, + "loss": 0.6221, + "step": 16110 + }, + { + "epoch": 0.9001312959186524, + "grad_norm": 0.4402511715888977, + "learning_rate": 0.0005519385925593904, + "loss": 0.4657, + "step": 16111 + }, + { + "epoch": 0.9001871665223343, + "grad_norm": 0.3293793797492981, + "learning_rate": 0.0005519105782160466, + "loss": 0.482, + "step": 16112 + }, + { + "epoch": 0.9002430371260162, + "grad_norm": 0.4683966636657715, + "learning_rate": 0.0005518825638727028, + "loss": 0.3883, + "step": 16113 + }, + { + "epoch": 0.900298907729698, + "grad_norm": 0.4524246156215668, + "learning_rate": 0.0005518545495293591, + "loss": 0.4075, + "step": 16114 + }, + { + "epoch": 0.9003547783333798, + "grad_norm": 0.4922575354576111, + "learning_rate": 0.0005518265351860152, + "loss": 0.4964, + "step": 16115 + }, + { + "epoch": 0.9004106489370618, + "grad_norm": 0.5367220044136047, + "learning_rate": 0.0005517985208426714, + "loss": 0.5242, + "step": 16116 + }, + { + "epoch": 0.9004665195407436, + "grad_norm": 0.5088585615158081, + "learning_rate": 0.0005517705064993277, + "loss": 0.4199, + "step": 16117 + }, + { + "epoch": 0.9005223901444255, + "grad_norm": 1.0236070156097412, + "learning_rate": 0.0005517424921559839, + "loss": 0.4872, + "step": 16118 + }, + { + "epoch": 0.9005782607481074, + "grad_norm": 0.41640010476112366, + "learning_rate": 0.0005517144778126401, + "loss": 0.4747, + "step": 16119 + }, + { + "epoch": 0.9006341313517893, + "grad_norm": 0.6453925967216492, + "learning_rate": 0.0005516864634692963, + "loss": 0.4581, + "step": 16120 + }, + { + "epoch": 0.9006900019554711, + "grad_norm": 0.4647132456302643, + "learning_rate": 0.0005516584491259526, + "loss": 0.4606, + "step": 16121 + }, + { + "epoch": 0.900745872559153, + "grad_norm": 1.0425384044647217, + "learning_rate": 0.0005516304347826087, + "loss": 0.485, + "step": 16122 + }, + { + "epoch": 0.9008017431628349, + "grad_norm": 0.50307297706604, + "learning_rate": 0.0005516024204392649, + "loss": 0.5135, + "step": 16123 + }, + { + "epoch": 0.9008576137665167, + "grad_norm": 0.4797559976577759, + "learning_rate": 0.0005515744060959211, + "loss": 0.5234, + "step": 16124 + }, + { + "epoch": 0.9009134843701986, + "grad_norm": 1.9473345279693604, + "learning_rate": 0.0005515463917525774, + "loss": 0.4029, + "step": 16125 + }, + { + "epoch": 0.9009693549738805, + "grad_norm": 0.48266881704330444, + "learning_rate": 0.0005515183774092335, + "loss": 0.3642, + "step": 16126 + }, + { + "epoch": 0.9010252255775624, + "grad_norm": 1.2739180326461792, + "learning_rate": 0.0005514903630658897, + "loss": 0.5391, + "step": 16127 + }, + { + "epoch": 0.9010810961812442, + "grad_norm": 1.174964189529419, + "learning_rate": 0.0005514623487225459, + "loss": 0.3567, + "step": 16128 + }, + { + "epoch": 0.9011369667849262, + "grad_norm": 1.592631220817566, + "learning_rate": 0.0005514343343792022, + "loss": 0.3328, + "step": 16129 + }, + { + "epoch": 0.901192837388608, + "grad_norm": 0.5605488419532776, + "learning_rate": 0.0005514063200358583, + "loss": 0.4419, + "step": 16130 + }, + { + "epoch": 0.9012487079922898, + "grad_norm": 0.3686579465866089, + "learning_rate": 0.0005513783056925145, + "loss": 0.4085, + "step": 16131 + }, + { + "epoch": 0.9013045785959717, + "grad_norm": 0.49798282980918884, + "learning_rate": 0.0005513502913491708, + "loss": 0.504, + "step": 16132 + }, + { + "epoch": 0.9013604491996536, + "grad_norm": 0.4310070276260376, + "learning_rate": 0.000551322277005827, + "loss": 0.5465, + "step": 16133 + }, + { + "epoch": 0.9014163198033355, + "grad_norm": 0.3788164258003235, + "learning_rate": 0.0005512942626624832, + "loss": 0.4879, + "step": 16134 + }, + { + "epoch": 0.9014721904070173, + "grad_norm": 2.306981086730957, + "learning_rate": 0.0005512662483191394, + "loss": 0.4659, + "step": 16135 + }, + { + "epoch": 0.9015280610106993, + "grad_norm": 0.48127010464668274, + "learning_rate": 0.0005512382339757957, + "loss": 0.5029, + "step": 16136 + }, + { + "epoch": 0.9015839316143811, + "grad_norm": 0.6584022641181946, + "learning_rate": 0.0005512102196324519, + "loss": 0.3952, + "step": 16137 + }, + { + "epoch": 0.901639802218063, + "grad_norm": 0.3581736385822296, + "learning_rate": 0.000551182205289108, + "loss": 0.3328, + "step": 16138 + }, + { + "epoch": 0.9016956728217448, + "grad_norm": 0.808049201965332, + "learning_rate": 0.0005511541909457643, + "loss": 0.6212, + "step": 16139 + }, + { + "epoch": 0.9017515434254267, + "grad_norm": 1.9523288011550903, + "learning_rate": 0.0005511261766024205, + "loss": 0.6403, + "step": 16140 + }, + { + "epoch": 0.9018074140291086, + "grad_norm": 0.8837628364562988, + "learning_rate": 0.0005510981622590766, + "loss": 0.4888, + "step": 16141 + }, + { + "epoch": 0.9018632846327904, + "grad_norm": 0.8269013166427612, + "learning_rate": 0.0005510701479157328, + "loss": 0.5, + "step": 16142 + }, + { + "epoch": 0.9019191552364724, + "grad_norm": 0.5855947732925415, + "learning_rate": 0.0005510421335723891, + "loss": 0.5556, + "step": 16143 + }, + { + "epoch": 0.9019750258401542, + "grad_norm": 0.7237779498100281, + "learning_rate": 0.0005510141192290453, + "loss": 0.4635, + "step": 16144 + }, + { + "epoch": 0.902030896443836, + "grad_norm": 0.3597396910190582, + "learning_rate": 0.0005509861048857014, + "loss": 0.5308, + "step": 16145 + }, + { + "epoch": 0.902086767047518, + "grad_norm": 0.4284672737121582, + "learning_rate": 0.0005509580905423577, + "loss": 0.4677, + "step": 16146 + }, + { + "epoch": 0.9021426376511998, + "grad_norm": 0.4158279597759247, + "learning_rate": 0.0005509300761990139, + "loss": 0.35, + "step": 16147 + }, + { + "epoch": 0.9021985082548817, + "grad_norm": 0.5588980317115784, + "learning_rate": 0.0005509020618556701, + "loss": 0.469, + "step": 16148 + }, + { + "epoch": 0.9022543788585635, + "grad_norm": 0.36384886503219604, + "learning_rate": 0.0005508740475123262, + "loss": 0.3517, + "step": 16149 + }, + { + "epoch": 0.9023102494622455, + "grad_norm": 1.7219589948654175, + "learning_rate": 0.0005508460331689826, + "loss": 0.7764, + "step": 16150 + }, + { + "epoch": 0.9023661200659273, + "grad_norm": 0.443052738904953, + "learning_rate": 0.0005508180188256388, + "loss": 0.3849, + "step": 16151 + }, + { + "epoch": 0.9024219906696092, + "grad_norm": 0.4508678913116455, + "learning_rate": 0.000550790004482295, + "loss": 0.3974, + "step": 16152 + }, + { + "epoch": 0.9024778612732911, + "grad_norm": 0.8677716851234436, + "learning_rate": 0.0005507619901389512, + "loss": 0.5605, + "step": 16153 + }, + { + "epoch": 0.9025337318769729, + "grad_norm": 0.6627373695373535, + "learning_rate": 0.0005507339757956074, + "loss": 0.4366, + "step": 16154 + }, + { + "epoch": 0.9025896024806548, + "grad_norm": 0.8753539323806763, + "learning_rate": 0.0005507059614522636, + "loss": 0.4912, + "step": 16155 + }, + { + "epoch": 0.9026454730843366, + "grad_norm": 3.543874740600586, + "learning_rate": 0.0005506779471089198, + "loss": 0.4991, + "step": 16156 + }, + { + "epoch": 0.9027013436880186, + "grad_norm": 0.7585520148277283, + "learning_rate": 0.000550649932765576, + "loss": 0.3715, + "step": 16157 + }, + { + "epoch": 0.9027572142917004, + "grad_norm": 0.4524703025817871, + "learning_rate": 0.0005506219184222322, + "loss": 0.3476, + "step": 16158 + }, + { + "epoch": 0.9028130848953823, + "grad_norm": 0.583713710308075, + "learning_rate": 0.0005505939040788884, + "loss": 0.6099, + "step": 16159 + }, + { + "epoch": 0.9028689554990642, + "grad_norm": 0.7144328355789185, + "learning_rate": 0.0005505658897355445, + "loss": 0.5011, + "step": 16160 + }, + { + "epoch": 0.902924826102746, + "grad_norm": 0.5369505286216736, + "learning_rate": 0.0005505378753922008, + "loss": 0.5173, + "step": 16161 + }, + { + "epoch": 0.9029806967064279, + "grad_norm": 0.5189807415008545, + "learning_rate": 0.000550509861048857, + "loss": 0.6176, + "step": 16162 + }, + { + "epoch": 0.9030365673101098, + "grad_norm": 0.388011634349823, + "learning_rate": 0.0005504818467055132, + "loss": 0.4158, + "step": 16163 + }, + { + "epoch": 0.9030924379137917, + "grad_norm": 0.45745953917503357, + "learning_rate": 0.0005504538323621694, + "loss": 0.4879, + "step": 16164 + }, + { + "epoch": 0.9031483085174735, + "grad_norm": 0.6208248138427734, + "learning_rate": 0.0005504258180188256, + "loss": 0.5658, + "step": 16165 + }, + { + "epoch": 0.9032041791211554, + "grad_norm": 0.382365345954895, + "learning_rate": 0.0005503978036754818, + "loss": 0.3712, + "step": 16166 + }, + { + "epoch": 0.9032600497248373, + "grad_norm": 0.5338571667671204, + "learning_rate": 0.000550369789332138, + "loss": 0.455, + "step": 16167 + }, + { + "epoch": 0.9033159203285192, + "grad_norm": 0.5821576118469238, + "learning_rate": 0.0005503417749887943, + "loss": 0.4662, + "step": 16168 + }, + { + "epoch": 0.903371790932201, + "grad_norm": 0.4655402600765228, + "learning_rate": 0.0005503137606454505, + "loss": 0.3611, + "step": 16169 + }, + { + "epoch": 0.9034276615358829, + "grad_norm": 0.4905712306499481, + "learning_rate": 0.0005502857463021067, + "loss": 0.4052, + "step": 16170 + }, + { + "epoch": 0.9034835321395648, + "grad_norm": 1.3888847827911377, + "learning_rate": 0.000550257731958763, + "loss": 0.4034, + "step": 16171 + }, + { + "epoch": 0.9035394027432466, + "grad_norm": 0.4994279444217682, + "learning_rate": 0.0005502297176154191, + "loss": 0.484, + "step": 16172 + }, + { + "epoch": 0.9035952733469285, + "grad_norm": 0.3796749711036682, + "learning_rate": 0.0005502017032720753, + "loss": 0.4509, + "step": 16173 + }, + { + "epoch": 0.9036511439506104, + "grad_norm": 1.9028922319412231, + "learning_rate": 0.0005501736889287315, + "loss": 0.3925, + "step": 16174 + }, + { + "epoch": 0.9037070145542923, + "grad_norm": 0.5131769180297852, + "learning_rate": 0.0005501456745853878, + "loss": 0.5971, + "step": 16175 + }, + { + "epoch": 0.9037628851579741, + "grad_norm": 0.4477365016937256, + "learning_rate": 0.0005501176602420439, + "loss": 0.4207, + "step": 16176 + }, + { + "epoch": 0.903818755761656, + "grad_norm": 0.7178270816802979, + "learning_rate": 0.0005500896458987001, + "loss": 0.3891, + "step": 16177 + }, + { + "epoch": 0.9038746263653379, + "grad_norm": 1.7003569602966309, + "learning_rate": 0.0005500616315553564, + "loss": 0.4016, + "step": 16178 + }, + { + "epoch": 0.9039304969690197, + "grad_norm": 0.4362263083457947, + "learning_rate": 0.0005500336172120126, + "loss": 0.4364, + "step": 16179 + }, + { + "epoch": 0.9039863675727017, + "grad_norm": 0.5964049696922302, + "learning_rate": 0.0005500056028686687, + "loss": 0.5033, + "step": 16180 + }, + { + "epoch": 0.9040422381763835, + "grad_norm": 0.7405802011489868, + "learning_rate": 0.0005499775885253249, + "loss": 0.4414, + "step": 16181 + }, + { + "epoch": 0.9040981087800654, + "grad_norm": 0.4541977643966675, + "learning_rate": 0.0005499495741819812, + "loss": 0.4718, + "step": 16182 + }, + { + "epoch": 0.9041539793837472, + "grad_norm": 0.41367921233177185, + "learning_rate": 0.0005499215598386373, + "loss": 0.3166, + "step": 16183 + }, + { + "epoch": 0.9042098499874291, + "grad_norm": 0.6425495743751526, + "learning_rate": 0.0005498935454952936, + "loss": 0.4305, + "step": 16184 + }, + { + "epoch": 0.904265720591111, + "grad_norm": 0.43169254064559937, + "learning_rate": 0.0005498655311519499, + "loss": 0.4346, + "step": 16185 + }, + { + "epoch": 0.9043215911947928, + "grad_norm": 0.37898364663124084, + "learning_rate": 0.0005498375168086061, + "loss": 0.4417, + "step": 16186 + }, + { + "epoch": 0.9043774617984748, + "grad_norm": 0.38514333963394165, + "learning_rate": 0.0005498095024652622, + "loss": 0.3941, + "step": 16187 + }, + { + "epoch": 0.9044333324021566, + "grad_norm": 0.6643823385238647, + "learning_rate": 0.0005497814881219184, + "loss": 0.4626, + "step": 16188 + }, + { + "epoch": 0.9044892030058385, + "grad_norm": 0.8208126425743103, + "learning_rate": 0.0005497534737785747, + "loss": 0.4674, + "step": 16189 + }, + { + "epoch": 0.9045450736095203, + "grad_norm": 1.6581814289093018, + "learning_rate": 0.0005497254594352309, + "loss": 0.4407, + "step": 16190 + }, + { + "epoch": 0.9046009442132023, + "grad_norm": 0.3896174132823944, + "learning_rate": 0.000549697445091887, + "loss": 0.4409, + "step": 16191 + }, + { + "epoch": 0.9046568148168841, + "grad_norm": 0.5777064561843872, + "learning_rate": 0.0005496694307485433, + "loss": 0.4481, + "step": 16192 + }, + { + "epoch": 0.9047126854205659, + "grad_norm": 0.5532296299934387, + "learning_rate": 0.0005496414164051995, + "loss": 0.4919, + "step": 16193 + }, + { + "epoch": 0.9047685560242479, + "grad_norm": 0.46235689520835876, + "learning_rate": 0.0005496134020618557, + "loss": 0.4193, + "step": 16194 + }, + { + "epoch": 0.9048244266279297, + "grad_norm": 0.4743284285068512, + "learning_rate": 0.0005495853877185118, + "loss": 0.3597, + "step": 16195 + }, + { + "epoch": 0.9048802972316116, + "grad_norm": 0.5099502205848694, + "learning_rate": 0.0005495573733751681, + "loss": 0.4444, + "step": 16196 + }, + { + "epoch": 0.9049361678352935, + "grad_norm": 0.29832062125205994, + "learning_rate": 0.0005495293590318243, + "loss": 0.3762, + "step": 16197 + }, + { + "epoch": 0.9049920384389754, + "grad_norm": 0.4942917227745056, + "learning_rate": 0.0005495013446884805, + "loss": 0.4385, + "step": 16198 + }, + { + "epoch": 0.9050479090426572, + "grad_norm": 0.5459005236625671, + "learning_rate": 0.0005494733303451366, + "loss": 0.5889, + "step": 16199 + }, + { + "epoch": 0.905103779646339, + "grad_norm": 0.4161585867404938, + "learning_rate": 0.000549445316001793, + "loss": 0.4142, + "step": 16200 + }, + { + "epoch": 0.905159650250021, + "grad_norm": 0.4775237441062927, + "learning_rate": 0.0005494173016584492, + "loss": 0.4142, + "step": 16201 + }, + { + "epoch": 0.9052155208537028, + "grad_norm": 0.38917604088783264, + "learning_rate": 0.0005493892873151053, + "loss": 0.4516, + "step": 16202 + }, + { + "epoch": 0.9052713914573847, + "grad_norm": 0.6256323456764221, + "learning_rate": 0.0005493612729717616, + "loss": 0.4159, + "step": 16203 + }, + { + "epoch": 0.9053272620610666, + "grad_norm": 0.6033145189285278, + "learning_rate": 0.0005493332586284178, + "loss": 0.5419, + "step": 16204 + }, + { + "epoch": 0.9053831326647485, + "grad_norm": 0.3735802471637726, + "learning_rate": 0.000549305244285074, + "loss": 0.347, + "step": 16205 + }, + { + "epoch": 0.9054390032684303, + "grad_norm": 0.3873332738876343, + "learning_rate": 0.0005492772299417301, + "loss": 0.5414, + "step": 16206 + }, + { + "epoch": 0.9054948738721121, + "grad_norm": 0.4950906038284302, + "learning_rate": 0.0005492492155983864, + "loss": 0.4893, + "step": 16207 + }, + { + "epoch": 0.9055507444757941, + "grad_norm": 0.5245281457901001, + "learning_rate": 0.0005492212012550426, + "loss": 0.424, + "step": 16208 + }, + { + "epoch": 0.9056066150794759, + "grad_norm": 0.41432151198387146, + "learning_rate": 0.0005491931869116988, + "loss": 0.3751, + "step": 16209 + }, + { + "epoch": 0.9056624856831578, + "grad_norm": 1.0052951574325562, + "learning_rate": 0.000549165172568355, + "loss": 0.542, + "step": 16210 + }, + { + "epoch": 0.9057183562868397, + "grad_norm": 0.5491342544555664, + "learning_rate": 0.0005491371582250112, + "loss": 0.5571, + "step": 16211 + }, + { + "epoch": 0.9057742268905216, + "grad_norm": 0.5978744626045227, + "learning_rate": 0.0005491091438816674, + "loss": 0.5079, + "step": 16212 + }, + { + "epoch": 0.9058300974942034, + "grad_norm": 0.564390242099762, + "learning_rate": 0.0005490811295383236, + "loss": 0.3534, + "step": 16213 + }, + { + "epoch": 0.9058859680978854, + "grad_norm": 0.5390400886535645, + "learning_rate": 0.0005490531151949798, + "loss": 0.4311, + "step": 16214 + }, + { + "epoch": 0.9059418387015672, + "grad_norm": 0.5850682854652405, + "learning_rate": 0.000549025100851636, + "loss": 0.4386, + "step": 16215 + }, + { + "epoch": 0.905997709305249, + "grad_norm": 0.40524232387542725, + "learning_rate": 0.0005489970865082922, + "loss": 0.3281, + "step": 16216 + }, + { + "epoch": 0.9060535799089309, + "grad_norm": 0.4370447099208832, + "learning_rate": 0.0005489690721649486, + "loss": 0.454, + "step": 16217 + }, + { + "epoch": 0.9061094505126128, + "grad_norm": 4.512547969818115, + "learning_rate": 0.0005489410578216047, + "loss": 0.4632, + "step": 16218 + }, + { + "epoch": 0.9061653211162947, + "grad_norm": 0.36643186211586, + "learning_rate": 0.0005489130434782609, + "loss": 0.3505, + "step": 16219 + }, + { + "epoch": 0.9062211917199765, + "grad_norm": 0.3515947461128235, + "learning_rate": 0.0005488850291349171, + "loss": 0.4094, + "step": 16220 + }, + { + "epoch": 0.9062770623236585, + "grad_norm": 0.44268956780433655, + "learning_rate": 0.0005488570147915734, + "loss": 0.4455, + "step": 16221 + }, + { + "epoch": 0.9063329329273403, + "grad_norm": 0.9644373655319214, + "learning_rate": 0.0005488290004482295, + "loss": 0.5191, + "step": 16222 + }, + { + "epoch": 0.9063888035310221, + "grad_norm": 0.3954141139984131, + "learning_rate": 0.0005488009861048857, + "loss": 0.3955, + "step": 16223 + }, + { + "epoch": 0.906444674134704, + "grad_norm": 0.3575811982154846, + "learning_rate": 0.000548772971761542, + "loss": 0.3552, + "step": 16224 + }, + { + "epoch": 0.9065005447383859, + "grad_norm": 0.4260707497596741, + "learning_rate": 0.0005487449574181981, + "loss": 0.5445, + "step": 16225 + }, + { + "epoch": 0.9065564153420678, + "grad_norm": 0.5555064082145691, + "learning_rate": 0.0005487169430748543, + "loss": 0.5266, + "step": 16226 + }, + { + "epoch": 0.9066122859457496, + "grad_norm": 0.32211264967918396, + "learning_rate": 0.0005486889287315105, + "loss": 0.3386, + "step": 16227 + }, + { + "epoch": 0.9066681565494316, + "grad_norm": 0.664693295955658, + "learning_rate": 0.0005486609143881668, + "loss": 0.4353, + "step": 16228 + }, + { + "epoch": 0.9067240271531134, + "grad_norm": 0.8305293917655945, + "learning_rate": 0.0005486329000448229, + "loss": 0.3986, + "step": 16229 + }, + { + "epoch": 0.9067798977567952, + "grad_norm": 0.8363955020904541, + "learning_rate": 0.0005486048857014791, + "loss": 0.5075, + "step": 16230 + }, + { + "epoch": 0.9068357683604772, + "grad_norm": 0.5620028376579285, + "learning_rate": 0.0005485768713581354, + "loss": 0.447, + "step": 16231 + }, + { + "epoch": 0.906891638964159, + "grad_norm": 1.5475029945373535, + "learning_rate": 0.0005485488570147916, + "loss": 0.2321, + "step": 16232 + }, + { + "epoch": 0.9069475095678409, + "grad_norm": 0.4270790219306946, + "learning_rate": 0.0005485208426714477, + "loss": 0.388, + "step": 16233 + }, + { + "epoch": 0.9070033801715227, + "grad_norm": 0.42549335956573486, + "learning_rate": 0.000548492828328104, + "loss": 0.4629, + "step": 16234 + }, + { + "epoch": 0.9070592507752047, + "grad_norm": 0.9378711581230164, + "learning_rate": 0.0005484648139847603, + "loss": 0.4291, + "step": 16235 + }, + { + "epoch": 0.9071151213788865, + "grad_norm": 0.7091695070266724, + "learning_rate": 0.0005484367996414165, + "loss": 0.3924, + "step": 16236 + }, + { + "epoch": 0.9071709919825683, + "grad_norm": 0.37537428736686707, + "learning_rate": 0.0005484087852980726, + "loss": 0.318, + "step": 16237 + }, + { + "epoch": 0.9072268625862503, + "grad_norm": 0.5040545463562012, + "learning_rate": 0.0005483807709547288, + "loss": 0.4221, + "step": 16238 + }, + { + "epoch": 0.9072827331899321, + "grad_norm": 0.6021084189414978, + "learning_rate": 0.0005483527566113851, + "loss": 0.3425, + "step": 16239 + }, + { + "epoch": 0.907338603793614, + "grad_norm": 0.9014391899108887, + "learning_rate": 0.0005483247422680413, + "loss": 0.4786, + "step": 16240 + }, + { + "epoch": 0.9073944743972958, + "grad_norm": 0.7037850022315979, + "learning_rate": 0.0005482967279246974, + "loss": 0.5142, + "step": 16241 + }, + { + "epoch": 0.9074503450009778, + "grad_norm": 0.4211156964302063, + "learning_rate": 0.0005482687135813537, + "loss": 0.4584, + "step": 16242 + }, + { + "epoch": 0.9075062156046596, + "grad_norm": 0.665021538734436, + "learning_rate": 0.0005482406992380099, + "loss": 0.4923, + "step": 16243 + }, + { + "epoch": 0.9075620862083414, + "grad_norm": 0.4094330966472626, + "learning_rate": 0.000548212684894666, + "loss": 0.4598, + "step": 16244 + }, + { + "epoch": 0.9076179568120234, + "grad_norm": 0.4049864113330841, + "learning_rate": 0.0005481846705513222, + "loss": 0.4407, + "step": 16245 + }, + { + "epoch": 0.9076738274157052, + "grad_norm": 0.4581248164176941, + "learning_rate": 0.0005481566562079785, + "loss": 0.4166, + "step": 16246 + }, + { + "epoch": 0.9077296980193871, + "grad_norm": 0.5206487774848938, + "learning_rate": 0.0005481286418646347, + "loss": 0.396, + "step": 16247 + }, + { + "epoch": 0.9077855686230689, + "grad_norm": 0.4354107677936554, + "learning_rate": 0.0005481006275212908, + "loss": 0.4731, + "step": 16248 + }, + { + "epoch": 0.9078414392267509, + "grad_norm": 1.5217713117599487, + "learning_rate": 0.0005480726131779471, + "loss": 0.4496, + "step": 16249 + }, + { + "epoch": 0.9078973098304327, + "grad_norm": 0.7017619609832764, + "learning_rate": 0.0005480445988346034, + "loss": 0.5395, + "step": 16250 + }, + { + "epoch": 0.9079531804341145, + "grad_norm": 0.5292048454284668, + "learning_rate": 0.0005480165844912596, + "loss": 0.4078, + "step": 16251 + }, + { + "epoch": 0.9080090510377965, + "grad_norm": 0.3053964674472809, + "learning_rate": 0.0005479885701479157, + "loss": 0.2638, + "step": 16252 + }, + { + "epoch": 0.9080649216414783, + "grad_norm": 0.5904061794281006, + "learning_rate": 0.000547960555804572, + "loss": 0.3978, + "step": 16253 + }, + { + "epoch": 0.9081207922451602, + "grad_norm": 0.5044125318527222, + "learning_rate": 0.0005479325414612282, + "loss": 0.5119, + "step": 16254 + }, + { + "epoch": 0.9081766628488421, + "grad_norm": 0.8480036854743958, + "learning_rate": 0.0005479045271178844, + "loss": 0.4881, + "step": 16255 + }, + { + "epoch": 0.908232533452524, + "grad_norm": 0.42585262656211853, + "learning_rate": 0.0005478765127745406, + "loss": 0.4943, + "step": 16256 + }, + { + "epoch": 0.9082884040562058, + "grad_norm": 1.877628207206726, + "learning_rate": 0.0005478484984311968, + "loss": 0.421, + "step": 16257 + }, + { + "epoch": 0.9083442746598877, + "grad_norm": 0.4252227544784546, + "learning_rate": 0.000547820484087853, + "loss": 0.5253, + "step": 16258 + }, + { + "epoch": 0.9084001452635696, + "grad_norm": 0.41062939167022705, + "learning_rate": 0.0005477924697445092, + "loss": 0.4594, + "step": 16259 + }, + { + "epoch": 0.9084560158672514, + "grad_norm": 0.3366455137729645, + "learning_rate": 0.0005477644554011654, + "loss": 0.4231, + "step": 16260 + }, + { + "epoch": 0.9085118864709333, + "grad_norm": 0.47494930028915405, + "learning_rate": 0.0005477364410578216, + "loss": 0.5064, + "step": 16261 + }, + { + "epoch": 0.9085677570746152, + "grad_norm": 0.44242778420448303, + "learning_rate": 0.0005477084267144778, + "loss": 0.5005, + "step": 16262 + }, + { + "epoch": 0.9086236276782971, + "grad_norm": 0.8019917011260986, + "learning_rate": 0.0005476804123711341, + "loss": 0.4403, + "step": 16263 + }, + { + "epoch": 0.9086794982819789, + "grad_norm": 0.5826294422149658, + "learning_rate": 0.0005476523980277902, + "loss": 0.4815, + "step": 16264 + }, + { + "epoch": 0.9087353688856608, + "grad_norm": 0.4641910493373871, + "learning_rate": 0.0005476243836844464, + "loss": 0.5078, + "step": 16265 + }, + { + "epoch": 0.9087912394893427, + "grad_norm": 1.5344514846801758, + "learning_rate": 0.0005475963693411026, + "loss": 0.3812, + "step": 16266 + }, + { + "epoch": 0.9088471100930245, + "grad_norm": 0.47340652346611023, + "learning_rate": 0.0005475683549977588, + "loss": 0.3816, + "step": 16267 + }, + { + "epoch": 0.9089029806967064, + "grad_norm": 0.751927375793457, + "learning_rate": 0.0005475403406544151, + "loss": 0.5674, + "step": 16268 + }, + { + "epoch": 0.9089588513003883, + "grad_norm": 1.2583779096603394, + "learning_rate": 0.0005475123263110713, + "loss": 0.4293, + "step": 16269 + }, + { + "epoch": 0.9090147219040702, + "grad_norm": 0.6767417788505554, + "learning_rate": 0.0005474843119677276, + "loss": 0.5743, + "step": 16270 + }, + { + "epoch": 0.909070592507752, + "grad_norm": 0.3895435929298401, + "learning_rate": 0.0005474562976243837, + "loss": 0.3745, + "step": 16271 + }, + { + "epoch": 0.909126463111434, + "grad_norm": 0.5040707588195801, + "learning_rate": 0.0005474282832810399, + "loss": 0.6857, + "step": 16272 + }, + { + "epoch": 0.9091823337151158, + "grad_norm": 0.5367717742919922, + "learning_rate": 0.0005474002689376961, + "loss": 0.4857, + "step": 16273 + }, + { + "epoch": 0.9092382043187976, + "grad_norm": 1.0114457607269287, + "learning_rate": 0.0005473722545943524, + "loss": 0.4775, + "step": 16274 + }, + { + "epoch": 0.9092940749224795, + "grad_norm": 1.061826229095459, + "learning_rate": 0.0005473442402510085, + "loss": 0.5288, + "step": 16275 + }, + { + "epoch": 0.9093499455261614, + "grad_norm": 0.5313809514045715, + "learning_rate": 0.0005473162259076647, + "loss": 0.4416, + "step": 16276 + }, + { + "epoch": 0.9094058161298433, + "grad_norm": 0.6007213592529297, + "learning_rate": 0.0005472882115643209, + "loss": 0.5203, + "step": 16277 + }, + { + "epoch": 0.9094616867335251, + "grad_norm": 1.9363266229629517, + "learning_rate": 0.0005472601972209772, + "loss": 0.5258, + "step": 16278 + }, + { + "epoch": 0.9095175573372071, + "grad_norm": 0.430054634809494, + "learning_rate": 0.0005472321828776333, + "loss": 0.359, + "step": 16279 + }, + { + "epoch": 0.9095734279408889, + "grad_norm": 0.6194666028022766, + "learning_rate": 0.0005472041685342895, + "loss": 0.445, + "step": 16280 + }, + { + "epoch": 0.9096292985445708, + "grad_norm": 0.40350887179374695, + "learning_rate": 0.0005471761541909458, + "loss": 0.4117, + "step": 16281 + }, + { + "epoch": 0.9096851691482526, + "grad_norm": 0.6580944061279297, + "learning_rate": 0.000547148139847602, + "loss": 0.481, + "step": 16282 + }, + { + "epoch": 0.9097410397519345, + "grad_norm": 0.3859778940677643, + "learning_rate": 0.0005471201255042581, + "loss": 0.423, + "step": 16283 + }, + { + "epoch": 0.9097969103556164, + "grad_norm": 5.030828475952148, + "learning_rate": 0.0005470921111609143, + "loss": 0.5656, + "step": 16284 + }, + { + "epoch": 0.9098527809592982, + "grad_norm": 1.1176328659057617, + "learning_rate": 0.0005470640968175707, + "loss": 0.6058, + "step": 16285 + }, + { + "epoch": 0.9099086515629802, + "grad_norm": 0.5661559104919434, + "learning_rate": 0.0005470360824742269, + "loss": 0.4129, + "step": 16286 + }, + { + "epoch": 0.909964522166662, + "grad_norm": 0.5328323841094971, + "learning_rate": 0.000547008068130883, + "loss": 0.4556, + "step": 16287 + }, + { + "epoch": 0.9100203927703439, + "grad_norm": 1.0796958208084106, + "learning_rate": 0.0005469800537875393, + "loss": 0.5029, + "step": 16288 + }, + { + "epoch": 0.9100762633740258, + "grad_norm": 0.543998658657074, + "learning_rate": 0.0005469520394441955, + "loss": 0.4943, + "step": 16289 + }, + { + "epoch": 0.9101321339777076, + "grad_norm": 0.5761914849281311, + "learning_rate": 0.0005469240251008516, + "loss": 0.4378, + "step": 16290 + }, + { + "epoch": 0.9101880045813895, + "grad_norm": 0.40355706214904785, + "learning_rate": 0.0005468960107575078, + "loss": 0.5318, + "step": 16291 + }, + { + "epoch": 0.9102438751850713, + "grad_norm": 0.5257022976875305, + "learning_rate": 0.0005468679964141641, + "loss": 0.3727, + "step": 16292 + }, + { + "epoch": 0.9102997457887533, + "grad_norm": 0.5615121722221375, + "learning_rate": 0.0005468399820708203, + "loss": 0.283, + "step": 16293 + }, + { + "epoch": 0.9103556163924351, + "grad_norm": 0.5849764943122864, + "learning_rate": 0.0005468119677274764, + "loss": 0.3786, + "step": 16294 + }, + { + "epoch": 0.910411486996117, + "grad_norm": 0.5208074450492859, + "learning_rate": 0.0005467839533841327, + "loss": 0.3983, + "step": 16295 + }, + { + "epoch": 0.9104673575997989, + "grad_norm": 0.5235249400138855, + "learning_rate": 0.0005467559390407889, + "loss": 0.5272, + "step": 16296 + }, + { + "epoch": 0.9105232282034808, + "grad_norm": 0.5722655653953552, + "learning_rate": 0.0005467279246974451, + "loss": 0.4031, + "step": 16297 + }, + { + "epoch": 0.9105790988071626, + "grad_norm": 0.42282360792160034, + "learning_rate": 0.0005466999103541012, + "loss": 0.4229, + "step": 16298 + }, + { + "epoch": 0.9106349694108444, + "grad_norm": 0.4990462064743042, + "learning_rate": 0.0005466718960107575, + "loss": 0.4341, + "step": 16299 + }, + { + "epoch": 0.9106908400145264, + "grad_norm": 0.4012435972690582, + "learning_rate": 0.0005466438816674138, + "loss": 0.4065, + "step": 16300 + }, + { + "epoch": 0.9107467106182082, + "grad_norm": 0.69905024766922, + "learning_rate": 0.00054661586732407, + "loss": 0.5334, + "step": 16301 + }, + { + "epoch": 0.9108025812218901, + "grad_norm": 0.5711586475372314, + "learning_rate": 0.0005465878529807262, + "loss": 0.4617, + "step": 16302 + }, + { + "epoch": 0.910858451825572, + "grad_norm": 0.43020474910736084, + "learning_rate": 0.0005465598386373824, + "loss": 0.5243, + "step": 16303 + }, + { + "epoch": 0.9109143224292539, + "grad_norm": 0.5101274251937866, + "learning_rate": 0.0005465318242940386, + "loss": 0.4482, + "step": 16304 + }, + { + "epoch": 0.9109701930329357, + "grad_norm": 1.6524381637573242, + "learning_rate": 0.0005465038099506948, + "loss": 0.4707, + "step": 16305 + }, + { + "epoch": 0.9110260636366176, + "grad_norm": 0.4378310441970825, + "learning_rate": 0.000546475795607351, + "loss": 0.4181, + "step": 16306 + }, + { + "epoch": 0.9110819342402995, + "grad_norm": 2.6578989028930664, + "learning_rate": 0.0005464477812640072, + "loss": 0.4947, + "step": 16307 + }, + { + "epoch": 0.9111378048439813, + "grad_norm": 0.8064612746238708, + "learning_rate": 0.0005464197669206634, + "loss": 0.561, + "step": 16308 + }, + { + "epoch": 0.9111936754476632, + "grad_norm": 0.7501070499420166, + "learning_rate": 0.0005463917525773195, + "loss": 0.4579, + "step": 16309 + }, + { + "epoch": 0.9112495460513451, + "grad_norm": 1.7316964864730835, + "learning_rate": 0.0005463637382339758, + "loss": 0.4665, + "step": 16310 + }, + { + "epoch": 0.911305416655027, + "grad_norm": 0.6044100522994995, + "learning_rate": 0.000546335723890632, + "loss": 0.5278, + "step": 16311 + }, + { + "epoch": 0.9113612872587088, + "grad_norm": 0.4820186495780945, + "learning_rate": 0.0005463077095472882, + "loss": 0.4138, + "step": 16312 + }, + { + "epoch": 0.9114171578623907, + "grad_norm": 0.41069164872169495, + "learning_rate": 0.0005462796952039444, + "loss": 0.4938, + "step": 16313 + }, + { + "epoch": 0.9114730284660726, + "grad_norm": 0.4627922475337982, + "learning_rate": 0.0005462516808606006, + "loss": 0.4371, + "step": 16314 + }, + { + "epoch": 0.9115288990697544, + "grad_norm": 0.37290453910827637, + "learning_rate": 0.0005462236665172568, + "loss": 0.3902, + "step": 16315 + }, + { + "epoch": 0.9115847696734363, + "grad_norm": 0.43658581376075745, + "learning_rate": 0.000546195652173913, + "loss": 0.4929, + "step": 16316 + }, + { + "epoch": 0.9116406402771182, + "grad_norm": 0.48671141266822815, + "learning_rate": 0.0005461676378305692, + "loss": 0.3814, + "step": 16317 + }, + { + "epoch": 0.9116965108808001, + "grad_norm": 0.5328053832054138, + "learning_rate": 0.0005461396234872255, + "loss": 0.4114, + "step": 16318 + }, + { + "epoch": 0.9117523814844819, + "grad_norm": 0.5264790058135986, + "learning_rate": 0.0005461116091438817, + "loss": 0.3708, + "step": 16319 + }, + { + "epoch": 0.9118082520881639, + "grad_norm": 0.5780777931213379, + "learning_rate": 0.000546083594800538, + "loss": 0.4307, + "step": 16320 + }, + { + "epoch": 0.9118641226918457, + "grad_norm": 0.35207968950271606, + "learning_rate": 0.0005460555804571941, + "loss": 0.3109, + "step": 16321 + }, + { + "epoch": 0.9119199932955275, + "grad_norm": 1.2995572090148926, + "learning_rate": 0.0005460275661138503, + "loss": 0.5994, + "step": 16322 + }, + { + "epoch": 0.9119758638992095, + "grad_norm": 0.38874930143356323, + "learning_rate": 0.0005459995517705065, + "loss": 0.4502, + "step": 16323 + }, + { + "epoch": 0.9120317345028913, + "grad_norm": 0.476474791765213, + "learning_rate": 0.0005459715374271628, + "loss": 0.4798, + "step": 16324 + }, + { + "epoch": 0.9120876051065732, + "grad_norm": 0.7723808288574219, + "learning_rate": 0.0005459435230838189, + "loss": 0.42, + "step": 16325 + }, + { + "epoch": 0.912143475710255, + "grad_norm": 0.4171563386917114, + "learning_rate": 0.0005459155087404751, + "loss": 0.3824, + "step": 16326 + }, + { + "epoch": 0.912199346313937, + "grad_norm": 0.6888163685798645, + "learning_rate": 0.0005458874943971314, + "loss": 0.3972, + "step": 16327 + }, + { + "epoch": 0.9122552169176188, + "grad_norm": 0.3483378291130066, + "learning_rate": 0.0005458594800537876, + "loss": 0.4627, + "step": 16328 + }, + { + "epoch": 0.9123110875213006, + "grad_norm": 1.3028945922851562, + "learning_rate": 0.0005458314657104437, + "loss": 0.5083, + "step": 16329 + }, + { + "epoch": 0.9123669581249826, + "grad_norm": 0.42183443903923035, + "learning_rate": 0.0005458034513670999, + "loss": 0.3444, + "step": 16330 + }, + { + "epoch": 0.9124228287286644, + "grad_norm": 1.6731088161468506, + "learning_rate": 0.0005457754370237562, + "loss": 0.4909, + "step": 16331 + }, + { + "epoch": 0.9124786993323463, + "grad_norm": 0.6874727606773376, + "learning_rate": 0.0005457474226804123, + "loss": 0.443, + "step": 16332 + }, + { + "epoch": 0.9125345699360281, + "grad_norm": 0.31641095876693726, + "learning_rate": 0.0005457194083370685, + "loss": 0.3782, + "step": 16333 + }, + { + "epoch": 0.9125904405397101, + "grad_norm": 0.7122689485549927, + "learning_rate": 0.0005456913939937249, + "loss": 0.4347, + "step": 16334 + }, + { + "epoch": 0.9126463111433919, + "grad_norm": 0.41536396741867065, + "learning_rate": 0.0005456633796503811, + "loss": 0.3582, + "step": 16335 + }, + { + "epoch": 0.9127021817470737, + "grad_norm": 0.6959275007247925, + "learning_rate": 0.0005456353653070372, + "loss": 0.4416, + "step": 16336 + }, + { + "epoch": 0.9127580523507557, + "grad_norm": 5.154533863067627, + "learning_rate": 0.0005456073509636934, + "loss": 0.4444, + "step": 16337 + }, + { + "epoch": 0.9128139229544375, + "grad_norm": 0.5343023538589478, + "learning_rate": 0.0005455793366203497, + "loss": 0.4396, + "step": 16338 + }, + { + "epoch": 0.9128697935581194, + "grad_norm": 1.0277392864227295, + "learning_rate": 0.0005455513222770059, + "loss": 0.4046, + "step": 16339 + }, + { + "epoch": 0.9129256641618013, + "grad_norm": 1.4774497747421265, + "learning_rate": 0.000545523307933662, + "loss": 0.4201, + "step": 16340 + }, + { + "epoch": 0.9129815347654832, + "grad_norm": 0.967558741569519, + "learning_rate": 0.0005454952935903182, + "loss": 0.5332, + "step": 16341 + }, + { + "epoch": 0.913037405369165, + "grad_norm": 1.0190798044204712, + "learning_rate": 0.0005454672792469745, + "loss": 0.4226, + "step": 16342 + }, + { + "epoch": 0.9130932759728468, + "grad_norm": 0.5983763933181763, + "learning_rate": 0.0005454392649036307, + "loss": 0.4857, + "step": 16343 + }, + { + "epoch": 0.9131491465765288, + "grad_norm": 0.5348984003067017, + "learning_rate": 0.0005454112505602868, + "loss": 0.4489, + "step": 16344 + }, + { + "epoch": 0.9132050171802106, + "grad_norm": 0.4096996486186981, + "learning_rate": 0.0005453832362169431, + "loss": 0.4215, + "step": 16345 + }, + { + "epoch": 0.9132608877838925, + "grad_norm": 0.3965628743171692, + "learning_rate": 0.0005453552218735993, + "loss": 0.4514, + "step": 16346 + }, + { + "epoch": 0.9133167583875744, + "grad_norm": 0.8100513219833374, + "learning_rate": 0.0005453272075302555, + "loss": 0.3514, + "step": 16347 + }, + { + "epoch": 0.9133726289912563, + "grad_norm": 1.606446623802185, + "learning_rate": 0.0005452991931869116, + "loss": 0.4595, + "step": 16348 + }, + { + "epoch": 0.9134284995949381, + "grad_norm": 0.3691753149032593, + "learning_rate": 0.0005452711788435679, + "loss": 0.4525, + "step": 16349 + }, + { + "epoch": 0.9134843701986199, + "grad_norm": 0.7339809536933899, + "learning_rate": 0.0005452431645002241, + "loss": 0.4033, + "step": 16350 + }, + { + "epoch": 0.9135402408023019, + "grad_norm": 0.7920724749565125, + "learning_rate": 0.0005452151501568802, + "loss": 0.538, + "step": 16351 + }, + { + "epoch": 0.9135961114059837, + "grad_norm": 0.5306304097175598, + "learning_rate": 0.0005451871358135366, + "loss": 0.4758, + "step": 16352 + }, + { + "epoch": 0.9136519820096656, + "grad_norm": 0.38625216484069824, + "learning_rate": 0.0005451591214701928, + "loss": 0.5297, + "step": 16353 + }, + { + "epoch": 0.9137078526133475, + "grad_norm": 2.37540340423584, + "learning_rate": 0.000545131107126849, + "loss": 0.4518, + "step": 16354 + }, + { + "epoch": 0.9137637232170294, + "grad_norm": 0.8976691961288452, + "learning_rate": 0.0005451030927835051, + "loss": 0.4069, + "step": 16355 + }, + { + "epoch": 0.9138195938207112, + "grad_norm": 6.812985897064209, + "learning_rate": 0.0005450750784401614, + "loss": 0.3512, + "step": 16356 + }, + { + "epoch": 0.9138754644243932, + "grad_norm": 0.44661903381347656, + "learning_rate": 0.0005450470640968176, + "loss": 0.4169, + "step": 16357 + }, + { + "epoch": 0.913931335028075, + "grad_norm": 0.40267327427864075, + "learning_rate": 0.0005450190497534738, + "loss": 0.4465, + "step": 16358 + }, + { + "epoch": 0.9139872056317568, + "grad_norm": 0.3900797963142395, + "learning_rate": 0.00054499103541013, + "loss": 0.5354, + "step": 16359 + }, + { + "epoch": 0.9140430762354387, + "grad_norm": 0.6367079615592957, + "learning_rate": 0.0005449630210667862, + "loss": 0.3787, + "step": 16360 + }, + { + "epoch": 0.9140989468391206, + "grad_norm": 0.4503241777420044, + "learning_rate": 0.0005449350067234424, + "loss": 0.3928, + "step": 16361 + }, + { + "epoch": 0.9141548174428025, + "grad_norm": 0.30212682485580444, + "learning_rate": 0.0005449069923800986, + "loss": 0.3257, + "step": 16362 + }, + { + "epoch": 0.9142106880464843, + "grad_norm": 0.38889238238334656, + "learning_rate": 0.0005448789780367548, + "loss": 0.4136, + "step": 16363 + }, + { + "epoch": 0.9142665586501663, + "grad_norm": 0.4472683370113373, + "learning_rate": 0.000544850963693411, + "loss": 0.4451, + "step": 16364 + }, + { + "epoch": 0.9143224292538481, + "grad_norm": 0.514809787273407, + "learning_rate": 0.0005448229493500672, + "loss": 0.486, + "step": 16365 + }, + { + "epoch": 0.9143782998575299, + "grad_norm": 0.40600183606147766, + "learning_rate": 0.0005447949350067236, + "loss": 0.3864, + "step": 16366 + }, + { + "epoch": 0.9144341704612118, + "grad_norm": 3.2774510383605957, + "learning_rate": 0.0005447669206633796, + "loss": 0.4084, + "step": 16367 + }, + { + "epoch": 0.9144900410648937, + "grad_norm": 0.38268372416496277, + "learning_rate": 0.0005447389063200359, + "loss": 0.4061, + "step": 16368 + }, + { + "epoch": 0.9145459116685756, + "grad_norm": 0.5240675806999207, + "learning_rate": 0.0005447108919766921, + "loss": 0.4983, + "step": 16369 + }, + { + "epoch": 0.9146017822722574, + "grad_norm": 0.5835674405097961, + "learning_rate": 0.0005446828776333484, + "loss": 0.3824, + "step": 16370 + }, + { + "epoch": 0.9146576528759394, + "grad_norm": 0.3980778157711029, + "learning_rate": 0.0005446548632900045, + "loss": 0.4039, + "step": 16371 + }, + { + "epoch": 0.9147135234796212, + "grad_norm": 0.884310245513916, + "learning_rate": 0.0005446268489466607, + "loss": 0.6863, + "step": 16372 + }, + { + "epoch": 0.914769394083303, + "grad_norm": 0.8806193470954895, + "learning_rate": 0.000544598834603317, + "loss": 0.4479, + "step": 16373 + }, + { + "epoch": 0.914825264686985, + "grad_norm": 0.33905091881752014, + "learning_rate": 0.0005445708202599731, + "loss": 0.3832, + "step": 16374 + }, + { + "epoch": 0.9148811352906668, + "grad_norm": 0.5678017139434814, + "learning_rate": 0.0005445428059166293, + "loss": 0.4939, + "step": 16375 + }, + { + "epoch": 0.9149370058943487, + "grad_norm": 0.5251827836036682, + "learning_rate": 0.0005445147915732855, + "loss": 0.4259, + "step": 16376 + }, + { + "epoch": 0.9149928764980305, + "grad_norm": 0.4974574148654938, + "learning_rate": 0.0005444867772299418, + "loss": 0.5907, + "step": 16377 + }, + { + "epoch": 0.9150487471017125, + "grad_norm": 0.4571557641029358, + "learning_rate": 0.0005444587628865979, + "loss": 0.351, + "step": 16378 + }, + { + "epoch": 0.9151046177053943, + "grad_norm": 0.39718517661094666, + "learning_rate": 0.0005444307485432541, + "loss": 0.416, + "step": 16379 + }, + { + "epoch": 0.9151604883090761, + "grad_norm": 0.8324798941612244, + "learning_rate": 0.0005444027341999103, + "loss": 0.5323, + "step": 16380 + }, + { + "epoch": 0.9152163589127581, + "grad_norm": 0.3679129481315613, + "learning_rate": 0.0005443747198565666, + "loss": 0.3773, + "step": 16381 + }, + { + "epoch": 0.9152722295164399, + "grad_norm": 0.5576088428497314, + "learning_rate": 0.0005443467055132227, + "loss": 0.5098, + "step": 16382 + }, + { + "epoch": 0.9153281001201218, + "grad_norm": 0.383737176656723, + "learning_rate": 0.0005443186911698789, + "loss": 0.4677, + "step": 16383 + }, + { + "epoch": 0.9153839707238036, + "grad_norm": 1.1476750373840332, + "learning_rate": 0.0005442906768265353, + "loss": 0.4584, + "step": 16384 + }, + { + "epoch": 0.9154398413274856, + "grad_norm": 0.4140738546848297, + "learning_rate": 0.0005442626624831915, + "loss": 0.4181, + "step": 16385 + }, + { + "epoch": 0.9154957119311674, + "grad_norm": 0.5585008859634399, + "learning_rate": 0.0005442346481398476, + "loss": 0.3879, + "step": 16386 + }, + { + "epoch": 0.9155515825348493, + "grad_norm": 0.5126986503601074, + "learning_rate": 0.0005442066337965038, + "loss": 0.4683, + "step": 16387 + }, + { + "epoch": 0.9156074531385312, + "grad_norm": 0.3510649800300598, + "learning_rate": 0.0005441786194531601, + "loss": 0.4085, + "step": 16388 + }, + { + "epoch": 0.915663323742213, + "grad_norm": 0.9253585934638977, + "learning_rate": 0.0005441506051098163, + "loss": 0.3964, + "step": 16389 + }, + { + "epoch": 0.9157191943458949, + "grad_norm": 0.642717182636261, + "learning_rate": 0.0005441225907664724, + "loss": 0.5147, + "step": 16390 + }, + { + "epoch": 0.9157750649495768, + "grad_norm": 0.8008230924606323, + "learning_rate": 0.0005440945764231287, + "loss": 0.4737, + "step": 16391 + }, + { + "epoch": 0.9158309355532587, + "grad_norm": 1.2158892154693604, + "learning_rate": 0.0005440665620797849, + "loss": 0.5125, + "step": 16392 + }, + { + "epoch": 0.9158868061569405, + "grad_norm": 0.455946683883667, + "learning_rate": 0.000544038547736441, + "loss": 0.4417, + "step": 16393 + }, + { + "epoch": 0.9159426767606224, + "grad_norm": 1.4124783277511597, + "learning_rate": 0.0005440105333930972, + "loss": 0.4877, + "step": 16394 + }, + { + "epoch": 0.9159985473643043, + "grad_norm": 0.4254326820373535, + "learning_rate": 0.0005439825190497535, + "loss": 0.5171, + "step": 16395 + }, + { + "epoch": 0.9160544179679861, + "grad_norm": 1.712852954864502, + "learning_rate": 0.0005439545047064097, + "loss": 0.4298, + "step": 16396 + }, + { + "epoch": 0.916110288571668, + "grad_norm": 0.7597902417182922, + "learning_rate": 0.0005439264903630658, + "loss": 0.3774, + "step": 16397 + }, + { + "epoch": 0.9161661591753499, + "grad_norm": 0.7281989455223083, + "learning_rate": 0.0005438984760197221, + "loss": 0.3282, + "step": 16398 + }, + { + "epoch": 0.9162220297790318, + "grad_norm": 0.6501371264457703, + "learning_rate": 0.0005438704616763783, + "loss": 0.4726, + "step": 16399 + }, + { + "epoch": 0.9162779003827136, + "grad_norm": 0.43886521458625793, + "learning_rate": 0.0005438424473330345, + "loss": 0.5389, + "step": 16400 + }, + { + "epoch": 0.9163337709863955, + "grad_norm": 0.6076676249504089, + "learning_rate": 0.0005438144329896906, + "loss": 0.4394, + "step": 16401 + }, + { + "epoch": 0.9163896415900774, + "grad_norm": 0.577519416809082, + "learning_rate": 0.000543786418646347, + "loss": 0.6034, + "step": 16402 + }, + { + "epoch": 0.9164455121937592, + "grad_norm": 0.6613641381263733, + "learning_rate": 0.0005437584043030032, + "loss": 0.5056, + "step": 16403 + }, + { + "epoch": 0.9165013827974411, + "grad_norm": 0.5470593571662903, + "learning_rate": 0.0005437303899596594, + "loss": 0.4256, + "step": 16404 + }, + { + "epoch": 0.916557253401123, + "grad_norm": 0.4614315927028656, + "learning_rate": 0.0005437023756163156, + "loss": 0.4111, + "step": 16405 + }, + { + "epoch": 0.9166131240048049, + "grad_norm": 0.768923282623291, + "learning_rate": 0.0005436743612729718, + "loss": 0.4333, + "step": 16406 + }, + { + "epoch": 0.9166689946084867, + "grad_norm": 0.4516122043132782, + "learning_rate": 0.000543646346929628, + "loss": 0.4865, + "step": 16407 + }, + { + "epoch": 0.9167248652121687, + "grad_norm": 1.6142879724502563, + "learning_rate": 0.0005436183325862842, + "loss": 0.4104, + "step": 16408 + }, + { + "epoch": 0.9167807358158505, + "grad_norm": 1.86721932888031, + "learning_rate": 0.0005435903182429404, + "loss": 0.4279, + "step": 16409 + }, + { + "epoch": 0.9168366064195324, + "grad_norm": 1.1316927671432495, + "learning_rate": 0.0005435623038995966, + "loss": 0.4335, + "step": 16410 + }, + { + "epoch": 0.9168924770232142, + "grad_norm": 0.49073314666748047, + "learning_rate": 0.0005435342895562528, + "loss": 0.4216, + "step": 16411 + }, + { + "epoch": 0.9169483476268961, + "grad_norm": 1.0092583894729614, + "learning_rate": 0.0005435062752129091, + "loss": 0.5079, + "step": 16412 + }, + { + "epoch": 0.917004218230578, + "grad_norm": 0.6172062158584595, + "learning_rate": 0.0005434782608695652, + "loss": 0.4722, + "step": 16413 + }, + { + "epoch": 0.9170600888342598, + "grad_norm": 0.5979495644569397, + "learning_rate": 0.0005434502465262214, + "loss": 0.4498, + "step": 16414 + }, + { + "epoch": 0.9171159594379418, + "grad_norm": 0.45302438735961914, + "learning_rate": 0.0005434222321828776, + "loss": 0.4979, + "step": 16415 + }, + { + "epoch": 0.9171718300416236, + "grad_norm": 0.415734201669693, + "learning_rate": 0.0005433942178395338, + "loss": 0.4463, + "step": 16416 + }, + { + "epoch": 0.9172277006453055, + "grad_norm": 2.88279128074646, + "learning_rate": 0.00054336620349619, + "loss": 0.5971, + "step": 16417 + }, + { + "epoch": 0.9172835712489873, + "grad_norm": 1.7696630954742432, + "learning_rate": 0.0005433381891528463, + "loss": 0.4116, + "step": 16418 + }, + { + "epoch": 0.9173394418526692, + "grad_norm": 0.4204019010066986, + "learning_rate": 0.0005433101748095025, + "loss": 0.4689, + "step": 16419 + }, + { + "epoch": 0.9173953124563511, + "grad_norm": 0.6589786410331726, + "learning_rate": 0.0005432821604661587, + "loss": 0.4909, + "step": 16420 + }, + { + "epoch": 0.9174511830600329, + "grad_norm": 0.5119803547859192, + "learning_rate": 0.0005432541461228149, + "loss": 0.4451, + "step": 16421 + }, + { + "epoch": 0.9175070536637149, + "grad_norm": 0.43899232149124146, + "learning_rate": 0.0005432261317794711, + "loss": 0.3992, + "step": 16422 + }, + { + "epoch": 0.9175629242673967, + "grad_norm": 0.46731624007225037, + "learning_rate": 0.0005431981174361274, + "loss": 0.4025, + "step": 16423 + }, + { + "epoch": 0.9176187948710786, + "grad_norm": 0.6723484396934509, + "learning_rate": 0.0005431701030927835, + "loss": 0.4388, + "step": 16424 + }, + { + "epoch": 0.9176746654747605, + "grad_norm": 2.7450125217437744, + "learning_rate": 0.0005431420887494397, + "loss": 0.578, + "step": 16425 + }, + { + "epoch": 0.9177305360784423, + "grad_norm": 0.555880606174469, + "learning_rate": 0.0005431140744060959, + "loss": 0.476, + "step": 16426 + }, + { + "epoch": 0.9177864066821242, + "grad_norm": 1.2420578002929688, + "learning_rate": 0.0005430860600627522, + "loss": 0.4822, + "step": 16427 + }, + { + "epoch": 0.917842277285806, + "grad_norm": 0.5807508230209351, + "learning_rate": 0.0005430580457194083, + "loss": 0.4223, + "step": 16428 + }, + { + "epoch": 0.917898147889488, + "grad_norm": 0.5539101362228394, + "learning_rate": 0.0005430300313760645, + "loss": 0.5087, + "step": 16429 + }, + { + "epoch": 0.9179540184931698, + "grad_norm": 0.3622729480266571, + "learning_rate": 0.0005430020170327208, + "loss": 0.4047, + "step": 16430 + }, + { + "epoch": 0.9180098890968517, + "grad_norm": 1.0249507427215576, + "learning_rate": 0.000542974002689377, + "loss": 0.401, + "step": 16431 + }, + { + "epoch": 0.9180657597005336, + "grad_norm": 0.8602946400642395, + "learning_rate": 0.0005429459883460331, + "loss": 0.4186, + "step": 16432 + }, + { + "epoch": 0.9181216303042155, + "grad_norm": 0.4983123242855072, + "learning_rate": 0.0005429179740026893, + "loss": 0.3879, + "step": 16433 + }, + { + "epoch": 0.9181775009078973, + "grad_norm": 0.6619552373886108, + "learning_rate": 0.0005428899596593457, + "loss": 0.4027, + "step": 16434 + }, + { + "epoch": 0.9182333715115791, + "grad_norm": 1.117605209350586, + "learning_rate": 0.0005428619453160019, + "loss": 0.444, + "step": 16435 + }, + { + "epoch": 0.9182892421152611, + "grad_norm": 0.5029239654541016, + "learning_rate": 0.000542833930972658, + "loss": 0.4249, + "step": 16436 + }, + { + "epoch": 0.9183451127189429, + "grad_norm": 0.904411256313324, + "learning_rate": 0.0005428059166293143, + "loss": 0.341, + "step": 16437 + }, + { + "epoch": 0.9184009833226248, + "grad_norm": 0.7440609335899353, + "learning_rate": 0.0005427779022859705, + "loss": 0.4667, + "step": 16438 + }, + { + "epoch": 0.9184568539263067, + "grad_norm": 0.5782729983329773, + "learning_rate": 0.0005427498879426266, + "loss": 0.4985, + "step": 16439 + }, + { + "epoch": 0.9185127245299886, + "grad_norm": 0.42581868171691895, + "learning_rate": 0.0005427218735992828, + "loss": 0.4138, + "step": 16440 + }, + { + "epoch": 0.9185685951336704, + "grad_norm": 0.5854861736297607, + "learning_rate": 0.0005426938592559391, + "loss": 0.3729, + "step": 16441 + }, + { + "epoch": 0.9186244657373523, + "grad_norm": 0.4443536400794983, + "learning_rate": 0.0005426658449125953, + "loss": 0.3627, + "step": 16442 + }, + { + "epoch": 0.9186803363410342, + "grad_norm": 0.3530850112438202, + "learning_rate": 0.0005426378305692514, + "loss": 0.3374, + "step": 16443 + }, + { + "epoch": 0.918736206944716, + "grad_norm": 0.4050285220146179, + "learning_rate": 0.0005426098162259077, + "loss": 0.3716, + "step": 16444 + }, + { + "epoch": 0.9187920775483979, + "grad_norm": 0.4311608374118805, + "learning_rate": 0.0005425818018825639, + "loss": 0.3877, + "step": 16445 + }, + { + "epoch": 0.9188479481520798, + "grad_norm": 1.3905998468399048, + "learning_rate": 0.0005425537875392201, + "loss": 0.4066, + "step": 16446 + }, + { + "epoch": 0.9189038187557617, + "grad_norm": 0.8597705364227295, + "learning_rate": 0.0005425257731958762, + "loss": 0.4272, + "step": 16447 + }, + { + "epoch": 0.9189596893594435, + "grad_norm": 0.4125775992870331, + "learning_rate": 0.0005424977588525325, + "loss": 0.4212, + "step": 16448 + }, + { + "epoch": 0.9190155599631255, + "grad_norm": 0.38239866495132446, + "learning_rate": 0.0005424697445091887, + "loss": 0.3262, + "step": 16449 + }, + { + "epoch": 0.9190714305668073, + "grad_norm": 0.6901271343231201, + "learning_rate": 0.000542441730165845, + "loss": 0.3841, + "step": 16450 + }, + { + "epoch": 0.9191273011704891, + "grad_norm": 1.8323049545288086, + "learning_rate": 0.000542413715822501, + "loss": 0.4428, + "step": 16451 + }, + { + "epoch": 0.919183171774171, + "grad_norm": 1.4580286741256714, + "learning_rate": 0.0005423857014791574, + "loss": 0.4992, + "step": 16452 + }, + { + "epoch": 0.9192390423778529, + "grad_norm": 0.45149216055870056, + "learning_rate": 0.0005423576871358136, + "loss": 0.374, + "step": 16453 + }, + { + "epoch": 0.9192949129815348, + "grad_norm": 0.56484055519104, + "learning_rate": 0.0005423296727924698, + "loss": 0.4243, + "step": 16454 + }, + { + "epoch": 0.9193507835852166, + "grad_norm": 1.6142446994781494, + "learning_rate": 0.000542301658449126, + "loss": 0.3544, + "step": 16455 + }, + { + "epoch": 0.9194066541888986, + "grad_norm": 0.5852084159851074, + "learning_rate": 0.0005422736441057822, + "loss": 0.4487, + "step": 16456 + }, + { + "epoch": 0.9194625247925804, + "grad_norm": 1.5041674375534058, + "learning_rate": 0.0005422456297624384, + "loss": 0.4946, + "step": 16457 + }, + { + "epoch": 0.9195183953962622, + "grad_norm": 0.4737708568572998, + "learning_rate": 0.0005422176154190945, + "loss": 0.4703, + "step": 16458 + }, + { + "epoch": 0.9195742659999442, + "grad_norm": 0.6187135577201843, + "learning_rate": 0.0005421896010757508, + "loss": 0.4061, + "step": 16459 + }, + { + "epoch": 0.919630136603626, + "grad_norm": 0.487163245677948, + "learning_rate": 0.000542161586732407, + "loss": 0.5333, + "step": 16460 + }, + { + "epoch": 0.9196860072073079, + "grad_norm": 0.403615266084671, + "learning_rate": 0.0005421335723890632, + "loss": 0.3617, + "step": 16461 + }, + { + "epoch": 0.9197418778109897, + "grad_norm": 0.4758142828941345, + "learning_rate": 0.0005421055580457194, + "loss": 0.4522, + "step": 16462 + }, + { + "epoch": 0.9197977484146717, + "grad_norm": 0.5798641443252563, + "learning_rate": 0.0005420775437023756, + "loss": 0.512, + "step": 16463 + }, + { + "epoch": 0.9198536190183535, + "grad_norm": 1.2298184633255005, + "learning_rate": 0.0005420495293590318, + "loss": 0.4438, + "step": 16464 + }, + { + "epoch": 0.9199094896220353, + "grad_norm": 0.6741176843643188, + "learning_rate": 0.000542021515015688, + "loss": 0.4579, + "step": 16465 + }, + { + "epoch": 0.9199653602257173, + "grad_norm": 0.37432751059532166, + "learning_rate": 0.0005419935006723442, + "loss": 0.3875, + "step": 16466 + }, + { + "epoch": 0.9200212308293991, + "grad_norm": 0.40795472264289856, + "learning_rate": 0.0005419654863290004, + "loss": 0.368, + "step": 16467 + }, + { + "epoch": 0.920077101433081, + "grad_norm": 0.6980853080749512, + "learning_rate": 0.0005419374719856567, + "loss": 0.3661, + "step": 16468 + }, + { + "epoch": 0.9201329720367628, + "grad_norm": 0.5021161437034607, + "learning_rate": 0.000541909457642313, + "loss": 0.3704, + "step": 16469 + }, + { + "epoch": 0.9201888426404448, + "grad_norm": 1.4646373987197876, + "learning_rate": 0.0005418814432989691, + "loss": 0.7058, + "step": 16470 + }, + { + "epoch": 0.9202447132441266, + "grad_norm": 0.40842509269714355, + "learning_rate": 0.0005418534289556253, + "loss": 0.389, + "step": 16471 + }, + { + "epoch": 0.9203005838478084, + "grad_norm": 0.40044689178466797, + "learning_rate": 0.0005418254146122815, + "loss": 0.4714, + "step": 16472 + }, + { + "epoch": 0.9203564544514904, + "grad_norm": 1.0619627237319946, + "learning_rate": 0.0005417974002689378, + "loss": 0.3965, + "step": 16473 + }, + { + "epoch": 0.9204123250551722, + "grad_norm": 0.5522732138633728, + "learning_rate": 0.0005417693859255939, + "loss": 0.3583, + "step": 16474 + }, + { + "epoch": 0.9204681956588541, + "grad_norm": 0.5502511262893677, + "learning_rate": 0.0005417413715822501, + "loss": 0.3814, + "step": 16475 + }, + { + "epoch": 0.9205240662625359, + "grad_norm": 0.38499927520751953, + "learning_rate": 0.0005417133572389064, + "loss": 0.4365, + "step": 16476 + }, + { + "epoch": 0.9205799368662179, + "grad_norm": 0.7272552251815796, + "learning_rate": 0.0005416853428955626, + "loss": 0.3831, + "step": 16477 + }, + { + "epoch": 0.9206358074698997, + "grad_norm": 0.7598942518234253, + "learning_rate": 0.0005416573285522187, + "loss": 0.4557, + "step": 16478 + }, + { + "epoch": 0.9206916780735815, + "grad_norm": 0.4187074303627014, + "learning_rate": 0.0005416293142088749, + "loss": 0.4208, + "step": 16479 + }, + { + "epoch": 0.9207475486772635, + "grad_norm": 0.47162166237831116, + "learning_rate": 0.0005416012998655312, + "loss": 0.5303, + "step": 16480 + }, + { + "epoch": 0.9208034192809453, + "grad_norm": 0.35181987285614014, + "learning_rate": 0.0005415732855221873, + "loss": 0.47, + "step": 16481 + }, + { + "epoch": 0.9208592898846272, + "grad_norm": 0.5237725377082825, + "learning_rate": 0.0005415452711788435, + "loss": 0.5859, + "step": 16482 + }, + { + "epoch": 0.9209151604883091, + "grad_norm": 1.050572395324707, + "learning_rate": 0.0005415172568354998, + "loss": 0.3141, + "step": 16483 + }, + { + "epoch": 0.920971031091991, + "grad_norm": 0.5289076566696167, + "learning_rate": 0.000541489242492156, + "loss": 0.4479, + "step": 16484 + }, + { + "epoch": 0.9210269016956728, + "grad_norm": 0.8035727739334106, + "learning_rate": 0.0005414612281488122, + "loss": 0.4139, + "step": 16485 + }, + { + "epoch": 0.9210827722993546, + "grad_norm": 0.8388510346412659, + "learning_rate": 0.0005414332138054684, + "loss": 0.5724, + "step": 16486 + }, + { + "epoch": 0.9211386429030366, + "grad_norm": 2.2555930614471436, + "learning_rate": 0.0005414051994621247, + "loss": 0.472, + "step": 16487 + }, + { + "epoch": 0.9211945135067184, + "grad_norm": 0.8609325289726257, + "learning_rate": 0.0005413771851187809, + "loss": 0.3998, + "step": 16488 + }, + { + "epoch": 0.9212503841104003, + "grad_norm": 0.5394794940948486, + "learning_rate": 0.000541349170775437, + "loss": 0.6127, + "step": 16489 + }, + { + "epoch": 0.9213062547140822, + "grad_norm": 0.5843639373779297, + "learning_rate": 0.0005413211564320932, + "loss": 0.3985, + "step": 16490 + }, + { + "epoch": 0.9213621253177641, + "grad_norm": 0.49711543321609497, + "learning_rate": 0.0005412931420887495, + "loss": 0.4632, + "step": 16491 + }, + { + "epoch": 0.9214179959214459, + "grad_norm": 0.5771260261535645, + "learning_rate": 0.0005412651277454057, + "loss": 0.5619, + "step": 16492 + }, + { + "epoch": 0.9214738665251277, + "grad_norm": 0.47584566473960876, + "learning_rate": 0.0005412371134020618, + "loss": 0.3718, + "step": 16493 + }, + { + "epoch": 0.9215297371288097, + "grad_norm": 0.4877389669418335, + "learning_rate": 0.0005412090990587181, + "loss": 0.4167, + "step": 16494 + }, + { + "epoch": 0.9215856077324915, + "grad_norm": 0.3583289384841919, + "learning_rate": 0.0005411810847153743, + "loss": 0.3913, + "step": 16495 + }, + { + "epoch": 0.9216414783361734, + "grad_norm": 0.35816431045532227, + "learning_rate": 0.0005411530703720305, + "loss": 0.4078, + "step": 16496 + }, + { + "epoch": 0.9216973489398553, + "grad_norm": 0.4589630365371704, + "learning_rate": 0.0005411250560286866, + "loss": 0.4565, + "step": 16497 + }, + { + "epoch": 0.9217532195435372, + "grad_norm": 0.39870455861091614, + "learning_rate": 0.0005410970416853429, + "loss": 0.2978, + "step": 16498 + }, + { + "epoch": 0.921809090147219, + "grad_norm": 1.3162717819213867, + "learning_rate": 0.0005410690273419991, + "loss": 0.5442, + "step": 16499 + }, + { + "epoch": 0.921864960750901, + "grad_norm": 0.4703967273235321, + "learning_rate": 0.0005410410129986552, + "loss": 0.3988, + "step": 16500 + }, + { + "epoch": 0.921864960750901, + "eval_cer": 0.08798402670943668, + "eval_loss": 0.3339237868785858, + "eval_runtime": 56.5595, + "eval_samples_per_second": 80.234, + "eval_steps_per_second": 5.021, + "eval_wer": 0.3467780229602525, + "step": 16500 + }, + { + "epoch": 0.9219208313545828, + "grad_norm": 0.934485673904419, + "learning_rate": 0.0005410129986553116, + "loss": 0.3384, + "step": 16501 + }, + { + "epoch": 0.9219767019582646, + "grad_norm": 0.534551739692688, + "learning_rate": 0.0005409849843119678, + "loss": 0.3611, + "step": 16502 + }, + { + "epoch": 0.9220325725619465, + "grad_norm": 0.3441932201385498, + "learning_rate": 0.000540956969968624, + "loss": 0.3995, + "step": 16503 + }, + { + "epoch": 0.9220884431656284, + "grad_norm": 0.644707441329956, + "learning_rate": 0.0005409289556252801, + "loss": 0.3635, + "step": 16504 + }, + { + "epoch": 0.9221443137693103, + "grad_norm": 0.5123574137687683, + "learning_rate": 0.0005409009412819364, + "loss": 0.4119, + "step": 16505 + }, + { + "epoch": 0.9222001843729921, + "grad_norm": 0.8041903972625732, + "learning_rate": 0.0005408729269385926, + "loss": 0.4728, + "step": 16506 + }, + { + "epoch": 0.9222560549766741, + "grad_norm": 0.42784059047698975, + "learning_rate": 0.0005408449125952488, + "loss": 0.3716, + "step": 16507 + }, + { + "epoch": 0.9223119255803559, + "grad_norm": 0.40506497025489807, + "learning_rate": 0.000540816898251905, + "loss": 0.4703, + "step": 16508 + }, + { + "epoch": 0.9223677961840377, + "grad_norm": 0.48493054509162903, + "learning_rate": 0.0005407888839085612, + "loss": 0.4904, + "step": 16509 + }, + { + "epoch": 0.9224236667877196, + "grad_norm": 0.47927701473236084, + "learning_rate": 0.0005407608695652174, + "loss": 0.4721, + "step": 16510 + }, + { + "epoch": 0.9224795373914015, + "grad_norm": 0.3778323531150818, + "learning_rate": 0.0005407328552218736, + "loss": 0.3461, + "step": 16511 + }, + { + "epoch": 0.9225354079950834, + "grad_norm": 0.39428791403770447, + "learning_rate": 0.0005407048408785298, + "loss": 0.391, + "step": 16512 + }, + { + "epoch": 0.9225912785987652, + "grad_norm": 0.37731269001960754, + "learning_rate": 0.000540676826535186, + "loss": 0.386, + "step": 16513 + }, + { + "epoch": 0.9226471492024472, + "grad_norm": 0.3969413638114929, + "learning_rate": 0.0005406488121918422, + "loss": 0.3781, + "step": 16514 + }, + { + "epoch": 0.922703019806129, + "grad_norm": 0.43767204880714417, + "learning_rate": 0.0005406207978484985, + "loss": 0.3631, + "step": 16515 + }, + { + "epoch": 0.9227588904098109, + "grad_norm": 0.5744378566741943, + "learning_rate": 0.0005405927835051546, + "loss": 0.4646, + "step": 16516 + }, + { + "epoch": 0.9228147610134928, + "grad_norm": 0.6505669951438904, + "learning_rate": 0.0005405647691618108, + "loss": 0.474, + "step": 16517 + }, + { + "epoch": 0.9228706316171746, + "grad_norm": 0.5552017092704773, + "learning_rate": 0.000540536754818467, + "loss": 0.4283, + "step": 16518 + }, + { + "epoch": 0.9229265022208565, + "grad_norm": 0.45205414295196533, + "learning_rate": 0.0005405087404751234, + "loss": 0.4341, + "step": 16519 + }, + { + "epoch": 0.9229823728245383, + "grad_norm": 0.38347524404525757, + "learning_rate": 0.0005404807261317795, + "loss": 0.435, + "step": 16520 + }, + { + "epoch": 0.9230382434282203, + "grad_norm": 0.43003198504447937, + "learning_rate": 0.0005404527117884357, + "loss": 0.4097, + "step": 16521 + }, + { + "epoch": 0.9230941140319021, + "grad_norm": 0.40923863649368286, + "learning_rate": 0.0005404246974450919, + "loss": 0.4834, + "step": 16522 + }, + { + "epoch": 0.923149984635584, + "grad_norm": 0.6301288604736328, + "learning_rate": 0.0005403966831017481, + "loss": 0.3787, + "step": 16523 + }, + { + "epoch": 0.9232058552392659, + "grad_norm": 0.5871738791465759, + "learning_rate": 0.0005403686687584043, + "loss": 0.4615, + "step": 16524 + }, + { + "epoch": 0.9232617258429477, + "grad_norm": 0.7243672609329224, + "learning_rate": 0.0005403406544150605, + "loss": 0.5933, + "step": 16525 + }, + { + "epoch": 0.9233175964466296, + "grad_norm": 0.43270525336265564, + "learning_rate": 0.0005403126400717168, + "loss": 0.4956, + "step": 16526 + }, + { + "epoch": 0.9233734670503114, + "grad_norm": 0.7798130512237549, + "learning_rate": 0.0005402846257283729, + "loss": 0.3844, + "step": 16527 + }, + { + "epoch": 0.9234293376539934, + "grad_norm": 0.5117434859275818, + "learning_rate": 0.0005402566113850291, + "loss": 0.5813, + "step": 16528 + }, + { + "epoch": 0.9234852082576752, + "grad_norm": 0.6334757208824158, + "learning_rate": 0.0005402285970416853, + "loss": 0.4019, + "step": 16529 + }, + { + "epoch": 0.9235410788613571, + "grad_norm": 0.4222647547721863, + "learning_rate": 0.0005402005826983416, + "loss": 0.5222, + "step": 16530 + }, + { + "epoch": 0.923596949465039, + "grad_norm": 0.8255357146263123, + "learning_rate": 0.0005401725683549977, + "loss": 0.4158, + "step": 16531 + }, + { + "epoch": 0.9236528200687208, + "grad_norm": 0.6738642454147339, + "learning_rate": 0.0005401445540116539, + "loss": 0.5044, + "step": 16532 + }, + { + "epoch": 0.9237086906724027, + "grad_norm": 0.7056590914726257, + "learning_rate": 0.0005401165396683102, + "loss": 0.487, + "step": 16533 + }, + { + "epoch": 0.9237645612760846, + "grad_norm": 0.9877554774284363, + "learning_rate": 0.0005400885253249665, + "loss": 0.6157, + "step": 16534 + }, + { + "epoch": 0.9238204318797665, + "grad_norm": 0.487841933965683, + "learning_rate": 0.0005400605109816226, + "loss": 0.4914, + "step": 16535 + }, + { + "epoch": 0.9238763024834483, + "grad_norm": 0.5482023358345032, + "learning_rate": 0.0005400324966382788, + "loss": 0.4975, + "step": 16536 + }, + { + "epoch": 0.9239321730871302, + "grad_norm": 0.40906184911727905, + "learning_rate": 0.0005400044822949351, + "loss": 0.4396, + "step": 16537 + }, + { + "epoch": 0.9239880436908121, + "grad_norm": 0.4947115480899811, + "learning_rate": 0.0005399764679515913, + "loss": 0.4778, + "step": 16538 + }, + { + "epoch": 0.924043914294494, + "grad_norm": 0.8987926244735718, + "learning_rate": 0.0005399484536082474, + "loss": 0.4101, + "step": 16539 + }, + { + "epoch": 0.9240997848981758, + "grad_norm": 0.466025710105896, + "learning_rate": 0.0005399204392649037, + "loss": 0.4693, + "step": 16540 + }, + { + "epoch": 0.9241556555018577, + "grad_norm": 1.299497127532959, + "learning_rate": 0.0005398924249215599, + "loss": 0.4533, + "step": 16541 + }, + { + "epoch": 0.9242115261055396, + "grad_norm": 0.48243850469589233, + "learning_rate": 0.000539864410578216, + "loss": 0.4089, + "step": 16542 + }, + { + "epoch": 0.9242673967092214, + "grad_norm": 0.5779135227203369, + "learning_rate": 0.0005398363962348722, + "loss": 0.5944, + "step": 16543 + }, + { + "epoch": 0.9243232673129033, + "grad_norm": 0.453779935836792, + "learning_rate": 0.0005398083818915285, + "loss": 0.4885, + "step": 16544 + }, + { + "epoch": 0.9243791379165852, + "grad_norm": 0.47792288661003113, + "learning_rate": 0.0005397803675481847, + "loss": 0.3732, + "step": 16545 + }, + { + "epoch": 0.924435008520267, + "grad_norm": 0.6214803457260132, + "learning_rate": 0.0005397523532048408, + "loss": 0.5715, + "step": 16546 + }, + { + "epoch": 0.9244908791239489, + "grad_norm": 0.48363950848579407, + "learning_rate": 0.0005397243388614971, + "loss": 0.4452, + "step": 16547 + }, + { + "epoch": 0.9245467497276308, + "grad_norm": 1.3722056150436401, + "learning_rate": 0.0005396963245181533, + "loss": 0.4118, + "step": 16548 + }, + { + "epoch": 0.9246026203313127, + "grad_norm": 0.4400820732116699, + "learning_rate": 0.0005396683101748095, + "loss": 0.4167, + "step": 16549 + }, + { + "epoch": 0.9246584909349945, + "grad_norm": 1.0675668716430664, + "learning_rate": 0.0005396402958314656, + "loss": 0.4715, + "step": 16550 + }, + { + "epoch": 0.9247143615386765, + "grad_norm": 0.41022399067878723, + "learning_rate": 0.000539612281488122, + "loss": 0.4145, + "step": 16551 + }, + { + "epoch": 0.9247702321423583, + "grad_norm": 0.5051771998405457, + "learning_rate": 0.0005395842671447782, + "loss": 0.3656, + "step": 16552 + }, + { + "epoch": 0.9248261027460402, + "grad_norm": 1.350921869277954, + "learning_rate": 0.0005395562528014344, + "loss": 0.4254, + "step": 16553 + }, + { + "epoch": 0.924881973349722, + "grad_norm": 0.46366000175476074, + "learning_rate": 0.0005395282384580905, + "loss": 0.4842, + "step": 16554 + }, + { + "epoch": 0.924937843953404, + "grad_norm": 3.28879714012146, + "learning_rate": 0.0005395002241147468, + "loss": 0.4924, + "step": 16555 + }, + { + "epoch": 0.9249937145570858, + "grad_norm": 0.8653109073638916, + "learning_rate": 0.000539472209771403, + "loss": 0.5912, + "step": 16556 + }, + { + "epoch": 0.9250495851607676, + "grad_norm": 0.9015624523162842, + "learning_rate": 0.0005394441954280592, + "loss": 0.5658, + "step": 16557 + }, + { + "epoch": 0.9251054557644496, + "grad_norm": 0.4043443202972412, + "learning_rate": 0.0005394161810847154, + "loss": 0.3696, + "step": 16558 + }, + { + "epoch": 0.9251613263681314, + "grad_norm": 0.41734349727630615, + "learning_rate": 0.0005393881667413716, + "loss": 0.4244, + "step": 16559 + }, + { + "epoch": 0.9252171969718133, + "grad_norm": 0.4197598993778229, + "learning_rate": 0.0005393601523980278, + "loss": 0.4216, + "step": 16560 + }, + { + "epoch": 0.9252730675754951, + "grad_norm": 1.0665857791900635, + "learning_rate": 0.000539332138054684, + "loss": 0.5798, + "step": 16561 + }, + { + "epoch": 0.925328938179177, + "grad_norm": 0.40224412083625793, + "learning_rate": 0.0005393041237113402, + "loss": 0.407, + "step": 16562 + }, + { + "epoch": 0.9253848087828589, + "grad_norm": 0.4520391821861267, + "learning_rate": 0.0005392761093679964, + "loss": 0.3873, + "step": 16563 + }, + { + "epoch": 0.9254406793865407, + "grad_norm": 0.8155240416526794, + "learning_rate": 0.0005392480950246526, + "loss": 0.4359, + "step": 16564 + }, + { + "epoch": 0.9254965499902227, + "grad_norm": 0.4967643916606903, + "learning_rate": 0.0005392200806813088, + "loss": 0.4846, + "step": 16565 + }, + { + "epoch": 0.9255524205939045, + "grad_norm": 1.1532496213912964, + "learning_rate": 0.000539192066337965, + "loss": 0.5198, + "step": 16566 + }, + { + "epoch": 0.9256082911975864, + "grad_norm": 0.49225661158561707, + "learning_rate": 0.0005391640519946212, + "loss": 0.5427, + "step": 16567 + }, + { + "epoch": 0.9256641618012683, + "grad_norm": 0.5488972663879395, + "learning_rate": 0.0005391360376512775, + "loss": 0.4601, + "step": 16568 + }, + { + "epoch": 0.9257200324049502, + "grad_norm": 0.5888493061065674, + "learning_rate": 0.0005391080233079337, + "loss": 0.4289, + "step": 16569 + }, + { + "epoch": 0.925775903008632, + "grad_norm": 0.44796037673950195, + "learning_rate": 0.0005390800089645899, + "loss": 0.4031, + "step": 16570 + }, + { + "epoch": 0.9258317736123138, + "grad_norm": 0.972216010093689, + "learning_rate": 0.0005390519946212461, + "loss": 0.498, + "step": 16571 + }, + { + "epoch": 0.9258876442159958, + "grad_norm": 0.4864712357521057, + "learning_rate": 0.0005390239802779024, + "loss": 0.4194, + "step": 16572 + }, + { + "epoch": 0.9259435148196776, + "grad_norm": 0.4264053404331207, + "learning_rate": 0.0005389959659345585, + "loss": 0.429, + "step": 16573 + }, + { + "epoch": 0.9259993854233595, + "grad_norm": 0.46359604597091675, + "learning_rate": 0.0005389679515912147, + "loss": 0.4402, + "step": 16574 + }, + { + "epoch": 0.9260552560270414, + "grad_norm": 0.44144386053085327, + "learning_rate": 0.0005389399372478709, + "loss": 0.5452, + "step": 16575 + }, + { + "epoch": 0.9261111266307233, + "grad_norm": 0.4894416928291321, + "learning_rate": 0.0005389119229045272, + "loss": 0.4661, + "step": 16576 + }, + { + "epoch": 0.9261669972344051, + "grad_norm": 0.4638812243938446, + "learning_rate": 0.0005388839085611833, + "loss": 0.4657, + "step": 16577 + }, + { + "epoch": 0.9262228678380869, + "grad_norm": 0.5452649593353271, + "learning_rate": 0.0005388558942178395, + "loss": 0.5339, + "step": 16578 + }, + { + "epoch": 0.9262787384417689, + "grad_norm": 0.5960823893547058, + "learning_rate": 0.0005388278798744958, + "loss": 0.5733, + "step": 16579 + }, + { + "epoch": 0.9263346090454507, + "grad_norm": 0.49958762526512146, + "learning_rate": 0.000538799865531152, + "loss": 0.5055, + "step": 16580 + }, + { + "epoch": 0.9263904796491326, + "grad_norm": 0.49632006883621216, + "learning_rate": 0.0005387718511878081, + "loss": 0.4561, + "step": 16581 + }, + { + "epoch": 0.9264463502528145, + "grad_norm": 0.4054170250892639, + "learning_rate": 0.0005387438368444643, + "loss": 0.4451, + "step": 16582 + }, + { + "epoch": 0.9265022208564964, + "grad_norm": 0.6269485354423523, + "learning_rate": 0.0005387158225011206, + "loss": 0.3471, + "step": 16583 + }, + { + "epoch": 0.9265580914601782, + "grad_norm": 0.5846990346908569, + "learning_rate": 0.0005386878081577767, + "loss": 0.4153, + "step": 16584 + }, + { + "epoch": 0.9266139620638602, + "grad_norm": 1.1454466581344604, + "learning_rate": 0.000538659793814433, + "loss": 0.3735, + "step": 16585 + }, + { + "epoch": 0.926669832667542, + "grad_norm": 0.44073715806007385, + "learning_rate": 0.0005386317794710893, + "loss": 0.5573, + "step": 16586 + }, + { + "epoch": 0.9267257032712238, + "grad_norm": 0.579658031463623, + "learning_rate": 0.0005386037651277455, + "loss": 0.528, + "step": 16587 + }, + { + "epoch": 0.9267815738749057, + "grad_norm": 0.39550480246543884, + "learning_rate": 0.0005385757507844016, + "loss": 0.3743, + "step": 16588 + }, + { + "epoch": 0.9268374444785876, + "grad_norm": 0.5488869547843933, + "learning_rate": 0.0005385477364410578, + "loss": 0.5799, + "step": 16589 + }, + { + "epoch": 0.9268933150822695, + "grad_norm": 0.3605436384677887, + "learning_rate": 0.0005385197220977141, + "loss": 0.2986, + "step": 16590 + }, + { + "epoch": 0.9269491856859513, + "grad_norm": 0.3895554840564728, + "learning_rate": 0.0005384917077543703, + "loss": 0.37, + "step": 16591 + }, + { + "epoch": 0.9270050562896333, + "grad_norm": 0.38902121782302856, + "learning_rate": 0.0005384636934110264, + "loss": 0.4847, + "step": 16592 + }, + { + "epoch": 0.9270609268933151, + "grad_norm": 0.7825896739959717, + "learning_rate": 0.0005384356790676826, + "loss": 0.4557, + "step": 16593 + }, + { + "epoch": 0.9271167974969969, + "grad_norm": 0.391802579164505, + "learning_rate": 0.0005384076647243389, + "loss": 0.4139, + "step": 16594 + }, + { + "epoch": 0.9271726681006788, + "grad_norm": 0.5482645630836487, + "learning_rate": 0.0005383796503809951, + "loss": 0.5262, + "step": 16595 + }, + { + "epoch": 0.9272285387043607, + "grad_norm": 0.6528567671775818, + "learning_rate": 0.0005383516360376512, + "loss": 0.4333, + "step": 16596 + }, + { + "epoch": 0.9272844093080426, + "grad_norm": 0.5100598931312561, + "learning_rate": 0.0005383236216943075, + "loss": 0.3474, + "step": 16597 + }, + { + "epoch": 0.9273402799117244, + "grad_norm": 0.48973584175109863, + "learning_rate": 0.0005382956073509637, + "loss": 0.4476, + "step": 16598 + }, + { + "epoch": 0.9273961505154064, + "grad_norm": 0.501772940158844, + "learning_rate": 0.0005382675930076199, + "loss": 0.596, + "step": 16599 + }, + { + "epoch": 0.9274520211190882, + "grad_norm": 0.7230068445205688, + "learning_rate": 0.000538239578664276, + "loss": 0.395, + "step": 16600 + }, + { + "epoch": 0.92750789172277, + "grad_norm": 1.4617596864700317, + "learning_rate": 0.0005382115643209324, + "loss": 0.503, + "step": 16601 + }, + { + "epoch": 0.927563762326452, + "grad_norm": 0.6529628038406372, + "learning_rate": 0.0005381835499775886, + "loss": 0.6158, + "step": 16602 + }, + { + "epoch": 0.9276196329301338, + "grad_norm": 0.5286274552345276, + "learning_rate": 0.0005381555356342448, + "loss": 0.4244, + "step": 16603 + }, + { + "epoch": 0.9276755035338157, + "grad_norm": 0.5403916239738464, + "learning_rate": 0.000538127521290901, + "loss": 0.5439, + "step": 16604 + }, + { + "epoch": 0.9277313741374975, + "grad_norm": 0.630158007144928, + "learning_rate": 0.0005380995069475572, + "loss": 0.4584, + "step": 16605 + }, + { + "epoch": 0.9277872447411795, + "grad_norm": 1.1748936176300049, + "learning_rate": 0.0005380714926042134, + "loss": 0.5495, + "step": 16606 + }, + { + "epoch": 0.9278431153448613, + "grad_norm": 0.3907076418399811, + "learning_rate": 0.0005380434782608695, + "loss": 0.4057, + "step": 16607 + }, + { + "epoch": 0.9278989859485431, + "grad_norm": 0.6038150191307068, + "learning_rate": 0.0005380154639175258, + "loss": 0.5323, + "step": 16608 + }, + { + "epoch": 0.9279548565522251, + "grad_norm": 0.7126844525337219, + "learning_rate": 0.000537987449574182, + "loss": 0.3404, + "step": 16609 + }, + { + "epoch": 0.9280107271559069, + "grad_norm": 0.3775217831134796, + "learning_rate": 0.0005379594352308382, + "loss": 0.3505, + "step": 16610 + }, + { + "epoch": 0.9280665977595888, + "grad_norm": 0.44356486201286316, + "learning_rate": 0.0005379314208874944, + "loss": 0.3547, + "step": 16611 + }, + { + "epoch": 0.9281224683632706, + "grad_norm": 0.42825251817703247, + "learning_rate": 0.0005379034065441506, + "loss": 0.387, + "step": 16612 + }, + { + "epoch": 0.9281783389669526, + "grad_norm": 0.5000155568122864, + "learning_rate": 0.0005378753922008068, + "loss": 0.4472, + "step": 16613 + }, + { + "epoch": 0.9282342095706344, + "grad_norm": 0.4745391607284546, + "learning_rate": 0.000537847377857463, + "loss": 0.4456, + "step": 16614 + }, + { + "epoch": 0.9282900801743162, + "grad_norm": 0.32537177205085754, + "learning_rate": 0.0005378193635141192, + "loss": 0.3365, + "step": 16615 + }, + { + "epoch": 0.9283459507779982, + "grad_norm": 0.38380712270736694, + "learning_rate": 0.0005377913491707754, + "loss": 0.357, + "step": 16616 + }, + { + "epoch": 0.92840182138168, + "grad_norm": 0.3907821476459503, + "learning_rate": 0.0005377633348274316, + "loss": 0.3859, + "step": 16617 + }, + { + "epoch": 0.9284576919853619, + "grad_norm": 1.171518325805664, + "learning_rate": 0.000537735320484088, + "loss": 0.6605, + "step": 16618 + }, + { + "epoch": 0.9285135625890438, + "grad_norm": 0.6451008319854736, + "learning_rate": 0.0005377073061407441, + "loss": 0.5185, + "step": 16619 + }, + { + "epoch": 0.9285694331927257, + "grad_norm": 1.1153432130813599, + "learning_rate": 0.0005376792917974003, + "loss": 0.3305, + "step": 16620 + }, + { + "epoch": 0.9286253037964075, + "grad_norm": 0.9468196034431458, + "learning_rate": 0.0005376512774540565, + "loss": 0.4104, + "step": 16621 + }, + { + "epoch": 0.9286811744000893, + "grad_norm": 1.2126922607421875, + "learning_rate": 0.0005376232631107128, + "loss": 0.3208, + "step": 16622 + }, + { + "epoch": 0.9287370450037713, + "grad_norm": 1.0483343601226807, + "learning_rate": 0.0005375952487673689, + "loss": 0.4038, + "step": 16623 + }, + { + "epoch": 0.9287929156074531, + "grad_norm": 0.4782596826553345, + "learning_rate": 0.0005375672344240251, + "loss": 0.4117, + "step": 16624 + }, + { + "epoch": 0.928848786211135, + "grad_norm": 0.8571441769599915, + "learning_rate": 0.0005375392200806814, + "loss": 0.4356, + "step": 16625 + }, + { + "epoch": 0.9289046568148169, + "grad_norm": 0.5105167627334595, + "learning_rate": 0.0005375112057373376, + "loss": 0.4655, + "step": 16626 + }, + { + "epoch": 0.9289605274184988, + "grad_norm": 1.7337969541549683, + "learning_rate": 0.0005374831913939937, + "loss": 0.44, + "step": 16627 + }, + { + "epoch": 0.9290163980221806, + "grad_norm": 0.6504356861114502, + "learning_rate": 0.0005374551770506499, + "loss": 0.4984, + "step": 16628 + }, + { + "epoch": 0.9290722686258625, + "grad_norm": 0.5243789553642273, + "learning_rate": 0.0005374271627073062, + "loss": 0.3655, + "step": 16629 + }, + { + "epoch": 0.9291281392295444, + "grad_norm": 0.7790745496749878, + "learning_rate": 0.0005373991483639623, + "loss": 0.456, + "step": 16630 + }, + { + "epoch": 0.9291840098332262, + "grad_norm": 0.46315130591392517, + "learning_rate": 0.0005373711340206185, + "loss": 0.5384, + "step": 16631 + }, + { + "epoch": 0.9292398804369081, + "grad_norm": 0.7847682237625122, + "learning_rate": 0.0005373431196772747, + "loss": 0.3525, + "step": 16632 + }, + { + "epoch": 0.92929575104059, + "grad_norm": 0.5124447345733643, + "learning_rate": 0.000537315105333931, + "loss": 0.4831, + "step": 16633 + }, + { + "epoch": 0.9293516216442719, + "grad_norm": 0.502463161945343, + "learning_rate": 0.0005372870909905871, + "loss": 0.425, + "step": 16634 + }, + { + "epoch": 0.9294074922479537, + "grad_norm": 0.5329257845878601, + "learning_rate": 0.0005372590766472433, + "loss": 0.4887, + "step": 16635 + }, + { + "epoch": 0.9294633628516357, + "grad_norm": 0.7273159623146057, + "learning_rate": 0.0005372310623038997, + "loss": 0.4571, + "step": 16636 + }, + { + "epoch": 0.9295192334553175, + "grad_norm": 0.5538855791091919, + "learning_rate": 0.0005372030479605559, + "loss": 0.5112, + "step": 16637 + }, + { + "epoch": 0.9295751040589993, + "grad_norm": 0.6078312397003174, + "learning_rate": 0.000537175033617212, + "loss": 0.4288, + "step": 16638 + }, + { + "epoch": 0.9296309746626812, + "grad_norm": 0.47821831703186035, + "learning_rate": 0.0005371470192738682, + "loss": 0.4685, + "step": 16639 + }, + { + "epoch": 0.9296868452663631, + "grad_norm": 0.5448079705238342, + "learning_rate": 0.0005371190049305245, + "loss": 0.3755, + "step": 16640 + }, + { + "epoch": 0.929742715870045, + "grad_norm": 0.6443954110145569, + "learning_rate": 0.0005370909905871807, + "loss": 0.5783, + "step": 16641 + }, + { + "epoch": 0.9297985864737268, + "grad_norm": 0.4803454875946045, + "learning_rate": 0.0005370629762438368, + "loss": 0.4286, + "step": 16642 + }, + { + "epoch": 0.9298544570774088, + "grad_norm": 0.4248979091644287, + "learning_rate": 0.0005370349619004931, + "loss": 0.4343, + "step": 16643 + }, + { + "epoch": 0.9299103276810906, + "grad_norm": 0.36979904770851135, + "learning_rate": 0.0005370069475571493, + "loss": 0.292, + "step": 16644 + }, + { + "epoch": 0.9299661982847724, + "grad_norm": 0.43331658840179443, + "learning_rate": 0.0005369789332138055, + "loss": 0.4305, + "step": 16645 + }, + { + "epoch": 0.9300220688884543, + "grad_norm": 0.7776563763618469, + "learning_rate": 0.0005369509188704616, + "loss": 0.4637, + "step": 16646 + }, + { + "epoch": 0.9300779394921362, + "grad_norm": 0.5636840462684631, + "learning_rate": 0.0005369229045271179, + "loss": 0.5041, + "step": 16647 + }, + { + "epoch": 0.9301338100958181, + "grad_norm": 0.6604616641998291, + "learning_rate": 0.0005368948901837741, + "loss": 0.4815, + "step": 16648 + }, + { + "epoch": 0.9301896806994999, + "grad_norm": 0.8293275833129883, + "learning_rate": 0.0005368668758404302, + "loss": 0.5729, + "step": 16649 + }, + { + "epoch": 0.9302455513031819, + "grad_norm": 0.4371901750564575, + "learning_rate": 0.0005368388614970865, + "loss": 0.3528, + "step": 16650 + }, + { + "epoch": 0.9303014219068637, + "grad_norm": 0.48611873388290405, + "learning_rate": 0.0005368108471537428, + "loss": 0.4637, + "step": 16651 + }, + { + "epoch": 0.9303572925105456, + "grad_norm": 1.8902888298034668, + "learning_rate": 0.000536782832810399, + "loss": 0.4235, + "step": 16652 + }, + { + "epoch": 0.9304131631142275, + "grad_norm": 0.5797620415687561, + "learning_rate": 0.000536754818467055, + "loss": 0.4832, + "step": 16653 + }, + { + "epoch": 0.9304690337179093, + "grad_norm": 1.4497413635253906, + "learning_rate": 0.0005367268041237114, + "loss": 0.4604, + "step": 16654 + }, + { + "epoch": 0.9305249043215912, + "grad_norm": 0.6777113676071167, + "learning_rate": 0.0005366987897803676, + "loss": 0.4585, + "step": 16655 + }, + { + "epoch": 0.930580774925273, + "grad_norm": 0.5377295613288879, + "learning_rate": 0.0005366707754370238, + "loss": 0.4942, + "step": 16656 + }, + { + "epoch": 0.930636645528955, + "grad_norm": 0.8134835958480835, + "learning_rate": 0.00053664276109368, + "loss": 0.4628, + "step": 16657 + }, + { + "epoch": 0.9306925161326368, + "grad_norm": 0.7346286177635193, + "learning_rate": 0.0005366147467503362, + "loss": 0.4761, + "step": 16658 + }, + { + "epoch": 0.9307483867363187, + "grad_norm": 0.42100948095321655, + "learning_rate": 0.0005365867324069924, + "loss": 0.3741, + "step": 16659 + }, + { + "epoch": 0.9308042573400006, + "grad_norm": 0.6152949333190918, + "learning_rate": 0.0005365587180636486, + "loss": 0.4866, + "step": 16660 + }, + { + "epoch": 0.9308601279436824, + "grad_norm": 0.6617674231529236, + "learning_rate": 0.0005365307037203048, + "loss": 0.4424, + "step": 16661 + }, + { + "epoch": 0.9309159985473643, + "grad_norm": 0.9567841291427612, + "learning_rate": 0.000536502689376961, + "loss": 0.6439, + "step": 16662 + }, + { + "epoch": 0.9309718691510461, + "grad_norm": 0.596537172794342, + "learning_rate": 0.0005364746750336172, + "loss": 0.4364, + "step": 16663 + }, + { + "epoch": 0.9310277397547281, + "grad_norm": 0.5103528499603271, + "learning_rate": 0.0005364466606902735, + "loss": 0.3754, + "step": 16664 + }, + { + "epoch": 0.9310836103584099, + "grad_norm": 0.5135740041732788, + "learning_rate": 0.0005364186463469296, + "loss": 0.4731, + "step": 16665 + }, + { + "epoch": 0.9311394809620918, + "grad_norm": 1.4921151399612427, + "learning_rate": 0.0005363906320035858, + "loss": 0.4973, + "step": 16666 + }, + { + "epoch": 0.9311953515657737, + "grad_norm": 0.456569105386734, + "learning_rate": 0.000536362617660242, + "loss": 0.4001, + "step": 16667 + }, + { + "epoch": 0.9312512221694556, + "grad_norm": 0.5161044001579285, + "learning_rate": 0.0005363346033168984, + "loss": 0.4226, + "step": 16668 + }, + { + "epoch": 0.9313070927731374, + "grad_norm": 0.6118404865264893, + "learning_rate": 0.0005363065889735545, + "loss": 0.5083, + "step": 16669 + }, + { + "epoch": 0.9313629633768193, + "grad_norm": 1.6293392181396484, + "learning_rate": 0.0005362785746302107, + "loss": 0.5821, + "step": 16670 + }, + { + "epoch": 0.9314188339805012, + "grad_norm": 1.8658653497695923, + "learning_rate": 0.0005362505602868669, + "loss": 0.4681, + "step": 16671 + }, + { + "epoch": 0.931474704584183, + "grad_norm": 0.44582536816596985, + "learning_rate": 0.0005362225459435231, + "loss": 0.5089, + "step": 16672 + }, + { + "epoch": 0.9315305751878649, + "grad_norm": 0.7737982869148254, + "learning_rate": 0.0005361945316001793, + "loss": 0.3935, + "step": 16673 + }, + { + "epoch": 0.9315864457915468, + "grad_norm": 0.4361270070075989, + "learning_rate": 0.0005361665172568355, + "loss": 0.4005, + "step": 16674 + }, + { + "epoch": 0.9316423163952287, + "grad_norm": 0.7784770727157593, + "learning_rate": 0.0005361385029134918, + "loss": 0.3839, + "step": 16675 + }, + { + "epoch": 0.9316981869989105, + "grad_norm": 0.7204387784004211, + "learning_rate": 0.0005361104885701479, + "loss": 0.5478, + "step": 16676 + }, + { + "epoch": 0.9317540576025924, + "grad_norm": 0.4455748200416565, + "learning_rate": 0.0005360824742268041, + "loss": 0.4305, + "step": 16677 + }, + { + "epoch": 0.9318099282062743, + "grad_norm": 0.5484066605567932, + "learning_rate": 0.0005360544598834603, + "loss": 0.389, + "step": 16678 + }, + { + "epoch": 0.9318657988099561, + "grad_norm": 0.7339828610420227, + "learning_rate": 0.0005360264455401166, + "loss": 0.4844, + "step": 16679 + }, + { + "epoch": 0.931921669413638, + "grad_norm": 0.6747909784317017, + "learning_rate": 0.0005359984311967727, + "loss": 0.5009, + "step": 16680 + }, + { + "epoch": 0.9319775400173199, + "grad_norm": 0.4696013629436493, + "learning_rate": 0.0005359704168534289, + "loss": 0.4586, + "step": 16681 + }, + { + "epoch": 0.9320334106210018, + "grad_norm": 4.13216495513916, + "learning_rate": 0.0005359424025100852, + "loss": 0.4335, + "step": 16682 + }, + { + "epoch": 0.9320892812246836, + "grad_norm": 0.4098418951034546, + "learning_rate": 0.0005359143881667414, + "loss": 0.4553, + "step": 16683 + }, + { + "epoch": 0.9321451518283655, + "grad_norm": 0.6282769441604614, + "learning_rate": 0.0005358863738233975, + "loss": 0.4311, + "step": 16684 + }, + { + "epoch": 0.9322010224320474, + "grad_norm": 4.36563777923584, + "learning_rate": 0.0005358583594800537, + "loss": 0.5613, + "step": 16685 + }, + { + "epoch": 0.9322568930357292, + "grad_norm": 0.7101638317108154, + "learning_rate": 0.0005358303451367101, + "loss": 0.3265, + "step": 16686 + }, + { + "epoch": 0.9323127636394111, + "grad_norm": 0.5149050951004028, + "learning_rate": 0.0005358023307933663, + "loss": 0.5014, + "step": 16687 + }, + { + "epoch": 0.932368634243093, + "grad_norm": 0.7562249898910522, + "learning_rate": 0.0005357743164500224, + "loss": 0.4892, + "step": 16688 + }, + { + "epoch": 0.9324245048467749, + "grad_norm": 0.3806728422641754, + "learning_rate": 0.0005357463021066787, + "loss": 0.3896, + "step": 16689 + }, + { + "epoch": 0.9324803754504567, + "grad_norm": 0.9142682552337646, + "learning_rate": 0.0005357182877633349, + "loss": 0.4058, + "step": 16690 + }, + { + "epoch": 0.9325362460541387, + "grad_norm": 0.42047953605651855, + "learning_rate": 0.000535690273419991, + "loss": 0.502, + "step": 16691 + }, + { + "epoch": 0.9325921166578205, + "grad_norm": 0.993699848651886, + "learning_rate": 0.0005356622590766472, + "loss": 0.4718, + "step": 16692 + }, + { + "epoch": 0.9326479872615023, + "grad_norm": 0.41910696029663086, + "learning_rate": 0.0005356342447333035, + "loss": 0.4368, + "step": 16693 + }, + { + "epoch": 0.9327038578651843, + "grad_norm": 0.344553679227829, + "learning_rate": 0.0005356062303899597, + "loss": 0.4357, + "step": 16694 + }, + { + "epoch": 0.9327597284688661, + "grad_norm": 13.774626731872559, + "learning_rate": 0.0005355782160466158, + "loss": 0.4586, + "step": 16695 + }, + { + "epoch": 0.932815599072548, + "grad_norm": 0.536346435546875, + "learning_rate": 0.0005355502017032721, + "loss": 0.5387, + "step": 16696 + }, + { + "epoch": 0.9328714696762298, + "grad_norm": 2.87544584274292, + "learning_rate": 0.0005355221873599283, + "loss": 0.4882, + "step": 16697 + }, + { + "epoch": 0.9329273402799118, + "grad_norm": 1.6501249074935913, + "learning_rate": 0.0005354941730165845, + "loss": 0.5318, + "step": 16698 + }, + { + "epoch": 0.9329832108835936, + "grad_norm": 0.7299724817276001, + "learning_rate": 0.0005354661586732406, + "loss": 0.3693, + "step": 16699 + }, + { + "epoch": 0.9330390814872754, + "grad_norm": 0.48290514945983887, + "learning_rate": 0.0005354381443298969, + "loss": 0.4423, + "step": 16700 + }, + { + "epoch": 0.9330949520909574, + "grad_norm": 0.35856887698173523, + "learning_rate": 0.0005354101299865531, + "loss": 0.3857, + "step": 16701 + }, + { + "epoch": 0.9331508226946392, + "grad_norm": 0.39228537678718567, + "learning_rate": 0.0005353821156432094, + "loss": 0.406, + "step": 16702 + }, + { + "epoch": 0.9332066932983211, + "grad_norm": 0.6066995859146118, + "learning_rate": 0.0005353541012998655, + "loss": 0.4162, + "step": 16703 + }, + { + "epoch": 0.9332625639020029, + "grad_norm": 0.4968808591365814, + "learning_rate": 0.0005353260869565218, + "loss": 0.4249, + "step": 16704 + }, + { + "epoch": 0.9333184345056849, + "grad_norm": 0.9211657047271729, + "learning_rate": 0.000535298072613178, + "loss": 0.4886, + "step": 16705 + }, + { + "epoch": 0.9333743051093667, + "grad_norm": 0.40891149640083313, + "learning_rate": 0.0005352700582698342, + "loss": 0.3887, + "step": 16706 + }, + { + "epoch": 0.9334301757130485, + "grad_norm": 0.5151289701461792, + "learning_rate": 0.0005352420439264904, + "loss": 0.4093, + "step": 16707 + }, + { + "epoch": 0.9334860463167305, + "grad_norm": 0.49429023265838623, + "learning_rate": 0.0005352140295831466, + "loss": 0.4388, + "step": 16708 + }, + { + "epoch": 0.9335419169204123, + "grad_norm": 0.5832515358924866, + "learning_rate": 0.0005351860152398028, + "loss": 0.4423, + "step": 16709 + }, + { + "epoch": 0.9335977875240942, + "grad_norm": 0.5387508273124695, + "learning_rate": 0.000535158000896459, + "loss": 0.5063, + "step": 16710 + }, + { + "epoch": 0.9336536581277761, + "grad_norm": 0.6965650916099548, + "learning_rate": 0.0005351299865531152, + "loss": 0.4818, + "step": 16711 + }, + { + "epoch": 0.933709528731458, + "grad_norm": 0.5097090601921082, + "learning_rate": 0.0005351019722097714, + "loss": 0.281, + "step": 16712 + }, + { + "epoch": 0.9337653993351398, + "grad_norm": 0.47043749690055847, + "learning_rate": 0.0005350739578664276, + "loss": 0.43, + "step": 16713 + }, + { + "epoch": 0.9338212699388216, + "grad_norm": 0.6336639523506165, + "learning_rate": 0.0005350459435230838, + "loss": 0.4132, + "step": 16714 + }, + { + "epoch": 0.9338771405425036, + "grad_norm": 0.5444793701171875, + "learning_rate": 0.00053501792917974, + "loss": 0.5521, + "step": 16715 + }, + { + "epoch": 0.9339330111461854, + "grad_norm": 0.5271199345588684, + "learning_rate": 0.0005349899148363962, + "loss": 0.3987, + "step": 16716 + }, + { + "epoch": 0.9339888817498673, + "grad_norm": 0.40868672728538513, + "learning_rate": 0.0005349619004930524, + "loss": 0.3841, + "step": 16717 + }, + { + "epoch": 0.9340447523535492, + "grad_norm": 0.3986000418663025, + "learning_rate": 0.0005349338861497086, + "loss": 0.326, + "step": 16718 + }, + { + "epoch": 0.9341006229572311, + "grad_norm": 0.6643268465995789, + "learning_rate": 0.0005349058718063649, + "loss": 0.4359, + "step": 16719 + }, + { + "epoch": 0.9341564935609129, + "grad_norm": 0.5228086709976196, + "learning_rate": 0.0005348778574630211, + "loss": 0.3945, + "step": 16720 + }, + { + "epoch": 0.9342123641645947, + "grad_norm": 0.6778776049613953, + "learning_rate": 0.0005348498431196774, + "loss": 0.3942, + "step": 16721 + }, + { + "epoch": 0.9342682347682767, + "grad_norm": 0.5614202618598938, + "learning_rate": 0.0005348218287763335, + "loss": 0.5397, + "step": 16722 + }, + { + "epoch": 0.9343241053719585, + "grad_norm": 0.5066548585891724, + "learning_rate": 0.0005347938144329897, + "loss": 0.4843, + "step": 16723 + }, + { + "epoch": 0.9343799759756404, + "grad_norm": 1.0353808403015137, + "learning_rate": 0.0005347658000896459, + "loss": 0.4415, + "step": 16724 + }, + { + "epoch": 0.9344358465793223, + "grad_norm": 0.3893764615058899, + "learning_rate": 0.0005347377857463022, + "loss": 0.4279, + "step": 16725 + }, + { + "epoch": 0.9344917171830042, + "grad_norm": 0.9617054462432861, + "learning_rate": 0.0005347097714029583, + "loss": 0.3146, + "step": 16726 + }, + { + "epoch": 0.934547587786686, + "grad_norm": 0.6063379049301147, + "learning_rate": 0.0005346817570596145, + "loss": 0.5048, + "step": 16727 + }, + { + "epoch": 0.934603458390368, + "grad_norm": 0.44396403431892395, + "learning_rate": 0.0005346537427162708, + "loss": 0.4596, + "step": 16728 + }, + { + "epoch": 0.9346593289940498, + "grad_norm": 0.6512336134910583, + "learning_rate": 0.000534625728372927, + "loss": 0.5077, + "step": 16729 + }, + { + "epoch": 0.9347151995977316, + "grad_norm": 0.3997063636779785, + "learning_rate": 0.0005345977140295831, + "loss": 0.3798, + "step": 16730 + }, + { + "epoch": 0.9347710702014135, + "grad_norm": 0.6824104189872742, + "learning_rate": 0.0005345696996862393, + "loss": 0.7759, + "step": 16731 + }, + { + "epoch": 0.9348269408050954, + "grad_norm": 1.1082366704940796, + "learning_rate": 0.0005345416853428956, + "loss": 0.4123, + "step": 16732 + }, + { + "epoch": 0.9348828114087773, + "grad_norm": 0.6713558435440063, + "learning_rate": 0.0005345136709995517, + "loss": 0.3742, + "step": 16733 + }, + { + "epoch": 0.9349386820124591, + "grad_norm": 1.2803065776824951, + "learning_rate": 0.0005344856566562079, + "loss": 0.4534, + "step": 16734 + }, + { + "epoch": 0.9349945526161411, + "grad_norm": 4.4289960861206055, + "learning_rate": 0.0005344576423128641, + "loss": 0.4262, + "step": 16735 + }, + { + "epoch": 0.9350504232198229, + "grad_norm": 0.727642297744751, + "learning_rate": 0.0005344296279695205, + "loss": 0.5332, + "step": 16736 + }, + { + "epoch": 0.9351062938235047, + "grad_norm": 0.44323211908340454, + "learning_rate": 0.0005344016136261766, + "loss": 0.4347, + "step": 16737 + }, + { + "epoch": 0.9351621644271866, + "grad_norm": 0.5987011790275574, + "learning_rate": 0.0005343735992828328, + "loss": 0.4554, + "step": 16738 + }, + { + "epoch": 0.9352180350308685, + "grad_norm": 0.559917151927948, + "learning_rate": 0.0005343455849394891, + "loss": 0.476, + "step": 16739 + }, + { + "epoch": 0.9352739056345504, + "grad_norm": 0.8911064267158508, + "learning_rate": 0.0005343175705961453, + "loss": 0.6329, + "step": 16740 + }, + { + "epoch": 0.9353297762382322, + "grad_norm": 1.3721773624420166, + "learning_rate": 0.0005342895562528014, + "loss": 0.4505, + "step": 16741 + }, + { + "epoch": 0.9353856468419142, + "grad_norm": 0.5446447134017944, + "learning_rate": 0.0005342615419094576, + "loss": 0.4586, + "step": 16742 + }, + { + "epoch": 0.935441517445596, + "grad_norm": 0.928949236869812, + "learning_rate": 0.0005342335275661139, + "loss": 0.5616, + "step": 16743 + }, + { + "epoch": 0.9354973880492778, + "grad_norm": 0.547827422618866, + "learning_rate": 0.0005342055132227701, + "loss": 0.4312, + "step": 16744 + }, + { + "epoch": 0.9355532586529598, + "grad_norm": 0.4301369786262512, + "learning_rate": 0.0005341774988794262, + "loss": 0.3931, + "step": 16745 + }, + { + "epoch": 0.9356091292566416, + "grad_norm": 0.5070576667785645, + "learning_rate": 0.0005341494845360825, + "loss": 0.493, + "step": 16746 + }, + { + "epoch": 0.9356649998603235, + "grad_norm": 1.6446126699447632, + "learning_rate": 0.0005341214701927387, + "loss": 0.4137, + "step": 16747 + }, + { + "epoch": 0.9357208704640053, + "grad_norm": 0.5437508821487427, + "learning_rate": 0.0005340934558493949, + "loss": 0.3998, + "step": 16748 + }, + { + "epoch": 0.9357767410676873, + "grad_norm": 0.609572172164917, + "learning_rate": 0.000534065441506051, + "loss": 0.3166, + "step": 16749 + }, + { + "epoch": 0.9358326116713691, + "grad_norm": 0.4567686915397644, + "learning_rate": 0.0005340374271627073, + "loss": 0.3777, + "step": 16750 + }, + { + "epoch": 0.935888482275051, + "grad_norm": 0.5495761036872864, + "learning_rate": 0.0005340094128193635, + "loss": 0.4407, + "step": 16751 + }, + { + "epoch": 0.9359443528787329, + "grad_norm": 0.8215982913970947, + "learning_rate": 0.0005339813984760198, + "loss": 0.3856, + "step": 16752 + }, + { + "epoch": 0.9360002234824147, + "grad_norm": 6.3641862869262695, + "learning_rate": 0.000533953384132676, + "loss": 0.4567, + "step": 16753 + }, + { + "epoch": 0.9360560940860966, + "grad_norm": 0.6247122883796692, + "learning_rate": 0.0005339253697893322, + "loss": 0.4683, + "step": 16754 + }, + { + "epoch": 0.9361119646897784, + "grad_norm": 10.704270362854004, + "learning_rate": 0.0005338973554459884, + "loss": 0.4432, + "step": 16755 + }, + { + "epoch": 0.9361678352934604, + "grad_norm": 1.9110504388809204, + "learning_rate": 0.0005338693411026445, + "loss": 0.3916, + "step": 16756 + }, + { + "epoch": 0.9362237058971422, + "grad_norm": 0.5290900468826294, + "learning_rate": 0.0005338413267593008, + "loss": 0.5312, + "step": 16757 + }, + { + "epoch": 0.936279576500824, + "grad_norm": 0.7395563721656799, + "learning_rate": 0.000533813312415957, + "loss": 0.5321, + "step": 16758 + }, + { + "epoch": 0.936335447104506, + "grad_norm": 0.6094143986701965, + "learning_rate": 0.0005337852980726132, + "loss": 0.5216, + "step": 16759 + }, + { + "epoch": 0.9363913177081878, + "grad_norm": 0.4035341143608093, + "learning_rate": 0.0005337572837292694, + "loss": 0.4101, + "step": 16760 + }, + { + "epoch": 0.9364471883118697, + "grad_norm": 0.5257874131202698, + "learning_rate": 0.0005337292693859256, + "loss": 0.4397, + "step": 16761 + }, + { + "epoch": 0.9365030589155516, + "grad_norm": 0.41845938563346863, + "learning_rate": 0.0005337012550425818, + "loss": 0.4476, + "step": 16762 + }, + { + "epoch": 0.9365589295192335, + "grad_norm": 0.4891835153102875, + "learning_rate": 0.000533673240699238, + "loss": 0.4123, + "step": 16763 + }, + { + "epoch": 0.9366148001229153, + "grad_norm": 0.3604746162891388, + "learning_rate": 0.0005336452263558942, + "loss": 0.4601, + "step": 16764 + }, + { + "epoch": 0.9366706707265972, + "grad_norm": 0.6685338020324707, + "learning_rate": 0.0005336172120125504, + "loss": 0.4042, + "step": 16765 + }, + { + "epoch": 0.9367265413302791, + "grad_norm": 0.5050128102302551, + "learning_rate": 0.0005335891976692066, + "loss": 0.3659, + "step": 16766 + }, + { + "epoch": 0.936782411933961, + "grad_norm": 0.5219718217849731, + "learning_rate": 0.000533561183325863, + "loss": 0.4581, + "step": 16767 + }, + { + "epoch": 0.9368382825376428, + "grad_norm": 0.452588826417923, + "learning_rate": 0.000533533168982519, + "loss": 0.361, + "step": 16768 + }, + { + "epoch": 0.9368941531413247, + "grad_norm": 1.0708277225494385, + "learning_rate": 0.0005335051546391753, + "loss": 0.4374, + "step": 16769 + }, + { + "epoch": 0.9369500237450066, + "grad_norm": 0.5494502782821655, + "learning_rate": 0.0005334771402958315, + "loss": 0.4279, + "step": 16770 + }, + { + "epoch": 0.9370058943486884, + "grad_norm": 0.561293363571167, + "learning_rate": 0.0005334491259524878, + "loss": 0.3892, + "step": 16771 + }, + { + "epoch": 0.9370617649523703, + "grad_norm": 0.4952652156352997, + "learning_rate": 0.0005334211116091439, + "loss": 0.4244, + "step": 16772 + }, + { + "epoch": 0.9371176355560522, + "grad_norm": 0.8728615045547485, + "learning_rate": 0.0005333930972658001, + "loss": 0.4298, + "step": 16773 + }, + { + "epoch": 0.937173506159734, + "grad_norm": 0.488473504781723, + "learning_rate": 0.0005333650829224563, + "loss": 0.5039, + "step": 16774 + }, + { + "epoch": 0.9372293767634159, + "grad_norm": 0.66480553150177, + "learning_rate": 0.0005333370685791126, + "loss": 0.5125, + "step": 16775 + }, + { + "epoch": 0.9372852473670978, + "grad_norm": 1.0720471143722534, + "learning_rate": 0.0005333090542357687, + "loss": 0.4102, + "step": 16776 + }, + { + "epoch": 0.9373411179707797, + "grad_norm": 0.7593253254890442, + "learning_rate": 0.0005332810398924249, + "loss": 0.4947, + "step": 16777 + }, + { + "epoch": 0.9373969885744615, + "grad_norm": 0.9109123349189758, + "learning_rate": 0.0005332530255490812, + "loss": 0.5242, + "step": 16778 + }, + { + "epoch": 0.9374528591781435, + "grad_norm": 0.6524176001548767, + "learning_rate": 0.0005332250112057373, + "loss": 0.4712, + "step": 16779 + }, + { + "epoch": 0.9375087297818253, + "grad_norm": 2.4642815589904785, + "learning_rate": 0.0005331969968623935, + "loss": 0.4923, + "step": 16780 + }, + { + "epoch": 0.9375646003855072, + "grad_norm": 1.0103284120559692, + "learning_rate": 0.0005331689825190497, + "loss": 0.5031, + "step": 16781 + }, + { + "epoch": 0.937620470989189, + "grad_norm": 0.3367442488670349, + "learning_rate": 0.000533140968175706, + "loss": 0.3699, + "step": 16782 + }, + { + "epoch": 0.9376763415928709, + "grad_norm": 0.5808252096176147, + "learning_rate": 0.0005331129538323621, + "loss": 0.5024, + "step": 16783 + }, + { + "epoch": 0.9377322121965528, + "grad_norm": 0.4368676543235779, + "learning_rate": 0.0005330849394890183, + "loss": 0.4209, + "step": 16784 + }, + { + "epoch": 0.9377880828002346, + "grad_norm": 1.076412320137024, + "learning_rate": 0.0005330569251456747, + "loss": 0.4172, + "step": 16785 + }, + { + "epoch": 0.9378439534039166, + "grad_norm": 0.41078877449035645, + "learning_rate": 0.0005330289108023309, + "loss": 0.4983, + "step": 16786 + }, + { + "epoch": 0.9378998240075984, + "grad_norm": 0.5638454556465149, + "learning_rate": 0.000533000896458987, + "loss": 0.3611, + "step": 16787 + }, + { + "epoch": 0.9379556946112803, + "grad_norm": 2.1463217735290527, + "learning_rate": 0.0005329728821156432, + "loss": 0.4347, + "step": 16788 + }, + { + "epoch": 0.9380115652149621, + "grad_norm": 0.5903355479240417, + "learning_rate": 0.0005329448677722995, + "loss": 0.4374, + "step": 16789 + }, + { + "epoch": 0.938067435818644, + "grad_norm": 0.6110395789146423, + "learning_rate": 0.0005329168534289557, + "loss": 0.5487, + "step": 16790 + }, + { + "epoch": 0.9381233064223259, + "grad_norm": 0.5178189277648926, + "learning_rate": 0.0005328888390856118, + "loss": 0.4387, + "step": 16791 + }, + { + "epoch": 0.9381791770260077, + "grad_norm": 0.752644956111908, + "learning_rate": 0.0005328608247422681, + "loss": 0.4912, + "step": 16792 + }, + { + "epoch": 0.9382350476296897, + "grad_norm": 0.558466911315918, + "learning_rate": 0.0005328328103989243, + "loss": 0.6265, + "step": 16793 + }, + { + "epoch": 0.9382909182333715, + "grad_norm": 0.5235158205032349, + "learning_rate": 0.0005328047960555805, + "loss": 0.4268, + "step": 16794 + }, + { + "epoch": 0.9383467888370534, + "grad_norm": 1.3287615776062012, + "learning_rate": 0.0005327767817122366, + "loss": 0.4074, + "step": 16795 + }, + { + "epoch": 0.9384026594407353, + "grad_norm": 0.4541247487068176, + "learning_rate": 0.0005327487673688929, + "loss": 0.4763, + "step": 16796 + }, + { + "epoch": 0.9384585300444171, + "grad_norm": 0.568727433681488, + "learning_rate": 0.0005327207530255491, + "loss": 0.3972, + "step": 16797 + }, + { + "epoch": 0.938514400648099, + "grad_norm": 0.34807461500167847, + "learning_rate": 0.0005326927386822052, + "loss": 0.3404, + "step": 16798 + }, + { + "epoch": 0.9385702712517808, + "grad_norm": 2.2186954021453857, + "learning_rate": 0.0005326647243388615, + "loss": 0.4296, + "step": 16799 + }, + { + "epoch": 0.9386261418554628, + "grad_norm": 0.4489508867263794, + "learning_rate": 0.0005326367099955177, + "loss": 0.4486, + "step": 16800 + }, + { + "epoch": 0.9386820124591446, + "grad_norm": 0.42232680320739746, + "learning_rate": 0.000532608695652174, + "loss": 0.323, + "step": 16801 + }, + { + "epoch": 0.9387378830628265, + "grad_norm": 0.6761463284492493, + "learning_rate": 0.00053258068130883, + "loss": 0.424, + "step": 16802 + }, + { + "epoch": 0.9387937536665084, + "grad_norm": 0.6879476308822632, + "learning_rate": 0.0005325526669654864, + "loss": 0.3826, + "step": 16803 + }, + { + "epoch": 0.9388496242701903, + "grad_norm": 0.3713017404079437, + "learning_rate": 0.0005325246526221426, + "loss": 0.5003, + "step": 16804 + }, + { + "epoch": 0.9389054948738721, + "grad_norm": 0.5846715569496155, + "learning_rate": 0.0005324966382787988, + "loss": 0.5621, + "step": 16805 + }, + { + "epoch": 0.9389613654775539, + "grad_norm": 0.5269457697868347, + "learning_rate": 0.0005324686239354549, + "loss": 0.5023, + "step": 16806 + }, + { + "epoch": 0.9390172360812359, + "grad_norm": 0.460488885641098, + "learning_rate": 0.0005324406095921112, + "loss": 0.4871, + "step": 16807 + }, + { + "epoch": 0.9390731066849177, + "grad_norm": 0.6328038573265076, + "learning_rate": 0.0005324125952487674, + "loss": 0.492, + "step": 16808 + }, + { + "epoch": 0.9391289772885996, + "grad_norm": 0.6945175528526306, + "learning_rate": 0.0005323845809054236, + "loss": 0.4223, + "step": 16809 + }, + { + "epoch": 0.9391848478922815, + "grad_norm": 0.6484053730964661, + "learning_rate": 0.0005323565665620798, + "loss": 0.378, + "step": 16810 + }, + { + "epoch": 0.9392407184959634, + "grad_norm": 1.2562551498413086, + "learning_rate": 0.000532328552218736, + "loss": 0.4647, + "step": 16811 + }, + { + "epoch": 0.9392965890996452, + "grad_norm": 0.5780539512634277, + "learning_rate": 0.0005323005378753922, + "loss": 0.4106, + "step": 16812 + }, + { + "epoch": 0.9393524597033271, + "grad_norm": 1.348942756652832, + "learning_rate": 0.0005322725235320484, + "loss": 0.4014, + "step": 16813 + }, + { + "epoch": 0.939408330307009, + "grad_norm": 0.8753619194030762, + "learning_rate": 0.0005322445091887046, + "loss": 0.5666, + "step": 16814 + }, + { + "epoch": 0.9394642009106908, + "grad_norm": 0.980251669883728, + "learning_rate": 0.0005322164948453608, + "loss": 0.5888, + "step": 16815 + }, + { + "epoch": 0.9395200715143727, + "grad_norm": 1.9358680248260498, + "learning_rate": 0.000532188480502017, + "loss": 0.4087, + "step": 16816 + }, + { + "epoch": 0.9395759421180546, + "grad_norm": 0.4535771906375885, + "learning_rate": 0.0005321604661586733, + "loss": 0.3557, + "step": 16817 + }, + { + "epoch": 0.9396318127217365, + "grad_norm": 0.4498709738254547, + "learning_rate": 0.0005321324518153294, + "loss": 0.4101, + "step": 16818 + }, + { + "epoch": 0.9396876833254183, + "grad_norm": 14.091222763061523, + "learning_rate": 0.0005321044374719857, + "loss": 0.3595, + "step": 16819 + }, + { + "epoch": 0.9397435539291003, + "grad_norm": 0.4427432715892792, + "learning_rate": 0.0005320764231286419, + "loss": 0.4626, + "step": 16820 + }, + { + "epoch": 0.9397994245327821, + "grad_norm": 0.3874290883541107, + "learning_rate": 0.0005320484087852981, + "loss": 0.3798, + "step": 16821 + }, + { + "epoch": 0.9398552951364639, + "grad_norm": 0.6097959280014038, + "learning_rate": 0.0005320203944419543, + "loss": 0.4304, + "step": 16822 + }, + { + "epoch": 0.9399111657401458, + "grad_norm": 0.40335923433303833, + "learning_rate": 0.0005319923800986105, + "loss": 0.4987, + "step": 16823 + }, + { + "epoch": 0.9399670363438277, + "grad_norm": 2.23004150390625, + "learning_rate": 0.0005319643657552668, + "loss": 0.445, + "step": 16824 + }, + { + "epoch": 0.9400229069475096, + "grad_norm": 0.5112172961235046, + "learning_rate": 0.0005319363514119229, + "loss": 0.4601, + "step": 16825 + }, + { + "epoch": 0.9400787775511914, + "grad_norm": 0.31863778829574585, + "learning_rate": 0.0005319083370685791, + "loss": 0.3491, + "step": 16826 + }, + { + "epoch": 0.9401346481548734, + "grad_norm": 0.5667107105255127, + "learning_rate": 0.0005318803227252353, + "loss": 0.495, + "step": 16827 + }, + { + "epoch": 0.9401905187585552, + "grad_norm": 0.6899587512016296, + "learning_rate": 0.0005318523083818916, + "loss": 0.363, + "step": 16828 + }, + { + "epoch": 0.940246389362237, + "grad_norm": 0.392936646938324, + "learning_rate": 0.0005318242940385477, + "loss": 0.3577, + "step": 16829 + }, + { + "epoch": 0.940302259965919, + "grad_norm": 0.5817273855209351, + "learning_rate": 0.0005317962796952039, + "loss": 0.4041, + "step": 16830 + }, + { + "epoch": 0.9403581305696008, + "grad_norm": 1.0046311616897583, + "learning_rate": 0.0005317682653518602, + "loss": 0.4523, + "step": 16831 + }, + { + "epoch": 0.9404140011732827, + "grad_norm": 0.4787006974220276, + "learning_rate": 0.0005317402510085164, + "loss": 0.4447, + "step": 16832 + }, + { + "epoch": 0.9404698717769645, + "grad_norm": 0.4067326784133911, + "learning_rate": 0.0005317122366651725, + "loss": 0.4828, + "step": 16833 + }, + { + "epoch": 0.9405257423806465, + "grad_norm": 1.3731275796890259, + "learning_rate": 0.0005316842223218287, + "loss": 0.5517, + "step": 16834 + }, + { + "epoch": 0.9405816129843283, + "grad_norm": 3.0183968544006348, + "learning_rate": 0.000531656207978485, + "loss": 0.4596, + "step": 16835 + }, + { + "epoch": 0.9406374835880101, + "grad_norm": 0.4108138382434845, + "learning_rate": 0.0005316281936351413, + "loss": 0.3999, + "step": 16836 + }, + { + "epoch": 0.9406933541916921, + "grad_norm": 0.5455321669578552, + "learning_rate": 0.0005316001792917974, + "loss": 0.354, + "step": 16837 + }, + { + "epoch": 0.9407492247953739, + "grad_norm": 4.662905216217041, + "learning_rate": 0.0005315721649484537, + "loss": 0.4722, + "step": 16838 + }, + { + "epoch": 0.9408050953990558, + "grad_norm": 0.4539664089679718, + "learning_rate": 0.0005315441506051099, + "loss": 0.377, + "step": 16839 + }, + { + "epoch": 0.9408609660027376, + "grad_norm": 0.9644096493721008, + "learning_rate": 0.000531516136261766, + "loss": 0.6732, + "step": 16840 + }, + { + "epoch": 0.9409168366064196, + "grad_norm": 0.7508513331413269, + "learning_rate": 0.0005314881219184222, + "loss": 0.4014, + "step": 16841 + }, + { + "epoch": 0.9409727072101014, + "grad_norm": 0.6110619306564331, + "learning_rate": 0.0005314601075750785, + "loss": 0.4742, + "step": 16842 + }, + { + "epoch": 0.9410285778137832, + "grad_norm": 0.43170639872550964, + "learning_rate": 0.0005314320932317347, + "loss": 0.471, + "step": 16843 + }, + { + "epoch": 0.9410844484174652, + "grad_norm": 1.3612964153289795, + "learning_rate": 0.0005314040788883908, + "loss": 0.4681, + "step": 16844 + }, + { + "epoch": 0.941140319021147, + "grad_norm": 0.3787463307380676, + "learning_rate": 0.000531376064545047, + "loss": 0.4507, + "step": 16845 + }, + { + "epoch": 0.9411961896248289, + "grad_norm": 0.39381468296051025, + "learning_rate": 0.0005313480502017033, + "loss": 0.4523, + "step": 16846 + }, + { + "epoch": 0.9412520602285108, + "grad_norm": 0.5111168622970581, + "learning_rate": 0.0005313200358583595, + "loss": 0.443, + "step": 16847 + }, + { + "epoch": 0.9413079308321927, + "grad_norm": 0.7639251351356506, + "learning_rate": 0.0005312920215150156, + "loss": 0.3915, + "step": 16848 + }, + { + "epoch": 0.9413638014358745, + "grad_norm": 0.5430421233177185, + "learning_rate": 0.0005312640071716719, + "loss": 0.5768, + "step": 16849 + }, + { + "epoch": 0.9414196720395563, + "grad_norm": 0.5444962382316589, + "learning_rate": 0.0005312359928283281, + "loss": 0.4394, + "step": 16850 + }, + { + "epoch": 0.9414755426432383, + "grad_norm": 0.4322383999824524, + "learning_rate": 0.0005312079784849843, + "loss": 0.3676, + "step": 16851 + }, + { + "epoch": 0.9415314132469201, + "grad_norm": 0.6101773381233215, + "learning_rate": 0.0005311799641416404, + "loss": 0.419, + "step": 16852 + }, + { + "epoch": 0.941587283850602, + "grad_norm": 0.48380333185195923, + "learning_rate": 0.0005311519497982968, + "loss": 0.4261, + "step": 16853 + }, + { + "epoch": 0.9416431544542839, + "grad_norm": 0.3896867632865906, + "learning_rate": 0.000531123935454953, + "loss": 0.3892, + "step": 16854 + }, + { + "epoch": 0.9416990250579658, + "grad_norm": 0.4644891321659088, + "learning_rate": 0.0005310959211116092, + "loss": 0.383, + "step": 16855 + }, + { + "epoch": 0.9417548956616476, + "grad_norm": 0.5698118209838867, + "learning_rate": 0.0005310679067682654, + "loss": 0.4968, + "step": 16856 + }, + { + "epoch": 0.9418107662653294, + "grad_norm": 1.346196174621582, + "learning_rate": 0.0005310398924249216, + "loss": 0.3752, + "step": 16857 + }, + { + "epoch": 0.9418666368690114, + "grad_norm": 2.1576826572418213, + "learning_rate": 0.0005310118780815778, + "loss": 0.349, + "step": 16858 + }, + { + "epoch": 0.9419225074726932, + "grad_norm": 0.5634671449661255, + "learning_rate": 0.000530983863738234, + "loss": 0.5365, + "step": 16859 + }, + { + "epoch": 0.9419783780763751, + "grad_norm": 0.396259069442749, + "learning_rate": 0.0005309558493948902, + "loss": 0.3792, + "step": 16860 + }, + { + "epoch": 0.942034248680057, + "grad_norm": 0.47993481159210205, + "learning_rate": 0.0005309278350515464, + "loss": 0.3802, + "step": 16861 + }, + { + "epoch": 0.9420901192837389, + "grad_norm": 0.4439360201358795, + "learning_rate": 0.0005308998207082026, + "loss": 0.4077, + "step": 16862 + }, + { + "epoch": 0.9421459898874207, + "grad_norm": 0.41603338718414307, + "learning_rate": 0.0005308718063648588, + "loss": 0.372, + "step": 16863 + }, + { + "epoch": 0.9422018604911027, + "grad_norm": 0.7314110994338989, + "learning_rate": 0.000530843792021515, + "loss": 0.4308, + "step": 16864 + }, + { + "epoch": 0.9422577310947845, + "grad_norm": 4.318739891052246, + "learning_rate": 0.0005308157776781712, + "loss": 0.3898, + "step": 16865 + }, + { + "epoch": 0.9423136016984663, + "grad_norm": 0.525908350944519, + "learning_rate": 0.0005307877633348274, + "loss": 0.4709, + "step": 16866 + }, + { + "epoch": 0.9423694723021482, + "grad_norm": 0.3143058121204376, + "learning_rate": 0.0005307597489914836, + "loss": 0.3428, + "step": 16867 + }, + { + "epoch": 0.9424253429058301, + "grad_norm": 0.37318453192710876, + "learning_rate": 0.0005307317346481398, + "loss": 0.4451, + "step": 16868 + }, + { + "epoch": 0.942481213509512, + "grad_norm": 0.37382006645202637, + "learning_rate": 0.000530703720304796, + "loss": 0.3896, + "step": 16869 + }, + { + "epoch": 0.9425370841131938, + "grad_norm": 0.5024839639663696, + "learning_rate": 0.0005306757059614524, + "loss": 0.5294, + "step": 16870 + }, + { + "epoch": 0.9425929547168758, + "grad_norm": 3.2826650142669678, + "learning_rate": 0.0005306476916181085, + "loss": 0.4947, + "step": 16871 + }, + { + "epoch": 0.9426488253205576, + "grad_norm": 1.2569952011108398, + "learning_rate": 0.0005306196772747647, + "loss": 0.4617, + "step": 16872 + }, + { + "epoch": 0.9427046959242394, + "grad_norm": 0.46315085887908936, + "learning_rate": 0.0005305916629314209, + "loss": 0.3957, + "step": 16873 + }, + { + "epoch": 0.9427605665279213, + "grad_norm": 0.47259607911109924, + "learning_rate": 0.0005305636485880772, + "loss": 0.4939, + "step": 16874 + }, + { + "epoch": 0.9428164371316032, + "grad_norm": 1.1792738437652588, + "learning_rate": 0.0005305356342447333, + "loss": 0.458, + "step": 16875 + }, + { + "epoch": 0.9428723077352851, + "grad_norm": 6.640029430389404, + "learning_rate": 0.0005305076199013895, + "loss": 0.3984, + "step": 16876 + }, + { + "epoch": 0.9429281783389669, + "grad_norm": 0.3928070366382599, + "learning_rate": 0.0005304796055580458, + "loss": 0.431, + "step": 16877 + }, + { + "epoch": 0.9429840489426489, + "grad_norm": 1.5763498544692993, + "learning_rate": 0.000530451591214702, + "loss": 0.39, + "step": 16878 + }, + { + "epoch": 0.9430399195463307, + "grad_norm": 1.1209403276443481, + "learning_rate": 0.0005304235768713581, + "loss": 0.5541, + "step": 16879 + }, + { + "epoch": 0.9430957901500125, + "grad_norm": 0.3282720148563385, + "learning_rate": 0.0005303955625280143, + "loss": 0.4133, + "step": 16880 + }, + { + "epoch": 0.9431516607536945, + "grad_norm": 0.6758965849876404, + "learning_rate": 0.0005303675481846706, + "loss": 0.3822, + "step": 16881 + }, + { + "epoch": 0.9432075313573763, + "grad_norm": 0.35582759976387024, + "learning_rate": 0.0005303395338413267, + "loss": 0.4042, + "step": 16882 + }, + { + "epoch": 0.9432634019610582, + "grad_norm": 0.47266843914985657, + "learning_rate": 0.0005303115194979829, + "loss": 0.4292, + "step": 16883 + }, + { + "epoch": 0.94331927256474, + "grad_norm": 4.289495944976807, + "learning_rate": 0.0005302835051546391, + "loss": 0.3212, + "step": 16884 + }, + { + "epoch": 0.943375143168422, + "grad_norm": 0.41963958740234375, + "learning_rate": 0.0005302554908112955, + "loss": 0.4207, + "step": 16885 + }, + { + "epoch": 0.9434310137721038, + "grad_norm": 0.858343780040741, + "learning_rate": 0.0005302274764679516, + "loss": 0.4851, + "step": 16886 + }, + { + "epoch": 0.9434868843757857, + "grad_norm": 0.4502648413181305, + "learning_rate": 0.0005301994621246078, + "loss": 0.4282, + "step": 16887 + }, + { + "epoch": 0.9435427549794676, + "grad_norm": 0.5772939324378967, + "learning_rate": 0.0005301714477812641, + "loss": 0.5322, + "step": 16888 + }, + { + "epoch": 0.9435986255831494, + "grad_norm": 0.48183631896972656, + "learning_rate": 0.0005301434334379203, + "loss": 0.3625, + "step": 16889 + }, + { + "epoch": 0.9436544961868313, + "grad_norm": 0.6886517405509949, + "learning_rate": 0.0005301154190945764, + "loss": 0.3364, + "step": 16890 + }, + { + "epoch": 0.9437103667905131, + "grad_norm": 3.6032936573028564, + "learning_rate": 0.0005300874047512326, + "loss": 0.3844, + "step": 16891 + }, + { + "epoch": 0.9437662373941951, + "grad_norm": 0.5310473442077637, + "learning_rate": 0.0005300593904078889, + "loss": 0.4898, + "step": 16892 + }, + { + "epoch": 0.9438221079978769, + "grad_norm": 0.47311660647392273, + "learning_rate": 0.0005300313760645451, + "loss": 0.4568, + "step": 16893 + }, + { + "epoch": 0.9438779786015588, + "grad_norm": 0.5691388845443726, + "learning_rate": 0.0005300033617212012, + "loss": 0.4516, + "step": 16894 + }, + { + "epoch": 0.9439338492052407, + "grad_norm": 3.7123966217041016, + "learning_rate": 0.0005299753473778575, + "loss": 0.4465, + "step": 16895 + }, + { + "epoch": 0.9439897198089225, + "grad_norm": 0.4685850441455841, + "learning_rate": 0.0005299473330345137, + "loss": 0.4634, + "step": 16896 + }, + { + "epoch": 0.9440455904126044, + "grad_norm": 0.5601983070373535, + "learning_rate": 0.0005299193186911699, + "loss": 0.5632, + "step": 16897 + }, + { + "epoch": 0.9441014610162863, + "grad_norm": 0.6465625762939453, + "learning_rate": 0.000529891304347826, + "loss": 0.5112, + "step": 16898 + }, + { + "epoch": 0.9441573316199682, + "grad_norm": 0.675508439540863, + "learning_rate": 0.0005298632900044823, + "loss": 0.6045, + "step": 16899 + }, + { + "epoch": 0.94421320222365, + "grad_norm": 0.7966699004173279, + "learning_rate": 0.0005298352756611385, + "loss": 0.4835, + "step": 16900 + }, + { + "epoch": 0.9442690728273319, + "grad_norm": 1.5204923152923584, + "learning_rate": 0.0005298072613177947, + "loss": 0.4515, + "step": 16901 + }, + { + "epoch": 0.9443249434310138, + "grad_norm": 0.4180886149406433, + "learning_rate": 0.000529779246974451, + "loss": 0.3858, + "step": 16902 + }, + { + "epoch": 0.9443808140346956, + "grad_norm": 0.8921626806259155, + "learning_rate": 0.0005297512326311072, + "loss": 0.4075, + "step": 16903 + }, + { + "epoch": 0.9444366846383775, + "grad_norm": 0.45971930027008057, + "learning_rate": 0.0005297232182877634, + "loss": 0.4337, + "step": 16904 + }, + { + "epoch": 0.9444925552420594, + "grad_norm": 1.1212407350540161, + "learning_rate": 0.0005296952039444195, + "loss": 0.6301, + "step": 16905 + }, + { + "epoch": 0.9445484258457413, + "grad_norm": 0.389585018157959, + "learning_rate": 0.0005296671896010758, + "loss": 0.5583, + "step": 16906 + }, + { + "epoch": 0.9446042964494231, + "grad_norm": 0.50372713804245, + "learning_rate": 0.000529639175257732, + "loss": 0.442, + "step": 16907 + }, + { + "epoch": 0.944660167053105, + "grad_norm": 0.6619626879692078, + "learning_rate": 0.0005296111609143882, + "loss": 0.4816, + "step": 16908 + }, + { + "epoch": 0.9447160376567869, + "grad_norm": 0.40433570742607117, + "learning_rate": 0.0005295831465710444, + "loss": 0.4885, + "step": 16909 + }, + { + "epoch": 0.9447719082604688, + "grad_norm": 0.46292686462402344, + "learning_rate": 0.0005295551322277006, + "loss": 0.4214, + "step": 16910 + }, + { + "epoch": 0.9448277788641506, + "grad_norm": 0.5438519716262817, + "learning_rate": 0.0005295271178843568, + "loss": 0.4315, + "step": 16911 + }, + { + "epoch": 0.9448836494678325, + "grad_norm": 0.46739596128463745, + "learning_rate": 0.000529499103541013, + "loss": 0.35, + "step": 16912 + }, + { + "epoch": 0.9449395200715144, + "grad_norm": 1.3238426446914673, + "learning_rate": 0.0005294710891976692, + "loss": 0.4969, + "step": 16913 + }, + { + "epoch": 0.9449953906751962, + "grad_norm": 0.6249369978904724, + "learning_rate": 0.0005294430748543254, + "loss": 0.4143, + "step": 16914 + }, + { + "epoch": 0.9450512612788781, + "grad_norm": 0.5513806939125061, + "learning_rate": 0.0005294150605109816, + "loss": 0.4685, + "step": 16915 + }, + { + "epoch": 0.94510713188256, + "grad_norm": 0.4687545895576477, + "learning_rate": 0.0005293870461676378, + "loss": 0.325, + "step": 16916 + }, + { + "epoch": 0.9451630024862419, + "grad_norm": 0.640082836151123, + "learning_rate": 0.000529359031824294, + "loss": 0.4959, + "step": 16917 + }, + { + "epoch": 0.9452188730899237, + "grad_norm": 1.1994624137878418, + "learning_rate": 0.0005293310174809502, + "loss": 0.3858, + "step": 16918 + }, + { + "epoch": 0.9452747436936056, + "grad_norm": 0.8430876731872559, + "learning_rate": 0.0005293030031376065, + "loss": 0.3534, + "step": 16919 + }, + { + "epoch": 0.9453306142972875, + "grad_norm": 0.5122107267379761, + "learning_rate": 0.0005292749887942628, + "loss": 0.4336, + "step": 16920 + }, + { + "epoch": 0.9453864849009693, + "grad_norm": 0.5310891270637512, + "learning_rate": 0.0005292469744509189, + "loss": 0.3987, + "step": 16921 + }, + { + "epoch": 0.9454423555046513, + "grad_norm": 0.42041701078414917, + "learning_rate": 0.0005292189601075751, + "loss": 0.4322, + "step": 16922 + }, + { + "epoch": 0.9454982261083331, + "grad_norm": 0.4636755585670471, + "learning_rate": 0.0005291909457642313, + "loss": 0.4523, + "step": 16923 + }, + { + "epoch": 0.945554096712015, + "grad_norm": 0.8836562037467957, + "learning_rate": 0.0005291629314208875, + "loss": 0.5232, + "step": 16924 + }, + { + "epoch": 0.9456099673156968, + "grad_norm": 0.6014986038208008, + "learning_rate": 0.0005291349170775437, + "loss": 0.3628, + "step": 16925 + }, + { + "epoch": 0.9456658379193787, + "grad_norm": 0.4376332759857178, + "learning_rate": 0.0005291069027341999, + "loss": 0.4804, + "step": 16926 + }, + { + "epoch": 0.9457217085230606, + "grad_norm": 0.42759501934051514, + "learning_rate": 0.0005290788883908562, + "loss": 0.4407, + "step": 16927 + }, + { + "epoch": 0.9457775791267424, + "grad_norm": 0.312051922082901, + "learning_rate": 0.0005290508740475123, + "loss": 0.374, + "step": 16928 + }, + { + "epoch": 0.9458334497304244, + "grad_norm": 0.5194003582000732, + "learning_rate": 0.0005290228597041685, + "loss": 0.424, + "step": 16929 + }, + { + "epoch": 0.9458893203341062, + "grad_norm": 0.46532800793647766, + "learning_rate": 0.0005289948453608247, + "loss": 0.4787, + "step": 16930 + }, + { + "epoch": 0.9459451909377881, + "grad_norm": 2.2270219326019287, + "learning_rate": 0.000528966831017481, + "loss": 0.3627, + "step": 16931 + }, + { + "epoch": 0.9460010615414699, + "grad_norm": 2.9272232055664062, + "learning_rate": 0.0005289388166741371, + "loss": 0.4036, + "step": 16932 + }, + { + "epoch": 0.9460569321451519, + "grad_norm": 0.8695594668388367, + "learning_rate": 0.0005289108023307933, + "loss": 0.4886, + "step": 16933 + }, + { + "epoch": 0.9461128027488337, + "grad_norm": 2.0327982902526855, + "learning_rate": 0.0005288827879874496, + "loss": 0.4465, + "step": 16934 + }, + { + "epoch": 0.9461686733525155, + "grad_norm": 0.581994354724884, + "learning_rate": 0.0005288547736441059, + "loss": 0.4843, + "step": 16935 + }, + { + "epoch": 0.9462245439561975, + "grad_norm": 0.4776178300380707, + "learning_rate": 0.000528826759300762, + "loss": 0.4159, + "step": 16936 + }, + { + "epoch": 0.9462804145598793, + "grad_norm": 0.5736627578735352, + "learning_rate": 0.0005287987449574182, + "loss": 0.4362, + "step": 16937 + }, + { + "epoch": 0.9463362851635612, + "grad_norm": 0.39928945899009705, + "learning_rate": 0.0005287707306140745, + "loss": 0.3874, + "step": 16938 + }, + { + "epoch": 0.9463921557672431, + "grad_norm": 0.4660637080669403, + "learning_rate": 0.0005287427162707307, + "loss": 0.4365, + "step": 16939 + }, + { + "epoch": 0.946448026370925, + "grad_norm": 0.3784901797771454, + "learning_rate": 0.0005287147019273868, + "loss": 0.4644, + "step": 16940 + }, + { + "epoch": 0.9465038969746068, + "grad_norm": 0.37795066833496094, + "learning_rate": 0.0005286866875840431, + "loss": 0.3481, + "step": 16941 + }, + { + "epoch": 0.9465597675782886, + "grad_norm": 0.6980178952217102, + "learning_rate": 0.0005286586732406993, + "loss": 0.5353, + "step": 16942 + }, + { + "epoch": 0.9466156381819706, + "grad_norm": 0.8605262637138367, + "learning_rate": 0.0005286306588973555, + "loss": 0.5267, + "step": 16943 + }, + { + "epoch": 0.9466715087856524, + "grad_norm": 0.46472182869911194, + "learning_rate": 0.0005286026445540116, + "loss": 0.3858, + "step": 16944 + }, + { + "epoch": 0.9467273793893343, + "grad_norm": 0.39177852869033813, + "learning_rate": 0.0005285746302106679, + "loss": 0.4567, + "step": 16945 + }, + { + "epoch": 0.9467832499930162, + "grad_norm": 0.673615574836731, + "learning_rate": 0.0005285466158673241, + "loss": 0.435, + "step": 16946 + }, + { + "epoch": 0.9468391205966981, + "grad_norm": 0.7678151726722717, + "learning_rate": 0.0005285186015239802, + "loss": 0.483, + "step": 16947 + }, + { + "epoch": 0.9468949912003799, + "grad_norm": 0.38197439908981323, + "learning_rate": 0.0005284905871806364, + "loss": 0.4802, + "step": 16948 + }, + { + "epoch": 0.9469508618040617, + "grad_norm": 0.722059428691864, + "learning_rate": 0.0005284625728372927, + "loss": 0.4766, + "step": 16949 + }, + { + "epoch": 0.9470067324077437, + "grad_norm": 0.5109726786613464, + "learning_rate": 0.0005284345584939489, + "loss": 0.465, + "step": 16950 + }, + { + "epoch": 0.9470626030114255, + "grad_norm": 0.4120565950870514, + "learning_rate": 0.000528406544150605, + "loss": 0.5725, + "step": 16951 + }, + { + "epoch": 0.9471184736151074, + "grad_norm": 0.5449019074440002, + "learning_rate": 0.0005283785298072614, + "loss": 0.4115, + "step": 16952 + }, + { + "epoch": 0.9471743442187893, + "grad_norm": 1.1640737056732178, + "learning_rate": 0.0005283505154639176, + "loss": 0.4443, + "step": 16953 + }, + { + "epoch": 0.9472302148224712, + "grad_norm": 0.6808809041976929, + "learning_rate": 0.0005283225011205738, + "loss": 0.3758, + "step": 16954 + }, + { + "epoch": 0.947286085426153, + "grad_norm": 0.5132268667221069, + "learning_rate": 0.0005282944867772299, + "loss": 0.4741, + "step": 16955 + }, + { + "epoch": 0.947341956029835, + "grad_norm": 0.48207589983940125, + "learning_rate": 0.0005282664724338862, + "loss": 0.4065, + "step": 16956 + }, + { + "epoch": 0.9473978266335168, + "grad_norm": 0.8223482370376587, + "learning_rate": 0.0005282384580905424, + "loss": 0.7807, + "step": 16957 + }, + { + "epoch": 0.9474536972371986, + "grad_norm": 0.5071449279785156, + "learning_rate": 0.0005282104437471986, + "loss": 0.3593, + "step": 16958 + }, + { + "epoch": 0.9475095678408805, + "grad_norm": 0.6504672169685364, + "learning_rate": 0.0005281824294038548, + "loss": 0.5148, + "step": 16959 + }, + { + "epoch": 0.9475654384445624, + "grad_norm": 0.6954883933067322, + "learning_rate": 0.000528154415060511, + "loss": 0.3695, + "step": 16960 + }, + { + "epoch": 0.9476213090482443, + "grad_norm": 0.4131867289543152, + "learning_rate": 0.0005281264007171672, + "loss": 0.4348, + "step": 16961 + }, + { + "epoch": 0.9476771796519261, + "grad_norm": 0.47196757793426514, + "learning_rate": 0.0005280983863738234, + "loss": 0.396, + "step": 16962 + }, + { + "epoch": 0.9477330502556081, + "grad_norm": 0.46444064378738403, + "learning_rate": 0.0005280703720304796, + "loss": 0.4405, + "step": 16963 + }, + { + "epoch": 0.9477889208592899, + "grad_norm": 0.5234575867652893, + "learning_rate": 0.0005280423576871358, + "loss": 0.4278, + "step": 16964 + }, + { + "epoch": 0.9478447914629717, + "grad_norm": 0.5690653324127197, + "learning_rate": 0.000528014343343792, + "loss": 0.4729, + "step": 16965 + }, + { + "epoch": 0.9479006620666536, + "grad_norm": 0.4626995325088501, + "learning_rate": 0.0005279863290004483, + "loss": 0.4596, + "step": 16966 + }, + { + "epoch": 0.9479565326703355, + "grad_norm": 0.5207066535949707, + "learning_rate": 0.0005279583146571044, + "loss": 0.5092, + "step": 16967 + }, + { + "epoch": 0.9480124032740174, + "grad_norm": 0.4017251133918762, + "learning_rate": 0.0005279303003137606, + "loss": 0.4448, + "step": 16968 + }, + { + "epoch": 0.9480682738776992, + "grad_norm": 0.6946123242378235, + "learning_rate": 0.0005279022859704168, + "loss": 0.5458, + "step": 16969 + }, + { + "epoch": 0.9481241444813812, + "grad_norm": 0.4121588468551636, + "learning_rate": 0.0005278742716270731, + "loss": 0.3838, + "step": 16970 + }, + { + "epoch": 0.948180015085063, + "grad_norm": 0.5486553311347961, + "learning_rate": 0.0005278462572837293, + "loss": 0.513, + "step": 16971 + }, + { + "epoch": 0.9482358856887448, + "grad_norm": 0.4249953627586365, + "learning_rate": 0.0005278182429403855, + "loss": 0.4511, + "step": 16972 + }, + { + "epoch": 0.9482917562924268, + "grad_norm": 0.39828914403915405, + "learning_rate": 0.0005277902285970418, + "loss": 0.5207, + "step": 16973 + }, + { + "epoch": 0.9483476268961086, + "grad_norm": 1.5794708728790283, + "learning_rate": 0.0005277622142536979, + "loss": 0.4087, + "step": 16974 + }, + { + "epoch": 0.9484034974997905, + "grad_norm": 0.3753533363342285, + "learning_rate": 0.0005277341999103541, + "loss": 0.4345, + "step": 16975 + }, + { + "epoch": 0.9484593681034723, + "grad_norm": 0.36351919174194336, + "learning_rate": 0.0005277061855670103, + "loss": 0.389, + "step": 16976 + }, + { + "epoch": 0.9485152387071543, + "grad_norm": 1.356364130973816, + "learning_rate": 0.0005276781712236666, + "loss": 0.5141, + "step": 16977 + }, + { + "epoch": 0.9485711093108361, + "grad_norm": 0.47384893894195557, + "learning_rate": 0.0005276501568803227, + "loss": 0.5268, + "step": 16978 + }, + { + "epoch": 0.9486269799145179, + "grad_norm": 0.4894532263278961, + "learning_rate": 0.0005276221425369789, + "loss": 0.4441, + "step": 16979 + }, + { + "epoch": 0.9486828505181999, + "grad_norm": 0.9875316023826599, + "learning_rate": 0.0005275941281936352, + "loss": 0.4626, + "step": 16980 + }, + { + "epoch": 0.9487387211218817, + "grad_norm": 0.6344906687736511, + "learning_rate": 0.0005275661138502914, + "loss": 0.4496, + "step": 16981 + }, + { + "epoch": 0.9487945917255636, + "grad_norm": 0.4931802749633789, + "learning_rate": 0.0005275380995069475, + "loss": 0.4475, + "step": 16982 + }, + { + "epoch": 0.9488504623292454, + "grad_norm": 1.3117711544036865, + "learning_rate": 0.0005275100851636037, + "loss": 0.3953, + "step": 16983 + }, + { + "epoch": 0.9489063329329274, + "grad_norm": 0.44629624485969543, + "learning_rate": 0.00052748207082026, + "loss": 0.4805, + "step": 16984 + }, + { + "epoch": 0.9489622035366092, + "grad_norm": 0.4137640595436096, + "learning_rate": 0.0005274540564769163, + "loss": 0.3748, + "step": 16985 + }, + { + "epoch": 0.949018074140291, + "grad_norm": 7.662415027618408, + "learning_rate": 0.0005274260421335723, + "loss": 0.3785, + "step": 16986 + }, + { + "epoch": 0.949073944743973, + "grad_norm": 0.487405925989151, + "learning_rate": 0.0005273980277902286, + "loss": 0.5228, + "step": 16987 + }, + { + "epoch": 0.9491298153476548, + "grad_norm": 0.561176598072052, + "learning_rate": 0.0005273700134468849, + "loss": 0.3789, + "step": 16988 + }, + { + "epoch": 0.9491856859513367, + "grad_norm": 0.4569690227508545, + "learning_rate": 0.000527341999103541, + "loss": 0.442, + "step": 16989 + }, + { + "epoch": 0.9492415565550186, + "grad_norm": 0.8086124658584595, + "learning_rate": 0.0005273139847601972, + "loss": 0.446, + "step": 16990 + }, + { + "epoch": 0.9492974271587005, + "grad_norm": 0.45089101791381836, + "learning_rate": 0.0005272859704168535, + "loss": 0.4091, + "step": 16991 + }, + { + "epoch": 0.9493532977623823, + "grad_norm": 0.4545641839504242, + "learning_rate": 0.0005272579560735097, + "loss": 0.5701, + "step": 16992 + }, + { + "epoch": 0.9494091683660641, + "grad_norm": 0.4990524649620056, + "learning_rate": 0.0005272299417301658, + "loss": 0.3468, + "step": 16993 + }, + { + "epoch": 0.9494650389697461, + "grad_norm": 0.43628573417663574, + "learning_rate": 0.000527201927386822, + "loss": 0.4185, + "step": 16994 + }, + { + "epoch": 0.9495209095734279, + "grad_norm": 0.6802843809127808, + "learning_rate": 0.0005271739130434783, + "loss": 0.4633, + "step": 16995 + }, + { + "epoch": 0.9495767801771098, + "grad_norm": 0.46935248374938965, + "learning_rate": 0.0005271458987001345, + "loss": 0.3483, + "step": 16996 + }, + { + "epoch": 0.9496326507807917, + "grad_norm": 0.9855479598045349, + "learning_rate": 0.0005271178843567906, + "loss": 0.4237, + "step": 16997 + }, + { + "epoch": 0.9496885213844736, + "grad_norm": 0.4183889627456665, + "learning_rate": 0.0005270898700134469, + "loss": 0.4815, + "step": 16998 + }, + { + "epoch": 0.9497443919881554, + "grad_norm": 0.3625223636627197, + "learning_rate": 0.0005270618556701031, + "loss": 0.3513, + "step": 16999 + }, + { + "epoch": 0.9498002625918373, + "grad_norm": 0.6226467490196228, + "learning_rate": 0.0005270338413267593, + "loss": 0.5077, + "step": 17000 + }, + { + "epoch": 0.9498002625918373, + "eval_cer": 0.08840408933695569, + "eval_loss": 0.3322269320487976, + "eval_runtime": 55.9793, + "eval_samples_per_second": 81.066, + "eval_steps_per_second": 5.073, + "eval_wer": 0.3529231027066283, + "step": 17000 + }, + { + "epoch": 0.9498561331955192, + "grad_norm": 0.37055909633636475, + "learning_rate": 0.0005270058269834154, + "loss": 0.3816, + "step": 17001 + }, + { + "epoch": 0.949912003799201, + "grad_norm": 0.33256497979164124, + "learning_rate": 0.0005269778126400718, + "loss": 0.3891, + "step": 17002 + }, + { + "epoch": 0.9499678744028829, + "grad_norm": 0.46190017461776733, + "learning_rate": 0.000526949798296728, + "loss": 0.3876, + "step": 17003 + }, + { + "epoch": 0.9500237450065648, + "grad_norm": 0.5256910920143127, + "learning_rate": 0.0005269217839533842, + "loss": 0.5066, + "step": 17004 + }, + { + "epoch": 0.9500796156102467, + "grad_norm": 0.6030314564704895, + "learning_rate": 0.0005268937696100404, + "loss": 0.434, + "step": 17005 + }, + { + "epoch": 0.9501354862139285, + "grad_norm": 0.5385448932647705, + "learning_rate": 0.0005268657552666966, + "loss": 0.3979, + "step": 17006 + }, + { + "epoch": 0.9501913568176105, + "grad_norm": 2.026538133621216, + "learning_rate": 0.0005268377409233528, + "loss": 0.4269, + "step": 17007 + }, + { + "epoch": 0.9502472274212923, + "grad_norm": 0.43505677580833435, + "learning_rate": 0.000526809726580009, + "loss": 0.4703, + "step": 17008 + }, + { + "epoch": 0.9503030980249741, + "grad_norm": 0.41358375549316406, + "learning_rate": 0.0005267817122366652, + "loss": 0.498, + "step": 17009 + }, + { + "epoch": 0.950358968628656, + "grad_norm": 0.5012059211730957, + "learning_rate": 0.0005267536978933214, + "loss": 0.4761, + "step": 17010 + }, + { + "epoch": 0.9504148392323379, + "grad_norm": 0.31651076674461365, + "learning_rate": 0.0005267256835499776, + "loss": 0.3111, + "step": 17011 + }, + { + "epoch": 0.9504707098360198, + "grad_norm": 0.35721060633659363, + "learning_rate": 0.0005266976692066338, + "loss": 0.3688, + "step": 17012 + }, + { + "epoch": 0.9505265804397016, + "grad_norm": 0.47452929615974426, + "learning_rate": 0.00052666965486329, + "loss": 0.4401, + "step": 17013 + }, + { + "epoch": 0.9505824510433836, + "grad_norm": 1.0229560136795044, + "learning_rate": 0.0005266416405199462, + "loss": 0.4753, + "step": 17014 + }, + { + "epoch": 0.9506383216470654, + "grad_norm": 0.4153377115726471, + "learning_rate": 0.0005266136261766024, + "loss": 0.423, + "step": 17015 + }, + { + "epoch": 0.9506941922507472, + "grad_norm": 0.3994157314300537, + "learning_rate": 0.0005265856118332586, + "loss": 0.4874, + "step": 17016 + }, + { + "epoch": 0.9507500628544291, + "grad_norm": 0.5483185052871704, + "learning_rate": 0.0005265575974899148, + "loss": 0.5006, + "step": 17017 + }, + { + "epoch": 0.950805933458111, + "grad_norm": 1.3403699398040771, + "learning_rate": 0.000526529583146571, + "loss": 0.3772, + "step": 17018 + }, + { + "epoch": 0.9508618040617929, + "grad_norm": 0.6062575578689575, + "learning_rate": 0.0005265015688032274, + "loss": 0.4511, + "step": 17019 + }, + { + "epoch": 0.9509176746654747, + "grad_norm": 0.42496490478515625, + "learning_rate": 0.0005264735544598835, + "loss": 0.437, + "step": 17020 + }, + { + "epoch": 0.9509735452691567, + "grad_norm": 3.2498364448547363, + "learning_rate": 0.0005264455401165397, + "loss": 0.4468, + "step": 17021 + }, + { + "epoch": 0.9510294158728385, + "grad_norm": 0.33544284105300903, + "learning_rate": 0.0005264175257731959, + "loss": 0.4428, + "step": 17022 + }, + { + "epoch": 0.9510852864765204, + "grad_norm": 0.48609110713005066, + "learning_rate": 0.0005263895114298522, + "loss": 0.4389, + "step": 17023 + }, + { + "epoch": 0.9511411570802023, + "grad_norm": 0.45292505621910095, + "learning_rate": 0.0005263614970865083, + "loss": 0.4055, + "step": 17024 + }, + { + "epoch": 0.9511970276838841, + "grad_norm": 0.41849300265312195, + "learning_rate": 0.0005263334827431645, + "loss": 0.4124, + "step": 17025 + }, + { + "epoch": 0.951252898287566, + "grad_norm": 0.40073397755622864, + "learning_rate": 0.0005263054683998207, + "loss": 0.4579, + "step": 17026 + }, + { + "epoch": 0.9513087688912478, + "grad_norm": 0.384394109249115, + "learning_rate": 0.000526277454056477, + "loss": 0.4313, + "step": 17027 + }, + { + "epoch": 0.9513646394949298, + "grad_norm": 0.3578680157661438, + "learning_rate": 0.0005262494397131331, + "loss": 0.4023, + "step": 17028 + }, + { + "epoch": 0.9514205100986116, + "grad_norm": 0.4703078866004944, + "learning_rate": 0.0005262214253697893, + "loss": 0.4783, + "step": 17029 + }, + { + "epoch": 0.9514763807022935, + "grad_norm": 0.4232523441314697, + "learning_rate": 0.0005261934110264456, + "loss": 0.4875, + "step": 17030 + }, + { + "epoch": 0.9515322513059754, + "grad_norm": 1.6248359680175781, + "learning_rate": 0.0005261653966831017, + "loss": 0.5158, + "step": 17031 + }, + { + "epoch": 0.9515881219096572, + "grad_norm": 0.3754744231700897, + "learning_rate": 0.0005261373823397579, + "loss": 0.4916, + "step": 17032 + }, + { + "epoch": 0.9516439925133391, + "grad_norm": 0.45071834325790405, + "learning_rate": 0.0005261093679964141, + "loss": 0.6265, + "step": 17033 + }, + { + "epoch": 0.9516998631170209, + "grad_norm": 0.3481754660606384, + "learning_rate": 0.0005260813536530704, + "loss": 0.4277, + "step": 17034 + }, + { + "epoch": 0.9517557337207029, + "grad_norm": 1.168172001838684, + "learning_rate": 0.0005260533393097265, + "loss": 0.4428, + "step": 17035 + }, + { + "epoch": 0.9518116043243847, + "grad_norm": 0.4499598741531372, + "learning_rate": 0.0005260253249663827, + "loss": 0.5569, + "step": 17036 + }, + { + "epoch": 0.9518674749280666, + "grad_norm": 0.4413292109966278, + "learning_rate": 0.0005259973106230391, + "loss": 0.4322, + "step": 17037 + }, + { + "epoch": 0.9519233455317485, + "grad_norm": 0.3872607946395874, + "learning_rate": 0.0005259692962796953, + "loss": 0.3826, + "step": 17038 + }, + { + "epoch": 0.9519792161354304, + "grad_norm": 1.6453983783721924, + "learning_rate": 0.0005259412819363514, + "loss": 0.4607, + "step": 17039 + }, + { + "epoch": 0.9520350867391122, + "grad_norm": 0.721885085105896, + "learning_rate": 0.0005259132675930076, + "loss": 0.5539, + "step": 17040 + }, + { + "epoch": 0.9520909573427941, + "grad_norm": 1.8035509586334229, + "learning_rate": 0.0005258852532496639, + "loss": 0.3708, + "step": 17041 + }, + { + "epoch": 0.952146827946476, + "grad_norm": 0.7800284028053284, + "learning_rate": 0.0005258572389063201, + "loss": 0.4279, + "step": 17042 + }, + { + "epoch": 0.9522026985501578, + "grad_norm": 0.5233300924301147, + "learning_rate": 0.0005258292245629762, + "loss": 0.5446, + "step": 17043 + }, + { + "epoch": 0.9522585691538397, + "grad_norm": 0.41592568159103394, + "learning_rate": 0.0005258012102196325, + "loss": 0.5411, + "step": 17044 + }, + { + "epoch": 0.9523144397575216, + "grad_norm": 0.4370788335800171, + "learning_rate": 0.0005257731958762887, + "loss": 0.3945, + "step": 17045 + }, + { + "epoch": 0.9523703103612035, + "grad_norm": 0.46973544359207153, + "learning_rate": 0.0005257451815329449, + "loss": 0.4814, + "step": 17046 + }, + { + "epoch": 0.9524261809648853, + "grad_norm": 0.5698665976524353, + "learning_rate": 0.000525717167189601, + "loss": 0.3981, + "step": 17047 + }, + { + "epoch": 0.9524820515685672, + "grad_norm": 1.7502076625823975, + "learning_rate": 0.0005256891528462573, + "loss": 0.391, + "step": 17048 + }, + { + "epoch": 0.9525379221722491, + "grad_norm": 0.4070487320423126, + "learning_rate": 0.0005256611385029135, + "loss": 0.3974, + "step": 17049 + }, + { + "epoch": 0.9525937927759309, + "grad_norm": 0.4533950686454773, + "learning_rate": 0.0005256331241595697, + "loss": 0.6034, + "step": 17050 + }, + { + "epoch": 0.9526496633796128, + "grad_norm": 8.828580856323242, + "learning_rate": 0.0005256051098162259, + "loss": 0.4971, + "step": 17051 + }, + { + "epoch": 0.9527055339832947, + "grad_norm": 0.6027664542198181, + "learning_rate": 0.0005255770954728821, + "loss": 0.3361, + "step": 17052 + }, + { + "epoch": 0.9527614045869766, + "grad_norm": 0.4003477990627289, + "learning_rate": 0.0005255490811295384, + "loss": 0.3752, + "step": 17053 + }, + { + "epoch": 0.9528172751906584, + "grad_norm": 0.4464861750602722, + "learning_rate": 0.0005255210667861945, + "loss": 0.5153, + "step": 17054 + }, + { + "epoch": 0.9528731457943403, + "grad_norm": 0.5746559500694275, + "learning_rate": 0.0005254930524428508, + "loss": 0.4544, + "step": 17055 + }, + { + "epoch": 0.9529290163980222, + "grad_norm": 1.2236591577529907, + "learning_rate": 0.000525465038099507, + "loss": 0.4707, + "step": 17056 + }, + { + "epoch": 0.952984887001704, + "grad_norm": 0.6182731986045837, + "learning_rate": 0.0005254370237561632, + "loss": 0.5127, + "step": 17057 + }, + { + "epoch": 0.953040757605386, + "grad_norm": 0.4507296681404114, + "learning_rate": 0.0005254090094128193, + "loss": 0.4053, + "step": 17058 + }, + { + "epoch": 0.9530966282090678, + "grad_norm": 0.31048256158828735, + "learning_rate": 0.0005253809950694756, + "loss": 0.3691, + "step": 17059 + }, + { + "epoch": 0.9531524988127497, + "grad_norm": 4.825919151306152, + "learning_rate": 0.0005253529807261318, + "loss": 0.4703, + "step": 17060 + }, + { + "epoch": 0.9532083694164315, + "grad_norm": 0.5245521068572998, + "learning_rate": 0.000525324966382788, + "loss": 0.3749, + "step": 17061 + }, + { + "epoch": 0.9532642400201135, + "grad_norm": 0.5478882193565369, + "learning_rate": 0.0005252969520394442, + "loss": 0.4788, + "step": 17062 + }, + { + "epoch": 0.9533201106237953, + "grad_norm": 2.2906808853149414, + "learning_rate": 0.0005252689376961004, + "loss": 0.4424, + "step": 17063 + }, + { + "epoch": 0.9533759812274771, + "grad_norm": 0.7993190884590149, + "learning_rate": 0.0005252409233527566, + "loss": 0.3919, + "step": 17064 + }, + { + "epoch": 0.9534318518311591, + "grad_norm": 0.5269777774810791, + "learning_rate": 0.0005252129090094128, + "loss": 0.4809, + "step": 17065 + }, + { + "epoch": 0.9534877224348409, + "grad_norm": 1.0030944347381592, + "learning_rate": 0.000525184894666069, + "loss": 0.3601, + "step": 17066 + }, + { + "epoch": 0.9535435930385228, + "grad_norm": 0.38343724608421326, + "learning_rate": 0.0005251568803227252, + "loss": 0.3984, + "step": 17067 + }, + { + "epoch": 0.9535994636422046, + "grad_norm": 0.7500091195106506, + "learning_rate": 0.0005251288659793814, + "loss": 0.5586, + "step": 17068 + }, + { + "epoch": 0.9536553342458866, + "grad_norm": 0.5802963972091675, + "learning_rate": 0.0005251008516360378, + "loss": 0.4487, + "step": 17069 + }, + { + "epoch": 0.9537112048495684, + "grad_norm": 0.5148309469223022, + "learning_rate": 0.0005250728372926939, + "loss": 0.5072, + "step": 17070 + }, + { + "epoch": 0.9537670754532502, + "grad_norm": 0.35455483198165894, + "learning_rate": 0.0005250448229493501, + "loss": 0.4035, + "step": 17071 + }, + { + "epoch": 0.9538229460569322, + "grad_norm": 0.45160359144210815, + "learning_rate": 0.0005250168086060063, + "loss": 0.4409, + "step": 17072 + }, + { + "epoch": 0.953878816660614, + "grad_norm": 0.48467692732810974, + "learning_rate": 0.0005249887942626625, + "loss": 0.5825, + "step": 17073 + }, + { + "epoch": 0.9539346872642959, + "grad_norm": 0.44691431522369385, + "learning_rate": 0.0005249607799193187, + "loss": 0.5255, + "step": 17074 + }, + { + "epoch": 0.9539905578679778, + "grad_norm": 0.44830387830734253, + "learning_rate": 0.0005249327655759749, + "loss": 0.486, + "step": 17075 + }, + { + "epoch": 0.9540464284716597, + "grad_norm": 1.0720757246017456, + "learning_rate": 0.0005249047512326312, + "loss": 0.5553, + "step": 17076 + }, + { + "epoch": 0.9541022990753415, + "grad_norm": 0.34121569991111755, + "learning_rate": 0.0005248767368892873, + "loss": 0.4091, + "step": 17077 + }, + { + "epoch": 0.9541581696790233, + "grad_norm": 1.9427484273910522, + "learning_rate": 0.0005248487225459435, + "loss": 0.3364, + "step": 17078 + }, + { + "epoch": 0.9542140402827053, + "grad_norm": 0.9358727931976318, + "learning_rate": 0.0005248207082025997, + "loss": 0.378, + "step": 17079 + }, + { + "epoch": 0.9542699108863871, + "grad_norm": 0.5077941417694092, + "learning_rate": 0.000524792693859256, + "loss": 0.398, + "step": 17080 + }, + { + "epoch": 0.954325781490069, + "grad_norm": 0.5041634440422058, + "learning_rate": 0.0005247646795159121, + "loss": 0.4185, + "step": 17081 + }, + { + "epoch": 0.9543816520937509, + "grad_norm": 0.43175047636032104, + "learning_rate": 0.0005247366651725683, + "loss": 0.4945, + "step": 17082 + }, + { + "epoch": 0.9544375226974328, + "grad_norm": 4.625693321228027, + "learning_rate": 0.0005247086508292246, + "loss": 0.3649, + "step": 17083 + }, + { + "epoch": 0.9544933933011146, + "grad_norm": 0.47412344813346863, + "learning_rate": 0.0005246806364858808, + "loss": 0.4164, + "step": 17084 + }, + { + "epoch": 0.9545492639047964, + "grad_norm": 0.4007730484008789, + "learning_rate": 0.0005246526221425369, + "loss": 0.45, + "step": 17085 + }, + { + "epoch": 0.9546051345084784, + "grad_norm": 0.4325503408908844, + "learning_rate": 0.0005246246077991931, + "loss": 0.4274, + "step": 17086 + }, + { + "epoch": 0.9546610051121602, + "grad_norm": 0.42037296295166016, + "learning_rate": 0.0005245965934558495, + "loss": 0.3411, + "step": 17087 + }, + { + "epoch": 0.9547168757158421, + "grad_norm": 0.45300808548927307, + "learning_rate": 0.0005245685791125057, + "loss": 0.3485, + "step": 17088 + }, + { + "epoch": 0.954772746319524, + "grad_norm": 0.489963173866272, + "learning_rate": 0.0005245405647691618, + "loss": 0.4828, + "step": 17089 + }, + { + "epoch": 0.9548286169232059, + "grad_norm": 0.5109121203422546, + "learning_rate": 0.0005245125504258181, + "loss": 0.3985, + "step": 17090 + }, + { + "epoch": 0.9548844875268877, + "grad_norm": 0.4923774302005768, + "learning_rate": 0.0005244845360824743, + "loss": 0.4509, + "step": 17091 + }, + { + "epoch": 0.9549403581305697, + "grad_norm": 0.9806409478187561, + "learning_rate": 0.0005244565217391305, + "loss": 0.4756, + "step": 17092 + }, + { + "epoch": 0.9549962287342515, + "grad_norm": 0.4521627426147461, + "learning_rate": 0.0005244285073957866, + "loss": 0.3542, + "step": 17093 + }, + { + "epoch": 0.9550520993379333, + "grad_norm": 0.7121152877807617, + "learning_rate": 0.0005244004930524429, + "loss": 0.5923, + "step": 17094 + }, + { + "epoch": 0.9551079699416152, + "grad_norm": 0.4211820363998413, + "learning_rate": 0.0005243724787090991, + "loss": 0.4358, + "step": 17095 + }, + { + "epoch": 0.9551638405452971, + "grad_norm": 0.5078240633010864, + "learning_rate": 0.0005243444643657552, + "loss": 0.4441, + "step": 17096 + }, + { + "epoch": 0.955219711148979, + "grad_norm": 0.5045053362846375, + "learning_rate": 0.0005243164500224114, + "loss": 0.4289, + "step": 17097 + }, + { + "epoch": 0.9552755817526608, + "grad_norm": 0.6429579257965088, + "learning_rate": 0.0005242884356790677, + "loss": 0.4229, + "step": 17098 + }, + { + "epoch": 0.9553314523563428, + "grad_norm": 1.0399346351623535, + "learning_rate": 0.0005242604213357239, + "loss": 0.3091, + "step": 17099 + }, + { + "epoch": 0.9553873229600246, + "grad_norm": 0.40263450145721436, + "learning_rate": 0.00052423240699238, + "loss": 0.4238, + "step": 17100 + }, + { + "epoch": 0.9554431935637064, + "grad_norm": 1.3264541625976562, + "learning_rate": 0.0005242043926490363, + "loss": 0.4662, + "step": 17101 + }, + { + "epoch": 0.9554990641673883, + "grad_norm": 1.2342501878738403, + "learning_rate": 0.0005241763783056925, + "loss": 0.4349, + "step": 17102 + }, + { + "epoch": 0.9555549347710702, + "grad_norm": 0.43239209055900574, + "learning_rate": 0.0005241483639623488, + "loss": 0.3834, + "step": 17103 + }, + { + "epoch": 0.9556108053747521, + "grad_norm": 0.35759010910987854, + "learning_rate": 0.0005241203496190049, + "loss": 0.3842, + "step": 17104 + }, + { + "epoch": 0.9556666759784339, + "grad_norm": 1.0316342115402222, + "learning_rate": 0.0005240923352756612, + "loss": 0.4511, + "step": 17105 + }, + { + "epoch": 0.9557225465821159, + "grad_norm": 0.5351161360740662, + "learning_rate": 0.0005240643209323174, + "loss": 0.4499, + "step": 17106 + }, + { + "epoch": 0.9557784171857977, + "grad_norm": 0.5189643502235413, + "learning_rate": 0.0005240363065889736, + "loss": 0.3955, + "step": 17107 + }, + { + "epoch": 0.9558342877894795, + "grad_norm": 0.6263545155525208, + "learning_rate": 0.0005240082922456298, + "loss": 0.4404, + "step": 17108 + }, + { + "epoch": 0.9558901583931615, + "grad_norm": 1.0870105028152466, + "learning_rate": 0.000523980277902286, + "loss": 0.3936, + "step": 17109 + }, + { + "epoch": 0.9559460289968433, + "grad_norm": 0.45859667658805847, + "learning_rate": 0.0005239522635589422, + "loss": 0.3554, + "step": 17110 + }, + { + "epoch": 0.9560018996005252, + "grad_norm": 0.4608641564846039, + "learning_rate": 0.0005239242492155984, + "loss": 0.479, + "step": 17111 + }, + { + "epoch": 0.956057770204207, + "grad_norm": 0.4515528082847595, + "learning_rate": 0.0005238962348722546, + "loss": 0.4078, + "step": 17112 + }, + { + "epoch": 0.956113640807889, + "grad_norm": 0.635575532913208, + "learning_rate": 0.0005238682205289108, + "loss": 0.45, + "step": 17113 + }, + { + "epoch": 0.9561695114115708, + "grad_norm": 0.6363956332206726, + "learning_rate": 0.000523840206185567, + "loss": 0.4927, + "step": 17114 + }, + { + "epoch": 0.9562253820152526, + "grad_norm": 1.8787108659744263, + "learning_rate": 0.0005238121918422233, + "loss": 0.3731, + "step": 17115 + }, + { + "epoch": 0.9562812526189346, + "grad_norm": 0.9484518766403198, + "learning_rate": 0.0005237841774988794, + "loss": 0.3881, + "step": 17116 + }, + { + "epoch": 0.9563371232226164, + "grad_norm": 0.4749348759651184, + "learning_rate": 0.0005237561631555356, + "loss": 0.4315, + "step": 17117 + }, + { + "epoch": 0.9563929938262983, + "grad_norm": 0.6451200246810913, + "learning_rate": 0.0005237281488121918, + "loss": 0.4739, + "step": 17118 + }, + { + "epoch": 0.9564488644299801, + "grad_norm": 0.6864706873893738, + "learning_rate": 0.000523700134468848, + "loss": 0.5757, + "step": 17119 + }, + { + "epoch": 0.9565047350336621, + "grad_norm": 0.605396568775177, + "learning_rate": 0.0005236721201255043, + "loss": 0.4914, + "step": 17120 + }, + { + "epoch": 0.9565606056373439, + "grad_norm": 0.49911201000213623, + "learning_rate": 0.0005236441057821605, + "loss": 0.4293, + "step": 17121 + }, + { + "epoch": 0.9566164762410257, + "grad_norm": 0.4521538019180298, + "learning_rate": 0.0005236160914388168, + "loss": 0.4017, + "step": 17122 + }, + { + "epoch": 0.9566723468447077, + "grad_norm": 0.4780918061733246, + "learning_rate": 0.0005235880770954729, + "loss": 0.4591, + "step": 17123 + }, + { + "epoch": 0.9567282174483895, + "grad_norm": 0.3456032872200012, + "learning_rate": 0.0005235600627521291, + "loss": 0.4158, + "step": 17124 + }, + { + "epoch": 0.9567840880520714, + "grad_norm": 0.4341253340244293, + "learning_rate": 0.0005235320484087853, + "loss": 0.3591, + "step": 17125 + }, + { + "epoch": 0.9568399586557532, + "grad_norm": 0.7231445908546448, + "learning_rate": 0.0005235040340654416, + "loss": 0.5194, + "step": 17126 + }, + { + "epoch": 0.9568958292594352, + "grad_norm": 0.6784906983375549, + "learning_rate": 0.0005234760197220977, + "loss": 0.4239, + "step": 17127 + }, + { + "epoch": 0.956951699863117, + "grad_norm": 1.2055819034576416, + "learning_rate": 0.0005234480053787539, + "loss": 0.4065, + "step": 17128 + }, + { + "epoch": 0.9570075704667989, + "grad_norm": 15.788054466247559, + "learning_rate": 0.0005234199910354101, + "loss": 0.483, + "step": 17129 + }, + { + "epoch": 0.9570634410704808, + "grad_norm": Infinity, + "learning_rate": 0.0005234199910354101, + "loss": 0.4318, + "step": 17130 + }, + { + "epoch": 0.9571193116741626, + "grad_norm": 0.8815737366676331, + "learning_rate": 0.0005233919766920664, + "loss": 0.5276, + "step": 17131 + }, + { + "epoch": 0.9571751822778445, + "grad_norm": 0.551853597164154, + "learning_rate": 0.0005233639623487225, + "loss": 0.438, + "step": 17132 + }, + { + "epoch": 0.9572310528815264, + "grad_norm": 0.8791380524635315, + "learning_rate": 0.0005233359480053787, + "loss": 0.4747, + "step": 17133 + }, + { + "epoch": 0.9572869234852083, + "grad_norm": 0.3781752586364746, + "learning_rate": 0.000523307933662035, + "loss": 0.4173, + "step": 17134 + }, + { + "epoch": 0.9573427940888901, + "grad_norm": 11.430985450744629, + "learning_rate": 0.0005232799193186912, + "loss": 0.4303, + "step": 17135 + }, + { + "epoch": 0.957398664692572, + "grad_norm": 1.8764374256134033, + "learning_rate": 0.0005232519049753473, + "loss": 0.5036, + "step": 17136 + }, + { + "epoch": 0.9574545352962539, + "grad_norm": 0.8740784525871277, + "learning_rate": 0.0005232238906320035, + "loss": 0.5259, + "step": 17137 + }, + { + "epoch": 0.9575104058999357, + "grad_norm": 8.604635238647461, + "learning_rate": 0.0005231958762886599, + "loss": 0.4679, + "step": 17138 + }, + { + "epoch": 0.9575662765036176, + "grad_norm": 1.5325303077697754, + "learning_rate": 0.000523167861945316, + "loss": 0.4403, + "step": 17139 + }, + { + "epoch": 0.9576221471072995, + "grad_norm": 0.6495340466499329, + "learning_rate": 0.0005231398476019722, + "loss": 0.4085, + "step": 17140 + }, + { + "epoch": 0.9576780177109814, + "grad_norm": 0.9098686575889587, + "learning_rate": 0.0005231118332586285, + "loss": 0.4552, + "step": 17141 + }, + { + "epoch": 0.9577338883146632, + "grad_norm": 0.5095335245132446, + "learning_rate": 0.0005230838189152847, + "loss": 0.4483, + "step": 17142 + }, + { + "epoch": 0.9577897589183451, + "grad_norm": 0.45895957946777344, + "learning_rate": 0.0005230558045719408, + "loss": 0.4159, + "step": 17143 + }, + { + "epoch": 0.957845629522027, + "grad_norm": 0.5063461661338806, + "learning_rate": 0.000523027790228597, + "loss": 0.5553, + "step": 17144 + }, + { + "epoch": 0.9579015001257088, + "grad_norm": 0.4340939521789551, + "learning_rate": 0.0005229997758852533, + "loss": 0.4128, + "step": 17145 + }, + { + "epoch": 0.9579573707293907, + "grad_norm": 0.4449552595615387, + "learning_rate": 0.0005229717615419095, + "loss": 0.4161, + "step": 17146 + }, + { + "epoch": 0.9580132413330726, + "grad_norm": 0.5758281350135803, + "learning_rate": 0.0005229437471985656, + "loss": 0.4114, + "step": 17147 + }, + { + "epoch": 0.9580691119367545, + "grad_norm": 0.4209095239639282, + "learning_rate": 0.0005229157328552219, + "loss": 0.4493, + "step": 17148 + }, + { + "epoch": 0.9581249825404363, + "grad_norm": 2.8667619228363037, + "learning_rate": 0.0005228877185118781, + "loss": 0.4452, + "step": 17149 + }, + { + "epoch": 0.9581808531441183, + "grad_norm": 0.5520362854003906, + "learning_rate": 0.0005228597041685343, + "loss": 0.3779, + "step": 17150 + }, + { + "epoch": 0.9582367237478001, + "grad_norm": 0.4085758328437805, + "learning_rate": 0.0005228316898251904, + "loss": 0.4747, + "step": 17151 + }, + { + "epoch": 0.958292594351482, + "grad_norm": 1.9213663339614868, + "learning_rate": 0.0005228036754818467, + "loss": 0.3145, + "step": 17152 + }, + { + "epoch": 0.9583484649551638, + "grad_norm": 0.3723694086074829, + "learning_rate": 0.000522775661138503, + "loss": 0.489, + "step": 17153 + }, + { + "epoch": 0.9584043355588457, + "grad_norm": 0.5807123780250549, + "learning_rate": 0.0005227476467951592, + "loss": 0.4397, + "step": 17154 + }, + { + "epoch": 0.9584602061625276, + "grad_norm": 0.499686062335968, + "learning_rate": 0.0005227196324518154, + "loss": 0.5496, + "step": 17155 + }, + { + "epoch": 0.9585160767662094, + "grad_norm": 0.6991187930107117, + "learning_rate": 0.0005226916181084716, + "loss": 0.4195, + "step": 17156 + }, + { + "epoch": 0.9585719473698914, + "grad_norm": 1.0362284183502197, + "learning_rate": 0.0005226636037651278, + "loss": 0.5252, + "step": 17157 + }, + { + "epoch": 0.9586278179735732, + "grad_norm": 0.4228332042694092, + "learning_rate": 0.000522635589421784, + "loss": 0.4607, + "step": 17158 + }, + { + "epoch": 0.958683688577255, + "grad_norm": 0.6698426604270935, + "learning_rate": 0.0005226075750784402, + "loss": 0.4817, + "step": 17159 + }, + { + "epoch": 0.9587395591809369, + "grad_norm": 0.619735598564148, + "learning_rate": 0.0005225795607350964, + "loss": 0.369, + "step": 17160 + }, + { + "epoch": 0.9587954297846188, + "grad_norm": 0.5029610991477966, + "learning_rate": 0.0005225515463917526, + "loss": 0.4, + "step": 17161 + }, + { + "epoch": 0.9588513003883007, + "grad_norm": 0.5941880345344543, + "learning_rate": 0.0005225235320484087, + "loss": 0.405, + "step": 17162 + }, + { + "epoch": 0.9589071709919825, + "grad_norm": 0.7919477224349976, + "learning_rate": 0.000522495517705065, + "loss": 0.4047, + "step": 17163 + }, + { + "epoch": 0.9589630415956645, + "grad_norm": 0.6445735692977905, + "learning_rate": 0.0005224675033617212, + "loss": 0.4346, + "step": 17164 + }, + { + "epoch": 0.9590189121993463, + "grad_norm": 0.3947991132736206, + "learning_rate": 0.0005224394890183774, + "loss": 0.4552, + "step": 17165 + }, + { + "epoch": 0.9590747828030282, + "grad_norm": 0.5493260622024536, + "learning_rate": 0.0005224114746750336, + "loss": 0.4403, + "step": 17166 + }, + { + "epoch": 0.9591306534067101, + "grad_norm": 0.6520140171051025, + "learning_rate": 0.0005223834603316898, + "loss": 0.4871, + "step": 17167 + }, + { + "epoch": 0.959186524010392, + "grad_norm": 0.5440923571586609, + "learning_rate": 0.000522355445988346, + "loss": 0.4041, + "step": 17168 + }, + { + "epoch": 0.9592423946140738, + "grad_norm": 0.8625870943069458, + "learning_rate": 0.0005223274316450022, + "loss": 0.2951, + "step": 17169 + }, + { + "epoch": 0.9592982652177556, + "grad_norm": 0.6925824284553528, + "learning_rate": 0.0005222994173016584, + "loss": 0.4974, + "step": 17170 + }, + { + "epoch": 0.9593541358214376, + "grad_norm": 0.7450929284095764, + "learning_rate": 0.0005222714029583147, + "loss": 0.4695, + "step": 17171 + }, + { + "epoch": 0.9594100064251194, + "grad_norm": 0.8340209126472473, + "learning_rate": 0.0005222433886149709, + "loss": 0.4501, + "step": 17172 + }, + { + "epoch": 0.9594658770288013, + "grad_norm": 1.676870584487915, + "learning_rate": 0.0005222153742716272, + "loss": 0.4696, + "step": 17173 + }, + { + "epoch": 0.9595217476324832, + "grad_norm": 0.6014431118965149, + "learning_rate": 0.0005221873599282833, + "loss": 0.5162, + "step": 17174 + }, + { + "epoch": 0.959577618236165, + "grad_norm": 0.4349935054779053, + "learning_rate": 0.0005221593455849395, + "loss": 0.5245, + "step": 17175 + }, + { + "epoch": 0.9596334888398469, + "grad_norm": 0.42558538913726807, + "learning_rate": 0.0005221313312415957, + "loss": 0.4966, + "step": 17176 + }, + { + "epoch": 0.9596893594435287, + "grad_norm": 0.5082433819770813, + "learning_rate": 0.000522103316898252, + "loss": 0.3949, + "step": 17177 + }, + { + "epoch": 0.9597452300472107, + "grad_norm": 0.3884952962398529, + "learning_rate": 0.0005220753025549081, + "loss": 0.4135, + "step": 17178 + }, + { + "epoch": 0.9598011006508925, + "grad_norm": 0.4782969653606415, + "learning_rate": 0.0005220472882115643, + "loss": 0.4902, + "step": 17179 + }, + { + "epoch": 0.9598569712545744, + "grad_norm": 0.526742160320282, + "learning_rate": 0.0005220192738682206, + "loss": 0.4396, + "step": 17180 + }, + { + "epoch": 0.9599128418582563, + "grad_norm": 1.0100321769714355, + "learning_rate": 0.0005219912595248767, + "loss": 0.4247, + "step": 17181 + }, + { + "epoch": 0.9599687124619382, + "grad_norm": 0.5757957100868225, + "learning_rate": 0.0005219632451815329, + "loss": 0.4014, + "step": 17182 + }, + { + "epoch": 0.96002458306562, + "grad_norm": 0.3941096365451813, + "learning_rate": 0.0005219352308381891, + "loss": 0.414, + "step": 17183 + }, + { + "epoch": 0.960080453669302, + "grad_norm": 0.4047401547431946, + "learning_rate": 0.0005219072164948454, + "loss": 0.3224, + "step": 17184 + }, + { + "epoch": 0.9601363242729838, + "grad_norm": 0.53811115026474, + "learning_rate": 0.0005218792021515015, + "loss": 0.4273, + "step": 17185 + }, + { + "epoch": 0.9601921948766656, + "grad_norm": 0.806690514087677, + "learning_rate": 0.0005218511878081577, + "loss": 0.6946, + "step": 17186 + }, + { + "epoch": 0.9602480654803475, + "grad_norm": 0.5350417494773865, + "learning_rate": 0.000521823173464814, + "loss": 0.3548, + "step": 17187 + }, + { + "epoch": 0.9603039360840294, + "grad_norm": 0.5022185444831848, + "learning_rate": 0.0005217951591214703, + "loss": 0.4333, + "step": 17188 + }, + { + "epoch": 0.9603598066877113, + "grad_norm": 0.45765551924705505, + "learning_rate": 0.0005217671447781264, + "loss": 0.4795, + "step": 17189 + }, + { + "epoch": 0.9604156772913931, + "grad_norm": 0.4995516240596771, + "learning_rate": 0.0005217391304347826, + "loss": 0.4577, + "step": 17190 + }, + { + "epoch": 0.960471547895075, + "grad_norm": 1.4494258165359497, + "learning_rate": 0.0005217111160914389, + "loss": 0.5639, + "step": 17191 + }, + { + "epoch": 0.9605274184987569, + "grad_norm": 0.42523080110549927, + "learning_rate": 0.0005216831017480951, + "loss": 0.3294, + "step": 17192 + }, + { + "epoch": 0.9605832891024387, + "grad_norm": 0.5306354761123657, + "learning_rate": 0.0005216550874047512, + "loss": 0.5154, + "step": 17193 + }, + { + "epoch": 0.9606391597061206, + "grad_norm": 0.5577884912490845, + "learning_rate": 0.0005216270730614075, + "loss": 0.3844, + "step": 17194 + }, + { + "epoch": 0.9606950303098025, + "grad_norm": 0.6969954967498779, + "learning_rate": 0.0005215990587180637, + "loss": 0.374, + "step": 17195 + }, + { + "epoch": 0.9607509009134844, + "grad_norm": 0.6692930459976196, + "learning_rate": 0.0005215710443747199, + "loss": 0.3371, + "step": 17196 + }, + { + "epoch": 0.9608067715171662, + "grad_norm": 0.5717565417289734, + "learning_rate": 0.000521543030031376, + "loss": 0.604, + "step": 17197 + }, + { + "epoch": 0.9608626421208482, + "grad_norm": 0.7392762899398804, + "learning_rate": 0.0005215150156880323, + "loss": 0.4506, + "step": 17198 + }, + { + "epoch": 0.96091851272453, + "grad_norm": 0.36200976371765137, + "learning_rate": 0.0005214870013446885, + "loss": 0.4973, + "step": 17199 + }, + { + "epoch": 0.9609743833282118, + "grad_norm": 0.48262348771095276, + "learning_rate": 0.0005214589870013447, + "loss": 0.529, + "step": 17200 + }, + { + "epoch": 0.9610302539318938, + "grad_norm": 0.7167149782180786, + "learning_rate": 0.0005214309726580008, + "loss": 0.6066, + "step": 17201 + }, + { + "epoch": 0.9610861245355756, + "grad_norm": 4.362346172332764, + "learning_rate": 0.0005214029583146571, + "loss": 0.4188, + "step": 17202 + }, + { + "epoch": 0.9611419951392575, + "grad_norm": 0.4748220145702362, + "learning_rate": 0.0005213749439713133, + "loss": 0.4722, + "step": 17203 + }, + { + "epoch": 0.9611978657429393, + "grad_norm": 0.46272915601730347, + "learning_rate": 0.0005213469296279694, + "loss": 0.4264, + "step": 17204 + }, + { + "epoch": 0.9612537363466213, + "grad_norm": 3.2897708415985107, + "learning_rate": 0.0005213189152846258, + "loss": 0.4433, + "step": 17205 + }, + { + "epoch": 0.9613096069503031, + "grad_norm": 0.6164407730102539, + "learning_rate": 0.000521290900941282, + "loss": 0.392, + "step": 17206 + }, + { + "epoch": 0.9613654775539849, + "grad_norm": 0.6163704991340637, + "learning_rate": 0.0005212628865979382, + "loss": 0.3822, + "step": 17207 + }, + { + "epoch": 0.9614213481576669, + "grad_norm": 0.43602070212364197, + "learning_rate": 0.0005212348722545943, + "loss": 0.4584, + "step": 17208 + }, + { + "epoch": 0.9614772187613487, + "grad_norm": 0.8693935871124268, + "learning_rate": 0.0005212068579112506, + "loss": 0.4454, + "step": 17209 + }, + { + "epoch": 0.9615330893650306, + "grad_norm": 0.46474790573120117, + "learning_rate": 0.0005211788435679068, + "loss": 0.4024, + "step": 17210 + }, + { + "epoch": 0.9615889599687124, + "grad_norm": 0.8614341616630554, + "learning_rate": 0.000521150829224563, + "loss": 0.5076, + "step": 17211 + }, + { + "epoch": 0.9616448305723944, + "grad_norm": 0.42853009700775146, + "learning_rate": 0.0005211228148812192, + "loss": 0.346, + "step": 17212 + }, + { + "epoch": 0.9617007011760762, + "grad_norm": 0.8958480954170227, + "learning_rate": 0.0005210948005378754, + "loss": 0.4209, + "step": 17213 + }, + { + "epoch": 0.961756571779758, + "grad_norm": 0.3864813446998596, + "learning_rate": 0.0005210667861945316, + "loss": 0.4272, + "step": 17214 + }, + { + "epoch": 0.96181244238344, + "grad_norm": 0.6139529943466187, + "learning_rate": 0.0005210387718511878, + "loss": 0.5407, + "step": 17215 + }, + { + "epoch": 0.9618683129871218, + "grad_norm": 1.145116925239563, + "learning_rate": 0.000521010757507844, + "loss": 0.4667, + "step": 17216 + }, + { + "epoch": 0.9619241835908037, + "grad_norm": 0.43921464681625366, + "learning_rate": 0.0005209827431645002, + "loss": 0.44, + "step": 17217 + }, + { + "epoch": 0.9619800541944856, + "grad_norm": 0.7424225211143494, + "learning_rate": 0.0005209547288211564, + "loss": 0.4812, + "step": 17218 + }, + { + "epoch": 0.9620359247981675, + "grad_norm": 0.5331676602363586, + "learning_rate": 0.0005209267144778127, + "loss": 0.4171, + "step": 17219 + }, + { + "epoch": 0.9620917954018493, + "grad_norm": 1.0546784400939941, + "learning_rate": 0.0005208987001344688, + "loss": 0.4606, + "step": 17220 + }, + { + "epoch": 0.9621476660055311, + "grad_norm": 0.7611668705940247, + "learning_rate": 0.000520870685791125, + "loss": 0.5089, + "step": 17221 + }, + { + "epoch": 0.9622035366092131, + "grad_norm": 1.0218212604522705, + "learning_rate": 0.0005208426714477813, + "loss": 0.4216, + "step": 17222 + }, + { + "epoch": 0.9622594072128949, + "grad_norm": 0.5616165995597839, + "learning_rate": 0.0005208146571044375, + "loss": 0.4553, + "step": 17223 + }, + { + "epoch": 0.9623152778165768, + "grad_norm": 0.5865601897239685, + "learning_rate": 0.0005207866427610937, + "loss": 0.4541, + "step": 17224 + }, + { + "epoch": 0.9623711484202587, + "grad_norm": 0.47823366522789, + "learning_rate": 0.0005207586284177499, + "loss": 0.3715, + "step": 17225 + }, + { + "epoch": 0.9624270190239406, + "grad_norm": 0.5800397992134094, + "learning_rate": 0.0005207306140744062, + "loss": 0.4654, + "step": 17226 + }, + { + "epoch": 0.9624828896276224, + "grad_norm": 0.661803126335144, + "learning_rate": 0.0005207025997310623, + "loss": 0.6001, + "step": 17227 + }, + { + "epoch": 0.9625387602313042, + "grad_norm": 0.6808329820632935, + "learning_rate": 0.0005206745853877185, + "loss": 0.3935, + "step": 17228 + }, + { + "epoch": 0.9625946308349862, + "grad_norm": 0.8476242423057556, + "learning_rate": 0.0005206465710443747, + "loss": 0.4084, + "step": 17229 + }, + { + "epoch": 0.962650501438668, + "grad_norm": 0.470468670129776, + "learning_rate": 0.000520618556701031, + "loss": 0.4639, + "step": 17230 + }, + { + "epoch": 0.9627063720423499, + "grad_norm": 0.5576077103614807, + "learning_rate": 0.0005205905423576871, + "loss": 0.3991, + "step": 17231 + }, + { + "epoch": 0.9627622426460318, + "grad_norm": 1.3218995332717896, + "learning_rate": 0.0005205625280143433, + "loss": 0.4064, + "step": 17232 + }, + { + "epoch": 0.9628181132497137, + "grad_norm": 0.6123731732368469, + "learning_rate": 0.0005205345136709996, + "loss": 0.4591, + "step": 17233 + }, + { + "epoch": 0.9628739838533955, + "grad_norm": 0.5384792685508728, + "learning_rate": 0.0005205064993276558, + "loss": 0.3702, + "step": 17234 + }, + { + "epoch": 0.9629298544570775, + "grad_norm": 0.5209780335426331, + "learning_rate": 0.0005204784849843119, + "loss": 0.4045, + "step": 17235 + }, + { + "epoch": 0.9629857250607593, + "grad_norm": 0.6628375053405762, + "learning_rate": 0.0005204504706409681, + "loss": 0.4916, + "step": 17236 + }, + { + "epoch": 0.9630415956644411, + "grad_norm": 0.3468601405620575, + "learning_rate": 0.0005204224562976245, + "loss": 0.3051, + "step": 17237 + }, + { + "epoch": 0.963097466268123, + "grad_norm": 0.49270951747894287, + "learning_rate": 0.0005203944419542807, + "loss": 0.442, + "step": 17238 + }, + { + "epoch": 0.9631533368718049, + "grad_norm": 0.5540347695350647, + "learning_rate": 0.0005203664276109368, + "loss": 0.5936, + "step": 17239 + }, + { + "epoch": 0.9632092074754868, + "grad_norm": 1.2834192514419556, + "learning_rate": 0.000520338413267593, + "loss": 0.5701, + "step": 17240 + }, + { + "epoch": 0.9632650780791686, + "grad_norm": 0.41044533252716064, + "learning_rate": 0.0005203103989242493, + "loss": 0.4657, + "step": 17241 + }, + { + "epoch": 0.9633209486828506, + "grad_norm": 0.5814249515533447, + "learning_rate": 0.0005202823845809055, + "loss": 0.4847, + "step": 17242 + }, + { + "epoch": 0.9633768192865324, + "grad_norm": 0.8546634316444397, + "learning_rate": 0.0005202543702375616, + "loss": 0.5312, + "step": 17243 + }, + { + "epoch": 0.9634326898902142, + "grad_norm": 0.649075448513031, + "learning_rate": 0.0005202263558942179, + "loss": 0.4454, + "step": 17244 + }, + { + "epoch": 0.9634885604938961, + "grad_norm": 0.4625098407268524, + "learning_rate": 0.0005201983415508741, + "loss": 0.4226, + "step": 17245 + }, + { + "epoch": 0.963544431097578, + "grad_norm": 3.6452746391296387, + "learning_rate": 0.0005201703272075302, + "loss": 0.3658, + "step": 17246 + }, + { + "epoch": 0.9636003017012599, + "grad_norm": 0.4734138250350952, + "learning_rate": 0.0005201423128641864, + "loss": 0.3755, + "step": 17247 + }, + { + "epoch": 0.9636561723049417, + "grad_norm": 0.45444628596305847, + "learning_rate": 0.0005201142985208427, + "loss": 0.4112, + "step": 17248 + }, + { + "epoch": 0.9637120429086237, + "grad_norm": 0.3693031668663025, + "learning_rate": 0.0005200862841774989, + "loss": 0.3968, + "step": 17249 + }, + { + "epoch": 0.9637679135123055, + "grad_norm": 0.7344266772270203, + "learning_rate": 0.000520058269834155, + "loss": 0.4553, + "step": 17250 + }, + { + "epoch": 0.9638237841159873, + "grad_norm": 0.40571048855781555, + "learning_rate": 0.0005200302554908113, + "loss": 0.3859, + "step": 17251 + }, + { + "epoch": 0.9638796547196693, + "grad_norm": 0.6711070537567139, + "learning_rate": 0.0005200022411474675, + "loss": 0.4948, + "step": 17252 + }, + { + "epoch": 0.9639355253233511, + "grad_norm": 0.5035292506217957, + "learning_rate": 0.0005199742268041237, + "loss": 0.3918, + "step": 17253 + }, + { + "epoch": 0.963991395927033, + "grad_norm": 0.4304666817188263, + "learning_rate": 0.0005199462124607798, + "loss": 0.3805, + "step": 17254 + }, + { + "epoch": 0.9640472665307148, + "grad_norm": 0.900791585445404, + "learning_rate": 0.0005199181981174362, + "loss": 0.5079, + "step": 17255 + }, + { + "epoch": 0.9641031371343968, + "grad_norm": 1.4875097274780273, + "learning_rate": 0.0005198901837740924, + "loss": 0.6334, + "step": 17256 + }, + { + "epoch": 0.9641590077380786, + "grad_norm": 0.39005568623542786, + "learning_rate": 0.0005198621694307486, + "loss": 0.4305, + "step": 17257 + }, + { + "epoch": 0.9642148783417605, + "grad_norm": 0.4604758620262146, + "learning_rate": 0.0005198341550874048, + "loss": 0.4727, + "step": 17258 + }, + { + "epoch": 0.9642707489454424, + "grad_norm": 0.7112866044044495, + "learning_rate": 0.000519806140744061, + "loss": 0.4638, + "step": 17259 + }, + { + "epoch": 0.9643266195491242, + "grad_norm": 0.40404847264289856, + "learning_rate": 0.0005197781264007172, + "loss": 0.4219, + "step": 17260 + }, + { + "epoch": 0.9643824901528061, + "grad_norm": 0.3784729242324829, + "learning_rate": 0.0005197501120573734, + "loss": 0.3927, + "step": 17261 + }, + { + "epoch": 0.9644383607564879, + "grad_norm": 0.5004622340202332, + "learning_rate": 0.0005197220977140296, + "loss": 0.4461, + "step": 17262 + }, + { + "epoch": 0.9644942313601699, + "grad_norm": 0.3839932084083557, + "learning_rate": 0.0005196940833706858, + "loss": 0.4117, + "step": 17263 + }, + { + "epoch": 0.9645501019638517, + "grad_norm": 0.3532012104988098, + "learning_rate": 0.000519666069027342, + "loss": 0.4438, + "step": 17264 + }, + { + "epoch": 0.9646059725675336, + "grad_norm": 0.5254203677177429, + "learning_rate": 0.0005196380546839982, + "loss": 0.5079, + "step": 17265 + }, + { + "epoch": 0.9646618431712155, + "grad_norm": 0.39281243085861206, + "learning_rate": 0.0005196100403406544, + "loss": 0.4217, + "step": 17266 + }, + { + "epoch": 0.9647177137748973, + "grad_norm": 0.4448240399360657, + "learning_rate": 0.0005195820259973106, + "loss": 0.4083, + "step": 17267 + }, + { + "epoch": 0.9647735843785792, + "grad_norm": 0.34404462575912476, + "learning_rate": 0.0005195540116539668, + "loss": 0.4465, + "step": 17268 + }, + { + "epoch": 0.9648294549822611, + "grad_norm": 0.415787011384964, + "learning_rate": 0.000519525997310623, + "loss": 0.4249, + "step": 17269 + }, + { + "epoch": 0.964885325585943, + "grad_norm": 0.3876666724681854, + "learning_rate": 0.0005194979829672792, + "loss": 0.4281, + "step": 17270 + }, + { + "epoch": 0.9649411961896248, + "grad_norm": 0.48603811860084534, + "learning_rate": 0.0005194699686239355, + "loss": 0.4528, + "step": 17271 + }, + { + "epoch": 0.9649970667933067, + "grad_norm": 0.3219534754753113, + "learning_rate": 0.0005194419542805917, + "loss": 0.349, + "step": 17272 + }, + { + "epoch": 0.9650529373969886, + "grad_norm": 0.5017842650413513, + "learning_rate": 0.0005194139399372479, + "loss": 0.3398, + "step": 17273 + }, + { + "epoch": 0.9651088080006704, + "grad_norm": 1.1114709377288818, + "learning_rate": 0.0005193859255939041, + "loss": 0.3894, + "step": 17274 + }, + { + "epoch": 0.9651646786043523, + "grad_norm": 1.2671458721160889, + "learning_rate": 0.0005193579112505603, + "loss": 0.3724, + "step": 17275 + }, + { + "epoch": 0.9652205492080342, + "grad_norm": 0.5772609114646912, + "learning_rate": 0.0005193298969072166, + "loss": 0.5492, + "step": 17276 + }, + { + "epoch": 0.9652764198117161, + "grad_norm": 0.9044739603996277, + "learning_rate": 0.0005193018825638727, + "loss": 0.5476, + "step": 17277 + }, + { + "epoch": 0.9653322904153979, + "grad_norm": 0.4353660047054291, + "learning_rate": 0.0005192738682205289, + "loss": 0.4195, + "step": 17278 + }, + { + "epoch": 0.9653881610190798, + "grad_norm": 0.46696341037750244, + "learning_rate": 0.0005192458538771851, + "loss": 0.4456, + "step": 17279 + }, + { + "epoch": 0.9654440316227617, + "grad_norm": 0.7237882018089294, + "learning_rate": 0.0005192178395338414, + "loss": 0.3791, + "step": 17280 + }, + { + "epoch": 0.9654999022264436, + "grad_norm": 0.5471433997154236, + "learning_rate": 0.0005191898251904975, + "loss": 0.4635, + "step": 17281 + }, + { + "epoch": 0.9655557728301254, + "grad_norm": 0.6604092121124268, + "learning_rate": 0.0005191618108471537, + "loss": 0.4616, + "step": 17282 + }, + { + "epoch": 0.9656116434338073, + "grad_norm": 0.5308656692504883, + "learning_rate": 0.00051913379650381, + "loss": 0.3663, + "step": 17283 + }, + { + "epoch": 0.9656675140374892, + "grad_norm": 1.036009430885315, + "learning_rate": 0.0005191057821604662, + "loss": 0.4426, + "step": 17284 + }, + { + "epoch": 0.965723384641171, + "grad_norm": 0.5423424243927002, + "learning_rate": 0.0005190777678171223, + "loss": 0.4666, + "step": 17285 + }, + { + "epoch": 0.965779255244853, + "grad_norm": 0.6737112998962402, + "learning_rate": 0.0005190497534737785, + "loss": 0.4321, + "step": 17286 + }, + { + "epoch": 0.9658351258485348, + "grad_norm": 0.4128410518169403, + "learning_rate": 0.0005190217391304349, + "loss": 0.3543, + "step": 17287 + }, + { + "epoch": 0.9658909964522167, + "grad_norm": 0.4381726384162903, + "learning_rate": 0.000518993724787091, + "loss": 0.4319, + "step": 17288 + }, + { + "epoch": 0.9659468670558985, + "grad_norm": 0.4450848698616028, + "learning_rate": 0.0005189657104437472, + "loss": 0.3702, + "step": 17289 + }, + { + "epoch": 0.9660027376595804, + "grad_norm": 0.45573917031288147, + "learning_rate": 0.0005189376961004035, + "loss": 0.4621, + "step": 17290 + }, + { + "epoch": 0.9660586082632623, + "grad_norm": 0.3826253414154053, + "learning_rate": 0.0005189096817570597, + "loss": 0.5286, + "step": 17291 + }, + { + "epoch": 0.9661144788669441, + "grad_norm": 0.9885721206665039, + "learning_rate": 0.0005188816674137158, + "loss": 0.5367, + "step": 17292 + }, + { + "epoch": 0.9661703494706261, + "grad_norm": 0.36320751905441284, + "learning_rate": 0.000518853653070372, + "loss": 0.5096, + "step": 17293 + }, + { + "epoch": 0.9662262200743079, + "grad_norm": 0.4959706664085388, + "learning_rate": 0.0005188256387270283, + "loss": 0.4252, + "step": 17294 + }, + { + "epoch": 0.9662820906779898, + "grad_norm": 0.5386580228805542, + "learning_rate": 0.0005187976243836845, + "loss": 0.4287, + "step": 17295 + }, + { + "epoch": 0.9663379612816716, + "grad_norm": 3.0614683628082275, + "learning_rate": 0.0005187696100403406, + "loss": 0.484, + "step": 17296 + }, + { + "epoch": 0.9663938318853535, + "grad_norm": 0.41045185923576355, + "learning_rate": 0.0005187415956969969, + "loss": 0.4136, + "step": 17297 + }, + { + "epoch": 0.9664497024890354, + "grad_norm": 0.4605971574783325, + "learning_rate": 0.0005187135813536531, + "loss": 0.4496, + "step": 17298 + }, + { + "epoch": 0.9665055730927172, + "grad_norm": 0.800525963306427, + "learning_rate": 0.0005186855670103093, + "loss": 0.5213, + "step": 17299 + }, + { + "epoch": 0.9665614436963992, + "grad_norm": 2.7242980003356934, + "learning_rate": 0.0005186575526669654, + "loss": 0.4361, + "step": 17300 + }, + { + "epoch": 0.966617314300081, + "grad_norm": 0.5764836072921753, + "learning_rate": 0.0005186295383236217, + "loss": 0.468, + "step": 17301 + }, + { + "epoch": 0.9666731849037629, + "grad_norm": 0.995600700378418, + "learning_rate": 0.0005186015239802779, + "loss": 0.47, + "step": 17302 + }, + { + "epoch": 0.9667290555074448, + "grad_norm": 0.4519959092140198, + "learning_rate": 0.0005185735096369341, + "loss": 0.4995, + "step": 17303 + }, + { + "epoch": 0.9667849261111267, + "grad_norm": 0.5684428215026855, + "learning_rate": 0.0005185454952935902, + "loss": 0.3877, + "step": 17304 + }, + { + "epoch": 0.9668407967148085, + "grad_norm": 0.524537205696106, + "learning_rate": 0.0005185174809502466, + "loss": 0.3872, + "step": 17305 + }, + { + "epoch": 0.9668966673184903, + "grad_norm": 0.44505608081817627, + "learning_rate": 0.0005184894666069028, + "loss": 0.5128, + "step": 17306 + }, + { + "epoch": 0.9669525379221723, + "grad_norm": 0.4474577307701111, + "learning_rate": 0.000518461452263559, + "loss": 0.5271, + "step": 17307 + }, + { + "epoch": 0.9670084085258541, + "grad_norm": 0.37446415424346924, + "learning_rate": 0.0005184334379202152, + "loss": 0.3303, + "step": 17308 + }, + { + "epoch": 0.967064279129536, + "grad_norm": 0.37544044852256775, + "learning_rate": 0.0005184054235768714, + "loss": 0.4099, + "step": 17309 + }, + { + "epoch": 0.9671201497332179, + "grad_norm": 3.628206253051758, + "learning_rate": 0.0005183774092335276, + "loss": 0.37, + "step": 17310 + }, + { + "epoch": 0.9671760203368998, + "grad_norm": 0.45080795884132385, + "learning_rate": 0.0005183493948901837, + "loss": 0.4371, + "step": 17311 + }, + { + "epoch": 0.9672318909405816, + "grad_norm": 0.3915563225746155, + "learning_rate": 0.00051832138054684, + "loss": 0.3731, + "step": 17312 + }, + { + "epoch": 0.9672877615442634, + "grad_norm": 0.43676599860191345, + "learning_rate": 0.0005182933662034962, + "loss": 0.4542, + "step": 17313 + }, + { + "epoch": 0.9673436321479454, + "grad_norm": 0.34777671098709106, + "learning_rate": 0.0005182653518601524, + "loss": 0.3891, + "step": 17314 + }, + { + "epoch": 0.9673995027516272, + "grad_norm": 0.6327469348907471, + "learning_rate": 0.0005182373375168086, + "loss": 0.4492, + "step": 17315 + }, + { + "epoch": 0.9674553733553091, + "grad_norm": 0.42368417978286743, + "learning_rate": 0.0005182093231734648, + "loss": 0.3723, + "step": 17316 + }, + { + "epoch": 0.967511243958991, + "grad_norm": 0.44452333450317383, + "learning_rate": 0.000518181308830121, + "loss": 0.5151, + "step": 17317 + }, + { + "epoch": 0.9675671145626729, + "grad_norm": 0.3373429477214813, + "learning_rate": 0.0005181532944867772, + "loss": 0.4314, + "step": 17318 + }, + { + "epoch": 0.9676229851663547, + "grad_norm": 0.5156334042549133, + "learning_rate": 0.0005181252801434334, + "loss": 0.4053, + "step": 17319 + }, + { + "epoch": 0.9676788557700366, + "grad_norm": 0.4157906472682953, + "learning_rate": 0.0005180972658000896, + "loss": 0.5043, + "step": 17320 + }, + { + "epoch": 0.9677347263737185, + "grad_norm": 0.35986873507499695, + "learning_rate": 0.0005180692514567459, + "loss": 0.3986, + "step": 17321 + }, + { + "epoch": 0.9677905969774003, + "grad_norm": 0.7739237546920776, + "learning_rate": 0.0005180412371134022, + "loss": 0.4587, + "step": 17322 + }, + { + "epoch": 0.9678464675810822, + "grad_norm": 0.422702431678772, + "learning_rate": 0.0005180132227700583, + "loss": 0.4629, + "step": 17323 + }, + { + "epoch": 0.9679023381847641, + "grad_norm": 0.3431282639503479, + "learning_rate": 0.0005179852084267145, + "loss": 0.4978, + "step": 17324 + }, + { + "epoch": 0.967958208788446, + "grad_norm": 0.5443609952926636, + "learning_rate": 0.0005179571940833707, + "loss": 0.3568, + "step": 17325 + }, + { + "epoch": 0.9680140793921278, + "grad_norm": 0.32538822293281555, + "learning_rate": 0.000517929179740027, + "loss": 0.3862, + "step": 17326 + }, + { + "epoch": 0.9680699499958098, + "grad_norm": 0.45348215103149414, + "learning_rate": 0.0005179011653966831, + "loss": 0.3834, + "step": 17327 + }, + { + "epoch": 0.9681258205994916, + "grad_norm": 0.3518955409526825, + "learning_rate": 0.0005178731510533393, + "loss": 0.3865, + "step": 17328 + }, + { + "epoch": 0.9681816912031734, + "grad_norm": 0.8201083540916443, + "learning_rate": 0.0005178451367099956, + "loss": 0.5792, + "step": 17329 + }, + { + "epoch": 0.9682375618068553, + "grad_norm": 0.4362366795539856, + "learning_rate": 0.0005178171223666517, + "loss": 0.4195, + "step": 17330 + }, + { + "epoch": 0.9682934324105372, + "grad_norm": 0.6178483366966248, + "learning_rate": 0.0005177891080233079, + "loss": 0.4551, + "step": 17331 + }, + { + "epoch": 0.9683493030142191, + "grad_norm": 0.45065218210220337, + "learning_rate": 0.0005177610936799641, + "loss": 0.5683, + "step": 17332 + }, + { + "epoch": 0.9684051736179009, + "grad_norm": 0.44094201922416687, + "learning_rate": 0.0005177330793366204, + "loss": 0.3492, + "step": 17333 + }, + { + "epoch": 0.9684610442215829, + "grad_norm": 0.5259076952934265, + "learning_rate": 0.0005177050649932765, + "loss": 0.5238, + "step": 17334 + }, + { + "epoch": 0.9685169148252647, + "grad_norm": 0.4957464337348938, + "learning_rate": 0.0005176770506499327, + "loss": 0.5907, + "step": 17335 + }, + { + "epoch": 0.9685727854289465, + "grad_norm": 0.47585153579711914, + "learning_rate": 0.000517649036306589, + "loss": 0.4064, + "step": 17336 + }, + { + "epoch": 0.9686286560326285, + "grad_norm": 0.721619188785553, + "learning_rate": 0.0005176210219632453, + "loss": 0.4162, + "step": 17337 + }, + { + "epoch": 0.9686845266363103, + "grad_norm": 0.440001517534256, + "learning_rate": 0.0005175930076199013, + "loss": 0.4252, + "step": 17338 + }, + { + "epoch": 0.9687403972399922, + "grad_norm": 1.0307329893112183, + "learning_rate": 0.0005175649932765576, + "loss": 0.3739, + "step": 17339 + }, + { + "epoch": 0.968796267843674, + "grad_norm": 0.8608969449996948, + "learning_rate": 0.0005175369789332139, + "loss": 0.467, + "step": 17340 + }, + { + "epoch": 0.968852138447356, + "grad_norm": 0.9922165870666504, + "learning_rate": 0.0005175089645898701, + "loss": 0.4518, + "step": 17341 + }, + { + "epoch": 0.9689080090510378, + "grad_norm": 0.564311146736145, + "learning_rate": 0.0005174809502465262, + "loss": 0.4647, + "step": 17342 + }, + { + "epoch": 0.9689638796547196, + "grad_norm": 0.4266946017742157, + "learning_rate": 0.0005174529359031824, + "loss": 0.4211, + "step": 17343 + }, + { + "epoch": 0.9690197502584016, + "grad_norm": 0.7368376851081848, + "learning_rate": 0.0005174249215598387, + "loss": 0.3437, + "step": 17344 + }, + { + "epoch": 0.9690756208620834, + "grad_norm": 0.680635392665863, + "learning_rate": 0.0005173969072164949, + "loss": 0.5129, + "step": 17345 + }, + { + "epoch": 0.9691314914657653, + "grad_norm": 0.5327621698379517, + "learning_rate": 0.000517368892873151, + "loss": 0.3878, + "step": 17346 + }, + { + "epoch": 0.9691873620694471, + "grad_norm": 0.4302014410495758, + "learning_rate": 0.0005173408785298073, + "loss": 0.4906, + "step": 17347 + }, + { + "epoch": 0.9692432326731291, + "grad_norm": 1.1569302082061768, + "learning_rate": 0.0005173128641864635, + "loss": 0.457, + "step": 17348 + }, + { + "epoch": 0.9692991032768109, + "grad_norm": 0.414275199174881, + "learning_rate": 0.0005172848498431197, + "loss": 0.4828, + "step": 17349 + }, + { + "epoch": 0.9693549738804927, + "grad_norm": 0.5252900719642639, + "learning_rate": 0.0005172568354997758, + "loss": 0.4492, + "step": 17350 + }, + { + "epoch": 0.9694108444841747, + "grad_norm": 16.11258316040039, + "learning_rate": 0.0005172288211564321, + "loss": 0.5263, + "step": 17351 + }, + { + "epoch": 0.9694667150878565, + "grad_norm": 0.514038622379303, + "learning_rate": 0.0005172008068130883, + "loss": 0.4018, + "step": 17352 + }, + { + "epoch": 0.9695225856915384, + "grad_norm": 0.5205409526824951, + "learning_rate": 0.0005171727924697444, + "loss": 0.4297, + "step": 17353 + }, + { + "epoch": 0.9695784562952202, + "grad_norm": 0.7178422212600708, + "learning_rate": 0.0005171447781264008, + "loss": 0.4398, + "step": 17354 + }, + { + "epoch": 0.9696343268989022, + "grad_norm": 0.42390623688697815, + "learning_rate": 0.000517116763783057, + "loss": 0.4286, + "step": 17355 + }, + { + "epoch": 0.969690197502584, + "grad_norm": 0.5301945805549622, + "learning_rate": 0.0005170887494397132, + "loss": 0.5629, + "step": 17356 + }, + { + "epoch": 0.9697460681062658, + "grad_norm": 0.3826347887516022, + "learning_rate": 0.0005170607350963693, + "loss": 0.4084, + "step": 17357 + }, + { + "epoch": 0.9698019387099478, + "grad_norm": 0.33248457312583923, + "learning_rate": 0.0005170327207530256, + "loss": 0.3096, + "step": 17358 + }, + { + "epoch": 0.9698578093136296, + "grad_norm": 0.563514232635498, + "learning_rate": 0.0005170047064096818, + "loss": 0.5083, + "step": 17359 + }, + { + "epoch": 0.9699136799173115, + "grad_norm": 0.5281736254692078, + "learning_rate": 0.000516976692066338, + "loss": 0.491, + "step": 17360 + }, + { + "epoch": 0.9699695505209934, + "grad_norm": 0.46237388253211975, + "learning_rate": 0.0005169486777229942, + "loss": 0.5521, + "step": 17361 + }, + { + "epoch": 0.9700254211246753, + "grad_norm": 0.7334683537483215, + "learning_rate": 0.0005169206633796504, + "loss": 0.4291, + "step": 17362 + }, + { + "epoch": 0.9700812917283571, + "grad_norm": 0.5309751033782959, + "learning_rate": 0.0005168926490363066, + "loss": 0.4551, + "step": 17363 + }, + { + "epoch": 0.970137162332039, + "grad_norm": 0.8423299789428711, + "learning_rate": 0.0005168646346929628, + "loss": 0.3717, + "step": 17364 + }, + { + "epoch": 0.9701930329357209, + "grad_norm": 0.37114059925079346, + "learning_rate": 0.000516836620349619, + "loss": 0.4777, + "step": 17365 + }, + { + "epoch": 0.9702489035394027, + "grad_norm": 0.5375327467918396, + "learning_rate": 0.0005168086060062752, + "loss": 0.494, + "step": 17366 + }, + { + "epoch": 0.9703047741430846, + "grad_norm": 0.41715267300605774, + "learning_rate": 0.0005167805916629314, + "loss": 0.4231, + "step": 17367 + }, + { + "epoch": 0.9703606447467665, + "grad_norm": 8.411275863647461, + "learning_rate": 0.0005167525773195877, + "loss": 0.4958, + "step": 17368 + }, + { + "epoch": 0.9704165153504484, + "grad_norm": 0.6246569156646729, + "learning_rate": 0.0005167245629762438, + "loss": 0.4493, + "step": 17369 + }, + { + "epoch": 0.9704723859541302, + "grad_norm": 0.6883368492126465, + "learning_rate": 0.0005166965486329, + "loss": 0.5238, + "step": 17370 + }, + { + "epoch": 0.970528256557812, + "grad_norm": 0.364326149225235, + "learning_rate": 0.0005166685342895562, + "loss": 0.4192, + "step": 17371 + }, + { + "epoch": 0.970584127161494, + "grad_norm": 0.6030899286270142, + "learning_rate": 0.0005166405199462125, + "loss": 0.2714, + "step": 17372 + }, + { + "epoch": 0.9706399977651758, + "grad_norm": 0.6610691547393799, + "learning_rate": 0.0005166125056028687, + "loss": 0.5372, + "step": 17373 + }, + { + "epoch": 0.9706958683688577, + "grad_norm": 0.5351042747497559, + "learning_rate": 0.0005165844912595249, + "loss": 0.3762, + "step": 17374 + }, + { + "epoch": 0.9707517389725396, + "grad_norm": 0.4668678343296051, + "learning_rate": 0.0005165564769161812, + "loss": 0.4381, + "step": 17375 + }, + { + "epoch": 0.9708076095762215, + "grad_norm": 0.4192667007446289, + "learning_rate": 0.0005165284625728373, + "loss": 0.4237, + "step": 17376 + }, + { + "epoch": 0.9708634801799033, + "grad_norm": 0.425924688577652, + "learning_rate": 0.0005165004482294935, + "loss": 0.3972, + "step": 17377 + }, + { + "epoch": 0.9709193507835853, + "grad_norm": 0.45322683453559875, + "learning_rate": 0.0005164724338861497, + "loss": 0.4707, + "step": 17378 + }, + { + "epoch": 0.9709752213872671, + "grad_norm": 1.9693360328674316, + "learning_rate": 0.000516444419542806, + "loss": 0.4259, + "step": 17379 + }, + { + "epoch": 0.971031091990949, + "grad_norm": 0.4892113506793976, + "learning_rate": 0.0005164164051994621, + "loss": 0.4137, + "step": 17380 + }, + { + "epoch": 0.9710869625946308, + "grad_norm": 0.6117229461669922, + "learning_rate": 0.0005163883908561183, + "loss": 0.489, + "step": 17381 + }, + { + "epoch": 0.9711428331983127, + "grad_norm": 1.3837441205978394, + "learning_rate": 0.0005163603765127745, + "loss": 0.6526, + "step": 17382 + }, + { + "epoch": 0.9711987038019946, + "grad_norm": 0.4620944559574127, + "learning_rate": 0.0005163323621694308, + "loss": 0.4325, + "step": 17383 + }, + { + "epoch": 0.9712545744056764, + "grad_norm": 0.40864089131355286, + "learning_rate": 0.0005163043478260869, + "loss": 0.3766, + "step": 17384 + }, + { + "epoch": 0.9713104450093584, + "grad_norm": 0.3328981101512909, + "learning_rate": 0.0005162763334827431, + "loss": 0.4124, + "step": 17385 + }, + { + "epoch": 0.9713663156130402, + "grad_norm": 0.41247525811195374, + "learning_rate": 0.0005162483191393994, + "loss": 0.3881, + "step": 17386 + }, + { + "epoch": 0.971422186216722, + "grad_norm": 0.5923580527305603, + "learning_rate": 0.0005162203047960557, + "loss": 0.4493, + "step": 17387 + }, + { + "epoch": 0.9714780568204039, + "grad_norm": 2.0903570652008057, + "learning_rate": 0.0005161922904527117, + "loss": 0.5005, + "step": 17388 + }, + { + "epoch": 0.9715339274240858, + "grad_norm": 0.46213629841804504, + "learning_rate": 0.000516164276109368, + "loss": 0.37, + "step": 17389 + }, + { + "epoch": 0.9715897980277677, + "grad_norm": 0.4720909595489502, + "learning_rate": 0.0005161362617660243, + "loss": 0.5596, + "step": 17390 + }, + { + "epoch": 0.9716456686314495, + "grad_norm": 0.46911805868148804, + "learning_rate": 0.0005161082474226805, + "loss": 0.4716, + "step": 17391 + }, + { + "epoch": 0.9717015392351315, + "grad_norm": 0.7062762379646301, + "learning_rate": 0.0005160802330793366, + "loss": 0.518, + "step": 17392 + }, + { + "epoch": 0.9717574098388133, + "grad_norm": 0.6345553994178772, + "learning_rate": 0.0005160522187359929, + "loss": 0.4268, + "step": 17393 + }, + { + "epoch": 0.9718132804424952, + "grad_norm": 0.6141601204872131, + "learning_rate": 0.0005160242043926491, + "loss": 0.4818, + "step": 17394 + }, + { + "epoch": 0.9718691510461771, + "grad_norm": 0.3617633283138275, + "learning_rate": 0.0005159961900493052, + "loss": 0.4322, + "step": 17395 + }, + { + "epoch": 0.971925021649859, + "grad_norm": 0.36717283725738525, + "learning_rate": 0.0005159681757059614, + "loss": 0.3689, + "step": 17396 + }, + { + "epoch": 0.9719808922535408, + "grad_norm": 0.9269907474517822, + "learning_rate": 0.0005159401613626177, + "loss": 0.4289, + "step": 17397 + }, + { + "epoch": 0.9720367628572226, + "grad_norm": 0.49043241143226624, + "learning_rate": 0.0005159121470192739, + "loss": 0.4165, + "step": 17398 + }, + { + "epoch": 0.9720926334609046, + "grad_norm": 0.5054203867912292, + "learning_rate": 0.00051588413267593, + "loss": 0.5454, + "step": 17399 + }, + { + "epoch": 0.9721485040645864, + "grad_norm": 0.7116101384162903, + "learning_rate": 0.0005158561183325863, + "loss": 0.5497, + "step": 17400 + }, + { + "epoch": 0.9722043746682683, + "grad_norm": 0.4123230576515198, + "learning_rate": 0.0005158281039892425, + "loss": 0.3853, + "step": 17401 + }, + { + "epoch": 0.9722602452719502, + "grad_norm": 0.29924100637435913, + "learning_rate": 0.0005158000896458987, + "loss": 0.3336, + "step": 17402 + }, + { + "epoch": 0.972316115875632, + "grad_norm": 0.44395899772644043, + "learning_rate": 0.0005157720753025548, + "loss": 0.3388, + "step": 17403 + }, + { + "epoch": 0.9723719864793139, + "grad_norm": 0.33414700627326965, + "learning_rate": 0.0005157440609592111, + "loss": 0.4247, + "step": 17404 + }, + { + "epoch": 0.9724278570829957, + "grad_norm": 0.36453157663345337, + "learning_rate": 0.0005157160466158674, + "loss": 0.3397, + "step": 17405 + }, + { + "epoch": 0.9724837276866777, + "grad_norm": 0.4807857275009155, + "learning_rate": 0.0005156880322725236, + "loss": 0.4859, + "step": 17406 + }, + { + "epoch": 0.9725395982903595, + "grad_norm": 0.8305355310440063, + "learning_rate": 0.0005156600179291798, + "loss": 0.4749, + "step": 17407 + }, + { + "epoch": 0.9725954688940414, + "grad_norm": 0.7397358417510986, + "learning_rate": 0.000515632003585836, + "loss": 0.3799, + "step": 17408 + }, + { + "epoch": 0.9726513394977233, + "grad_norm": 0.37905174493789673, + "learning_rate": 0.0005156039892424922, + "loss": 0.5075, + "step": 17409 + }, + { + "epoch": 0.9727072101014052, + "grad_norm": 0.6986328959465027, + "learning_rate": 0.0005155759748991484, + "loss": 0.4588, + "step": 17410 + }, + { + "epoch": 0.972763080705087, + "grad_norm": 0.40772703289985657, + "learning_rate": 0.0005155479605558046, + "loss": 0.4066, + "step": 17411 + }, + { + "epoch": 0.9728189513087689, + "grad_norm": 0.47316524386405945, + "learning_rate": 0.0005155199462124608, + "loss": 0.4119, + "step": 17412 + }, + { + "epoch": 0.9728748219124508, + "grad_norm": 0.6408177018165588, + "learning_rate": 0.000515491931869117, + "loss": 0.3852, + "step": 17413 + }, + { + "epoch": 0.9729306925161326, + "grad_norm": 0.417909175157547, + "learning_rate": 0.0005154639175257731, + "loss": 0.3484, + "step": 17414 + }, + { + "epoch": 0.9729865631198145, + "grad_norm": 0.5474541783332825, + "learning_rate": 0.0005154359031824294, + "loss": 0.4361, + "step": 17415 + }, + { + "epoch": 0.9730424337234964, + "grad_norm": 0.6183429956436157, + "learning_rate": 0.0005154078888390856, + "loss": 0.455, + "step": 17416 + }, + { + "epoch": 0.9730983043271783, + "grad_norm": 0.9575892686843872, + "learning_rate": 0.0005153798744957418, + "loss": 0.4568, + "step": 17417 + }, + { + "epoch": 0.9731541749308601, + "grad_norm": 0.7467257976531982, + "learning_rate": 0.000515351860152398, + "loss": 0.4195, + "step": 17418 + }, + { + "epoch": 0.973210045534542, + "grad_norm": 0.8092653155326843, + "learning_rate": 0.0005153238458090542, + "loss": 0.5053, + "step": 17419 + }, + { + "epoch": 0.9732659161382239, + "grad_norm": 0.36290207505226135, + "learning_rate": 0.0005152958314657104, + "loss": 0.4096, + "step": 17420 + }, + { + "epoch": 0.9733217867419057, + "grad_norm": 0.40937796235084534, + "learning_rate": 0.0005152678171223666, + "loss": 0.4485, + "step": 17421 + }, + { + "epoch": 0.9733776573455876, + "grad_norm": 0.6533880233764648, + "learning_rate": 0.0005152398027790229, + "loss": 0.4862, + "step": 17422 + }, + { + "epoch": 0.9734335279492695, + "grad_norm": 0.5149346590042114, + "learning_rate": 0.0005152117884356791, + "loss": 0.469, + "step": 17423 + }, + { + "epoch": 0.9734893985529514, + "grad_norm": 0.5619905591011047, + "learning_rate": 0.0005151837740923353, + "loss": 0.5123, + "step": 17424 + }, + { + "epoch": 0.9735452691566332, + "grad_norm": 0.6163181662559509, + "learning_rate": 0.0005151557597489916, + "loss": 0.3733, + "step": 17425 + }, + { + "epoch": 0.9736011397603151, + "grad_norm": 0.48503541946411133, + "learning_rate": 0.0005151277454056477, + "loss": 0.44, + "step": 17426 + }, + { + "epoch": 0.973657010363997, + "grad_norm": 0.37304753065109253, + "learning_rate": 0.0005150997310623039, + "loss": 0.4116, + "step": 17427 + }, + { + "epoch": 0.9737128809676788, + "grad_norm": 0.41259124875068665, + "learning_rate": 0.0005150717167189601, + "loss": 0.5228, + "step": 17428 + }, + { + "epoch": 0.9737687515713608, + "grad_norm": 0.6556482315063477, + "learning_rate": 0.0005150437023756164, + "loss": 0.3382, + "step": 17429 + }, + { + "epoch": 0.9738246221750426, + "grad_norm": 0.6428855657577515, + "learning_rate": 0.0005150156880322725, + "loss": 0.4238, + "step": 17430 + }, + { + "epoch": 0.9738804927787245, + "grad_norm": 1.3827089071273804, + "learning_rate": 0.0005149876736889287, + "loss": 0.3207, + "step": 17431 + }, + { + "epoch": 0.9739363633824063, + "grad_norm": 0.4565766453742981, + "learning_rate": 0.000514959659345585, + "loss": 0.4511, + "step": 17432 + }, + { + "epoch": 0.9739922339860883, + "grad_norm": 0.43937647342681885, + "learning_rate": 0.0005149316450022412, + "loss": 0.4024, + "step": 17433 + }, + { + "epoch": 0.9740481045897701, + "grad_norm": 0.6068854331970215, + "learning_rate": 0.0005149036306588973, + "loss": 0.5161, + "step": 17434 + }, + { + "epoch": 0.9741039751934519, + "grad_norm": 0.4925512671470642, + "learning_rate": 0.0005148756163155535, + "loss": 0.4528, + "step": 17435 + }, + { + "epoch": 0.9741598457971339, + "grad_norm": 0.41443127393722534, + "learning_rate": 0.0005148476019722098, + "loss": 0.509, + "step": 17436 + }, + { + "epoch": 0.9742157164008157, + "grad_norm": 0.4405592381954193, + "learning_rate": 0.0005148195876288659, + "loss": 0.4551, + "step": 17437 + }, + { + "epoch": 0.9742715870044976, + "grad_norm": 0.4208815097808838, + "learning_rate": 0.0005147915732855221, + "loss": 0.4878, + "step": 17438 + }, + { + "epoch": 0.9743274576081794, + "grad_norm": 0.5243644714355469, + "learning_rate": 0.0005147635589421785, + "loss": 0.4879, + "step": 17439 + }, + { + "epoch": 0.9743833282118614, + "grad_norm": 0.36486613750457764, + "learning_rate": 0.0005147355445988347, + "loss": 0.3747, + "step": 17440 + }, + { + "epoch": 0.9744391988155432, + "grad_norm": 0.37741050124168396, + "learning_rate": 0.0005147075302554908, + "loss": 0.418, + "step": 17441 + }, + { + "epoch": 0.974495069419225, + "grad_norm": 0.31645435094833374, + "learning_rate": 0.000514679515912147, + "loss": 0.4248, + "step": 17442 + }, + { + "epoch": 0.974550940022907, + "grad_norm": 0.3500012755393982, + "learning_rate": 0.0005146515015688033, + "loss": 0.3514, + "step": 17443 + }, + { + "epoch": 0.9746068106265888, + "grad_norm": 1.8384281396865845, + "learning_rate": 0.0005146234872254595, + "loss": 0.3666, + "step": 17444 + }, + { + "epoch": 0.9746626812302707, + "grad_norm": 0.5217940211296082, + "learning_rate": 0.0005145954728821156, + "loss": 0.5776, + "step": 17445 + }, + { + "epoch": 0.9747185518339526, + "grad_norm": 0.44261208176612854, + "learning_rate": 0.0005145674585387719, + "loss": 0.3366, + "step": 17446 + }, + { + "epoch": 0.9747744224376345, + "grad_norm": 4.453639030456543, + "learning_rate": 0.0005145394441954281, + "loss": 0.4178, + "step": 17447 + }, + { + "epoch": 0.9748302930413163, + "grad_norm": 1.378798484802246, + "learning_rate": 0.0005145114298520843, + "loss": 0.4183, + "step": 17448 + }, + { + "epoch": 0.9748861636449981, + "grad_norm": 0.677188515663147, + "learning_rate": 0.0005144834155087404, + "loss": 0.447, + "step": 17449 + }, + { + "epoch": 0.9749420342486801, + "grad_norm": 0.6188548803329468, + "learning_rate": 0.0005144554011653967, + "loss": 0.3799, + "step": 17450 + }, + { + "epoch": 0.9749979048523619, + "grad_norm": 0.37866583466529846, + "learning_rate": 0.0005144273868220529, + "loss": 0.4054, + "step": 17451 + }, + { + "epoch": 0.9750537754560438, + "grad_norm": 0.3580942451953888, + "learning_rate": 0.0005143993724787091, + "loss": 0.4032, + "step": 17452 + }, + { + "epoch": 0.9751096460597257, + "grad_norm": 0.6104447245597839, + "learning_rate": 0.0005143713581353652, + "loss": 0.4177, + "step": 17453 + }, + { + "epoch": 0.9751655166634076, + "grad_norm": 0.4426863193511963, + "learning_rate": 0.0005143433437920215, + "loss": 0.4251, + "step": 17454 + }, + { + "epoch": 0.9752213872670894, + "grad_norm": 0.4767497479915619, + "learning_rate": 0.0005143153294486778, + "loss": 0.4367, + "step": 17455 + }, + { + "epoch": 0.9752772578707712, + "grad_norm": 0.4612554609775543, + "learning_rate": 0.000514287315105334, + "loss": 0.475, + "step": 17456 + }, + { + "epoch": 0.9753331284744532, + "grad_norm": 0.4634662866592407, + "learning_rate": 0.0005142593007619902, + "loss": 0.4021, + "step": 17457 + }, + { + "epoch": 0.975388999078135, + "grad_norm": 0.4707720875740051, + "learning_rate": 0.0005142312864186464, + "loss": 0.4166, + "step": 17458 + }, + { + "epoch": 0.9754448696818169, + "grad_norm": 0.3622208535671234, + "learning_rate": 0.0005142032720753026, + "loss": 0.4089, + "step": 17459 + }, + { + "epoch": 0.9755007402854988, + "grad_norm": 0.5264747142791748, + "learning_rate": 0.0005141752577319587, + "loss": 0.5231, + "step": 17460 + }, + { + "epoch": 0.9755566108891807, + "grad_norm": 0.7160525321960449, + "learning_rate": 0.000514147243388615, + "loss": 0.4002, + "step": 17461 + }, + { + "epoch": 0.9756124814928625, + "grad_norm": 0.46628543734550476, + "learning_rate": 0.0005141192290452712, + "loss": 0.3679, + "step": 17462 + }, + { + "epoch": 0.9756683520965445, + "grad_norm": 1.1952030658721924, + "learning_rate": 0.0005140912147019274, + "loss": 0.421, + "step": 17463 + }, + { + "epoch": 0.9757242227002263, + "grad_norm": 7.899128437042236, + "learning_rate": 0.0005140632003585836, + "loss": 0.3487, + "step": 17464 + }, + { + "epoch": 0.9757800933039081, + "grad_norm": 0.7102341651916504, + "learning_rate": 0.0005140351860152398, + "loss": 0.4885, + "step": 17465 + }, + { + "epoch": 0.97583596390759, + "grad_norm": 0.7398027181625366, + "learning_rate": 0.000514007171671896, + "loss": 0.4346, + "step": 17466 + }, + { + "epoch": 0.9758918345112719, + "grad_norm": 0.3949776291847229, + "learning_rate": 0.0005139791573285522, + "loss": 0.4628, + "step": 17467 + }, + { + "epoch": 0.9759477051149538, + "grad_norm": 0.4239189028739929, + "learning_rate": 0.0005139511429852084, + "loss": 0.5016, + "step": 17468 + }, + { + "epoch": 0.9760035757186356, + "grad_norm": 0.43221694231033325, + "learning_rate": 0.0005139231286418646, + "loss": 0.4474, + "step": 17469 + }, + { + "epoch": 0.9760594463223176, + "grad_norm": 0.5666254162788391, + "learning_rate": 0.0005138951142985208, + "loss": 0.4483, + "step": 17470 + }, + { + "epoch": 0.9761153169259994, + "grad_norm": 0.42019185423851013, + "learning_rate": 0.0005138670999551772, + "loss": 0.3549, + "step": 17471 + }, + { + "epoch": 0.9761711875296812, + "grad_norm": 0.8804469108581543, + "learning_rate": 0.0005138390856118333, + "loss": 0.3965, + "step": 17472 + }, + { + "epoch": 0.9762270581333631, + "grad_norm": 0.4795267581939697, + "learning_rate": 0.0005138110712684895, + "loss": 0.4794, + "step": 17473 + }, + { + "epoch": 0.976282928737045, + "grad_norm": 0.43349355459213257, + "learning_rate": 0.0005137830569251457, + "loss": 0.3983, + "step": 17474 + }, + { + "epoch": 0.9763387993407269, + "grad_norm": 0.5260438919067383, + "learning_rate": 0.000513755042581802, + "loss": 0.5411, + "step": 17475 + }, + { + "epoch": 0.9763946699444087, + "grad_norm": 0.46766531467437744, + "learning_rate": 0.0005137270282384581, + "loss": 0.5494, + "step": 17476 + }, + { + "epoch": 0.9764505405480907, + "grad_norm": 0.3714561462402344, + "learning_rate": 0.0005136990138951143, + "loss": 0.3961, + "step": 17477 + }, + { + "epoch": 0.9765064111517725, + "grad_norm": 0.48652133345603943, + "learning_rate": 0.0005136709995517706, + "loss": 0.4426, + "step": 17478 + }, + { + "epoch": 0.9765622817554543, + "grad_norm": 1.4523992538452148, + "learning_rate": 0.0005136429852084267, + "loss": 0.4645, + "step": 17479 + }, + { + "epoch": 0.9766181523591363, + "grad_norm": 0.4652807116508484, + "learning_rate": 0.0005136149708650829, + "loss": 0.4212, + "step": 17480 + }, + { + "epoch": 0.9766740229628181, + "grad_norm": 0.47916093468666077, + "learning_rate": 0.0005135869565217391, + "loss": 0.5044, + "step": 17481 + }, + { + "epoch": 0.9767298935665, + "grad_norm": 0.4286094307899475, + "learning_rate": 0.0005135589421783954, + "loss": 0.386, + "step": 17482 + }, + { + "epoch": 0.9767857641701818, + "grad_norm": 0.5935458540916443, + "learning_rate": 0.0005135309278350515, + "loss": 0.4597, + "step": 17483 + }, + { + "epoch": 0.9768416347738638, + "grad_norm": 0.510100245475769, + "learning_rate": 0.0005135029134917077, + "loss": 0.4451, + "step": 17484 + }, + { + "epoch": 0.9768975053775456, + "grad_norm": 0.35244524478912354, + "learning_rate": 0.0005134748991483639, + "loss": 0.3685, + "step": 17485 + }, + { + "epoch": 0.9769533759812274, + "grad_norm": 0.5962564945220947, + "learning_rate": 0.0005134468848050202, + "loss": 0.4429, + "step": 17486 + }, + { + "epoch": 0.9770092465849094, + "grad_norm": 0.8429818749427795, + "learning_rate": 0.0005134188704616763, + "loss": 0.4364, + "step": 17487 + }, + { + "epoch": 0.9770651171885912, + "grad_norm": 1.4989994764328003, + "learning_rate": 0.0005133908561183325, + "loss": 0.4153, + "step": 17488 + }, + { + "epoch": 0.9771209877922731, + "grad_norm": 0.49760565161705017, + "learning_rate": 0.0005133628417749889, + "loss": 0.4027, + "step": 17489 + }, + { + "epoch": 0.9771768583959549, + "grad_norm": 0.513431191444397, + "learning_rate": 0.0005133348274316451, + "loss": 0.4778, + "step": 17490 + }, + { + "epoch": 0.9772327289996369, + "grad_norm": 1.2786411046981812, + "learning_rate": 0.0005133068130883012, + "loss": 0.466, + "step": 17491 + }, + { + "epoch": 0.9772885996033187, + "grad_norm": 2.694061279296875, + "learning_rate": 0.0005132787987449574, + "loss": 0.4298, + "step": 17492 + }, + { + "epoch": 0.9773444702070005, + "grad_norm": 0.5198598504066467, + "learning_rate": 0.0005132507844016137, + "loss": 0.3629, + "step": 17493 + }, + { + "epoch": 0.9774003408106825, + "grad_norm": 1.8641821146011353, + "learning_rate": 0.0005132227700582699, + "loss": 0.5171, + "step": 17494 + }, + { + "epoch": 0.9774562114143643, + "grad_norm": 0.6839872002601624, + "learning_rate": 0.000513194755714926, + "loss": 0.3188, + "step": 17495 + }, + { + "epoch": 0.9775120820180462, + "grad_norm": 0.643166184425354, + "learning_rate": 0.0005131667413715823, + "loss": 0.5739, + "step": 17496 + }, + { + "epoch": 0.9775679526217281, + "grad_norm": 0.4557409882545471, + "learning_rate": 0.0005131387270282385, + "loss": 0.3595, + "step": 17497 + }, + { + "epoch": 0.97762382322541, + "grad_norm": 0.6638292670249939, + "learning_rate": 0.0005131107126848947, + "loss": 0.6266, + "step": 17498 + }, + { + "epoch": 0.9776796938290918, + "grad_norm": 0.6755972504615784, + "learning_rate": 0.0005130826983415508, + "loss": 0.3883, + "step": 17499 + }, + { + "epoch": 0.9777355644327737, + "grad_norm": 0.3321804106235504, + "learning_rate": 0.0005130546839982071, + "loss": 0.4159, + "step": 17500 + }, + { + "epoch": 0.9777355644327737, + "eval_cer": 0.08851319651293466, + "eval_loss": 0.3319922387599945, + "eval_runtime": 56.1856, + "eval_samples_per_second": 80.768, + "eval_steps_per_second": 5.055, + "eval_wer": 0.3508561213373928, + "step": 17500 + }, + { + "epoch": 0.9777914350364556, + "grad_norm": 0.45425498485565186, + "learning_rate": 0.0005130266696548633, + "loss": 0.4723, + "step": 17501 + }, + { + "epoch": 0.9778473056401374, + "grad_norm": 3.6380791664123535, + "learning_rate": 0.0005129986553115194, + "loss": 0.4973, + "step": 17502 + }, + { + "epoch": 0.9779031762438193, + "grad_norm": 0.8997290134429932, + "learning_rate": 0.0005129706409681757, + "loss": 0.3086, + "step": 17503 + }, + { + "epoch": 0.9779590468475012, + "grad_norm": 3.5547282695770264, + "learning_rate": 0.000512942626624832, + "loss": 0.3956, + "step": 17504 + }, + { + "epoch": 0.9780149174511831, + "grad_norm": 0.5897759795188904, + "learning_rate": 0.0005129146122814882, + "loss": 0.3487, + "step": 17505 + }, + { + "epoch": 0.9780707880548649, + "grad_norm": 0.5524156093597412, + "learning_rate": 0.0005128865979381443, + "loss": 0.4569, + "step": 17506 + }, + { + "epoch": 0.9781266586585468, + "grad_norm": 1.5168935060501099, + "learning_rate": 0.0005128585835948006, + "loss": 0.3637, + "step": 17507 + }, + { + "epoch": 0.9781825292622287, + "grad_norm": 14.724503517150879, + "learning_rate": 0.0005128305692514568, + "loss": 0.4415, + "step": 17508 + }, + { + "epoch": 0.9782383998659105, + "grad_norm": 0.6128382086753845, + "learning_rate": 0.000512802554908113, + "loss": 0.4607, + "step": 17509 + }, + { + "epoch": 0.9782942704695924, + "grad_norm": 0.6350130438804626, + "learning_rate": 0.0005127745405647692, + "loss": 0.5434, + "step": 17510 + }, + { + "epoch": 0.9783501410732743, + "grad_norm": 0.8980945944786072, + "learning_rate": 0.0005127465262214254, + "loss": 0.4742, + "step": 17511 + }, + { + "epoch": 0.9784060116769562, + "grad_norm": 0.37990355491638184, + "learning_rate": 0.0005127185118780816, + "loss": 0.5772, + "step": 17512 + }, + { + "epoch": 0.978461882280638, + "grad_norm": 0.5887312889099121, + "learning_rate": 0.0005126904975347378, + "loss": 0.5986, + "step": 17513 + }, + { + "epoch": 0.97851775288432, + "grad_norm": 0.44618552923202515, + "learning_rate": 0.000512662483191394, + "loss": 0.4979, + "step": 17514 + }, + { + "epoch": 0.9785736234880018, + "grad_norm": 2.3406121730804443, + "learning_rate": 0.0005126344688480502, + "loss": 0.4667, + "step": 17515 + }, + { + "epoch": 0.9786294940916836, + "grad_norm": 0.5148364901542664, + "learning_rate": 0.0005126064545047064, + "loss": 0.5049, + "step": 17516 + }, + { + "epoch": 0.9786853646953655, + "grad_norm": 0.7555166482925415, + "learning_rate": 0.0005125784401613627, + "loss": 0.4131, + "step": 17517 + }, + { + "epoch": 0.9787412352990474, + "grad_norm": 1.2179100513458252, + "learning_rate": 0.0005125504258180188, + "loss": 0.5379, + "step": 17518 + }, + { + "epoch": 0.9787971059027293, + "grad_norm": 0.4029715061187744, + "learning_rate": 0.000512522411474675, + "loss": 0.5397, + "step": 17519 + }, + { + "epoch": 0.9788529765064111, + "grad_norm": 1.953614354133606, + "learning_rate": 0.0005124943971313312, + "loss": 0.3155, + "step": 17520 + }, + { + "epoch": 0.9789088471100931, + "grad_norm": 0.5093328356742859, + "learning_rate": 0.0005124663827879874, + "loss": 0.3951, + "step": 17521 + }, + { + "epoch": 0.9789647177137749, + "grad_norm": 0.5797253847122192, + "learning_rate": 0.0005124383684446437, + "loss": 0.4504, + "step": 17522 + }, + { + "epoch": 0.9790205883174568, + "grad_norm": 0.47098079323768616, + "learning_rate": 0.0005124103541012999, + "loss": 0.3839, + "step": 17523 + }, + { + "epoch": 0.9790764589211386, + "grad_norm": 0.5285418033599854, + "learning_rate": 0.0005123823397579561, + "loss": 0.4755, + "step": 17524 + }, + { + "epoch": 0.9791323295248205, + "grad_norm": 0.3600643277168274, + "learning_rate": 0.0005123543254146123, + "loss": 0.5224, + "step": 17525 + }, + { + "epoch": 0.9791882001285024, + "grad_norm": 0.9905405044555664, + "learning_rate": 0.0005123263110712685, + "loss": 0.3882, + "step": 17526 + }, + { + "epoch": 0.9792440707321842, + "grad_norm": 0.3860851228237152, + "learning_rate": 0.0005122982967279247, + "loss": 0.4747, + "step": 17527 + }, + { + "epoch": 0.9792999413358662, + "grad_norm": 0.5381686091423035, + "learning_rate": 0.000512270282384581, + "loss": 0.3857, + "step": 17528 + }, + { + "epoch": 0.979355811939548, + "grad_norm": 0.6703985333442688, + "learning_rate": 0.0005122422680412371, + "loss": 0.4077, + "step": 17529 + }, + { + "epoch": 0.9794116825432299, + "grad_norm": 0.8609031438827515, + "learning_rate": 0.0005122142536978933, + "loss": 0.4681, + "step": 17530 + }, + { + "epoch": 0.9794675531469118, + "grad_norm": 0.48135003447532654, + "learning_rate": 0.0005121862393545495, + "loss": 0.4725, + "step": 17531 + }, + { + "epoch": 0.9795234237505936, + "grad_norm": 0.37932518124580383, + "learning_rate": 0.0005121582250112058, + "loss": 0.3982, + "step": 17532 + }, + { + "epoch": 0.9795792943542755, + "grad_norm": 0.42115405201911926, + "learning_rate": 0.0005121302106678619, + "loss": 0.447, + "step": 17533 + }, + { + "epoch": 0.9796351649579573, + "grad_norm": 0.960780680179596, + "learning_rate": 0.0005121021963245181, + "loss": 0.3937, + "step": 17534 + }, + { + "epoch": 0.9796910355616393, + "grad_norm": 0.41979196667671204, + "learning_rate": 0.0005120741819811744, + "loss": 0.4596, + "step": 17535 + }, + { + "epoch": 0.9797469061653211, + "grad_norm": 0.4665622115135193, + "learning_rate": 0.0005120461676378306, + "loss": 0.4491, + "step": 17536 + }, + { + "epoch": 0.979802776769003, + "grad_norm": 0.43986496329307556, + "learning_rate": 0.0005120181532944867, + "loss": 0.3996, + "step": 17537 + }, + { + "epoch": 0.9798586473726849, + "grad_norm": 0.7370406985282898, + "learning_rate": 0.0005119901389511429, + "loss": 0.5257, + "step": 17538 + }, + { + "epoch": 0.9799145179763668, + "grad_norm": 0.4847760498523712, + "learning_rate": 0.0005119621246077993, + "loss": 0.4356, + "step": 17539 + }, + { + "epoch": 0.9799703885800486, + "grad_norm": 1.3418480157852173, + "learning_rate": 0.0005119341102644555, + "loss": 0.3497, + "step": 17540 + }, + { + "epoch": 0.9800262591837304, + "grad_norm": 0.521504282951355, + "learning_rate": 0.0005119060959211116, + "loss": 0.5655, + "step": 17541 + }, + { + "epoch": 0.9800821297874124, + "grad_norm": 0.5172722935676575, + "learning_rate": 0.0005118780815777679, + "loss": 0.4238, + "step": 17542 + }, + { + "epoch": 0.9801380003910942, + "grad_norm": 0.6656014919281006, + "learning_rate": 0.0005118500672344241, + "loss": 0.4165, + "step": 17543 + }, + { + "epoch": 0.9801938709947761, + "grad_norm": 0.5190698504447937, + "learning_rate": 0.0005118220528910802, + "loss": 0.4207, + "step": 17544 + }, + { + "epoch": 0.980249741598458, + "grad_norm": 0.3774193823337555, + "learning_rate": 0.0005117940385477364, + "loss": 0.4327, + "step": 17545 + }, + { + "epoch": 0.9803056122021399, + "grad_norm": 0.42671701312065125, + "learning_rate": 0.0005117660242043927, + "loss": 0.4288, + "step": 17546 + }, + { + "epoch": 0.9803614828058217, + "grad_norm": 0.3998306095600128, + "learning_rate": 0.0005117380098610489, + "loss": 0.4169, + "step": 17547 + }, + { + "epoch": 0.9804173534095036, + "grad_norm": 0.8629884719848633, + "learning_rate": 0.000511709995517705, + "loss": 0.3884, + "step": 17548 + }, + { + "epoch": 0.9804732240131855, + "grad_norm": 0.6858118772506714, + "learning_rate": 0.0005116819811743613, + "loss": 0.3854, + "step": 17549 + }, + { + "epoch": 0.9805290946168673, + "grad_norm": 0.579622745513916, + "learning_rate": 0.0005116539668310175, + "loss": 0.692, + "step": 17550 + }, + { + "epoch": 0.9805849652205492, + "grad_norm": 0.6594758629798889, + "learning_rate": 0.0005116259524876737, + "loss": 0.523, + "step": 17551 + }, + { + "epoch": 0.9806408358242311, + "grad_norm": 0.46926650404930115, + "learning_rate": 0.0005115979381443298, + "loss": 0.4055, + "step": 17552 + }, + { + "epoch": 0.980696706427913, + "grad_norm": 0.3931312561035156, + "learning_rate": 0.0005115699238009861, + "loss": 0.4316, + "step": 17553 + }, + { + "epoch": 0.9807525770315948, + "grad_norm": 0.35349413752555847, + "learning_rate": 0.0005115419094576423, + "loss": 0.3884, + "step": 17554 + }, + { + "epoch": 0.9808084476352767, + "grad_norm": 0.47134825587272644, + "learning_rate": 0.0005115138951142986, + "loss": 0.4108, + "step": 17555 + }, + { + "epoch": 0.9808643182389586, + "grad_norm": 2.1264731884002686, + "learning_rate": 0.0005114858807709547, + "loss": 0.4831, + "step": 17556 + }, + { + "epoch": 0.9809201888426404, + "grad_norm": 0.4905335009098053, + "learning_rate": 0.000511457866427611, + "loss": 0.4246, + "step": 17557 + }, + { + "epoch": 0.9809760594463223, + "grad_norm": 0.5579735040664673, + "learning_rate": 0.0005114298520842672, + "loss": 0.366, + "step": 17558 + }, + { + "epoch": 0.9810319300500042, + "grad_norm": 0.4439331293106079, + "learning_rate": 0.0005114018377409234, + "loss": 0.5908, + "step": 17559 + }, + { + "epoch": 0.9810878006536861, + "grad_norm": 0.4541410803794861, + "learning_rate": 0.0005113738233975796, + "loss": 0.4069, + "step": 17560 + }, + { + "epoch": 0.9811436712573679, + "grad_norm": 2.6133437156677246, + "learning_rate": 0.0005113458090542358, + "loss": 0.5242, + "step": 17561 + }, + { + "epoch": 0.9811995418610499, + "grad_norm": 0.5690885782241821, + "learning_rate": 0.000511317794710892, + "loss": 0.4421, + "step": 17562 + }, + { + "epoch": 0.9812554124647317, + "grad_norm": 0.4211975932121277, + "learning_rate": 0.0005112897803675481, + "loss": 0.4614, + "step": 17563 + }, + { + "epoch": 0.9813112830684135, + "grad_norm": 3.13775372505188, + "learning_rate": 0.0005112617660242044, + "loss": 0.4585, + "step": 17564 + }, + { + "epoch": 0.9813671536720955, + "grad_norm": 0.6028986573219299, + "learning_rate": 0.0005112337516808606, + "loss": 0.4284, + "step": 17565 + }, + { + "epoch": 0.9814230242757773, + "grad_norm": 0.3594239056110382, + "learning_rate": 0.0005112057373375168, + "loss": 0.3951, + "step": 17566 + }, + { + "epoch": 0.9814788948794592, + "grad_norm": 7.182921886444092, + "learning_rate": 0.000511177722994173, + "loss": 0.5456, + "step": 17567 + }, + { + "epoch": 0.981534765483141, + "grad_norm": 0.553452730178833, + "learning_rate": 0.0005111497086508292, + "loss": 0.5216, + "step": 17568 + }, + { + "epoch": 0.981590636086823, + "grad_norm": 0.30397114157676697, + "learning_rate": 0.0005111216943074854, + "loss": 0.3496, + "step": 17569 + }, + { + "epoch": 0.9816465066905048, + "grad_norm": 0.3113769590854645, + "learning_rate": 0.0005110936799641416, + "loss": 0.3666, + "step": 17570 + }, + { + "epoch": 0.9817023772941866, + "grad_norm": 0.8412176966667175, + "learning_rate": 0.0005110656656207978, + "loss": 0.6616, + "step": 17571 + }, + { + "epoch": 0.9817582478978686, + "grad_norm": 0.5748639702796936, + "learning_rate": 0.000511037651277454, + "loss": 0.4095, + "step": 17572 + }, + { + "epoch": 0.9818141185015504, + "grad_norm": 4.367847919464111, + "learning_rate": 0.0005110096369341103, + "loss": 0.487, + "step": 17573 + }, + { + "epoch": 0.9818699891052323, + "grad_norm": 0.9045568704605103, + "learning_rate": 0.0005109816225907666, + "loss": 0.3777, + "step": 17574 + }, + { + "epoch": 0.9819258597089141, + "grad_norm": 0.32650694251060486, + "learning_rate": 0.0005109536082474227, + "loss": 0.4177, + "step": 17575 + }, + { + "epoch": 0.9819817303125961, + "grad_norm": 0.45016422867774963, + "learning_rate": 0.0005109255939040789, + "loss": 0.4194, + "step": 17576 + }, + { + "epoch": 0.9820376009162779, + "grad_norm": 0.3825277090072632, + "learning_rate": 0.0005108975795607351, + "loss": 0.4149, + "step": 17577 + }, + { + "epoch": 0.9820934715199597, + "grad_norm": 2.916433334350586, + "learning_rate": 0.0005108695652173914, + "loss": 0.3256, + "step": 17578 + }, + { + "epoch": 0.9821493421236417, + "grad_norm": 0.41703373193740845, + "learning_rate": 0.0005108415508740475, + "loss": 0.4667, + "step": 17579 + }, + { + "epoch": 0.9822052127273235, + "grad_norm": 0.5475935339927673, + "learning_rate": 0.0005108135365307037, + "loss": 0.4068, + "step": 17580 + }, + { + "epoch": 0.9822610833310054, + "grad_norm": 0.6851330399513245, + "learning_rate": 0.00051078552218736, + "loss": 0.5721, + "step": 17581 + }, + { + "epoch": 0.9823169539346872, + "grad_norm": 1.014081358909607, + "learning_rate": 0.0005107575078440162, + "loss": 0.398, + "step": 17582 + }, + { + "epoch": 0.9823728245383692, + "grad_norm": 0.9683920741081238, + "learning_rate": 0.0005107294935006723, + "loss": 0.3981, + "step": 17583 + }, + { + "epoch": 0.982428695142051, + "grad_norm": 1.1215826272964478, + "learning_rate": 0.0005107014791573285, + "loss": 0.3444, + "step": 17584 + }, + { + "epoch": 0.9824845657457328, + "grad_norm": 0.4845680892467499, + "learning_rate": 0.0005106734648139848, + "loss": 0.3479, + "step": 17585 + }, + { + "epoch": 0.9825404363494148, + "grad_norm": 0.3422548472881317, + "learning_rate": 0.0005106454504706409, + "loss": 0.5161, + "step": 17586 + }, + { + "epoch": 0.9825963069530966, + "grad_norm": 0.36987459659576416, + "learning_rate": 0.0005106174361272971, + "loss": 0.3305, + "step": 17587 + }, + { + "epoch": 0.9826521775567785, + "grad_norm": 0.3242086172103882, + "learning_rate": 0.0005105894217839535, + "loss": 0.3695, + "step": 17588 + }, + { + "epoch": 0.9827080481604604, + "grad_norm": 0.8411484360694885, + "learning_rate": 0.0005105614074406097, + "loss": 0.6131, + "step": 17589 + }, + { + "epoch": 0.9827639187641423, + "grad_norm": 0.39959365129470825, + "learning_rate": 0.0005105333930972658, + "loss": 0.4657, + "step": 17590 + }, + { + "epoch": 0.9828197893678241, + "grad_norm": 0.4263031482696533, + "learning_rate": 0.000510505378753922, + "loss": 0.4014, + "step": 17591 + }, + { + "epoch": 0.9828756599715059, + "grad_norm": 1.937611699104309, + "learning_rate": 0.0005104773644105783, + "loss": 0.5345, + "step": 17592 + }, + { + "epoch": 0.9829315305751879, + "grad_norm": 1.812680959701538, + "learning_rate": 0.0005104493500672345, + "loss": 0.3935, + "step": 17593 + }, + { + "epoch": 0.9829874011788697, + "grad_norm": 0.47413021326065063, + "learning_rate": 0.0005104213357238906, + "loss": 0.5047, + "step": 17594 + }, + { + "epoch": 0.9830432717825516, + "grad_norm": 0.5270043015480042, + "learning_rate": 0.0005103933213805468, + "loss": 0.4995, + "step": 17595 + }, + { + "epoch": 0.9830991423862335, + "grad_norm": 0.5083218216896057, + "learning_rate": 0.0005103653070372031, + "loss": 0.4613, + "step": 17596 + }, + { + "epoch": 0.9831550129899154, + "grad_norm": 0.5994338989257812, + "learning_rate": 0.0005103372926938593, + "loss": 0.3655, + "step": 17597 + }, + { + "epoch": 0.9832108835935972, + "grad_norm": 0.3914313316345215, + "learning_rate": 0.0005103092783505154, + "loss": 0.4197, + "step": 17598 + }, + { + "epoch": 0.983266754197279, + "grad_norm": 0.4253346621990204, + "learning_rate": 0.0005102812640071717, + "loss": 0.5207, + "step": 17599 + }, + { + "epoch": 0.983322624800961, + "grad_norm": 0.4200710952281952, + "learning_rate": 0.0005102532496638279, + "loss": 0.4972, + "step": 17600 + }, + { + "epoch": 0.9833784954046428, + "grad_norm": 0.6226274371147156, + "learning_rate": 0.0005102252353204841, + "loss": 0.6018, + "step": 17601 + }, + { + "epoch": 0.9834343660083247, + "grad_norm": 0.46964579820632935, + "learning_rate": 0.0005101972209771402, + "loss": 0.4693, + "step": 17602 + }, + { + "epoch": 0.9834902366120066, + "grad_norm": 0.47990697622299194, + "learning_rate": 0.0005101692066337965, + "loss": 0.4682, + "step": 17603 + }, + { + "epoch": 0.9835461072156885, + "grad_norm": 0.454912394285202, + "learning_rate": 0.0005101411922904527, + "loss": 0.4807, + "step": 17604 + }, + { + "epoch": 0.9836019778193703, + "grad_norm": 0.5642811059951782, + "learning_rate": 0.0005101131779471088, + "loss": 0.3977, + "step": 17605 + }, + { + "epoch": 0.9836578484230523, + "grad_norm": 2.9967386722564697, + "learning_rate": 0.0005100851636037652, + "loss": 0.673, + "step": 17606 + }, + { + "epoch": 0.9837137190267341, + "grad_norm": 0.495842844247818, + "learning_rate": 0.0005100571492604214, + "loss": 0.55, + "step": 17607 + }, + { + "epoch": 0.9837695896304159, + "grad_norm": 0.43258902430534363, + "learning_rate": 0.0005100291349170776, + "loss": 0.4355, + "step": 17608 + }, + { + "epoch": 0.9838254602340978, + "grad_norm": 0.44449347257614136, + "learning_rate": 0.0005100011205737337, + "loss": 0.53, + "step": 17609 + }, + { + "epoch": 0.9838813308377797, + "grad_norm": 0.563256025314331, + "learning_rate": 0.00050997310623039, + "loss": 0.4599, + "step": 17610 + }, + { + "epoch": 0.9839372014414616, + "grad_norm": 0.4086563289165497, + "learning_rate": 0.0005099450918870462, + "loss": 0.3746, + "step": 17611 + }, + { + "epoch": 0.9839930720451434, + "grad_norm": 0.40284985303878784, + "learning_rate": 0.0005099170775437024, + "loss": 0.3517, + "step": 17612 + }, + { + "epoch": 0.9840489426488254, + "grad_norm": 1.3889538049697876, + "learning_rate": 0.0005098890632003586, + "loss": 0.4375, + "step": 17613 + }, + { + "epoch": 0.9841048132525072, + "grad_norm": 1.4592355489730835, + "learning_rate": 0.0005098610488570148, + "loss": 0.5296, + "step": 17614 + }, + { + "epoch": 0.984160683856189, + "grad_norm": 0.31921184062957764, + "learning_rate": 0.000509833034513671, + "loss": 0.3671, + "step": 17615 + }, + { + "epoch": 0.9842165544598709, + "grad_norm": 0.944848358631134, + "learning_rate": 0.0005098050201703272, + "loss": 0.6722, + "step": 17616 + }, + { + "epoch": 0.9842724250635528, + "grad_norm": 0.7304681539535522, + "learning_rate": 0.0005097770058269834, + "loss": 0.5819, + "step": 17617 + }, + { + "epoch": 0.9843282956672347, + "grad_norm": 1.8088382482528687, + "learning_rate": 0.0005097489914836396, + "loss": 0.3504, + "step": 17618 + }, + { + "epoch": 0.9843841662709165, + "grad_norm": 0.4538138210773468, + "learning_rate": 0.0005097209771402958, + "loss": 0.4322, + "step": 17619 + }, + { + "epoch": 0.9844400368745985, + "grad_norm": 0.5650497078895569, + "learning_rate": 0.0005096929627969521, + "loss": 0.3808, + "step": 17620 + }, + { + "epoch": 0.9844959074782803, + "grad_norm": 0.7978362441062927, + "learning_rate": 0.0005096649484536082, + "loss": 0.4201, + "step": 17621 + }, + { + "epoch": 0.9845517780819621, + "grad_norm": 0.4551921486854553, + "learning_rate": 0.0005096369341102645, + "loss": 0.426, + "step": 17622 + }, + { + "epoch": 0.9846076486856441, + "grad_norm": 0.4915648102760315, + "learning_rate": 0.0005096089197669207, + "loss": 0.3822, + "step": 17623 + }, + { + "epoch": 0.9846635192893259, + "grad_norm": 0.6043376326560974, + "learning_rate": 0.000509580905423577, + "loss": 0.4249, + "step": 17624 + }, + { + "epoch": 0.9847193898930078, + "grad_norm": 0.5190222859382629, + "learning_rate": 0.0005095528910802331, + "loss": 0.4603, + "step": 17625 + }, + { + "epoch": 0.9847752604966896, + "grad_norm": 1.0896633863449097, + "learning_rate": 0.0005095248767368893, + "loss": 0.6023, + "step": 17626 + }, + { + "epoch": 0.9848311311003716, + "grad_norm": 0.4752677381038666, + "learning_rate": 0.0005094968623935456, + "loss": 0.4021, + "step": 17627 + }, + { + "epoch": 0.9848870017040534, + "grad_norm": 12.241194725036621, + "learning_rate": 0.0005094688480502017, + "loss": 0.3764, + "step": 17628 + }, + { + "epoch": 0.9849428723077353, + "grad_norm": 2.2721781730651855, + "learning_rate": 0.0005094408337068579, + "loss": 0.5153, + "step": 17629 + }, + { + "epoch": 0.9849987429114172, + "grad_norm": 0.526464581489563, + "learning_rate": 0.0005094128193635141, + "loss": 0.4807, + "step": 17630 + }, + { + "epoch": 0.985054613515099, + "grad_norm": 0.38028383255004883, + "learning_rate": 0.0005093848050201704, + "loss": 0.3336, + "step": 17631 + }, + { + "epoch": 0.9851104841187809, + "grad_norm": 0.3764721155166626, + "learning_rate": 0.0005093567906768265, + "loss": 0.4008, + "step": 17632 + }, + { + "epoch": 0.9851663547224627, + "grad_norm": 0.3968941867351532, + "learning_rate": 0.0005093287763334827, + "loss": 0.3725, + "step": 17633 + }, + { + "epoch": 0.9852222253261447, + "grad_norm": 0.49403122067451477, + "learning_rate": 0.0005093007619901389, + "loss": 0.3475, + "step": 17634 + }, + { + "epoch": 0.9852780959298265, + "grad_norm": 0.6095128059387207, + "learning_rate": 0.0005092727476467952, + "loss": 0.4791, + "step": 17635 + }, + { + "epoch": 0.9853339665335084, + "grad_norm": 0.5854508280754089, + "learning_rate": 0.0005092447333034513, + "loss": 0.4144, + "step": 17636 + }, + { + "epoch": 0.9853898371371903, + "grad_norm": 0.9015842080116272, + "learning_rate": 0.0005092167189601075, + "loss": 0.4098, + "step": 17637 + }, + { + "epoch": 0.9854457077408721, + "grad_norm": 0.5372818112373352, + "learning_rate": 0.0005091887046167639, + "loss": 0.5368, + "step": 17638 + }, + { + "epoch": 0.985501578344554, + "grad_norm": 0.3367762267589569, + "learning_rate": 0.0005091606902734201, + "loss": 0.4068, + "step": 17639 + }, + { + "epoch": 0.9855574489482359, + "grad_norm": 0.6455354690551758, + "learning_rate": 0.0005091326759300762, + "loss": 0.3651, + "step": 17640 + }, + { + "epoch": 0.9856133195519178, + "grad_norm": 0.5598285794258118, + "learning_rate": 0.0005091046615867324, + "loss": 0.3535, + "step": 17641 + }, + { + "epoch": 0.9856691901555996, + "grad_norm": 0.5724907517433167, + "learning_rate": 0.0005090766472433887, + "loss": 0.5614, + "step": 17642 + }, + { + "epoch": 0.9857250607592815, + "grad_norm": 0.4723854064941406, + "learning_rate": 0.0005090486329000449, + "loss": 0.5183, + "step": 17643 + }, + { + "epoch": 0.9857809313629634, + "grad_norm": 0.6381345391273499, + "learning_rate": 0.000509020618556701, + "loss": 0.5272, + "step": 17644 + }, + { + "epoch": 0.9858368019666452, + "grad_norm": 0.5646612644195557, + "learning_rate": 0.0005089926042133573, + "loss": 0.5119, + "step": 17645 + }, + { + "epoch": 0.9858926725703271, + "grad_norm": 0.8969854116439819, + "learning_rate": 0.0005089645898700135, + "loss": 0.5219, + "step": 17646 + }, + { + "epoch": 0.985948543174009, + "grad_norm": 0.42240262031555176, + "learning_rate": 0.0005089365755266697, + "loss": 0.4051, + "step": 17647 + }, + { + "epoch": 0.9860044137776909, + "grad_norm": 0.40728679299354553, + "learning_rate": 0.0005089085611833258, + "loss": 0.52, + "step": 17648 + }, + { + "epoch": 0.9860602843813727, + "grad_norm": 0.5244276523590088, + "learning_rate": 0.0005088805468399821, + "loss": 0.4167, + "step": 17649 + }, + { + "epoch": 0.9861161549850546, + "grad_norm": 0.44820237159729004, + "learning_rate": 0.0005088525324966383, + "loss": 0.5032, + "step": 17650 + }, + { + "epoch": 0.9861720255887365, + "grad_norm": 1.0735301971435547, + "learning_rate": 0.0005088245181532944, + "loss": 0.3092, + "step": 17651 + }, + { + "epoch": 0.9862278961924184, + "grad_norm": 0.4525391757488251, + "learning_rate": 0.0005087965038099507, + "loss": 0.3361, + "step": 17652 + }, + { + "epoch": 0.9862837667961002, + "grad_norm": 0.4879629909992218, + "learning_rate": 0.0005087684894666069, + "loss": 0.4391, + "step": 17653 + }, + { + "epoch": 0.9863396373997821, + "grad_norm": 0.37217023968696594, + "learning_rate": 0.0005087404751232631, + "loss": 0.5156, + "step": 17654 + }, + { + "epoch": 0.986395508003464, + "grad_norm": 0.35424721240997314, + "learning_rate": 0.0005087124607799192, + "loss": 0.4014, + "step": 17655 + }, + { + "epoch": 0.9864513786071458, + "grad_norm": 0.6076816916465759, + "learning_rate": 0.0005086844464365756, + "loss": 0.5406, + "step": 17656 + }, + { + "epoch": 0.9865072492108278, + "grad_norm": 0.4061864912509918, + "learning_rate": 0.0005086564320932318, + "loss": 0.4522, + "step": 17657 + }, + { + "epoch": 0.9865631198145096, + "grad_norm": 0.493636816740036, + "learning_rate": 0.000508628417749888, + "loss": 0.4069, + "step": 17658 + }, + { + "epoch": 0.9866189904181915, + "grad_norm": 0.4420832097530365, + "learning_rate": 0.0005086004034065442, + "loss": 0.5243, + "step": 17659 + }, + { + "epoch": 0.9866748610218733, + "grad_norm": 0.9527454376220703, + "learning_rate": 0.0005085723890632004, + "loss": 0.5303, + "step": 17660 + }, + { + "epoch": 0.9867307316255552, + "grad_norm": 0.641545295715332, + "learning_rate": 0.0005085443747198566, + "loss": 0.4412, + "step": 17661 + }, + { + "epoch": 0.9867866022292371, + "grad_norm": 0.5733144879341125, + "learning_rate": 0.0005085163603765128, + "loss": 0.4113, + "step": 17662 + }, + { + "epoch": 0.9868424728329189, + "grad_norm": 0.34478759765625, + "learning_rate": 0.000508488346033169, + "loss": 0.4134, + "step": 17663 + }, + { + "epoch": 0.9868983434366009, + "grad_norm": 0.46360722184181213, + "learning_rate": 0.0005084603316898252, + "loss": 0.4756, + "step": 17664 + }, + { + "epoch": 0.9869542140402827, + "grad_norm": 0.4279220700263977, + "learning_rate": 0.0005084323173464814, + "loss": 0.4452, + "step": 17665 + }, + { + "epoch": 0.9870100846439646, + "grad_norm": 0.5097653865814209, + "learning_rate": 0.0005084043030031376, + "loss": 0.383, + "step": 17666 + }, + { + "epoch": 0.9870659552476464, + "grad_norm": 0.4760741889476776, + "learning_rate": 0.0005083762886597938, + "loss": 0.3905, + "step": 17667 + }, + { + "epoch": 0.9871218258513283, + "grad_norm": 0.5484346747398376, + "learning_rate": 0.00050834827431645, + "loss": 0.4256, + "step": 17668 + }, + { + "epoch": 0.9871776964550102, + "grad_norm": 0.5003387928009033, + "learning_rate": 0.0005083202599731062, + "loss": 0.4627, + "step": 17669 + }, + { + "epoch": 0.987233567058692, + "grad_norm": 0.5636060237884521, + "learning_rate": 0.0005082922456297624, + "loss": 0.3416, + "step": 17670 + }, + { + "epoch": 0.987289437662374, + "grad_norm": 0.7351133823394775, + "learning_rate": 0.0005082642312864186, + "loss": 0.4344, + "step": 17671 + }, + { + "epoch": 0.9873453082660558, + "grad_norm": 0.4527966380119324, + "learning_rate": 0.0005082362169430749, + "loss": 0.4062, + "step": 17672 + }, + { + "epoch": 0.9874011788697377, + "grad_norm": 0.3145250380039215, + "learning_rate": 0.0005082082025997311, + "loss": 0.3337, + "step": 17673 + }, + { + "epoch": 0.9874570494734196, + "grad_norm": 0.4705909192562103, + "learning_rate": 0.0005081801882563873, + "loss": 0.3645, + "step": 17674 + }, + { + "epoch": 0.9875129200771015, + "grad_norm": 0.7013692855834961, + "learning_rate": 0.0005081521739130435, + "loss": 0.447, + "step": 17675 + }, + { + "epoch": 0.9875687906807833, + "grad_norm": 0.4820777475833893, + "learning_rate": 0.0005081241595696997, + "loss": 0.5775, + "step": 17676 + }, + { + "epoch": 0.9876246612844651, + "grad_norm": 0.5567259192466736, + "learning_rate": 0.000508096145226356, + "loss": 0.3798, + "step": 17677 + }, + { + "epoch": 0.9876805318881471, + "grad_norm": 0.7183355093002319, + "learning_rate": 0.0005080681308830121, + "loss": 0.4406, + "step": 17678 + }, + { + "epoch": 0.9877364024918289, + "grad_norm": 1.6247446537017822, + "learning_rate": 0.0005080401165396683, + "loss": 0.5083, + "step": 17679 + }, + { + "epoch": 0.9877922730955108, + "grad_norm": 0.5631040930747986, + "learning_rate": 0.0005080121021963245, + "loss": 0.5401, + "step": 17680 + }, + { + "epoch": 0.9878481436991927, + "grad_norm": 0.47174379229545593, + "learning_rate": 0.0005079840878529808, + "loss": 0.5079, + "step": 17681 + }, + { + "epoch": 0.9879040143028746, + "grad_norm": 0.4928877353668213, + "learning_rate": 0.0005079560735096369, + "loss": 0.4884, + "step": 17682 + }, + { + "epoch": 0.9879598849065564, + "grad_norm": 0.48733681440353394, + "learning_rate": 0.0005079280591662931, + "loss": 0.3875, + "step": 17683 + }, + { + "epoch": 0.9880157555102382, + "grad_norm": 0.4600412845611572, + "learning_rate": 0.0005079000448229494, + "loss": 0.4582, + "step": 17684 + }, + { + "epoch": 0.9880716261139202, + "grad_norm": 1.3254210948944092, + "learning_rate": 0.0005078720304796056, + "loss": 0.4486, + "step": 17685 + }, + { + "epoch": 0.988127496717602, + "grad_norm": 0.4690702259540558, + "learning_rate": 0.0005078440161362617, + "loss": 0.4381, + "step": 17686 + }, + { + "epoch": 0.9881833673212839, + "grad_norm": 0.5350250601768494, + "learning_rate": 0.0005078160017929179, + "loss": 0.5146, + "step": 17687 + }, + { + "epoch": 0.9882392379249658, + "grad_norm": 0.6362314820289612, + "learning_rate": 0.0005077879874495743, + "loss": 0.4677, + "step": 17688 + }, + { + "epoch": 0.9882951085286477, + "grad_norm": 0.4016256630420685, + "learning_rate": 0.0005077599731062305, + "loss": 0.4173, + "step": 17689 + }, + { + "epoch": 0.9883509791323295, + "grad_norm": 0.4101105034351349, + "learning_rate": 0.0005077319587628866, + "loss": 0.4351, + "step": 17690 + }, + { + "epoch": 0.9884068497360115, + "grad_norm": 0.35114216804504395, + "learning_rate": 0.0005077039444195429, + "loss": 0.3502, + "step": 17691 + }, + { + "epoch": 0.9884627203396933, + "grad_norm": 0.4068067669868469, + "learning_rate": 0.0005076759300761991, + "loss": 0.4965, + "step": 17692 + }, + { + "epoch": 0.9885185909433751, + "grad_norm": 0.3855532705783844, + "learning_rate": 0.0005076479157328552, + "loss": 0.4325, + "step": 17693 + }, + { + "epoch": 0.988574461547057, + "grad_norm": 0.596009373664856, + "learning_rate": 0.0005076199013895114, + "loss": 0.7753, + "step": 17694 + }, + { + "epoch": 0.9886303321507389, + "grad_norm": 3.1243934631347656, + "learning_rate": 0.0005075918870461677, + "loss": 0.3944, + "step": 17695 + }, + { + "epoch": 0.9886862027544208, + "grad_norm": 5.456907749176025, + "learning_rate": 0.0005075638727028239, + "loss": 0.6004, + "step": 17696 + }, + { + "epoch": 0.9887420733581026, + "grad_norm": 0.9840179681777954, + "learning_rate": 0.00050753585835948, + "loss": 0.6, + "step": 17697 + }, + { + "epoch": 0.9887979439617846, + "grad_norm": 4.032726287841797, + "learning_rate": 0.0005075078440161362, + "loss": 0.5094, + "step": 17698 + }, + { + "epoch": 0.9888538145654664, + "grad_norm": 0.48030826449394226, + "learning_rate": 0.0005074798296727925, + "loss": 0.5098, + "step": 17699 + }, + { + "epoch": 0.9889096851691482, + "grad_norm": 0.9397158026695251, + "learning_rate": 0.0005074518153294487, + "loss": 0.4165, + "step": 17700 + }, + { + "epoch": 0.9889655557728301, + "grad_norm": 1.9564239978790283, + "learning_rate": 0.0005074238009861048, + "loss": 0.4434, + "step": 17701 + }, + { + "epoch": 0.989021426376512, + "grad_norm": 0.3711514472961426, + "learning_rate": 0.0005073957866427611, + "loss": 0.3749, + "step": 17702 + }, + { + "epoch": 0.9890772969801939, + "grad_norm": 0.5231796503067017, + "learning_rate": 0.0005073677722994173, + "loss": 0.4714, + "step": 17703 + }, + { + "epoch": 0.9891331675838757, + "grad_norm": 2.0530941486358643, + "learning_rate": 0.0005073397579560735, + "loss": 0.3497, + "step": 17704 + }, + { + "epoch": 0.9891890381875577, + "grad_norm": 0.36119213700294495, + "learning_rate": 0.0005073117436127296, + "loss": 0.421, + "step": 17705 + }, + { + "epoch": 0.9892449087912395, + "grad_norm": 0.38499510288238525, + "learning_rate": 0.000507283729269386, + "loss": 0.4716, + "step": 17706 + }, + { + "epoch": 0.9893007793949213, + "grad_norm": 0.35164961218833923, + "learning_rate": 0.0005072557149260422, + "loss": 0.3948, + "step": 17707 + }, + { + "epoch": 0.9893566499986033, + "grad_norm": 0.480901300907135, + "learning_rate": 0.0005072277005826984, + "loss": 0.4221, + "step": 17708 + }, + { + "epoch": 0.9894125206022851, + "grad_norm": 1.8901225328445435, + "learning_rate": 0.0005071996862393546, + "loss": 0.4935, + "step": 17709 + }, + { + "epoch": 0.989468391205967, + "grad_norm": 0.33440452814102173, + "learning_rate": 0.0005071716718960108, + "loss": 0.4118, + "step": 17710 + }, + { + "epoch": 0.9895242618096488, + "grad_norm": 1.2505066394805908, + "learning_rate": 0.000507143657552667, + "loss": 0.4568, + "step": 17711 + }, + { + "epoch": 0.9895801324133308, + "grad_norm": 0.7391903400421143, + "learning_rate": 0.0005071156432093231, + "loss": 0.4933, + "step": 17712 + }, + { + "epoch": 0.9896360030170126, + "grad_norm": 0.4353151321411133, + "learning_rate": 0.0005070876288659794, + "loss": 0.4239, + "step": 17713 + }, + { + "epoch": 0.9896918736206944, + "grad_norm": 0.8422449827194214, + "learning_rate": 0.0005070596145226356, + "loss": 0.4953, + "step": 17714 + }, + { + "epoch": 0.9897477442243764, + "grad_norm": 0.636716365814209, + "learning_rate": 0.0005070316001792918, + "loss": 0.4599, + "step": 17715 + }, + { + "epoch": 0.9898036148280582, + "grad_norm": 0.3433661162853241, + "learning_rate": 0.000507003585835948, + "loss": 0.3625, + "step": 17716 + }, + { + "epoch": 0.9898594854317401, + "grad_norm": 0.6203038692474365, + "learning_rate": 0.0005069755714926042, + "loss": 0.5012, + "step": 17717 + }, + { + "epoch": 0.9899153560354219, + "grad_norm": 0.5161101222038269, + "learning_rate": 0.0005069475571492604, + "loss": 0.5147, + "step": 17718 + }, + { + "epoch": 0.9899712266391039, + "grad_norm": 0.3814427852630615, + "learning_rate": 0.0005069195428059166, + "loss": 0.4117, + "step": 17719 + }, + { + "epoch": 0.9900270972427857, + "grad_norm": 0.28413867950439453, + "learning_rate": 0.0005068915284625728, + "loss": 0.3396, + "step": 17720 + }, + { + "epoch": 0.9900829678464675, + "grad_norm": 0.360182523727417, + "learning_rate": 0.000506863514119229, + "loss": 0.3469, + "step": 17721 + }, + { + "epoch": 0.9901388384501495, + "grad_norm": 0.4823872447013855, + "learning_rate": 0.0005068354997758852, + "loss": 0.5201, + "step": 17722 + }, + { + "epoch": 0.9901947090538313, + "grad_norm": 0.5739291906356812, + "learning_rate": 0.0005068074854325416, + "loss": 0.4927, + "step": 17723 + }, + { + "epoch": 0.9902505796575132, + "grad_norm": 0.340907484292984, + "learning_rate": 0.0005067794710891977, + "loss": 0.4332, + "step": 17724 + }, + { + "epoch": 0.9903064502611951, + "grad_norm": 1.0025509595870972, + "learning_rate": 0.0005067514567458539, + "loss": 0.384, + "step": 17725 + }, + { + "epoch": 0.990362320864877, + "grad_norm": 0.934185802936554, + "learning_rate": 0.0005067234424025101, + "loss": 0.5925, + "step": 17726 + }, + { + "epoch": 0.9904181914685588, + "grad_norm": 0.3243636190891266, + "learning_rate": 0.0005066954280591664, + "loss": 0.4123, + "step": 17727 + }, + { + "epoch": 0.9904740620722406, + "grad_norm": 0.5816631317138672, + "learning_rate": 0.0005066674137158225, + "loss": 0.3775, + "step": 17728 + }, + { + "epoch": 0.9905299326759226, + "grad_norm": 0.8288601040840149, + "learning_rate": 0.0005066393993724787, + "loss": 0.4241, + "step": 17729 + }, + { + "epoch": 0.9905858032796044, + "grad_norm": 0.6526883840560913, + "learning_rate": 0.000506611385029135, + "loss": 0.7217, + "step": 17730 + }, + { + "epoch": 0.9906416738832863, + "grad_norm": 0.5501333475112915, + "learning_rate": 0.0005065833706857912, + "loss": 0.4138, + "step": 17731 + }, + { + "epoch": 0.9906975444869682, + "grad_norm": 0.3975365459918976, + "learning_rate": 0.0005065553563424473, + "loss": 0.3898, + "step": 17732 + }, + { + "epoch": 0.9907534150906501, + "grad_norm": 2.1236178874969482, + "learning_rate": 0.0005065273419991035, + "loss": 0.4001, + "step": 17733 + }, + { + "epoch": 0.9908092856943319, + "grad_norm": 0.3863665461540222, + "learning_rate": 0.0005064993276557598, + "loss": 0.447, + "step": 17734 + }, + { + "epoch": 0.9908651562980137, + "grad_norm": 0.5555933713912964, + "learning_rate": 0.0005064713133124159, + "loss": 0.5518, + "step": 17735 + }, + { + "epoch": 0.9909210269016957, + "grad_norm": 1.420944333076477, + "learning_rate": 0.0005064432989690721, + "loss": 0.4525, + "step": 17736 + }, + { + "epoch": 0.9909768975053775, + "grad_norm": 0.5222059488296509, + "learning_rate": 0.0005064152846257283, + "loss": 0.5294, + "step": 17737 + }, + { + "epoch": 0.9910327681090594, + "grad_norm": 5.060547828674316, + "learning_rate": 0.0005063872702823847, + "loss": 0.4827, + "step": 17738 + }, + { + "epoch": 0.9910886387127413, + "grad_norm": 0.5102651119232178, + "learning_rate": 0.0005063592559390407, + "loss": 0.6015, + "step": 17739 + }, + { + "epoch": 0.9911445093164232, + "grad_norm": 0.5271468758583069, + "learning_rate": 0.000506331241595697, + "loss": 0.5059, + "step": 17740 + }, + { + "epoch": 0.991200379920105, + "grad_norm": 0.5278060436248779, + "learning_rate": 0.0005063032272523533, + "loss": 0.4051, + "step": 17741 + }, + { + "epoch": 0.991256250523787, + "grad_norm": 0.40830713510513306, + "learning_rate": 0.0005062752129090095, + "loss": 0.4593, + "step": 17742 + }, + { + "epoch": 0.9913121211274688, + "grad_norm": 0.4834471046924591, + "learning_rate": 0.0005062471985656656, + "loss": 0.4377, + "step": 17743 + }, + { + "epoch": 0.9913679917311506, + "grad_norm": 0.43179386854171753, + "learning_rate": 0.0005062191842223218, + "loss": 0.4085, + "step": 17744 + }, + { + "epoch": 0.9914238623348325, + "grad_norm": 1.3213344812393188, + "learning_rate": 0.0005061911698789781, + "loss": 0.5257, + "step": 17745 + }, + { + "epoch": 0.9914797329385144, + "grad_norm": 0.3839477598667145, + "learning_rate": 0.0005061631555356343, + "loss": 0.4288, + "step": 17746 + }, + { + "epoch": 0.9915356035421963, + "grad_norm": 2.9204564094543457, + "learning_rate": 0.0005061351411922904, + "loss": 0.4574, + "step": 17747 + }, + { + "epoch": 0.9915914741458781, + "grad_norm": 0.5278803110122681, + "learning_rate": 0.0005061071268489467, + "loss": 0.3662, + "step": 17748 + }, + { + "epoch": 0.9916473447495601, + "grad_norm": 0.443231463432312, + "learning_rate": 0.0005060791125056029, + "loss": 0.473, + "step": 17749 + }, + { + "epoch": 0.9917032153532419, + "grad_norm": 0.8135861158370972, + "learning_rate": 0.0005060510981622591, + "loss": 0.4753, + "step": 17750 + }, + { + "epoch": 0.9917590859569237, + "grad_norm": 0.3369845449924469, + "learning_rate": 0.0005060230838189152, + "loss": 0.3856, + "step": 17751 + }, + { + "epoch": 0.9918149565606056, + "grad_norm": 0.37702783942222595, + "learning_rate": 0.0005059950694755715, + "loss": 0.5167, + "step": 17752 + }, + { + "epoch": 0.9918708271642875, + "grad_norm": 0.3895067274570465, + "learning_rate": 0.0005059670551322277, + "loss": 0.3832, + "step": 17753 + }, + { + "epoch": 0.9919266977679694, + "grad_norm": 0.6529462933540344, + "learning_rate": 0.0005059390407888838, + "loss": 0.4466, + "step": 17754 + }, + { + "epoch": 0.9919825683716512, + "grad_norm": 0.4764333963394165, + "learning_rate": 0.0005059110264455401, + "loss": 0.4785, + "step": 17755 + }, + { + "epoch": 0.9920384389753332, + "grad_norm": 0.3675253391265869, + "learning_rate": 0.0005058830121021964, + "loss": 0.4152, + "step": 17756 + }, + { + "epoch": 0.992094309579015, + "grad_norm": 0.3967045843601227, + "learning_rate": 0.0005058549977588526, + "loss": 0.3416, + "step": 17757 + }, + { + "epoch": 0.9921501801826969, + "grad_norm": 0.5568209886550903, + "learning_rate": 0.0005058269834155087, + "loss": 0.4332, + "step": 17758 + }, + { + "epoch": 0.9922060507863788, + "grad_norm": 0.504359245300293, + "learning_rate": 0.000505798969072165, + "loss": 0.4517, + "step": 17759 + }, + { + "epoch": 0.9922619213900606, + "grad_norm": 0.4580940008163452, + "learning_rate": 0.0005057709547288212, + "loss": 0.4321, + "step": 17760 + }, + { + "epoch": 0.9923177919937425, + "grad_norm": 0.7205761075019836, + "learning_rate": 0.0005057429403854774, + "loss": 0.436, + "step": 17761 + }, + { + "epoch": 0.9923736625974243, + "grad_norm": 0.6187639832496643, + "learning_rate": 0.0005057149260421336, + "loss": 0.5091, + "step": 17762 + }, + { + "epoch": 0.9924295332011063, + "grad_norm": 0.6643950343132019, + "learning_rate": 0.0005056869116987898, + "loss": 0.56, + "step": 17763 + }, + { + "epoch": 0.9924854038047881, + "grad_norm": 0.44367071986198425, + "learning_rate": 0.000505658897355446, + "loss": 0.4795, + "step": 17764 + }, + { + "epoch": 0.99254127440847, + "grad_norm": 0.4544038772583008, + "learning_rate": 0.0005056308830121022, + "loss": 0.4227, + "step": 17765 + }, + { + "epoch": 0.9925971450121519, + "grad_norm": 0.45252320170402527, + "learning_rate": 0.0005056028686687584, + "loss": 0.4141, + "step": 17766 + }, + { + "epoch": 0.9926530156158337, + "grad_norm": 0.6000473499298096, + "learning_rate": 0.0005055748543254146, + "loss": 0.436, + "step": 17767 + }, + { + "epoch": 0.9927088862195156, + "grad_norm": 5.630963325500488, + "learning_rate": 0.0005055468399820708, + "loss": 0.3905, + "step": 17768 + }, + { + "epoch": 0.9927647568231974, + "grad_norm": 0.45847320556640625, + "learning_rate": 0.0005055188256387271, + "loss": 0.3584, + "step": 17769 + }, + { + "epoch": 0.9928206274268794, + "grad_norm": 3.365044593811035, + "learning_rate": 0.0005054908112953832, + "loss": 0.4148, + "step": 17770 + }, + { + "epoch": 0.9928764980305612, + "grad_norm": 0.46844327449798584, + "learning_rate": 0.0005054627969520394, + "loss": 0.5679, + "step": 17771 + }, + { + "epoch": 0.9929323686342431, + "grad_norm": 1.0261645317077637, + "learning_rate": 0.0005054347826086956, + "loss": 0.4927, + "step": 17772 + }, + { + "epoch": 0.992988239237925, + "grad_norm": 0.46149942278862, + "learning_rate": 0.000505406768265352, + "loss": 0.5394, + "step": 17773 + }, + { + "epoch": 0.9930441098416068, + "grad_norm": 0.371196448802948, + "learning_rate": 0.0005053787539220081, + "loss": 0.4864, + "step": 17774 + }, + { + "epoch": 0.9930999804452887, + "grad_norm": 0.4335106909275055, + "learning_rate": 0.0005053507395786643, + "loss": 0.4138, + "step": 17775 + }, + { + "epoch": 0.9931558510489706, + "grad_norm": 0.722576916217804, + "learning_rate": 0.0005053227252353205, + "loss": 0.5182, + "step": 17776 + }, + { + "epoch": 0.9932117216526525, + "grad_norm": 3.6740055084228516, + "learning_rate": 0.0005052947108919767, + "loss": 0.4957, + "step": 17777 + }, + { + "epoch": 0.9932675922563343, + "grad_norm": 0.979655385017395, + "learning_rate": 0.0005052666965486329, + "loss": 0.3341, + "step": 17778 + }, + { + "epoch": 0.9933234628600162, + "grad_norm": 0.44414499402046204, + "learning_rate": 0.0005052386822052891, + "loss": 0.3748, + "step": 17779 + }, + { + "epoch": 0.9933793334636981, + "grad_norm": 0.5602564811706543, + "learning_rate": 0.0005052106678619454, + "loss": 0.4463, + "step": 17780 + }, + { + "epoch": 0.99343520406738, + "grad_norm": 0.49204781651496887, + "learning_rate": 0.0005051826535186015, + "loss": 0.3814, + "step": 17781 + }, + { + "epoch": 0.9934910746710618, + "grad_norm": 0.3219582140445709, + "learning_rate": 0.0005051546391752577, + "loss": 0.4621, + "step": 17782 + }, + { + "epoch": 0.9935469452747437, + "grad_norm": 1.3394267559051514, + "learning_rate": 0.0005051266248319139, + "loss": 0.4518, + "step": 17783 + }, + { + "epoch": 0.9936028158784256, + "grad_norm": 0.44461819529533386, + "learning_rate": 0.0005050986104885702, + "loss": 0.453, + "step": 17784 + }, + { + "epoch": 0.9936586864821074, + "grad_norm": 0.47179633378982544, + "learning_rate": 0.0005050705961452263, + "loss": 0.3652, + "step": 17785 + }, + { + "epoch": 0.9937145570857893, + "grad_norm": 0.9033605456352234, + "learning_rate": 0.0005050425818018825, + "loss": 0.6077, + "step": 17786 + }, + { + "epoch": 0.9937704276894712, + "grad_norm": 0.44167736172676086, + "learning_rate": 0.0005050145674585388, + "loss": 0.4483, + "step": 17787 + }, + { + "epoch": 0.993826298293153, + "grad_norm": 0.458518385887146, + "learning_rate": 0.000504986553115195, + "loss": 0.4012, + "step": 17788 + }, + { + "epoch": 0.9938821688968349, + "grad_norm": 0.7863150835037231, + "learning_rate": 0.0005049585387718511, + "loss": 0.4456, + "step": 17789 + }, + { + "epoch": 0.9939380395005168, + "grad_norm": 0.3944619297981262, + "learning_rate": 0.0005049305244285074, + "loss": 0.4767, + "step": 17790 + }, + { + "epoch": 0.9939939101041987, + "grad_norm": 0.4200340211391449, + "learning_rate": 0.0005049025100851637, + "loss": 0.4312, + "step": 17791 + }, + { + "epoch": 0.9940497807078805, + "grad_norm": 0.3454003930091858, + "learning_rate": 0.0005048744957418199, + "loss": 0.4001, + "step": 17792 + }, + { + "epoch": 0.9941056513115624, + "grad_norm": 0.6579907536506653, + "learning_rate": 0.000504846481398476, + "loss": 0.4014, + "step": 17793 + }, + { + "epoch": 0.9941615219152443, + "grad_norm": 2.2757253646850586, + "learning_rate": 0.0005048184670551323, + "loss": 0.4374, + "step": 17794 + }, + { + "epoch": 0.9942173925189262, + "grad_norm": 0.5052655935287476, + "learning_rate": 0.0005047904527117885, + "loss": 0.4905, + "step": 17795 + }, + { + "epoch": 0.994273263122608, + "grad_norm": 0.5456076860427856, + "learning_rate": 0.0005047624383684447, + "loss": 0.4302, + "step": 17796 + }, + { + "epoch": 0.99432913372629, + "grad_norm": 0.7933659553527832, + "learning_rate": 0.0005047344240251008, + "loss": 0.4781, + "step": 17797 + }, + { + "epoch": 0.9943850043299718, + "grad_norm": 0.4655524790287018, + "learning_rate": 0.0005047064096817571, + "loss": 0.4711, + "step": 17798 + }, + { + "epoch": 0.9944408749336536, + "grad_norm": 0.7358472347259521, + "learning_rate": 0.0005046783953384133, + "loss": 0.51, + "step": 17799 + }, + { + "epoch": 0.9944967455373356, + "grad_norm": 0.48469942808151245, + "learning_rate": 0.0005046503809950694, + "loss": 0.48, + "step": 17800 + }, + { + "epoch": 0.9945526161410174, + "grad_norm": 0.41364586353302, + "learning_rate": 0.0005046223666517257, + "loss": 0.4601, + "step": 17801 + }, + { + "epoch": 0.9946084867446993, + "grad_norm": 4.496304988861084, + "learning_rate": 0.0005045943523083819, + "loss": 0.4581, + "step": 17802 + }, + { + "epoch": 0.9946643573483811, + "grad_norm": 0.7387206554412842, + "learning_rate": 0.0005045663379650381, + "loss": 0.4398, + "step": 17803 + }, + { + "epoch": 0.994720227952063, + "grad_norm": 0.46445488929748535, + "learning_rate": 0.0005045383236216942, + "loss": 0.5182, + "step": 17804 + }, + { + "epoch": 0.9947760985557449, + "grad_norm": 0.6511033177375793, + "learning_rate": 0.0005045103092783505, + "loss": 0.4348, + "step": 17805 + }, + { + "epoch": 0.9948319691594267, + "grad_norm": 0.46510785818099976, + "learning_rate": 0.0005044822949350068, + "loss": 0.4253, + "step": 17806 + }, + { + "epoch": 0.9948878397631087, + "grad_norm": 0.4601931571960449, + "learning_rate": 0.000504454280591663, + "loss": 0.409, + "step": 17807 + }, + { + "epoch": 0.9949437103667905, + "grad_norm": 0.5312385559082031, + "learning_rate": 0.0005044262662483191, + "loss": 0.4259, + "step": 17808 + }, + { + "epoch": 0.9949995809704724, + "grad_norm": 0.4327475428581238, + "learning_rate": 0.0005043982519049754, + "loss": 0.36, + "step": 17809 + }, + { + "epoch": 0.9950554515741542, + "grad_norm": 0.49435341358184814, + "learning_rate": 0.0005043702375616316, + "loss": 0.5611, + "step": 17810 + }, + { + "epoch": 0.9951113221778362, + "grad_norm": 0.46852362155914307, + "learning_rate": 0.0005043422232182878, + "loss": 0.4686, + "step": 17811 + }, + { + "epoch": 0.995167192781518, + "grad_norm": 0.3242324888706207, + "learning_rate": 0.000504314208874944, + "loss": 0.3862, + "step": 17812 + }, + { + "epoch": 0.9952230633851998, + "grad_norm": 0.6326824426651001, + "learning_rate": 0.0005042861945316002, + "loss": 0.4388, + "step": 17813 + }, + { + "epoch": 0.9952789339888818, + "grad_norm": 0.8404468297958374, + "learning_rate": 0.0005042581801882564, + "loss": 0.3493, + "step": 17814 + }, + { + "epoch": 0.9953348045925636, + "grad_norm": 1.341300368309021, + "learning_rate": 0.0005042301658449126, + "loss": 0.3814, + "step": 17815 + }, + { + "epoch": 0.9953906751962455, + "grad_norm": 0.48461228609085083, + "learning_rate": 0.0005042021515015688, + "loss": 0.4401, + "step": 17816 + }, + { + "epoch": 0.9954465457999274, + "grad_norm": 1.8348644971847534, + "learning_rate": 0.000504174137158225, + "loss": 0.4897, + "step": 17817 + }, + { + "epoch": 0.9955024164036093, + "grad_norm": 0.3491465449333191, + "learning_rate": 0.0005041461228148812, + "loss": 0.4884, + "step": 17818 + }, + { + "epoch": 0.9955582870072911, + "grad_norm": 0.49508556723594666, + "learning_rate": 0.0005041181084715374, + "loss": 0.3859, + "step": 17819 + }, + { + "epoch": 0.9956141576109729, + "grad_norm": 0.5938343405723572, + "learning_rate": 0.0005040900941281936, + "loss": 0.4142, + "step": 17820 + }, + { + "epoch": 0.9956700282146549, + "grad_norm": 0.5441776514053345, + "learning_rate": 0.0005040620797848498, + "loss": 0.4492, + "step": 17821 + }, + { + "epoch": 0.9957258988183367, + "grad_norm": 0.44318142533302307, + "learning_rate": 0.000504034065441506, + "loss": 0.4038, + "step": 17822 + }, + { + "epoch": 0.9957817694220186, + "grad_norm": 0.7051774859428406, + "learning_rate": 0.0005040060510981623, + "loss": 0.3548, + "step": 17823 + }, + { + "epoch": 0.9958376400257005, + "grad_norm": 0.7091790437698364, + "learning_rate": 0.0005039780367548185, + "loss": 0.5451, + "step": 17824 + }, + { + "epoch": 0.9958935106293824, + "grad_norm": 0.7516396641731262, + "learning_rate": 0.0005039500224114747, + "loss": 0.467, + "step": 17825 + }, + { + "epoch": 0.9959493812330642, + "grad_norm": 0.4183390140533447, + "learning_rate": 0.000503922008068131, + "loss": 0.4233, + "step": 17826 + }, + { + "epoch": 0.996005251836746, + "grad_norm": 2.2382090091705322, + "learning_rate": 0.0005038939937247871, + "loss": 0.3963, + "step": 17827 + }, + { + "epoch": 0.996061122440428, + "grad_norm": 0.4210306406021118, + "learning_rate": 0.0005038659793814433, + "loss": 0.3916, + "step": 17828 + }, + { + "epoch": 0.9961169930441098, + "grad_norm": 1.323868989944458, + "learning_rate": 0.0005038379650380995, + "loss": 0.412, + "step": 17829 + }, + { + "epoch": 0.9961728636477917, + "grad_norm": 0.5730027556419373, + "learning_rate": 0.0005038099506947558, + "loss": 0.5027, + "step": 17830 + }, + { + "epoch": 0.9962287342514736, + "grad_norm": 0.5043101906776428, + "learning_rate": 0.0005037819363514119, + "loss": 0.4896, + "step": 17831 + }, + { + "epoch": 0.9962846048551555, + "grad_norm": 0.5048205852508545, + "learning_rate": 0.0005037539220080681, + "loss": 0.5248, + "step": 17832 + }, + { + "epoch": 0.9963404754588373, + "grad_norm": 1.9959005117416382, + "learning_rate": 0.0005037259076647244, + "loss": 0.44, + "step": 17833 + }, + { + "epoch": 0.9963963460625193, + "grad_norm": 0.712533712387085, + "learning_rate": 0.0005036978933213806, + "loss": 0.4862, + "step": 17834 + }, + { + "epoch": 0.9964522166662011, + "grad_norm": 1.0090553760528564, + "learning_rate": 0.0005036698789780367, + "loss": 0.4834, + "step": 17835 + }, + { + "epoch": 0.9965080872698829, + "grad_norm": 0.34790852665901184, + "learning_rate": 0.0005036418646346929, + "loss": 0.361, + "step": 17836 + }, + { + "epoch": 0.9965639578735648, + "grad_norm": 0.4296223521232605, + "learning_rate": 0.0005036138502913492, + "loss": 0.3788, + "step": 17837 + }, + { + "epoch": 0.9966198284772467, + "grad_norm": 0.4489489495754242, + "learning_rate": 0.0005035858359480054, + "loss": 0.4997, + "step": 17838 + }, + { + "epoch": 0.9966756990809286, + "grad_norm": 0.5066514015197754, + "learning_rate": 0.0005035578216046615, + "loss": 0.4478, + "step": 17839 + }, + { + "epoch": 0.9967315696846104, + "grad_norm": 0.4113646149635315, + "learning_rate": 0.0005035298072613179, + "loss": 0.4751, + "step": 17840 + }, + { + "epoch": 0.9967874402882924, + "grad_norm": 0.4845876097679138, + "learning_rate": 0.0005035017929179741, + "loss": 0.3113, + "step": 17841 + }, + { + "epoch": 0.9968433108919742, + "grad_norm": 0.36890873312950134, + "learning_rate": 0.0005034737785746302, + "loss": 0.5517, + "step": 17842 + }, + { + "epoch": 0.996899181495656, + "grad_norm": 0.4794452488422394, + "learning_rate": 0.0005034457642312864, + "loss": 0.4698, + "step": 17843 + }, + { + "epoch": 0.9969550520993379, + "grad_norm": 0.6656431555747986, + "learning_rate": 0.0005034177498879427, + "loss": 0.5194, + "step": 17844 + }, + { + "epoch": 0.9970109227030198, + "grad_norm": 0.5159928202629089, + "learning_rate": 0.0005033897355445989, + "loss": 0.473, + "step": 17845 + }, + { + "epoch": 0.9970667933067017, + "grad_norm": 0.5138458013534546, + "learning_rate": 0.000503361721201255, + "loss": 0.4415, + "step": 17846 + }, + { + "epoch": 0.9971226639103835, + "grad_norm": 1.0434972047805786, + "learning_rate": 0.0005033337068579112, + "loss": 0.4982, + "step": 17847 + }, + { + "epoch": 0.9971785345140655, + "grad_norm": 0.5774309635162354, + "learning_rate": 0.0005033056925145675, + "loss": 0.4049, + "step": 17848 + }, + { + "epoch": 0.9972344051177473, + "grad_norm": 0.5036318898200989, + "learning_rate": 0.0005032776781712237, + "loss": 0.3986, + "step": 17849 + }, + { + "epoch": 0.9972902757214291, + "grad_norm": 8.789798736572266, + "learning_rate": 0.0005032496638278798, + "loss": 0.4696, + "step": 17850 + }, + { + "epoch": 0.9973461463251111, + "grad_norm": 0.36805322766304016, + "learning_rate": 0.0005032216494845361, + "loss": 0.4366, + "step": 17851 + }, + { + "epoch": 0.9974020169287929, + "grad_norm": 0.5906342267990112, + "learning_rate": 0.0005031936351411923, + "loss": 0.3886, + "step": 17852 + }, + { + "epoch": 0.9974578875324748, + "grad_norm": 0.6898302435874939, + "learning_rate": 0.0005031656207978485, + "loss": 0.557, + "step": 17853 + }, + { + "epoch": 0.9975137581361566, + "grad_norm": 0.5151469707489014, + "learning_rate": 0.0005031376064545046, + "loss": 0.4561, + "step": 17854 + }, + { + "epoch": 0.9975696287398386, + "grad_norm": 0.44518759846687317, + "learning_rate": 0.000503109592111161, + "loss": 0.4929, + "step": 17855 + }, + { + "epoch": 0.9976254993435204, + "grad_norm": 0.3929125964641571, + "learning_rate": 0.0005030815777678172, + "loss": 0.3981, + "step": 17856 + }, + { + "epoch": 0.9976813699472022, + "grad_norm": 0.3300730586051941, + "learning_rate": 0.0005030535634244734, + "loss": 0.3415, + "step": 17857 + }, + { + "epoch": 0.9977372405508842, + "grad_norm": 0.36458927392959595, + "learning_rate": 0.0005030255490811296, + "loss": 0.368, + "step": 17858 + }, + { + "epoch": 0.997793111154566, + "grad_norm": 1.3450015783309937, + "learning_rate": 0.0005029975347377858, + "loss": 0.4534, + "step": 17859 + }, + { + "epoch": 0.9978489817582479, + "grad_norm": 0.49592292308807373, + "learning_rate": 0.000502969520394442, + "loss": 0.5752, + "step": 17860 + }, + { + "epoch": 0.9979048523619297, + "grad_norm": 0.3562728762626648, + "learning_rate": 0.0005029415060510981, + "loss": 0.3912, + "step": 17861 + }, + { + "epoch": 0.9979607229656117, + "grad_norm": 1.580647587776184, + "learning_rate": 0.0005029134917077544, + "loss": 0.4712, + "step": 17862 + }, + { + "epoch": 0.9980165935692935, + "grad_norm": 0.5498859882354736, + "learning_rate": 0.0005028854773644106, + "loss": 0.4348, + "step": 17863 + }, + { + "epoch": 0.9980724641729753, + "grad_norm": 0.5235080122947693, + "learning_rate": 0.0005028574630210668, + "loss": 0.4463, + "step": 17864 + }, + { + "epoch": 0.9981283347766573, + "grad_norm": 0.9620032906532288, + "learning_rate": 0.000502829448677723, + "loss": 0.4942, + "step": 17865 + }, + { + "epoch": 0.9981842053803391, + "grad_norm": 0.4892062246799469, + "learning_rate": 0.0005028014343343792, + "loss": 0.4017, + "step": 17866 + }, + { + "epoch": 0.998240075984021, + "grad_norm": 0.36011773347854614, + "learning_rate": 0.0005027734199910354, + "loss": 0.347, + "step": 17867 + }, + { + "epoch": 0.9982959465877029, + "grad_norm": 0.3915625512599945, + "learning_rate": 0.0005027454056476916, + "loss": 0.3631, + "step": 17868 + }, + { + "epoch": 0.9983518171913848, + "grad_norm": 0.3095785975456238, + "learning_rate": 0.0005027173913043478, + "loss": 0.3694, + "step": 17869 + }, + { + "epoch": 0.9984076877950666, + "grad_norm": 0.4269053637981415, + "learning_rate": 0.000502689376961004, + "loss": 0.3591, + "step": 17870 + }, + { + "epoch": 0.9984635583987485, + "grad_norm": 0.46422263979911804, + "learning_rate": 0.0005026613626176602, + "loss": 0.3888, + "step": 17871 + }, + { + "epoch": 0.9985194290024304, + "grad_norm": 0.4080902934074402, + "learning_rate": 0.0005026333482743166, + "loss": 0.4721, + "step": 17872 + }, + { + "epoch": 0.9985752996061122, + "grad_norm": 0.46678343415260315, + "learning_rate": 0.0005026053339309727, + "loss": 0.478, + "step": 17873 + }, + { + "epoch": 0.9986311702097941, + "grad_norm": 0.7056798934936523, + "learning_rate": 0.0005025773195876289, + "loss": 0.539, + "step": 17874 + }, + { + "epoch": 0.998687040813476, + "grad_norm": 0.36755451560020447, + "learning_rate": 0.0005025493052442851, + "loss": 0.3647, + "step": 17875 + }, + { + "epoch": 0.9987429114171579, + "grad_norm": 0.46468421816825867, + "learning_rate": 0.0005025212909009414, + "loss": 0.5806, + "step": 17876 + }, + { + "epoch": 0.9987987820208397, + "grad_norm": 5.54908561706543, + "learning_rate": 0.0005024932765575975, + "loss": 0.613, + "step": 17877 + }, + { + "epoch": 0.9988546526245216, + "grad_norm": 0.39092764258384705, + "learning_rate": 0.0005024652622142537, + "loss": 0.3967, + "step": 17878 + }, + { + "epoch": 0.9989105232282035, + "grad_norm": 0.4766077399253845, + "learning_rate": 0.0005024372478709099, + "loss": 0.4577, + "step": 17879 + }, + { + "epoch": 0.9989663938318853, + "grad_norm": 0.8927488327026367, + "learning_rate": 0.0005024092335275662, + "loss": 0.6273, + "step": 17880 + }, + { + "epoch": 0.9990222644355672, + "grad_norm": 1.072130799293518, + "learning_rate": 0.0005023812191842223, + "loss": 0.6186, + "step": 17881 + }, + { + "epoch": 0.9990781350392491, + "grad_norm": 0.4757367670536041, + "learning_rate": 0.0005023532048408785, + "loss": 0.5148, + "step": 17882 + }, + { + "epoch": 0.999134005642931, + "grad_norm": 0.4403579831123352, + "learning_rate": 0.0005023251904975348, + "loss": 0.4165, + "step": 17883 + }, + { + "epoch": 0.9991898762466128, + "grad_norm": 0.6401516199111938, + "learning_rate": 0.0005022971761541909, + "loss": 0.4106, + "step": 17884 + }, + { + "epoch": 0.9992457468502948, + "grad_norm": 1.0280745029449463, + "learning_rate": 0.0005022691618108471, + "loss": 0.4989, + "step": 17885 + }, + { + "epoch": 0.9993016174539766, + "grad_norm": 0.3448118567466736, + "learning_rate": 0.0005022411474675033, + "loss": 0.4027, + "step": 17886 + }, + { + "epoch": 0.9993574880576584, + "grad_norm": 2.127251148223877, + "learning_rate": 0.0005022131331241596, + "loss": 0.5172, + "step": 17887 + }, + { + "epoch": 0.9994133586613403, + "grad_norm": 0.3182320296764374, + "learning_rate": 0.0005021851187808157, + "loss": 0.4005, + "step": 17888 + }, + { + "epoch": 0.9994692292650222, + "grad_norm": 0.5162581205368042, + "learning_rate": 0.0005021571044374719, + "loss": 0.438, + "step": 17889 + }, + { + "epoch": 0.9995250998687041, + "grad_norm": 0.3588276505470276, + "learning_rate": 0.0005021290900941283, + "loss": 0.4148, + "step": 17890 + }, + { + "epoch": 0.9995809704723859, + "grad_norm": 0.41627591848373413, + "learning_rate": 0.0005021010757507845, + "loss": 0.4085, + "step": 17891 + }, + { + "epoch": 0.9996368410760679, + "grad_norm": 0.5864758491516113, + "learning_rate": 0.0005020730614074406, + "loss": 0.5252, + "step": 17892 + }, + { + "epoch": 0.9996927116797497, + "grad_norm": 0.5606831908226013, + "learning_rate": 0.0005020450470640968, + "loss": 0.4145, + "step": 17893 + }, + { + "epoch": 0.9997485822834316, + "grad_norm": 0.39471495151519775, + "learning_rate": 0.0005020170327207531, + "loss": 0.4139, + "step": 17894 + }, + { + "epoch": 0.9998044528871134, + "grad_norm": 1.5852986574172974, + "learning_rate": 0.0005019890183774093, + "loss": 0.4672, + "step": 17895 + }, + { + "epoch": 0.9998603234907953, + "grad_norm": 0.325692355632782, + "learning_rate": 0.0005019610040340654, + "loss": 0.3887, + "step": 17896 + }, + { + "epoch": 0.9999161940944772, + "grad_norm": 0.8976548910140991, + "learning_rate": 0.0005019329896907217, + "loss": 0.3676, + "step": 17897 + }, + { + "epoch": 0.999972064698159, + "grad_norm": 0.4673980474472046, + "learning_rate": 0.0005019049753473779, + "loss": 0.3637, + "step": 17898 + }, + { + "epoch": 1.000055870603682, + "grad_norm": 0.5138245224952698, + "learning_rate": 0.0005018769610040341, + "loss": 0.5739, + "step": 17899 + }, + { + "epoch": 1.0001117412073637, + "grad_norm": 0.6027733087539673, + "learning_rate": 0.0005018489466606902, + "loss": 0.5205, + "step": 17900 + }, + { + "epoch": 1.0001676118110456, + "grad_norm": 0.4131118953227997, + "learning_rate": 0.0005018209323173465, + "loss": 0.4369, + "step": 17901 + }, + { + "epoch": 1.0002234824147276, + "grad_norm": 0.5411525964736938, + "learning_rate": 0.0005017929179740027, + "loss": 0.4213, + "step": 17902 + }, + { + "epoch": 1.0002793530184093, + "grad_norm": 0.3696529269218445, + "learning_rate": 0.0005017649036306588, + "loss": 0.4031, + "step": 17903 + }, + { + "epoch": 1.0003352236220913, + "grad_norm": 0.5179861187934875, + "learning_rate": 0.0005017368892873151, + "loss": 0.661, + "step": 17904 + }, + { + "epoch": 1.0003910942257732, + "grad_norm": 0.658402681350708, + "learning_rate": 0.0005017088749439713, + "loss": 0.4024, + "step": 17905 + }, + { + "epoch": 1.000446964829455, + "grad_norm": 0.4487610161304474, + "learning_rate": 0.0005016808606006276, + "loss": 0.3866, + "step": 17906 + }, + { + "epoch": 1.000502835433137, + "grad_norm": 0.6111307144165039, + "learning_rate": 0.0005016528462572837, + "loss": 0.3528, + "step": 17907 + }, + { + "epoch": 1.0005587060368186, + "grad_norm": 0.6617330312728882, + "learning_rate": 0.00050162483191394, + "loss": 0.5611, + "step": 17908 + }, + { + "epoch": 1.0006145766405006, + "grad_norm": 1.404229998588562, + "learning_rate": 0.0005015968175705962, + "loss": 0.5028, + "step": 17909 + }, + { + "epoch": 1.0006704472441825, + "grad_norm": 0.47040680050849915, + "learning_rate": 0.0005015688032272524, + "loss": 0.5165, + "step": 17910 + }, + { + "epoch": 1.0007263178478643, + "grad_norm": 2.2809197902679443, + "learning_rate": 0.0005015407888839085, + "loss": 0.3684, + "step": 17911 + }, + { + "epoch": 1.0007821884515462, + "grad_norm": 1.339878797531128, + "learning_rate": 0.0005015127745405648, + "loss": 0.5632, + "step": 17912 + }, + { + "epoch": 1.0008380590552282, + "grad_norm": 0.8332105875015259, + "learning_rate": 0.000501484760197221, + "loss": 0.4764, + "step": 17913 + }, + { + "epoch": 1.0008939296589099, + "grad_norm": 0.5567039847373962, + "learning_rate": 0.0005014567458538772, + "loss": 0.4504, + "step": 17914 + }, + { + "epoch": 1.0009498002625918, + "grad_norm": 0.6805049180984497, + "learning_rate": 0.0005014287315105334, + "loss": 0.435, + "step": 17915 + }, + { + "epoch": 1.0010056708662738, + "grad_norm": 0.48981645703315735, + "learning_rate": 0.0005014007171671896, + "loss": 0.4393, + "step": 17916 + }, + { + "epoch": 1.0010615414699555, + "grad_norm": 0.432752788066864, + "learning_rate": 0.0005013727028238458, + "loss": 0.4125, + "step": 17917 + }, + { + "epoch": 1.0011174120736375, + "grad_norm": 0.43536442518234253, + "learning_rate": 0.000501344688480502, + "loss": 0.4834, + "step": 17918 + }, + { + "epoch": 1.0011732826773194, + "grad_norm": 0.8386384844779968, + "learning_rate": 0.0005013166741371582, + "loss": 0.684, + "step": 17919 + }, + { + "epoch": 1.0012291532810011, + "grad_norm": 0.5621084570884705, + "learning_rate": 0.0005012886597938144, + "loss": 0.4186, + "step": 17920 + }, + { + "epoch": 1.001285023884683, + "grad_norm": 0.38822224736213684, + "learning_rate": 0.0005012606454504706, + "loss": 0.3811, + "step": 17921 + }, + { + "epoch": 1.001340894488365, + "grad_norm": 0.4951607584953308, + "learning_rate": 0.000501232631107127, + "loss": 0.4585, + "step": 17922 + }, + { + "epoch": 1.0013967650920468, + "grad_norm": 0.7676005363464355, + "learning_rate": 0.000501204616763783, + "loss": 0.4136, + "step": 17923 + }, + { + "epoch": 1.0014526356957287, + "grad_norm": 0.5295408964157104, + "learning_rate": 0.0005011766024204393, + "loss": 0.341, + "step": 17924 + }, + { + "epoch": 1.0015085062994105, + "grad_norm": 0.5481603741645813, + "learning_rate": 0.0005011485880770955, + "loss": 0.3806, + "step": 17925 + }, + { + "epoch": 1.0015643769030924, + "grad_norm": 1.5353974103927612, + "learning_rate": 0.0005011205737337517, + "loss": 0.4347, + "step": 17926 + }, + { + "epoch": 1.0016202475067744, + "grad_norm": 0.5801703929901123, + "learning_rate": 0.0005010925593904079, + "loss": 0.5371, + "step": 17927 + }, + { + "epoch": 1.001676118110456, + "grad_norm": 0.9258538484573364, + "learning_rate": 0.0005010645450470641, + "loss": 0.5297, + "step": 17928 + }, + { + "epoch": 1.001731988714138, + "grad_norm": 0.4018459618091583, + "learning_rate": 0.0005010365307037204, + "loss": 0.3419, + "step": 17929 + }, + { + "epoch": 1.00178785931782, + "grad_norm": 19.3364315032959, + "learning_rate": 0.0005010085163603765, + "loss": 0.4907, + "step": 17930 + }, + { + "epoch": 1.0018437299215017, + "grad_norm": 0.8678176403045654, + "learning_rate": 0.0005009805020170327, + "loss": 0.4023, + "step": 17931 + }, + { + "epoch": 1.0018996005251837, + "grad_norm": 0.6769725680351257, + "learning_rate": 0.0005009524876736889, + "loss": 0.4552, + "step": 17932 + }, + { + "epoch": 1.0019554711288656, + "grad_norm": 0.669452428817749, + "learning_rate": 0.0005009244733303452, + "loss": 0.3917, + "step": 17933 + }, + { + "epoch": 1.0020113417325474, + "grad_norm": 0.39489108324050903, + "learning_rate": 0.0005008964589870013, + "loss": 0.4572, + "step": 17934 + }, + { + "epoch": 1.0020672123362293, + "grad_norm": 10.21764087677002, + "learning_rate": 0.0005008684446436575, + "loss": 0.3661, + "step": 17935 + }, + { + "epoch": 1.0021230829399113, + "grad_norm": 0.5150909423828125, + "learning_rate": 0.0005008404303003138, + "loss": 0.4053, + "step": 17936 + }, + { + "epoch": 1.002178953543593, + "grad_norm": 0.5748830437660217, + "learning_rate": 0.00050081241595697, + "loss": 0.3986, + "step": 17937 + }, + { + "epoch": 1.002234824147275, + "grad_norm": 0.631584644317627, + "learning_rate": 0.0005007844016136261, + "loss": 0.4395, + "step": 17938 + }, + { + "epoch": 1.002290694750957, + "grad_norm": 0.5148150324821472, + "learning_rate": 0.0005007563872702823, + "loss": 0.3515, + "step": 17939 + }, + { + "epoch": 1.0023465653546386, + "grad_norm": 0.5785627365112305, + "learning_rate": 0.0005007283729269387, + "loss": 0.464, + "step": 17940 + }, + { + "epoch": 1.0024024359583206, + "grad_norm": 0.613819420337677, + "learning_rate": 0.0005007003585835949, + "loss": 0.3968, + "step": 17941 + }, + { + "epoch": 1.0024583065620023, + "grad_norm": 0.5907513499259949, + "learning_rate": 0.000500672344240251, + "loss": 0.5387, + "step": 17942 + }, + { + "epoch": 1.0025141771656843, + "grad_norm": 0.5706262588500977, + "learning_rate": 0.0005006443298969073, + "loss": 0.4924, + "step": 17943 + }, + { + "epoch": 1.0025700477693662, + "grad_norm": 0.49362704157829285, + "learning_rate": 0.0005006163155535635, + "loss": 0.4393, + "step": 17944 + }, + { + "epoch": 1.002625918373048, + "grad_norm": 0.9701772928237915, + "learning_rate": 0.0005005883012102196, + "loss": 0.4191, + "step": 17945 + }, + { + "epoch": 1.0026817889767299, + "grad_norm": 0.6064215302467346, + "learning_rate": 0.0005005602868668758, + "loss": 0.5729, + "step": 17946 + }, + { + "epoch": 1.0027376595804118, + "grad_norm": 0.5773648619651794, + "learning_rate": 0.0005005322725235321, + "loss": 0.4223, + "step": 17947 + }, + { + "epoch": 1.0027935301840936, + "grad_norm": 0.41647183895111084, + "learning_rate": 0.0005005042581801883, + "loss": 0.4357, + "step": 17948 + }, + { + "epoch": 1.0028494007877755, + "grad_norm": 0.9069080352783203, + "learning_rate": 0.0005004762438368444, + "loss": 0.3654, + "step": 17949 + }, + { + "epoch": 1.0029052713914575, + "grad_norm": 0.4761938452720642, + "learning_rate": 0.0005004482294935006, + "loss": 0.4382, + "step": 17950 + }, + { + "epoch": 1.0029611419951392, + "grad_norm": 0.4416482448577881, + "learning_rate": 0.0005004202151501569, + "loss": 0.4198, + "step": 17951 + }, + { + "epoch": 1.0030170125988211, + "grad_norm": 0.4265572428703308, + "learning_rate": 0.0005003922008068131, + "loss": 0.3001, + "step": 17952 + }, + { + "epoch": 1.003072883202503, + "grad_norm": 0.6397790312767029, + "learning_rate": 0.0005003641864634692, + "loss": 0.4702, + "step": 17953 + }, + { + "epoch": 1.0031287538061848, + "grad_norm": 0.5708019137382507, + "learning_rate": 0.0005003361721201255, + "loss": 0.4953, + "step": 17954 + }, + { + "epoch": 1.0031846244098668, + "grad_norm": 0.46864748001098633, + "learning_rate": 0.0005003081577767817, + "loss": 0.4632, + "step": 17955 + }, + { + "epoch": 1.0032404950135487, + "grad_norm": 0.5435078740119934, + "learning_rate": 0.000500280143433438, + "loss": 0.3819, + "step": 17956 + }, + { + "epoch": 1.0032963656172305, + "grad_norm": 0.47986212372779846, + "learning_rate": 0.000500252129090094, + "loss": 0.4727, + "step": 17957 + }, + { + "epoch": 1.0033522362209124, + "grad_norm": 0.9720260500907898, + "learning_rate": 0.0005002241147467504, + "loss": 0.4624, + "step": 17958 + }, + { + "epoch": 1.0034081068245941, + "grad_norm": 0.39070045948028564, + "learning_rate": 0.0005001961004034066, + "loss": 0.4559, + "step": 17959 + }, + { + "epoch": 1.003463977428276, + "grad_norm": 0.5389635562896729, + "learning_rate": 0.0005001680860600628, + "loss": 0.4729, + "step": 17960 + }, + { + "epoch": 1.003519848031958, + "grad_norm": 0.6589836478233337, + "learning_rate": 0.000500140071716719, + "loss": 0.5231, + "step": 17961 + }, + { + "epoch": 1.0035757186356398, + "grad_norm": 0.6115561127662659, + "learning_rate": 0.0005001120573733752, + "loss": 0.4295, + "step": 17962 + }, + { + "epoch": 1.0036315892393217, + "grad_norm": 1.0823957920074463, + "learning_rate": 0.0005000840430300314, + "loss": 0.5507, + "step": 17963 + }, + { + "epoch": 1.0036874598430037, + "grad_norm": 0.3943935036659241, + "learning_rate": 0.0005000560286866876, + "loss": 0.4414, + "step": 17964 + }, + { + "epoch": 1.0037433304466854, + "grad_norm": 1.4701447486877441, + "learning_rate": 0.0005000280143433438, + "loss": 0.4184, + "step": 17965 + }, + { + "epoch": 1.0037992010503674, + "grad_norm": 0.4794009029865265, + "learning_rate": 0.0005, + "loss": 0.5481, + "step": 17966 + }, + { + "epoch": 1.0038550716540493, + "grad_norm": 0.8504229187965393, + "learning_rate": 0.0004999719856566562, + "loss": 0.4074, + "step": 17967 + }, + { + "epoch": 1.003910942257731, + "grad_norm": 0.5864220857620239, + "learning_rate": 0.0004999439713133124, + "loss": 0.3767, + "step": 17968 + }, + { + "epoch": 1.003966812861413, + "grad_norm": 0.38640251755714417, + "learning_rate": 0.0004999159569699686, + "loss": 0.3756, + "step": 17969 + }, + { + "epoch": 1.004022683465095, + "grad_norm": 0.48518410325050354, + "learning_rate": 0.0004998879426266248, + "loss": 0.3801, + "step": 17970 + }, + { + "epoch": 1.0040785540687767, + "grad_norm": 0.6979808807373047, + "learning_rate": 0.000499859928283281, + "loss": 0.4813, + "step": 17971 + }, + { + "epoch": 1.0041344246724586, + "grad_norm": 0.5809674859046936, + "learning_rate": 0.0004998319139399372, + "loss": 0.5153, + "step": 17972 + }, + { + "epoch": 1.0041902952761406, + "grad_norm": 0.42386123538017273, + "learning_rate": 0.0004998038995965935, + "loss": 0.4039, + "step": 17973 + }, + { + "epoch": 1.0042461658798223, + "grad_norm": 0.9027739763259888, + "learning_rate": 0.0004997758852532497, + "loss": 0.3574, + "step": 17974 + }, + { + "epoch": 1.0043020364835042, + "grad_norm": 0.5889331102371216, + "learning_rate": 0.0004997478709099059, + "loss": 0.4423, + "step": 17975 + }, + { + "epoch": 1.004357907087186, + "grad_norm": 0.40388238430023193, + "learning_rate": 0.0004997198565665621, + "loss": 0.4004, + "step": 17976 + }, + { + "epoch": 1.004413777690868, + "grad_norm": 2.401179313659668, + "learning_rate": 0.0004996918422232183, + "loss": 0.4002, + "step": 17977 + }, + { + "epoch": 1.0044696482945499, + "grad_norm": 0.4398198127746582, + "learning_rate": 0.0004996638278798745, + "loss": 0.4982, + "step": 17978 + }, + { + "epoch": 1.0045255188982316, + "grad_norm": 0.5585030913352966, + "learning_rate": 0.0004996358135365307, + "loss": 0.5418, + "step": 17979 + }, + { + "epoch": 1.0045813895019136, + "grad_norm": 0.33261838555336, + "learning_rate": 0.0004996077991931869, + "loss": 0.4251, + "step": 17980 + }, + { + "epoch": 1.0046372601055955, + "grad_norm": 0.44790366291999817, + "learning_rate": 0.0004995797848498431, + "loss": 0.4349, + "step": 17981 + }, + { + "epoch": 1.0046931307092772, + "grad_norm": 0.6918690204620361, + "learning_rate": 0.0004995517705064994, + "loss": 0.4293, + "step": 17982 + }, + { + "epoch": 1.0047490013129592, + "grad_norm": 7.7210259437561035, + "learning_rate": 0.0004995237561631555, + "loss": 0.6187, + "step": 17983 + }, + { + "epoch": 1.0048048719166411, + "grad_norm": 0.7767944931983948, + "learning_rate": 0.0004994957418198118, + "loss": 0.5535, + "step": 17984 + }, + { + "epoch": 1.0048607425203229, + "grad_norm": 0.4021340608596802, + "learning_rate": 0.0004994677274764679, + "loss": 0.5678, + "step": 17985 + }, + { + "epoch": 1.0049166131240048, + "grad_norm": 0.383186936378479, + "learning_rate": 0.0004994397131331242, + "loss": 0.4636, + "step": 17986 + }, + { + "epoch": 1.0049724837276868, + "grad_norm": 0.7872123718261719, + "learning_rate": 0.0004994116987897803, + "loss": 0.407, + "step": 17987 + }, + { + "epoch": 1.0050283543313685, + "grad_norm": 0.4941841959953308, + "learning_rate": 0.0004993836844464365, + "loss": 0.4904, + "step": 17988 + }, + { + "epoch": 1.0050842249350505, + "grad_norm": 0.42959752678871155, + "learning_rate": 0.0004993556701030927, + "loss": 0.4654, + "step": 17989 + }, + { + "epoch": 1.0051400955387324, + "grad_norm": 0.4603096544742584, + "learning_rate": 0.000499327655759749, + "loss": 0.48, + "step": 17990 + }, + { + "epoch": 1.0051959661424141, + "grad_norm": 0.4025799036026001, + "learning_rate": 0.0004992996414164053, + "loss": 0.3534, + "step": 17991 + }, + { + "epoch": 1.005251836746096, + "grad_norm": 0.5358416438102722, + "learning_rate": 0.0004992716270730614, + "loss": 0.4353, + "step": 17992 + }, + { + "epoch": 1.0053077073497778, + "grad_norm": 1.5889114141464233, + "learning_rate": 0.0004992436127297177, + "loss": 0.4504, + "step": 17993 + }, + { + "epoch": 1.0053635779534598, + "grad_norm": 0.4644075632095337, + "learning_rate": 0.0004992155983863738, + "loss": 0.4208, + "step": 17994 + }, + { + "epoch": 1.0054194485571417, + "grad_norm": 0.3453684449195862, + "learning_rate": 0.0004991875840430301, + "loss": 0.3587, + "step": 17995 + }, + { + "epoch": 1.0054753191608234, + "grad_norm": 0.8021454215049744, + "learning_rate": 0.0004991595696996862, + "loss": 0.4186, + "step": 17996 + }, + { + "epoch": 1.0055311897645054, + "grad_norm": 2.5746893882751465, + "learning_rate": 0.0004991315553563425, + "loss": 0.4099, + "step": 17997 + }, + { + "epoch": 1.0055870603681873, + "grad_norm": 0.6293035745620728, + "learning_rate": 0.0004991035410129987, + "loss": 0.3743, + "step": 17998 + }, + { + "epoch": 1.005642930971869, + "grad_norm": 0.4794500172138214, + "learning_rate": 0.0004990755266696549, + "loss": 0.435, + "step": 17999 + }, + { + "epoch": 1.005698801575551, + "grad_norm": 1.7492680549621582, + "learning_rate": 0.0004990475123263111, + "loss": 0.46, + "step": 18000 + }, + { + "epoch": 1.005698801575551, + "eval_cer": 0.08809313388541565, + "eval_loss": 0.3313485383987427, + "eval_runtime": 56.0926, + "eval_samples_per_second": 80.902, + "eval_steps_per_second": 5.063, + "eval_wer": 0.34691768386357924, + "step": 18000 + }, + { + "epoch": 1.005754672179233, + "grad_norm": 0.7862080335617065, + "learning_rate": 0.0004990194979829673, + "loss": 0.5451, + "step": 18001 + }, + { + "epoch": 1.0058105427829147, + "grad_norm": 0.572189450263977, + "learning_rate": 0.0004989914836396235, + "loss": 0.4369, + "step": 18002 + }, + { + "epoch": 1.0058664133865967, + "grad_norm": 1.5324534177780151, + "learning_rate": 0.0004989634692962797, + "loss": 0.4021, + "step": 18003 + }, + { + "epoch": 1.0059222839902786, + "grad_norm": 0.8501786589622498, + "learning_rate": 0.0004989354549529359, + "loss": 0.3999, + "step": 18004 + }, + { + "epoch": 1.0059781545939603, + "grad_norm": 0.3964289128780365, + "learning_rate": 0.0004989074406095921, + "loss": 0.3621, + "step": 18005 + }, + { + "epoch": 1.0060340251976423, + "grad_norm": 0.33013811707496643, + "learning_rate": 0.0004988794262662484, + "loss": 0.3898, + "step": 18006 + }, + { + "epoch": 1.0060898958013242, + "grad_norm": 0.48788273334503174, + "learning_rate": 0.0004988514119229046, + "loss": 0.4448, + "step": 18007 + }, + { + "epoch": 1.006145766405006, + "grad_norm": 0.4211832582950592, + "learning_rate": 0.0004988233975795608, + "loss": 0.3229, + "step": 18008 + }, + { + "epoch": 1.006201637008688, + "grad_norm": 0.38278862833976746, + "learning_rate": 0.000498795383236217, + "loss": 0.3204, + "step": 18009 + }, + { + "epoch": 1.0062575076123697, + "grad_norm": 0.8701338171958923, + "learning_rate": 0.0004987673688928732, + "loss": 0.4207, + "step": 18010 + }, + { + "epoch": 1.0063133782160516, + "grad_norm": 0.3949558138847351, + "learning_rate": 0.0004987393545495294, + "loss": 0.4401, + "step": 18011 + }, + { + "epoch": 1.0063692488197336, + "grad_norm": 0.4141845405101776, + "learning_rate": 0.0004987113402061856, + "loss": 0.373, + "step": 18012 + }, + { + "epoch": 1.0064251194234153, + "grad_norm": 0.4620506167411804, + "learning_rate": 0.0004986833258628418, + "loss": 0.3935, + "step": 18013 + }, + { + "epoch": 1.0064809900270972, + "grad_norm": 2.6094844341278076, + "learning_rate": 0.000498655311519498, + "loss": 0.483, + "step": 18014 + }, + { + "epoch": 1.0065368606307792, + "grad_norm": 0.48986852169036865, + "learning_rate": 0.0004986272971761542, + "loss": 0.4945, + "step": 18015 + }, + { + "epoch": 1.006592731234461, + "grad_norm": 0.5332025289535522, + "learning_rate": 0.0004985992828328104, + "loss": 0.369, + "step": 18016 + }, + { + "epoch": 1.0066486018381429, + "grad_norm": 0.5082410573959351, + "learning_rate": 0.0004985712684894666, + "loss": 0.4521, + "step": 18017 + }, + { + "epoch": 1.0067044724418248, + "grad_norm": 0.33739471435546875, + "learning_rate": 0.0004985432541461228, + "loss": 0.3933, + "step": 18018 + }, + { + "epoch": 1.0067603430455065, + "grad_norm": 1.4863418340682983, + "learning_rate": 0.000498515239802779, + "loss": 0.7272, + "step": 18019 + }, + { + "epoch": 1.0068162136491885, + "grad_norm": 0.4827415347099304, + "learning_rate": 0.0004984872254594352, + "loss": 0.547, + "step": 18020 + }, + { + "epoch": 1.0068720842528704, + "grad_norm": 0.3714950680732727, + "learning_rate": 0.0004984592111160914, + "loss": 0.4035, + "step": 18021 + }, + { + "epoch": 1.0069279548565522, + "grad_norm": 1.3299787044525146, + "learning_rate": 0.0004984311967727476, + "loss": 0.5954, + "step": 18022 + }, + { + "epoch": 1.0069838254602341, + "grad_norm": 0.5519317388534546, + "learning_rate": 0.0004984031824294039, + "loss": 0.4143, + "step": 18023 + }, + { + "epoch": 1.007039696063916, + "grad_norm": 0.8645349144935608, + "learning_rate": 0.0004983751680860601, + "loss": 0.3616, + "step": 18024 + }, + { + "epoch": 1.0070955666675978, + "grad_norm": 0.5340176820755005, + "learning_rate": 0.0004983471537427163, + "loss": 0.5826, + "step": 18025 + }, + { + "epoch": 1.0071514372712798, + "grad_norm": 0.40448251366615295, + "learning_rate": 0.0004983191393993725, + "loss": 0.5458, + "step": 18026 + }, + { + "epoch": 1.0072073078749615, + "grad_norm": 0.36338287591934204, + "learning_rate": 0.0004982911250560287, + "loss": 0.4714, + "step": 18027 + }, + { + "epoch": 1.0072631784786434, + "grad_norm": 1.274100422859192, + "learning_rate": 0.0004982631107126849, + "loss": 0.4585, + "step": 18028 + }, + { + "epoch": 1.0073190490823254, + "grad_norm": 0.35685858130455017, + "learning_rate": 0.0004982350963693411, + "loss": 0.3908, + "step": 18029 + }, + { + "epoch": 1.0073749196860071, + "grad_norm": 1.6104340553283691, + "learning_rate": 0.0004982070820259973, + "loss": 0.5584, + "step": 18030 + }, + { + "epoch": 1.007430790289689, + "grad_norm": 0.37906500697135925, + "learning_rate": 0.0004981790676826535, + "loss": 0.3851, + "step": 18031 + }, + { + "epoch": 1.007486660893371, + "grad_norm": 0.5978991389274597, + "learning_rate": 0.0004981510533393097, + "loss": 0.3685, + "step": 18032 + }, + { + "epoch": 1.0075425314970528, + "grad_norm": 0.6120970249176025, + "learning_rate": 0.0004981230389959659, + "loss": 0.6364, + "step": 18033 + }, + { + "epoch": 1.0075984021007347, + "grad_norm": 13.498554229736328, + "learning_rate": 0.0004980950246526221, + "loss": 0.4109, + "step": 18034 + }, + { + "epoch": 1.0076542727044167, + "grad_norm": 0.5517807602882385, + "learning_rate": 0.0004980670103092783, + "loss": 0.4701, + "step": 18035 + }, + { + "epoch": 1.0077101433080984, + "grad_norm": 0.6550309658050537, + "learning_rate": 0.0004980389959659345, + "loss": 0.6047, + "step": 18036 + }, + { + "epoch": 1.0077660139117803, + "grad_norm": 0.4880651831626892, + "learning_rate": 0.0004980109816225908, + "loss": 0.3997, + "step": 18037 + }, + { + "epoch": 1.0078218845154623, + "grad_norm": 0.4615819752216339, + "learning_rate": 0.0004979829672792469, + "loss": 0.374, + "step": 18038 + }, + { + "epoch": 1.007877755119144, + "grad_norm": 0.7574900388717651, + "learning_rate": 0.0004979549529359033, + "loss": 0.4906, + "step": 18039 + }, + { + "epoch": 1.007933625722826, + "grad_norm": 0.4080792963504791, + "learning_rate": 0.0004979269385925593, + "loss": 0.4319, + "step": 18040 + }, + { + "epoch": 1.007989496326508, + "grad_norm": 1.3909056186676025, + "learning_rate": 0.0004978989242492157, + "loss": 0.4952, + "step": 18041 + }, + { + "epoch": 1.0080453669301896, + "grad_norm": 0.34794163703918457, + "learning_rate": 0.0004978709099058718, + "loss": 0.4728, + "step": 18042 + }, + { + "epoch": 1.0081012375338716, + "grad_norm": 0.7897557616233826, + "learning_rate": 0.0004978428955625281, + "loss": 0.5761, + "step": 18043 + }, + { + "epoch": 1.0081571081375533, + "grad_norm": 0.7249839901924133, + "learning_rate": 0.0004978148812191842, + "loss": 0.4081, + "step": 18044 + }, + { + "epoch": 1.0082129787412353, + "grad_norm": 0.4275643527507782, + "learning_rate": 0.0004977868668758405, + "loss": 0.3575, + "step": 18045 + }, + { + "epoch": 1.0082688493449172, + "grad_norm": 0.353468120098114, + "learning_rate": 0.0004977588525324967, + "loss": 0.4126, + "step": 18046 + }, + { + "epoch": 1.008324719948599, + "grad_norm": 0.6711594462394714, + "learning_rate": 0.0004977308381891529, + "loss": 0.4658, + "step": 18047 + }, + { + "epoch": 1.008380590552281, + "grad_norm": 3.413087844848633, + "learning_rate": 0.0004977028238458091, + "loss": 0.3844, + "step": 18048 + }, + { + "epoch": 1.0084364611559629, + "grad_norm": 0.5608792304992676, + "learning_rate": 0.0004976748095024653, + "loss": 0.3667, + "step": 18049 + }, + { + "epoch": 1.0084923317596446, + "grad_norm": 0.5015581846237183, + "learning_rate": 0.0004976467951591215, + "loss": 0.5898, + "step": 18050 + }, + { + "epoch": 1.0085482023633265, + "grad_norm": 0.4385131299495697, + "learning_rate": 0.0004976187808157776, + "loss": 0.407, + "step": 18051 + }, + { + "epoch": 1.0086040729670085, + "grad_norm": 0.35579460859298706, + "learning_rate": 0.0004975907664724339, + "loss": 0.4415, + "step": 18052 + }, + { + "epoch": 1.0086599435706902, + "grad_norm": 0.4544835388660431, + "learning_rate": 0.0004975627521290901, + "loss": 0.3804, + "step": 18053 + }, + { + "epoch": 1.0087158141743722, + "grad_norm": 0.8403786420822144, + "learning_rate": 0.0004975347377857463, + "loss": 0.4302, + "step": 18054 + }, + { + "epoch": 1.0087716847780541, + "grad_norm": 0.4057299792766571, + "learning_rate": 0.0004975067234424025, + "loss": 0.3904, + "step": 18055 + }, + { + "epoch": 1.0088275553817359, + "grad_norm": 0.5640223622322083, + "learning_rate": 0.0004974787090990588, + "loss": 0.4543, + "step": 18056 + }, + { + "epoch": 1.0088834259854178, + "grad_norm": 0.30360716581344604, + "learning_rate": 0.000497450694755715, + "loss": 0.3139, + "step": 18057 + }, + { + "epoch": 1.0089392965890998, + "grad_norm": 0.6241030693054199, + "learning_rate": 0.0004974226804123712, + "loss": 0.6855, + "step": 18058 + }, + { + "epoch": 1.0089951671927815, + "grad_norm": 0.5085378289222717, + "learning_rate": 0.0004973946660690274, + "loss": 0.4988, + "step": 18059 + }, + { + "epoch": 1.0090510377964634, + "grad_norm": 0.4395587146282196, + "learning_rate": 0.0004973666517256836, + "loss": 0.38, + "step": 18060 + }, + { + "epoch": 1.0091069084001452, + "grad_norm": 0.5665132403373718, + "learning_rate": 0.0004973386373823398, + "loss": 0.4853, + "step": 18061 + }, + { + "epoch": 1.0091627790038271, + "grad_norm": 0.40105104446411133, + "learning_rate": 0.000497310623038996, + "loss": 0.3609, + "step": 18062 + }, + { + "epoch": 1.009218649607509, + "grad_norm": 0.4568201005458832, + "learning_rate": 0.0004972826086956522, + "loss": 0.3526, + "step": 18063 + }, + { + "epoch": 1.0092745202111908, + "grad_norm": 0.7819159626960754, + "learning_rate": 0.0004972545943523084, + "loss": 0.4209, + "step": 18064 + }, + { + "epoch": 1.0093303908148727, + "grad_norm": 0.5902245044708252, + "learning_rate": 0.0004972265800089646, + "loss": 0.4328, + "step": 18065 + }, + { + "epoch": 1.0093862614185547, + "grad_norm": 1.4297749996185303, + "learning_rate": 0.0004971985656656208, + "loss": 0.4328, + "step": 18066 + }, + { + "epoch": 1.0094421320222364, + "grad_norm": 0.4927610754966736, + "learning_rate": 0.000497170551322277, + "loss": 0.4397, + "step": 18067 + }, + { + "epoch": 1.0094980026259184, + "grad_norm": 0.44963228702545166, + "learning_rate": 0.0004971425369789332, + "loss": 0.4671, + "step": 18068 + }, + { + "epoch": 1.0095538732296003, + "grad_norm": 0.3458667993545532, + "learning_rate": 0.0004971145226355894, + "loss": 0.363, + "step": 18069 + }, + { + "epoch": 1.009609743833282, + "grad_norm": 0.6023271679878235, + "learning_rate": 0.0004970865082922456, + "loss": 0.3454, + "step": 18070 + }, + { + "epoch": 1.009665614436964, + "grad_norm": 0.2912629246711731, + "learning_rate": 0.0004970584939489018, + "loss": 0.345, + "step": 18071 + }, + { + "epoch": 1.009721485040646, + "grad_norm": 1.1206573247909546, + "learning_rate": 0.000497030479605558, + "loss": 0.5536, + "step": 18072 + }, + { + "epoch": 1.0097773556443277, + "grad_norm": 0.44196847081184387, + "learning_rate": 0.0004970024652622142, + "loss": 0.3835, + "step": 18073 + }, + { + "epoch": 1.0098332262480096, + "grad_norm": 0.35229822993278503, + "learning_rate": 0.0004969744509188705, + "loss": 0.3557, + "step": 18074 + }, + { + "epoch": 1.0098890968516916, + "grad_norm": 0.44763198494911194, + "learning_rate": 0.0004969464365755267, + "loss": 0.4737, + "step": 18075 + }, + { + "epoch": 1.0099449674553733, + "grad_norm": 0.3950290083885193, + "learning_rate": 0.0004969184222321829, + "loss": 0.4091, + "step": 18076 + }, + { + "epoch": 1.0100008380590553, + "grad_norm": 1.1919450759887695, + "learning_rate": 0.0004968904078888391, + "loss": 0.7522, + "step": 18077 + }, + { + "epoch": 1.010056708662737, + "grad_norm": 0.41085949540138245, + "learning_rate": 0.0004968623935454953, + "loss": 0.3522, + "step": 18078 + }, + { + "epoch": 1.010112579266419, + "grad_norm": 0.35860997438430786, + "learning_rate": 0.0004968343792021515, + "loss": 0.3611, + "step": 18079 + }, + { + "epoch": 1.010168449870101, + "grad_norm": 0.6458920836448669, + "learning_rate": 0.0004968063648588077, + "loss": 0.3898, + "step": 18080 + }, + { + "epoch": 1.0102243204737826, + "grad_norm": 0.3608826994895935, + "learning_rate": 0.0004967783505154639, + "loss": 0.3764, + "step": 18081 + }, + { + "epoch": 1.0102801910774646, + "grad_norm": 0.42067739367485046, + "learning_rate": 0.0004967503361721201, + "loss": 0.5213, + "step": 18082 + }, + { + "epoch": 1.0103360616811465, + "grad_norm": 0.5538346767425537, + "learning_rate": 0.0004967223218287763, + "loss": 0.3356, + "step": 18083 + }, + { + "epoch": 1.0103919322848283, + "grad_norm": 0.7849851846694946, + "learning_rate": 0.0004966943074854325, + "loss": 0.3029, + "step": 18084 + }, + { + "epoch": 1.0104478028885102, + "grad_norm": 0.5499436259269714, + "learning_rate": 0.0004966662931420888, + "loss": 0.5307, + "step": 18085 + }, + { + "epoch": 1.0105036734921922, + "grad_norm": 0.39805668592453003, + "learning_rate": 0.0004966382787987449, + "loss": 0.5086, + "step": 18086 + }, + { + "epoch": 1.010559544095874, + "grad_norm": 0.44753557443618774, + "learning_rate": 0.0004966102644554012, + "loss": 0.3705, + "step": 18087 + }, + { + "epoch": 1.0106154146995558, + "grad_norm": 0.3956845700740814, + "learning_rate": 0.0004965822501120573, + "loss": 0.3687, + "step": 18088 + }, + { + "epoch": 1.0106712853032378, + "grad_norm": 0.5121334195137024, + "learning_rate": 0.0004965542357687137, + "loss": 0.4202, + "step": 18089 + }, + { + "epoch": 1.0107271559069195, + "grad_norm": 2.7821872234344482, + "learning_rate": 0.0004965262214253697, + "loss": 0.3954, + "step": 18090 + }, + { + "epoch": 1.0107830265106015, + "grad_norm": 0.494444340467453, + "learning_rate": 0.0004964982070820261, + "loss": 0.4299, + "step": 18091 + }, + { + "epoch": 1.0108388971142834, + "grad_norm": 0.8788852691650391, + "learning_rate": 0.0004964701927386822, + "loss": 0.4214, + "step": 18092 + }, + { + "epoch": 1.0108947677179652, + "grad_norm": 0.47244441509246826, + "learning_rate": 0.0004964421783953385, + "loss": 0.4453, + "step": 18093 + }, + { + "epoch": 1.010950638321647, + "grad_norm": 0.4027687609195709, + "learning_rate": 0.0004964141640519947, + "loss": 0.3535, + "step": 18094 + }, + { + "epoch": 1.0110065089253288, + "grad_norm": 0.5314896106719971, + "learning_rate": 0.0004963861497086508, + "loss": 0.437, + "step": 18095 + }, + { + "epoch": 1.0110623795290108, + "grad_norm": 0.4761708974838257, + "learning_rate": 0.0004963581353653071, + "loss": 0.3078, + "step": 18096 + }, + { + "epoch": 1.0111182501326927, + "grad_norm": 0.694304883480072, + "learning_rate": 0.0004963301210219632, + "loss": 0.4095, + "step": 18097 + }, + { + "epoch": 1.0111741207363745, + "grad_norm": 0.5779362916946411, + "learning_rate": 0.0004963021066786195, + "loss": 0.5201, + "step": 18098 + }, + { + "epoch": 1.0112299913400564, + "grad_norm": 0.5204885601997375, + "learning_rate": 0.0004962740923352756, + "loss": 0.5535, + "step": 18099 + }, + { + "epoch": 1.0112858619437384, + "grad_norm": 2.0302343368530273, + "learning_rate": 0.0004962460779919319, + "loss": 0.4293, + "step": 18100 + }, + { + "epoch": 1.01134173254742, + "grad_norm": 0.9660356044769287, + "learning_rate": 0.0004962180636485881, + "loss": 0.439, + "step": 18101 + }, + { + "epoch": 1.011397603151102, + "grad_norm": 0.8166044354438782, + "learning_rate": 0.0004961900493052443, + "loss": 0.4244, + "step": 18102 + }, + { + "epoch": 1.011453473754784, + "grad_norm": 0.45879223942756653, + "learning_rate": 0.0004961620349619005, + "loss": 0.558, + "step": 18103 + }, + { + "epoch": 1.0115093443584657, + "grad_norm": 0.3514538109302521, + "learning_rate": 0.0004961340206185567, + "loss": 0.3742, + "step": 18104 + }, + { + "epoch": 1.0115652149621477, + "grad_norm": 0.4094843566417694, + "learning_rate": 0.0004961060062752129, + "loss": 0.3962, + "step": 18105 + }, + { + "epoch": 1.0116210855658296, + "grad_norm": 0.5265064239501953, + "learning_rate": 0.0004960779919318691, + "loss": 0.3759, + "step": 18106 + }, + { + "epoch": 1.0116769561695114, + "grad_norm": 0.7083197832107544, + "learning_rate": 0.0004960499775885254, + "loss": 0.4532, + "step": 18107 + }, + { + "epoch": 1.0117328267731933, + "grad_norm": 0.4080544114112854, + "learning_rate": 0.0004960219632451816, + "loss": 0.4373, + "step": 18108 + }, + { + "epoch": 1.011788697376875, + "grad_norm": 0.5497007369995117, + "learning_rate": 0.0004959939489018378, + "loss": 0.3411, + "step": 18109 + }, + { + "epoch": 1.011844567980557, + "grad_norm": 0.42736223340034485, + "learning_rate": 0.000495965934558494, + "loss": 0.4384, + "step": 18110 + }, + { + "epoch": 1.011900438584239, + "grad_norm": 0.4412011504173279, + "learning_rate": 0.0004959379202151502, + "loss": 0.4953, + "step": 18111 + }, + { + "epoch": 1.0119563091879207, + "grad_norm": 0.41814351081848145, + "learning_rate": 0.0004959099058718064, + "loss": 0.3907, + "step": 18112 + }, + { + "epoch": 1.0120121797916026, + "grad_norm": 0.4589410722255707, + "learning_rate": 0.0004958818915284626, + "loss": 0.2669, + "step": 18113 + }, + { + "epoch": 1.0120680503952846, + "grad_norm": 0.4663478434085846, + "learning_rate": 0.0004958538771851188, + "loss": 0.4484, + "step": 18114 + }, + { + "epoch": 1.0121239209989663, + "grad_norm": 0.4966093897819519, + "learning_rate": 0.000495825862841775, + "loss": 0.4615, + "step": 18115 + }, + { + "epoch": 1.0121797916026483, + "grad_norm": 0.34178125858306885, + "learning_rate": 0.0004957978484984312, + "loss": 0.381, + "step": 18116 + }, + { + "epoch": 1.0122356622063302, + "grad_norm": 0.5162151455879211, + "learning_rate": 0.0004957698341550874, + "loss": 0.3645, + "step": 18117 + }, + { + "epoch": 1.012291532810012, + "grad_norm": 0.4317743480205536, + "learning_rate": 0.0004957418198117436, + "loss": 0.4355, + "step": 18118 + }, + { + "epoch": 1.012347403413694, + "grad_norm": 0.7051270604133606, + "learning_rate": 0.0004957138054683998, + "loss": 0.4034, + "step": 18119 + }, + { + "epoch": 1.0124032740173758, + "grad_norm": 0.7128881812095642, + "learning_rate": 0.000495685791125056, + "loss": 0.367, + "step": 18120 + }, + { + "epoch": 1.0124591446210576, + "grad_norm": 0.37837842106819153, + "learning_rate": 0.0004956577767817122, + "loss": 0.3641, + "step": 18121 + }, + { + "epoch": 1.0125150152247395, + "grad_norm": 0.4411306381225586, + "learning_rate": 0.0004956297624383684, + "loss": 0.3955, + "step": 18122 + }, + { + "epoch": 1.0125708858284215, + "grad_norm": 1.0682740211486816, + "learning_rate": 0.0004956017480950246, + "loss": 0.4397, + "step": 18123 + }, + { + "epoch": 1.0126267564321032, + "grad_norm": 0.35829704999923706, + "learning_rate": 0.0004955737337516809, + "loss": 0.4383, + "step": 18124 + }, + { + "epoch": 1.0126826270357852, + "grad_norm": 0.42400285601615906, + "learning_rate": 0.0004955457194083371, + "loss": 0.4351, + "step": 18125 + }, + { + "epoch": 1.012738497639467, + "grad_norm": 0.4558722674846649, + "learning_rate": 0.0004955177050649933, + "loss": 0.4513, + "step": 18126 + }, + { + "epoch": 1.0127943682431488, + "grad_norm": 0.43528538942337036, + "learning_rate": 0.0004954896907216495, + "loss": 0.4359, + "step": 18127 + }, + { + "epoch": 1.0128502388468308, + "grad_norm": 0.6899311542510986, + "learning_rate": 0.0004954616763783057, + "loss": 0.5476, + "step": 18128 + }, + { + "epoch": 1.0129061094505125, + "grad_norm": 0.6298003792762756, + "learning_rate": 0.0004954336620349619, + "loss": 0.4154, + "step": 18129 + }, + { + "epoch": 1.0129619800541945, + "grad_norm": 1.2831366062164307, + "learning_rate": 0.0004954056476916181, + "loss": 0.3839, + "step": 18130 + }, + { + "epoch": 1.0130178506578764, + "grad_norm": 0.5483906865119934, + "learning_rate": 0.0004953776333482743, + "loss": 0.5257, + "step": 18131 + }, + { + "epoch": 1.0130737212615581, + "grad_norm": 0.37659335136413574, + "learning_rate": 0.0004953496190049305, + "loss": 0.4896, + "step": 18132 + }, + { + "epoch": 1.01312959186524, + "grad_norm": 0.6583579778671265, + "learning_rate": 0.0004953216046615868, + "loss": 0.4475, + "step": 18133 + }, + { + "epoch": 1.013185462468922, + "grad_norm": 0.3411029279232025, + "learning_rate": 0.0004952935903182429, + "loss": 0.4221, + "step": 18134 + }, + { + "epoch": 1.0132413330726038, + "grad_norm": 0.4745800197124481, + "learning_rate": 0.0004952655759748992, + "loss": 0.4358, + "step": 18135 + }, + { + "epoch": 1.0132972036762857, + "grad_norm": 0.9332897067070007, + "learning_rate": 0.0004952375616315553, + "loss": 0.4629, + "step": 18136 + }, + { + "epoch": 1.0133530742799677, + "grad_norm": 0.4799421429634094, + "learning_rate": 0.0004952095472882115, + "loss": 0.6392, + "step": 18137 + }, + { + "epoch": 1.0134089448836494, + "grad_norm": 0.639891505241394, + "learning_rate": 0.0004951815329448677, + "loss": 0.4596, + "step": 18138 + }, + { + "epoch": 1.0134648154873314, + "grad_norm": 0.5377475619316101, + "learning_rate": 0.0004951535186015239, + "loss": 0.5476, + "step": 18139 + }, + { + "epoch": 1.0135206860910133, + "grad_norm": 0.5400976538658142, + "learning_rate": 0.0004951255042581803, + "loss": 0.4001, + "step": 18140 + }, + { + "epoch": 1.013576556694695, + "grad_norm": 0.43373122811317444, + "learning_rate": 0.0004950974899148364, + "loss": 0.447, + "step": 18141 + }, + { + "epoch": 1.013632427298377, + "grad_norm": 0.5187661051750183, + "learning_rate": 0.0004950694755714927, + "loss": 0.4413, + "step": 18142 + }, + { + "epoch": 1.0136882979020587, + "grad_norm": 0.49251627922058105, + "learning_rate": 0.0004950414612281488, + "loss": 0.4223, + "step": 18143 + }, + { + "epoch": 1.0137441685057407, + "grad_norm": 0.5896259546279907, + "learning_rate": 0.0004950134468848051, + "loss": 0.4392, + "step": 18144 + }, + { + "epoch": 1.0138000391094226, + "grad_norm": 0.4111991822719574, + "learning_rate": 0.0004949854325414612, + "loss": 0.3519, + "step": 18145 + }, + { + "epoch": 1.0138559097131044, + "grad_norm": 3.801828145980835, + "learning_rate": 0.0004949574181981175, + "loss": 0.4272, + "step": 18146 + }, + { + "epoch": 1.0139117803167863, + "grad_norm": 0.4572485089302063, + "learning_rate": 0.0004949294038547736, + "loss": 0.46, + "step": 18147 + }, + { + "epoch": 1.0139676509204683, + "grad_norm": 0.6517757177352905, + "learning_rate": 0.0004949013895114299, + "loss": 0.5464, + "step": 18148 + }, + { + "epoch": 1.01402352152415, + "grad_norm": 0.48967409133911133, + "learning_rate": 0.0004948733751680861, + "loss": 0.4876, + "step": 18149 + }, + { + "epoch": 1.014079392127832, + "grad_norm": 0.9091432690620422, + "learning_rate": 0.0004948453608247423, + "loss": 0.5582, + "step": 18150 + }, + { + "epoch": 1.0141352627315139, + "grad_norm": 0.42380619049072266, + "learning_rate": 0.0004948173464813985, + "loss": 0.487, + "step": 18151 + }, + { + "epoch": 1.0141911333351956, + "grad_norm": 0.46711355447769165, + "learning_rate": 0.0004947893321380547, + "loss": 0.361, + "step": 18152 + }, + { + "epoch": 1.0142470039388776, + "grad_norm": 0.8633831739425659, + "learning_rate": 0.0004947613177947109, + "loss": 0.3879, + "step": 18153 + }, + { + "epoch": 1.0143028745425595, + "grad_norm": 0.5408996939659119, + "learning_rate": 0.0004947333034513671, + "loss": 0.4337, + "step": 18154 + }, + { + "epoch": 1.0143587451462412, + "grad_norm": 1.0463485717773438, + "learning_rate": 0.0004947052891080233, + "loss": 0.5297, + "step": 18155 + }, + { + "epoch": 1.0144146157499232, + "grad_norm": 0.4079901874065399, + "learning_rate": 0.0004946772747646795, + "loss": 0.479, + "step": 18156 + }, + { + "epoch": 1.0144704863536051, + "grad_norm": 0.9289140701293945, + "learning_rate": 0.0004946492604213358, + "loss": 0.2988, + "step": 18157 + }, + { + "epoch": 1.0145263569572869, + "grad_norm": 0.691033661365509, + "learning_rate": 0.000494621246077992, + "loss": 0.362, + "step": 18158 + }, + { + "epoch": 1.0145822275609688, + "grad_norm": 0.5044304728507996, + "learning_rate": 0.0004945932317346482, + "loss": 0.4903, + "step": 18159 + }, + { + "epoch": 1.0146380981646506, + "grad_norm": 0.538696825504303, + "learning_rate": 0.0004945652173913044, + "loss": 0.3056, + "step": 18160 + }, + { + "epoch": 1.0146939687683325, + "grad_norm": 0.3487122058868408, + "learning_rate": 0.0004945372030479606, + "loss": 0.3389, + "step": 18161 + }, + { + "epoch": 1.0147498393720145, + "grad_norm": 0.4312290847301483, + "learning_rate": 0.0004945091887046168, + "loss": 0.443, + "step": 18162 + }, + { + "epoch": 1.0148057099756962, + "grad_norm": 0.3152795135974884, + "learning_rate": 0.000494481174361273, + "loss": 0.3892, + "step": 18163 + }, + { + "epoch": 1.0148615805793781, + "grad_norm": 0.5115416049957275, + "learning_rate": 0.0004944531600179292, + "loss": 0.4803, + "step": 18164 + }, + { + "epoch": 1.01491745118306, + "grad_norm": 0.5423107743263245, + "learning_rate": 0.0004944251456745854, + "loss": 0.4879, + "step": 18165 + }, + { + "epoch": 1.0149733217867418, + "grad_norm": 0.3080306649208069, + "learning_rate": 0.0004943971313312416, + "loss": 0.3329, + "step": 18166 + }, + { + "epoch": 1.0150291923904238, + "grad_norm": 0.6158595085144043, + "learning_rate": 0.0004943691169878978, + "loss": 0.3143, + "step": 18167 + }, + { + "epoch": 1.0150850629941057, + "grad_norm": 0.41898179054260254, + "learning_rate": 0.000494341102644554, + "loss": 0.5015, + "step": 18168 + }, + { + "epoch": 1.0151409335977875, + "grad_norm": 0.47712188959121704, + "learning_rate": 0.0004943130883012102, + "loss": 0.5063, + "step": 18169 + }, + { + "epoch": 1.0151968042014694, + "grad_norm": 0.5063532590866089, + "learning_rate": 0.0004942850739578664, + "loss": 0.3874, + "step": 18170 + }, + { + "epoch": 1.0152526748051514, + "grad_norm": 0.34306859970092773, + "learning_rate": 0.0004942570596145226, + "loss": 0.3348, + "step": 18171 + }, + { + "epoch": 1.015308545408833, + "grad_norm": 0.9199440479278564, + "learning_rate": 0.0004942290452711788, + "loss": 0.4476, + "step": 18172 + }, + { + "epoch": 1.015364416012515, + "grad_norm": 0.5973634123802185, + "learning_rate": 0.000494201030927835, + "loss": 0.4379, + "step": 18173 + }, + { + "epoch": 1.015420286616197, + "grad_norm": 0.533600926399231, + "learning_rate": 0.0004941730165844913, + "loss": 0.416, + "step": 18174 + }, + { + "epoch": 1.0154761572198787, + "grad_norm": 0.6180300116539001, + "learning_rate": 0.0004941450022411475, + "loss": 0.5063, + "step": 18175 + }, + { + "epoch": 1.0155320278235607, + "grad_norm": 0.47208499908447266, + "learning_rate": 0.0004941169878978037, + "loss": 0.4168, + "step": 18176 + }, + { + "epoch": 1.0155878984272424, + "grad_norm": 0.5021909475326538, + "learning_rate": 0.0004940889735544599, + "loss": 0.4508, + "step": 18177 + }, + { + "epoch": 1.0156437690309243, + "grad_norm": 0.45194703340530396, + "learning_rate": 0.0004940609592111161, + "loss": 0.4835, + "step": 18178 + }, + { + "epoch": 1.0156996396346063, + "grad_norm": 0.4708554446697235, + "learning_rate": 0.0004940329448677723, + "loss": 0.4032, + "step": 18179 + }, + { + "epoch": 1.015755510238288, + "grad_norm": 0.4081377387046814, + "learning_rate": 0.0004940049305244285, + "loss": 0.3879, + "step": 18180 + }, + { + "epoch": 1.01581138084197, + "grad_norm": 0.4356946349143982, + "learning_rate": 0.0004939769161810847, + "loss": 0.4379, + "step": 18181 + }, + { + "epoch": 1.015867251445652, + "grad_norm": 0.7051409482955933, + "learning_rate": 0.0004939489018377409, + "loss": 0.4629, + "step": 18182 + }, + { + "epoch": 1.0159231220493337, + "grad_norm": 0.44132333993911743, + "learning_rate": 0.0004939208874943971, + "loss": 0.3751, + "step": 18183 + }, + { + "epoch": 1.0159789926530156, + "grad_norm": 0.5553722977638245, + "learning_rate": 0.0004938928731510533, + "loss": 0.5263, + "step": 18184 + }, + { + "epoch": 1.0160348632566976, + "grad_norm": 0.45879659056663513, + "learning_rate": 0.0004938648588077095, + "loss": 0.4107, + "step": 18185 + }, + { + "epoch": 1.0160907338603793, + "grad_norm": 0.4241660535335541, + "learning_rate": 0.0004938368444643657, + "loss": 0.5057, + "step": 18186 + }, + { + "epoch": 1.0161466044640612, + "grad_norm": 0.2968285083770752, + "learning_rate": 0.0004938088301210219, + "loss": 0.2707, + "step": 18187 + }, + { + "epoch": 1.0162024750677432, + "grad_norm": 0.5195723176002502, + "learning_rate": 0.0004937808157776782, + "loss": 0.5162, + "step": 18188 + }, + { + "epoch": 1.016258345671425, + "grad_norm": 0.41790294647216797, + "learning_rate": 0.0004937528014343343, + "loss": 0.4772, + "step": 18189 + }, + { + "epoch": 1.0163142162751069, + "grad_norm": 2.371018648147583, + "learning_rate": 0.0004937247870909907, + "loss": 0.4274, + "step": 18190 + }, + { + "epoch": 1.0163700868787888, + "grad_norm": 0.3935621976852417, + "learning_rate": 0.0004936967727476468, + "loss": 0.47, + "step": 18191 + }, + { + "epoch": 1.0164259574824706, + "grad_norm": 0.3491103947162628, + "learning_rate": 0.0004936687584043031, + "loss": 0.4413, + "step": 18192 + }, + { + "epoch": 1.0164818280861525, + "grad_norm": 0.4500223994255066, + "learning_rate": 0.0004936407440609592, + "loss": 0.3761, + "step": 18193 + }, + { + "epoch": 1.0165376986898342, + "grad_norm": 0.7132715582847595, + "learning_rate": 0.0004936127297176155, + "loss": 0.4662, + "step": 18194 + }, + { + "epoch": 1.0165935692935162, + "grad_norm": 0.5490005016326904, + "learning_rate": 0.0004935847153742717, + "loss": 0.4519, + "step": 18195 + }, + { + "epoch": 1.0166494398971981, + "grad_norm": 0.34342965483665466, + "learning_rate": 0.0004935567010309279, + "loss": 0.3545, + "step": 18196 + }, + { + "epoch": 1.0167053105008799, + "grad_norm": 0.38159483671188354, + "learning_rate": 0.0004935286866875841, + "loss": 0.3945, + "step": 18197 + }, + { + "epoch": 1.0167611811045618, + "grad_norm": 0.6351414322853088, + "learning_rate": 0.0004935006723442403, + "loss": 0.4519, + "step": 18198 + }, + { + "epoch": 1.0168170517082438, + "grad_norm": 0.8706967234611511, + "learning_rate": 0.0004934726580008965, + "loss": 0.3955, + "step": 18199 + }, + { + "epoch": 1.0168729223119255, + "grad_norm": 1.4134197235107422, + "learning_rate": 0.0004934446436575526, + "loss": 0.4413, + "step": 18200 + }, + { + "epoch": 1.0169287929156074, + "grad_norm": 0.4088282585144043, + "learning_rate": 0.0004934166293142089, + "loss": 0.4503, + "step": 18201 + }, + { + "epoch": 1.0169846635192894, + "grad_norm": 2.4819045066833496, + "learning_rate": 0.000493388614970865, + "loss": 0.5369, + "step": 18202 + }, + { + "epoch": 1.0170405341229711, + "grad_norm": 0.48266172409057617, + "learning_rate": 0.0004933606006275213, + "loss": 0.3566, + "step": 18203 + }, + { + "epoch": 1.017096404726653, + "grad_norm": 4.306331157684326, + "learning_rate": 0.0004933325862841775, + "loss": 0.3788, + "step": 18204 + }, + { + "epoch": 1.017152275330335, + "grad_norm": 0.6325392723083496, + "learning_rate": 0.0004933045719408337, + "loss": 0.4336, + "step": 18205 + }, + { + "epoch": 1.0172081459340168, + "grad_norm": 0.4543397128582001, + "learning_rate": 0.00049327655759749, + "loss": 0.3927, + "step": 18206 + }, + { + "epoch": 1.0172640165376987, + "grad_norm": 0.6222689747810364, + "learning_rate": 0.0004932485432541462, + "loss": 0.5068, + "step": 18207 + }, + { + "epoch": 1.0173198871413807, + "grad_norm": 0.38135451078414917, + "learning_rate": 0.0004932205289108024, + "loss": 0.3804, + "step": 18208 + }, + { + "epoch": 1.0173757577450624, + "grad_norm": 0.6074116826057434, + "learning_rate": 0.0004931925145674586, + "loss": 0.3456, + "step": 18209 + }, + { + "epoch": 1.0174316283487443, + "grad_norm": 0.4221930503845215, + "learning_rate": 0.0004931645002241148, + "loss": 0.4473, + "step": 18210 + }, + { + "epoch": 1.017487498952426, + "grad_norm": 1.1766752004623413, + "learning_rate": 0.000493136485880771, + "loss": 0.44, + "step": 18211 + }, + { + "epoch": 1.017543369556108, + "grad_norm": 0.45175623893737793, + "learning_rate": 0.0004931084715374272, + "loss": 0.2789, + "step": 18212 + }, + { + "epoch": 1.01759924015979, + "grad_norm": 0.5849420428276062, + "learning_rate": 0.0004930804571940834, + "loss": 0.4322, + "step": 18213 + }, + { + "epoch": 1.0176551107634717, + "grad_norm": 0.8259017467498779, + "learning_rate": 0.0004930524428507396, + "loss": 0.7464, + "step": 18214 + }, + { + "epoch": 1.0177109813671537, + "grad_norm": 1.1637545824050903, + "learning_rate": 0.0004930244285073958, + "loss": 0.3384, + "step": 18215 + }, + { + "epoch": 1.0177668519708356, + "grad_norm": 0.6660046577453613, + "learning_rate": 0.000492996414164052, + "loss": 0.4137, + "step": 18216 + }, + { + "epoch": 1.0178227225745173, + "grad_norm": 0.5147814750671387, + "learning_rate": 0.0004929683998207082, + "loss": 0.4191, + "step": 18217 + }, + { + "epoch": 1.0178785931781993, + "grad_norm": 0.7535356879234314, + "learning_rate": 0.0004929403854773644, + "loss": 0.4057, + "step": 18218 + }, + { + "epoch": 1.0179344637818812, + "grad_norm": 1.348069190979004, + "learning_rate": 0.0004929123711340206, + "loss": 0.3404, + "step": 18219 + }, + { + "epoch": 1.017990334385563, + "grad_norm": 0.4638194441795349, + "learning_rate": 0.0004928843567906768, + "loss": 0.4938, + "step": 18220 + }, + { + "epoch": 1.018046204989245, + "grad_norm": 1.1600953340530396, + "learning_rate": 0.000492856342447333, + "loss": 0.3354, + "step": 18221 + }, + { + "epoch": 1.0181020755929269, + "grad_norm": 0.5507082939147949, + "learning_rate": 0.0004928283281039892, + "loss": 0.6498, + "step": 18222 + }, + { + "epoch": 1.0181579461966086, + "grad_norm": 0.3679966330528259, + "learning_rate": 0.0004928003137606454, + "loss": 0.3625, + "step": 18223 + }, + { + "epoch": 1.0182138168002905, + "grad_norm": 0.6273463368415833, + "learning_rate": 0.0004927722994173017, + "loss": 0.6315, + "step": 18224 + }, + { + "epoch": 1.0182696874039725, + "grad_norm": 0.3762637674808502, + "learning_rate": 0.0004927442850739579, + "loss": 0.3856, + "step": 18225 + }, + { + "epoch": 1.0183255580076542, + "grad_norm": 0.35311055183410645, + "learning_rate": 0.0004927162707306141, + "loss": 0.3908, + "step": 18226 + }, + { + "epoch": 1.0183814286113362, + "grad_norm": 0.8973369598388672, + "learning_rate": 0.0004926882563872703, + "loss": 0.408, + "step": 18227 + }, + { + "epoch": 1.018437299215018, + "grad_norm": 0.41963014006614685, + "learning_rate": 0.0004926602420439265, + "loss": 0.5304, + "step": 18228 + }, + { + "epoch": 1.0184931698186999, + "grad_norm": 0.6333062052726746, + "learning_rate": 0.0004926322277005827, + "loss": 0.4194, + "step": 18229 + }, + { + "epoch": 1.0185490404223818, + "grad_norm": 0.42156392335891724, + "learning_rate": 0.0004926042133572389, + "loss": 0.484, + "step": 18230 + }, + { + "epoch": 1.0186049110260635, + "grad_norm": 0.706224799156189, + "learning_rate": 0.0004925761990138951, + "loss": 0.4906, + "step": 18231 + }, + { + "epoch": 1.0186607816297455, + "grad_norm": 0.40232521295547485, + "learning_rate": 0.0004925481846705513, + "loss": 0.5374, + "step": 18232 + }, + { + "epoch": 1.0187166522334274, + "grad_norm": 0.4642275869846344, + "learning_rate": 0.0004925201703272075, + "loss": 0.5119, + "step": 18233 + }, + { + "epoch": 1.0187725228371092, + "grad_norm": 0.5265001058578491, + "learning_rate": 0.0004924921559838638, + "loss": 0.4123, + "step": 18234 + }, + { + "epoch": 1.0188283934407911, + "grad_norm": 0.5816701054573059, + "learning_rate": 0.0004924641416405199, + "loss": 0.4986, + "step": 18235 + }, + { + "epoch": 1.018884264044473, + "grad_norm": 5.30432653427124, + "learning_rate": 0.0004924361272971762, + "loss": 0.3643, + "step": 18236 + }, + { + "epoch": 1.0189401346481548, + "grad_norm": 0.5644680857658386, + "learning_rate": 0.0004924081129538323, + "loss": 0.5448, + "step": 18237 + }, + { + "epoch": 1.0189960052518368, + "grad_norm": 1.4965740442276, + "learning_rate": 0.0004923800986104886, + "loss": 0.3717, + "step": 18238 + }, + { + "epoch": 1.0190518758555187, + "grad_norm": 0.616021454334259, + "learning_rate": 0.0004923520842671447, + "loss": 0.4243, + "step": 18239 + }, + { + "epoch": 1.0191077464592004, + "grad_norm": 1.0997717380523682, + "learning_rate": 0.0004923240699238011, + "loss": 0.4282, + "step": 18240 + }, + { + "epoch": 1.0191636170628824, + "grad_norm": 0.591167688369751, + "learning_rate": 0.0004922960555804572, + "loss": 0.4664, + "step": 18241 + }, + { + "epoch": 1.0192194876665643, + "grad_norm": 0.7325365543365479, + "learning_rate": 0.0004922680412371135, + "loss": 0.4607, + "step": 18242 + }, + { + "epoch": 1.019275358270246, + "grad_norm": 0.5806012749671936, + "learning_rate": 0.0004922400268937697, + "loss": 0.4068, + "step": 18243 + }, + { + "epoch": 1.019331228873928, + "grad_norm": 0.5509853959083557, + "learning_rate": 0.0004922120125504258, + "loss": 0.5328, + "step": 18244 + }, + { + "epoch": 1.0193870994776097, + "grad_norm": 0.426459938287735, + "learning_rate": 0.0004921839982070821, + "loss": 0.4196, + "step": 18245 + }, + { + "epoch": 1.0194429700812917, + "grad_norm": 0.4595513939857483, + "learning_rate": 0.0004921559838637382, + "loss": 0.3801, + "step": 18246 + }, + { + "epoch": 1.0194988406849737, + "grad_norm": 0.3568066358566284, + "learning_rate": 0.0004921279695203945, + "loss": 0.4625, + "step": 18247 + }, + { + "epoch": 1.0195547112886554, + "grad_norm": 0.4676617980003357, + "learning_rate": 0.0004920999551770506, + "loss": 0.5683, + "step": 18248 + }, + { + "epoch": 1.0196105818923373, + "grad_norm": 0.44724899530410767, + "learning_rate": 0.0004920719408337069, + "loss": 0.3935, + "step": 18249 + }, + { + "epoch": 1.0196664524960193, + "grad_norm": 0.6228687167167664, + "learning_rate": 0.0004920439264903631, + "loss": 0.3839, + "step": 18250 + }, + { + "epoch": 1.019722323099701, + "grad_norm": 2.595865488052368, + "learning_rate": 0.0004920159121470193, + "loss": 0.4376, + "step": 18251 + }, + { + "epoch": 1.019778193703383, + "grad_norm": 0.5934461355209351, + "learning_rate": 0.0004919878978036755, + "loss": 0.3757, + "step": 18252 + }, + { + "epoch": 1.019834064307065, + "grad_norm": 0.5996135473251343, + "learning_rate": 0.0004919598834603317, + "loss": 0.4205, + "step": 18253 + }, + { + "epoch": 1.0198899349107466, + "grad_norm": 0.5517286658287048, + "learning_rate": 0.0004919318691169879, + "loss": 0.4585, + "step": 18254 + }, + { + "epoch": 1.0199458055144286, + "grad_norm": 0.3975768983364105, + "learning_rate": 0.0004919038547736441, + "loss": 0.3115, + "step": 18255 + }, + { + "epoch": 1.0200016761181105, + "grad_norm": 1.5961668491363525, + "learning_rate": 0.0004918758404303003, + "loss": 0.4765, + "step": 18256 + }, + { + "epoch": 1.0200575467217923, + "grad_norm": 0.6106342077255249, + "learning_rate": 0.0004918478260869566, + "loss": 0.4843, + "step": 18257 + }, + { + "epoch": 1.0201134173254742, + "grad_norm": 0.4710228443145752, + "learning_rate": 0.0004918198117436128, + "loss": 0.4349, + "step": 18258 + }, + { + "epoch": 1.0201692879291562, + "grad_norm": 0.588970959186554, + "learning_rate": 0.000491791797400269, + "loss": 0.4839, + "step": 18259 + }, + { + "epoch": 1.020225158532838, + "grad_norm": 0.6235185265541077, + "learning_rate": 0.0004917637830569252, + "loss": 0.4974, + "step": 18260 + }, + { + "epoch": 1.0202810291365199, + "grad_norm": 0.29602956771850586, + "learning_rate": 0.0004917357687135814, + "loss": 0.3093, + "step": 18261 + }, + { + "epoch": 1.0203368997402016, + "grad_norm": 0.42876315116882324, + "learning_rate": 0.0004917077543702376, + "loss": 0.3512, + "step": 18262 + }, + { + "epoch": 1.0203927703438835, + "grad_norm": 0.38048410415649414, + "learning_rate": 0.0004916797400268938, + "loss": 0.3213, + "step": 18263 + }, + { + "epoch": 1.0204486409475655, + "grad_norm": 2.133364200592041, + "learning_rate": 0.00049165172568355, + "loss": 0.3528, + "step": 18264 + }, + { + "epoch": 1.0205045115512472, + "grad_norm": 0.3566247820854187, + "learning_rate": 0.0004916237113402062, + "loss": 0.4234, + "step": 18265 + }, + { + "epoch": 1.0205603821549292, + "grad_norm": 0.808238685131073, + "learning_rate": 0.0004915956969968624, + "loss": 0.3186, + "step": 18266 + }, + { + "epoch": 1.0206162527586111, + "grad_norm": 0.8776907920837402, + "learning_rate": 0.0004915676826535186, + "loss": 0.4008, + "step": 18267 + }, + { + "epoch": 1.0206721233622928, + "grad_norm": 0.3378530740737915, + "learning_rate": 0.0004915396683101748, + "loss": 0.3935, + "step": 18268 + }, + { + "epoch": 1.0207279939659748, + "grad_norm": 0.3892776072025299, + "learning_rate": 0.000491511653966831, + "loss": 0.48, + "step": 18269 + }, + { + "epoch": 1.0207838645696568, + "grad_norm": 0.5381297469139099, + "learning_rate": 0.0004914836396234872, + "loss": 0.343, + "step": 18270 + }, + { + "epoch": 1.0208397351733385, + "grad_norm": 0.5764870047569275, + "learning_rate": 0.0004914556252801434, + "loss": 0.5472, + "step": 18271 + }, + { + "epoch": 1.0208956057770204, + "grad_norm": 0.3692784607410431, + "learning_rate": 0.0004914276109367996, + "loss": 0.4869, + "step": 18272 + }, + { + "epoch": 1.0209514763807024, + "grad_norm": 0.2791777551174164, + "learning_rate": 0.0004913995965934558, + "loss": 0.2937, + "step": 18273 + }, + { + "epoch": 1.0210073469843841, + "grad_norm": 0.4949378967285156, + "learning_rate": 0.000491371582250112, + "loss": 0.515, + "step": 18274 + }, + { + "epoch": 1.021063217588066, + "grad_norm": 0.37838655710220337, + "learning_rate": 0.0004913435679067683, + "loss": 0.3887, + "step": 18275 + }, + { + "epoch": 1.021119088191748, + "grad_norm": 0.5450097322463989, + "learning_rate": 0.0004913155535634245, + "loss": 0.517, + "step": 18276 + }, + { + "epoch": 1.0211749587954297, + "grad_norm": 0.41184622049331665, + "learning_rate": 0.0004912875392200807, + "loss": 0.4447, + "step": 18277 + }, + { + "epoch": 1.0212308293991117, + "grad_norm": 0.376992404460907, + "learning_rate": 0.0004912595248767369, + "loss": 0.3221, + "step": 18278 + }, + { + "epoch": 1.0212867000027934, + "grad_norm": 0.6617192625999451, + "learning_rate": 0.0004912315105333931, + "loss": 0.6047, + "step": 18279 + }, + { + "epoch": 1.0213425706064754, + "grad_norm": 0.669956624507904, + "learning_rate": 0.0004912034961900493, + "loss": 0.4973, + "step": 18280 + }, + { + "epoch": 1.0213984412101573, + "grad_norm": 0.477338582277298, + "learning_rate": 0.0004911754818467055, + "loss": 0.4975, + "step": 18281 + }, + { + "epoch": 1.021454311813839, + "grad_norm": 0.4232087731361389, + "learning_rate": 0.0004911474675033618, + "loss": 0.3475, + "step": 18282 + }, + { + "epoch": 1.021510182417521, + "grad_norm": 0.7397297620773315, + "learning_rate": 0.0004911194531600179, + "loss": 0.6633, + "step": 18283 + }, + { + "epoch": 1.021566053021203, + "grad_norm": 0.4600067436695099, + "learning_rate": 0.0004910914388166742, + "loss": 0.3613, + "step": 18284 + }, + { + "epoch": 1.0216219236248847, + "grad_norm": 0.357884019613266, + "learning_rate": 0.0004910634244733303, + "loss": 0.3965, + "step": 18285 + }, + { + "epoch": 1.0216777942285666, + "grad_norm": 1.0988987684249878, + "learning_rate": 0.0004910354101299865, + "loss": 0.4953, + "step": 18286 + }, + { + "epoch": 1.0217336648322486, + "grad_norm": 0.7013881802558899, + "learning_rate": 0.0004910073957866427, + "loss": 0.5476, + "step": 18287 + }, + { + "epoch": 1.0217895354359303, + "grad_norm": 0.5616387724876404, + "learning_rate": 0.0004909793814432989, + "loss": 0.4069, + "step": 18288 + }, + { + "epoch": 1.0218454060396123, + "grad_norm": 0.41604843735694885, + "learning_rate": 0.0004909513670999551, + "loss": 0.4862, + "step": 18289 + }, + { + "epoch": 1.0219012766432942, + "grad_norm": 0.43115371465682983, + "learning_rate": 0.0004909233527566113, + "loss": 0.4248, + "step": 18290 + }, + { + "epoch": 1.021957147246976, + "grad_norm": 0.8432316780090332, + "learning_rate": 0.0004908953384132677, + "loss": 0.429, + "step": 18291 + }, + { + "epoch": 1.022013017850658, + "grad_norm": 0.44093409180641174, + "learning_rate": 0.0004908673240699238, + "loss": 0.3502, + "step": 18292 + }, + { + "epoch": 1.0220688884543399, + "grad_norm": 1.8342872858047485, + "learning_rate": 0.0004908393097265801, + "loss": 0.4324, + "step": 18293 + }, + { + "epoch": 1.0221247590580216, + "grad_norm": 0.3906678855419159, + "learning_rate": 0.0004908112953832362, + "loss": 0.3765, + "step": 18294 + }, + { + "epoch": 1.0221806296617035, + "grad_norm": 0.3826470375061035, + "learning_rate": 0.0004907832810398925, + "loss": 0.4163, + "step": 18295 + }, + { + "epoch": 1.0222365002653853, + "grad_norm": 2.1980531215667725, + "learning_rate": 0.0004907552666965486, + "loss": 0.6116, + "step": 18296 + }, + { + "epoch": 1.0222923708690672, + "grad_norm": 11.090136528015137, + "learning_rate": 0.0004907272523532049, + "loss": 0.429, + "step": 18297 + }, + { + "epoch": 1.0223482414727492, + "grad_norm": 0.44439297914505005, + "learning_rate": 0.0004906992380098611, + "loss": 0.3976, + "step": 18298 + }, + { + "epoch": 1.022404112076431, + "grad_norm": 0.4758705794811249, + "learning_rate": 0.0004906712236665173, + "loss": 0.3471, + "step": 18299 + }, + { + "epoch": 1.0224599826801128, + "grad_norm": 0.4482257664203644, + "learning_rate": 0.0004906432093231735, + "loss": 0.4816, + "step": 18300 + }, + { + "epoch": 1.0225158532837948, + "grad_norm": 0.42189106345176697, + "learning_rate": 0.0004906151949798297, + "loss": 0.4224, + "step": 18301 + }, + { + "epoch": 1.0225717238874765, + "grad_norm": 0.8036267757415771, + "learning_rate": 0.0004905871806364859, + "loss": 0.487, + "step": 18302 + }, + { + "epoch": 1.0226275944911585, + "grad_norm": 0.4151196777820587, + "learning_rate": 0.0004905591662931421, + "loss": 0.4171, + "step": 18303 + }, + { + "epoch": 1.0226834650948404, + "grad_norm": 0.6541417241096497, + "learning_rate": 0.0004905311519497983, + "loss": 0.548, + "step": 18304 + }, + { + "epoch": 1.0227393356985222, + "grad_norm": 0.32593655586242676, + "learning_rate": 0.0004905031376064545, + "loss": 0.4832, + "step": 18305 + }, + { + "epoch": 1.022795206302204, + "grad_norm": 0.4568961560726166, + "learning_rate": 0.0004904751232631107, + "loss": 0.3999, + "step": 18306 + }, + { + "epoch": 1.022851076905886, + "grad_norm": 0.33737844228744507, + "learning_rate": 0.000490447108919767, + "loss": 0.4514, + "step": 18307 + }, + { + "epoch": 1.0229069475095678, + "grad_norm": 0.7564979791641235, + "learning_rate": 0.0004904190945764232, + "loss": 0.4386, + "step": 18308 + }, + { + "epoch": 1.0229628181132497, + "grad_norm": 0.4049661457538605, + "learning_rate": 0.0004903910802330794, + "loss": 0.4168, + "step": 18309 + }, + { + "epoch": 1.0230186887169317, + "grad_norm": 0.43395718932151794, + "learning_rate": 0.0004903630658897356, + "loss": 0.4894, + "step": 18310 + }, + { + "epoch": 1.0230745593206134, + "grad_norm": 0.5876695513725281, + "learning_rate": 0.0004903350515463918, + "loss": 0.5604, + "step": 18311 + }, + { + "epoch": 1.0231304299242954, + "grad_norm": 0.28659066557884216, + "learning_rate": 0.000490307037203048, + "loss": 0.3187, + "step": 18312 + }, + { + "epoch": 1.023186300527977, + "grad_norm": 0.3727923631668091, + "learning_rate": 0.0004902790228597042, + "loss": 0.3648, + "step": 18313 + }, + { + "epoch": 1.023242171131659, + "grad_norm": 0.6666594743728638, + "learning_rate": 0.0004902510085163604, + "loss": 0.5597, + "step": 18314 + }, + { + "epoch": 1.023298041735341, + "grad_norm": 0.4236935079097748, + "learning_rate": 0.0004902229941730166, + "loss": 0.6346, + "step": 18315 + }, + { + "epoch": 1.0233539123390227, + "grad_norm": 0.42554202675819397, + "learning_rate": 0.0004901949798296728, + "loss": 0.4573, + "step": 18316 + }, + { + "epoch": 1.0234097829427047, + "grad_norm": 0.5459070205688477, + "learning_rate": 0.000490166965486329, + "loss": 0.3974, + "step": 18317 + }, + { + "epoch": 1.0234656535463866, + "grad_norm": 0.5867787003517151, + "learning_rate": 0.0004901389511429852, + "loss": 0.479, + "step": 18318 + }, + { + "epoch": 1.0235215241500684, + "grad_norm": 0.5596860647201538, + "learning_rate": 0.0004901109367996414, + "loss": 0.4866, + "step": 18319 + }, + { + "epoch": 1.0235773947537503, + "grad_norm": 0.2795903980731964, + "learning_rate": 0.0004900829224562976, + "loss": 0.3508, + "step": 18320 + }, + { + "epoch": 1.0236332653574323, + "grad_norm": 0.5083131194114685, + "learning_rate": 0.0004900549081129538, + "loss": 0.5455, + "step": 18321 + }, + { + "epoch": 1.023689135961114, + "grad_norm": 0.38655564188957214, + "learning_rate": 0.00049002689376961, + "loss": 0.4259, + "step": 18322 + }, + { + "epoch": 1.023745006564796, + "grad_norm": 0.29039105772972107, + "learning_rate": 0.0004899988794262662, + "loss": 0.3402, + "step": 18323 + }, + { + "epoch": 1.023800877168478, + "grad_norm": 0.38251379132270813, + "learning_rate": 0.0004899708650829225, + "loss": 0.4157, + "step": 18324 + }, + { + "epoch": 1.0238567477721596, + "grad_norm": 0.40956568717956543, + "learning_rate": 0.0004899428507395787, + "loss": 0.3176, + "step": 18325 + }, + { + "epoch": 1.0239126183758416, + "grad_norm": 0.8397138714790344, + "learning_rate": 0.0004899148363962349, + "loss": 0.4472, + "step": 18326 + }, + { + "epoch": 1.0239684889795235, + "grad_norm": 0.6228715181350708, + "learning_rate": 0.0004898868220528911, + "loss": 0.5305, + "step": 18327 + }, + { + "epoch": 1.0240243595832053, + "grad_norm": 0.3444176912307739, + "learning_rate": 0.0004898588077095473, + "loss": 0.4142, + "step": 18328 + }, + { + "epoch": 1.0240802301868872, + "grad_norm": 0.500365674495697, + "learning_rate": 0.0004898307933662035, + "loss": 0.6432, + "step": 18329 + }, + { + "epoch": 1.024136100790569, + "grad_norm": 0.613463282585144, + "learning_rate": 0.0004898027790228597, + "loss": 0.5128, + "step": 18330 + }, + { + "epoch": 1.0241919713942509, + "grad_norm": 1.0016050338745117, + "learning_rate": 0.0004897747646795159, + "loss": 0.4694, + "step": 18331 + }, + { + "epoch": 1.0242478419979328, + "grad_norm": 0.5248172879219055, + "learning_rate": 0.0004897467503361721, + "loss": 0.3163, + "step": 18332 + }, + { + "epoch": 1.0243037126016146, + "grad_norm": 0.4389728307723999, + "learning_rate": 0.0004897187359928283, + "loss": 0.4247, + "step": 18333 + }, + { + "epoch": 1.0243595832052965, + "grad_norm": 0.42015162110328674, + "learning_rate": 0.0004896907216494845, + "loss": 0.3809, + "step": 18334 + }, + { + "epoch": 1.0244154538089785, + "grad_norm": 0.3805907368659973, + "learning_rate": 0.0004896627073061407, + "loss": 0.345, + "step": 18335 + }, + { + "epoch": 1.0244713244126602, + "grad_norm": 0.31370142102241516, + "learning_rate": 0.0004896346929627969, + "loss": 0.4131, + "step": 18336 + }, + { + "epoch": 1.0245271950163422, + "grad_norm": 0.43095844984054565, + "learning_rate": 0.0004896066786194532, + "loss": 0.4621, + "step": 18337 + }, + { + "epoch": 1.024583065620024, + "grad_norm": 0.45415449142456055, + "learning_rate": 0.0004895786642761093, + "loss": 0.4493, + "step": 18338 + }, + { + "epoch": 1.0246389362237058, + "grad_norm": 0.5715726613998413, + "learning_rate": 0.0004895506499327656, + "loss": 0.3901, + "step": 18339 + }, + { + "epoch": 1.0246948068273878, + "grad_norm": 0.3637644052505493, + "learning_rate": 0.0004895226355894217, + "loss": 0.4117, + "step": 18340 + }, + { + "epoch": 1.0247506774310697, + "grad_norm": 0.5649706721305847, + "learning_rate": 0.0004894946212460781, + "loss": 0.4459, + "step": 18341 + }, + { + "epoch": 1.0248065480347515, + "grad_norm": 0.3247135281562805, + "learning_rate": 0.0004894666069027342, + "loss": 0.37, + "step": 18342 + }, + { + "epoch": 1.0248624186384334, + "grad_norm": 0.4286007583141327, + "learning_rate": 0.0004894385925593905, + "loss": 0.4258, + "step": 18343 + }, + { + "epoch": 1.0249182892421154, + "grad_norm": 0.5528742671012878, + "learning_rate": 0.0004894105782160466, + "loss": 0.4479, + "step": 18344 + }, + { + "epoch": 1.024974159845797, + "grad_norm": 0.4469507038593292, + "learning_rate": 0.0004893825638727029, + "loss": 0.4647, + "step": 18345 + }, + { + "epoch": 1.025030030449479, + "grad_norm": 0.8789118528366089, + "learning_rate": 0.0004893545495293591, + "loss": 0.6421, + "step": 18346 + }, + { + "epoch": 1.0250859010531608, + "grad_norm": 0.3958839178085327, + "learning_rate": 0.0004893265351860153, + "loss": 0.4252, + "step": 18347 + }, + { + "epoch": 1.0251417716568427, + "grad_norm": 0.3755641579627991, + "learning_rate": 0.0004892985208426715, + "loss": 0.3353, + "step": 18348 + }, + { + "epoch": 1.0251976422605247, + "grad_norm": 0.4315613806247711, + "learning_rate": 0.0004892705064993276, + "loss": 0.4509, + "step": 18349 + }, + { + "epoch": 1.0252535128642064, + "grad_norm": 1.1365398168563843, + "learning_rate": 0.0004892424921559839, + "loss": 0.371, + "step": 18350 + }, + { + "epoch": 1.0253093834678884, + "grad_norm": 0.5229973793029785, + "learning_rate": 0.00048921447781264, + "loss": 0.3644, + "step": 18351 + }, + { + "epoch": 1.0253652540715703, + "grad_norm": 0.54414963722229, + "learning_rate": 0.0004891864634692963, + "loss": 0.4149, + "step": 18352 + }, + { + "epoch": 1.025421124675252, + "grad_norm": 1.187511920928955, + "learning_rate": 0.0004891584491259525, + "loss": 0.4351, + "step": 18353 + }, + { + "epoch": 1.025476995278934, + "grad_norm": 1.0698070526123047, + "learning_rate": 0.0004891304347826087, + "loss": 0.3158, + "step": 18354 + }, + { + "epoch": 1.025532865882616, + "grad_norm": 0.6194050312042236, + "learning_rate": 0.0004891024204392649, + "loss": 0.4016, + "step": 18355 + }, + { + "epoch": 1.0255887364862977, + "grad_norm": 0.3899744153022766, + "learning_rate": 0.0004890744060959211, + "loss": 0.3315, + "step": 18356 + }, + { + "epoch": 1.0256446070899796, + "grad_norm": 1.5306254625320435, + "learning_rate": 0.0004890463917525774, + "loss": 0.4216, + "step": 18357 + }, + { + "epoch": 1.0257004776936616, + "grad_norm": 0.3574649691581726, + "learning_rate": 0.0004890183774092336, + "loss": 0.3695, + "step": 18358 + }, + { + "epoch": 1.0257563482973433, + "grad_norm": 0.5308761596679688, + "learning_rate": 0.0004889903630658898, + "loss": 0.3704, + "step": 18359 + }, + { + "epoch": 1.0258122189010253, + "grad_norm": 0.863679826259613, + "learning_rate": 0.000488962348722546, + "loss": 0.5103, + "step": 18360 + }, + { + "epoch": 1.0258680895047072, + "grad_norm": 0.5394409894943237, + "learning_rate": 0.0004889343343792022, + "loss": 0.4194, + "step": 18361 + }, + { + "epoch": 1.025923960108389, + "grad_norm": 0.8748247027397156, + "learning_rate": 0.0004889063200358584, + "loss": 0.4753, + "step": 18362 + }, + { + "epoch": 1.0259798307120709, + "grad_norm": 0.3061869144439697, + "learning_rate": 0.0004888783056925146, + "loss": 0.3969, + "step": 18363 + }, + { + "epoch": 1.0260357013157526, + "grad_norm": 0.2986275851726532, + "learning_rate": 0.0004888502913491708, + "loss": 0.3456, + "step": 18364 + }, + { + "epoch": 1.0260915719194346, + "grad_norm": 2.6051909923553467, + "learning_rate": 0.000488822277005827, + "loss": 0.4215, + "step": 18365 + }, + { + "epoch": 1.0261474425231165, + "grad_norm": 0.5039915442466736, + "learning_rate": 0.0004887942626624832, + "loss": 0.4941, + "step": 18366 + }, + { + "epoch": 1.0262033131267982, + "grad_norm": 0.554992139339447, + "learning_rate": 0.0004887662483191394, + "loss": 0.4957, + "step": 18367 + }, + { + "epoch": 1.0262591837304802, + "grad_norm": 0.5127183794975281, + "learning_rate": 0.0004887382339757956, + "loss": 0.4083, + "step": 18368 + }, + { + "epoch": 1.0263150543341621, + "grad_norm": 0.6963492631912231, + "learning_rate": 0.0004887102196324518, + "loss": 0.3814, + "step": 18369 + }, + { + "epoch": 1.0263709249378439, + "grad_norm": 0.608310878276825, + "learning_rate": 0.000488682205289108, + "loss": 0.2788, + "step": 18370 + }, + { + "epoch": 1.0264267955415258, + "grad_norm": 0.9187970757484436, + "learning_rate": 0.0004886541909457642, + "loss": 0.4221, + "step": 18371 + }, + { + "epoch": 1.0264826661452078, + "grad_norm": 0.4683224856853485, + "learning_rate": 0.0004886261766024204, + "loss": 0.386, + "step": 18372 + }, + { + "epoch": 1.0265385367488895, + "grad_norm": 0.6152942180633545, + "learning_rate": 0.0004885981622590766, + "loss": 0.5359, + "step": 18373 + }, + { + "epoch": 1.0265944073525715, + "grad_norm": 0.5212556719779968, + "learning_rate": 0.0004885701479157329, + "loss": 0.4181, + "step": 18374 + }, + { + "epoch": 1.0266502779562534, + "grad_norm": 0.40746310353279114, + "learning_rate": 0.0004885421335723891, + "loss": 0.515, + "step": 18375 + }, + { + "epoch": 1.0267061485599351, + "grad_norm": 0.5200693607330322, + "learning_rate": 0.0004885141192290453, + "loss": 0.3426, + "step": 18376 + }, + { + "epoch": 1.026762019163617, + "grad_norm": 0.5146017670631409, + "learning_rate": 0.0004884861048857015, + "loss": 0.5106, + "step": 18377 + }, + { + "epoch": 1.026817889767299, + "grad_norm": 0.72695392370224, + "learning_rate": 0.0004884580905423577, + "loss": 0.493, + "step": 18378 + }, + { + "epoch": 1.0268737603709808, + "grad_norm": 0.37709927558898926, + "learning_rate": 0.0004884300761990139, + "loss": 0.4189, + "step": 18379 + }, + { + "epoch": 1.0269296309746627, + "grad_norm": 1.0177757740020752, + "learning_rate": 0.0004884020618556701, + "loss": 0.6137, + "step": 18380 + }, + { + "epoch": 1.0269855015783445, + "grad_norm": 0.4366978406906128, + "learning_rate": 0.0004883740475123263, + "loss": 0.5306, + "step": 18381 + }, + { + "epoch": 1.0270413721820264, + "grad_norm": 1.1087615489959717, + "learning_rate": 0.0004883460331689825, + "loss": 0.3248, + "step": 18382 + }, + { + "epoch": 1.0270972427857084, + "grad_norm": 1.1759092807769775, + "learning_rate": 0.0004883180188256387, + "loss": 0.4248, + "step": 18383 + }, + { + "epoch": 1.02715311338939, + "grad_norm": 0.7866088151931763, + "learning_rate": 0.0004882900044822949, + "loss": 0.5092, + "step": 18384 + }, + { + "epoch": 1.027208983993072, + "grad_norm": 0.549919068813324, + "learning_rate": 0.00048826199013895117, + "loss": 0.4105, + "step": 18385 + }, + { + "epoch": 1.027264854596754, + "grad_norm": 0.5900113582611084, + "learning_rate": 0.0004882339757956074, + "loss": 0.4851, + "step": 18386 + }, + { + "epoch": 1.0273207252004357, + "grad_norm": 1.6612813472747803, + "learning_rate": 0.0004882059614522636, + "loss": 0.3913, + "step": 18387 + }, + { + "epoch": 1.0273765958041177, + "grad_norm": 0.4552616477012634, + "learning_rate": 0.0004881779471089198, + "loss": 0.5505, + "step": 18388 + }, + { + "epoch": 1.0274324664077996, + "grad_norm": 0.42529451847076416, + "learning_rate": 0.000488149932765576, + "loss": 0.4753, + "step": 18389 + }, + { + "epoch": 1.0274883370114813, + "grad_norm": 0.6590661406517029, + "learning_rate": 0.0004881219184222322, + "loss": 0.587, + "step": 18390 + }, + { + "epoch": 1.0275442076151633, + "grad_norm": 0.692702054977417, + "learning_rate": 0.00048809390407888835, + "loss": 0.5287, + "step": 18391 + }, + { + "epoch": 1.0276000782188452, + "grad_norm": 0.48534977436065674, + "learning_rate": 0.0004880658897355446, + "loss": 0.3914, + "step": 18392 + }, + { + "epoch": 1.027655948822527, + "grad_norm": 0.5435926914215088, + "learning_rate": 0.0004880378753922008, + "loss": 0.4747, + "step": 18393 + }, + { + "epoch": 1.027711819426209, + "grad_norm": 0.3865984082221985, + "learning_rate": 0.000488009861048857, + "loss": 0.3956, + "step": 18394 + }, + { + "epoch": 1.0277676900298909, + "grad_norm": 0.3836868107318878, + "learning_rate": 0.0004879818467055132, + "loss": 0.3733, + "step": 18395 + }, + { + "epoch": 1.0278235606335726, + "grad_norm": 0.7251520156860352, + "learning_rate": 0.00048795383236216943, + "loss": 0.285, + "step": 18396 + }, + { + "epoch": 1.0278794312372546, + "grad_norm": 0.3656884431838989, + "learning_rate": 0.00048792581801882564, + "loss": 0.3788, + "step": 18397 + }, + { + "epoch": 1.0279353018409363, + "grad_norm": 0.353730708360672, + "learning_rate": 0.00048789780367548184, + "loss": 0.481, + "step": 18398 + }, + { + "epoch": 1.0279911724446182, + "grad_norm": 0.5185216665267944, + "learning_rate": 0.00048786978933213805, + "loss": 0.4147, + "step": 18399 + }, + { + "epoch": 1.0280470430483002, + "grad_norm": 5.544842720031738, + "learning_rate": 0.0004878417749887943, + "loss": 0.3774, + "step": 18400 + }, + { + "epoch": 1.028102913651982, + "grad_norm": 0.42562031745910645, + "learning_rate": 0.00048781376064545046, + "loss": 0.4413, + "step": 18401 + }, + { + "epoch": 1.0281587842556639, + "grad_norm": 0.35910913348197937, + "learning_rate": 0.0004877857463021067, + "loss": 0.4202, + "step": 18402 + }, + { + "epoch": 1.0282146548593458, + "grad_norm": 0.5164417028427124, + "learning_rate": 0.0004877577319587629, + "loss": 0.4492, + "step": 18403 + }, + { + "epoch": 1.0282705254630276, + "grad_norm": 0.3427884876728058, + "learning_rate": 0.00048772971761541913, + "loss": 0.4005, + "step": 18404 + }, + { + "epoch": 1.0283263960667095, + "grad_norm": 0.3677365779876709, + "learning_rate": 0.0004877017032720753, + "loss": 0.3819, + "step": 18405 + }, + { + "epoch": 1.0283822666703915, + "grad_norm": 0.39581888914108276, + "learning_rate": 0.00048767368892873154, + "loss": 0.3651, + "step": 18406 + }, + { + "epoch": 1.0284381372740732, + "grad_norm": 1.2201309204101562, + "learning_rate": 0.0004876456745853877, + "loss": 0.3408, + "step": 18407 + }, + { + "epoch": 1.0284940078777551, + "grad_norm": 0.4363892674446106, + "learning_rate": 0.00048761766024204396, + "loss": 0.3922, + "step": 18408 + }, + { + "epoch": 1.028549878481437, + "grad_norm": 0.6448490023612976, + "learning_rate": 0.00048758964589870016, + "loss": 0.3568, + "step": 18409 + }, + { + "epoch": 1.0286057490851188, + "grad_norm": 0.6012989282608032, + "learning_rate": 0.00048756163155535637, + "loss": 0.4649, + "step": 18410 + }, + { + "epoch": 1.0286616196888008, + "grad_norm": 0.5879793763160706, + "learning_rate": 0.0004875336172120126, + "loss": 0.3641, + "step": 18411 + }, + { + "epoch": 1.0287174902924827, + "grad_norm": 0.6036038398742676, + "learning_rate": 0.0004875056028686688, + "loss": 0.5648, + "step": 18412 + }, + { + "epoch": 1.0287733608961644, + "grad_norm": 0.36820700764656067, + "learning_rate": 0.000487477588525325, + "loss": 0.3684, + "step": 18413 + }, + { + "epoch": 1.0288292314998464, + "grad_norm": 0.3888430595397949, + "learning_rate": 0.00048744957418198114, + "loss": 0.3626, + "step": 18414 + }, + { + "epoch": 1.0288851021035281, + "grad_norm": 0.6240828633308411, + "learning_rate": 0.0004874215598386374, + "loss": 0.6012, + "step": 18415 + }, + { + "epoch": 1.02894097270721, + "grad_norm": 0.4409683644771576, + "learning_rate": 0.0004873935454952936, + "loss": 0.5558, + "step": 18416 + }, + { + "epoch": 1.028996843310892, + "grad_norm": 0.7500364780426025, + "learning_rate": 0.0004873655311519498, + "loss": 0.4663, + "step": 18417 + }, + { + "epoch": 1.0290527139145738, + "grad_norm": 0.4643923342227936, + "learning_rate": 0.000487337516808606, + "loss": 0.5235, + "step": 18418 + }, + { + "epoch": 1.0291085845182557, + "grad_norm": 0.4388252794742584, + "learning_rate": 0.0004873095024652622, + "loss": 0.5446, + "step": 18419 + }, + { + "epoch": 1.0291644551219377, + "grad_norm": 0.5699554085731506, + "learning_rate": 0.0004872814881219184, + "loss": 0.3983, + "step": 18420 + }, + { + "epoch": 1.0292203257256194, + "grad_norm": 0.3548429310321808, + "learning_rate": 0.00048725347377857463, + "loss": 0.2985, + "step": 18421 + }, + { + "epoch": 1.0292761963293013, + "grad_norm": 0.36227333545684814, + "learning_rate": 0.00048722545943523084, + "loss": 0.4474, + "step": 18422 + }, + { + "epoch": 1.0293320669329833, + "grad_norm": 0.9807329177856445, + "learning_rate": 0.00048719744509188704, + "loss": 0.4002, + "step": 18423 + }, + { + "epoch": 1.029387937536665, + "grad_norm": 0.3339606821537018, + "learning_rate": 0.00048716943074854325, + "loss": 0.4257, + "step": 18424 + }, + { + "epoch": 1.029443808140347, + "grad_norm": 0.5509966015815735, + "learning_rate": 0.0004871414164051995, + "loss": 0.4399, + "step": 18425 + }, + { + "epoch": 1.029499678744029, + "grad_norm": 0.4333062171936035, + "learning_rate": 0.00048711340206185566, + "loss": 0.5137, + "step": 18426 + }, + { + "epoch": 1.0295555493477107, + "grad_norm": 0.3742404878139496, + "learning_rate": 0.0004870853877185119, + "loss": 0.363, + "step": 18427 + }, + { + "epoch": 1.0296114199513926, + "grad_norm": 0.39711794257164, + "learning_rate": 0.00048705737337516807, + "loss": 0.5093, + "step": 18428 + }, + { + "epoch": 1.0296672905550746, + "grad_norm": 0.4005369246006012, + "learning_rate": 0.00048702935903182433, + "loss": 0.334, + "step": 18429 + }, + { + "epoch": 1.0297231611587563, + "grad_norm": 0.5211586952209473, + "learning_rate": 0.0004870013446884805, + "loss": 0.4131, + "step": 18430 + }, + { + "epoch": 1.0297790317624382, + "grad_norm": 0.7121951580047607, + "learning_rate": 0.00048697333034513674, + "loss": 0.3272, + "step": 18431 + }, + { + "epoch": 1.02983490236612, + "grad_norm": 0.48137328028678894, + "learning_rate": 0.00048694531600179295, + "loss": 0.4919, + "step": 18432 + }, + { + "epoch": 1.029890772969802, + "grad_norm": 0.5777752995491028, + "learning_rate": 0.00048691730165844916, + "loss": 0.4256, + "step": 18433 + }, + { + "epoch": 1.0299466435734839, + "grad_norm": 0.41728147864341736, + "learning_rate": 0.00048688928731510536, + "loss": 0.4116, + "step": 18434 + }, + { + "epoch": 1.0300025141771656, + "grad_norm": 0.3200973868370056, + "learning_rate": 0.0004868612729717615, + "loss": 0.3425, + "step": 18435 + }, + { + "epoch": 1.0300583847808475, + "grad_norm": 0.45763182640075684, + "learning_rate": 0.0004868332586284178, + "loss": 0.3041, + "step": 18436 + }, + { + "epoch": 1.0301142553845295, + "grad_norm": 0.4730863869190216, + "learning_rate": 0.0004868052442850739, + "loss": 0.5047, + "step": 18437 + }, + { + "epoch": 1.0301701259882112, + "grad_norm": 0.449588418006897, + "learning_rate": 0.0004867772299417302, + "loss": 0.433, + "step": 18438 + }, + { + "epoch": 1.0302259965918932, + "grad_norm": 0.466396689414978, + "learning_rate": 0.00048674921559838634, + "loss": 0.4402, + "step": 18439 + }, + { + "epoch": 1.0302818671955751, + "grad_norm": 1.2438244819641113, + "learning_rate": 0.0004867212012550426, + "loss": 0.5292, + "step": 18440 + }, + { + "epoch": 1.0303377377992569, + "grad_norm": 0.4514647126197815, + "learning_rate": 0.0004866931869116988, + "loss": 0.421, + "step": 18441 + }, + { + "epoch": 1.0303936084029388, + "grad_norm": 0.33642587065696716, + "learning_rate": 0.000486665172568355, + "loss": 0.3845, + "step": 18442 + }, + { + "epoch": 1.0304494790066208, + "grad_norm": 0.5011821389198303, + "learning_rate": 0.0004866371582250112, + "loss": 0.3879, + "step": 18443 + }, + { + "epoch": 1.0305053496103025, + "grad_norm": 0.44357559084892273, + "learning_rate": 0.0004866091438816674, + "loss": 0.4332, + "step": 18444 + }, + { + "epoch": 1.0305612202139844, + "grad_norm": 0.4807109832763672, + "learning_rate": 0.0004865811295383236, + "loss": 0.4705, + "step": 18445 + }, + { + "epoch": 1.0306170908176664, + "grad_norm": 0.3525863289833069, + "learning_rate": 0.00048655311519497983, + "loss": 0.3788, + "step": 18446 + }, + { + "epoch": 1.0306729614213481, + "grad_norm": 0.9526371955871582, + "learning_rate": 0.00048652510085163604, + "loss": 0.4365, + "step": 18447 + }, + { + "epoch": 1.03072883202503, + "grad_norm": 3.645042657852173, + "learning_rate": 0.0004864970865082923, + "loss": 0.4103, + "step": 18448 + }, + { + "epoch": 1.0307847026287118, + "grad_norm": 0.30325260758399963, + "learning_rate": 0.00048646907216494845, + "loss": 0.3135, + "step": 18449 + }, + { + "epoch": 1.0308405732323938, + "grad_norm": 0.6390168070793152, + "learning_rate": 0.0004864410578216047, + "loss": 0.4607, + "step": 18450 + }, + { + "epoch": 1.0308964438360757, + "grad_norm": 0.41691431403160095, + "learning_rate": 0.00048641304347826086, + "loss": 0.3549, + "step": 18451 + }, + { + "epoch": 1.0309523144397574, + "grad_norm": 0.4489305317401886, + "learning_rate": 0.0004863850291349171, + "loss": 0.5532, + "step": 18452 + }, + { + "epoch": 1.0310081850434394, + "grad_norm": 0.3592822849750519, + "learning_rate": 0.00048635701479157327, + "loss": 0.3534, + "step": 18453 + }, + { + "epoch": 1.0310640556471213, + "grad_norm": 0.472087562084198, + "learning_rate": 0.00048632900044822953, + "loss": 0.5613, + "step": 18454 + }, + { + "epoch": 1.031119926250803, + "grad_norm": 0.8062137365341187, + "learning_rate": 0.00048630098610488574, + "loss": 0.558, + "step": 18455 + }, + { + "epoch": 1.031175796854485, + "grad_norm": 0.35769978165626526, + "learning_rate": 0.0004862729717615419, + "loss": 0.3583, + "step": 18456 + }, + { + "epoch": 1.031231667458167, + "grad_norm": 1.298961877822876, + "learning_rate": 0.00048624495741819815, + "loss": 0.5267, + "step": 18457 + }, + { + "epoch": 1.0312875380618487, + "grad_norm": 0.7452054023742676, + "learning_rate": 0.0004862169430748543, + "loss": 0.4646, + "step": 18458 + }, + { + "epoch": 1.0313434086655306, + "grad_norm": 0.7730370163917542, + "learning_rate": 0.00048618892873151056, + "loss": 0.5965, + "step": 18459 + }, + { + "epoch": 1.0313992792692126, + "grad_norm": 0.5816314220428467, + "learning_rate": 0.0004861609143881667, + "loss": 0.4196, + "step": 18460 + }, + { + "epoch": 1.0314551498728943, + "grad_norm": 0.38286396861076355, + "learning_rate": 0.000486132900044823, + "loss": 0.4473, + "step": 18461 + }, + { + "epoch": 1.0315110204765763, + "grad_norm": 0.47528076171875, + "learning_rate": 0.0004861048857014791, + "loss": 0.4287, + "step": 18462 + }, + { + "epoch": 1.0315668910802582, + "grad_norm": 0.7049185633659363, + "learning_rate": 0.0004860768713581354, + "loss": 0.4569, + "step": 18463 + }, + { + "epoch": 1.03162276168394, + "grad_norm": 0.37496355175971985, + "learning_rate": 0.0004860488570147916, + "loss": 0.4302, + "step": 18464 + }, + { + "epoch": 1.031678632287622, + "grad_norm": 0.4467948377132416, + "learning_rate": 0.0004860208426714478, + "loss": 0.5102, + "step": 18465 + }, + { + "epoch": 1.0317345028913036, + "grad_norm": 0.7421972751617432, + "learning_rate": 0.000485992828328104, + "loss": 0.4047, + "step": 18466 + }, + { + "epoch": 1.0317903734949856, + "grad_norm": 0.6005853414535522, + "learning_rate": 0.0004859648139847602, + "loss": 0.4374, + "step": 18467 + }, + { + "epoch": 1.0318462440986675, + "grad_norm": 0.41686776280403137, + "learning_rate": 0.0004859367996414164, + "loss": 0.3071, + "step": 18468 + }, + { + "epoch": 1.0319021147023493, + "grad_norm": 1.592700719833374, + "learning_rate": 0.0004859087852980726, + "loss": 0.5396, + "step": 18469 + }, + { + "epoch": 1.0319579853060312, + "grad_norm": 0.5326283574104309, + "learning_rate": 0.0004858807709547288, + "loss": 0.4501, + "step": 18470 + }, + { + "epoch": 1.0320138559097132, + "grad_norm": 0.35380297899246216, + "learning_rate": 0.0004858527566113851, + "loss": 0.4116, + "step": 18471 + }, + { + "epoch": 1.032069726513395, + "grad_norm": 0.4675070643424988, + "learning_rate": 0.00048582474226804124, + "loss": 0.4434, + "step": 18472 + }, + { + "epoch": 1.0321255971170769, + "grad_norm": 0.49225160479545593, + "learning_rate": 0.0004857967279246975, + "loss": 0.4305, + "step": 18473 + }, + { + "epoch": 1.0321814677207588, + "grad_norm": 0.47256505489349365, + "learning_rate": 0.00048576871358135365, + "loss": 0.4446, + "step": 18474 + }, + { + "epoch": 1.0322373383244405, + "grad_norm": 0.3427252471446991, + "learning_rate": 0.0004857406992380099, + "loss": 0.4129, + "step": 18475 + }, + { + "epoch": 1.0322932089281225, + "grad_norm": 0.5824837684631348, + "learning_rate": 0.00048571268489466606, + "loss": 0.4899, + "step": 18476 + }, + { + "epoch": 1.0323490795318044, + "grad_norm": 0.45712071657180786, + "learning_rate": 0.00048568467055132227, + "loss": 0.4238, + "step": 18477 + }, + { + "epoch": 1.0324049501354862, + "grad_norm": 0.4238644540309906, + "learning_rate": 0.00048565665620797847, + "loss": 0.3893, + "step": 18478 + }, + { + "epoch": 1.0324608207391681, + "grad_norm": 0.8821075558662415, + "learning_rate": 0.0004856286418646347, + "loss": 0.4086, + "step": 18479 + }, + { + "epoch": 1.0325166913428498, + "grad_norm": 0.3799939751625061, + "learning_rate": 0.00048560062752129094, + "loss": 0.3922, + "step": 18480 + }, + { + "epoch": 1.0325725619465318, + "grad_norm": 0.3731380105018616, + "learning_rate": 0.0004855726131779471, + "loss": 0.4877, + "step": 18481 + }, + { + "epoch": 1.0326284325502137, + "grad_norm": 0.7395439147949219, + "learning_rate": 0.00048554459883460335, + "loss": 0.5938, + "step": 18482 + }, + { + "epoch": 1.0326843031538955, + "grad_norm": 0.38706910610198975, + "learning_rate": 0.0004855165844912595, + "loss": 0.4033, + "step": 18483 + }, + { + "epoch": 1.0327401737575774, + "grad_norm": 0.2858664095401764, + "learning_rate": 0.00048548857014791576, + "loss": 0.3568, + "step": 18484 + }, + { + "epoch": 1.0327960443612594, + "grad_norm": 0.9824135303497314, + "learning_rate": 0.0004854605558045719, + "loss": 0.442, + "step": 18485 + }, + { + "epoch": 1.032851914964941, + "grad_norm": 0.5056923031806946, + "learning_rate": 0.00048543254146122817, + "loss": 0.4021, + "step": 18486 + }, + { + "epoch": 1.032907785568623, + "grad_norm": 2.4833974838256836, + "learning_rate": 0.0004854045271178844, + "loss": 0.4237, + "step": 18487 + }, + { + "epoch": 1.032963656172305, + "grad_norm": 0.487551748752594, + "learning_rate": 0.0004853765127745406, + "loss": 0.4451, + "step": 18488 + }, + { + "epoch": 1.0330195267759867, + "grad_norm": 0.4239400625228882, + "learning_rate": 0.0004853484984311968, + "loss": 0.3816, + "step": 18489 + }, + { + "epoch": 1.0330753973796687, + "grad_norm": 0.43590644001960754, + "learning_rate": 0.000485320484087853, + "loss": 0.4508, + "step": 18490 + }, + { + "epoch": 1.0331312679833506, + "grad_norm": 0.46437501907348633, + "learning_rate": 0.0004852924697445092, + "loss": 0.3619, + "step": 18491 + }, + { + "epoch": 1.0331871385870324, + "grad_norm": 1.6679961681365967, + "learning_rate": 0.0004852644554011654, + "loss": 0.5497, + "step": 18492 + }, + { + "epoch": 1.0332430091907143, + "grad_norm": 0.6817772388458252, + "learning_rate": 0.0004852364410578216, + "loss": 0.3434, + "step": 18493 + }, + { + "epoch": 1.0332988797943963, + "grad_norm": 0.42623016238212585, + "learning_rate": 0.0004852084267144778, + "loss": 0.4379, + "step": 18494 + }, + { + "epoch": 1.033354750398078, + "grad_norm": 0.3503974378108978, + "learning_rate": 0.000485180412371134, + "loss": 0.3698, + "step": 18495 + }, + { + "epoch": 1.03341062100176, + "grad_norm": 0.6980582475662231, + "learning_rate": 0.0004851523980277903, + "loss": 0.428, + "step": 18496 + }, + { + "epoch": 1.033466491605442, + "grad_norm": 1.6316431760787964, + "learning_rate": 0.00048512438368444644, + "loss": 0.4914, + "step": 18497 + }, + { + "epoch": 1.0335223622091236, + "grad_norm": 0.6057857871055603, + "learning_rate": 0.00048509636934110264, + "loss": 0.4379, + "step": 18498 + }, + { + "epoch": 1.0335782328128056, + "grad_norm": 0.488147109746933, + "learning_rate": 0.00048506835499775885, + "loss": 0.475, + "step": 18499 + }, + { + "epoch": 1.0336341034164873, + "grad_norm": 0.49152839183807373, + "learning_rate": 0.00048504034065441505, + "loss": 0.4727, + "step": 18500 + }, + { + "epoch": 1.0336341034164873, + "eval_cer": 0.08929331282118425, + "eval_loss": 0.33201003074645996, + "eval_runtime": 56.4526, + "eval_samples_per_second": 80.386, + "eval_steps_per_second": 5.031, + "eval_wer": 0.358816792827016, + "step": 18500 + }, + { + "epoch": 1.0336899740201693, + "grad_norm": 0.9785820245742798, + "learning_rate": 0.00048501232631107126, + "loss": 0.338, + "step": 18501 + }, + { + "epoch": 1.0337458446238512, + "grad_norm": 1.2491443157196045, + "learning_rate": 0.00048498431196772747, + "loss": 0.358, + "step": 18502 + }, + { + "epoch": 1.033801715227533, + "grad_norm": 0.6541601419448853, + "learning_rate": 0.0004849562976243837, + "loss": 0.5009, + "step": 18503 + }, + { + "epoch": 1.033857585831215, + "grad_norm": 0.4336085915565491, + "learning_rate": 0.0004849282832810399, + "loss": 0.4076, + "step": 18504 + }, + { + "epoch": 1.0339134564348968, + "grad_norm": 0.4650174677371979, + "learning_rate": 0.00048490026893769614, + "loss": 0.505, + "step": 18505 + }, + { + "epoch": 1.0339693270385786, + "grad_norm": 28.11886215209961, + "learning_rate": 0.0004848722545943523, + "loss": 0.4672, + "step": 18506 + }, + { + "epoch": 1.0340251976422605, + "grad_norm": 0.5450652241706848, + "learning_rate": 0.00048484424025100855, + "loss": 0.4513, + "step": 18507 + }, + { + "epoch": 1.0340810682459425, + "grad_norm": 0.47506365180015564, + "learning_rate": 0.0004848162259076647, + "loss": 0.5225, + "step": 18508 + }, + { + "epoch": 1.0341369388496242, + "grad_norm": 0.4979197084903717, + "learning_rate": 0.00048478821156432096, + "loss": 0.4095, + "step": 18509 + }, + { + "epoch": 1.0341928094533062, + "grad_norm": 0.48170191049575806, + "learning_rate": 0.0004847601972209771, + "loss": 0.3745, + "step": 18510 + }, + { + "epoch": 1.0342486800569881, + "grad_norm": 0.5263825058937073, + "learning_rate": 0.00048473218287763337, + "loss": 0.4093, + "step": 18511 + }, + { + "epoch": 1.0343045506606698, + "grad_norm": 0.4062644839286804, + "learning_rate": 0.0004847041685342896, + "loss": 0.5604, + "step": 18512 + }, + { + "epoch": 1.0343604212643518, + "grad_norm": 0.5789686441421509, + "learning_rate": 0.0004846761541909458, + "loss": 0.3606, + "step": 18513 + }, + { + "epoch": 1.0344162918680335, + "grad_norm": 1.36076021194458, + "learning_rate": 0.000484648139847602, + "loss": 0.4604, + "step": 18514 + }, + { + "epoch": 1.0344721624717155, + "grad_norm": 1.3601237535476685, + "learning_rate": 0.0004846201255042582, + "loss": 0.4999, + "step": 18515 + }, + { + "epoch": 1.0345280330753974, + "grad_norm": 0.3765154778957367, + "learning_rate": 0.0004845921111609144, + "loss": 0.3631, + "step": 18516 + }, + { + "epoch": 1.0345839036790792, + "grad_norm": 0.721697211265564, + "learning_rate": 0.0004845640968175706, + "loss": 0.3951, + "step": 18517 + }, + { + "epoch": 1.034639774282761, + "grad_norm": 0.5431519150733948, + "learning_rate": 0.0004845360824742268, + "loss": 0.5405, + "step": 18518 + }, + { + "epoch": 1.034695644886443, + "grad_norm": 0.7616831064224243, + "learning_rate": 0.000484508068130883, + "loss": 0.4477, + "step": 18519 + }, + { + "epoch": 1.0347515154901248, + "grad_norm": 0.37584686279296875, + "learning_rate": 0.0004844800537875392, + "loss": 0.451, + "step": 18520 + }, + { + "epoch": 1.0348073860938067, + "grad_norm": 0.3200860619544983, + "learning_rate": 0.00048445203944419543, + "loss": 0.3593, + "step": 18521 + }, + { + "epoch": 1.0348632566974887, + "grad_norm": 2.065326690673828, + "learning_rate": 0.00048442402510085164, + "loss": 0.4665, + "step": 18522 + }, + { + "epoch": 1.0349191273011704, + "grad_norm": 0.4688323140144348, + "learning_rate": 0.00048439601075750784, + "loss": 0.4328, + "step": 18523 + }, + { + "epoch": 1.0349749979048524, + "grad_norm": 0.6515554785728455, + "learning_rate": 0.00048436799641416405, + "loss": 0.4124, + "step": 18524 + }, + { + "epoch": 1.0350308685085343, + "grad_norm": 0.6302104592323303, + "learning_rate": 0.00048433998207082025, + "loss": 0.4866, + "step": 18525 + }, + { + "epoch": 1.035086739112216, + "grad_norm": 0.398929238319397, + "learning_rate": 0.0004843119677274765, + "loss": 0.4845, + "step": 18526 + }, + { + "epoch": 1.035142609715898, + "grad_norm": 0.3970988690853119, + "learning_rate": 0.00048428395338413267, + "loss": 0.3714, + "step": 18527 + }, + { + "epoch": 1.03519848031958, + "grad_norm": 0.4041551351547241, + "learning_rate": 0.0004842559390407889, + "loss": 0.5448, + "step": 18528 + }, + { + "epoch": 1.0352543509232617, + "grad_norm": 0.3649987280368805, + "learning_rate": 0.0004842279246974451, + "loss": 0.3242, + "step": 18529 + }, + { + "epoch": 1.0353102215269436, + "grad_norm": 1.3700679540634155, + "learning_rate": 0.00048419991035410134, + "loss": 0.5722, + "step": 18530 + }, + { + "epoch": 1.0353660921306256, + "grad_norm": 0.6782779097557068, + "learning_rate": 0.0004841718960107575, + "loss": 0.3741, + "step": 18531 + }, + { + "epoch": 1.0354219627343073, + "grad_norm": 0.37052121758461, + "learning_rate": 0.00048414388166741375, + "loss": 0.3114, + "step": 18532 + }, + { + "epoch": 1.0354778333379893, + "grad_norm": 1.752467155456543, + "learning_rate": 0.0004841158673240699, + "loss": 0.8508, + "step": 18533 + }, + { + "epoch": 1.035533703941671, + "grad_norm": 0.46586281061172485, + "learning_rate": 0.00048408785298072616, + "loss": 0.5163, + "step": 18534 + }, + { + "epoch": 1.035589574545353, + "grad_norm": 0.5419819355010986, + "learning_rate": 0.00048405983863738237, + "loss": 0.4506, + "step": 18535 + }, + { + "epoch": 1.035645445149035, + "grad_norm": 0.4377056658267975, + "learning_rate": 0.00048403182429403857, + "loss": 0.346, + "step": 18536 + }, + { + "epoch": 1.0357013157527166, + "grad_norm": 0.4109164774417877, + "learning_rate": 0.0004840038099506948, + "loss": 0.3958, + "step": 18537 + }, + { + "epoch": 1.0357571863563986, + "grad_norm": 10.369301795959473, + "learning_rate": 0.000483975795607351, + "loss": 0.3999, + "step": 18538 + }, + { + "epoch": 1.0358130569600805, + "grad_norm": 0.40018758177757263, + "learning_rate": 0.0004839477812640072, + "loss": 0.4537, + "step": 18539 + }, + { + "epoch": 1.0358689275637623, + "grad_norm": 0.825415849685669, + "learning_rate": 0.00048391976692066334, + "loss": 0.3536, + "step": 18540 + }, + { + "epoch": 1.0359247981674442, + "grad_norm": 0.5416344404220581, + "learning_rate": 0.0004838917525773196, + "loss": 0.5466, + "step": 18541 + }, + { + "epoch": 1.0359806687711262, + "grad_norm": 0.7457669973373413, + "learning_rate": 0.0004838637382339758, + "loss": 0.4603, + "step": 18542 + }, + { + "epoch": 1.0360365393748079, + "grad_norm": 0.6654112935066223, + "learning_rate": 0.000483835723890632, + "loss": 0.4198, + "step": 18543 + }, + { + "epoch": 1.0360924099784898, + "grad_norm": 0.6578369736671448, + "learning_rate": 0.0004838077095472882, + "loss": 0.4844, + "step": 18544 + }, + { + "epoch": 1.0361482805821718, + "grad_norm": 0.5652469396591187, + "learning_rate": 0.0004837796952039444, + "loss": 0.3283, + "step": 18545 + }, + { + "epoch": 1.0362041511858535, + "grad_norm": 0.5065485835075378, + "learning_rate": 0.00048375168086060063, + "loss": 0.448, + "step": 18546 + }, + { + "epoch": 1.0362600217895355, + "grad_norm": 2.55049991607666, + "learning_rate": 0.00048372366651725684, + "loss": 0.5168, + "step": 18547 + }, + { + "epoch": 1.0363158923932172, + "grad_norm": 0.46358683705329895, + "learning_rate": 0.00048369565217391304, + "loss": 0.3263, + "step": 18548 + }, + { + "epoch": 1.0363717629968991, + "grad_norm": 0.5978980660438538, + "learning_rate": 0.00048366763783056925, + "loss": 0.3738, + "step": 18549 + }, + { + "epoch": 1.036427633600581, + "grad_norm": 0.5092008709907532, + "learning_rate": 0.00048363962348722545, + "loss": 0.4902, + "step": 18550 + }, + { + "epoch": 1.0364835042042628, + "grad_norm": 0.5102955102920532, + "learning_rate": 0.0004836116091438817, + "loss": 0.4492, + "step": 18551 + }, + { + "epoch": 1.0365393748079448, + "grad_norm": 0.4279729127883911, + "learning_rate": 0.00048358359480053787, + "loss": 0.5028, + "step": 18552 + }, + { + "epoch": 1.0365952454116267, + "grad_norm": 0.8160298466682434, + "learning_rate": 0.0004835555804571941, + "loss": 0.3706, + "step": 18553 + }, + { + "epoch": 1.0366511160153085, + "grad_norm": 0.54111647605896, + "learning_rate": 0.0004835275661138503, + "loss": 0.4289, + "step": 18554 + }, + { + "epoch": 1.0367069866189904, + "grad_norm": 0.5637398362159729, + "learning_rate": 0.00048349955177050654, + "loss": 0.5935, + "step": 18555 + }, + { + "epoch": 1.0367628572226724, + "grad_norm": 0.4210165739059448, + "learning_rate": 0.0004834715374271627, + "loss": 0.3868, + "step": 18556 + }, + { + "epoch": 1.036818727826354, + "grad_norm": 0.60826176404953, + "learning_rate": 0.00048344352308381895, + "loss": 0.5335, + "step": 18557 + }, + { + "epoch": 1.036874598430036, + "grad_norm": 0.4419252276420593, + "learning_rate": 0.00048341550874047515, + "loss": 0.3767, + "step": 18558 + }, + { + "epoch": 1.036930469033718, + "grad_norm": 0.6025627255439758, + "learning_rate": 0.00048338749439713136, + "loss": 0.4417, + "step": 18559 + }, + { + "epoch": 1.0369863396373997, + "grad_norm": 0.4035542905330658, + "learning_rate": 0.00048335948005378757, + "loss": 0.3154, + "step": 18560 + }, + { + "epoch": 1.0370422102410817, + "grad_norm": 0.41656261682510376, + "learning_rate": 0.0004833314657104437, + "loss": 0.4191, + "step": 18561 + }, + { + "epoch": 1.0370980808447636, + "grad_norm": 0.4423770308494568, + "learning_rate": 0.0004833034513671, + "loss": 0.4825, + "step": 18562 + }, + { + "epoch": 1.0371539514484454, + "grad_norm": 0.28834274411201477, + "learning_rate": 0.00048327543702375613, + "loss": 0.3962, + "step": 18563 + }, + { + "epoch": 1.0372098220521273, + "grad_norm": 0.6504805088043213, + "learning_rate": 0.0004832474226804124, + "loss": 0.529, + "step": 18564 + }, + { + "epoch": 1.0372656926558093, + "grad_norm": 0.4614419937133789, + "learning_rate": 0.00048321940833706854, + "loss": 0.3147, + "step": 18565 + }, + { + "epoch": 1.037321563259491, + "grad_norm": 0.4219275116920471, + "learning_rate": 0.0004831913939937248, + "loss": 0.5335, + "step": 18566 + }, + { + "epoch": 1.037377433863173, + "grad_norm": 3.4315359592437744, + "learning_rate": 0.000483163379650381, + "loss": 0.4452, + "step": 18567 + }, + { + "epoch": 1.0374333044668547, + "grad_norm": 0.404437780380249, + "learning_rate": 0.0004831353653070372, + "loss": 0.4001, + "step": 18568 + }, + { + "epoch": 1.0374891750705366, + "grad_norm": 0.4064788222312927, + "learning_rate": 0.0004831073509636934, + "loss": 0.2904, + "step": 18569 + }, + { + "epoch": 1.0375450456742186, + "grad_norm": 0.5065802931785583, + "learning_rate": 0.0004830793366203496, + "loss": 0.3623, + "step": 18570 + }, + { + "epoch": 1.0376009162779003, + "grad_norm": 0.6776336431503296, + "learning_rate": 0.00048305132227700583, + "loss": 0.4619, + "step": 18571 + }, + { + "epoch": 1.0376567868815822, + "grad_norm": 0.33705973625183105, + "learning_rate": 0.00048302330793366204, + "loss": 0.4186, + "step": 18572 + }, + { + "epoch": 1.0377126574852642, + "grad_norm": 0.828400731086731, + "learning_rate": 0.00048299529359031824, + "loss": 0.3031, + "step": 18573 + }, + { + "epoch": 1.037768528088946, + "grad_norm": 0.4916454255580902, + "learning_rate": 0.0004829672792469745, + "loss": 0.6326, + "step": 18574 + }, + { + "epoch": 1.0378243986926279, + "grad_norm": 0.47016215324401855, + "learning_rate": 0.00048293926490363065, + "loss": 0.4207, + "step": 18575 + }, + { + "epoch": 1.0378802692963098, + "grad_norm": 0.42693018913269043, + "learning_rate": 0.0004829112505602869, + "loss": 0.4122, + "step": 18576 + }, + { + "epoch": 1.0379361398999916, + "grad_norm": 0.5451803207397461, + "learning_rate": 0.00048288323621694306, + "loss": 0.444, + "step": 18577 + }, + { + "epoch": 1.0379920105036735, + "grad_norm": 1.165772557258606, + "learning_rate": 0.0004828552218735993, + "loss": 0.301, + "step": 18578 + }, + { + "epoch": 1.0380478811073555, + "grad_norm": 0.43640536069869995, + "learning_rate": 0.0004828272075302555, + "loss": 0.4321, + "step": 18579 + }, + { + "epoch": 1.0381037517110372, + "grad_norm": 1.0727429389953613, + "learning_rate": 0.00048279919318691174, + "loss": 0.4453, + "step": 18580 + }, + { + "epoch": 1.0381596223147191, + "grad_norm": 0.47160932421684265, + "learning_rate": 0.00048277117884356794, + "loss": 0.3651, + "step": 18581 + }, + { + "epoch": 1.0382154929184009, + "grad_norm": 0.5758702754974365, + "learning_rate": 0.00048274316450022415, + "loss": 0.4163, + "step": 18582 + }, + { + "epoch": 1.0382713635220828, + "grad_norm": 0.43246328830718994, + "learning_rate": 0.00048271515015688035, + "loss": 0.4041, + "step": 18583 + }, + { + "epoch": 1.0383272341257648, + "grad_norm": 0.48376569151878357, + "learning_rate": 0.0004826871358135365, + "loss": 0.4478, + "step": 18584 + }, + { + "epoch": 1.0383831047294465, + "grad_norm": 0.6066864728927612, + "learning_rate": 0.00048265912147019277, + "loss": 0.386, + "step": 18585 + }, + { + "epoch": 1.0384389753331285, + "grad_norm": 0.6829167604446411, + "learning_rate": 0.0004826311071268489, + "loss": 0.4549, + "step": 18586 + }, + { + "epoch": 1.0384948459368104, + "grad_norm": 1.9914984703063965, + "learning_rate": 0.0004826030927835052, + "loss": 0.5274, + "step": 18587 + }, + { + "epoch": 1.0385507165404921, + "grad_norm": 0.5458638668060303, + "learning_rate": 0.00048257507844016133, + "loss": 0.3969, + "step": 18588 + }, + { + "epoch": 1.038606587144174, + "grad_norm": 0.7171047329902649, + "learning_rate": 0.0004825470640968176, + "loss": 0.4501, + "step": 18589 + }, + { + "epoch": 1.038662457747856, + "grad_norm": 0.6643500924110413, + "learning_rate": 0.0004825190497534738, + "loss": 0.5314, + "step": 18590 + }, + { + "epoch": 1.0387183283515378, + "grad_norm": 0.3522263467311859, + "learning_rate": 0.00048249103541013, + "loss": 0.371, + "step": 18591 + }, + { + "epoch": 1.0387741989552197, + "grad_norm": 0.3166092336177826, + "learning_rate": 0.0004824630210667862, + "loss": 0.3443, + "step": 18592 + }, + { + "epoch": 1.0388300695589017, + "grad_norm": 0.5201511383056641, + "learning_rate": 0.0004824350067234424, + "loss": 0.3604, + "step": 18593 + }, + { + "epoch": 1.0388859401625834, + "grad_norm": 1.6393499374389648, + "learning_rate": 0.0004824069923800986, + "loss": 0.4389, + "step": 18594 + }, + { + "epoch": 1.0389418107662654, + "grad_norm": 0.46234580874443054, + "learning_rate": 0.0004823789780367548, + "loss": 0.4492, + "step": 18595 + }, + { + "epoch": 1.0389976813699473, + "grad_norm": 0.6296025514602661, + "learning_rate": 0.00048235096369341103, + "loss": 0.4352, + "step": 18596 + }, + { + "epoch": 1.039053551973629, + "grad_norm": 1.0487267971038818, + "learning_rate": 0.0004823229493500673, + "loss": 0.5177, + "step": 18597 + }, + { + "epoch": 1.039109422577311, + "grad_norm": 0.5471785664558411, + "learning_rate": 0.00048229493500672344, + "loss": 0.6104, + "step": 18598 + }, + { + "epoch": 1.039165293180993, + "grad_norm": 0.40385156869888306, + "learning_rate": 0.0004822669206633797, + "loss": 0.4426, + "step": 18599 + }, + { + "epoch": 1.0392211637846747, + "grad_norm": 0.760958194732666, + "learning_rate": 0.00048223890632003585, + "loss": 0.4146, + "step": 18600 + }, + { + "epoch": 1.0392770343883566, + "grad_norm": 0.4715145528316498, + "learning_rate": 0.0004822108919766921, + "loss": 0.3968, + "step": 18601 + }, + { + "epoch": 1.0393329049920383, + "grad_norm": 1.09650456905365, + "learning_rate": 0.00048218287763334826, + "loss": 0.3775, + "step": 18602 + }, + { + "epoch": 1.0393887755957203, + "grad_norm": 3.643892288208008, + "learning_rate": 0.0004821548632900045, + "loss": 0.5226, + "step": 18603 + }, + { + "epoch": 1.0394446461994022, + "grad_norm": 0.5416861772537231, + "learning_rate": 0.0004821268489466607, + "loss": 0.2895, + "step": 18604 + }, + { + "epoch": 1.039500516803084, + "grad_norm": 0.4593907594680786, + "learning_rate": 0.0004820988346033169, + "loss": 0.3894, + "step": 18605 + }, + { + "epoch": 1.039556387406766, + "grad_norm": 1.7538046836853027, + "learning_rate": 0.00048207082025997314, + "loss": 0.4371, + "step": 18606 + }, + { + "epoch": 1.0396122580104479, + "grad_norm": 0.4485698938369751, + "learning_rate": 0.0004820428059166293, + "loss": 0.3436, + "step": 18607 + }, + { + "epoch": 1.0396681286141296, + "grad_norm": 0.4382074773311615, + "learning_rate": 0.00048201479157328555, + "loss": 0.4837, + "step": 18608 + }, + { + "epoch": 1.0397239992178116, + "grad_norm": 3.602306604385376, + "learning_rate": 0.0004819867772299417, + "loss": 0.4187, + "step": 18609 + }, + { + "epoch": 1.0397798698214935, + "grad_norm": 0.47913509607315063, + "learning_rate": 0.00048195876288659796, + "loss": 0.5046, + "step": 18610 + }, + { + "epoch": 1.0398357404251752, + "grad_norm": 1.4039959907531738, + "learning_rate": 0.0004819307485432541, + "loss": 0.533, + "step": 18611 + }, + { + "epoch": 1.0398916110288572, + "grad_norm": 0.5239292979240417, + "learning_rate": 0.0004819027341999104, + "loss": 0.4563, + "step": 18612 + }, + { + "epoch": 1.0399474816325391, + "grad_norm": 0.4519944489002228, + "learning_rate": 0.0004818747198565666, + "loss": 0.47, + "step": 18613 + }, + { + "epoch": 1.0400033522362209, + "grad_norm": 0.5875957012176514, + "learning_rate": 0.0004818467055132228, + "loss": 0.4055, + "step": 18614 + }, + { + "epoch": 1.0400592228399028, + "grad_norm": 0.6692652702331543, + "learning_rate": 0.000481818691169879, + "loss": 0.3455, + "step": 18615 + }, + { + "epoch": 1.0401150934435845, + "grad_norm": 0.6027116775512695, + "learning_rate": 0.0004817906768265352, + "loss": 0.3906, + "step": 18616 + }, + { + "epoch": 1.0401709640472665, + "grad_norm": 0.711663007736206, + "learning_rate": 0.0004817626624831914, + "loss": 0.3945, + "step": 18617 + }, + { + "epoch": 1.0402268346509485, + "grad_norm": 0.8334952592849731, + "learning_rate": 0.0004817346481398476, + "loss": 0.3456, + "step": 18618 + }, + { + "epoch": 1.0402827052546302, + "grad_norm": 0.4690101444721222, + "learning_rate": 0.0004817066337965038, + "loss": 0.4487, + "step": 18619 + }, + { + "epoch": 1.0403385758583121, + "grad_norm": 0.9330597519874573, + "learning_rate": 0.00048167861945316, + "loss": 0.4118, + "step": 18620 + }, + { + "epoch": 1.040394446461994, + "grad_norm": 0.5505359172821045, + "learning_rate": 0.00048165060510981623, + "loss": 0.4239, + "step": 18621 + }, + { + "epoch": 1.0404503170656758, + "grad_norm": 0.5882265567779541, + "learning_rate": 0.0004816225907664725, + "loss": 0.4046, + "step": 18622 + }, + { + "epoch": 1.0405061876693578, + "grad_norm": 0.6024518609046936, + "learning_rate": 0.00048159457642312864, + "loss": 0.5154, + "step": 18623 + }, + { + "epoch": 1.0405620582730397, + "grad_norm": 0.3310547471046448, + "learning_rate": 0.0004815665620797849, + "loss": 0.3365, + "step": 18624 + }, + { + "epoch": 1.0406179288767214, + "grad_norm": 0.6984590291976929, + "learning_rate": 0.00048153854773644105, + "loss": 0.4158, + "step": 18625 + }, + { + "epoch": 1.0406737994804034, + "grad_norm": 0.6249784231185913, + "learning_rate": 0.00048151053339309726, + "loss": 0.4123, + "step": 18626 + }, + { + "epoch": 1.0407296700840853, + "grad_norm": 0.4711768627166748, + "learning_rate": 0.00048148251904975346, + "loss": 0.5026, + "step": 18627 + }, + { + "epoch": 1.040785540687767, + "grad_norm": 1.74514901638031, + "learning_rate": 0.00048145450470640967, + "loss": 0.3829, + "step": 18628 + }, + { + "epoch": 1.040841411291449, + "grad_norm": 0.7389633655548096, + "learning_rate": 0.00048142649036306593, + "loss": 0.4514, + "step": 18629 + }, + { + "epoch": 1.040897281895131, + "grad_norm": 0.6281452178955078, + "learning_rate": 0.0004813984760197221, + "loss": 0.5846, + "step": 18630 + }, + { + "epoch": 1.0409531524988127, + "grad_norm": 0.6679859757423401, + "learning_rate": 0.00048137046167637834, + "loss": 0.4395, + "step": 18631 + }, + { + "epoch": 1.0410090231024947, + "grad_norm": 0.5902581214904785, + "learning_rate": 0.0004813424473330345, + "loss": 0.5146, + "step": 18632 + }, + { + "epoch": 1.0410648937061766, + "grad_norm": 0.494890034198761, + "learning_rate": 0.00048131443298969075, + "loss": 0.3807, + "step": 18633 + }, + { + "epoch": 1.0411207643098583, + "grad_norm": 0.6368703246116638, + "learning_rate": 0.0004812864186463469, + "loss": 0.6306, + "step": 18634 + }, + { + "epoch": 1.0411766349135403, + "grad_norm": 2.1396749019622803, + "learning_rate": 0.00048125840430300316, + "loss": 0.4149, + "step": 18635 + }, + { + "epoch": 1.041232505517222, + "grad_norm": 0.729987382888794, + "learning_rate": 0.0004812303899596593, + "loss": 0.4939, + "step": 18636 + }, + { + "epoch": 1.041288376120904, + "grad_norm": 0.41790395975112915, + "learning_rate": 0.0004812023756163156, + "loss": 0.3532, + "step": 18637 + }, + { + "epoch": 1.041344246724586, + "grad_norm": 0.7253421545028687, + "learning_rate": 0.0004811743612729718, + "loss": 0.4763, + "step": 18638 + }, + { + "epoch": 1.0414001173282676, + "grad_norm": 1.4959256649017334, + "learning_rate": 0.000481146346929628, + "loss": 0.3955, + "step": 18639 + }, + { + "epoch": 1.0414559879319496, + "grad_norm": 0.6363739967346191, + "learning_rate": 0.0004811183325862842, + "loss": 0.3724, + "step": 18640 + }, + { + "epoch": 1.0415118585356316, + "grad_norm": 0.41158244013786316, + "learning_rate": 0.0004810903182429404, + "loss": 0.4683, + "step": 18641 + }, + { + "epoch": 1.0415677291393133, + "grad_norm": 1.541137933731079, + "learning_rate": 0.0004810623038995966, + "loss": 0.464, + "step": 18642 + }, + { + "epoch": 1.0416235997429952, + "grad_norm": 0.48146647214889526, + "learning_rate": 0.0004810342895562528, + "loss": 0.5395, + "step": 18643 + }, + { + "epoch": 1.0416794703466772, + "grad_norm": 0.4784839451313019, + "learning_rate": 0.000481006275212909, + "loss": 0.452, + "step": 18644 + }, + { + "epoch": 1.041735340950359, + "grad_norm": 1.990951657295227, + "learning_rate": 0.0004809782608695653, + "loss": 0.4928, + "step": 18645 + }, + { + "epoch": 1.0417912115540409, + "grad_norm": 0.5901873707771301, + "learning_rate": 0.00048095024652622143, + "loss": 0.3945, + "step": 18646 + }, + { + "epoch": 1.0418470821577228, + "grad_norm": 0.6251290440559387, + "learning_rate": 0.00048092223218287763, + "loss": 0.458, + "step": 18647 + }, + { + "epoch": 1.0419029527614045, + "grad_norm": 0.46032872796058655, + "learning_rate": 0.00048089421783953384, + "loss": 0.4131, + "step": 18648 + }, + { + "epoch": 1.0419588233650865, + "grad_norm": 0.6241039633750916, + "learning_rate": 0.00048086620349619005, + "loss": 0.3246, + "step": 18649 + }, + { + "epoch": 1.0420146939687682, + "grad_norm": 0.37582913041114807, + "learning_rate": 0.00048083818915284625, + "loss": 0.4333, + "step": 18650 + }, + { + "epoch": 1.0420705645724502, + "grad_norm": 0.41807687282562256, + "learning_rate": 0.00048081017480950246, + "loss": 0.3973, + "step": 18651 + }, + { + "epoch": 1.0421264351761321, + "grad_norm": 0.6966021060943604, + "learning_rate": 0.0004807821604661587, + "loss": 0.3585, + "step": 18652 + }, + { + "epoch": 1.0421823057798139, + "grad_norm": 0.48061373829841614, + "learning_rate": 0.00048075414612281487, + "loss": 0.4401, + "step": 18653 + }, + { + "epoch": 1.0422381763834958, + "grad_norm": 1.1187968254089355, + "learning_rate": 0.00048072613177947113, + "loss": 0.4167, + "step": 18654 + }, + { + "epoch": 1.0422940469871778, + "grad_norm": 0.7075284719467163, + "learning_rate": 0.0004806981174361273, + "loss": 0.3474, + "step": 18655 + }, + { + "epoch": 1.0423499175908595, + "grad_norm": 0.37790316343307495, + "learning_rate": 0.00048067010309278354, + "loss": 0.3407, + "step": 18656 + }, + { + "epoch": 1.0424057881945414, + "grad_norm": 0.32616713643074036, + "learning_rate": 0.0004806420887494397, + "loss": 0.4342, + "step": 18657 + }, + { + "epoch": 1.0424616587982234, + "grad_norm": 1.2943382263183594, + "learning_rate": 0.00048061407440609595, + "loss": 0.6435, + "step": 18658 + }, + { + "epoch": 1.0425175294019051, + "grad_norm": 0.9644662737846375, + "learning_rate": 0.0004805860600627521, + "loss": 0.5262, + "step": 18659 + }, + { + "epoch": 1.042573400005587, + "grad_norm": 0.6121543049812317, + "learning_rate": 0.00048055804571940836, + "loss": 0.4684, + "step": 18660 + }, + { + "epoch": 1.042629270609269, + "grad_norm": 0.3658926486968994, + "learning_rate": 0.00048053003137606457, + "loss": 0.4321, + "step": 18661 + }, + { + "epoch": 1.0426851412129508, + "grad_norm": 1.581866979598999, + "learning_rate": 0.0004805020170327208, + "loss": 0.4929, + "step": 18662 + }, + { + "epoch": 1.0427410118166327, + "grad_norm": 9.740697860717773, + "learning_rate": 0.000480474002689377, + "loss": 0.4566, + "step": 18663 + }, + { + "epoch": 1.0427968824203147, + "grad_norm": 0.5561107397079468, + "learning_rate": 0.0004804459883460332, + "loss": 0.4343, + "step": 18664 + }, + { + "epoch": 1.0428527530239964, + "grad_norm": 1.29888117313385, + "learning_rate": 0.0004804179740026894, + "loss": 0.4827, + "step": 18665 + }, + { + "epoch": 1.0429086236276783, + "grad_norm": 0.47080904245376587, + "learning_rate": 0.0004803899596593456, + "loss": 0.4003, + "step": 18666 + }, + { + "epoch": 1.0429644942313603, + "grad_norm": 0.6795974969863892, + "learning_rate": 0.0004803619453160018, + "loss": 0.5515, + "step": 18667 + }, + { + "epoch": 1.043020364835042, + "grad_norm": 0.42534661293029785, + "learning_rate": 0.000480333930972658, + "loss": 0.369, + "step": 18668 + }, + { + "epoch": 1.043076235438724, + "grad_norm": 0.4894769489765167, + "learning_rate": 0.0004803059166293142, + "loss": 0.4076, + "step": 18669 + }, + { + "epoch": 1.0431321060424057, + "grad_norm": 0.5072007179260254, + "learning_rate": 0.0004802779022859704, + "loss": 0.4518, + "step": 18670 + }, + { + "epoch": 1.0431879766460876, + "grad_norm": 0.6717013716697693, + "learning_rate": 0.00048024988794262663, + "loss": 0.4178, + "step": 18671 + }, + { + "epoch": 1.0432438472497696, + "grad_norm": 0.7375001907348633, + "learning_rate": 0.00048022187359928283, + "loss": 0.4654, + "step": 18672 + }, + { + "epoch": 1.0432997178534513, + "grad_norm": 0.48381751775741577, + "learning_rate": 0.00048019385925593904, + "loss": 0.3965, + "step": 18673 + }, + { + "epoch": 1.0433555884571333, + "grad_norm": 1.1559373140335083, + "learning_rate": 0.00048016584491259525, + "loss": 0.3006, + "step": 18674 + }, + { + "epoch": 1.0434114590608152, + "grad_norm": 0.6550506949424744, + "learning_rate": 0.00048013783056925145, + "loss": 0.4878, + "step": 18675 + }, + { + "epoch": 1.043467329664497, + "grad_norm": 0.6550810933113098, + "learning_rate": 0.00048010981622590766, + "loss": 0.4182, + "step": 18676 + }, + { + "epoch": 1.043523200268179, + "grad_norm": 0.5490190982818604, + "learning_rate": 0.0004800818018825639, + "loss": 0.4772, + "step": 18677 + }, + { + "epoch": 1.0435790708718609, + "grad_norm": 0.9096829891204834, + "learning_rate": 0.00048005378753922007, + "loss": 0.4449, + "step": 18678 + }, + { + "epoch": 1.0436349414755426, + "grad_norm": 0.5936029553413391, + "learning_rate": 0.00048002577319587633, + "loss": 0.4308, + "step": 18679 + }, + { + "epoch": 1.0436908120792245, + "grad_norm": 0.7989597320556641, + "learning_rate": 0.0004799977588525325, + "loss": 0.4437, + "step": 18680 + }, + { + "epoch": 1.0437466826829065, + "grad_norm": 0.3830178380012512, + "learning_rate": 0.00047996974450918874, + "loss": 0.3768, + "step": 18681 + }, + { + "epoch": 1.0438025532865882, + "grad_norm": 0.3824549615383148, + "learning_rate": 0.0004799417301658449, + "loss": 0.3715, + "step": 18682 + }, + { + "epoch": 1.0438584238902702, + "grad_norm": 0.4114072918891907, + "learning_rate": 0.00047991371582250115, + "loss": 0.3989, + "step": 18683 + }, + { + "epoch": 1.043914294493952, + "grad_norm": 0.4499852657318115, + "learning_rate": 0.00047988570147915736, + "loss": 0.4615, + "step": 18684 + }, + { + "epoch": 1.0439701650976339, + "grad_norm": 0.9713336825370789, + "learning_rate": 0.00047985768713581356, + "loss": 0.4146, + "step": 18685 + }, + { + "epoch": 1.0440260357013158, + "grad_norm": 0.49885305762290955, + "learning_rate": 0.00047982967279246977, + "loss": 0.4579, + "step": 18686 + }, + { + "epoch": 1.0440819063049975, + "grad_norm": 0.37668344378471375, + "learning_rate": 0.000479801658449126, + "loss": 0.3781, + "step": 18687 + }, + { + "epoch": 1.0441377769086795, + "grad_norm": 0.48520782589912415, + "learning_rate": 0.0004797736441057822, + "loss": 0.4956, + "step": 18688 + }, + { + "epoch": 1.0441936475123614, + "grad_norm": 0.6440392136573792, + "learning_rate": 0.00047974562976243833, + "loss": 0.3807, + "step": 18689 + }, + { + "epoch": 1.0442495181160432, + "grad_norm": 0.548438549041748, + "learning_rate": 0.0004797176154190946, + "loss": 0.5621, + "step": 18690 + }, + { + "epoch": 1.0443053887197251, + "grad_norm": 0.3919421434402466, + "learning_rate": 0.00047968960107575074, + "loss": 0.4069, + "step": 18691 + }, + { + "epoch": 1.044361259323407, + "grad_norm": 0.4021190404891968, + "learning_rate": 0.000479661586732407, + "loss": 0.4995, + "step": 18692 + }, + { + "epoch": 1.0444171299270888, + "grad_norm": 0.578088104724884, + "learning_rate": 0.0004796335723890632, + "loss": 0.3248, + "step": 18693 + }, + { + "epoch": 1.0444730005307707, + "grad_norm": 0.5828781127929688, + "learning_rate": 0.0004796055580457194, + "loss": 0.5198, + "step": 18694 + }, + { + "epoch": 1.0445288711344527, + "grad_norm": 1.651731252670288, + "learning_rate": 0.0004795775437023756, + "loss": 0.3732, + "step": 18695 + }, + { + "epoch": 1.0445847417381344, + "grad_norm": 0.4110654592514038, + "learning_rate": 0.00047954952935903183, + "loss": 0.4994, + "step": 18696 + }, + { + "epoch": 1.0446406123418164, + "grad_norm": 0.67000812292099, + "learning_rate": 0.00047952151501568803, + "loss": 0.5517, + "step": 18697 + }, + { + "epoch": 1.0446964829454983, + "grad_norm": 0.4897032380104065, + "learning_rate": 0.00047949350067234424, + "loss": 0.3617, + "step": 18698 + }, + { + "epoch": 1.04475235354918, + "grad_norm": 0.30829596519470215, + "learning_rate": 0.00047946548632900045, + "loss": 0.344, + "step": 18699 + }, + { + "epoch": 1.044808224152862, + "grad_norm": 0.47663959860801697, + "learning_rate": 0.0004794374719856567, + "loss": 0.461, + "step": 18700 + }, + { + "epoch": 1.0448640947565437, + "grad_norm": 0.4919140040874481, + "learning_rate": 0.00047940945764231286, + "loss": 0.414, + "step": 18701 + }, + { + "epoch": 1.0449199653602257, + "grad_norm": 0.6439734101295471, + "learning_rate": 0.0004793814432989691, + "loss": 0.4614, + "step": 18702 + }, + { + "epoch": 1.0449758359639076, + "grad_norm": 0.44980475306510925, + "learning_rate": 0.00047935342895562527, + "loss": 0.4352, + "step": 18703 + }, + { + "epoch": 1.0450317065675894, + "grad_norm": 2.723039150238037, + "learning_rate": 0.00047932541461228153, + "loss": 0.5928, + "step": 18704 + }, + { + "epoch": 1.0450875771712713, + "grad_norm": 1.4558357000350952, + "learning_rate": 0.0004792974002689377, + "loss": 0.4384, + "step": 18705 + }, + { + "epoch": 1.0451434477749533, + "grad_norm": 0.6160163879394531, + "learning_rate": 0.00047926938592559394, + "loss": 0.4859, + "step": 18706 + }, + { + "epoch": 1.045199318378635, + "grad_norm": 0.43106135725975037, + "learning_rate": 0.0004792413715822501, + "loss": 0.4711, + "step": 18707 + }, + { + "epoch": 1.045255188982317, + "grad_norm": 0.5141469836235046, + "learning_rate": 0.00047921335723890635, + "loss": 0.426, + "step": 18708 + }, + { + "epoch": 1.045311059585999, + "grad_norm": 0.4179990291595459, + "learning_rate": 0.00047918534289556256, + "loss": 0.3815, + "step": 18709 + }, + { + "epoch": 1.0453669301896806, + "grad_norm": 0.7228434085845947, + "learning_rate": 0.0004791573285522187, + "loss": 0.4049, + "step": 18710 + }, + { + "epoch": 1.0454228007933626, + "grad_norm": 1.538797378540039, + "learning_rate": 0.00047912931420887497, + "loss": 0.4239, + "step": 18711 + }, + { + "epoch": 1.0454786713970445, + "grad_norm": 0.4937976598739624, + "learning_rate": 0.0004791012998655311, + "loss": 0.4268, + "step": 18712 + }, + { + "epoch": 1.0455345420007263, + "grad_norm": 0.7679262161254883, + "learning_rate": 0.0004790732855221874, + "loss": 0.5199, + "step": 18713 + }, + { + "epoch": 1.0455904126044082, + "grad_norm": 0.5343531966209412, + "learning_rate": 0.00047904527117884353, + "loss": 0.4493, + "step": 18714 + }, + { + "epoch": 1.0456462832080902, + "grad_norm": 0.35031047463417053, + "learning_rate": 0.0004790172568354998, + "loss": 0.4269, + "step": 18715 + }, + { + "epoch": 1.045702153811772, + "grad_norm": 0.4482930898666382, + "learning_rate": 0.000478989242492156, + "loss": 0.3711, + "step": 18716 + }, + { + "epoch": 1.0457580244154538, + "grad_norm": 0.947471022605896, + "learning_rate": 0.0004789612281488122, + "loss": 0.4554, + "step": 18717 + }, + { + "epoch": 1.0458138950191356, + "grad_norm": 0.602116584777832, + "learning_rate": 0.0004789332138054684, + "loss": 0.5392, + "step": 18718 + }, + { + "epoch": 1.0458697656228175, + "grad_norm": 0.47295260429382324, + "learning_rate": 0.0004789051994621246, + "loss": 0.473, + "step": 18719 + }, + { + "epoch": 1.0459256362264995, + "grad_norm": 0.38058000802993774, + "learning_rate": 0.0004788771851187808, + "loss": 0.5404, + "step": 18720 + }, + { + "epoch": 1.0459815068301812, + "grad_norm": 0.32752376794815063, + "learning_rate": 0.00047884917077543703, + "loss": 0.3241, + "step": 18721 + }, + { + "epoch": 1.0460373774338632, + "grad_norm": 0.5334653258323669, + "learning_rate": 0.00047882115643209323, + "loss": 0.484, + "step": 18722 + }, + { + "epoch": 1.046093248037545, + "grad_norm": 0.36232590675354004, + "learning_rate": 0.0004787931420887495, + "loss": 0.4235, + "step": 18723 + }, + { + "epoch": 1.0461491186412268, + "grad_norm": 0.779760479927063, + "learning_rate": 0.00047876512774540564, + "loss": 0.4446, + "step": 18724 + }, + { + "epoch": 1.0462049892449088, + "grad_norm": 0.4755397439002991, + "learning_rate": 0.0004787371134020619, + "loss": 0.3629, + "step": 18725 + }, + { + "epoch": 1.0462608598485907, + "grad_norm": 0.5351112484931946, + "learning_rate": 0.00047870909905871806, + "loss": 0.5512, + "step": 18726 + }, + { + "epoch": 1.0463167304522725, + "grad_norm": 0.5567600131034851, + "learning_rate": 0.0004786810847153743, + "loss": 0.432, + "step": 18727 + }, + { + "epoch": 1.0463726010559544, + "grad_norm": 0.5163243412971497, + "learning_rate": 0.00047865307037203047, + "loss": 0.4739, + "step": 18728 + }, + { + "epoch": 1.0464284716596364, + "grad_norm": 0.4279181659221649, + "learning_rate": 0.00047862505602868673, + "loss": 0.4544, + "step": 18729 + }, + { + "epoch": 1.046484342263318, + "grad_norm": 0.5327653884887695, + "learning_rate": 0.0004785970416853429, + "loss": 0.5462, + "step": 18730 + }, + { + "epoch": 1.046540212867, + "grad_norm": 0.3819056451320648, + "learning_rate": 0.0004785690273419991, + "loss": 0.4133, + "step": 18731 + }, + { + "epoch": 1.046596083470682, + "grad_norm": 0.5392715334892273, + "learning_rate": 0.00047854101299865535, + "loss": 0.4981, + "step": 18732 + }, + { + "epoch": 1.0466519540743637, + "grad_norm": 0.5470923781394958, + "learning_rate": 0.0004785129986553115, + "loss": 0.683, + "step": 18733 + }, + { + "epoch": 1.0467078246780457, + "grad_norm": 0.5539746880531311, + "learning_rate": 0.00047848498431196776, + "loss": 0.4058, + "step": 18734 + }, + { + "epoch": 1.0467636952817274, + "grad_norm": 0.9461345076560974, + "learning_rate": 0.0004784569699686239, + "loss": 0.4756, + "step": 18735 + }, + { + "epoch": 1.0468195658854094, + "grad_norm": 0.8636103868484497, + "learning_rate": 0.00047842895562528017, + "loss": 0.4201, + "step": 18736 + }, + { + "epoch": 1.0468754364890913, + "grad_norm": 0.45916247367858887, + "learning_rate": 0.0004784009412819363, + "loss": 0.5351, + "step": 18737 + }, + { + "epoch": 1.046931307092773, + "grad_norm": 0.9831225275993347, + "learning_rate": 0.0004783729269385926, + "loss": 0.3514, + "step": 18738 + }, + { + "epoch": 1.046987177696455, + "grad_norm": 0.7937163710594177, + "learning_rate": 0.0004783449125952488, + "loss": 0.4454, + "step": 18739 + }, + { + "epoch": 1.047043048300137, + "grad_norm": 0.47684982419013977, + "learning_rate": 0.000478316898251905, + "loss": 0.395, + "step": 18740 + }, + { + "epoch": 1.0470989189038187, + "grad_norm": 0.3566957712173462, + "learning_rate": 0.0004782888839085612, + "loss": 0.4182, + "step": 18741 + }, + { + "epoch": 1.0471547895075006, + "grad_norm": 0.5086924433708191, + "learning_rate": 0.0004782608695652174, + "loss": 0.4105, + "step": 18742 + }, + { + "epoch": 1.0472106601111826, + "grad_norm": 0.4490985870361328, + "learning_rate": 0.0004782328552218736, + "loss": 0.3611, + "step": 18743 + }, + { + "epoch": 1.0472665307148643, + "grad_norm": 0.4974822700023651, + "learning_rate": 0.0004782048408785298, + "loss": 0.3351, + "step": 18744 + }, + { + "epoch": 1.0473224013185463, + "grad_norm": 1.0591868162155151, + "learning_rate": 0.000478176826535186, + "loss": 0.4557, + "step": 18745 + }, + { + "epoch": 1.0473782719222282, + "grad_norm": 0.5602611899375916, + "learning_rate": 0.0004781488121918422, + "loss": 0.3632, + "step": 18746 + }, + { + "epoch": 1.04743414252591, + "grad_norm": 0.6405073404312134, + "learning_rate": 0.00047812079784849843, + "loss": 0.3552, + "step": 18747 + }, + { + "epoch": 1.047490013129592, + "grad_norm": 0.5389959216117859, + "learning_rate": 0.0004780927835051547, + "loss": 0.4209, + "step": 18748 + }, + { + "epoch": 1.0475458837332738, + "grad_norm": 0.5472221970558167, + "learning_rate": 0.00047806476916181084, + "loss": 0.4473, + "step": 18749 + }, + { + "epoch": 1.0476017543369556, + "grad_norm": 0.48878422379493713, + "learning_rate": 0.0004780367548184671, + "loss": 0.634, + "step": 18750 + }, + { + "epoch": 1.0476576249406375, + "grad_norm": 0.61836177110672, + "learning_rate": 0.00047800874047512326, + "loss": 0.5367, + "step": 18751 + }, + { + "epoch": 1.0477134955443193, + "grad_norm": 0.5719289779663086, + "learning_rate": 0.0004779807261317795, + "loss": 0.54, + "step": 18752 + }, + { + "epoch": 1.0477693661480012, + "grad_norm": 0.5046599507331848, + "learning_rate": 0.00047795271178843567, + "loss": 0.34, + "step": 18753 + }, + { + "epoch": 1.0478252367516832, + "grad_norm": 0.7342709302902222, + "learning_rate": 0.0004779246974450919, + "loss": 0.6982, + "step": 18754 + }, + { + "epoch": 1.0478811073553649, + "grad_norm": 0.334354430437088, + "learning_rate": 0.00047789668310174813, + "loss": 0.4911, + "step": 18755 + }, + { + "epoch": 1.0479369779590468, + "grad_norm": 0.5584256052970886, + "learning_rate": 0.0004778686687584043, + "loss": 0.4348, + "step": 18756 + }, + { + "epoch": 1.0479928485627288, + "grad_norm": 0.5301147699356079, + "learning_rate": 0.00047784065441506055, + "loss": 0.4166, + "step": 18757 + }, + { + "epoch": 1.0480487191664105, + "grad_norm": 0.45054468512535095, + "learning_rate": 0.0004778126400717167, + "loss": 0.3511, + "step": 18758 + }, + { + "epoch": 1.0481045897700925, + "grad_norm": 0.45747116208076477, + "learning_rate": 0.00047778462572837296, + "loss": 0.4387, + "step": 18759 + }, + { + "epoch": 1.0481604603737744, + "grad_norm": 0.44585245847702026, + "learning_rate": 0.0004777566113850291, + "loss": 0.552, + "step": 18760 + }, + { + "epoch": 1.0482163309774561, + "grad_norm": 0.5349293947219849, + "learning_rate": 0.00047772859704168537, + "loss": 0.4677, + "step": 18761 + }, + { + "epoch": 1.048272201581138, + "grad_norm": 0.6619677543640137, + "learning_rate": 0.0004777005826983415, + "loss": 0.4287, + "step": 18762 + }, + { + "epoch": 1.04832807218482, + "grad_norm": 0.4826405346393585, + "learning_rate": 0.0004776725683549978, + "loss": 0.5598, + "step": 18763 + }, + { + "epoch": 1.0483839427885018, + "grad_norm": 1.504992961883545, + "learning_rate": 0.000477644554011654, + "loss": 0.5017, + "step": 18764 + }, + { + "epoch": 1.0484398133921837, + "grad_norm": 0.4987986087799072, + "learning_rate": 0.0004776165396683102, + "loss": 0.3599, + "step": 18765 + }, + { + "epoch": 1.0484956839958657, + "grad_norm": 0.37009409070014954, + "learning_rate": 0.0004775885253249664, + "loss": 0.4219, + "step": 18766 + }, + { + "epoch": 1.0485515545995474, + "grad_norm": 0.48728707432746887, + "learning_rate": 0.0004775605109816226, + "loss": 0.4329, + "step": 18767 + }, + { + "epoch": 1.0486074252032294, + "grad_norm": 0.6005893349647522, + "learning_rate": 0.0004775324966382788, + "loss": 0.4697, + "step": 18768 + }, + { + "epoch": 1.048663295806911, + "grad_norm": 0.4784763753414154, + "learning_rate": 0.000477504482294935, + "loss": 0.4179, + "step": 18769 + }, + { + "epoch": 1.048719166410593, + "grad_norm": 0.5708923935890198, + "learning_rate": 0.0004774764679515912, + "loss": 0.5676, + "step": 18770 + }, + { + "epoch": 1.048775037014275, + "grad_norm": 0.4546228349208832, + "learning_rate": 0.0004774484536082475, + "loss": 0.385, + "step": 18771 + }, + { + "epoch": 1.0488309076179567, + "grad_norm": 0.42185789346694946, + "learning_rate": 0.00047742043926490363, + "loss": 0.3221, + "step": 18772 + }, + { + "epoch": 1.0488867782216387, + "grad_norm": 0.5242276787757874, + "learning_rate": 0.0004773924249215599, + "loss": 0.4205, + "step": 18773 + }, + { + "epoch": 1.0489426488253206, + "grad_norm": 0.5128833055496216, + "learning_rate": 0.00047736441057821604, + "loss": 0.4507, + "step": 18774 + }, + { + "epoch": 1.0489985194290024, + "grad_norm": 0.5429149866104126, + "learning_rate": 0.00047733639623487225, + "loss": 0.5097, + "step": 18775 + }, + { + "epoch": 1.0490543900326843, + "grad_norm": 0.7148167490959167, + "learning_rate": 0.00047730838189152846, + "loss": 0.3855, + "step": 18776 + }, + { + "epoch": 1.0491102606363663, + "grad_norm": 0.7040285468101501, + "learning_rate": 0.00047728036754818466, + "loss": 0.6614, + "step": 18777 + }, + { + "epoch": 1.049166131240048, + "grad_norm": 0.4618641138076782, + "learning_rate": 0.0004772523532048409, + "loss": 0.4279, + "step": 18778 + }, + { + "epoch": 1.04922200184373, + "grad_norm": 0.4032747447490692, + "learning_rate": 0.0004772243388614971, + "loss": 0.4799, + "step": 18779 + }, + { + "epoch": 1.0492778724474119, + "grad_norm": 0.6372403502464294, + "learning_rate": 0.00047719632451815333, + "loss": 0.6505, + "step": 18780 + }, + { + "epoch": 1.0493337430510936, + "grad_norm": 1.3909331560134888, + "learning_rate": 0.0004771683101748095, + "loss": 0.5338, + "step": 18781 + }, + { + "epoch": 1.0493896136547756, + "grad_norm": 2.8606395721435547, + "learning_rate": 0.00047714029583146574, + "loss": 0.4137, + "step": 18782 + }, + { + "epoch": 1.0494454842584575, + "grad_norm": 0.5967307090759277, + "learning_rate": 0.0004771122814881219, + "loss": 0.5454, + "step": 18783 + }, + { + "epoch": 1.0495013548621392, + "grad_norm": 0.5983384847640991, + "learning_rate": 0.00047708426714477816, + "loss": 0.4288, + "step": 18784 + }, + { + "epoch": 1.0495572254658212, + "grad_norm": 0.4293185770511627, + "learning_rate": 0.0004770562528014343, + "loss": 0.4568, + "step": 18785 + }, + { + "epoch": 1.049613096069503, + "grad_norm": 0.683964729309082, + "learning_rate": 0.00047702823845809057, + "loss": 0.4898, + "step": 18786 + }, + { + "epoch": 1.0496689666731849, + "grad_norm": 0.46563708782196045, + "learning_rate": 0.0004770002241147468, + "loss": 0.292, + "step": 18787 + }, + { + "epoch": 1.0497248372768668, + "grad_norm": 0.5549793243408203, + "learning_rate": 0.000476972209771403, + "loss": 0.4624, + "step": 18788 + }, + { + "epoch": 1.0497807078805486, + "grad_norm": 0.5111228227615356, + "learning_rate": 0.0004769441954280592, + "loss": 0.4353, + "step": 18789 + }, + { + "epoch": 1.0498365784842305, + "grad_norm": 0.5460866093635559, + "learning_rate": 0.0004769161810847154, + "loss": 0.4302, + "step": 18790 + }, + { + "epoch": 1.0498924490879125, + "grad_norm": 0.37364548444747925, + "learning_rate": 0.0004768881667413716, + "loss": 0.4262, + "step": 18791 + }, + { + "epoch": 1.0499483196915942, + "grad_norm": 0.6230331659317017, + "learning_rate": 0.0004768601523980278, + "loss": 0.407, + "step": 18792 + }, + { + "epoch": 1.0500041902952761, + "grad_norm": 0.5425530076026917, + "learning_rate": 0.000476832138054684, + "loss": 0.4143, + "step": 18793 + }, + { + "epoch": 1.050060060898958, + "grad_norm": 0.474527508020401, + "learning_rate": 0.00047680412371134027, + "loss": 0.4128, + "step": 18794 + }, + { + "epoch": 1.0501159315026398, + "grad_norm": 0.5973784923553467, + "learning_rate": 0.0004767761093679964, + "loss": 0.4833, + "step": 18795 + }, + { + "epoch": 1.0501718021063218, + "grad_norm": 0.4510686695575714, + "learning_rate": 0.0004767480950246526, + "loss": 0.3614, + "step": 18796 + }, + { + "epoch": 1.0502276727100037, + "grad_norm": 0.6724164485931396, + "learning_rate": 0.00047672008068130883, + "loss": 0.6479, + "step": 18797 + }, + { + "epoch": 1.0502835433136855, + "grad_norm": 0.3915543258190155, + "learning_rate": 0.00047669206633796504, + "loss": 0.4394, + "step": 18798 + }, + { + "epoch": 1.0503394139173674, + "grad_norm": 0.40560635924339294, + "learning_rate": 0.00047666405199462124, + "loss": 0.462, + "step": 18799 + }, + { + "epoch": 1.0503952845210494, + "grad_norm": 0.3084856867790222, + "learning_rate": 0.00047663603765127745, + "loss": 0.3855, + "step": 18800 + }, + { + "epoch": 1.050451155124731, + "grad_norm": 0.3926101326942444, + "learning_rate": 0.00047660802330793366, + "loss": 0.4472, + "step": 18801 + }, + { + "epoch": 1.050507025728413, + "grad_norm": 0.8383581638336182, + "learning_rate": 0.00047658000896458986, + "loss": 0.5185, + "step": 18802 + }, + { + "epoch": 1.0505628963320948, + "grad_norm": 1.0286388397216797, + "learning_rate": 0.0004765519946212461, + "loss": 0.4681, + "step": 18803 + }, + { + "epoch": 1.0506187669357767, + "grad_norm": 0.5436263680458069, + "learning_rate": 0.00047652398027790227, + "loss": 0.5586, + "step": 18804 + }, + { + "epoch": 1.0506746375394587, + "grad_norm": 0.7734765410423279, + "learning_rate": 0.00047649596593455853, + "loss": 0.3301, + "step": 18805 + }, + { + "epoch": 1.0507305081431404, + "grad_norm": 0.5934070348739624, + "learning_rate": 0.0004764679515912147, + "loss": 0.4825, + "step": 18806 + }, + { + "epoch": 1.0507863787468223, + "grad_norm": 0.7238785028457642, + "learning_rate": 0.00047643993724787094, + "loss": 0.3436, + "step": 18807 + }, + { + "epoch": 1.0508422493505043, + "grad_norm": 1.4132771492004395, + "learning_rate": 0.0004764119229045271, + "loss": 0.3626, + "step": 18808 + }, + { + "epoch": 1.050898119954186, + "grad_norm": 0.5839511156082153, + "learning_rate": 0.00047638390856118336, + "loss": 0.3561, + "step": 18809 + }, + { + "epoch": 1.050953990557868, + "grad_norm": 1.0725263357162476, + "learning_rate": 0.00047635589421783956, + "loss": 0.3796, + "step": 18810 + }, + { + "epoch": 1.05100986116155, + "grad_norm": 0.3960241973400116, + "learning_rate": 0.00047632787987449577, + "loss": 0.3974, + "step": 18811 + }, + { + "epoch": 1.0510657317652317, + "grad_norm": 0.6362894773483276, + "learning_rate": 0.000476299865531152, + "loss": 0.4451, + "step": 18812 + }, + { + "epoch": 1.0511216023689136, + "grad_norm": 0.4252646565437317, + "learning_rate": 0.0004762718511878082, + "loss": 0.5552, + "step": 18813 + }, + { + "epoch": 1.0511774729725956, + "grad_norm": 0.42754021286964417, + "learning_rate": 0.0004762438368444644, + "loss": 0.3768, + "step": 18814 + }, + { + "epoch": 1.0512333435762773, + "grad_norm": 0.7870817184448242, + "learning_rate": 0.0004762158225011206, + "loss": 0.4294, + "step": 18815 + }, + { + "epoch": 1.0512892141799592, + "grad_norm": 0.5032487511634827, + "learning_rate": 0.0004761878081577768, + "loss": 0.4244, + "step": 18816 + }, + { + "epoch": 1.0513450847836412, + "grad_norm": 0.3409149646759033, + "learning_rate": 0.00047615979381443295, + "loss": 0.393, + "step": 18817 + }, + { + "epoch": 1.051400955387323, + "grad_norm": 0.46733441948890686, + "learning_rate": 0.0004761317794710892, + "loss": 0.4733, + "step": 18818 + }, + { + "epoch": 1.0514568259910049, + "grad_norm": 0.5265946388244629, + "learning_rate": 0.0004761037651277454, + "loss": 0.4687, + "step": 18819 + }, + { + "epoch": 1.0515126965946866, + "grad_norm": 2.710669755935669, + "learning_rate": 0.0004760757507844016, + "loss": 0.4202, + "step": 18820 + }, + { + "epoch": 1.0515685671983686, + "grad_norm": 0.48173174262046814, + "learning_rate": 0.0004760477364410578, + "loss": 0.3378, + "step": 18821 + }, + { + "epoch": 1.0516244378020505, + "grad_norm": 0.5201056003570557, + "learning_rate": 0.00047601972209771403, + "loss": 0.3507, + "step": 18822 + }, + { + "epoch": 1.0516803084057322, + "grad_norm": 0.5970332622528076, + "learning_rate": 0.00047599170775437024, + "loss": 0.6164, + "step": 18823 + }, + { + "epoch": 1.0517361790094142, + "grad_norm": 0.4298827052116394, + "learning_rate": 0.00047596369341102644, + "loss": 0.4732, + "step": 18824 + }, + { + "epoch": 1.0517920496130961, + "grad_norm": 0.4927917420864105, + "learning_rate": 0.00047593567906768265, + "loss": 0.4818, + "step": 18825 + }, + { + "epoch": 1.0518479202167779, + "grad_norm": 0.4428502917289734, + "learning_rate": 0.0004759076647243389, + "loss": 0.3592, + "step": 18826 + }, + { + "epoch": 1.0519037908204598, + "grad_norm": 0.7147983312606812, + "learning_rate": 0.00047587965038099506, + "loss": 0.4009, + "step": 18827 + }, + { + "epoch": 1.0519596614241418, + "grad_norm": 0.49417340755462646, + "learning_rate": 0.0004758516360376513, + "loss": 0.3757, + "step": 18828 + }, + { + "epoch": 1.0520155320278235, + "grad_norm": 0.6322794556617737, + "learning_rate": 0.00047582362169430747, + "loss": 0.4481, + "step": 18829 + }, + { + "epoch": 1.0520714026315054, + "grad_norm": 9.044120788574219, + "learning_rate": 0.00047579560735096373, + "loss": 0.3976, + "step": 18830 + }, + { + "epoch": 1.0521272732351874, + "grad_norm": 0.47033485770225525, + "learning_rate": 0.0004757675930076199, + "loss": 0.4401, + "step": 18831 + }, + { + "epoch": 1.0521831438388691, + "grad_norm": 0.3825990557670593, + "learning_rate": 0.00047573957866427614, + "loss": 0.412, + "step": 18832 + }, + { + "epoch": 1.052239014442551, + "grad_norm": 0.5271837115287781, + "learning_rate": 0.0004757115643209323, + "loss": 0.4637, + "step": 18833 + }, + { + "epoch": 1.052294885046233, + "grad_norm": 0.7323696613311768, + "learning_rate": 0.00047568354997758856, + "loss": 0.503, + "step": 18834 + }, + { + "epoch": 1.0523507556499148, + "grad_norm": 0.3949071764945984, + "learning_rate": 0.00047565553563424476, + "loss": 0.4252, + "step": 18835 + }, + { + "epoch": 1.0524066262535967, + "grad_norm": 0.49573108553886414, + "learning_rate": 0.00047562752129090097, + "loss": 0.4207, + "step": 18836 + }, + { + "epoch": 1.0524624968572784, + "grad_norm": 1.4940191507339478, + "learning_rate": 0.0004755995069475572, + "loss": 0.3923, + "step": 18837 + }, + { + "epoch": 1.0525183674609604, + "grad_norm": 0.4679076671600342, + "learning_rate": 0.0004755714926042133, + "loss": 0.3969, + "step": 18838 + }, + { + "epoch": 1.0525742380646423, + "grad_norm": 0.47445276379585266, + "learning_rate": 0.0004755434782608696, + "loss": 0.4552, + "step": 18839 + }, + { + "epoch": 1.052630108668324, + "grad_norm": 2.5238921642303467, + "learning_rate": 0.00047551546391752574, + "loss": 0.3684, + "step": 18840 + }, + { + "epoch": 1.052685979272006, + "grad_norm": 0.7207733988761902, + "learning_rate": 0.000475487449574182, + "loss": 0.4098, + "step": 18841 + }, + { + "epoch": 1.052741849875688, + "grad_norm": 0.49211063981056213, + "learning_rate": 0.0004754594352308382, + "loss": 0.4857, + "step": 18842 + }, + { + "epoch": 1.0527977204793697, + "grad_norm": 0.3836842477321625, + "learning_rate": 0.0004754314208874944, + "loss": 0.3933, + "step": 18843 + }, + { + "epoch": 1.0528535910830517, + "grad_norm": 1.2818759679794312, + "learning_rate": 0.0004754034065441506, + "loss": 0.3859, + "step": 18844 + }, + { + "epoch": 1.0529094616867336, + "grad_norm": 0.48059526085853577, + "learning_rate": 0.0004753753922008068, + "loss": 0.4902, + "step": 18845 + }, + { + "epoch": 1.0529653322904153, + "grad_norm": 0.5000432729721069, + "learning_rate": 0.000475347377857463, + "loss": 0.5008, + "step": 18846 + }, + { + "epoch": 1.0530212028940973, + "grad_norm": 0.5768634080886841, + "learning_rate": 0.00047531936351411923, + "loss": 0.4119, + "step": 18847 + }, + { + "epoch": 1.0530770734977792, + "grad_norm": 0.5414196848869324, + "learning_rate": 0.00047529134917077544, + "loss": 0.4069, + "step": 18848 + }, + { + "epoch": 1.053132944101461, + "grad_norm": 0.4136102795600891, + "learning_rate": 0.0004752633348274317, + "loss": 0.4169, + "step": 18849 + }, + { + "epoch": 1.053188814705143, + "grad_norm": 0.44429922103881836, + "learning_rate": 0.00047523532048408785, + "loss": 0.4464, + "step": 18850 + }, + { + "epoch": 1.0532446853088249, + "grad_norm": 0.4036077558994293, + "learning_rate": 0.0004752073061407441, + "loss": 0.4551, + "step": 18851 + }, + { + "epoch": 1.0533005559125066, + "grad_norm": 0.4114793837070465, + "learning_rate": 0.00047517929179740026, + "loss": 0.5108, + "step": 18852 + }, + { + "epoch": 1.0533564265161885, + "grad_norm": 0.3573899269104004, + "learning_rate": 0.0004751512774540565, + "loss": 0.3274, + "step": 18853 + }, + { + "epoch": 1.0534122971198703, + "grad_norm": 0.7139743566513062, + "learning_rate": 0.00047512326311071267, + "loss": 0.3851, + "step": 18854 + }, + { + "epoch": 1.0534681677235522, + "grad_norm": 0.43691307306289673, + "learning_rate": 0.00047509524876736893, + "loss": 0.467, + "step": 18855 + }, + { + "epoch": 1.0535240383272342, + "grad_norm": 0.4550175070762634, + "learning_rate": 0.0004750672344240251, + "loss": 0.3603, + "step": 18856 + }, + { + "epoch": 1.053579908930916, + "grad_norm": 0.5113556385040283, + "learning_rate": 0.00047503922008068134, + "loss": 0.4471, + "step": 18857 + }, + { + "epoch": 1.0536357795345979, + "grad_norm": 1.6291159391403198, + "learning_rate": 0.00047501120573733755, + "loss": 0.5024, + "step": 18858 + }, + { + "epoch": 1.0536916501382798, + "grad_norm": 0.32302534580230713, + "learning_rate": 0.0004749831913939937, + "loss": 0.3826, + "step": 18859 + }, + { + "epoch": 1.0537475207419615, + "grad_norm": 0.7058364152908325, + "learning_rate": 0.00047495517705064996, + "loss": 0.395, + "step": 18860 + }, + { + "epoch": 1.0538033913456435, + "grad_norm": 0.4765687584877014, + "learning_rate": 0.0004749271627073061, + "loss": 0.3621, + "step": 18861 + }, + { + "epoch": 1.0538592619493254, + "grad_norm": 1.1215077638626099, + "learning_rate": 0.00047489914836396237, + "loss": 0.5657, + "step": 18862 + }, + { + "epoch": 1.0539151325530072, + "grad_norm": 0.4911588728427887, + "learning_rate": 0.0004748711340206185, + "loss": 0.4097, + "step": 18863 + }, + { + "epoch": 1.0539710031566891, + "grad_norm": 0.6239192485809326, + "learning_rate": 0.0004748431196772748, + "loss": 0.4697, + "step": 18864 + }, + { + "epoch": 1.054026873760371, + "grad_norm": 3.070132255554199, + "learning_rate": 0.000474815105333931, + "loss": 0.4621, + "step": 18865 + }, + { + "epoch": 1.0540827443640528, + "grad_norm": 0.5816750526428223, + "learning_rate": 0.0004747870909905872, + "loss": 0.4542, + "step": 18866 + }, + { + "epoch": 1.0541386149677348, + "grad_norm": 0.36892515420913696, + "learning_rate": 0.0004747590766472434, + "loss": 0.5031, + "step": 18867 + }, + { + "epoch": 1.0541944855714167, + "grad_norm": 0.48201608657836914, + "learning_rate": 0.0004747310623038996, + "loss": 0.397, + "step": 18868 + }, + { + "epoch": 1.0542503561750984, + "grad_norm": 0.3732408583164215, + "learning_rate": 0.0004747030479605558, + "loss": 0.453, + "step": 18869 + }, + { + "epoch": 1.0543062267787804, + "grad_norm": 0.4507982134819031, + "learning_rate": 0.000474675033617212, + "loss": 0.4244, + "step": 18870 + }, + { + "epoch": 1.0543620973824621, + "grad_norm": 1.3560519218444824, + "learning_rate": 0.0004746470192738682, + "loss": 0.5405, + "step": 18871 + }, + { + "epoch": 1.054417967986144, + "grad_norm": 4.364433288574219, + "learning_rate": 0.00047461900493052443, + "loss": 1.5126, + "step": 18872 + }, + { + "epoch": 1.054473838589826, + "grad_norm": 2.731079578399658, + "learning_rate": 0.00047459099058718064, + "loss": 0.3523, + "step": 18873 + }, + { + "epoch": 1.0545297091935077, + "grad_norm": 0.45923012495040894, + "learning_rate": 0.0004745629762438369, + "loss": 0.3937, + "step": 18874 + }, + { + "epoch": 1.0545855797971897, + "grad_norm": 0.8828878402709961, + "learning_rate": 0.00047453496190049305, + "loss": 0.4313, + "step": 18875 + }, + { + "epoch": 1.0546414504008716, + "grad_norm": 1.253139853477478, + "learning_rate": 0.0004745069475571493, + "loss": 0.4628, + "step": 18876 + }, + { + "epoch": 1.0546973210045534, + "grad_norm": 0.8594611883163452, + "learning_rate": 0.00047447893321380546, + "loss": 0.4436, + "step": 18877 + }, + { + "epoch": 1.0547531916082353, + "grad_norm": 1.2789102792739868, + "learning_rate": 0.0004744509188704617, + "loss": 0.4912, + "step": 18878 + }, + { + "epoch": 1.0548090622119173, + "grad_norm": 0.5054759979248047, + "learning_rate": 0.00047442290452711787, + "loss": 0.4468, + "step": 18879 + }, + { + "epoch": 1.054864932815599, + "grad_norm": 3.3459315299987793, + "learning_rate": 0.0004743948901837741, + "loss": 0.4098, + "step": 18880 + }, + { + "epoch": 1.054920803419281, + "grad_norm": 0.3827361464500427, + "learning_rate": 0.00047436687584043034, + "loss": 0.3547, + "step": 18881 + }, + { + "epoch": 1.054976674022963, + "grad_norm": 0.32320305705070496, + "learning_rate": 0.0004743388614970865, + "loss": 0.4104, + "step": 18882 + }, + { + "epoch": 1.0550325446266446, + "grad_norm": 5.16923189163208, + "learning_rate": 0.00047431084715374275, + "loss": 0.387, + "step": 18883 + }, + { + "epoch": 1.0550884152303266, + "grad_norm": 0.5752010941505432, + "learning_rate": 0.0004742828328103989, + "loss": 0.3794, + "step": 18884 + }, + { + "epoch": 1.0551442858340085, + "grad_norm": 0.40869176387786865, + "learning_rate": 0.00047425481846705516, + "loss": 0.4643, + "step": 18885 + }, + { + "epoch": 1.0552001564376903, + "grad_norm": 5.163275241851807, + "learning_rate": 0.0004742268041237113, + "loss": 0.4967, + "step": 18886 + }, + { + "epoch": 1.0552560270413722, + "grad_norm": 2.5970730781555176, + "learning_rate": 0.00047419878978036757, + "loss": 0.401, + "step": 18887 + }, + { + "epoch": 1.055311897645054, + "grad_norm": 0.42739418148994446, + "learning_rate": 0.0004741707754370237, + "loss": 0.4572, + "step": 18888 + }, + { + "epoch": 1.055367768248736, + "grad_norm": 0.6670829653739929, + "learning_rate": 0.00047414276109368, + "loss": 0.3649, + "step": 18889 + }, + { + "epoch": 1.0554236388524179, + "grad_norm": 0.544702410697937, + "learning_rate": 0.0004741147467503362, + "loss": 0.4113, + "step": 18890 + }, + { + "epoch": 1.0554795094560996, + "grad_norm": 0.3624964654445648, + "learning_rate": 0.0004740867324069924, + "loss": 0.362, + "step": 18891 + }, + { + "epoch": 1.0555353800597815, + "grad_norm": 0.7064487934112549, + "learning_rate": 0.0004740587180636486, + "loss": 0.4619, + "step": 18892 + }, + { + "epoch": 1.0555912506634635, + "grad_norm": 0.9215566515922546, + "learning_rate": 0.0004740307037203048, + "loss": 0.3922, + "step": 18893 + }, + { + "epoch": 1.0556471212671452, + "grad_norm": 0.6216382384300232, + "learning_rate": 0.000474002689376961, + "loss": 0.4344, + "step": 18894 + }, + { + "epoch": 1.0557029918708272, + "grad_norm": 1.1207354068756104, + "learning_rate": 0.0004739746750336172, + "loss": 0.5112, + "step": 18895 + }, + { + "epoch": 1.0557588624745091, + "grad_norm": 0.4004688560962677, + "learning_rate": 0.0004739466606902734, + "loss": 0.4603, + "step": 18896 + }, + { + "epoch": 1.0558147330781908, + "grad_norm": 0.5643457770347595, + "learning_rate": 0.0004739186463469297, + "loss": 0.5885, + "step": 18897 + }, + { + "epoch": 1.0558706036818728, + "grad_norm": 0.4283551871776581, + "learning_rate": 0.00047389063200358584, + "loss": 0.4723, + "step": 18898 + }, + { + "epoch": 1.0559264742855548, + "grad_norm": 0.4791354238986969, + "learning_rate": 0.0004738626176602421, + "loss": 0.3974, + "step": 18899 + }, + { + "epoch": 1.0559823448892365, + "grad_norm": 0.4278687536716461, + "learning_rate": 0.00047383460331689825, + "loss": 0.504, + "step": 18900 + }, + { + "epoch": 1.0560382154929184, + "grad_norm": 0.5316165685653687, + "learning_rate": 0.00047380658897355445, + "loss": 0.4491, + "step": 18901 + }, + { + "epoch": 1.0560940860966004, + "grad_norm": 0.5350983142852783, + "learning_rate": 0.00047377857463021066, + "loss": 0.3704, + "step": 18902 + }, + { + "epoch": 1.056149956700282, + "grad_norm": 0.5240693092346191, + "learning_rate": 0.00047375056028686687, + "loss": 0.4991, + "step": 18903 + }, + { + "epoch": 1.056205827303964, + "grad_norm": 1.1815378665924072, + "learning_rate": 0.00047372254594352307, + "loss": 0.5141, + "step": 18904 + }, + { + "epoch": 1.0562616979076458, + "grad_norm": 0.36245599389076233, + "learning_rate": 0.0004736945316001793, + "loss": 0.3948, + "step": 18905 + }, + { + "epoch": 1.0563175685113277, + "grad_norm": 0.5495826005935669, + "learning_rate": 0.00047366651725683554, + "loss": 0.4483, + "step": 18906 + }, + { + "epoch": 1.0563734391150097, + "grad_norm": 0.5938689708709717, + "learning_rate": 0.0004736385029134917, + "loss": 0.4166, + "step": 18907 + }, + { + "epoch": 1.0564293097186914, + "grad_norm": 0.3367507755756378, + "learning_rate": 0.00047361048857014795, + "loss": 0.4293, + "step": 18908 + }, + { + "epoch": 1.0564851803223734, + "grad_norm": 0.4854956269264221, + "learning_rate": 0.0004735824742268041, + "loss": 0.4481, + "step": 18909 + }, + { + "epoch": 1.0565410509260553, + "grad_norm": 0.3690897524356842, + "learning_rate": 0.00047355445988346036, + "loss": 0.4483, + "step": 18910 + }, + { + "epoch": 1.056596921529737, + "grad_norm": 0.6562878489494324, + "learning_rate": 0.0004735264455401165, + "loss": 0.4805, + "step": 18911 + }, + { + "epoch": 1.056652792133419, + "grad_norm": 0.6709489226341248, + "learning_rate": 0.00047349843119677277, + "loss": 0.3404, + "step": 18912 + }, + { + "epoch": 1.056708662737101, + "grad_norm": 0.5013763904571533, + "learning_rate": 0.000473470416853429, + "loss": 0.4237, + "step": 18913 + }, + { + "epoch": 1.0567645333407827, + "grad_norm": 0.5777496695518494, + "learning_rate": 0.0004734424025100852, + "loss": 0.3907, + "step": 18914 + }, + { + "epoch": 1.0568204039444646, + "grad_norm": 0.6634140014648438, + "learning_rate": 0.0004734143881667414, + "loss": 0.3717, + "step": 18915 + }, + { + "epoch": 1.0568762745481466, + "grad_norm": 0.5094257593154907, + "learning_rate": 0.0004733863738233976, + "loss": 0.4478, + "step": 18916 + }, + { + "epoch": 1.0569321451518283, + "grad_norm": 0.45782479643821716, + "learning_rate": 0.0004733583594800538, + "loss": 0.5134, + "step": 18917 + }, + { + "epoch": 1.0569880157555103, + "grad_norm": 0.46013346314430237, + "learning_rate": 0.00047333034513671, + "loss": 0.3941, + "step": 18918 + }, + { + "epoch": 1.057043886359192, + "grad_norm": 1.8129990100860596, + "learning_rate": 0.0004733023307933662, + "loss": 0.468, + "step": 18919 + }, + { + "epoch": 1.057099756962874, + "grad_norm": 0.37430068850517273, + "learning_rate": 0.00047327431645002247, + "loss": 0.3658, + "step": 18920 + }, + { + "epoch": 1.057155627566556, + "grad_norm": 0.603175938129425, + "learning_rate": 0.0004732463021066786, + "loss": 0.3934, + "step": 18921 + }, + { + "epoch": 1.0572114981702376, + "grad_norm": 0.4523930847644806, + "learning_rate": 0.0004732182877633349, + "loss": 0.3889, + "step": 18922 + }, + { + "epoch": 1.0572673687739196, + "grad_norm": 0.4330889582633972, + "learning_rate": 0.00047319027341999104, + "loss": 0.4351, + "step": 18923 + }, + { + "epoch": 1.0573232393776015, + "grad_norm": 0.40877968072891235, + "learning_rate": 0.00047316225907664724, + "loss": 0.3694, + "step": 18924 + }, + { + "epoch": 1.0573791099812833, + "grad_norm": 0.4389800429344177, + "learning_rate": 0.00047313424473330345, + "loss": 0.4025, + "step": 18925 + }, + { + "epoch": 1.0574349805849652, + "grad_norm": 0.33923691511154175, + "learning_rate": 0.00047310623038995965, + "loss": 0.3803, + "step": 18926 + }, + { + "epoch": 1.0574908511886472, + "grad_norm": 0.3571331799030304, + "learning_rate": 0.00047307821604661586, + "loss": 0.3759, + "step": 18927 + }, + { + "epoch": 1.057546721792329, + "grad_norm": 2.212381601333618, + "learning_rate": 0.00047305020170327206, + "loss": 0.4153, + "step": 18928 + }, + { + "epoch": 1.0576025923960108, + "grad_norm": 0.5466653108596802, + "learning_rate": 0.0004730221873599283, + "loss": 0.3455, + "step": 18929 + }, + { + "epoch": 1.0576584629996928, + "grad_norm": 0.41525083780288696, + "learning_rate": 0.0004729941730165845, + "loss": 0.4585, + "step": 18930 + }, + { + "epoch": 1.0577143336033745, + "grad_norm": 0.4005703330039978, + "learning_rate": 0.00047296615867324074, + "loss": 0.4414, + "step": 18931 + }, + { + "epoch": 1.0577702042070565, + "grad_norm": 0.43249085545539856, + "learning_rate": 0.0004729381443298969, + "loss": 0.4597, + "step": 18932 + }, + { + "epoch": 1.0578260748107384, + "grad_norm": 0.49422451853752136, + "learning_rate": 0.00047291012998655315, + "loss": 0.4652, + "step": 18933 + }, + { + "epoch": 1.0578819454144202, + "grad_norm": 0.8420789241790771, + "learning_rate": 0.0004728821156432093, + "loss": 0.4074, + "step": 18934 + }, + { + "epoch": 1.057937816018102, + "grad_norm": 0.4420829713344574, + "learning_rate": 0.00047285410129986556, + "loss": 0.4294, + "step": 18935 + }, + { + "epoch": 1.057993686621784, + "grad_norm": 0.4528072476387024, + "learning_rate": 0.00047282608695652177, + "loss": 0.3419, + "step": 18936 + }, + { + "epoch": 1.0580495572254658, + "grad_norm": 0.5550611019134521, + "learning_rate": 0.00047279807261317797, + "loss": 0.5226, + "step": 18937 + }, + { + "epoch": 1.0581054278291477, + "grad_norm": 0.6153182983398438, + "learning_rate": 0.0004727700582698342, + "loss": 0.3664, + "step": 18938 + }, + { + "epoch": 1.0581612984328295, + "grad_norm": 0.367912620306015, + "learning_rate": 0.0004727420439264904, + "loss": 0.3141, + "step": 18939 + }, + { + "epoch": 1.0582171690365114, + "grad_norm": 0.5046834945678711, + "learning_rate": 0.0004727140295831466, + "loss": 0.4605, + "step": 18940 + }, + { + "epoch": 1.0582730396401934, + "grad_norm": 0.7094347476959229, + "learning_rate": 0.0004726860152398028, + "loss": 0.4797, + "step": 18941 + }, + { + "epoch": 1.058328910243875, + "grad_norm": 0.5954566597938538, + "learning_rate": 0.000472658000896459, + "loss": 0.4426, + "step": 18942 + }, + { + "epoch": 1.058384780847557, + "grad_norm": 1.2859909534454346, + "learning_rate": 0.0004726299865531152, + "loss": 0.524, + "step": 18943 + }, + { + "epoch": 1.058440651451239, + "grad_norm": 0.6151472330093384, + "learning_rate": 0.0004726019722097714, + "loss": 0.4745, + "step": 18944 + }, + { + "epoch": 1.0584965220549207, + "grad_norm": 0.4627450108528137, + "learning_rate": 0.0004725739578664276, + "loss": 0.5074, + "step": 18945 + }, + { + "epoch": 1.0585523926586027, + "grad_norm": 0.6445800065994263, + "learning_rate": 0.0004725459435230838, + "loss": 0.415, + "step": 18946 + }, + { + "epoch": 1.0586082632622846, + "grad_norm": 0.44622236490249634, + "learning_rate": 0.00047251792917974003, + "loss": 0.5144, + "step": 18947 + }, + { + "epoch": 1.0586641338659664, + "grad_norm": 0.44077906012535095, + "learning_rate": 0.00047248991483639624, + "loss": 0.3564, + "step": 18948 + }, + { + "epoch": 1.0587200044696483, + "grad_norm": 0.40648967027664185, + "learning_rate": 0.00047246190049305244, + "loss": 0.3889, + "step": 18949 + }, + { + "epoch": 1.0587758750733303, + "grad_norm": 0.35695183277130127, + "learning_rate": 0.00047243388614970865, + "loss": 0.417, + "step": 18950 + }, + { + "epoch": 1.058831745677012, + "grad_norm": 0.48829320073127747, + "learning_rate": 0.00047240587180636485, + "loss": 0.3734, + "step": 18951 + }, + { + "epoch": 1.058887616280694, + "grad_norm": 0.4388054311275482, + "learning_rate": 0.0004723778574630211, + "loss": 0.3359, + "step": 18952 + }, + { + "epoch": 1.0589434868843757, + "grad_norm": 3.711817741394043, + "learning_rate": 0.00047234984311967726, + "loss": 0.5107, + "step": 18953 + }, + { + "epoch": 1.0589993574880576, + "grad_norm": 0.4133788049221039, + "learning_rate": 0.0004723218287763335, + "loss": 0.4806, + "step": 18954 + }, + { + "epoch": 1.0590552280917396, + "grad_norm": 1.0148907899856567, + "learning_rate": 0.0004722938144329897, + "loss": 0.3351, + "step": 18955 + }, + { + "epoch": 1.0591110986954213, + "grad_norm": 0.4966649115085602, + "learning_rate": 0.00047226580008964594, + "loss": 0.41, + "step": 18956 + }, + { + "epoch": 1.0591669692991033, + "grad_norm": 0.5871896743774414, + "learning_rate": 0.0004722377857463021, + "loss": 0.4047, + "step": 18957 + }, + { + "epoch": 1.0592228399027852, + "grad_norm": 0.527472972869873, + "learning_rate": 0.00047220977140295835, + "loss": 0.3811, + "step": 18958 + }, + { + "epoch": 1.059278710506467, + "grad_norm": 0.43413761258125305, + "learning_rate": 0.0004721817570596145, + "loss": 0.3766, + "step": 18959 + }, + { + "epoch": 1.0593345811101489, + "grad_norm": 3.1049604415893555, + "learning_rate": 0.00047215374271627076, + "loss": 0.3332, + "step": 18960 + }, + { + "epoch": 1.0593904517138308, + "grad_norm": 0.4346156120300293, + "learning_rate": 0.00047212572837292697, + "loss": 0.4183, + "step": 18961 + }, + { + "epoch": 1.0594463223175126, + "grad_norm": 0.4224885404109955, + "learning_rate": 0.00047209771402958317, + "loss": 0.3492, + "step": 18962 + }, + { + "epoch": 1.0595021929211945, + "grad_norm": 0.5537341237068176, + "learning_rate": 0.0004720696996862394, + "loss": 0.4451, + "step": 18963 + }, + { + "epoch": 1.0595580635248765, + "grad_norm": 0.3047366738319397, + "learning_rate": 0.0004720416853428956, + "loss": 0.3992, + "step": 18964 + }, + { + "epoch": 1.0596139341285582, + "grad_norm": 0.4819236993789673, + "learning_rate": 0.0004720136709995518, + "loss": 0.4249, + "step": 18965 + }, + { + "epoch": 1.0596698047322402, + "grad_norm": 0.42456117272377014, + "learning_rate": 0.00047198565665620794, + "loss": 0.4611, + "step": 18966 + }, + { + "epoch": 1.059725675335922, + "grad_norm": 0.962077796459198, + "learning_rate": 0.0004719576423128642, + "loss": 0.4948, + "step": 18967 + }, + { + "epoch": 1.0597815459396038, + "grad_norm": 2.821742057800293, + "learning_rate": 0.0004719296279695204, + "loss": 0.4995, + "step": 18968 + }, + { + "epoch": 1.0598374165432858, + "grad_norm": 0.8279415965080261, + "learning_rate": 0.0004719016136261766, + "loss": 0.4617, + "step": 18969 + }, + { + "epoch": 1.0598932871469677, + "grad_norm": 0.36394205689430237, + "learning_rate": 0.0004718735992828328, + "loss": 0.3782, + "step": 18970 + }, + { + "epoch": 1.0599491577506495, + "grad_norm": 0.3483678102493286, + "learning_rate": 0.000471845584939489, + "loss": 0.3539, + "step": 18971 + }, + { + "epoch": 1.0600050283543314, + "grad_norm": 0.40406352281570435, + "learning_rate": 0.00047181757059614523, + "loss": 0.4389, + "step": 18972 + }, + { + "epoch": 1.0600608989580131, + "grad_norm": 0.6954027414321899, + "learning_rate": 0.00047178955625280143, + "loss": 0.5934, + "step": 18973 + }, + { + "epoch": 1.060116769561695, + "grad_norm": 0.6551207304000854, + "learning_rate": 0.00047176154190945764, + "loss": 0.4759, + "step": 18974 + }, + { + "epoch": 1.060172640165377, + "grad_norm": 0.4315655529499054, + "learning_rate": 0.0004717335275661139, + "loss": 0.362, + "step": 18975 + }, + { + "epoch": 1.0602285107690588, + "grad_norm": 0.3813748359680176, + "learning_rate": 0.00047170551322277005, + "loss": 0.329, + "step": 18976 + }, + { + "epoch": 1.0602843813727407, + "grad_norm": 0.36004647612571716, + "learning_rate": 0.0004716774988794263, + "loss": 0.2935, + "step": 18977 + }, + { + "epoch": 1.0603402519764227, + "grad_norm": 0.9714619517326355, + "learning_rate": 0.00047164948453608246, + "loss": 0.4835, + "step": 18978 + }, + { + "epoch": 1.0603961225801044, + "grad_norm": 0.6757155656814575, + "learning_rate": 0.0004716214701927387, + "loss": 0.4049, + "step": 18979 + }, + { + "epoch": 1.0604519931837864, + "grad_norm": 0.46617642045021057, + "learning_rate": 0.0004715934558493949, + "loss": 0.4723, + "step": 18980 + }, + { + "epoch": 1.0605078637874683, + "grad_norm": 0.9406777620315552, + "learning_rate": 0.00047156544150605114, + "loss": 0.4134, + "step": 18981 + }, + { + "epoch": 1.06056373439115, + "grad_norm": 0.5439208149909973, + "learning_rate": 0.0004715374271627073, + "loss": 0.5274, + "step": 18982 + }, + { + "epoch": 1.060619604994832, + "grad_norm": 22.360485076904297, + "learning_rate": 0.00047150941281936355, + "loss": 0.4184, + "step": 18983 + }, + { + "epoch": 1.060675475598514, + "grad_norm": 0.6239106059074402, + "learning_rate": 0.00047148139847601975, + "loss": 0.4521, + "step": 18984 + }, + { + "epoch": 1.0607313462021957, + "grad_norm": 2.7091729640960693, + "learning_rate": 0.00047145338413267596, + "loss": 0.4432, + "step": 18985 + }, + { + "epoch": 1.0607872168058776, + "grad_norm": 0.43604496121406555, + "learning_rate": 0.00047142536978933216, + "loss": 0.5963, + "step": 18986 + }, + { + "epoch": 1.0608430874095593, + "grad_norm": 0.6625362038612366, + "learning_rate": 0.0004713973554459883, + "loss": 0.6199, + "step": 18987 + }, + { + "epoch": 1.0608989580132413, + "grad_norm": 0.504162609577179, + "learning_rate": 0.0004713693411026446, + "loss": 0.4153, + "step": 18988 + }, + { + "epoch": 1.0609548286169233, + "grad_norm": 0.5316917300224304, + "learning_rate": 0.00047134132675930073, + "loss": 0.4481, + "step": 18989 + }, + { + "epoch": 1.061010699220605, + "grad_norm": 0.6578286290168762, + "learning_rate": 0.000471313312415957, + "loss": 0.4502, + "step": 18990 + }, + { + "epoch": 1.061066569824287, + "grad_norm": 0.7683237195014954, + "learning_rate": 0.0004712852980726132, + "loss": 0.4497, + "step": 18991 + }, + { + "epoch": 1.0611224404279689, + "grad_norm": 0.8077331781387329, + "learning_rate": 0.0004712572837292694, + "loss": 0.5582, + "step": 18992 + }, + { + "epoch": 1.0611783110316506, + "grad_norm": 0.6959805488586426, + "learning_rate": 0.0004712292693859256, + "loss": 0.5205, + "step": 18993 + }, + { + "epoch": 1.0612341816353326, + "grad_norm": 0.8733076453208923, + "learning_rate": 0.0004712012550425818, + "loss": 0.4563, + "step": 18994 + }, + { + "epoch": 1.0612900522390145, + "grad_norm": 0.4732275903224945, + "learning_rate": 0.000471173240699238, + "loss": 0.4147, + "step": 18995 + }, + { + "epoch": 1.0613459228426962, + "grad_norm": 0.6825861930847168, + "learning_rate": 0.0004711452263558942, + "loss": 0.4645, + "step": 18996 + }, + { + "epoch": 1.0614017934463782, + "grad_norm": 0.43150895833969116, + "learning_rate": 0.00047111721201255043, + "loss": 0.4866, + "step": 18997 + }, + { + "epoch": 1.0614576640500601, + "grad_norm": 1.4568625688552856, + "learning_rate": 0.00047108919766920663, + "loss": 0.5029, + "step": 18998 + }, + { + "epoch": 1.0615135346537419, + "grad_norm": 0.5000104308128357, + "learning_rate": 0.00047106118332586284, + "loss": 0.3418, + "step": 18999 + }, + { + "epoch": 1.0615694052574238, + "grad_norm": 0.45157113671302795, + "learning_rate": 0.0004710331689825191, + "loss": 0.5366, + "step": 19000 + }, + { + "epoch": 1.0615694052574238, + "eval_cer": 0.08906964311042738, + "eval_loss": 0.32922545075416565, + "eval_runtime": 56.5921, + "eval_samples_per_second": 80.188, + "eval_steps_per_second": 5.018, + "eval_wer": 0.3518896120220106, + "step": 19000 + }, + { + "epoch": 1.0616252758611058, + "grad_norm": 0.3632162809371948, + "learning_rate": 0.00047100515463917525, + "loss": 0.3509, + "step": 19001 + }, + { + "epoch": 1.0616811464647875, + "grad_norm": 0.6629639267921448, + "learning_rate": 0.0004709771402958315, + "loss": 0.4323, + "step": 19002 + }, + { + "epoch": 1.0617370170684695, + "grad_norm": 0.7091209292411804, + "learning_rate": 0.00047094912595248766, + "loss": 0.4944, + "step": 19003 + }, + { + "epoch": 1.0617928876721514, + "grad_norm": 0.449959397315979, + "learning_rate": 0.0004709211116091439, + "loss": 0.4847, + "step": 19004 + }, + { + "epoch": 1.0618487582758331, + "grad_norm": 0.5430222153663635, + "learning_rate": 0.0004708930972658001, + "loss": 0.6581, + "step": 19005 + }, + { + "epoch": 1.061904628879515, + "grad_norm": 0.4418654143810272, + "learning_rate": 0.00047086508292245634, + "loss": 0.4416, + "step": 19006 + }, + { + "epoch": 1.0619604994831968, + "grad_norm": 1.7632747888565063, + "learning_rate": 0.00047083706857911254, + "loss": 0.5143, + "step": 19007 + }, + { + "epoch": 1.0620163700868788, + "grad_norm": 0.5772262215614319, + "learning_rate": 0.0004708090542357687, + "loss": 0.3923, + "step": 19008 + }, + { + "epoch": 1.0620722406905607, + "grad_norm": 1.4120577573776245, + "learning_rate": 0.00047078103989242495, + "loss": 0.4974, + "step": 19009 + }, + { + "epoch": 1.0621281112942424, + "grad_norm": 0.42404767870903015, + "learning_rate": 0.0004707530255490811, + "loss": 0.5023, + "step": 19010 + }, + { + "epoch": 1.0621839818979244, + "grad_norm": 0.5632519721984863, + "learning_rate": 0.00047072501120573736, + "loss": 0.4106, + "step": 19011 + }, + { + "epoch": 1.0622398525016064, + "grad_norm": 0.42158234119415283, + "learning_rate": 0.0004706969968623935, + "loss": 0.3633, + "step": 19012 + }, + { + "epoch": 1.062295723105288, + "grad_norm": 0.8381600975990295, + "learning_rate": 0.0004706689825190498, + "loss": 0.3691, + "step": 19013 + }, + { + "epoch": 1.06235159370897, + "grad_norm": 0.7270371913909912, + "learning_rate": 0.00047064096817570593, + "loss": 0.4929, + "step": 19014 + }, + { + "epoch": 1.062407464312652, + "grad_norm": 0.45198607444763184, + "learning_rate": 0.0004706129538323622, + "loss": 0.4649, + "step": 19015 + }, + { + "epoch": 1.0624633349163337, + "grad_norm": 0.4889337122440338, + "learning_rate": 0.0004705849394890184, + "loss": 0.4682, + "step": 19016 + }, + { + "epoch": 1.0625192055200157, + "grad_norm": 1.1254832744598389, + "learning_rate": 0.0004705569251456746, + "loss": 0.4263, + "step": 19017 + }, + { + "epoch": 1.0625750761236976, + "grad_norm": 0.5061700344085693, + "learning_rate": 0.0004705289108023308, + "loss": 0.3842, + "step": 19018 + }, + { + "epoch": 1.0626309467273793, + "grad_norm": 1.0202901363372803, + "learning_rate": 0.000470500896458987, + "loss": 0.3624, + "step": 19019 + }, + { + "epoch": 1.0626868173310613, + "grad_norm": 0.45950788259506226, + "learning_rate": 0.0004704728821156432, + "loss": 0.4598, + "step": 19020 + }, + { + "epoch": 1.062742687934743, + "grad_norm": 2.230480194091797, + "learning_rate": 0.0004704448677722994, + "loss": 0.4119, + "step": 19021 + }, + { + "epoch": 1.062798558538425, + "grad_norm": 0.8773832321166992, + "learning_rate": 0.00047041685342895563, + "loss": 0.3493, + "step": 19022 + }, + { + "epoch": 1.062854429142107, + "grad_norm": 0.2728404700756073, + "learning_rate": 0.0004703888390856119, + "loss": 0.3067, + "step": 19023 + }, + { + "epoch": 1.0629102997457887, + "grad_norm": 0.427693635225296, + "learning_rate": 0.00047036082474226804, + "loss": 0.4752, + "step": 19024 + }, + { + "epoch": 1.0629661703494706, + "grad_norm": 0.44622474908828735, + "learning_rate": 0.0004703328103989243, + "loss": 0.4277, + "step": 19025 + }, + { + "epoch": 1.0630220409531526, + "grad_norm": 0.37049826979637146, + "learning_rate": 0.00047030479605558045, + "loss": 0.4035, + "step": 19026 + }, + { + "epoch": 1.0630779115568343, + "grad_norm": 0.5607179403305054, + "learning_rate": 0.0004702767817122367, + "loss": 0.4727, + "step": 19027 + }, + { + "epoch": 1.0631337821605162, + "grad_norm": 0.39228180050849915, + "learning_rate": 0.00047024876736889286, + "loss": 0.3764, + "step": 19028 + }, + { + "epoch": 1.0631896527641982, + "grad_norm": 0.5572718381881714, + "learning_rate": 0.00047022075302554907, + "loss": 0.4871, + "step": 19029 + }, + { + "epoch": 1.06324552336788, + "grad_norm": 0.4839877784252167, + "learning_rate": 0.0004701927386822053, + "loss": 0.3686, + "step": 19030 + }, + { + "epoch": 1.0633013939715619, + "grad_norm": 0.39232704043388367, + "learning_rate": 0.0004701647243388615, + "loss": 0.3981, + "step": 19031 + }, + { + "epoch": 1.0633572645752438, + "grad_norm": 0.5128938555717468, + "learning_rate": 0.00047013670999551774, + "loss": 0.4915, + "step": 19032 + }, + { + "epoch": 1.0634131351789256, + "grad_norm": 0.5505936145782471, + "learning_rate": 0.0004701086956521739, + "loss": 0.4357, + "step": 19033 + }, + { + "epoch": 1.0634690057826075, + "grad_norm": 0.6115055084228516, + "learning_rate": 0.00047008068130883015, + "loss": 0.4632, + "step": 19034 + }, + { + "epoch": 1.0635248763862895, + "grad_norm": 0.686732292175293, + "learning_rate": 0.0004700526669654863, + "loss": 0.4361, + "step": 19035 + }, + { + "epoch": 1.0635807469899712, + "grad_norm": 0.5025471448898315, + "learning_rate": 0.00047002465262214256, + "loss": 0.4709, + "step": 19036 + }, + { + "epoch": 1.0636366175936531, + "grad_norm": 0.4747249484062195, + "learning_rate": 0.0004699966382787987, + "loss": 0.4933, + "step": 19037 + }, + { + "epoch": 1.063692488197335, + "grad_norm": 1.628391981124878, + "learning_rate": 0.000469968623935455, + "loss": 0.4796, + "step": 19038 + }, + { + "epoch": 1.0637483588010168, + "grad_norm": 0.4161963164806366, + "learning_rate": 0.0004699406095921112, + "loss": 0.5474, + "step": 19039 + }, + { + "epoch": 1.0638042294046988, + "grad_norm": 0.537339985370636, + "learning_rate": 0.0004699125952487674, + "loss": 0.4298, + "step": 19040 + }, + { + "epoch": 1.0638601000083805, + "grad_norm": 0.5719856023788452, + "learning_rate": 0.0004698845809054236, + "loss": 0.3882, + "step": 19041 + }, + { + "epoch": 1.0639159706120624, + "grad_norm": 0.464463472366333, + "learning_rate": 0.0004698565665620798, + "loss": 0.4834, + "step": 19042 + }, + { + "epoch": 1.0639718412157444, + "grad_norm": 0.44188597798347473, + "learning_rate": 0.000469828552218736, + "loss": 0.4335, + "step": 19043 + }, + { + "epoch": 1.0640277118194261, + "grad_norm": 0.6571291089057922, + "learning_rate": 0.0004698005378753922, + "loss": 0.4457, + "step": 19044 + }, + { + "epoch": 1.064083582423108, + "grad_norm": 3.662980079650879, + "learning_rate": 0.0004697725235320484, + "loss": 0.3603, + "step": 19045 + }, + { + "epoch": 1.06413945302679, + "grad_norm": 0.6702707409858704, + "learning_rate": 0.0004697445091887047, + "loss": 0.4723, + "step": 19046 + }, + { + "epoch": 1.0641953236304718, + "grad_norm": 0.5668670535087585, + "learning_rate": 0.00046971649484536083, + "loss": 0.4578, + "step": 19047 + }, + { + "epoch": 1.0642511942341537, + "grad_norm": 0.4592207372188568, + "learning_rate": 0.0004696884805020171, + "loss": 0.4329, + "step": 19048 + }, + { + "epoch": 1.0643070648378357, + "grad_norm": 0.4785517752170563, + "learning_rate": 0.00046966046615867324, + "loss": 0.5123, + "step": 19049 + }, + { + "epoch": 1.0643629354415174, + "grad_norm": 0.7176587581634521, + "learning_rate": 0.00046963245181532945, + "loss": 0.4396, + "step": 19050 + }, + { + "epoch": 1.0644188060451993, + "grad_norm": 0.5749958157539368, + "learning_rate": 0.00046960443747198565, + "loss": 0.4282, + "step": 19051 + }, + { + "epoch": 1.0644746766488813, + "grad_norm": 0.48078131675720215, + "learning_rate": 0.00046957642312864186, + "loss": 0.51, + "step": 19052 + }, + { + "epoch": 1.064530547252563, + "grad_norm": 0.394440233707428, + "learning_rate": 0.00046954840878529806, + "loss": 0.3782, + "step": 19053 + }, + { + "epoch": 1.064586417856245, + "grad_norm": 0.4333800971508026, + "learning_rate": 0.00046952039444195427, + "loss": 0.4179, + "step": 19054 + }, + { + "epoch": 1.0646422884599267, + "grad_norm": 0.6772512197494507, + "learning_rate": 0.00046949238009861053, + "loss": 0.5218, + "step": 19055 + }, + { + "epoch": 1.0646981590636087, + "grad_norm": 0.4532347321510315, + "learning_rate": 0.0004694643657552667, + "loss": 0.4143, + "step": 19056 + }, + { + "epoch": 1.0647540296672906, + "grad_norm": 0.5239144563674927, + "learning_rate": 0.00046943635141192294, + "loss": 0.3479, + "step": 19057 + }, + { + "epoch": 1.0648099002709723, + "grad_norm": 3.839121103286743, + "learning_rate": 0.0004694083370685791, + "loss": 0.308, + "step": 19058 + }, + { + "epoch": 1.0648657708746543, + "grad_norm": 0.39629557728767395, + "learning_rate": 0.00046938032272523535, + "loss": 0.355, + "step": 19059 + }, + { + "epoch": 1.0649216414783362, + "grad_norm": 0.801535964012146, + "learning_rate": 0.0004693523083818915, + "loss": 0.4261, + "step": 19060 + }, + { + "epoch": 1.064977512082018, + "grad_norm": 0.7102916240692139, + "learning_rate": 0.00046932429403854776, + "loss": 0.4139, + "step": 19061 + }, + { + "epoch": 1.0650333826857, + "grad_norm": 0.45415282249450684, + "learning_rate": 0.00046929627969520397, + "loss": 0.3223, + "step": 19062 + }, + { + "epoch": 1.0650892532893819, + "grad_norm": 0.656502366065979, + "learning_rate": 0.0004692682653518602, + "loss": 0.4537, + "step": 19063 + }, + { + "epoch": 1.0651451238930636, + "grad_norm": 0.42250362038612366, + "learning_rate": 0.0004692402510085164, + "loss": 0.4233, + "step": 19064 + }, + { + "epoch": 1.0652009944967455, + "grad_norm": 0.4176709055900574, + "learning_rate": 0.0004692122366651726, + "loss": 0.3643, + "step": 19065 + }, + { + "epoch": 1.0652568651004275, + "grad_norm": 0.6078409552574158, + "learning_rate": 0.0004691842223218288, + "loss": 0.4158, + "step": 19066 + }, + { + "epoch": 1.0653127357041092, + "grad_norm": 0.47346171736717224, + "learning_rate": 0.000469156207978485, + "loss": 0.4131, + "step": 19067 + }, + { + "epoch": 1.0653686063077912, + "grad_norm": 0.6188529133796692, + "learning_rate": 0.0004691281936351412, + "loss": 0.4808, + "step": 19068 + }, + { + "epoch": 1.0654244769114731, + "grad_norm": 0.5338703989982605, + "learning_rate": 0.0004691001792917974, + "loss": 0.3877, + "step": 19069 + }, + { + "epoch": 1.0654803475151549, + "grad_norm": 0.6458069086074829, + "learning_rate": 0.0004690721649484536, + "loss": 0.4478, + "step": 19070 + }, + { + "epoch": 1.0655362181188368, + "grad_norm": 0.6658934950828552, + "learning_rate": 0.0004690441506051098, + "loss": 0.3837, + "step": 19071 + }, + { + "epoch": 1.0655920887225188, + "grad_norm": 0.3945033848285675, + "learning_rate": 0.00046901613626176603, + "loss": 0.4457, + "step": 19072 + }, + { + "epoch": 1.0656479593262005, + "grad_norm": 0.5888072848320007, + "learning_rate": 0.00046898812191842223, + "loss": 0.5675, + "step": 19073 + }, + { + "epoch": 1.0657038299298824, + "grad_norm": 0.4511852562427521, + "learning_rate": 0.00046896010757507844, + "loss": 0.4274, + "step": 19074 + }, + { + "epoch": 1.0657597005335642, + "grad_norm": 0.4402541220188141, + "learning_rate": 0.00046893209323173464, + "loss": 0.4082, + "step": 19075 + }, + { + "epoch": 1.0658155711372461, + "grad_norm": 1.0783482789993286, + "learning_rate": 0.00046890407888839085, + "loss": 0.3856, + "step": 19076 + }, + { + "epoch": 1.065871441740928, + "grad_norm": 0.31434863805770874, + "learning_rate": 0.00046887606454504706, + "loss": 0.3964, + "step": 19077 + }, + { + "epoch": 1.0659273123446098, + "grad_norm": 0.8519948720932007, + "learning_rate": 0.0004688480502017033, + "loss": 0.3664, + "step": 19078 + }, + { + "epoch": 1.0659831829482918, + "grad_norm": 0.536806583404541, + "learning_rate": 0.00046882003585835947, + "loss": 0.5102, + "step": 19079 + }, + { + "epoch": 1.0660390535519737, + "grad_norm": 1.5293291807174683, + "learning_rate": 0.00046879202151501573, + "loss": 0.4524, + "step": 19080 + }, + { + "epoch": 1.0660949241556554, + "grad_norm": 0.7288993000984192, + "learning_rate": 0.0004687640071716719, + "loss": 0.4815, + "step": 19081 + }, + { + "epoch": 1.0661507947593374, + "grad_norm": 0.7840500473976135, + "learning_rate": 0.00046873599282832814, + "loss": 0.6892, + "step": 19082 + }, + { + "epoch": 1.0662066653630193, + "grad_norm": 0.609002947807312, + "learning_rate": 0.0004687079784849843, + "loss": 0.4375, + "step": 19083 + }, + { + "epoch": 1.066262535966701, + "grad_norm": 0.5538574457168579, + "learning_rate": 0.00046867996414164055, + "loss": 0.3986, + "step": 19084 + }, + { + "epoch": 1.066318406570383, + "grad_norm": 0.722525417804718, + "learning_rate": 0.0004686519497982967, + "loss": 0.4428, + "step": 19085 + }, + { + "epoch": 1.066374277174065, + "grad_norm": 0.4917444884777069, + "learning_rate": 0.00046862393545495296, + "loss": 0.4359, + "step": 19086 + }, + { + "epoch": 1.0664301477777467, + "grad_norm": 0.7359322905540466, + "learning_rate": 0.00046859592111160917, + "loss": 0.4021, + "step": 19087 + }, + { + "epoch": 1.0664860183814286, + "grad_norm": 0.7320356965065002, + "learning_rate": 0.0004685679067682654, + "loss": 0.5165, + "step": 19088 + }, + { + "epoch": 1.0665418889851104, + "grad_norm": 0.5935863852500916, + "learning_rate": 0.0004685398924249216, + "loss": 0.522, + "step": 19089 + }, + { + "epoch": 1.0665977595887923, + "grad_norm": 0.4462428689002991, + "learning_rate": 0.0004685118780815778, + "loss": 0.5185, + "step": 19090 + }, + { + "epoch": 1.0666536301924743, + "grad_norm": 0.5540143251419067, + "learning_rate": 0.000468483863738234, + "loss": 0.3492, + "step": 19091 + }, + { + "epoch": 1.066709500796156, + "grad_norm": 0.5132013559341431, + "learning_rate": 0.0004684558493948902, + "loss": 0.4853, + "step": 19092 + }, + { + "epoch": 1.066765371399838, + "grad_norm": 0.9025536775588989, + "learning_rate": 0.0004684278350515464, + "loss": 0.5287, + "step": 19093 + }, + { + "epoch": 1.06682124200352, + "grad_norm": 0.5305168628692627, + "learning_rate": 0.0004683998207082026, + "loss": 0.4525, + "step": 19094 + }, + { + "epoch": 1.0668771126072016, + "grad_norm": 3.2670881748199463, + "learning_rate": 0.0004683718063648588, + "loss": 0.5151, + "step": 19095 + }, + { + "epoch": 1.0669329832108836, + "grad_norm": 1.065873146057129, + "learning_rate": 0.000468343792021515, + "loss": 0.4188, + "step": 19096 + }, + { + "epoch": 1.0669888538145655, + "grad_norm": 0.38083481788635254, + "learning_rate": 0.0004683157776781712, + "loss": 0.3692, + "step": 19097 + }, + { + "epoch": 1.0670447244182473, + "grad_norm": 0.3723178803920746, + "learning_rate": 0.00046828776333482743, + "loss": 0.417, + "step": 19098 + }, + { + "epoch": 1.0671005950219292, + "grad_norm": 0.9759877324104309, + "learning_rate": 0.00046825974899148364, + "loss": 0.345, + "step": 19099 + }, + { + "epoch": 1.0671564656256112, + "grad_norm": 0.5452614426612854, + "learning_rate": 0.00046823173464813984, + "loss": 0.4598, + "step": 19100 + }, + { + "epoch": 1.067212336229293, + "grad_norm": 0.5011054873466492, + "learning_rate": 0.00046820372030479605, + "loss": 0.4379, + "step": 19101 + }, + { + "epoch": 1.0672682068329749, + "grad_norm": 0.7086650729179382, + "learning_rate": 0.00046817570596145226, + "loss": 0.5196, + "step": 19102 + }, + { + "epoch": 1.0673240774366568, + "grad_norm": 2.8060805797576904, + "learning_rate": 0.0004681476916181085, + "loss": 0.4235, + "step": 19103 + }, + { + "epoch": 1.0673799480403385, + "grad_norm": 0.3339446783065796, + "learning_rate": 0.00046811967727476467, + "loss": 0.3536, + "step": 19104 + }, + { + "epoch": 1.0674358186440205, + "grad_norm": 0.5482609272003174, + "learning_rate": 0.00046809166293142093, + "loss": 0.3627, + "step": 19105 + }, + { + "epoch": 1.0674916892477024, + "grad_norm": 0.46309661865234375, + "learning_rate": 0.0004680636485880771, + "loss": 0.4158, + "step": 19106 + }, + { + "epoch": 1.0675475598513842, + "grad_norm": 0.37120679020881653, + "learning_rate": 0.00046803563424473334, + "loss": 0.3822, + "step": 19107 + }, + { + "epoch": 1.0676034304550661, + "grad_norm": 0.8327468037605286, + "learning_rate": 0.0004680076199013895, + "loss": 0.4385, + "step": 19108 + }, + { + "epoch": 1.0676593010587478, + "grad_norm": 0.591620922088623, + "learning_rate": 0.00046797960555804575, + "loss": 0.4284, + "step": 19109 + }, + { + "epoch": 1.0677151716624298, + "grad_norm": 0.8385320901870728, + "learning_rate": 0.00046795159121470196, + "loss": 0.4187, + "step": 19110 + }, + { + "epoch": 1.0677710422661117, + "grad_norm": 0.4993055462837219, + "learning_rate": 0.00046792357687135816, + "loss": 0.4383, + "step": 19111 + }, + { + "epoch": 1.0678269128697935, + "grad_norm": 0.8878640532493591, + "learning_rate": 0.00046789556252801437, + "loss": 0.4239, + "step": 19112 + }, + { + "epoch": 1.0678827834734754, + "grad_norm": 0.6735202670097351, + "learning_rate": 0.0004678675481846706, + "loss": 0.3725, + "step": 19113 + }, + { + "epoch": 1.0679386540771574, + "grad_norm": 0.6986836194992065, + "learning_rate": 0.0004678395338413268, + "loss": 0.4225, + "step": 19114 + }, + { + "epoch": 1.067994524680839, + "grad_norm": 0.3996610641479492, + "learning_rate": 0.00046781151949798293, + "loss": 0.4635, + "step": 19115 + }, + { + "epoch": 1.068050395284521, + "grad_norm": 0.5499026775360107, + "learning_rate": 0.0004677835051546392, + "loss": 0.3973, + "step": 19116 + }, + { + "epoch": 1.068106265888203, + "grad_norm": 1.769737958908081, + "learning_rate": 0.00046775549081129534, + "loss": 0.3645, + "step": 19117 + }, + { + "epoch": 1.0681621364918847, + "grad_norm": 0.49206283688545227, + "learning_rate": 0.0004677274764679516, + "loss": 0.4336, + "step": 19118 + }, + { + "epoch": 1.0682180070955667, + "grad_norm": 0.5063198208808899, + "learning_rate": 0.0004676994621246078, + "loss": 0.5132, + "step": 19119 + }, + { + "epoch": 1.0682738776992486, + "grad_norm": 0.5296167135238647, + "learning_rate": 0.000467671447781264, + "loss": 0.5258, + "step": 19120 + }, + { + "epoch": 1.0683297483029304, + "grad_norm": 0.460145503282547, + "learning_rate": 0.0004676434334379202, + "loss": 0.4503, + "step": 19121 + }, + { + "epoch": 1.0683856189066123, + "grad_norm": 0.7409366369247437, + "learning_rate": 0.0004676154190945764, + "loss": 0.4749, + "step": 19122 + }, + { + "epoch": 1.068441489510294, + "grad_norm": 0.3335546553134918, + "learning_rate": 0.00046758740475123263, + "loss": 0.4516, + "step": 19123 + }, + { + "epoch": 1.068497360113976, + "grad_norm": 0.637395441532135, + "learning_rate": 0.00046755939040788884, + "loss": 0.4424, + "step": 19124 + }, + { + "epoch": 1.068553230717658, + "grad_norm": 0.3412025570869446, + "learning_rate": 0.00046753137606454504, + "loss": 0.3952, + "step": 19125 + }, + { + "epoch": 1.0686091013213397, + "grad_norm": 0.45451921224594116, + "learning_rate": 0.0004675033617212013, + "loss": 0.4805, + "step": 19126 + }, + { + "epoch": 1.0686649719250216, + "grad_norm": 0.45005518198013306, + "learning_rate": 0.00046747534737785746, + "loss": 0.4751, + "step": 19127 + }, + { + "epoch": 1.0687208425287036, + "grad_norm": 0.5249797105789185, + "learning_rate": 0.0004674473330345137, + "loss": 0.4647, + "step": 19128 + }, + { + "epoch": 1.0687767131323853, + "grad_norm": 0.5483509302139282, + "learning_rate": 0.00046741931869116987, + "loss": 0.433, + "step": 19129 + }, + { + "epoch": 1.0688325837360673, + "grad_norm": 0.7431477904319763, + "learning_rate": 0.00046739130434782613, + "loss": 0.4946, + "step": 19130 + }, + { + "epoch": 1.0688884543397492, + "grad_norm": 0.4165104031562805, + "learning_rate": 0.0004673632900044823, + "loss": 0.4113, + "step": 19131 + }, + { + "epoch": 1.068944324943431, + "grad_norm": 0.3706815242767334, + "learning_rate": 0.00046733527566113854, + "loss": 0.3801, + "step": 19132 + }, + { + "epoch": 1.069000195547113, + "grad_norm": 0.7433362007141113, + "learning_rate": 0.00046730726131779474, + "loss": 0.4971, + "step": 19133 + }, + { + "epoch": 1.0690560661507948, + "grad_norm": 5.635871410369873, + "learning_rate": 0.00046727924697445095, + "loss": 0.5048, + "step": 19134 + }, + { + "epoch": 1.0691119367544766, + "grad_norm": 0.44141367077827454, + "learning_rate": 0.00046725123263110716, + "loss": 0.3332, + "step": 19135 + }, + { + "epoch": 1.0691678073581585, + "grad_norm": 1.1504639387130737, + "learning_rate": 0.0004672232182877633, + "loss": 0.4229, + "step": 19136 + }, + { + "epoch": 1.0692236779618405, + "grad_norm": 0.32115134596824646, + "learning_rate": 0.00046719520394441957, + "loss": 0.3254, + "step": 19137 + }, + { + "epoch": 1.0692795485655222, + "grad_norm": 0.6208354234695435, + "learning_rate": 0.0004671671896010757, + "loss": 0.7812, + "step": 19138 + }, + { + "epoch": 1.0693354191692042, + "grad_norm": 0.4657175838947296, + "learning_rate": 0.000467139175257732, + "loss": 0.4851, + "step": 19139 + }, + { + "epoch": 1.0693912897728861, + "grad_norm": 0.540298581123352, + "learning_rate": 0.00046711116091438813, + "loss": 0.2866, + "step": 19140 + }, + { + "epoch": 1.0694471603765678, + "grad_norm": 0.8642581105232239, + "learning_rate": 0.0004670831465710444, + "loss": 0.4252, + "step": 19141 + }, + { + "epoch": 1.0695030309802498, + "grad_norm": 0.792280375957489, + "learning_rate": 0.0004670551322277006, + "loss": 0.4057, + "step": 19142 + }, + { + "epoch": 1.0695589015839315, + "grad_norm": 0.5590070486068726, + "learning_rate": 0.0004670271178843568, + "loss": 0.3733, + "step": 19143 + }, + { + "epoch": 1.0696147721876135, + "grad_norm": 0.35182318091392517, + "learning_rate": 0.000466999103541013, + "loss": 0.4591, + "step": 19144 + }, + { + "epoch": 1.0696706427912954, + "grad_norm": 0.4550420641899109, + "learning_rate": 0.0004669710891976692, + "loss": 0.5572, + "step": 19145 + }, + { + "epoch": 1.0697265133949772, + "grad_norm": 0.4175336956977844, + "learning_rate": 0.0004669430748543254, + "loss": 0.4092, + "step": 19146 + }, + { + "epoch": 1.069782383998659, + "grad_norm": 0.4675232172012329, + "learning_rate": 0.0004669150605109816, + "loss": 0.3683, + "step": 19147 + }, + { + "epoch": 1.069838254602341, + "grad_norm": 0.3926471769809723, + "learning_rate": 0.00046688704616763783, + "loss": 0.438, + "step": 19148 + }, + { + "epoch": 1.0698941252060228, + "grad_norm": 1.0295047760009766, + "learning_rate": 0.0004668590318242941, + "loss": 0.3446, + "step": 19149 + }, + { + "epoch": 1.0699499958097047, + "grad_norm": 0.5490304827690125, + "learning_rate": 0.00046683101748095024, + "loss": 0.5284, + "step": 19150 + }, + { + "epoch": 1.0700058664133867, + "grad_norm": 0.6160852313041687, + "learning_rate": 0.0004668030031376065, + "loss": 0.5077, + "step": 19151 + }, + { + "epoch": 1.0700617370170684, + "grad_norm": 0.41175681352615356, + "learning_rate": 0.00046677498879426266, + "loss": 0.3468, + "step": 19152 + }, + { + "epoch": 1.0701176076207504, + "grad_norm": 0.3669334352016449, + "learning_rate": 0.0004667469744509189, + "loss": 0.3977, + "step": 19153 + }, + { + "epoch": 1.0701734782244323, + "grad_norm": 1.5389671325683594, + "learning_rate": 0.00046671896010757507, + "loss": 0.4402, + "step": 19154 + }, + { + "epoch": 1.070229348828114, + "grad_norm": 2.0968756675720215, + "learning_rate": 0.0004666909457642313, + "loss": 0.4258, + "step": 19155 + }, + { + "epoch": 1.070285219431796, + "grad_norm": 0.3962308466434479, + "learning_rate": 0.0004666629314208875, + "loss": 0.4077, + "step": 19156 + }, + { + "epoch": 1.0703410900354777, + "grad_norm": 0.3720594644546509, + "learning_rate": 0.0004666349170775437, + "loss": 0.3927, + "step": 19157 + }, + { + "epoch": 1.0703969606391597, + "grad_norm": 0.6276147365570068, + "learning_rate": 0.00046660690273419994, + "loss": 0.481, + "step": 19158 + }, + { + "epoch": 1.0704528312428416, + "grad_norm": 0.48747244477272034, + "learning_rate": 0.0004665788883908561, + "loss": 0.384, + "step": 19159 + }, + { + "epoch": 1.0705087018465234, + "grad_norm": 0.5234811902046204, + "learning_rate": 0.00046655087404751236, + "loss": 0.4344, + "step": 19160 + }, + { + "epoch": 1.0705645724502053, + "grad_norm": 0.3726770877838135, + "learning_rate": 0.0004665228597041685, + "loss": 0.4693, + "step": 19161 + }, + { + "epoch": 1.0706204430538873, + "grad_norm": 0.3987727761268616, + "learning_rate": 0.00046649484536082477, + "loss": 0.5777, + "step": 19162 + }, + { + "epoch": 1.070676313657569, + "grad_norm": 0.5494532585144043, + "learning_rate": 0.0004664668310174809, + "loss": 0.5112, + "step": 19163 + }, + { + "epoch": 1.070732184261251, + "grad_norm": 0.5667769908905029, + "learning_rate": 0.0004664388166741372, + "loss": 0.4531, + "step": 19164 + }, + { + "epoch": 1.070788054864933, + "grad_norm": 0.5176324248313904, + "learning_rate": 0.0004664108023307934, + "loss": 0.4874, + "step": 19165 + }, + { + "epoch": 1.0708439254686146, + "grad_norm": 0.4993896782398224, + "learning_rate": 0.0004663827879874496, + "loss": 0.6492, + "step": 19166 + }, + { + "epoch": 1.0708997960722966, + "grad_norm": 0.8115918040275574, + "learning_rate": 0.0004663547736441058, + "loss": 0.621, + "step": 19167 + }, + { + "epoch": 1.0709556666759785, + "grad_norm": 1.4177565574645996, + "learning_rate": 0.000466326759300762, + "loss": 0.4708, + "step": 19168 + }, + { + "epoch": 1.0710115372796603, + "grad_norm": 0.4554324150085449, + "learning_rate": 0.0004662987449574182, + "loss": 0.343, + "step": 19169 + }, + { + "epoch": 1.0710674078833422, + "grad_norm": 0.5020360946655273, + "learning_rate": 0.0004662707306140744, + "loss": 0.5625, + "step": 19170 + }, + { + "epoch": 1.0711232784870242, + "grad_norm": 0.4005953371524811, + "learning_rate": 0.0004662427162707306, + "loss": 0.369, + "step": 19171 + }, + { + "epoch": 1.0711791490907059, + "grad_norm": 0.5670609474182129, + "learning_rate": 0.0004662147019273868, + "loss": 0.4977, + "step": 19172 + }, + { + "epoch": 1.0712350196943878, + "grad_norm": 0.563781201839447, + "learning_rate": 0.00046618668758404303, + "loss": 0.4678, + "step": 19173 + }, + { + "epoch": 1.0712908902980698, + "grad_norm": 0.5663554072380066, + "learning_rate": 0.0004661586732406993, + "loss": 0.3953, + "step": 19174 + }, + { + "epoch": 1.0713467609017515, + "grad_norm": 0.5001922249794006, + "learning_rate": 0.00046613065889735544, + "loss": 0.557, + "step": 19175 + }, + { + "epoch": 1.0714026315054335, + "grad_norm": 0.6829466223716736, + "learning_rate": 0.0004661026445540117, + "loss": 0.3609, + "step": 19176 + }, + { + "epoch": 1.0714585021091152, + "grad_norm": 0.3736323118209839, + "learning_rate": 0.00046607463021066785, + "loss": 0.4734, + "step": 19177 + }, + { + "epoch": 1.0715143727127971, + "grad_norm": 0.4413478374481201, + "learning_rate": 0.00046604661586732406, + "loss": 0.4815, + "step": 19178 + }, + { + "epoch": 1.071570243316479, + "grad_norm": 0.4166807234287262, + "learning_rate": 0.00046601860152398027, + "loss": 0.3489, + "step": 19179 + }, + { + "epoch": 1.0716261139201608, + "grad_norm": 1.2434476613998413, + "learning_rate": 0.00046599058718063647, + "loss": 0.3887, + "step": 19180 + }, + { + "epoch": 1.0716819845238428, + "grad_norm": 0.5246871113777161, + "learning_rate": 0.00046596257283729273, + "loss": 0.4801, + "step": 19181 + }, + { + "epoch": 1.0717378551275247, + "grad_norm": 0.4270648956298828, + "learning_rate": 0.0004659345584939489, + "loss": 0.4782, + "step": 19182 + }, + { + "epoch": 1.0717937257312065, + "grad_norm": 0.587573766708374, + "learning_rate": 0.00046590654415060514, + "loss": 0.5163, + "step": 19183 + }, + { + "epoch": 1.0718495963348884, + "grad_norm": 0.5407771468162537, + "learning_rate": 0.0004658785298072613, + "loss": 0.5948, + "step": 19184 + }, + { + "epoch": 1.0719054669385704, + "grad_norm": 0.515678882598877, + "learning_rate": 0.00046585051546391756, + "loss": 0.4238, + "step": 19185 + }, + { + "epoch": 1.071961337542252, + "grad_norm": 0.47132471203804016, + "learning_rate": 0.0004658225011205737, + "loss": 0.5347, + "step": 19186 + }, + { + "epoch": 1.072017208145934, + "grad_norm": 0.39185309410095215, + "learning_rate": 0.00046579448677722997, + "loss": 0.4228, + "step": 19187 + }, + { + "epoch": 1.072073078749616, + "grad_norm": 0.5517094135284424, + "learning_rate": 0.0004657664724338861, + "loss": 0.3796, + "step": 19188 + }, + { + "epoch": 1.0721289493532977, + "grad_norm": 0.5772870182991028, + "learning_rate": 0.0004657384580905424, + "loss": 0.4258, + "step": 19189 + }, + { + "epoch": 1.0721848199569797, + "grad_norm": 0.44342339038848877, + "learning_rate": 0.0004657104437471986, + "loss": 0.3584, + "step": 19190 + }, + { + "epoch": 1.0722406905606614, + "grad_norm": 0.35883939266204834, + "learning_rate": 0.0004656824294038548, + "loss": 0.4231, + "step": 19191 + }, + { + "epoch": 1.0722965611643434, + "grad_norm": 1.5956077575683594, + "learning_rate": 0.000465654415060511, + "loss": 0.4785, + "step": 19192 + }, + { + "epoch": 1.0723524317680253, + "grad_norm": 0.32695022225379944, + "learning_rate": 0.0004656264007171672, + "loss": 0.477, + "step": 19193 + }, + { + "epoch": 1.072408302371707, + "grad_norm": 0.36537134647369385, + "learning_rate": 0.0004655983863738234, + "loss": 0.4521, + "step": 19194 + }, + { + "epoch": 1.072464172975389, + "grad_norm": 0.3431412875652313, + "learning_rate": 0.0004655703720304796, + "loss": 0.3912, + "step": 19195 + }, + { + "epoch": 1.072520043579071, + "grad_norm": 0.39231228828430176, + "learning_rate": 0.0004655423576871358, + "loss": 0.3776, + "step": 19196 + }, + { + "epoch": 1.0725759141827527, + "grad_norm": 0.4640977084636688, + "learning_rate": 0.0004655143433437921, + "loss": 0.4065, + "step": 19197 + }, + { + "epoch": 1.0726317847864346, + "grad_norm": 0.4167291820049286, + "learning_rate": 0.00046548632900044823, + "loss": 0.4823, + "step": 19198 + }, + { + "epoch": 1.0726876553901166, + "grad_norm": 0.35848909616470337, + "learning_rate": 0.00046545831465710444, + "loss": 0.354, + "step": 19199 + }, + { + "epoch": 1.0727435259937983, + "grad_norm": 0.5537431836128235, + "learning_rate": 0.00046543030031376064, + "loss": 0.4387, + "step": 19200 + }, + { + "epoch": 1.0727993965974802, + "grad_norm": 0.5311104655265808, + "learning_rate": 0.00046540228597041685, + "loss": 0.4416, + "step": 19201 + }, + { + "epoch": 1.0728552672011622, + "grad_norm": 0.3923008143901825, + "learning_rate": 0.00046537427162707305, + "loss": 0.4413, + "step": 19202 + }, + { + "epoch": 1.072911137804844, + "grad_norm": 0.48321333527565, + "learning_rate": 0.00046534625728372926, + "loss": 0.4501, + "step": 19203 + }, + { + "epoch": 1.0729670084085259, + "grad_norm": 0.4820128083229065, + "learning_rate": 0.0004653182429403855, + "loss": 0.4561, + "step": 19204 + }, + { + "epoch": 1.0730228790122078, + "grad_norm": 0.41609883308410645, + "learning_rate": 0.00046529022859704167, + "loss": 0.3194, + "step": 19205 + }, + { + "epoch": 1.0730787496158896, + "grad_norm": 0.5538482666015625, + "learning_rate": 0.00046526221425369793, + "loss": 0.3821, + "step": 19206 + }, + { + "epoch": 1.0731346202195715, + "grad_norm": 0.3439604341983795, + "learning_rate": 0.0004652341999103541, + "loss": 0.4082, + "step": 19207 + }, + { + "epoch": 1.0731904908232535, + "grad_norm": 0.5508474111557007, + "learning_rate": 0.00046520618556701034, + "loss": 0.4863, + "step": 19208 + }, + { + "epoch": 1.0732463614269352, + "grad_norm": 0.37987324595451355, + "learning_rate": 0.0004651781712236665, + "loss": 0.3651, + "step": 19209 + }, + { + "epoch": 1.0733022320306171, + "grad_norm": 2.460773229598999, + "learning_rate": 0.00046515015688032276, + "loss": 0.3896, + "step": 19210 + }, + { + "epoch": 1.0733581026342989, + "grad_norm": 0.3591419458389282, + "learning_rate": 0.0004651221425369789, + "loss": 0.3935, + "step": 19211 + }, + { + "epoch": 1.0734139732379808, + "grad_norm": 0.5133680105209351, + "learning_rate": 0.00046509412819363517, + "loss": 0.5436, + "step": 19212 + }, + { + "epoch": 1.0734698438416628, + "grad_norm": 1.783042550086975, + "learning_rate": 0.00046506611385029137, + "loss": 0.4543, + "step": 19213 + }, + { + "epoch": 1.0735257144453445, + "grad_norm": 0.5488684177398682, + "learning_rate": 0.0004650380995069476, + "loss": 0.451, + "step": 19214 + }, + { + "epoch": 1.0735815850490265, + "grad_norm": 0.4693807363510132, + "learning_rate": 0.0004650100851636038, + "loss": 0.3571, + "step": 19215 + }, + { + "epoch": 1.0736374556527084, + "grad_norm": 0.3960597515106201, + "learning_rate": 0.00046498207082026, + "loss": 0.4407, + "step": 19216 + }, + { + "epoch": 1.0736933262563901, + "grad_norm": 0.4764605164527893, + "learning_rate": 0.0004649540564769162, + "loss": 0.5017, + "step": 19217 + }, + { + "epoch": 1.073749196860072, + "grad_norm": 2.981023073196411, + "learning_rate": 0.0004649260421335724, + "loss": 0.4764, + "step": 19218 + }, + { + "epoch": 1.073805067463754, + "grad_norm": 1.9867037534713745, + "learning_rate": 0.0004648980277902286, + "loss": 0.4197, + "step": 19219 + }, + { + "epoch": 1.0738609380674358, + "grad_norm": 0.7274839878082275, + "learning_rate": 0.0004648700134468848, + "loss": 0.427, + "step": 19220 + }, + { + "epoch": 1.0739168086711177, + "grad_norm": 0.3977402448654175, + "learning_rate": 0.000464841999103541, + "loss": 0.3924, + "step": 19221 + }, + { + "epoch": 1.0739726792747994, + "grad_norm": 1.2260198593139648, + "learning_rate": 0.0004648139847601972, + "loss": 0.5199, + "step": 19222 + }, + { + "epoch": 1.0740285498784814, + "grad_norm": 0.3422035276889801, + "learning_rate": 0.00046478597041685343, + "loss": 0.3221, + "step": 19223 + }, + { + "epoch": 1.0740844204821633, + "grad_norm": 0.5074552893638611, + "learning_rate": 0.00046475795607350964, + "loss": 0.5001, + "step": 19224 + }, + { + "epoch": 1.074140291085845, + "grad_norm": 0.44784218072891235, + "learning_rate": 0.00046472994173016584, + "loss": 0.4082, + "step": 19225 + }, + { + "epoch": 1.074196161689527, + "grad_norm": 0.41920575499534607, + "learning_rate": 0.00046470192738682205, + "loss": 0.4899, + "step": 19226 + }, + { + "epoch": 1.074252032293209, + "grad_norm": 0.5953748226165771, + "learning_rate": 0.00046467391304347825, + "loss": 0.6132, + "step": 19227 + }, + { + "epoch": 1.0743079028968907, + "grad_norm": 0.45389607548713684, + "learning_rate": 0.00046464589870013446, + "loss": 0.344, + "step": 19228 + }, + { + "epoch": 1.0743637735005727, + "grad_norm": 0.6183366179466248, + "learning_rate": 0.0004646178843567907, + "loss": 0.5391, + "step": 19229 + }, + { + "epoch": 1.0744196441042546, + "grad_norm": 0.34619662165641785, + "learning_rate": 0.00046458987001344687, + "loss": 0.4509, + "step": 19230 + }, + { + "epoch": 1.0744755147079363, + "grad_norm": 0.3965436518192291, + "learning_rate": 0.00046456185567010313, + "loss": 0.4487, + "step": 19231 + }, + { + "epoch": 1.0745313853116183, + "grad_norm": 1.0786430835723877, + "learning_rate": 0.0004645338413267593, + "loss": 0.3795, + "step": 19232 + }, + { + "epoch": 1.0745872559153002, + "grad_norm": 3.1722891330718994, + "learning_rate": 0.00046450582698341554, + "loss": 0.5011, + "step": 19233 + }, + { + "epoch": 1.074643126518982, + "grad_norm": 0.6075801849365234, + "learning_rate": 0.0004644778126400717, + "loss": 0.4432, + "step": 19234 + }, + { + "epoch": 1.074698997122664, + "grad_norm": 2.0335853099823, + "learning_rate": 0.00046444979829672795, + "loss": 0.385, + "step": 19235 + }, + { + "epoch": 1.0747548677263459, + "grad_norm": 0.5540317893028259, + "learning_rate": 0.00046442178395338416, + "loss": 0.4215, + "step": 19236 + }, + { + "epoch": 1.0748107383300276, + "grad_norm": 0.48569196462631226, + "learning_rate": 0.00046439376961004037, + "loss": 0.5072, + "step": 19237 + }, + { + "epoch": 1.0748666089337096, + "grad_norm": 0.8109817504882812, + "learning_rate": 0.00046436575526669657, + "loss": 0.3963, + "step": 19238 + }, + { + "epoch": 1.0749224795373915, + "grad_norm": 0.4423699378967285, + "learning_rate": 0.0004643377409233528, + "loss": 0.3558, + "step": 19239 + }, + { + "epoch": 1.0749783501410732, + "grad_norm": 6.15464448928833, + "learning_rate": 0.000464309726580009, + "loss": 0.3082, + "step": 19240 + }, + { + "epoch": 1.0750342207447552, + "grad_norm": 0.9550255537033081, + "learning_rate": 0.00046428171223666514, + "loss": 0.4715, + "step": 19241 + }, + { + "epoch": 1.075090091348437, + "grad_norm": 0.46587610244750977, + "learning_rate": 0.0004642536978933214, + "loss": 0.3598, + "step": 19242 + }, + { + "epoch": 1.0751459619521189, + "grad_norm": 0.4063360095024109, + "learning_rate": 0.00046422568354997755, + "loss": 0.398, + "step": 19243 + }, + { + "epoch": 1.0752018325558008, + "grad_norm": 0.5073213577270508, + "learning_rate": 0.0004641976692066338, + "loss": 0.4673, + "step": 19244 + }, + { + "epoch": 1.0752577031594825, + "grad_norm": 0.4873636066913605, + "learning_rate": 0.00046416965486329, + "loss": 0.3829, + "step": 19245 + }, + { + "epoch": 1.0753135737631645, + "grad_norm": 0.29499348998069763, + "learning_rate": 0.0004641416405199462, + "loss": 0.386, + "step": 19246 + }, + { + "epoch": 1.0753694443668464, + "grad_norm": 0.30003622174263, + "learning_rate": 0.0004641136261766024, + "loss": 0.4478, + "step": 19247 + }, + { + "epoch": 1.0754253149705282, + "grad_norm": 0.4493175446987152, + "learning_rate": 0.00046408561183325863, + "loss": 0.5148, + "step": 19248 + }, + { + "epoch": 1.0754811855742101, + "grad_norm": 0.807591438293457, + "learning_rate": 0.00046405759748991484, + "loss": 0.4569, + "step": 19249 + }, + { + "epoch": 1.075537056177892, + "grad_norm": 0.4200882613658905, + "learning_rate": 0.00046402958314657104, + "loss": 0.599, + "step": 19250 + }, + { + "epoch": 1.0755929267815738, + "grad_norm": 1.840232253074646, + "learning_rate": 0.00046400156880322725, + "loss": 0.4419, + "step": 19251 + }, + { + "epoch": 1.0756487973852558, + "grad_norm": 0.3758127987384796, + "learning_rate": 0.0004639735544598835, + "loss": 0.4292, + "step": 19252 + }, + { + "epoch": 1.0757046679889377, + "grad_norm": 0.42899900674819946, + "learning_rate": 0.00046394554011653966, + "loss": 0.4364, + "step": 19253 + }, + { + "epoch": 1.0757605385926194, + "grad_norm": 0.44510000944137573, + "learning_rate": 0.0004639175257731959, + "loss": 0.4664, + "step": 19254 + }, + { + "epoch": 1.0758164091963014, + "grad_norm": 0.6034139394760132, + "learning_rate": 0.00046388951142985207, + "loss": 0.3882, + "step": 19255 + }, + { + "epoch": 1.0758722797999831, + "grad_norm": 4.347630977630615, + "learning_rate": 0.00046386149708650833, + "loss": 0.4628, + "step": 19256 + }, + { + "epoch": 1.075928150403665, + "grad_norm": 0.49710339307785034, + "learning_rate": 0.0004638334827431645, + "loss": 0.309, + "step": 19257 + }, + { + "epoch": 1.075984021007347, + "grad_norm": 0.35745784640312195, + "learning_rate": 0.00046380546839982074, + "loss": 0.4188, + "step": 19258 + }, + { + "epoch": 1.0760398916110288, + "grad_norm": 0.4306316077709198, + "learning_rate": 0.00046377745405647695, + "loss": 0.3991, + "step": 19259 + }, + { + "epoch": 1.0760957622147107, + "grad_norm": 0.5213345289230347, + "learning_rate": 0.00046374943971313315, + "loss": 0.5127, + "step": 19260 + }, + { + "epoch": 1.0761516328183927, + "grad_norm": 0.6184742450714111, + "learning_rate": 0.00046372142536978936, + "loss": 0.4293, + "step": 19261 + }, + { + "epoch": 1.0762075034220744, + "grad_norm": 0.35856398940086365, + "learning_rate": 0.00046369341102644557, + "loss": 0.4136, + "step": 19262 + }, + { + "epoch": 1.0762633740257563, + "grad_norm": 0.4704131782054901, + "learning_rate": 0.00046366539668310177, + "loss": 0.4868, + "step": 19263 + }, + { + "epoch": 1.0763192446294383, + "grad_norm": 1.1181254386901855, + "learning_rate": 0.0004636373823397579, + "loss": 0.4518, + "step": 19264 + }, + { + "epoch": 1.07637511523312, + "grad_norm": 0.426300048828125, + "learning_rate": 0.0004636093679964142, + "loss": 0.4859, + "step": 19265 + }, + { + "epoch": 1.076430985836802, + "grad_norm": 0.8179359436035156, + "learning_rate": 0.00046358135365307034, + "loss": 0.3844, + "step": 19266 + }, + { + "epoch": 1.076486856440484, + "grad_norm": 0.42657628655433655, + "learning_rate": 0.0004635533393097266, + "loss": 0.3797, + "step": 19267 + }, + { + "epoch": 1.0765427270441656, + "grad_norm": 0.44726505875587463, + "learning_rate": 0.0004635253249663828, + "loss": 0.4637, + "step": 19268 + }, + { + "epoch": 1.0765985976478476, + "grad_norm": 0.624509871006012, + "learning_rate": 0.000463497310623039, + "loss": 0.407, + "step": 19269 + }, + { + "epoch": 1.0766544682515296, + "grad_norm": 0.6315744519233704, + "learning_rate": 0.0004634692962796952, + "loss": 0.4907, + "step": 19270 + }, + { + "epoch": 1.0767103388552113, + "grad_norm": 0.5123518705368042, + "learning_rate": 0.0004634412819363514, + "loss": 0.373, + "step": 19271 + }, + { + "epoch": 1.0767662094588932, + "grad_norm": 0.39483514428138733, + "learning_rate": 0.0004634132675930076, + "loss": 0.4183, + "step": 19272 + }, + { + "epoch": 1.0768220800625752, + "grad_norm": 0.7714408040046692, + "learning_rate": 0.00046338525324966383, + "loss": 0.4529, + "step": 19273 + }, + { + "epoch": 1.076877950666257, + "grad_norm": 0.3778264820575714, + "learning_rate": 0.00046335723890632004, + "loss": 0.3358, + "step": 19274 + }, + { + "epoch": 1.0769338212699389, + "grad_norm": 0.5957016944885254, + "learning_rate": 0.0004633292245629763, + "loss": 0.5305, + "step": 19275 + }, + { + "epoch": 1.0769896918736206, + "grad_norm": 0.3547683656215668, + "learning_rate": 0.00046330121021963245, + "loss": 0.4813, + "step": 19276 + }, + { + "epoch": 1.0770455624773025, + "grad_norm": 0.41202250123023987, + "learning_rate": 0.0004632731958762887, + "loss": 0.4309, + "step": 19277 + }, + { + "epoch": 1.0771014330809845, + "grad_norm": 0.7247516512870789, + "learning_rate": 0.00046324518153294486, + "loss": 0.4971, + "step": 19278 + }, + { + "epoch": 1.0771573036846662, + "grad_norm": 0.5319809913635254, + "learning_rate": 0.0004632171671896011, + "loss": 0.4776, + "step": 19279 + }, + { + "epoch": 1.0772131742883482, + "grad_norm": 0.5031575560569763, + "learning_rate": 0.00046318915284625727, + "loss": 0.4106, + "step": 19280 + }, + { + "epoch": 1.0772690448920301, + "grad_norm": 0.6737425923347473, + "learning_rate": 0.00046316113850291353, + "loss": 0.5216, + "step": 19281 + }, + { + "epoch": 1.0773249154957119, + "grad_norm": 0.5406571626663208, + "learning_rate": 0.0004631331241595697, + "loss": 0.4595, + "step": 19282 + }, + { + "epoch": 1.0773807860993938, + "grad_norm": 0.4288605749607086, + "learning_rate": 0.00046310510981622594, + "loss": 0.3895, + "step": 19283 + }, + { + "epoch": 1.0774366567030758, + "grad_norm": 0.586567759513855, + "learning_rate": 0.00046307709547288215, + "loss": 0.4548, + "step": 19284 + }, + { + "epoch": 1.0774925273067575, + "grad_norm": 0.5273689031600952, + "learning_rate": 0.0004630490811295383, + "loss": 0.4452, + "step": 19285 + }, + { + "epoch": 1.0775483979104394, + "grad_norm": 0.7741566300392151, + "learning_rate": 0.00046302106678619456, + "loss": 0.3764, + "step": 19286 + }, + { + "epoch": 1.0776042685141214, + "grad_norm": 0.39839109778404236, + "learning_rate": 0.0004629930524428507, + "loss": 0.4461, + "step": 19287 + }, + { + "epoch": 1.0776601391178031, + "grad_norm": 0.5674782991409302, + "learning_rate": 0.00046296503809950697, + "loss": 0.3897, + "step": 19288 + }, + { + "epoch": 1.077716009721485, + "grad_norm": 0.2891356647014618, + "learning_rate": 0.0004629370237561631, + "loss": 0.3215, + "step": 19289 + }, + { + "epoch": 1.0777718803251668, + "grad_norm": 0.646867573261261, + "learning_rate": 0.0004629090094128194, + "loss": 0.6611, + "step": 19290 + }, + { + "epoch": 1.0778277509288487, + "grad_norm": 0.4267713725566864, + "learning_rate": 0.0004628809950694756, + "loss": 0.4587, + "step": 19291 + }, + { + "epoch": 1.0778836215325307, + "grad_norm": 0.3995933532714844, + "learning_rate": 0.0004628529807261318, + "loss": 0.4255, + "step": 19292 + }, + { + "epoch": 1.0779394921362124, + "grad_norm": 0.7155071496963501, + "learning_rate": 0.000462824966382788, + "loss": 0.5059, + "step": 19293 + }, + { + "epoch": 1.0779953627398944, + "grad_norm": 3.759474277496338, + "learning_rate": 0.0004627969520394442, + "loss": 0.4258, + "step": 19294 + }, + { + "epoch": 1.0780512333435763, + "grad_norm": 0.3544466495513916, + "learning_rate": 0.0004627689376961004, + "loss": 0.4453, + "step": 19295 + }, + { + "epoch": 1.078107103947258, + "grad_norm": 0.6510049700737, + "learning_rate": 0.0004627409233527566, + "loss": 0.5122, + "step": 19296 + }, + { + "epoch": 1.07816297455094, + "grad_norm": 0.7269936203956604, + "learning_rate": 0.0004627129090094128, + "loss": 0.5098, + "step": 19297 + }, + { + "epoch": 1.078218845154622, + "grad_norm": 0.5124948620796204, + "learning_rate": 0.00046268489466606903, + "loss": 0.3834, + "step": 19298 + }, + { + "epoch": 1.0782747157583037, + "grad_norm": 0.5372222065925598, + "learning_rate": 0.00046265688032272524, + "loss": 0.4715, + "step": 19299 + }, + { + "epoch": 1.0783305863619856, + "grad_norm": 0.5032718181610107, + "learning_rate": 0.0004626288659793815, + "loss": 0.5076, + "step": 19300 + }, + { + "epoch": 1.0783864569656676, + "grad_norm": 0.3611942231655121, + "learning_rate": 0.00046260085163603765, + "loss": 0.4622, + "step": 19301 + }, + { + "epoch": 1.0784423275693493, + "grad_norm": 2.4532527923583984, + "learning_rate": 0.0004625728372926939, + "loss": 0.3908, + "step": 19302 + }, + { + "epoch": 1.0784981981730313, + "grad_norm": 0.5424118638038635, + "learning_rate": 0.00046254482294935006, + "loss": 0.4242, + "step": 19303 + }, + { + "epoch": 1.0785540687767132, + "grad_norm": 0.6443061232566833, + "learning_rate": 0.0004625168086060063, + "loss": 0.4369, + "step": 19304 + }, + { + "epoch": 1.078609939380395, + "grad_norm": 0.43167659640312195, + "learning_rate": 0.00046248879426266247, + "loss": 0.4539, + "step": 19305 + }, + { + "epoch": 1.078665809984077, + "grad_norm": 0.5178162455558777, + "learning_rate": 0.0004624607799193187, + "loss": 0.4283, + "step": 19306 + }, + { + "epoch": 1.0787216805877589, + "grad_norm": 0.36421799659729004, + "learning_rate": 0.00046243276557597494, + "loss": 0.4239, + "step": 19307 + }, + { + "epoch": 1.0787775511914406, + "grad_norm": 0.4816279709339142, + "learning_rate": 0.0004624047512326311, + "loss": 0.4641, + "step": 19308 + }, + { + "epoch": 1.0788334217951225, + "grad_norm": 0.49721619486808777, + "learning_rate": 0.00046237673688928735, + "loss": 0.3256, + "step": 19309 + }, + { + "epoch": 1.0788892923988043, + "grad_norm": 0.48730015754699707, + "learning_rate": 0.0004623487225459435, + "loss": 0.3847, + "step": 19310 + }, + { + "epoch": 1.0789451630024862, + "grad_norm": 0.6341501474380493, + "learning_rate": 0.00046232070820259976, + "loss": 0.4566, + "step": 19311 + }, + { + "epoch": 1.0790010336061682, + "grad_norm": 0.537011981010437, + "learning_rate": 0.0004622926938592559, + "loss": 0.446, + "step": 19312 + }, + { + "epoch": 1.07905690420985, + "grad_norm": 0.8977299332618713, + "learning_rate": 0.00046226467951591217, + "loss": 0.7184, + "step": 19313 + }, + { + "epoch": 1.0791127748135318, + "grad_norm": 0.6184439063072205, + "learning_rate": 0.0004622366651725683, + "loss": 0.4647, + "step": 19314 + }, + { + "epoch": 1.0791686454172138, + "grad_norm": 0.38316184282302856, + "learning_rate": 0.0004622086508292246, + "loss": 0.4463, + "step": 19315 + }, + { + "epoch": 1.0792245160208955, + "grad_norm": 0.5500901937484741, + "learning_rate": 0.0004621806364858808, + "loss": 0.4452, + "step": 19316 + }, + { + "epoch": 1.0792803866245775, + "grad_norm": 0.47580617666244507, + "learning_rate": 0.000462152622142537, + "loss": 0.4148, + "step": 19317 + }, + { + "epoch": 1.0793362572282594, + "grad_norm": 0.9134114384651184, + "learning_rate": 0.0004621246077991932, + "loss": 0.3367, + "step": 19318 + }, + { + "epoch": 1.0793921278319412, + "grad_norm": 0.6919000148773193, + "learning_rate": 0.0004620965934558494, + "loss": 0.5874, + "step": 19319 + }, + { + "epoch": 1.0794479984356231, + "grad_norm": 0.36038514971733093, + "learning_rate": 0.0004620685791125056, + "loss": 0.3716, + "step": 19320 + }, + { + "epoch": 1.079503869039305, + "grad_norm": 0.5457484126091003, + "learning_rate": 0.0004620405647691618, + "loss": 0.361, + "step": 19321 + }, + { + "epoch": 1.0795597396429868, + "grad_norm": 0.7032338976860046, + "learning_rate": 0.000462012550425818, + "loss": 0.4493, + "step": 19322 + }, + { + "epoch": 1.0796156102466687, + "grad_norm": 0.4463641345500946, + "learning_rate": 0.0004619845360824743, + "loss": 0.3527, + "step": 19323 + }, + { + "epoch": 1.0796714808503505, + "grad_norm": 1.1218087673187256, + "learning_rate": 0.00046195652173913043, + "loss": 0.541, + "step": 19324 + }, + { + "epoch": 1.0797273514540324, + "grad_norm": 0.6436823606491089, + "learning_rate": 0.0004619285073957867, + "loss": 0.5425, + "step": 19325 + }, + { + "epoch": 1.0797832220577144, + "grad_norm": 0.4054082930088043, + "learning_rate": 0.00046190049305244285, + "loss": 0.3959, + "step": 19326 + }, + { + "epoch": 1.079839092661396, + "grad_norm": 0.3177461624145508, + "learning_rate": 0.00046187247870909905, + "loss": 0.4527, + "step": 19327 + }, + { + "epoch": 1.079894963265078, + "grad_norm": 0.4492526352405548, + "learning_rate": 0.00046184446436575526, + "loss": 0.5236, + "step": 19328 + }, + { + "epoch": 1.07995083386876, + "grad_norm": 0.3537319004535675, + "learning_rate": 0.00046181645002241146, + "loss": 0.4347, + "step": 19329 + }, + { + "epoch": 1.0800067044724417, + "grad_norm": 0.7541767358779907, + "learning_rate": 0.0004617884356790677, + "loss": 0.3902, + "step": 19330 + }, + { + "epoch": 1.0800625750761237, + "grad_norm": 0.6442145705223083, + "learning_rate": 0.0004617604213357239, + "loss": 0.3551, + "step": 19331 + }, + { + "epoch": 1.0801184456798056, + "grad_norm": 0.4586670398712158, + "learning_rate": 0.00046173240699238014, + "loss": 0.378, + "step": 19332 + }, + { + "epoch": 1.0801743162834874, + "grad_norm": 0.3923884928226471, + "learning_rate": 0.0004617043926490363, + "loss": 0.4584, + "step": 19333 + }, + { + "epoch": 1.0802301868871693, + "grad_norm": 0.34636828303337097, + "learning_rate": 0.00046167637830569255, + "loss": 0.3489, + "step": 19334 + }, + { + "epoch": 1.0802860574908513, + "grad_norm": 0.3926423490047455, + "learning_rate": 0.0004616483639623487, + "loss": 0.431, + "step": 19335 + }, + { + "epoch": 1.080341928094533, + "grad_norm": 0.8551467061042786, + "learning_rate": 0.00046162034961900496, + "loss": 0.4443, + "step": 19336 + }, + { + "epoch": 1.080397798698215, + "grad_norm": 0.41621533036231995, + "learning_rate": 0.0004615923352756611, + "loss": 0.4632, + "step": 19337 + }, + { + "epoch": 1.080453669301897, + "grad_norm": 0.47700178623199463, + "learning_rate": 0.00046156432093231737, + "loss": 0.3393, + "step": 19338 + }, + { + "epoch": 1.0805095399055786, + "grad_norm": 1.0141637325286865, + "learning_rate": 0.0004615363065889736, + "loss": 0.4745, + "step": 19339 + }, + { + "epoch": 1.0805654105092606, + "grad_norm": 0.40931713581085205, + "learning_rate": 0.0004615082922456298, + "loss": 0.4141, + "step": 19340 + }, + { + "epoch": 1.0806212811129425, + "grad_norm": 0.4874081313610077, + "learning_rate": 0.000461480277902286, + "loss": 0.5242, + "step": 19341 + }, + { + "epoch": 1.0806771517166243, + "grad_norm": 0.5071318745613098, + "learning_rate": 0.0004614522635589422, + "loss": 0.3528, + "step": 19342 + }, + { + "epoch": 1.0807330223203062, + "grad_norm": 0.475042462348938, + "learning_rate": 0.0004614242492155984, + "loss": 0.4999, + "step": 19343 + }, + { + "epoch": 1.080788892923988, + "grad_norm": 0.6380023956298828, + "learning_rate": 0.0004613962348722546, + "loss": 0.4081, + "step": 19344 + }, + { + "epoch": 1.08084476352767, + "grad_norm": 0.5182833671569824, + "learning_rate": 0.0004613682205289108, + "loss": 0.596, + "step": 19345 + }, + { + "epoch": 1.0809006341313518, + "grad_norm": 0.5509362816810608, + "learning_rate": 0.00046134020618556707, + "loss": 0.5723, + "step": 19346 + }, + { + "epoch": 1.0809565047350336, + "grad_norm": 0.3429674208164215, + "learning_rate": 0.0004613121918422232, + "loss": 0.3771, + "step": 19347 + }, + { + "epoch": 1.0810123753387155, + "grad_norm": 0.5612035989761353, + "learning_rate": 0.00046128417749887943, + "loss": 0.3762, + "step": 19348 + }, + { + "epoch": 1.0810682459423975, + "grad_norm": 0.31504255533218384, + "learning_rate": 0.00046125616315553563, + "loss": 0.4383, + "step": 19349 + }, + { + "epoch": 1.0811241165460792, + "grad_norm": 0.5366840362548828, + "learning_rate": 0.00046122814881219184, + "loss": 0.4787, + "step": 19350 + }, + { + "epoch": 1.0811799871497612, + "grad_norm": 1.7383967638015747, + "learning_rate": 0.00046120013446884805, + "loss": 0.5102, + "step": 19351 + }, + { + "epoch": 1.081235857753443, + "grad_norm": 0.3421078026294708, + "learning_rate": 0.00046117212012550425, + "loss": 0.4711, + "step": 19352 + }, + { + "epoch": 1.0812917283571248, + "grad_norm": 0.5557044744491577, + "learning_rate": 0.00046114410578216046, + "loss": 0.4128, + "step": 19353 + }, + { + "epoch": 1.0813475989608068, + "grad_norm": 0.4307330250740051, + "learning_rate": 0.00046111609143881666, + "loss": 0.372, + "step": 19354 + }, + { + "epoch": 1.0814034695644887, + "grad_norm": 0.5932120084762573, + "learning_rate": 0.0004610880770954729, + "loss": 0.4394, + "step": 19355 + }, + { + "epoch": 1.0814593401681705, + "grad_norm": 0.4078221321105957, + "learning_rate": 0.0004610600627521291, + "loss": 0.448, + "step": 19356 + }, + { + "epoch": 1.0815152107718524, + "grad_norm": 0.34890738129615784, + "learning_rate": 0.00046103204840878534, + "loss": 0.3536, + "step": 19357 + }, + { + "epoch": 1.0815710813755341, + "grad_norm": 2.1108860969543457, + "learning_rate": 0.0004610040340654415, + "loss": 0.4115, + "step": 19358 + }, + { + "epoch": 1.081626951979216, + "grad_norm": 0.45427650213241577, + "learning_rate": 0.00046097601972209775, + "loss": 0.4114, + "step": 19359 + }, + { + "epoch": 1.081682822582898, + "grad_norm": 0.46077612042427063, + "learning_rate": 0.0004609480053787539, + "loss": 0.3728, + "step": 19360 + }, + { + "epoch": 1.0817386931865798, + "grad_norm": 0.3683410882949829, + "learning_rate": 0.00046091999103541016, + "loss": 0.2933, + "step": 19361 + }, + { + "epoch": 1.0817945637902617, + "grad_norm": 0.6513609290122986, + "learning_rate": 0.00046089197669206636, + "loss": 0.3798, + "step": 19362 + }, + { + "epoch": 1.0818504343939437, + "grad_norm": 1.2137638330459595, + "learning_rate": 0.00046086396234872257, + "loss": 0.469, + "step": 19363 + }, + { + "epoch": 1.0819063049976254, + "grad_norm": 0.7605241537094116, + "learning_rate": 0.0004608359480053788, + "loss": 0.3893, + "step": 19364 + }, + { + "epoch": 1.0819621756013074, + "grad_norm": 0.550901472568512, + "learning_rate": 0.000460807933662035, + "loss": 0.3914, + "step": 19365 + }, + { + "epoch": 1.0820180462049893, + "grad_norm": 0.4887843132019043, + "learning_rate": 0.0004607799193186912, + "loss": 0.4455, + "step": 19366 + }, + { + "epoch": 1.082073916808671, + "grad_norm": 0.5288654565811157, + "learning_rate": 0.0004607519049753474, + "loss": 0.3969, + "step": 19367 + }, + { + "epoch": 1.082129787412353, + "grad_norm": 0.573356568813324, + "learning_rate": 0.0004607238906320036, + "loss": 0.4747, + "step": 19368 + }, + { + "epoch": 1.082185658016035, + "grad_norm": 0.8962626457214355, + "learning_rate": 0.00046069587628865975, + "loss": 0.4893, + "step": 19369 + }, + { + "epoch": 1.0822415286197167, + "grad_norm": 1.0832732915878296, + "learning_rate": 0.000460667861945316, + "loss": 0.4245, + "step": 19370 + }, + { + "epoch": 1.0822973992233986, + "grad_norm": 0.39540350437164307, + "learning_rate": 0.0004606398476019722, + "loss": 0.3492, + "step": 19371 + }, + { + "epoch": 1.0823532698270806, + "grad_norm": 0.38337400555610657, + "learning_rate": 0.0004606118332586284, + "loss": 0.4127, + "step": 19372 + }, + { + "epoch": 1.0824091404307623, + "grad_norm": 0.45403853058815, + "learning_rate": 0.00046058381891528463, + "loss": 0.4154, + "step": 19373 + }, + { + "epoch": 1.0824650110344443, + "grad_norm": 0.7723836898803711, + "learning_rate": 0.00046055580457194083, + "loss": 0.5764, + "step": 19374 + }, + { + "epoch": 1.0825208816381262, + "grad_norm": 0.6113090515136719, + "learning_rate": 0.00046052779022859704, + "loss": 0.517, + "step": 19375 + }, + { + "epoch": 1.082576752241808, + "grad_norm": 0.4116278886795044, + "learning_rate": 0.00046049977588525325, + "loss": 0.3634, + "step": 19376 + }, + { + "epoch": 1.08263262284549, + "grad_norm": 0.6791595220565796, + "learning_rate": 0.00046047176154190945, + "loss": 0.3797, + "step": 19377 + }, + { + "epoch": 1.0826884934491716, + "grad_norm": 0.49543866515159607, + "learning_rate": 0.0004604437471985657, + "loss": 0.5359, + "step": 19378 + }, + { + "epoch": 1.0827443640528536, + "grad_norm": 0.3925161361694336, + "learning_rate": 0.00046041573285522186, + "loss": 0.4293, + "step": 19379 + }, + { + "epoch": 1.0828002346565355, + "grad_norm": 0.8010302782058716, + "learning_rate": 0.0004603877185118781, + "loss": 0.4301, + "step": 19380 + }, + { + "epoch": 1.0828561052602172, + "grad_norm": 1.771966814994812, + "learning_rate": 0.0004603597041685343, + "loss": 0.4297, + "step": 19381 + }, + { + "epoch": 1.0829119758638992, + "grad_norm": 0.5244864821434021, + "learning_rate": 0.00046033168982519053, + "loss": 0.3997, + "step": 19382 + }, + { + "epoch": 1.0829678464675812, + "grad_norm": 0.34325316548347473, + "learning_rate": 0.0004603036754818467, + "loss": 0.3722, + "step": 19383 + }, + { + "epoch": 1.0830237170712629, + "grad_norm": 0.5068625211715698, + "learning_rate": 0.00046027566113850295, + "loss": 0.4207, + "step": 19384 + }, + { + "epoch": 1.0830795876749448, + "grad_norm": 0.3552395701408386, + "learning_rate": 0.0004602476467951591, + "loss": 0.4673, + "step": 19385 + }, + { + "epoch": 1.0831354582786268, + "grad_norm": 0.41711440682411194, + "learning_rate": 0.00046021963245181536, + "loss": 0.3585, + "step": 19386 + }, + { + "epoch": 1.0831913288823085, + "grad_norm": 0.6108904480934143, + "learning_rate": 0.00046019161810847156, + "loss": 0.3613, + "step": 19387 + }, + { + "epoch": 1.0832471994859905, + "grad_norm": 0.46436333656311035, + "learning_rate": 0.00046016360376512777, + "loss": 0.3028, + "step": 19388 + }, + { + "epoch": 1.0833030700896724, + "grad_norm": 0.4145281910896301, + "learning_rate": 0.000460135589421784, + "loss": 0.5171, + "step": 19389 + }, + { + "epoch": 1.0833589406933541, + "grad_norm": 6.097256660461426, + "learning_rate": 0.00046010757507844013, + "loss": 0.5124, + "step": 19390 + }, + { + "epoch": 1.083414811297036, + "grad_norm": 0.859180212020874, + "learning_rate": 0.0004600795607350964, + "loss": 0.4534, + "step": 19391 + }, + { + "epoch": 1.0834706819007178, + "grad_norm": 0.39868611097335815, + "learning_rate": 0.00046005154639175254, + "loss": 0.5259, + "step": 19392 + }, + { + "epoch": 1.0835265525043998, + "grad_norm": 0.41302385926246643, + "learning_rate": 0.0004600235320484088, + "loss": 0.3805, + "step": 19393 + }, + { + "epoch": 1.0835824231080817, + "grad_norm": 0.32587364315986633, + "learning_rate": 0.000459995517705065, + "loss": 0.4184, + "step": 19394 + }, + { + "epoch": 1.0836382937117635, + "grad_norm": 0.4171140193939209, + "learning_rate": 0.0004599675033617212, + "loss": 0.459, + "step": 19395 + }, + { + "epoch": 1.0836941643154454, + "grad_norm": 0.42654988169670105, + "learning_rate": 0.0004599394890183774, + "loss": 0.408, + "step": 19396 + }, + { + "epoch": 1.0837500349191274, + "grad_norm": 0.4097774624824524, + "learning_rate": 0.0004599114746750336, + "loss": 0.4014, + "step": 19397 + }, + { + "epoch": 1.083805905522809, + "grad_norm": 0.571841835975647, + "learning_rate": 0.00045988346033168983, + "loss": 0.3808, + "step": 19398 + }, + { + "epoch": 1.083861776126491, + "grad_norm": 0.40053820610046387, + "learning_rate": 0.00045985544598834603, + "loss": 0.3475, + "step": 19399 + }, + { + "epoch": 1.083917646730173, + "grad_norm": 0.5500874519348145, + "learning_rate": 0.00045982743164500224, + "loss": 0.2994, + "step": 19400 + }, + { + "epoch": 1.0839735173338547, + "grad_norm": 2.3806307315826416, + "learning_rate": 0.0004597994173016585, + "loss": 0.4264, + "step": 19401 + }, + { + "epoch": 1.0840293879375367, + "grad_norm": 0.602206826210022, + "learning_rate": 0.00045977140295831465, + "loss": 0.3972, + "step": 19402 + }, + { + "epoch": 1.0840852585412186, + "grad_norm": 0.4672526717185974, + "learning_rate": 0.0004597433886149709, + "loss": 0.389, + "step": 19403 + }, + { + "epoch": 1.0841411291449004, + "grad_norm": 0.7695478200912476, + "learning_rate": 0.00045971537427162706, + "loss": 0.3473, + "step": 19404 + }, + { + "epoch": 1.0841969997485823, + "grad_norm": 0.7442868947982788, + "learning_rate": 0.0004596873599282833, + "loss": 0.4539, + "step": 19405 + }, + { + "epoch": 1.0842528703522643, + "grad_norm": 0.8429576754570007, + "learning_rate": 0.0004596593455849395, + "loss": 0.3782, + "step": 19406 + }, + { + "epoch": 1.084308740955946, + "grad_norm": 0.3897945284843445, + "learning_rate": 0.00045963133124159573, + "loss": 0.3898, + "step": 19407 + }, + { + "epoch": 1.084364611559628, + "grad_norm": 0.3653881251811981, + "learning_rate": 0.0004596033168982519, + "loss": 0.3853, + "step": 19408 + }, + { + "epoch": 1.0844204821633099, + "grad_norm": 0.48017022013664246, + "learning_rate": 0.00045957530255490815, + "loss": 0.3531, + "step": 19409 + }, + { + "epoch": 1.0844763527669916, + "grad_norm": 0.46152999997138977, + "learning_rate": 0.00045954728821156435, + "loss": 0.3865, + "step": 19410 + }, + { + "epoch": 1.0845322233706736, + "grad_norm": 0.37047210335731506, + "learning_rate": 0.0004595192738682205, + "loss": 0.3481, + "step": 19411 + }, + { + "epoch": 1.0845880939743553, + "grad_norm": 1.0747509002685547, + "learning_rate": 0.00045949125952487676, + "loss": 0.3875, + "step": 19412 + }, + { + "epoch": 1.0846439645780372, + "grad_norm": 1.7693735361099243, + "learning_rate": 0.0004594632451815329, + "loss": 0.4516, + "step": 19413 + }, + { + "epoch": 1.0846998351817192, + "grad_norm": 0.5604301691055298, + "learning_rate": 0.0004594352308381892, + "loss": 0.4505, + "step": 19414 + }, + { + "epoch": 1.084755705785401, + "grad_norm": 0.5268290042877197, + "learning_rate": 0.0004594072164948453, + "loss": 0.5398, + "step": 19415 + }, + { + "epoch": 1.0848115763890829, + "grad_norm": 0.48359215259552, + "learning_rate": 0.0004593792021515016, + "loss": 0.4318, + "step": 19416 + }, + { + "epoch": 1.0848674469927648, + "grad_norm": 0.5314176678657532, + "learning_rate": 0.0004593511878081578, + "loss": 0.6057, + "step": 19417 + }, + { + "epoch": 1.0849233175964466, + "grad_norm": 0.523352861404419, + "learning_rate": 0.000459323173464814, + "loss": 0.6289, + "step": 19418 + }, + { + "epoch": 1.0849791882001285, + "grad_norm": 0.41284245252609253, + "learning_rate": 0.0004592951591214702, + "loss": 0.4756, + "step": 19419 + }, + { + "epoch": 1.0850350588038105, + "grad_norm": 0.4672359228134155, + "learning_rate": 0.0004592671447781264, + "loss": 0.4134, + "step": 19420 + }, + { + "epoch": 1.0850909294074922, + "grad_norm": 12.13719367980957, + "learning_rate": 0.0004592391304347826, + "loss": 0.4952, + "step": 19421 + }, + { + "epoch": 1.0851468000111741, + "grad_norm": 0.4412294626235962, + "learning_rate": 0.0004592111160914388, + "loss": 0.4747, + "step": 19422 + }, + { + "epoch": 1.085202670614856, + "grad_norm": 0.48301881551742554, + "learning_rate": 0.00045918310174809503, + "loss": 0.4372, + "step": 19423 + }, + { + "epoch": 1.0852585412185378, + "grad_norm": 3.952888011932373, + "learning_rate": 0.00045915508740475123, + "loss": 0.4546, + "step": 19424 + }, + { + "epoch": 1.0853144118222198, + "grad_norm": 0.5878037810325623, + "learning_rate": 0.00045912707306140744, + "loss": 0.4676, + "step": 19425 + }, + { + "epoch": 1.0853702824259015, + "grad_norm": 0.6182078123092651, + "learning_rate": 0.0004590990587180637, + "loss": 0.461, + "step": 19426 + }, + { + "epoch": 1.0854261530295835, + "grad_norm": 0.45193150639533997, + "learning_rate": 0.00045907104437471985, + "loss": 0.4451, + "step": 19427 + }, + { + "epoch": 1.0854820236332654, + "grad_norm": 0.5326424241065979, + "learning_rate": 0.0004590430300313761, + "loss": 0.4382, + "step": 19428 + }, + { + "epoch": 1.0855378942369471, + "grad_norm": 0.4475398361682892, + "learning_rate": 0.00045901501568803226, + "loss": 0.3752, + "step": 19429 + }, + { + "epoch": 1.085593764840629, + "grad_norm": 0.5718649625778198, + "learning_rate": 0.0004589870013446885, + "loss": 0.5111, + "step": 19430 + }, + { + "epoch": 1.085649635444311, + "grad_norm": 0.657557487487793, + "learning_rate": 0.0004589589870013447, + "loss": 0.4904, + "step": 19431 + }, + { + "epoch": 1.0857055060479928, + "grad_norm": 0.43080055713653564, + "learning_rate": 0.00045893097265800093, + "loss": 0.3269, + "step": 19432 + }, + { + "epoch": 1.0857613766516747, + "grad_norm": 0.39890143275260925, + "learning_rate": 0.00045890295831465714, + "loss": 0.4731, + "step": 19433 + }, + { + "epoch": 1.0858172472553567, + "grad_norm": 0.30651259422302246, + "learning_rate": 0.0004588749439713133, + "loss": 0.3069, + "step": 19434 + }, + { + "epoch": 1.0858731178590384, + "grad_norm": 1.2654685974121094, + "learning_rate": 0.00045884692962796955, + "loss": 0.6002, + "step": 19435 + }, + { + "epoch": 1.0859289884627203, + "grad_norm": 0.6239588260650635, + "learning_rate": 0.0004588189152846257, + "loss": 0.5217, + "step": 19436 + }, + { + "epoch": 1.0859848590664023, + "grad_norm": 0.622877836227417, + "learning_rate": 0.00045879090094128196, + "loss": 0.4357, + "step": 19437 + }, + { + "epoch": 1.086040729670084, + "grad_norm": 0.3613218665122986, + "learning_rate": 0.0004587628865979381, + "loss": 0.453, + "step": 19438 + }, + { + "epoch": 1.086096600273766, + "grad_norm": 0.4058077037334442, + "learning_rate": 0.0004587348722545944, + "loss": 0.3729, + "step": 19439 + }, + { + "epoch": 1.086152470877448, + "grad_norm": 0.643984317779541, + "learning_rate": 0.0004587068579112505, + "loss": 0.5505, + "step": 19440 + }, + { + "epoch": 1.0862083414811297, + "grad_norm": 0.7507803440093994, + "learning_rate": 0.0004586788435679068, + "loss": 0.4237, + "step": 19441 + }, + { + "epoch": 1.0862642120848116, + "grad_norm": 3.7929139137268066, + "learning_rate": 0.000458650829224563, + "loss": 0.4326, + "step": 19442 + }, + { + "epoch": 1.0863200826884936, + "grad_norm": 0.6678999066352844, + "learning_rate": 0.0004586228148812192, + "loss": 0.3695, + "step": 19443 + }, + { + "epoch": 1.0863759532921753, + "grad_norm": 0.6433736681938171, + "learning_rate": 0.0004585948005378754, + "loss": 0.3713, + "step": 19444 + }, + { + "epoch": 1.0864318238958572, + "grad_norm": 0.42513662576675415, + "learning_rate": 0.0004585667861945316, + "loss": 0.4379, + "step": 19445 + }, + { + "epoch": 1.086487694499539, + "grad_norm": 1.4313859939575195, + "learning_rate": 0.0004585387718511878, + "loss": 0.4539, + "step": 19446 + }, + { + "epoch": 1.086543565103221, + "grad_norm": 0.4140970706939697, + "learning_rate": 0.000458510757507844, + "loss": 0.3592, + "step": 19447 + }, + { + "epoch": 1.0865994357069029, + "grad_norm": 0.48800191283226013, + "learning_rate": 0.00045848274316450023, + "loss": 0.4574, + "step": 19448 + }, + { + "epoch": 1.0866553063105846, + "grad_norm": 0.5730022192001343, + "learning_rate": 0.0004584547288211565, + "loss": 0.6091, + "step": 19449 + }, + { + "epoch": 1.0867111769142666, + "grad_norm": 0.41120386123657227, + "learning_rate": 0.00045842671447781264, + "loss": 0.3962, + "step": 19450 + }, + { + "epoch": 1.0867670475179485, + "grad_norm": 0.5400816798210144, + "learning_rate": 0.0004583987001344689, + "loss": 0.3834, + "step": 19451 + }, + { + "epoch": 1.0868229181216302, + "grad_norm": 0.47792235016822815, + "learning_rate": 0.00045837068579112505, + "loss": 0.4035, + "step": 19452 + }, + { + "epoch": 1.0868787887253122, + "grad_norm": 0.4146227240562439, + "learning_rate": 0.0004583426714477813, + "loss": 0.4021, + "step": 19453 + }, + { + "epoch": 1.0869346593289941, + "grad_norm": 0.38369935750961304, + "learning_rate": 0.00045831465710443746, + "loss": 0.3957, + "step": 19454 + }, + { + "epoch": 1.0869905299326759, + "grad_norm": 0.3994603455066681, + "learning_rate": 0.00045828664276109367, + "loss": 0.3361, + "step": 19455 + }, + { + "epoch": 1.0870464005363578, + "grad_norm": 0.3867556154727936, + "learning_rate": 0.00045825862841774993, + "loss": 0.3982, + "step": 19456 + }, + { + "epoch": 1.0871022711400398, + "grad_norm": 0.3616637587547302, + "learning_rate": 0.0004582306140744061, + "loss": 0.4158, + "step": 19457 + }, + { + "epoch": 1.0871581417437215, + "grad_norm": 0.42498713731765747, + "learning_rate": 0.00045820259973106234, + "loss": 0.4507, + "step": 19458 + }, + { + "epoch": 1.0872140123474034, + "grad_norm": 4.2367262840271, + "learning_rate": 0.0004581745853877185, + "loss": 0.4858, + "step": 19459 + }, + { + "epoch": 1.0872698829510852, + "grad_norm": 0.7230790257453918, + "learning_rate": 0.00045814657104437475, + "loss": 0.4155, + "step": 19460 + }, + { + "epoch": 1.0873257535547671, + "grad_norm": 0.38270702958106995, + "learning_rate": 0.0004581185567010309, + "loss": 0.4893, + "step": 19461 + }, + { + "epoch": 1.087381624158449, + "grad_norm": 0.46371662616729736, + "learning_rate": 0.00045809054235768716, + "loss": 0.3845, + "step": 19462 + }, + { + "epoch": 1.0874374947621308, + "grad_norm": 2.364112138748169, + "learning_rate": 0.0004580625280143433, + "loss": 0.4578, + "step": 19463 + }, + { + "epoch": 1.0874933653658128, + "grad_norm": 0.6127414107322693, + "learning_rate": 0.0004580345136709996, + "loss": 0.4469, + "step": 19464 + }, + { + "epoch": 1.0875492359694947, + "grad_norm": 1.106184959411621, + "learning_rate": 0.0004580064993276558, + "loss": 0.4511, + "step": 19465 + }, + { + "epoch": 1.0876051065731764, + "grad_norm": 0.3790120780467987, + "learning_rate": 0.000457978484984312, + "loss": 0.3874, + "step": 19466 + }, + { + "epoch": 1.0876609771768584, + "grad_norm": 0.4226793050765991, + "learning_rate": 0.0004579504706409682, + "loss": 0.3822, + "step": 19467 + }, + { + "epoch": 1.0877168477805403, + "grad_norm": 0.566379725933075, + "learning_rate": 0.0004579224562976244, + "loss": 0.4041, + "step": 19468 + }, + { + "epoch": 1.087772718384222, + "grad_norm": 0.48969724774360657, + "learning_rate": 0.0004578944419542806, + "loss": 0.4748, + "step": 19469 + }, + { + "epoch": 1.087828588987904, + "grad_norm": 0.49936342239379883, + "learning_rate": 0.0004578664276109368, + "loss": 0.4357, + "step": 19470 + }, + { + "epoch": 1.087884459591586, + "grad_norm": 0.3116365373134613, + "learning_rate": 0.000457838413267593, + "loss": 0.2969, + "step": 19471 + }, + { + "epoch": 1.0879403301952677, + "grad_norm": 0.40517884492874146, + "learning_rate": 0.0004578103989242493, + "loss": 0.4686, + "step": 19472 + }, + { + "epoch": 1.0879962007989497, + "grad_norm": 0.4231486916542053, + "learning_rate": 0.0004577823845809054, + "loss": 0.3835, + "step": 19473 + }, + { + "epoch": 1.0880520714026316, + "grad_norm": 0.9137679934501648, + "learning_rate": 0.0004577543702375617, + "loss": 0.4874, + "step": 19474 + }, + { + "epoch": 1.0881079420063133, + "grad_norm": 0.5660406351089478, + "learning_rate": 0.00045772635589421784, + "loss": 0.5752, + "step": 19475 + }, + { + "epoch": 1.0881638126099953, + "grad_norm": 0.7401509284973145, + "learning_rate": 0.00045769834155087404, + "loss": 0.4647, + "step": 19476 + }, + { + "epoch": 1.0882196832136772, + "grad_norm": 0.4582013189792633, + "learning_rate": 0.00045767032720753025, + "loss": 0.3826, + "step": 19477 + }, + { + "epoch": 1.088275553817359, + "grad_norm": 0.48891860246658325, + "learning_rate": 0.00045764231286418646, + "loss": 0.3743, + "step": 19478 + }, + { + "epoch": 1.088331424421041, + "grad_norm": 0.48586755990982056, + "learning_rate": 0.00045761429852084266, + "loss": 0.3786, + "step": 19479 + }, + { + "epoch": 1.0883872950247226, + "grad_norm": 0.4168640673160553, + "learning_rate": 0.00045758628417749887, + "loss": 0.4062, + "step": 19480 + }, + { + "epoch": 1.0884431656284046, + "grad_norm": 0.3742636740207672, + "learning_rate": 0.00045755826983415513, + "loss": 0.4606, + "step": 19481 + }, + { + "epoch": 1.0884990362320865, + "grad_norm": 0.798538327217102, + "learning_rate": 0.0004575302554908113, + "loss": 0.4175, + "step": 19482 + }, + { + "epoch": 1.0885549068357683, + "grad_norm": 2.2356009483337402, + "learning_rate": 0.00045750224114746754, + "loss": 0.4774, + "step": 19483 + }, + { + "epoch": 1.0886107774394502, + "grad_norm": 0.6212634444236755, + "learning_rate": 0.0004574742268041237, + "loss": 0.4378, + "step": 19484 + }, + { + "epoch": 1.0886666480431322, + "grad_norm": 0.3940389156341553, + "learning_rate": 0.00045744621246077995, + "loss": 0.3345, + "step": 19485 + }, + { + "epoch": 1.088722518646814, + "grad_norm": 0.4694225490093231, + "learning_rate": 0.0004574181981174361, + "loss": 0.4365, + "step": 19486 + }, + { + "epoch": 1.0887783892504959, + "grad_norm": 0.40518349409103394, + "learning_rate": 0.00045739018377409236, + "loss": 0.4008, + "step": 19487 + }, + { + "epoch": 1.0888342598541778, + "grad_norm": 0.36529451608657837, + "learning_rate": 0.00045736216943074857, + "loss": 0.4326, + "step": 19488 + }, + { + "epoch": 1.0888901304578595, + "grad_norm": 0.45307719707489014, + "learning_rate": 0.0004573341550874048, + "loss": 0.3709, + "step": 19489 + }, + { + "epoch": 1.0889460010615415, + "grad_norm": 0.5831955671310425, + "learning_rate": 0.000457306140744061, + "loss": 0.4026, + "step": 19490 + }, + { + "epoch": 1.0890018716652234, + "grad_norm": 0.6113218665122986, + "learning_rate": 0.0004572781264007172, + "loss": 0.4573, + "step": 19491 + }, + { + "epoch": 1.0890577422689052, + "grad_norm": 0.5052394270896912, + "learning_rate": 0.0004572501120573734, + "loss": 0.3286, + "step": 19492 + }, + { + "epoch": 1.0891136128725871, + "grad_norm": 0.4205191135406494, + "learning_rate": 0.0004572220977140296, + "loss": 0.4001, + "step": 19493 + }, + { + "epoch": 1.0891694834762689, + "grad_norm": 0.5979551672935486, + "learning_rate": 0.0004571940833706858, + "loss": 0.4528, + "step": 19494 + }, + { + "epoch": 1.0892253540799508, + "grad_norm": 0.427619993686676, + "learning_rate": 0.000457166069027342, + "loss": 0.37, + "step": 19495 + }, + { + "epoch": 1.0892812246836328, + "grad_norm": 0.49746620655059814, + "learning_rate": 0.0004571380546839982, + "loss": 0.4702, + "step": 19496 + }, + { + "epoch": 1.0893370952873145, + "grad_norm": 0.5719438791275024, + "learning_rate": 0.0004571100403406544, + "loss": 0.5401, + "step": 19497 + }, + { + "epoch": 1.0893929658909964, + "grad_norm": 3.220961093902588, + "learning_rate": 0.0004570820259973106, + "loss": 0.4628, + "step": 19498 + }, + { + "epoch": 1.0894488364946784, + "grad_norm": 0.488925963640213, + "learning_rate": 0.00045705401165396683, + "loss": 0.4612, + "step": 19499 + }, + { + "epoch": 1.0895047070983601, + "grad_norm": 1.0968319177627563, + "learning_rate": 0.00045702599731062304, + "loss": 0.7299, + "step": 19500 + }, + { + "epoch": 1.0895047070983601, + "eval_cer": 0.08818041962619881, + "eval_loss": 0.3291606903076172, + "eval_runtime": 56.6047, + "eval_samples_per_second": 80.17, + "eval_steps_per_second": 5.017, + "eval_wer": 0.35177788329934917, + "step": 19500 + }, + { + "epoch": 1.089560577702042, + "grad_norm": 0.6281291246414185, + "learning_rate": 0.00045699798296727924, + "loss": 0.6179, + "step": 19501 + }, + { + "epoch": 1.089616448305724, + "grad_norm": 0.42950165271759033, + "learning_rate": 0.00045696996862393545, + "loss": 0.5707, + "step": 19502 + }, + { + "epoch": 1.0896723189094057, + "grad_norm": 1.5872550010681152, + "learning_rate": 0.00045694195428059166, + "loss": 0.5505, + "step": 19503 + }, + { + "epoch": 1.0897281895130877, + "grad_norm": 0.3719429671764374, + "learning_rate": 0.0004569139399372479, + "loss": 0.4069, + "step": 19504 + }, + { + "epoch": 1.0897840601167696, + "grad_norm": 0.34997299313545227, + "learning_rate": 0.00045688592559390407, + "loss": 0.3453, + "step": 19505 + }, + { + "epoch": 1.0898399307204514, + "grad_norm": 0.6326642632484436, + "learning_rate": 0.0004568579112505603, + "loss": 0.3801, + "step": 19506 + }, + { + "epoch": 1.0898958013241333, + "grad_norm": 0.46285223960876465, + "learning_rate": 0.0004568298969072165, + "loss": 0.4393, + "step": 19507 + }, + { + "epoch": 1.0899516719278153, + "grad_norm": 0.3746400475502014, + "learning_rate": 0.00045680188256387274, + "loss": 0.4344, + "step": 19508 + }, + { + "epoch": 1.090007542531497, + "grad_norm": 0.4613138735294342, + "learning_rate": 0.0004567738682205289, + "loss": 0.5826, + "step": 19509 + }, + { + "epoch": 1.090063413135179, + "grad_norm": 0.7739976048469543, + "learning_rate": 0.00045674585387718515, + "loss": 0.432, + "step": 19510 + }, + { + "epoch": 1.090119283738861, + "grad_norm": 0.5308918952941895, + "learning_rate": 0.0004567178395338413, + "loss": 0.4177, + "step": 19511 + }, + { + "epoch": 1.0901751543425426, + "grad_norm": 0.4611074924468994, + "learning_rate": 0.00045668982519049756, + "loss": 0.5392, + "step": 19512 + }, + { + "epoch": 1.0902310249462246, + "grad_norm": 0.3961004316806793, + "learning_rate": 0.00045666181084715377, + "loss": 0.4498, + "step": 19513 + }, + { + "epoch": 1.0902868955499063, + "grad_norm": 0.3958984613418579, + "learning_rate": 0.00045663379650381, + "loss": 0.3797, + "step": 19514 + }, + { + "epoch": 1.0903427661535883, + "grad_norm": 0.4458449184894562, + "learning_rate": 0.0004566057821604662, + "loss": 0.4652, + "step": 19515 + }, + { + "epoch": 1.0903986367572702, + "grad_norm": 0.7972674369812012, + "learning_rate": 0.0004565777678171224, + "loss": 0.3831, + "step": 19516 + }, + { + "epoch": 1.090454507360952, + "grad_norm": 0.38797345757484436, + "learning_rate": 0.0004565497534737786, + "loss": 0.4085, + "step": 19517 + }, + { + "epoch": 1.090510377964634, + "grad_norm": 0.49537405371665955, + "learning_rate": 0.00045652173913043474, + "loss": 0.3883, + "step": 19518 + }, + { + "epoch": 1.0905662485683159, + "grad_norm": 0.468987375497818, + "learning_rate": 0.000456493724787091, + "loss": 0.4892, + "step": 19519 + }, + { + "epoch": 1.0906221191719976, + "grad_norm": 0.40774989128112793, + "learning_rate": 0.0004564657104437472, + "loss": 0.4067, + "step": 19520 + }, + { + "epoch": 1.0906779897756795, + "grad_norm": 0.5556089282035828, + "learning_rate": 0.0004564376961004034, + "loss": 0.344, + "step": 19521 + }, + { + "epoch": 1.0907338603793615, + "grad_norm": 4.741180896759033, + "learning_rate": 0.0004564096817570596, + "loss": 0.4724, + "step": 19522 + }, + { + "epoch": 1.0907897309830432, + "grad_norm": 0.3891945481300354, + "learning_rate": 0.0004563816674137158, + "loss": 0.3877, + "step": 19523 + }, + { + "epoch": 1.0908456015867252, + "grad_norm": 0.5329642295837402, + "learning_rate": 0.00045635365307037203, + "loss": 0.4365, + "step": 19524 + }, + { + "epoch": 1.0909014721904071, + "grad_norm": 0.357785701751709, + "learning_rate": 0.00045632563872702824, + "loss": 0.4274, + "step": 19525 + }, + { + "epoch": 1.0909573427940888, + "grad_norm": 0.39677175879478455, + "learning_rate": 0.00045629762438368444, + "loss": 0.4414, + "step": 19526 + }, + { + "epoch": 1.0910132133977708, + "grad_norm": 0.34202855825424194, + "learning_rate": 0.0004562696100403407, + "loss": 0.3797, + "step": 19527 + }, + { + "epoch": 1.0910690840014525, + "grad_norm": 0.42772042751312256, + "learning_rate": 0.00045624159569699685, + "loss": 0.4651, + "step": 19528 + }, + { + "epoch": 1.0911249546051345, + "grad_norm": 9.646693229675293, + "learning_rate": 0.0004562135813536531, + "loss": 0.3432, + "step": 19529 + }, + { + "epoch": 1.0911808252088164, + "grad_norm": 0.4922870397567749, + "learning_rate": 0.00045618556701030927, + "loss": 0.566, + "step": 19530 + }, + { + "epoch": 1.0912366958124982, + "grad_norm": 1.7158242464065552, + "learning_rate": 0.0004561575526669655, + "loss": 0.4854, + "step": 19531 + }, + { + "epoch": 1.09129256641618, + "grad_norm": 0.6403611898422241, + "learning_rate": 0.0004561295383236217, + "loss": 0.5236, + "step": 19532 + }, + { + "epoch": 1.091348437019862, + "grad_norm": 1.294723391532898, + "learning_rate": 0.00045610152398027794, + "loss": 0.4034, + "step": 19533 + }, + { + "epoch": 1.0914043076235438, + "grad_norm": 0.3881399929523468, + "learning_rate": 0.0004560735096369341, + "loss": 0.3439, + "step": 19534 + }, + { + "epoch": 1.0914601782272257, + "grad_norm": 0.5107452869415283, + "learning_rate": 0.00045604549529359035, + "loss": 0.4309, + "step": 19535 + }, + { + "epoch": 1.0915160488309077, + "grad_norm": 0.5677504539489746, + "learning_rate": 0.00045601748095024656, + "loss": 0.4244, + "step": 19536 + }, + { + "epoch": 1.0915719194345894, + "grad_norm": 0.6334575414657593, + "learning_rate": 0.00045598946660690276, + "loss": 0.5674, + "step": 19537 + }, + { + "epoch": 1.0916277900382714, + "grad_norm": 0.3621346652507782, + "learning_rate": 0.00045596145226355897, + "loss": 0.3361, + "step": 19538 + }, + { + "epoch": 1.0916836606419533, + "grad_norm": 0.3647255003452301, + "learning_rate": 0.0004559334379202151, + "loss": 0.435, + "step": 19539 + }, + { + "epoch": 1.091739531245635, + "grad_norm": 0.41461485624313354, + "learning_rate": 0.0004559054235768714, + "loss": 0.3376, + "step": 19540 + }, + { + "epoch": 1.091795401849317, + "grad_norm": 0.3627227246761322, + "learning_rate": 0.00045587740923352753, + "loss": 0.3401, + "step": 19541 + }, + { + "epoch": 1.091851272452999, + "grad_norm": 0.4324037432670593, + "learning_rate": 0.0004558493948901838, + "loss": 0.4629, + "step": 19542 + }, + { + "epoch": 1.0919071430566807, + "grad_norm": 0.40098729729652405, + "learning_rate": 0.00045582138054684, + "loss": 0.3537, + "step": 19543 + }, + { + "epoch": 1.0919630136603626, + "grad_norm": 0.4586426913738251, + "learning_rate": 0.0004557933662034962, + "loss": 0.5498, + "step": 19544 + }, + { + "epoch": 1.0920188842640446, + "grad_norm": 0.5332663059234619, + "learning_rate": 0.0004557653518601524, + "loss": 0.3564, + "step": 19545 + }, + { + "epoch": 1.0920747548677263, + "grad_norm": 0.5124722719192505, + "learning_rate": 0.0004557373375168086, + "loss": 0.4887, + "step": 19546 + }, + { + "epoch": 1.0921306254714083, + "grad_norm": 0.38945069909095764, + "learning_rate": 0.0004557093231734648, + "loss": 0.4694, + "step": 19547 + }, + { + "epoch": 1.09218649607509, + "grad_norm": 0.5450897216796875, + "learning_rate": 0.000455681308830121, + "loss": 0.4593, + "step": 19548 + }, + { + "epoch": 1.092242366678772, + "grad_norm": 0.3554268181324005, + "learning_rate": 0.00045565329448677723, + "loss": 0.4479, + "step": 19549 + }, + { + "epoch": 1.092298237282454, + "grad_norm": 0.7337926626205444, + "learning_rate": 0.00045562528014343344, + "loss": 0.5158, + "step": 19550 + }, + { + "epoch": 1.0923541078861356, + "grad_norm": 0.5659620761871338, + "learning_rate": 0.00045559726580008964, + "loss": 0.3396, + "step": 19551 + }, + { + "epoch": 1.0924099784898176, + "grad_norm": 0.42387762665748596, + "learning_rate": 0.0004555692514567459, + "loss": 0.4838, + "step": 19552 + }, + { + "epoch": 1.0924658490934995, + "grad_norm": 0.548486590385437, + "learning_rate": 0.00045554123711340205, + "loss": 0.4267, + "step": 19553 + }, + { + "epoch": 1.0925217196971813, + "grad_norm": 0.7935689091682434, + "learning_rate": 0.0004555132227700583, + "loss": 0.4286, + "step": 19554 + }, + { + "epoch": 1.0925775903008632, + "grad_norm": 1.0256390571594238, + "learning_rate": 0.00045548520842671447, + "loss": 0.5049, + "step": 19555 + }, + { + "epoch": 1.0926334609045452, + "grad_norm": 0.5775803327560425, + "learning_rate": 0.0004554571940833707, + "loss": 0.3409, + "step": 19556 + }, + { + "epoch": 1.092689331508227, + "grad_norm": 0.6405089497566223, + "learning_rate": 0.0004554291797400269, + "loss": 0.4098, + "step": 19557 + }, + { + "epoch": 1.0927452021119088, + "grad_norm": 0.4300435483455658, + "learning_rate": 0.00045540116539668314, + "loss": 0.3639, + "step": 19558 + }, + { + "epoch": 1.0928010727155908, + "grad_norm": 0.537250816822052, + "learning_rate": 0.00045537315105333934, + "loss": 0.5165, + "step": 19559 + }, + { + "epoch": 1.0928569433192725, + "grad_norm": 3.401348352432251, + "learning_rate": 0.0004553451367099955, + "loss": 0.4475, + "step": 19560 + }, + { + "epoch": 1.0929128139229545, + "grad_norm": 0.3646322190761566, + "learning_rate": 0.00045531712236665176, + "loss": 0.3491, + "step": 19561 + }, + { + "epoch": 1.0929686845266362, + "grad_norm": 0.431272953748703, + "learning_rate": 0.0004552891080233079, + "loss": 0.4187, + "step": 19562 + }, + { + "epoch": 1.0930245551303182, + "grad_norm": 0.41462984681129456, + "learning_rate": 0.00045526109367996417, + "loss": 0.4422, + "step": 19563 + }, + { + "epoch": 1.093080425734, + "grad_norm": 0.48021960258483887, + "learning_rate": 0.0004552330793366203, + "loss": 0.4035, + "step": 19564 + }, + { + "epoch": 1.0931362963376818, + "grad_norm": 0.6781880259513855, + "learning_rate": 0.0004552050649932766, + "loss": 0.3992, + "step": 19565 + }, + { + "epoch": 1.0931921669413638, + "grad_norm": 0.45848289132118225, + "learning_rate": 0.00045517705064993273, + "loss": 0.4348, + "step": 19566 + }, + { + "epoch": 1.0932480375450457, + "grad_norm": 0.42103150486946106, + "learning_rate": 0.000455149036306589, + "loss": 0.4476, + "step": 19567 + }, + { + "epoch": 1.0933039081487275, + "grad_norm": 0.38293683528900146, + "learning_rate": 0.0004551210219632452, + "loss": 0.3834, + "step": 19568 + }, + { + "epoch": 1.0933597787524094, + "grad_norm": 0.7783222198486328, + "learning_rate": 0.0004550930076199014, + "loss": 0.3581, + "step": 19569 + }, + { + "epoch": 1.0934156493560914, + "grad_norm": 0.6755350232124329, + "learning_rate": 0.0004550649932765576, + "loss": 0.6379, + "step": 19570 + }, + { + "epoch": 1.093471519959773, + "grad_norm": 0.5105915665626526, + "learning_rate": 0.0004550369789332138, + "loss": 0.3858, + "step": 19571 + }, + { + "epoch": 1.093527390563455, + "grad_norm": 0.5327539443969727, + "learning_rate": 0.00045500896458987, + "loss": 0.4295, + "step": 19572 + }, + { + "epoch": 1.093583261167137, + "grad_norm": 0.37190064787864685, + "learning_rate": 0.0004549809502465262, + "loss": 0.34, + "step": 19573 + }, + { + "epoch": 1.0936391317708187, + "grad_norm": 0.42814186215400696, + "learning_rate": 0.00045495293590318243, + "loss": 0.3656, + "step": 19574 + }, + { + "epoch": 1.0936950023745007, + "grad_norm": 0.42481422424316406, + "learning_rate": 0.0004549249215598387, + "loss": 0.4502, + "step": 19575 + }, + { + "epoch": 1.0937508729781826, + "grad_norm": 0.8303750157356262, + "learning_rate": 0.00045489690721649484, + "loss": 0.4086, + "step": 19576 + }, + { + "epoch": 1.0938067435818644, + "grad_norm": 0.44617676734924316, + "learning_rate": 0.0004548688928731511, + "loss": 0.431, + "step": 19577 + }, + { + "epoch": 1.0938626141855463, + "grad_norm": 0.43851056694984436, + "learning_rate": 0.00045484087852980725, + "loss": 0.4095, + "step": 19578 + }, + { + "epoch": 1.0939184847892283, + "grad_norm": 0.609786331653595, + "learning_rate": 0.0004548128641864635, + "loss": 0.367, + "step": 19579 + }, + { + "epoch": 1.09397435539291, + "grad_norm": 0.4363541007041931, + "learning_rate": 0.00045478484984311967, + "loss": 0.3647, + "step": 19580 + }, + { + "epoch": 1.094030225996592, + "grad_norm": 0.3334774374961853, + "learning_rate": 0.00045475683549977587, + "loss": 0.4403, + "step": 19581 + }, + { + "epoch": 1.0940860966002737, + "grad_norm": 0.5282602906227112, + "learning_rate": 0.0004547288211564321, + "loss": 0.5295, + "step": 19582 + }, + { + "epoch": 1.0941419672039556, + "grad_norm": 0.4652773439884186, + "learning_rate": 0.0004547008068130883, + "loss": 0.4555, + "step": 19583 + }, + { + "epoch": 1.0941978378076376, + "grad_norm": 0.4471600353717804, + "learning_rate": 0.00045467279246974454, + "loss": 0.5176, + "step": 19584 + }, + { + "epoch": 1.0942537084113193, + "grad_norm": 0.7212729454040527, + "learning_rate": 0.0004546447781264007, + "loss": 0.4775, + "step": 19585 + }, + { + "epoch": 1.0943095790150013, + "grad_norm": 0.4560784101486206, + "learning_rate": 0.00045461676378305695, + "loss": 0.5056, + "step": 19586 + }, + { + "epoch": 1.0943654496186832, + "grad_norm": 0.38773655891418457, + "learning_rate": 0.0004545887494397131, + "loss": 0.3425, + "step": 19587 + }, + { + "epoch": 1.094421320222365, + "grad_norm": 0.4570973813533783, + "learning_rate": 0.00045456073509636937, + "loss": 0.4278, + "step": 19588 + }, + { + "epoch": 1.0944771908260469, + "grad_norm": 1.2752594947814941, + "learning_rate": 0.0004545327207530255, + "loss": 0.4504, + "step": 19589 + }, + { + "epoch": 1.0945330614297288, + "grad_norm": 0.8481886386871338, + "learning_rate": 0.0004545047064096818, + "loss": 0.5046, + "step": 19590 + }, + { + "epoch": 1.0945889320334106, + "grad_norm": 0.5596429109573364, + "learning_rate": 0.000454476692066338, + "loss": 0.5136, + "step": 19591 + }, + { + "epoch": 1.0946448026370925, + "grad_norm": 0.6796489953994751, + "learning_rate": 0.0004544486777229942, + "loss": 0.4586, + "step": 19592 + }, + { + "epoch": 1.0947006732407745, + "grad_norm": 0.4904266893863678, + "learning_rate": 0.0004544206633796504, + "loss": 0.4334, + "step": 19593 + }, + { + "epoch": 1.0947565438444562, + "grad_norm": 0.4406459331512451, + "learning_rate": 0.0004543926490363066, + "loss": 0.3219, + "step": 19594 + }, + { + "epoch": 1.0948124144481381, + "grad_norm": 28.005796432495117, + "learning_rate": 0.0004543646346929628, + "loss": 0.3694, + "step": 19595 + }, + { + "epoch": 1.0948682850518199, + "grad_norm": 0.5869415402412415, + "learning_rate": 0.000454336620349619, + "loss": 0.5317, + "step": 19596 + }, + { + "epoch": 1.0949241556555018, + "grad_norm": 0.7924975156784058, + "learning_rate": 0.0004543086060062752, + "loss": 0.4603, + "step": 19597 + }, + { + "epoch": 1.0949800262591838, + "grad_norm": 0.46711406111717224, + "learning_rate": 0.0004542805916629315, + "loss": 0.4189, + "step": 19598 + }, + { + "epoch": 1.0950358968628655, + "grad_norm": 0.6233173608779907, + "learning_rate": 0.00045425257731958763, + "loss": 0.5059, + "step": 19599 + }, + { + "epoch": 1.0950917674665475, + "grad_norm": 0.5664699673652649, + "learning_rate": 0.0004542245629762439, + "loss": 0.5143, + "step": 19600 + }, + { + "epoch": 1.0951476380702294, + "grad_norm": 0.6996640563011169, + "learning_rate": 0.00045419654863290004, + "loss": 0.4714, + "step": 19601 + }, + { + "epoch": 1.0952035086739111, + "grad_norm": 1.288151741027832, + "learning_rate": 0.0004541685342895563, + "loss": 0.3917, + "step": 19602 + }, + { + "epoch": 1.095259379277593, + "grad_norm": 0.7906299829483032, + "learning_rate": 0.00045414051994621245, + "loss": 0.3867, + "step": 19603 + }, + { + "epoch": 1.095315249881275, + "grad_norm": 0.37314295768737793, + "learning_rate": 0.00045411250560286866, + "loss": 0.3741, + "step": 19604 + }, + { + "epoch": 1.0953711204849568, + "grad_norm": 0.4824608266353607, + "learning_rate": 0.00045408449125952487, + "loss": 0.4438, + "step": 19605 + }, + { + "epoch": 1.0954269910886387, + "grad_norm": 1.0555307865142822, + "learning_rate": 0.00045405647691618107, + "loss": 0.4832, + "step": 19606 + }, + { + "epoch": 1.0954828616923207, + "grad_norm": 0.46513575315475464, + "learning_rate": 0.00045402846257283733, + "loss": 0.6375, + "step": 19607 + }, + { + "epoch": 1.0955387322960024, + "grad_norm": 0.580273449420929, + "learning_rate": 0.0004540004482294935, + "loss": 0.3277, + "step": 19608 + }, + { + "epoch": 1.0955946028996844, + "grad_norm": 0.43177178502082825, + "learning_rate": 0.00045397243388614974, + "loss": 0.4971, + "step": 19609 + }, + { + "epoch": 1.0956504735033663, + "grad_norm": 0.5409917831420898, + "learning_rate": 0.0004539444195428059, + "loss": 0.4099, + "step": 19610 + }, + { + "epoch": 1.095706344107048, + "grad_norm": 0.48909759521484375, + "learning_rate": 0.00045391640519946215, + "loss": 0.4662, + "step": 19611 + }, + { + "epoch": 1.09576221471073, + "grad_norm": 0.3385774791240692, + "learning_rate": 0.0004538883908561183, + "loss": 0.4402, + "step": 19612 + }, + { + "epoch": 1.095818085314412, + "grad_norm": 1.2612265348434448, + "learning_rate": 0.00045386037651277457, + "loss": 0.486, + "step": 19613 + }, + { + "epoch": 1.0958739559180937, + "grad_norm": 0.8204753398895264, + "learning_rate": 0.00045383236216943077, + "loss": 0.3939, + "step": 19614 + }, + { + "epoch": 1.0959298265217756, + "grad_norm": 0.753221333026886, + "learning_rate": 0.000453804347826087, + "loss": 0.463, + "step": 19615 + }, + { + "epoch": 1.0959856971254573, + "grad_norm": 1.1649911403656006, + "learning_rate": 0.0004537763334827432, + "loss": 0.4107, + "step": 19616 + }, + { + "epoch": 1.0960415677291393, + "grad_norm": 0.8716086149215698, + "learning_rate": 0.0004537483191393994, + "loss": 0.52, + "step": 19617 + }, + { + "epoch": 1.0960974383328213, + "grad_norm": 0.4394444227218628, + "learning_rate": 0.0004537203047960556, + "loss": 0.3952, + "step": 19618 + }, + { + "epoch": 1.096153308936503, + "grad_norm": 4.621092319488525, + "learning_rate": 0.0004536922904527118, + "loss": 0.5526, + "step": 19619 + }, + { + "epoch": 1.096209179540185, + "grad_norm": 0.5830351710319519, + "learning_rate": 0.000453664276109368, + "loss": 0.4563, + "step": 19620 + }, + { + "epoch": 1.0962650501438669, + "grad_norm": 0.3866783678531647, + "learning_rate": 0.0004536362617660242, + "loss": 0.4555, + "step": 19621 + }, + { + "epoch": 1.0963209207475486, + "grad_norm": 1.6238287687301636, + "learning_rate": 0.0004536082474226804, + "loss": 0.4617, + "step": 19622 + }, + { + "epoch": 1.0963767913512306, + "grad_norm": 0.9132727384567261, + "learning_rate": 0.0004535802330793367, + "loss": 0.3051, + "step": 19623 + }, + { + "epoch": 1.0964326619549125, + "grad_norm": 0.37168416380882263, + "learning_rate": 0.00045355221873599283, + "loss": 0.3632, + "step": 19624 + }, + { + "epoch": 1.0964885325585942, + "grad_norm": 0.3616362512111664, + "learning_rate": 0.00045352420439264904, + "loss": 0.4312, + "step": 19625 + }, + { + "epoch": 1.0965444031622762, + "grad_norm": 0.38160228729248047, + "learning_rate": 0.00045349619004930524, + "loss": 0.461, + "step": 19626 + }, + { + "epoch": 1.0966002737659581, + "grad_norm": 0.5318437218666077, + "learning_rate": 0.00045346817570596145, + "loss": 0.3957, + "step": 19627 + }, + { + "epoch": 1.0966561443696399, + "grad_norm": 1.2857364416122437, + "learning_rate": 0.00045344016136261765, + "loss": 0.4845, + "step": 19628 + }, + { + "epoch": 1.0967120149733218, + "grad_norm": 0.40936845541000366, + "learning_rate": 0.00045341214701927386, + "loss": 0.5053, + "step": 19629 + }, + { + "epoch": 1.0967678855770036, + "grad_norm": 0.424205482006073, + "learning_rate": 0.0004533841326759301, + "loss": 0.4904, + "step": 19630 + }, + { + "epoch": 1.0968237561806855, + "grad_norm": 1.2906092405319214, + "learning_rate": 0.00045335611833258627, + "loss": 0.5014, + "step": 19631 + }, + { + "epoch": 1.0968796267843675, + "grad_norm": 0.40481212735176086, + "learning_rate": 0.00045332810398924253, + "loss": 0.4138, + "step": 19632 + }, + { + "epoch": 1.0969354973880492, + "grad_norm": 0.465501993894577, + "learning_rate": 0.0004533000896458987, + "loss": 0.4329, + "step": 19633 + }, + { + "epoch": 1.0969913679917311, + "grad_norm": 0.42561352252960205, + "learning_rate": 0.00045327207530255494, + "loss": 0.4963, + "step": 19634 + }, + { + "epoch": 1.097047238595413, + "grad_norm": 0.36058250069618225, + "learning_rate": 0.0004532440609592111, + "loss": 0.4157, + "step": 19635 + }, + { + "epoch": 1.0971031091990948, + "grad_norm": 0.9731425046920776, + "learning_rate": 0.00045321604661586735, + "loss": 0.3981, + "step": 19636 + }, + { + "epoch": 1.0971589798027768, + "grad_norm": 0.37958523631095886, + "learning_rate": 0.0004531880322725235, + "loss": 0.4493, + "step": 19637 + }, + { + "epoch": 1.0972148504064587, + "grad_norm": 0.42793938517570496, + "learning_rate": 0.00045316001792917977, + "loss": 0.4829, + "step": 19638 + }, + { + "epoch": 1.0972707210101404, + "grad_norm": 0.6708852052688599, + "learning_rate": 0.00045313200358583597, + "loss": 0.4397, + "step": 19639 + }, + { + "epoch": 1.0973265916138224, + "grad_norm": 0.582474946975708, + "learning_rate": 0.0004531039892424922, + "loss": 0.3775, + "step": 19640 + }, + { + "epoch": 1.0973824622175044, + "grad_norm": 0.4538704752922058, + "learning_rate": 0.0004530759748991484, + "loss": 0.4007, + "step": 19641 + }, + { + "epoch": 1.097438332821186, + "grad_norm": 0.46093180775642395, + "learning_rate": 0.0004530479605558046, + "loss": 0.4006, + "step": 19642 + }, + { + "epoch": 1.097494203424868, + "grad_norm": 0.4217790961265564, + "learning_rate": 0.0004530199462124608, + "loss": 0.4282, + "step": 19643 + }, + { + "epoch": 1.09755007402855, + "grad_norm": 0.8671864867210388, + "learning_rate": 0.000452991931869117, + "loss": 0.5238, + "step": 19644 + }, + { + "epoch": 1.0976059446322317, + "grad_norm": 2.507982015609741, + "learning_rate": 0.0004529639175257732, + "loss": 0.3558, + "step": 19645 + }, + { + "epoch": 1.0976618152359137, + "grad_norm": 0.35633528232574463, + "learning_rate": 0.0004529359031824294, + "loss": 0.4679, + "step": 19646 + }, + { + "epoch": 1.0977176858395956, + "grad_norm": 0.5248463749885559, + "learning_rate": 0.0004529078888390856, + "loss": 0.4119, + "step": 19647 + }, + { + "epoch": 1.0977735564432773, + "grad_norm": 0.4898277223110199, + "learning_rate": 0.0004528798744957418, + "loss": 0.4957, + "step": 19648 + }, + { + "epoch": 1.0978294270469593, + "grad_norm": 0.37652650475502014, + "learning_rate": 0.00045285186015239803, + "loss": 0.332, + "step": 19649 + }, + { + "epoch": 1.097885297650641, + "grad_norm": 0.8924794793128967, + "learning_rate": 0.00045282384580905424, + "loss": 0.3487, + "step": 19650 + }, + { + "epoch": 1.097941168254323, + "grad_norm": 0.4595866799354553, + "learning_rate": 0.00045279583146571044, + "loss": 0.5002, + "step": 19651 + }, + { + "epoch": 1.097997038858005, + "grad_norm": 0.4850884675979614, + "learning_rate": 0.00045276781712236665, + "loss": 0.5175, + "step": 19652 + }, + { + "epoch": 1.0980529094616867, + "grad_norm": 0.348032683134079, + "learning_rate": 0.0004527398027790229, + "loss": 0.3302, + "step": 19653 + }, + { + "epoch": 1.0981087800653686, + "grad_norm": 0.4110800623893738, + "learning_rate": 0.00045271178843567906, + "loss": 0.4177, + "step": 19654 + }, + { + "epoch": 1.0981646506690506, + "grad_norm": 0.48468655347824097, + "learning_rate": 0.0004526837740923353, + "loss": 0.5476, + "step": 19655 + }, + { + "epoch": 1.0982205212727323, + "grad_norm": 0.6985436677932739, + "learning_rate": 0.00045265575974899147, + "loss": 0.5085, + "step": 19656 + }, + { + "epoch": 1.0982763918764142, + "grad_norm": 0.3505488932132721, + "learning_rate": 0.00045262774540564773, + "loss": 0.3547, + "step": 19657 + }, + { + "epoch": 1.0983322624800962, + "grad_norm": 0.4968234896659851, + "learning_rate": 0.0004525997310623039, + "loss": 0.4865, + "step": 19658 + }, + { + "epoch": 1.098388133083778, + "grad_norm": 0.42628052830696106, + "learning_rate": 0.00045257171671896014, + "loss": 0.4144, + "step": 19659 + }, + { + "epoch": 1.0984440036874599, + "grad_norm": 0.45053884387016296, + "learning_rate": 0.0004525437023756163, + "loss": 0.3775, + "step": 19660 + }, + { + "epoch": 1.0984998742911416, + "grad_norm": 4.686016082763672, + "learning_rate": 0.00045251568803227255, + "loss": 0.4902, + "step": 19661 + }, + { + "epoch": 1.0985557448948235, + "grad_norm": 0.4174436628818512, + "learning_rate": 0.00045248767368892876, + "loss": 0.4561, + "step": 19662 + }, + { + "epoch": 1.0986116154985055, + "grad_norm": 0.3336493968963623, + "learning_rate": 0.00045245965934558497, + "loss": 0.3521, + "step": 19663 + }, + { + "epoch": 1.0986674861021872, + "grad_norm": 0.40734466910362244, + "learning_rate": 0.00045243164500224117, + "loss": 0.4209, + "step": 19664 + }, + { + "epoch": 1.0987233567058692, + "grad_norm": 0.5798870325088501, + "learning_rate": 0.0004524036306588974, + "loss": 0.4723, + "step": 19665 + }, + { + "epoch": 1.0987792273095511, + "grad_norm": 0.3916803300380707, + "learning_rate": 0.0004523756163155536, + "loss": 0.4843, + "step": 19666 + }, + { + "epoch": 1.0988350979132329, + "grad_norm": 0.3067859411239624, + "learning_rate": 0.00045234760197220973, + "loss": 0.3894, + "step": 19667 + }, + { + "epoch": 1.0988909685169148, + "grad_norm": 0.8251456618309021, + "learning_rate": 0.000452319587628866, + "loss": 0.376, + "step": 19668 + }, + { + "epoch": 1.0989468391205968, + "grad_norm": 0.3818126320838928, + "learning_rate": 0.0004522915732855222, + "loss": 0.4219, + "step": 19669 + }, + { + "epoch": 1.0990027097242785, + "grad_norm": 0.5031989216804504, + "learning_rate": 0.0004522635589421784, + "loss": 0.4634, + "step": 19670 + }, + { + "epoch": 1.0990585803279604, + "grad_norm": 0.4229661524295807, + "learning_rate": 0.0004522355445988346, + "loss": 0.4466, + "step": 19671 + }, + { + "epoch": 1.0991144509316424, + "grad_norm": 0.7130590081214905, + "learning_rate": 0.0004522075302554908, + "loss": 0.3783, + "step": 19672 + }, + { + "epoch": 1.0991703215353241, + "grad_norm": 0.708164632320404, + "learning_rate": 0.000452179515912147, + "loss": 0.4649, + "step": 19673 + }, + { + "epoch": 1.099226192139006, + "grad_norm": 0.508095383644104, + "learning_rate": 0.00045215150156880323, + "loss": 0.614, + "step": 19674 + }, + { + "epoch": 1.099282062742688, + "grad_norm": 0.494967520236969, + "learning_rate": 0.00045212348722545944, + "loss": 0.4012, + "step": 19675 + }, + { + "epoch": 1.0993379333463698, + "grad_norm": 0.3441930413246155, + "learning_rate": 0.00045209547288211564, + "loss": 0.4159, + "step": 19676 + }, + { + "epoch": 1.0993938039500517, + "grad_norm": 0.5586714744567871, + "learning_rate": 0.00045206745853877185, + "loss": 0.4146, + "step": 19677 + }, + { + "epoch": 1.0994496745537337, + "grad_norm": 0.3284938633441925, + "learning_rate": 0.0004520394441954281, + "loss": 0.4024, + "step": 19678 + }, + { + "epoch": 1.0995055451574154, + "grad_norm": 2.294637680053711, + "learning_rate": 0.00045201142985208426, + "loss": 0.4517, + "step": 19679 + }, + { + "epoch": 1.0995614157610973, + "grad_norm": 0.4846992492675781, + "learning_rate": 0.0004519834155087405, + "loss": 0.4138, + "step": 19680 + }, + { + "epoch": 1.099617286364779, + "grad_norm": 1.2351447343826294, + "learning_rate": 0.00045195540116539667, + "loss": 0.3748, + "step": 19681 + }, + { + "epoch": 1.099673156968461, + "grad_norm": 0.4244135320186615, + "learning_rate": 0.00045192738682205293, + "loss": 0.3593, + "step": 19682 + }, + { + "epoch": 1.099729027572143, + "grad_norm": 0.393257737159729, + "learning_rate": 0.0004518993724787091, + "loss": 0.4408, + "step": 19683 + }, + { + "epoch": 1.0997848981758247, + "grad_norm": 0.33522143959999084, + "learning_rate": 0.00045187135813536534, + "loss": 0.412, + "step": 19684 + }, + { + "epoch": 1.0998407687795067, + "grad_norm": 0.45438340306282043, + "learning_rate": 0.00045184334379202155, + "loss": 0.3811, + "step": 19685 + }, + { + "epoch": 1.0998966393831886, + "grad_norm": 0.8971171975135803, + "learning_rate": 0.00045181532944867775, + "loss": 0.3445, + "step": 19686 + }, + { + "epoch": 1.0999525099868703, + "grad_norm": 0.7554340362548828, + "learning_rate": 0.00045178731510533396, + "loss": 0.4686, + "step": 19687 + }, + { + "epoch": 1.1000083805905523, + "grad_norm": 0.7306141257286072, + "learning_rate": 0.0004517593007619901, + "loss": 0.5241, + "step": 19688 + }, + { + "epoch": 1.1000642511942342, + "grad_norm": 0.7673091292381287, + "learning_rate": 0.00045173128641864637, + "loss": 0.4042, + "step": 19689 + }, + { + "epoch": 1.100120121797916, + "grad_norm": 0.48260727524757385, + "learning_rate": 0.0004517032720753025, + "loss": 0.458, + "step": 19690 + }, + { + "epoch": 1.100175992401598, + "grad_norm": 0.5385705828666687, + "learning_rate": 0.0004516752577319588, + "loss": 0.4373, + "step": 19691 + }, + { + "epoch": 1.1002318630052799, + "grad_norm": 1.2218860387802124, + "learning_rate": 0.00045164724338861493, + "loss": 0.461, + "step": 19692 + }, + { + "epoch": 1.1002877336089616, + "grad_norm": 0.8961328268051147, + "learning_rate": 0.0004516192290452712, + "loss": 0.4308, + "step": 19693 + }, + { + "epoch": 1.1003436042126435, + "grad_norm": 0.42159661650657654, + "learning_rate": 0.0004515912147019274, + "loss": 0.525, + "step": 19694 + }, + { + "epoch": 1.1003994748163253, + "grad_norm": 0.4586019814014435, + "learning_rate": 0.0004515632003585836, + "loss": 0.4498, + "step": 19695 + }, + { + "epoch": 1.1004553454200072, + "grad_norm": 1.641979455947876, + "learning_rate": 0.0004515351860152398, + "loss": 0.4402, + "step": 19696 + }, + { + "epoch": 1.1005112160236892, + "grad_norm": 0.4913528859615326, + "learning_rate": 0.000451507171671896, + "loss": 0.3227, + "step": 19697 + }, + { + "epoch": 1.100567086627371, + "grad_norm": 1.125427484512329, + "learning_rate": 0.0004514791573285522, + "loss": 0.4456, + "step": 19698 + }, + { + "epoch": 1.1006229572310529, + "grad_norm": 13.225260734558105, + "learning_rate": 0.00045145114298520843, + "loss": 0.3832, + "step": 19699 + }, + { + "epoch": 1.1006788278347348, + "grad_norm": 0.35853973031044006, + "learning_rate": 0.00045142312864186463, + "loss": 0.4654, + "step": 19700 + }, + { + "epoch": 1.1007346984384165, + "grad_norm": 0.45151910185813904, + "learning_rate": 0.0004513951142985209, + "loss": 0.4889, + "step": 19701 + }, + { + "epoch": 1.1007905690420985, + "grad_norm": 0.3384332060813904, + "learning_rate": 0.00045136709995517705, + "loss": 0.3944, + "step": 19702 + }, + { + "epoch": 1.1008464396457804, + "grad_norm": 0.7443218231201172, + "learning_rate": 0.0004513390856118333, + "loss": 0.4098, + "step": 19703 + }, + { + "epoch": 1.1009023102494622, + "grad_norm": 0.3930072486400604, + "learning_rate": 0.00045131107126848946, + "loss": 0.4605, + "step": 19704 + }, + { + "epoch": 1.1009581808531441, + "grad_norm": 0.47549745440483093, + "learning_rate": 0.0004512830569251457, + "loss": 0.4277, + "step": 19705 + }, + { + "epoch": 1.101014051456826, + "grad_norm": 3.9154767990112305, + "learning_rate": 0.00045125504258180187, + "loss": 0.5778, + "step": 19706 + }, + { + "epoch": 1.1010699220605078, + "grad_norm": 0.4446046054363251, + "learning_rate": 0.00045122702823845813, + "loss": 0.4627, + "step": 19707 + }, + { + "epoch": 1.1011257926641898, + "grad_norm": 0.6220989227294922, + "learning_rate": 0.0004511990138951143, + "loss": 0.5337, + "step": 19708 + }, + { + "epoch": 1.1011816632678717, + "grad_norm": 0.4353446960449219, + "learning_rate": 0.0004511709995517705, + "loss": 0.4466, + "step": 19709 + }, + { + "epoch": 1.1012375338715534, + "grad_norm": 0.4324088990688324, + "learning_rate": 0.00045114298520842675, + "loss": 0.4069, + "step": 19710 + }, + { + "epoch": 1.1012934044752354, + "grad_norm": 1.7233308553695679, + "learning_rate": 0.0004511149708650829, + "loss": 0.5595, + "step": 19711 + }, + { + "epoch": 1.1013492750789173, + "grad_norm": 0.43723100423812866, + "learning_rate": 0.00045108695652173916, + "loss": 0.369, + "step": 19712 + }, + { + "epoch": 1.101405145682599, + "grad_norm": 0.4595082998275757, + "learning_rate": 0.0004510589421783953, + "loss": 0.392, + "step": 19713 + }, + { + "epoch": 1.101461016286281, + "grad_norm": 0.3515032231807709, + "learning_rate": 0.00045103092783505157, + "loss": 0.3598, + "step": 19714 + }, + { + "epoch": 1.1015168868899627, + "grad_norm": 0.25838401913642883, + "learning_rate": 0.0004510029134917077, + "loss": 0.3455, + "step": 19715 + }, + { + "epoch": 1.1015727574936447, + "grad_norm": 0.479336142539978, + "learning_rate": 0.000450974899148364, + "loss": 0.4162, + "step": 19716 + }, + { + "epoch": 1.1016286280973266, + "grad_norm": 0.4136204421520233, + "learning_rate": 0.0004509468848050202, + "loss": 0.4921, + "step": 19717 + }, + { + "epoch": 1.1016844987010084, + "grad_norm": 0.47566261887550354, + "learning_rate": 0.0004509188704616764, + "loss": 0.4739, + "step": 19718 + }, + { + "epoch": 1.1017403693046903, + "grad_norm": 0.837925374507904, + "learning_rate": 0.0004508908561183326, + "loss": 0.4283, + "step": 19719 + }, + { + "epoch": 1.1017962399083723, + "grad_norm": 1.3036974668502808, + "learning_rate": 0.0004508628417749888, + "loss": 0.4979, + "step": 19720 + }, + { + "epoch": 1.101852110512054, + "grad_norm": 0.43540433049201965, + "learning_rate": 0.000450834827431645, + "loss": 0.3702, + "step": 19721 + }, + { + "epoch": 1.101907981115736, + "grad_norm": 0.5265597701072693, + "learning_rate": 0.0004508068130883012, + "loss": 0.4057, + "step": 19722 + }, + { + "epoch": 1.101963851719418, + "grad_norm": 0.787346601486206, + "learning_rate": 0.0004507787987449574, + "loss": 0.5753, + "step": 19723 + }, + { + "epoch": 1.1020197223230996, + "grad_norm": 0.45861342549324036, + "learning_rate": 0.0004507507844016137, + "loss": 0.3381, + "step": 19724 + }, + { + "epoch": 1.1020755929267816, + "grad_norm": 0.5499906539916992, + "learning_rate": 0.00045072277005826983, + "loss": 0.541, + "step": 19725 + }, + { + "epoch": 1.1021314635304635, + "grad_norm": 0.4626483917236328, + "learning_rate": 0.0004506947557149261, + "loss": 0.3828, + "step": 19726 + }, + { + "epoch": 1.1021873341341453, + "grad_norm": 0.4711328446865082, + "learning_rate": 0.00045066674137158225, + "loss": 0.5094, + "step": 19727 + }, + { + "epoch": 1.1022432047378272, + "grad_norm": 0.7510603070259094, + "learning_rate": 0.0004506387270282385, + "loss": 0.3908, + "step": 19728 + }, + { + "epoch": 1.102299075341509, + "grad_norm": 1.8992033004760742, + "learning_rate": 0.00045061071268489466, + "loss": 0.4094, + "step": 19729 + }, + { + "epoch": 1.102354945945191, + "grad_norm": 0.5269978046417236, + "learning_rate": 0.00045058269834155086, + "loss": 0.4098, + "step": 19730 + }, + { + "epoch": 1.1024108165488729, + "grad_norm": 0.5280738472938538, + "learning_rate": 0.00045055468399820707, + "loss": 0.4093, + "step": 19731 + }, + { + "epoch": 1.1024666871525546, + "grad_norm": 0.4923098087310791, + "learning_rate": 0.0004505266696548633, + "loss": 0.4864, + "step": 19732 + }, + { + "epoch": 1.1025225577562365, + "grad_norm": 0.4016968309879303, + "learning_rate": 0.00045049865531151954, + "loss": 0.4821, + "step": 19733 + }, + { + "epoch": 1.1025784283599185, + "grad_norm": 1.6780765056610107, + "learning_rate": 0.0004504706409681757, + "loss": 0.3932, + "step": 19734 + }, + { + "epoch": 1.1026342989636002, + "grad_norm": 0.4525280296802521, + "learning_rate": 0.00045044262662483195, + "loss": 0.3485, + "step": 19735 + }, + { + "epoch": 1.1026901695672822, + "grad_norm": 0.6096603274345398, + "learning_rate": 0.0004504146122814881, + "loss": 0.6241, + "step": 19736 + }, + { + "epoch": 1.1027460401709641, + "grad_norm": 4.815281867980957, + "learning_rate": 0.00045038659793814436, + "loss": 0.4331, + "step": 19737 + }, + { + "epoch": 1.1028019107746458, + "grad_norm": 0.46766695380210876, + "learning_rate": 0.0004503585835948005, + "loss": 0.4922, + "step": 19738 + }, + { + "epoch": 1.1028577813783278, + "grad_norm": 0.6871021389961243, + "learning_rate": 0.00045033056925145677, + "loss": 0.4477, + "step": 19739 + }, + { + "epoch": 1.1029136519820097, + "grad_norm": 0.7798712253570557, + "learning_rate": 0.000450302554908113, + "loss": 0.3619, + "step": 19740 + }, + { + "epoch": 1.1029695225856915, + "grad_norm": 0.3434516191482544, + "learning_rate": 0.0004502745405647692, + "loss": 0.4685, + "step": 19741 + }, + { + "epoch": 1.1030253931893734, + "grad_norm": 0.5315161347389221, + "learning_rate": 0.0004502465262214254, + "loss": 0.4445, + "step": 19742 + }, + { + "epoch": 1.1030812637930554, + "grad_norm": 0.6746530532836914, + "learning_rate": 0.0004502185118780816, + "loss": 0.4599, + "step": 19743 + }, + { + "epoch": 1.103137134396737, + "grad_norm": 0.444525808095932, + "learning_rate": 0.0004501904975347378, + "loss": 0.5978, + "step": 19744 + }, + { + "epoch": 1.103193005000419, + "grad_norm": 0.4934478998184204, + "learning_rate": 0.000450162483191394, + "loss": 0.4398, + "step": 19745 + }, + { + "epoch": 1.103248875604101, + "grad_norm": 0.47167274355888367, + "learning_rate": 0.0004501344688480502, + "loss": 0.3215, + "step": 19746 + }, + { + "epoch": 1.1033047462077827, + "grad_norm": 0.6104289889335632, + "learning_rate": 0.0004501064545047064, + "loss": 0.4082, + "step": 19747 + }, + { + "epoch": 1.1033606168114647, + "grad_norm": 0.4202827215194702, + "learning_rate": 0.0004500784401613626, + "loss": 0.4826, + "step": 19748 + }, + { + "epoch": 1.1034164874151464, + "grad_norm": 0.8236360549926758, + "learning_rate": 0.0004500504258180189, + "loss": 0.4012, + "step": 19749 + }, + { + "epoch": 1.1034723580188284, + "grad_norm": 0.5417159199714661, + "learning_rate": 0.00045002241147467503, + "loss": 0.4039, + "step": 19750 + }, + { + "epoch": 1.1035282286225103, + "grad_norm": 0.3806409239768982, + "learning_rate": 0.00044999439713133124, + "loss": 0.3762, + "step": 19751 + }, + { + "epoch": 1.103584099226192, + "grad_norm": 1.2087949514389038, + "learning_rate": 0.00044996638278798745, + "loss": 0.4234, + "step": 19752 + }, + { + "epoch": 1.103639969829874, + "grad_norm": 0.37900006771087646, + "learning_rate": 0.00044993836844464365, + "loss": 0.4468, + "step": 19753 + }, + { + "epoch": 1.103695840433556, + "grad_norm": 0.7269716858863831, + "learning_rate": 0.00044991035410129986, + "loss": 0.5034, + "step": 19754 + }, + { + "epoch": 1.1037517110372377, + "grad_norm": 0.6302278637886047, + "learning_rate": 0.00044988233975795606, + "loss": 0.477, + "step": 19755 + }, + { + "epoch": 1.1038075816409196, + "grad_norm": 0.49686458706855774, + "learning_rate": 0.0004498543254146123, + "loss": 0.4448, + "step": 19756 + }, + { + "epoch": 1.1038634522446016, + "grad_norm": 0.3234063684940338, + "learning_rate": 0.0004498263110712685, + "loss": 0.3964, + "step": 19757 + }, + { + "epoch": 1.1039193228482833, + "grad_norm": 0.56364905834198, + "learning_rate": 0.00044979829672792473, + "loss": 0.5549, + "step": 19758 + }, + { + "epoch": 1.1039751934519653, + "grad_norm": 0.3626877963542938, + "learning_rate": 0.0004497702823845809, + "loss": 0.471, + "step": 19759 + }, + { + "epoch": 1.1040310640556472, + "grad_norm": 0.34746044874191284, + "learning_rate": 0.00044974226804123715, + "loss": 0.4452, + "step": 19760 + }, + { + "epoch": 1.104086934659329, + "grad_norm": 0.5033230781555176, + "learning_rate": 0.0004497142536978933, + "loss": 0.4225, + "step": 19761 + }, + { + "epoch": 1.104142805263011, + "grad_norm": 0.42210641503334045, + "learning_rate": 0.00044968623935454956, + "loss": 0.4381, + "step": 19762 + }, + { + "epoch": 1.1041986758666926, + "grad_norm": 0.41470614075660706, + "learning_rate": 0.0004496582250112057, + "loss": 0.4692, + "step": 19763 + }, + { + "epoch": 1.1042545464703746, + "grad_norm": 0.4156918525695801, + "learning_rate": 0.00044963021066786197, + "loss": 0.47, + "step": 19764 + }, + { + "epoch": 1.1043104170740565, + "grad_norm": 0.8547062277793884, + "learning_rate": 0.0004496021963245182, + "loss": 0.6174, + "step": 19765 + }, + { + "epoch": 1.1043662876777383, + "grad_norm": 0.48659273982048035, + "learning_rate": 0.0004495741819811744, + "loss": 0.3813, + "step": 19766 + }, + { + "epoch": 1.1044221582814202, + "grad_norm": 0.330710232257843, + "learning_rate": 0.0004495461676378306, + "loss": 0.4371, + "step": 19767 + }, + { + "epoch": 1.1044780288851022, + "grad_norm": 0.6305220127105713, + "learning_rate": 0.0004495181532944868, + "loss": 0.4579, + "step": 19768 + }, + { + "epoch": 1.1045338994887839, + "grad_norm": 0.3831937313079834, + "learning_rate": 0.000449490138951143, + "loss": 0.3778, + "step": 19769 + }, + { + "epoch": 1.1045897700924658, + "grad_norm": 0.43506351113319397, + "learning_rate": 0.0004494621246077992, + "loss": 0.3918, + "step": 19770 + }, + { + "epoch": 1.1046456406961478, + "grad_norm": 0.35957249999046326, + "learning_rate": 0.0004494341102644554, + "loss": 0.3642, + "step": 19771 + }, + { + "epoch": 1.1047015112998295, + "grad_norm": 1.7005436420440674, + "learning_rate": 0.00044940609592111167, + "loss": 0.4362, + "step": 19772 + }, + { + "epoch": 1.1047573819035115, + "grad_norm": 1.7312259674072266, + "learning_rate": 0.0004493780815777678, + "loss": 0.4423, + "step": 19773 + }, + { + "epoch": 1.1048132525071934, + "grad_norm": 0.43091943860054016, + "learning_rate": 0.00044935006723442403, + "loss": 0.5511, + "step": 19774 + }, + { + "epoch": 1.1048691231108752, + "grad_norm": 0.7237133979797363, + "learning_rate": 0.00044932205289108023, + "loss": 0.4905, + "step": 19775 + }, + { + "epoch": 1.104924993714557, + "grad_norm": 0.3906311094760895, + "learning_rate": 0.00044929403854773644, + "loss": 0.3909, + "step": 19776 + }, + { + "epoch": 1.104980864318239, + "grad_norm": 0.3968881368637085, + "learning_rate": 0.00044926602420439265, + "loss": 0.534, + "step": 19777 + }, + { + "epoch": 1.1050367349219208, + "grad_norm": 1.0542092323303223, + "learning_rate": 0.00044923800986104885, + "loss": 0.4184, + "step": 19778 + }, + { + "epoch": 1.1050926055256027, + "grad_norm": 0.5145635008811951, + "learning_rate": 0.00044920999551770506, + "loss": 0.3909, + "step": 19779 + }, + { + "epoch": 1.1051484761292847, + "grad_norm": 0.7828555107116699, + "learning_rate": 0.00044918198117436126, + "loss": 0.4342, + "step": 19780 + }, + { + "epoch": 1.1052043467329664, + "grad_norm": 0.47401684522628784, + "learning_rate": 0.0004491539668310175, + "loss": 0.3143, + "step": 19781 + }, + { + "epoch": 1.1052602173366484, + "grad_norm": 0.3993418514728546, + "learning_rate": 0.0004491259524876737, + "loss": 0.4322, + "step": 19782 + }, + { + "epoch": 1.10531608794033, + "grad_norm": 0.6904125213623047, + "learning_rate": 0.00044909793814432993, + "loss": 0.4186, + "step": 19783 + }, + { + "epoch": 1.105371958544012, + "grad_norm": 0.4019426703453064, + "learning_rate": 0.0004490699238009861, + "loss": 0.4226, + "step": 19784 + }, + { + "epoch": 1.105427829147694, + "grad_norm": 0.4254942238330841, + "learning_rate": 0.00044904190945764235, + "loss": 0.4284, + "step": 19785 + }, + { + "epoch": 1.1054836997513757, + "grad_norm": 0.521727979183197, + "learning_rate": 0.0004490138951142985, + "loss": 0.3639, + "step": 19786 + }, + { + "epoch": 1.1055395703550577, + "grad_norm": 1.0123088359832764, + "learning_rate": 0.00044898588077095476, + "loss": 0.4082, + "step": 19787 + }, + { + "epoch": 1.1055954409587396, + "grad_norm": 0.9803475737571716, + "learning_rate": 0.00044895786642761096, + "loss": 0.345, + "step": 19788 + }, + { + "epoch": 1.1056513115624214, + "grad_norm": 0.6807441115379333, + "learning_rate": 0.00044892985208426717, + "loss": 0.3918, + "step": 19789 + }, + { + "epoch": 1.1057071821661033, + "grad_norm": 0.6339095234870911, + "learning_rate": 0.0004489018377409234, + "loss": 0.3549, + "step": 19790 + }, + { + "epoch": 1.1057630527697853, + "grad_norm": 0.5549077391624451, + "learning_rate": 0.0004488738233975796, + "loss": 0.4574, + "step": 19791 + }, + { + "epoch": 1.105818923373467, + "grad_norm": 0.35005220770835876, + "learning_rate": 0.0004488458090542358, + "loss": 0.4137, + "step": 19792 + }, + { + "epoch": 1.105874793977149, + "grad_norm": 0.7228426933288574, + "learning_rate": 0.000448817794710892, + "loss": 0.3005, + "step": 19793 + }, + { + "epoch": 1.105930664580831, + "grad_norm": 1.3253737688064575, + "learning_rate": 0.0004487897803675482, + "loss": 0.4205, + "step": 19794 + }, + { + "epoch": 1.1059865351845126, + "grad_norm": 1.2324162721633911, + "learning_rate": 0.00044876176602420435, + "loss": 0.5812, + "step": 19795 + }, + { + "epoch": 1.1060424057881946, + "grad_norm": 0.552370548248291, + "learning_rate": 0.0004487337516808606, + "loss": 0.397, + "step": 19796 + }, + { + "epoch": 1.1060982763918763, + "grad_norm": 0.40345892310142517, + "learning_rate": 0.0004487057373375168, + "loss": 0.4102, + "step": 19797 + }, + { + "epoch": 1.1061541469955583, + "grad_norm": 0.43415358662605286, + "learning_rate": 0.000448677722994173, + "loss": 0.4063, + "step": 19798 + }, + { + "epoch": 1.1062100175992402, + "grad_norm": 0.6444132924079895, + "learning_rate": 0.00044864970865082923, + "loss": 0.551, + "step": 19799 + }, + { + "epoch": 1.106265888202922, + "grad_norm": 0.4221161901950836, + "learning_rate": 0.00044862169430748543, + "loss": 0.5617, + "step": 19800 + }, + { + "epoch": 1.1063217588066039, + "grad_norm": 0.6196674108505249, + "learning_rate": 0.00044859367996414164, + "loss": 0.5369, + "step": 19801 + }, + { + "epoch": 1.1063776294102858, + "grad_norm": 0.3889618217945099, + "learning_rate": 0.00044856566562079784, + "loss": 0.4013, + "step": 19802 + }, + { + "epoch": 1.1064335000139676, + "grad_norm": 0.5923811793327332, + "learning_rate": 0.00044853765127745405, + "loss": 0.5629, + "step": 19803 + }, + { + "epoch": 1.1064893706176495, + "grad_norm": 0.41127511858940125, + "learning_rate": 0.0004485096369341103, + "loss": 0.375, + "step": 19804 + }, + { + "epoch": 1.1065452412213315, + "grad_norm": 0.7849301099777222, + "learning_rate": 0.00044848162259076646, + "loss": 0.4271, + "step": 19805 + }, + { + "epoch": 1.1066011118250132, + "grad_norm": 0.7401925921440125, + "learning_rate": 0.0004484536082474227, + "loss": 0.4478, + "step": 19806 + }, + { + "epoch": 1.1066569824286951, + "grad_norm": 5.659556865692139, + "learning_rate": 0.0004484255939040789, + "loss": 0.4007, + "step": 19807 + }, + { + "epoch": 1.106712853032377, + "grad_norm": 1.1390795707702637, + "learning_rate": 0.00044839757956073513, + "loss": 0.4188, + "step": 19808 + }, + { + "epoch": 1.1067687236360588, + "grad_norm": 0.512064516544342, + "learning_rate": 0.0004483695652173913, + "loss": 0.4259, + "step": 19809 + }, + { + "epoch": 1.1068245942397408, + "grad_norm": 0.4733828604221344, + "learning_rate": 0.00044834155087404755, + "loss": 0.4277, + "step": 19810 + }, + { + "epoch": 1.1068804648434227, + "grad_norm": 0.4203338325023651, + "learning_rate": 0.00044831353653070375, + "loss": 0.4096, + "step": 19811 + }, + { + "epoch": 1.1069363354471045, + "grad_norm": 0.39705514907836914, + "learning_rate": 0.00044828552218735996, + "loss": 0.3068, + "step": 19812 + }, + { + "epoch": 1.1069922060507864, + "grad_norm": 0.4222647249698639, + "learning_rate": 0.00044825750784401616, + "loss": 0.5283, + "step": 19813 + }, + { + "epoch": 1.1070480766544684, + "grad_norm": 0.42690131068229675, + "learning_rate": 0.00044822949350067237, + "loss": 0.4638, + "step": 19814 + }, + { + "epoch": 1.10710394725815, + "grad_norm": 0.5646849870681763, + "learning_rate": 0.0004482014791573286, + "loss": 0.4927, + "step": 19815 + }, + { + "epoch": 1.107159817861832, + "grad_norm": 0.5031330585479736, + "learning_rate": 0.0004481734648139847, + "loss": 0.377, + "step": 19816 + }, + { + "epoch": 1.1072156884655138, + "grad_norm": 0.44143325090408325, + "learning_rate": 0.000448145450470641, + "loss": 0.4365, + "step": 19817 + }, + { + "epoch": 1.1072715590691957, + "grad_norm": 0.5689504146575928, + "learning_rate": 0.00044811743612729714, + "loss": 0.7162, + "step": 19818 + }, + { + "epoch": 1.1073274296728777, + "grad_norm": 0.5287508368492126, + "learning_rate": 0.0004480894217839534, + "loss": 0.5402, + "step": 19819 + }, + { + "epoch": 1.1073833002765594, + "grad_norm": 0.40917113423347473, + "learning_rate": 0.0004480614074406096, + "loss": 0.4553, + "step": 19820 + }, + { + "epoch": 1.1074391708802414, + "grad_norm": 0.46447432041168213, + "learning_rate": 0.0004480333930972658, + "loss": 0.3759, + "step": 19821 + }, + { + "epoch": 1.1074950414839233, + "grad_norm": 0.405125230550766, + "learning_rate": 0.000448005378753922, + "loss": 0.4464, + "step": 19822 + }, + { + "epoch": 1.107550912087605, + "grad_norm": 0.752811849117279, + "learning_rate": 0.0004479773644105782, + "loss": 0.4678, + "step": 19823 + }, + { + "epoch": 1.107606782691287, + "grad_norm": 0.4256785809993744, + "learning_rate": 0.0004479493500672344, + "loss": 0.3984, + "step": 19824 + }, + { + "epoch": 1.107662653294969, + "grad_norm": 0.3952004909515381, + "learning_rate": 0.00044792133572389063, + "loss": 0.36, + "step": 19825 + }, + { + "epoch": 1.1077185238986507, + "grad_norm": 0.42111966013908386, + "learning_rate": 0.00044789332138054684, + "loss": 0.4351, + "step": 19826 + }, + { + "epoch": 1.1077743945023326, + "grad_norm": 0.5292633175849915, + "learning_rate": 0.0004478653070372031, + "loss": 0.5387, + "step": 19827 + }, + { + "epoch": 1.1078302651060146, + "grad_norm": 1.6792409420013428, + "learning_rate": 0.00044783729269385925, + "loss": 0.4653, + "step": 19828 + }, + { + "epoch": 1.1078861357096963, + "grad_norm": 0.4042298197746277, + "learning_rate": 0.0004478092783505155, + "loss": 0.4445, + "step": 19829 + }, + { + "epoch": 1.1079420063133782, + "grad_norm": 1.2308564186096191, + "learning_rate": 0.00044778126400717166, + "loss": 0.4932, + "step": 19830 + }, + { + "epoch": 1.10799787691706, + "grad_norm": 0.4534153938293457, + "learning_rate": 0.0004477532496638279, + "loss": 0.4318, + "step": 19831 + }, + { + "epoch": 1.108053747520742, + "grad_norm": 0.7793142199516296, + "learning_rate": 0.0004477252353204841, + "loss": 0.4476, + "step": 19832 + }, + { + "epoch": 1.1081096181244239, + "grad_norm": 0.5808934569358826, + "learning_rate": 0.00044769722097714033, + "loss": 0.4614, + "step": 19833 + }, + { + "epoch": 1.1081654887281056, + "grad_norm": 3.352155923843384, + "learning_rate": 0.0004476692066337965, + "loss": 0.4376, + "step": 19834 + }, + { + "epoch": 1.1082213593317876, + "grad_norm": 0.6043905019760132, + "learning_rate": 0.00044764119229045275, + "loss": 0.3703, + "step": 19835 + }, + { + "epoch": 1.1082772299354695, + "grad_norm": 0.4718884229660034, + "learning_rate": 0.00044761317794710895, + "loss": 0.5155, + "step": 19836 + }, + { + "epoch": 1.1083331005391512, + "grad_norm": 1.149834156036377, + "learning_rate": 0.0004475851636037651, + "loss": 0.499, + "step": 19837 + }, + { + "epoch": 1.1083889711428332, + "grad_norm": 0.41123297810554504, + "learning_rate": 0.00044755714926042136, + "loss": 0.3642, + "step": 19838 + }, + { + "epoch": 1.1084448417465151, + "grad_norm": 0.5839034914970398, + "learning_rate": 0.0004475291349170775, + "loss": 0.36, + "step": 19839 + }, + { + "epoch": 1.1085007123501969, + "grad_norm": 1.3502988815307617, + "learning_rate": 0.0004475011205737338, + "loss": 0.5124, + "step": 19840 + }, + { + "epoch": 1.1085565829538788, + "grad_norm": 1.1481295824050903, + "learning_rate": 0.0004474731062303899, + "loss": 0.5607, + "step": 19841 + }, + { + "epoch": 1.1086124535575608, + "grad_norm": 0.7058351039886475, + "learning_rate": 0.0004474450918870462, + "loss": 0.4324, + "step": 19842 + }, + { + "epoch": 1.1086683241612425, + "grad_norm": 1.0131568908691406, + "learning_rate": 0.0004474170775437024, + "loss": 0.3946, + "step": 19843 + }, + { + "epoch": 1.1087241947649245, + "grad_norm": 0.7804633975028992, + "learning_rate": 0.0004473890632003586, + "loss": 0.4262, + "step": 19844 + }, + { + "epoch": 1.1087800653686064, + "grad_norm": 0.6793167591094971, + "learning_rate": 0.0004473610488570148, + "loss": 0.456, + "step": 19845 + }, + { + "epoch": 1.1088359359722881, + "grad_norm": 0.5463383197784424, + "learning_rate": 0.000447333034513671, + "loss": 0.3516, + "step": 19846 + }, + { + "epoch": 1.10889180657597, + "grad_norm": 0.39011695981025696, + "learning_rate": 0.0004473050201703272, + "loss": 0.3772, + "step": 19847 + }, + { + "epoch": 1.108947677179652, + "grad_norm": 0.44870564341545105, + "learning_rate": 0.0004472770058269834, + "loss": 0.4761, + "step": 19848 + }, + { + "epoch": 1.1090035477833338, + "grad_norm": 0.47883540391921997, + "learning_rate": 0.0004472489914836396, + "loss": 0.3916, + "step": 19849 + }, + { + "epoch": 1.1090594183870157, + "grad_norm": 0.38711944222450256, + "learning_rate": 0.00044722097714029583, + "loss": 0.4438, + "step": 19850 + }, + { + "epoch": 1.1091152889906974, + "grad_norm": 0.6181144714355469, + "learning_rate": 0.00044719296279695204, + "loss": 0.3639, + "step": 19851 + }, + { + "epoch": 1.1091711595943794, + "grad_norm": 0.7975968718528748, + "learning_rate": 0.0004471649484536083, + "loss": 0.4367, + "step": 19852 + }, + { + "epoch": 1.1092270301980613, + "grad_norm": 1.234458088874817, + "learning_rate": 0.00044713693411026445, + "loss": 0.5751, + "step": 19853 + }, + { + "epoch": 1.109282900801743, + "grad_norm": 0.3739296495914459, + "learning_rate": 0.0004471089197669207, + "loss": 0.4359, + "step": 19854 + }, + { + "epoch": 1.109338771405425, + "grad_norm": 0.5284265279769897, + "learning_rate": 0.00044708090542357686, + "loss": 0.4706, + "step": 19855 + }, + { + "epoch": 1.109394642009107, + "grad_norm": 0.4258183240890503, + "learning_rate": 0.0004470528910802331, + "loss": 0.4771, + "step": 19856 + }, + { + "epoch": 1.1094505126127887, + "grad_norm": 0.4575117528438568, + "learning_rate": 0.0004470248767368893, + "loss": 0.4002, + "step": 19857 + }, + { + "epoch": 1.1095063832164707, + "grad_norm": 0.6182119846343994, + "learning_rate": 0.0004469968623935455, + "loss": 0.4424, + "step": 19858 + }, + { + "epoch": 1.1095622538201526, + "grad_norm": 1.3266000747680664, + "learning_rate": 0.00044696884805020174, + "loss": 0.4607, + "step": 19859 + }, + { + "epoch": 1.1096181244238343, + "grad_norm": 0.3513745963573456, + "learning_rate": 0.0004469408337068579, + "loss": 0.382, + "step": 19860 + }, + { + "epoch": 1.1096739950275163, + "grad_norm": 0.7137050628662109, + "learning_rate": 0.00044691281936351415, + "loss": 0.4807, + "step": 19861 + }, + { + "epoch": 1.1097298656311982, + "grad_norm": 8.849645614624023, + "learning_rate": 0.0004468848050201703, + "loss": 0.4965, + "step": 19862 + }, + { + "epoch": 1.10978573623488, + "grad_norm": 0.397641658782959, + "learning_rate": 0.00044685679067682656, + "loss": 0.3903, + "step": 19863 + }, + { + "epoch": 1.109841606838562, + "grad_norm": 0.541979193687439, + "learning_rate": 0.0004468287763334827, + "loss": 0.4121, + "step": 19864 + }, + { + "epoch": 1.1098974774422437, + "grad_norm": 0.5506418347358704, + "learning_rate": 0.000446800761990139, + "loss": 0.4742, + "step": 19865 + }, + { + "epoch": 1.1099533480459256, + "grad_norm": 0.5158501863479614, + "learning_rate": 0.0004467727476467951, + "loss": 0.3529, + "step": 19866 + }, + { + "epoch": 1.1100092186496076, + "grad_norm": 0.5719771385192871, + "learning_rate": 0.0004467447333034514, + "loss": 0.4374, + "step": 19867 + }, + { + "epoch": 1.1100650892532893, + "grad_norm": 0.4219617545604706, + "learning_rate": 0.0004467167189601076, + "loss": 0.409, + "step": 19868 + }, + { + "epoch": 1.1101209598569712, + "grad_norm": 0.4697694778442383, + "learning_rate": 0.0004466887046167638, + "loss": 0.3965, + "step": 19869 + }, + { + "epoch": 1.1101768304606532, + "grad_norm": 0.584130048751831, + "learning_rate": 0.00044666069027342, + "loss": 0.4069, + "step": 19870 + }, + { + "epoch": 1.110232701064335, + "grad_norm": 0.3423226773738861, + "learning_rate": 0.0004466326759300762, + "loss": 0.3963, + "step": 19871 + }, + { + "epoch": 1.1102885716680169, + "grad_norm": 0.37696459889411926, + "learning_rate": 0.0004466046615867324, + "loss": 0.3714, + "step": 19872 + }, + { + "epoch": 1.1103444422716988, + "grad_norm": 0.6303426027297974, + "learning_rate": 0.0004465766472433886, + "loss": 0.4713, + "step": 19873 + }, + { + "epoch": 1.1104003128753805, + "grad_norm": 0.5746414661407471, + "learning_rate": 0.0004465486329000448, + "loss": 0.4207, + "step": 19874 + }, + { + "epoch": 1.1104561834790625, + "grad_norm": 0.4625340700149536, + "learning_rate": 0.0004465206185567011, + "loss": 0.4912, + "step": 19875 + }, + { + "epoch": 1.1105120540827444, + "grad_norm": 0.5184044241905212, + "learning_rate": 0.00044649260421335724, + "loss": 0.3278, + "step": 19876 + }, + { + "epoch": 1.1105679246864262, + "grad_norm": 0.7626926302909851, + "learning_rate": 0.0004464645898700135, + "loss": 0.6733, + "step": 19877 + }, + { + "epoch": 1.1106237952901081, + "grad_norm": 0.5433309078216553, + "learning_rate": 0.00044643657552666965, + "loss": 0.4793, + "step": 19878 + }, + { + "epoch": 1.11067966589379, + "grad_norm": 0.6071287989616394, + "learning_rate": 0.00044640856118332586, + "loss": 0.5095, + "step": 19879 + }, + { + "epoch": 1.1107355364974718, + "grad_norm": 0.48836737871170044, + "learning_rate": 0.00044638054683998206, + "loss": 0.4783, + "step": 19880 + }, + { + "epoch": 1.1107914071011538, + "grad_norm": 0.3953384459018707, + "learning_rate": 0.00044635253249663827, + "loss": 0.4406, + "step": 19881 + }, + { + "epoch": 1.1108472777048357, + "grad_norm": 3.7148590087890625, + "learning_rate": 0.0004463245181532945, + "loss": 0.5235, + "step": 19882 + }, + { + "epoch": 1.1109031483085174, + "grad_norm": 1.3106836080551147, + "learning_rate": 0.0004462965038099507, + "loss": 0.4992, + "step": 19883 + }, + { + "epoch": 1.1109590189121994, + "grad_norm": 0.43632400035858154, + "learning_rate": 0.00044626848946660694, + "loss": 0.4357, + "step": 19884 + }, + { + "epoch": 1.1110148895158811, + "grad_norm": 0.644878089427948, + "learning_rate": 0.0004462404751232631, + "loss": 0.4573, + "step": 19885 + }, + { + "epoch": 1.111070760119563, + "grad_norm": 0.40030437707901, + "learning_rate": 0.00044621246077991935, + "loss": 0.3975, + "step": 19886 + }, + { + "epoch": 1.111126630723245, + "grad_norm": 0.513010561466217, + "learning_rate": 0.0004461844464365755, + "loss": 0.4675, + "step": 19887 + }, + { + "epoch": 1.1111825013269268, + "grad_norm": 1.8976777791976929, + "learning_rate": 0.00044615643209323176, + "loss": 0.4837, + "step": 19888 + }, + { + "epoch": 1.1112383719306087, + "grad_norm": 0.5581178665161133, + "learning_rate": 0.0004461284177498879, + "loss": 0.4771, + "step": 19889 + }, + { + "epoch": 1.1112942425342907, + "grad_norm": 0.4349944293498993, + "learning_rate": 0.0004461004034065442, + "loss": 0.3796, + "step": 19890 + }, + { + "epoch": 1.1113501131379724, + "grad_norm": 0.6147105097770691, + "learning_rate": 0.0004460723890632004, + "loss": 0.4003, + "step": 19891 + }, + { + "epoch": 1.1114059837416543, + "grad_norm": 0.48892030119895935, + "learning_rate": 0.0004460443747198566, + "loss": 0.4577, + "step": 19892 + }, + { + "epoch": 1.1114618543453363, + "grad_norm": 0.9737526774406433, + "learning_rate": 0.0004460163603765128, + "loss": 0.5914, + "step": 19893 + }, + { + "epoch": 1.111517724949018, + "grad_norm": 1.036501169204712, + "learning_rate": 0.000445988346033169, + "loss": 0.3878, + "step": 19894 + }, + { + "epoch": 1.1115735955527, + "grad_norm": 0.4183790981769562, + "learning_rate": 0.0004459603316898252, + "loss": 0.3625, + "step": 19895 + }, + { + "epoch": 1.111629466156382, + "grad_norm": 0.9512483477592468, + "learning_rate": 0.0004459323173464814, + "loss": 0.5765, + "step": 19896 + }, + { + "epoch": 1.1116853367600636, + "grad_norm": 0.5478219389915466, + "learning_rate": 0.0004459043030031376, + "loss": 0.3901, + "step": 19897 + }, + { + "epoch": 1.1117412073637456, + "grad_norm": 2.061380386352539, + "learning_rate": 0.0004458762886597939, + "loss": 0.3765, + "step": 19898 + }, + { + "epoch": 1.1117970779674273, + "grad_norm": 5.081721782684326, + "learning_rate": 0.00044584827431645, + "loss": 0.3804, + "step": 19899 + }, + { + "epoch": 1.1118529485711093, + "grad_norm": 2.6763839721679688, + "learning_rate": 0.00044582025997310623, + "loss": 0.4075, + "step": 19900 + }, + { + "epoch": 1.1119088191747912, + "grad_norm": 0.40840137004852295, + "learning_rate": 0.00044579224562976244, + "loss": 0.4839, + "step": 19901 + }, + { + "epoch": 1.111964689778473, + "grad_norm": 2.347533702850342, + "learning_rate": 0.00044576423128641864, + "loss": 0.3378, + "step": 19902 + }, + { + "epoch": 1.112020560382155, + "grad_norm": 0.4123016893863678, + "learning_rate": 0.00044573621694307485, + "loss": 0.4052, + "step": 19903 + }, + { + "epoch": 1.1120764309858369, + "grad_norm": 0.47231703996658325, + "learning_rate": 0.00044570820259973105, + "loss": 0.4738, + "step": 19904 + }, + { + "epoch": 1.1121323015895186, + "grad_norm": 0.8349965810775757, + "learning_rate": 0.00044568018825638726, + "loss": 0.378, + "step": 19905 + }, + { + "epoch": 1.1121881721932005, + "grad_norm": 2.6194655895233154, + "learning_rate": 0.00044565217391304347, + "loss": 0.4392, + "step": 19906 + }, + { + "epoch": 1.1122440427968825, + "grad_norm": 0.6084486842155457, + "learning_rate": 0.0004456241595696997, + "loss": 0.5661, + "step": 19907 + }, + { + "epoch": 1.1122999134005642, + "grad_norm": 0.3780638873577118, + "learning_rate": 0.0004455961452263559, + "loss": 0.36, + "step": 19908 + }, + { + "epoch": 1.1123557840042462, + "grad_norm": 0.7052932381629944, + "learning_rate": 0.00044556813088301214, + "loss": 0.4749, + "step": 19909 + }, + { + "epoch": 1.1124116546079281, + "grad_norm": 0.3740042448043823, + "learning_rate": 0.0004455401165396683, + "loss": 0.3296, + "step": 19910 + }, + { + "epoch": 1.1124675252116099, + "grad_norm": 1.545491099357605, + "learning_rate": 0.00044551210219632455, + "loss": 0.5314, + "step": 19911 + }, + { + "epoch": 1.1125233958152918, + "grad_norm": 0.4569876790046692, + "learning_rate": 0.0004454840878529807, + "loss": 0.3658, + "step": 19912 + }, + { + "epoch": 1.1125792664189738, + "grad_norm": 0.5053375959396362, + "learning_rate": 0.00044545607350963696, + "loss": 0.4727, + "step": 19913 + }, + { + "epoch": 1.1126351370226555, + "grad_norm": 0.5856257677078247, + "learning_rate": 0.00044542805916629317, + "loss": 0.5424, + "step": 19914 + }, + { + "epoch": 1.1126910076263374, + "grad_norm": 0.3576067388057709, + "learning_rate": 0.00044540004482294937, + "loss": 0.459, + "step": 19915 + }, + { + "epoch": 1.1127468782300194, + "grad_norm": 0.957303524017334, + "learning_rate": 0.0004453720304796056, + "loss": 0.3692, + "step": 19916 + }, + { + "epoch": 1.1128027488337011, + "grad_norm": 0.42888393998146057, + "learning_rate": 0.0004453440161362618, + "loss": 0.4139, + "step": 19917 + }, + { + "epoch": 1.112858619437383, + "grad_norm": 0.64570152759552, + "learning_rate": 0.000445316001792918, + "loss": 0.4063, + "step": 19918 + }, + { + "epoch": 1.1129144900410648, + "grad_norm": 0.6980811953544617, + "learning_rate": 0.0004452879874495742, + "loss": 0.3806, + "step": 19919 + }, + { + "epoch": 1.1129703606447467, + "grad_norm": 0.6899658441543579, + "learning_rate": 0.0004452599731062304, + "loss": 0.7101, + "step": 19920 + }, + { + "epoch": 1.1130262312484287, + "grad_norm": 0.4745479226112366, + "learning_rate": 0.00044523195876288655, + "loss": 0.3883, + "step": 19921 + }, + { + "epoch": 1.1130821018521104, + "grad_norm": 0.5159439444541931, + "learning_rate": 0.0004452039444195428, + "loss": 0.3561, + "step": 19922 + }, + { + "epoch": 1.1131379724557924, + "grad_norm": 0.47799891233444214, + "learning_rate": 0.000445175930076199, + "loss": 0.426, + "step": 19923 + }, + { + "epoch": 1.1131938430594743, + "grad_norm": 0.5052812695503235, + "learning_rate": 0.0004451479157328552, + "loss": 0.4718, + "step": 19924 + }, + { + "epoch": 1.113249713663156, + "grad_norm": 0.7952449917793274, + "learning_rate": 0.00044511990138951143, + "loss": 0.3306, + "step": 19925 + }, + { + "epoch": 1.113305584266838, + "grad_norm": 0.49266743659973145, + "learning_rate": 0.00044509188704616764, + "loss": 0.4853, + "step": 19926 + }, + { + "epoch": 1.11336145487052, + "grad_norm": 0.4133360981941223, + "learning_rate": 0.00044506387270282384, + "loss": 0.4292, + "step": 19927 + }, + { + "epoch": 1.1134173254742017, + "grad_norm": 0.4396034777164459, + "learning_rate": 0.00044503585835948005, + "loss": 0.5069, + "step": 19928 + }, + { + "epoch": 1.1134731960778836, + "grad_norm": 0.3765599727630615, + "learning_rate": 0.00044500784401613625, + "loss": 0.4007, + "step": 19929 + }, + { + "epoch": 1.1135290666815656, + "grad_norm": 0.5399726629257202, + "learning_rate": 0.0004449798296727925, + "loss": 0.3704, + "step": 19930 + }, + { + "epoch": 1.1135849372852473, + "grad_norm": 0.3475736081600189, + "learning_rate": 0.00044495181532944867, + "loss": 0.3736, + "step": 19931 + }, + { + "epoch": 1.1136408078889293, + "grad_norm": 0.4021284282207489, + "learning_rate": 0.0004449238009861049, + "loss": 0.4165, + "step": 19932 + }, + { + "epoch": 1.113696678492611, + "grad_norm": 1.1107367277145386, + "learning_rate": 0.0004448957866427611, + "loss": 0.4171, + "step": 19933 + }, + { + "epoch": 1.113752549096293, + "grad_norm": 0.620159924030304, + "learning_rate": 0.00044486777229941734, + "loss": 0.4903, + "step": 19934 + }, + { + "epoch": 1.113808419699975, + "grad_norm": 0.5911586284637451, + "learning_rate": 0.0004448397579560735, + "loss": 0.5284, + "step": 19935 + }, + { + "epoch": 1.1138642903036566, + "grad_norm": 0.40787196159362793, + "learning_rate": 0.00044481174361272975, + "loss": 0.3957, + "step": 19936 + }, + { + "epoch": 1.1139201609073386, + "grad_norm": 0.497977077960968, + "learning_rate": 0.00044478372926938596, + "loss": 0.4584, + "step": 19937 + }, + { + "epoch": 1.1139760315110205, + "grad_norm": 0.6699540019035339, + "learning_rate": 0.00044475571492604216, + "loss": 0.3625, + "step": 19938 + }, + { + "epoch": 1.1140319021147023, + "grad_norm": 0.3737075626850128, + "learning_rate": 0.00044472770058269837, + "loss": 0.3601, + "step": 19939 + }, + { + "epoch": 1.1140877727183842, + "grad_norm": 0.36192435026168823, + "learning_rate": 0.00044469968623935457, + "loss": 0.4114, + "step": 19940 + }, + { + "epoch": 1.1141436433220662, + "grad_norm": 0.5001122951507568, + "learning_rate": 0.0004446716718960108, + "loss": 0.4161, + "step": 19941 + }, + { + "epoch": 1.114199513925748, + "grad_norm": 0.4299237132072449, + "learning_rate": 0.000444643657552667, + "loss": 0.3594, + "step": 19942 + }, + { + "epoch": 1.1142553845294298, + "grad_norm": 0.8796808123588562, + "learning_rate": 0.0004446156432093232, + "loss": 0.5568, + "step": 19943 + }, + { + "epoch": 1.1143112551331118, + "grad_norm": 0.4980526566505432, + "learning_rate": 0.00044458762886597934, + "loss": 0.3925, + "step": 19944 + }, + { + "epoch": 1.1143671257367935, + "grad_norm": 0.5252367854118347, + "learning_rate": 0.0004445596145226356, + "loss": 0.6739, + "step": 19945 + }, + { + "epoch": 1.1144229963404755, + "grad_norm": 0.6410105228424072, + "learning_rate": 0.0004445316001792918, + "loss": 0.4349, + "step": 19946 + }, + { + "epoch": 1.1144788669441574, + "grad_norm": 0.4308195412158966, + "learning_rate": 0.000444503585835948, + "loss": 0.4297, + "step": 19947 + }, + { + "epoch": 1.1145347375478392, + "grad_norm": 10.36929702758789, + "learning_rate": 0.0004444755714926042, + "loss": 0.4327, + "step": 19948 + }, + { + "epoch": 1.1145906081515211, + "grad_norm": 0.3741483688354492, + "learning_rate": 0.0004444475571492604, + "loss": 0.4012, + "step": 19949 + }, + { + "epoch": 1.114646478755203, + "grad_norm": 0.5729476809501648, + "learning_rate": 0.00044441954280591663, + "loss": 0.3925, + "step": 19950 + }, + { + "epoch": 1.1147023493588848, + "grad_norm": 0.31191906332969666, + "learning_rate": 0.00044439152846257284, + "loss": 0.3044, + "step": 19951 + }, + { + "epoch": 1.1147582199625667, + "grad_norm": 0.4128904938697815, + "learning_rate": 0.00044436351411922904, + "loss": 0.4647, + "step": 19952 + }, + { + "epoch": 1.1148140905662485, + "grad_norm": 0.3641165494918823, + "learning_rate": 0.0004443354997758853, + "loss": 0.4268, + "step": 19953 + }, + { + "epoch": 1.1148699611699304, + "grad_norm": 2.2114946842193604, + "learning_rate": 0.00044430748543254145, + "loss": 0.4203, + "step": 19954 + }, + { + "epoch": 1.1149258317736124, + "grad_norm": 1.1938695907592773, + "learning_rate": 0.0004442794710891977, + "loss": 0.4288, + "step": 19955 + }, + { + "epoch": 1.114981702377294, + "grad_norm": 0.529112696647644, + "learning_rate": 0.00044425145674585387, + "loss": 0.4117, + "step": 19956 + }, + { + "epoch": 1.115037572980976, + "grad_norm": 0.5114923119544983, + "learning_rate": 0.0004442234424025101, + "loss": 0.4234, + "step": 19957 + }, + { + "epoch": 1.115093443584658, + "grad_norm": 0.3899543583393097, + "learning_rate": 0.0004441954280591663, + "loss": 0.5723, + "step": 19958 + }, + { + "epoch": 1.1151493141883397, + "grad_norm": 1.6496533155441284, + "learning_rate": 0.00044416741371582254, + "loss": 0.4411, + "step": 19959 + }, + { + "epoch": 1.1152051847920217, + "grad_norm": 0.3924984931945801, + "learning_rate": 0.0004441393993724787, + "loss": 0.3907, + "step": 19960 + }, + { + "epoch": 1.1152610553957036, + "grad_norm": 0.36604130268096924, + "learning_rate": 0.00044411138502913495, + "loss": 0.3705, + "step": 19961 + }, + { + "epoch": 1.1153169259993854, + "grad_norm": 0.48694881796836853, + "learning_rate": 0.00044408337068579115, + "loss": 0.3868, + "step": 19962 + }, + { + "epoch": 1.1153727966030673, + "grad_norm": 0.4688943028450012, + "learning_rate": 0.00044405535634244736, + "loss": 0.41, + "step": 19963 + }, + { + "epoch": 1.1154286672067493, + "grad_norm": 1.2107815742492676, + "learning_rate": 0.00044402734199910357, + "loss": 0.3252, + "step": 19964 + }, + { + "epoch": 1.115484537810431, + "grad_norm": 0.4335245192050934, + "learning_rate": 0.0004439993276557597, + "loss": 0.3841, + "step": 19965 + }, + { + "epoch": 1.115540408414113, + "grad_norm": 0.5237331390380859, + "learning_rate": 0.000443971313312416, + "loss": 0.5229, + "step": 19966 + }, + { + "epoch": 1.1155962790177947, + "grad_norm": 0.8448934555053711, + "learning_rate": 0.00044394329896907213, + "loss": 0.3816, + "step": 19967 + }, + { + "epoch": 1.1156521496214766, + "grad_norm": 0.5793457627296448, + "learning_rate": 0.0004439152846257284, + "loss": 0.4469, + "step": 19968 + }, + { + "epoch": 1.1157080202251586, + "grad_norm": 1.659447193145752, + "learning_rate": 0.0004438872702823846, + "loss": 0.507, + "step": 19969 + }, + { + "epoch": 1.1157638908288403, + "grad_norm": 0.5777044892311096, + "learning_rate": 0.0004438592559390408, + "loss": 0.4748, + "step": 19970 + }, + { + "epoch": 1.1158197614325223, + "grad_norm": 0.4643118381500244, + "learning_rate": 0.000443831241595697, + "loss": 0.4708, + "step": 19971 + }, + { + "epoch": 1.1158756320362042, + "grad_norm": 0.4239487648010254, + "learning_rate": 0.0004438032272523532, + "loss": 0.3892, + "step": 19972 + }, + { + "epoch": 1.115931502639886, + "grad_norm": 0.4022235870361328, + "learning_rate": 0.0004437752129090094, + "loss": 0.4451, + "step": 19973 + }, + { + "epoch": 1.115987373243568, + "grad_norm": 0.3395739197731018, + "learning_rate": 0.0004437471985656656, + "loss": 0.3439, + "step": 19974 + }, + { + "epoch": 1.1160432438472498, + "grad_norm": 0.4063016176223755, + "learning_rate": 0.00044371918422232183, + "loss": 0.3496, + "step": 19975 + }, + { + "epoch": 1.1160991144509316, + "grad_norm": 0.5112307071685791, + "learning_rate": 0.00044369116987897804, + "loss": 0.4165, + "step": 19976 + }, + { + "epoch": 1.1161549850546135, + "grad_norm": 0.5535897016525269, + "learning_rate": 0.00044366315553563424, + "loss": 0.384, + "step": 19977 + }, + { + "epoch": 1.1162108556582955, + "grad_norm": 0.3690154254436493, + "learning_rate": 0.0004436351411922905, + "loss": 0.4082, + "step": 19978 + }, + { + "epoch": 1.1162667262619772, + "grad_norm": 0.6509933471679688, + "learning_rate": 0.00044360712684894665, + "loss": 0.5006, + "step": 19979 + }, + { + "epoch": 1.1163225968656592, + "grad_norm": 0.4301048219203949, + "learning_rate": 0.0004435791125056029, + "loss": 0.3853, + "step": 19980 + }, + { + "epoch": 1.116378467469341, + "grad_norm": 0.9483281373977661, + "learning_rate": 0.00044355109816225907, + "loss": 0.4726, + "step": 19981 + }, + { + "epoch": 1.1164343380730228, + "grad_norm": 0.5963273048400879, + "learning_rate": 0.0004435230838189153, + "loss": 0.4198, + "step": 19982 + }, + { + "epoch": 1.1164902086767048, + "grad_norm": 0.4600369334220886, + "learning_rate": 0.0004434950694755715, + "loss": 0.3627, + "step": 19983 + }, + { + "epoch": 1.1165460792803867, + "grad_norm": 0.37426888942718506, + "learning_rate": 0.00044346705513222774, + "loss": 0.3592, + "step": 19984 + }, + { + "epoch": 1.1166019498840685, + "grad_norm": 0.5805225372314453, + "learning_rate": 0.00044343904078888394, + "loss": 0.5532, + "step": 19985 + }, + { + "epoch": 1.1166578204877504, + "grad_norm": 0.572267472743988, + "learning_rate": 0.0004434110264455401, + "loss": 0.4625, + "step": 19986 + }, + { + "epoch": 1.1167136910914321, + "grad_norm": 0.6061015725135803, + "learning_rate": 0.00044338301210219635, + "loss": 0.3389, + "step": 19987 + }, + { + "epoch": 1.116769561695114, + "grad_norm": 0.4652900993824005, + "learning_rate": 0.0004433549977588525, + "loss": 0.4721, + "step": 19988 + }, + { + "epoch": 1.116825432298796, + "grad_norm": 2.7015106678009033, + "learning_rate": 0.00044332698341550877, + "loss": 0.3854, + "step": 19989 + }, + { + "epoch": 1.1168813029024778, + "grad_norm": 0.4679594039916992, + "learning_rate": 0.0004432989690721649, + "loss": 0.3199, + "step": 19990 + }, + { + "epoch": 1.1169371735061597, + "grad_norm": 0.7866052389144897, + "learning_rate": 0.0004432709547288212, + "loss": 0.4632, + "step": 19991 + }, + { + "epoch": 1.1169930441098417, + "grad_norm": 0.43850040435791016, + "learning_rate": 0.00044324294038547733, + "loss": 0.3731, + "step": 19992 + }, + { + "epoch": 1.1170489147135234, + "grad_norm": 0.3592032492160797, + "learning_rate": 0.0004432149260421336, + "loss": 0.3605, + "step": 19993 + }, + { + "epoch": 1.1171047853172054, + "grad_norm": 0.48577335476875305, + "learning_rate": 0.0004431869116987898, + "loss": 0.3712, + "step": 19994 + }, + { + "epoch": 1.1171606559208873, + "grad_norm": 0.44025716185569763, + "learning_rate": 0.000443158897355446, + "loss": 0.4628, + "step": 19995 + }, + { + "epoch": 1.117216526524569, + "grad_norm": 0.4400652348995209, + "learning_rate": 0.0004431308830121022, + "loss": 0.3968, + "step": 19996 + }, + { + "epoch": 1.117272397128251, + "grad_norm": 0.4115810990333557, + "learning_rate": 0.0004431028686687584, + "loss": 0.3669, + "step": 19997 + }, + { + "epoch": 1.117328267731933, + "grad_norm": 0.961990237236023, + "learning_rate": 0.0004430748543254146, + "loss": 0.4249, + "step": 19998 + }, + { + "epoch": 1.1173841383356147, + "grad_norm": 0.6063356399536133, + "learning_rate": 0.0004430468399820708, + "loss": 0.4386, + "step": 19999 + }, + { + "epoch": 1.1174400089392966, + "grad_norm": 0.7029333710670471, + "learning_rate": 0.00044301882563872703, + "loss": 0.4359, + "step": 20000 + }, + { + "epoch": 1.1174400089392966, + "eval_cer": 0.08723664255398078, + "eval_loss": 0.32922637462615967, + "eval_runtime": 56.4886, + "eval_samples_per_second": 80.335, + "eval_steps_per_second": 5.028, + "eval_wer": 0.34700148040557527, + "step": 20000 + }, + { + "epoch": 1.1174958795429784, + "grad_norm": 0.5436995029449463, + "learning_rate": 0.0004429908112953833, + "loss": 0.4375, + "step": 20001 + }, + { + "epoch": 1.1175517501466603, + "grad_norm": 0.4022081792354584, + "learning_rate": 0.00044296279695203944, + "loss": 0.4963, + "step": 20002 + }, + { + "epoch": 1.1176076207503423, + "grad_norm": 0.5976927876472473, + "learning_rate": 0.0004429347826086957, + "loss": 0.4478, + "step": 20003 + }, + { + "epoch": 1.117663491354024, + "grad_norm": 0.4700738489627838, + "learning_rate": 0.00044290676826535185, + "loss": 0.411, + "step": 20004 + }, + { + "epoch": 1.117719361957706, + "grad_norm": 1.9874316453933716, + "learning_rate": 0.0004428787539220081, + "loss": 0.4892, + "step": 20005 + }, + { + "epoch": 1.1177752325613879, + "grad_norm": 5.740621089935303, + "learning_rate": 0.00044285073957866426, + "loss": 0.4285, + "step": 20006 + }, + { + "epoch": 1.1178311031650696, + "grad_norm": 0.5261066555976868, + "learning_rate": 0.00044282272523532047, + "loss": 0.4262, + "step": 20007 + }, + { + "epoch": 1.1178869737687516, + "grad_norm": 0.7938253879547119, + "learning_rate": 0.00044279471089197673, + "loss": 0.3453, + "step": 20008 + }, + { + "epoch": 1.1179428443724335, + "grad_norm": 8.160327911376953, + "learning_rate": 0.0004427666965486329, + "loss": 0.4933, + "step": 20009 + }, + { + "epoch": 1.1179987149761152, + "grad_norm": 1.3005993366241455, + "learning_rate": 0.00044273868220528914, + "loss": 0.4159, + "step": 20010 + }, + { + "epoch": 1.1180545855797972, + "grad_norm": 1.718984842300415, + "learning_rate": 0.0004427106678619453, + "loss": 0.4117, + "step": 20011 + }, + { + "epoch": 1.1181104561834792, + "grad_norm": 0.34770137071609497, + "learning_rate": 0.00044268265351860155, + "loss": 0.3514, + "step": 20012 + }, + { + "epoch": 1.1181663267871609, + "grad_norm": 1.4127305746078491, + "learning_rate": 0.0004426546391752577, + "loss": 0.5431, + "step": 20013 + }, + { + "epoch": 1.1182221973908428, + "grad_norm": 0.4334588944911957, + "learning_rate": 0.00044262662483191397, + "loss": 0.3119, + "step": 20014 + }, + { + "epoch": 1.1182780679945248, + "grad_norm": 0.4764646887779236, + "learning_rate": 0.0004425986104885701, + "loss": 0.4601, + "step": 20015 + }, + { + "epoch": 1.1183339385982065, + "grad_norm": 0.43807849287986755, + "learning_rate": 0.0004425705961452264, + "loss": 0.352, + "step": 20016 + }, + { + "epoch": 1.1183898092018885, + "grad_norm": 1.0822324752807617, + "learning_rate": 0.0004425425818018826, + "loss": 0.3574, + "step": 20017 + }, + { + "epoch": 1.1184456798055704, + "grad_norm": 0.5377005338668823, + "learning_rate": 0.0004425145674585388, + "loss": 0.5264, + "step": 20018 + }, + { + "epoch": 1.1185015504092521, + "grad_norm": 3.830247640609741, + "learning_rate": 0.000442486553115195, + "loss": 0.3669, + "step": 20019 + }, + { + "epoch": 1.118557421012934, + "grad_norm": 0.355447381734848, + "learning_rate": 0.0004424585387718512, + "loss": 0.3888, + "step": 20020 + }, + { + "epoch": 1.1186132916166158, + "grad_norm": 0.3872387707233429, + "learning_rate": 0.0004424305244285074, + "loss": 0.4769, + "step": 20021 + }, + { + "epoch": 1.1186691622202978, + "grad_norm": 1.0820597410202026, + "learning_rate": 0.0004424025100851636, + "loss": 0.7994, + "step": 20022 + }, + { + "epoch": 1.1187250328239797, + "grad_norm": 0.5894052386283875, + "learning_rate": 0.0004423744957418198, + "loss": 0.4455, + "step": 20023 + }, + { + "epoch": 1.1187809034276615, + "grad_norm": 3.1321356296539307, + "learning_rate": 0.0004423464813984761, + "loss": 0.3717, + "step": 20024 + }, + { + "epoch": 1.1188367740313434, + "grad_norm": 0.3962198495864868, + "learning_rate": 0.00044231846705513223, + "loss": 0.4665, + "step": 20025 + }, + { + "epoch": 1.1188926446350254, + "grad_norm": 0.4941116273403168, + "learning_rate": 0.0004422904527117885, + "loss": 0.6091, + "step": 20026 + }, + { + "epoch": 1.118948515238707, + "grad_norm": 0.43200209736824036, + "learning_rate": 0.00044226243836844464, + "loss": 0.5141, + "step": 20027 + }, + { + "epoch": 1.119004385842389, + "grad_norm": 0.36056825518608093, + "learning_rate": 0.00044223442402510085, + "loss": 0.5064, + "step": 20028 + }, + { + "epoch": 1.119060256446071, + "grad_norm": 0.35993680357933044, + "learning_rate": 0.00044220640968175705, + "loss": 0.3573, + "step": 20029 + }, + { + "epoch": 1.1191161270497527, + "grad_norm": 3.9201536178588867, + "learning_rate": 0.00044217839533841326, + "loss": 0.5169, + "step": 20030 + }, + { + "epoch": 1.1191719976534347, + "grad_norm": 0.9464017152786255, + "learning_rate": 0.00044215038099506946, + "loss": 0.4185, + "step": 20031 + }, + { + "epoch": 1.1192278682571166, + "grad_norm": 3.448793411254883, + "learning_rate": 0.00044212236665172567, + "loss": 0.3861, + "step": 20032 + }, + { + "epoch": 1.1192837388607983, + "grad_norm": 0.5226978659629822, + "learning_rate": 0.00044209435230838193, + "loss": 0.4339, + "step": 20033 + }, + { + "epoch": 1.1193396094644803, + "grad_norm": 1.0793752670288086, + "learning_rate": 0.0004420663379650381, + "loss": 0.4986, + "step": 20034 + }, + { + "epoch": 1.119395480068162, + "grad_norm": 0.5697492957115173, + "learning_rate": 0.00044203832362169434, + "loss": 0.4243, + "step": 20035 + }, + { + "epoch": 1.119451350671844, + "grad_norm": 0.4843428134918213, + "learning_rate": 0.0004420103092783505, + "loss": 0.4704, + "step": 20036 + }, + { + "epoch": 1.119507221275526, + "grad_norm": 0.4774492681026459, + "learning_rate": 0.00044198229493500675, + "loss": 0.4199, + "step": 20037 + }, + { + "epoch": 1.1195630918792077, + "grad_norm": 0.41296476125717163, + "learning_rate": 0.0004419542805916629, + "loss": 0.4689, + "step": 20038 + }, + { + "epoch": 1.1196189624828896, + "grad_norm": 0.5282856225967407, + "learning_rate": 0.00044192626624831917, + "loss": 0.4547, + "step": 20039 + }, + { + "epoch": 1.1196748330865716, + "grad_norm": 0.972586989402771, + "learning_rate": 0.00044189825190497537, + "loss": 0.3971, + "step": 20040 + }, + { + "epoch": 1.1197307036902533, + "grad_norm": 0.4145061671733856, + "learning_rate": 0.0004418702375616316, + "loss": 0.394, + "step": 20041 + }, + { + "epoch": 1.1197865742939352, + "grad_norm": 0.4057416319847107, + "learning_rate": 0.0004418422232182878, + "loss": 0.3705, + "step": 20042 + }, + { + "epoch": 1.1198424448976172, + "grad_norm": 0.5888508558273315, + "learning_rate": 0.000441814208874944, + "loss": 0.3861, + "step": 20043 + }, + { + "epoch": 1.119898315501299, + "grad_norm": 0.8136501908302307, + "learning_rate": 0.0004417861945316002, + "loss": 0.4117, + "step": 20044 + }, + { + "epoch": 1.1199541861049809, + "grad_norm": 0.3966708183288574, + "learning_rate": 0.0004417581801882564, + "loss": 0.39, + "step": 20045 + }, + { + "epoch": 1.1200100567086628, + "grad_norm": 0.5065010786056519, + "learning_rate": 0.0004417301658449126, + "loss": 0.3836, + "step": 20046 + }, + { + "epoch": 1.1200659273123446, + "grad_norm": 0.29769831895828247, + "learning_rate": 0.0004417021515015688, + "loss": 0.4088, + "step": 20047 + }, + { + "epoch": 1.1201217979160265, + "grad_norm": 0.4703199863433838, + "learning_rate": 0.000441674137158225, + "loss": 0.4838, + "step": 20048 + }, + { + "epoch": 1.1201776685197085, + "grad_norm": 0.47727838158607483, + "learning_rate": 0.0004416461228148812, + "loss": 0.4132, + "step": 20049 + }, + { + "epoch": 1.1202335391233902, + "grad_norm": 0.36109182238578796, + "learning_rate": 0.00044161810847153743, + "loss": 0.3887, + "step": 20050 + }, + { + "epoch": 1.1202894097270721, + "grad_norm": 0.559653639793396, + "learning_rate": 0.00044159009412819363, + "loss": 0.3838, + "step": 20051 + }, + { + "epoch": 1.120345280330754, + "grad_norm": 0.7454647421836853, + "learning_rate": 0.00044156207978484984, + "loss": 0.3665, + "step": 20052 + }, + { + "epoch": 1.1204011509344358, + "grad_norm": 1.0841844081878662, + "learning_rate": 0.00044153406544150605, + "loss": 0.4629, + "step": 20053 + }, + { + "epoch": 1.1204570215381178, + "grad_norm": 0.35207951068878174, + "learning_rate": 0.00044150605109816225, + "loss": 0.3459, + "step": 20054 + }, + { + "epoch": 1.1205128921417995, + "grad_norm": 0.38882583379745483, + "learning_rate": 0.00044147803675481846, + "loss": 0.3766, + "step": 20055 + }, + { + "epoch": 1.1205687627454815, + "grad_norm": 0.3268832862377167, + "learning_rate": 0.0004414500224114747, + "loss": 0.4514, + "step": 20056 + }, + { + "epoch": 1.1206246333491634, + "grad_norm": 0.6356741189956665, + "learning_rate": 0.00044142200806813087, + "loss": 0.4061, + "step": 20057 + }, + { + "epoch": 1.1206805039528451, + "grad_norm": 0.5938470959663391, + "learning_rate": 0.00044139399372478713, + "loss": 0.4668, + "step": 20058 + }, + { + "epoch": 1.120736374556527, + "grad_norm": 0.7623485326766968, + "learning_rate": 0.0004413659793814433, + "loss": 0.4411, + "step": 20059 + }, + { + "epoch": 1.120792245160209, + "grad_norm": 0.513039231300354, + "learning_rate": 0.00044133796503809954, + "loss": 0.4645, + "step": 20060 + }, + { + "epoch": 1.1208481157638908, + "grad_norm": 0.5383438467979431, + "learning_rate": 0.0004413099506947557, + "loss": 0.4283, + "step": 20061 + }, + { + "epoch": 1.1209039863675727, + "grad_norm": 0.3681895136833191, + "learning_rate": 0.00044128193635141195, + "loss": 0.3168, + "step": 20062 + }, + { + "epoch": 1.1209598569712547, + "grad_norm": 0.7265245318412781, + "learning_rate": 0.0004412539220080681, + "loss": 0.5118, + "step": 20063 + }, + { + "epoch": 1.1210157275749364, + "grad_norm": 0.310645192861557, + "learning_rate": 0.00044122590766472436, + "loss": 0.3606, + "step": 20064 + }, + { + "epoch": 1.1210715981786183, + "grad_norm": 2.189579486846924, + "learning_rate": 0.00044119789332138057, + "loss": 0.5446, + "step": 20065 + }, + { + "epoch": 1.1211274687823003, + "grad_norm": 0.5076457262039185, + "learning_rate": 0.0004411698789780368, + "loss": 0.4461, + "step": 20066 + }, + { + "epoch": 1.121183339385982, + "grad_norm": 0.37991806864738464, + "learning_rate": 0.000441141864634693, + "loss": 0.3263, + "step": 20067 + }, + { + "epoch": 1.121239209989664, + "grad_norm": 0.5035305023193359, + "learning_rate": 0.0004411138502913492, + "loss": 0.3719, + "step": 20068 + }, + { + "epoch": 1.1212950805933457, + "grad_norm": 0.6949422359466553, + "learning_rate": 0.0004410858359480054, + "loss": 0.6248, + "step": 20069 + }, + { + "epoch": 1.1213509511970277, + "grad_norm": 0.4439489543437958, + "learning_rate": 0.00044105782160466155, + "loss": 0.3587, + "step": 20070 + }, + { + "epoch": 1.1214068218007096, + "grad_norm": 0.5403600931167603, + "learning_rate": 0.0004410298072613178, + "loss": 0.4818, + "step": 20071 + }, + { + "epoch": 1.1214626924043913, + "grad_norm": 0.43164190649986267, + "learning_rate": 0.000441001792917974, + "loss": 0.4804, + "step": 20072 + }, + { + "epoch": 1.1215185630080733, + "grad_norm": 0.3538266718387604, + "learning_rate": 0.0004409737785746302, + "loss": 0.4455, + "step": 20073 + }, + { + "epoch": 1.1215744336117552, + "grad_norm": 0.35746556520462036, + "learning_rate": 0.0004409457642312864, + "loss": 0.4036, + "step": 20074 + }, + { + "epoch": 1.121630304215437, + "grad_norm": 2.160703182220459, + "learning_rate": 0.00044091774988794263, + "loss": 0.4427, + "step": 20075 + }, + { + "epoch": 1.121686174819119, + "grad_norm": 0.448702871799469, + "learning_rate": 0.00044088973554459883, + "loss": 0.3556, + "step": 20076 + }, + { + "epoch": 1.1217420454228009, + "grad_norm": 0.8050331473350525, + "learning_rate": 0.00044086172120125504, + "loss": 0.3875, + "step": 20077 + }, + { + "epoch": 1.1217979160264826, + "grad_norm": 0.49826565384864807, + "learning_rate": 0.00044083370685791125, + "loss": 0.4332, + "step": 20078 + }, + { + "epoch": 1.1218537866301646, + "grad_norm": 0.4915858805179596, + "learning_rate": 0.0004408056925145675, + "loss": 0.5254, + "step": 20079 + }, + { + "epoch": 1.1219096572338465, + "grad_norm": 0.6065822839736938, + "learning_rate": 0.00044077767817122366, + "loss": 0.6984, + "step": 20080 + }, + { + "epoch": 1.1219655278375282, + "grad_norm": 0.479390949010849, + "learning_rate": 0.0004407496638278799, + "loss": 0.4711, + "step": 20081 + }, + { + "epoch": 1.1220213984412102, + "grad_norm": 0.5806214809417725, + "learning_rate": 0.00044072164948453607, + "loss": 0.444, + "step": 20082 + }, + { + "epoch": 1.1220772690448921, + "grad_norm": 0.46934837102890015, + "learning_rate": 0.00044069363514119233, + "loss": 0.5005, + "step": 20083 + }, + { + "epoch": 1.1221331396485739, + "grad_norm": 0.6456646919250488, + "learning_rate": 0.0004406656207978485, + "loss": 0.4297, + "step": 20084 + }, + { + "epoch": 1.1221890102522558, + "grad_norm": 0.5276886224746704, + "learning_rate": 0.00044063760645450474, + "loss": 0.4597, + "step": 20085 + }, + { + "epoch": 1.1222448808559378, + "grad_norm": 0.6312508583068848, + "learning_rate": 0.0004406095921111609, + "loss": 0.7502, + "step": 20086 + }, + { + "epoch": 1.1223007514596195, + "grad_norm": 0.49736765027046204, + "learning_rate": 0.00044058157776781715, + "loss": 0.3258, + "step": 20087 + }, + { + "epoch": 1.1223566220633014, + "grad_norm": 0.7404358386993408, + "learning_rate": 0.00044055356342447336, + "loss": 0.5854, + "step": 20088 + }, + { + "epoch": 1.1224124926669832, + "grad_norm": 0.998863160610199, + "learning_rate": 0.00044052554908112956, + "loss": 0.5048, + "step": 20089 + }, + { + "epoch": 1.1224683632706651, + "grad_norm": 0.4946717917919159, + "learning_rate": 0.00044049753473778577, + "loss": 0.5036, + "step": 20090 + }, + { + "epoch": 1.122524233874347, + "grad_norm": 0.3947303295135498, + "learning_rate": 0.0004404695203944419, + "loss": 0.4792, + "step": 20091 + }, + { + "epoch": 1.1225801044780288, + "grad_norm": 1.2444053888320923, + "learning_rate": 0.0004404415060510982, + "loss": 0.4408, + "step": 20092 + }, + { + "epoch": 1.1226359750817108, + "grad_norm": 0.4253399968147278, + "learning_rate": 0.00044041349170775433, + "loss": 0.414, + "step": 20093 + }, + { + "epoch": 1.1226918456853927, + "grad_norm": 0.4046750068664551, + "learning_rate": 0.0004403854773644106, + "loss": 0.5568, + "step": 20094 + }, + { + "epoch": 1.1227477162890744, + "grad_norm": 0.39668840169906616, + "learning_rate": 0.0004403574630210668, + "loss": 0.471, + "step": 20095 + }, + { + "epoch": 1.1228035868927564, + "grad_norm": 0.4868960678577423, + "learning_rate": 0.000440329448677723, + "loss": 0.4315, + "step": 20096 + }, + { + "epoch": 1.1228594574964383, + "grad_norm": 0.7309813499450684, + "learning_rate": 0.0004403014343343792, + "loss": 0.5252, + "step": 20097 + }, + { + "epoch": 1.12291532810012, + "grad_norm": 0.35843175649642944, + "learning_rate": 0.0004402734199910354, + "loss": 0.4325, + "step": 20098 + }, + { + "epoch": 1.122971198703802, + "grad_norm": 0.5324684977531433, + "learning_rate": 0.0004402454056476916, + "loss": 0.4578, + "step": 20099 + }, + { + "epoch": 1.1230270693074837, + "grad_norm": 0.4769417643547058, + "learning_rate": 0.00044021739130434783, + "loss": 0.441, + "step": 20100 + }, + { + "epoch": 1.1230829399111657, + "grad_norm": 0.4176371395587921, + "learning_rate": 0.00044018937696100403, + "loss": 0.4981, + "step": 20101 + }, + { + "epoch": 1.1231388105148477, + "grad_norm": 0.38414037227630615, + "learning_rate": 0.00044016136261766024, + "loss": 0.3637, + "step": 20102 + }, + { + "epoch": 1.1231946811185294, + "grad_norm": 0.9423163533210754, + "learning_rate": 0.00044013334827431645, + "loss": 0.4618, + "step": 20103 + }, + { + "epoch": 1.1232505517222113, + "grad_norm": 0.39993491768836975, + "learning_rate": 0.0004401053339309727, + "loss": 0.4254, + "step": 20104 + }, + { + "epoch": 1.1233064223258933, + "grad_norm": 0.47464919090270996, + "learning_rate": 0.00044007731958762886, + "loss": 0.4608, + "step": 20105 + }, + { + "epoch": 1.123362292929575, + "grad_norm": 0.5021829009056091, + "learning_rate": 0.0004400493052442851, + "loss": 0.377, + "step": 20106 + }, + { + "epoch": 1.123418163533257, + "grad_norm": 0.5762068629264832, + "learning_rate": 0.00044002129090094127, + "loss": 0.4806, + "step": 20107 + }, + { + "epoch": 1.123474034136939, + "grad_norm": 0.5538597702980042, + "learning_rate": 0.00043999327655759753, + "loss": 0.4152, + "step": 20108 + }, + { + "epoch": 1.1235299047406206, + "grad_norm": 0.48910465836524963, + "learning_rate": 0.0004399652622142537, + "loss": 0.3513, + "step": 20109 + }, + { + "epoch": 1.1235857753443026, + "grad_norm": 0.48964670300483704, + "learning_rate": 0.00043993724787090994, + "loss": 0.348, + "step": 20110 + }, + { + "epoch": 1.1236416459479845, + "grad_norm": 0.44569075107574463, + "learning_rate": 0.00043990923352756615, + "loss": 0.5142, + "step": 20111 + }, + { + "epoch": 1.1236975165516663, + "grad_norm": 0.40189915895462036, + "learning_rate": 0.00043988121918422235, + "loss": 0.4041, + "step": 20112 + }, + { + "epoch": 1.1237533871553482, + "grad_norm": 0.34688347578048706, + "learning_rate": 0.00043985320484087856, + "loss": 0.3153, + "step": 20113 + }, + { + "epoch": 1.1238092577590302, + "grad_norm": 4.608960151672363, + "learning_rate": 0.0004398251904975347, + "loss": 0.4326, + "step": 20114 + }, + { + "epoch": 1.123865128362712, + "grad_norm": 0.5964322686195374, + "learning_rate": 0.00043979717615419097, + "loss": 0.4102, + "step": 20115 + }, + { + "epoch": 1.1239209989663939, + "grad_norm": 0.5715038180351257, + "learning_rate": 0.0004397691618108471, + "loss": 0.5081, + "step": 20116 + }, + { + "epoch": 1.1239768695700758, + "grad_norm": 0.5079628825187683, + "learning_rate": 0.0004397411474675034, + "loss": 0.4559, + "step": 20117 + }, + { + "epoch": 1.1240327401737575, + "grad_norm": 0.4283066689968109, + "learning_rate": 0.00043971313312415953, + "loss": 0.4457, + "step": 20118 + }, + { + "epoch": 1.1240886107774395, + "grad_norm": 0.8873963356018066, + "learning_rate": 0.0004396851187808158, + "loss": 0.4864, + "step": 20119 + }, + { + "epoch": 1.1241444813811214, + "grad_norm": 1.0644737482070923, + "learning_rate": 0.000439657104437472, + "loss": 0.3319, + "step": 20120 + }, + { + "epoch": 1.1242003519848032, + "grad_norm": 2.406130075454712, + "learning_rate": 0.0004396290900941282, + "loss": 0.368, + "step": 20121 + }, + { + "epoch": 1.1242562225884851, + "grad_norm": 0.3260580599308014, + "learning_rate": 0.0004396010757507844, + "loss": 0.3906, + "step": 20122 + }, + { + "epoch": 1.1243120931921669, + "grad_norm": 0.632070004940033, + "learning_rate": 0.0004395730614074406, + "loss": 0.501, + "step": 20123 + }, + { + "epoch": 1.1243679637958488, + "grad_norm": 0.42790094017982483, + "learning_rate": 0.0004395450470640968, + "loss": 0.3043, + "step": 20124 + }, + { + "epoch": 1.1244238343995308, + "grad_norm": 0.3654153645038605, + "learning_rate": 0.00043951703272075303, + "loss": 0.4189, + "step": 20125 + }, + { + "epoch": 1.1244797050032125, + "grad_norm": 0.442509263753891, + "learning_rate": 0.00043948901837740923, + "loss": 0.5562, + "step": 20126 + }, + { + "epoch": 1.1245355756068944, + "grad_norm": 0.42178380489349365, + "learning_rate": 0.0004394610040340655, + "loss": 0.5396, + "step": 20127 + }, + { + "epoch": 1.1245914462105764, + "grad_norm": 0.7774338722229004, + "learning_rate": 0.00043943298969072165, + "loss": 0.4604, + "step": 20128 + }, + { + "epoch": 1.1246473168142581, + "grad_norm": 0.7510708570480347, + "learning_rate": 0.0004394049753473779, + "loss": 0.4912, + "step": 20129 + }, + { + "epoch": 1.12470318741794, + "grad_norm": 0.41129371523857117, + "learning_rate": 0.00043937696100403406, + "loss": 0.4286, + "step": 20130 + }, + { + "epoch": 1.124759058021622, + "grad_norm": 1.251682996749878, + "learning_rate": 0.0004393489466606903, + "loss": 0.6681, + "step": 20131 + }, + { + "epoch": 1.1248149286253037, + "grad_norm": 0.4819381535053253, + "learning_rate": 0.00043932093231734647, + "loss": 0.5433, + "step": 20132 + }, + { + "epoch": 1.1248707992289857, + "grad_norm": 0.3822026252746582, + "learning_rate": 0.00043929291797400273, + "loss": 0.3823, + "step": 20133 + }, + { + "epoch": 1.1249266698326674, + "grad_norm": 0.589510440826416, + "learning_rate": 0.00043926490363065893, + "loss": 0.4938, + "step": 20134 + }, + { + "epoch": 1.1249825404363494, + "grad_norm": 0.6484969854354858, + "learning_rate": 0.0004392368892873151, + "loss": 0.6447, + "step": 20135 + }, + { + "epoch": 1.1250384110400313, + "grad_norm": 0.3802771270275116, + "learning_rate": 0.00043920887494397135, + "loss": 0.4162, + "step": 20136 + }, + { + "epoch": 1.125094281643713, + "grad_norm": 0.3961308002471924, + "learning_rate": 0.0004391808606006275, + "loss": 0.4167, + "step": 20137 + }, + { + "epoch": 1.125150152247395, + "grad_norm": 1.4671424627304077, + "learning_rate": 0.00043915284625728376, + "loss": 0.4718, + "step": 20138 + }, + { + "epoch": 1.125206022851077, + "grad_norm": 0.4054437577724457, + "learning_rate": 0.0004391248319139399, + "loss": 0.386, + "step": 20139 + }, + { + "epoch": 1.1252618934547587, + "grad_norm": 0.5810218453407288, + "learning_rate": 0.00043909681757059617, + "loss": 0.4354, + "step": 20140 + }, + { + "epoch": 1.1253177640584406, + "grad_norm": 1.6787824630737305, + "learning_rate": 0.0004390688032272523, + "loss": 0.5224, + "step": 20141 + }, + { + "epoch": 1.1253736346621226, + "grad_norm": 0.4002978503704071, + "learning_rate": 0.0004390407888839086, + "loss": 0.3606, + "step": 20142 + }, + { + "epoch": 1.1254295052658043, + "grad_norm": 0.3525383770465851, + "learning_rate": 0.0004390127745405648, + "loss": 0.4492, + "step": 20143 + }, + { + "epoch": 1.1254853758694863, + "grad_norm": 0.4374346435070038, + "learning_rate": 0.000438984760197221, + "loss": 0.3696, + "step": 20144 + }, + { + "epoch": 1.1255412464731682, + "grad_norm": 0.44880765676498413, + "learning_rate": 0.0004389567458538772, + "loss": 0.3906, + "step": 20145 + }, + { + "epoch": 1.12559711707685, + "grad_norm": 0.39351052045822144, + "learning_rate": 0.0004389287315105334, + "loss": 0.3784, + "step": 20146 + }, + { + "epoch": 1.125652987680532, + "grad_norm": 0.4356077015399933, + "learning_rate": 0.0004389007171671896, + "loss": 0.3882, + "step": 20147 + }, + { + "epoch": 1.1257088582842139, + "grad_norm": 0.29777199029922485, + "learning_rate": 0.0004388727028238458, + "loss": 0.3275, + "step": 20148 + }, + { + "epoch": 1.1257647288878956, + "grad_norm": 0.6534481644630432, + "learning_rate": 0.000438844688480502, + "loss": 0.5748, + "step": 20149 + }, + { + "epoch": 1.1258205994915775, + "grad_norm": 0.4154852330684662, + "learning_rate": 0.0004388166741371583, + "loss": 0.3354, + "step": 20150 + }, + { + "epoch": 1.1258764700952595, + "grad_norm": 1.5485479831695557, + "learning_rate": 0.00043878865979381443, + "loss": 0.4608, + "step": 20151 + }, + { + "epoch": 1.1259323406989412, + "grad_norm": 0.4349132478237152, + "learning_rate": 0.0004387606454504707, + "loss": 0.6305, + "step": 20152 + }, + { + "epoch": 1.1259882113026232, + "grad_norm": 0.5018962025642395, + "learning_rate": 0.00043873263110712684, + "loss": 0.3867, + "step": 20153 + }, + { + "epoch": 1.1260440819063051, + "grad_norm": 0.3164904713630676, + "learning_rate": 0.0004387046167637831, + "loss": 0.3242, + "step": 20154 + }, + { + "epoch": 1.1260999525099868, + "grad_norm": 0.47122934460639954, + "learning_rate": 0.00043867660242043926, + "loss": 0.5135, + "step": 20155 + }, + { + "epoch": 1.1261558231136688, + "grad_norm": 1.2040150165557861, + "learning_rate": 0.00043864858807709546, + "loss": 0.3233, + "step": 20156 + }, + { + "epoch": 1.1262116937173505, + "grad_norm": 0.40673068165779114, + "learning_rate": 0.00043862057373375167, + "loss": 0.4179, + "step": 20157 + }, + { + "epoch": 1.1262675643210325, + "grad_norm": 9.282291412353516, + "learning_rate": 0.0004385925593904079, + "loss": 0.4038, + "step": 20158 + }, + { + "epoch": 1.1263234349247144, + "grad_norm": 0.8324569463729858, + "learning_rate": 0.00043856454504706413, + "loss": 0.6548, + "step": 20159 + }, + { + "epoch": 1.1263793055283962, + "grad_norm": 0.7963978052139282, + "learning_rate": 0.0004385365307037203, + "loss": 0.4517, + "step": 20160 + }, + { + "epoch": 1.126435176132078, + "grad_norm": 0.40850841999053955, + "learning_rate": 0.00043850851636037655, + "loss": 0.3827, + "step": 20161 + }, + { + "epoch": 1.12649104673576, + "grad_norm": 0.5321620106697083, + "learning_rate": 0.0004384805020170327, + "loss": 0.6837, + "step": 20162 + }, + { + "epoch": 1.1265469173394418, + "grad_norm": 0.6155428290367126, + "learning_rate": 0.00043845248767368896, + "loss": 0.427, + "step": 20163 + }, + { + "epoch": 1.1266027879431237, + "grad_norm": 1.515643835067749, + "learning_rate": 0.0004384244733303451, + "loss": 0.4492, + "step": 20164 + }, + { + "epoch": 1.1266586585468057, + "grad_norm": 0.3522374629974365, + "learning_rate": 0.00043839645898700137, + "loss": 0.4019, + "step": 20165 + }, + { + "epoch": 1.1267145291504874, + "grad_norm": 0.5564371347427368, + "learning_rate": 0.0004383684446436576, + "loss": 0.441, + "step": 20166 + }, + { + "epoch": 1.1267703997541694, + "grad_norm": 0.40476953983306885, + "learning_rate": 0.0004383404303003138, + "loss": 0.496, + "step": 20167 + }, + { + "epoch": 1.126826270357851, + "grad_norm": 0.3750860393047333, + "learning_rate": 0.00043831241595697, + "loss": 0.3936, + "step": 20168 + }, + { + "epoch": 1.126882140961533, + "grad_norm": 0.3856782615184784, + "learning_rate": 0.0004382844016136262, + "loss": 0.3817, + "step": 20169 + }, + { + "epoch": 1.126938011565215, + "grad_norm": 0.4521386921405792, + "learning_rate": 0.0004382563872702824, + "loss": 0.3982, + "step": 20170 + }, + { + "epoch": 1.1269938821688967, + "grad_norm": 0.48070481419563293, + "learning_rate": 0.0004382283729269386, + "loss": 0.4501, + "step": 20171 + }, + { + "epoch": 1.1270497527725787, + "grad_norm": 0.4459656774997711, + "learning_rate": 0.0004382003585835948, + "loss": 0.3695, + "step": 20172 + }, + { + "epoch": 1.1271056233762606, + "grad_norm": 0.8026424050331116, + "learning_rate": 0.000438172344240251, + "loss": 0.4649, + "step": 20173 + }, + { + "epoch": 1.1271614939799424, + "grad_norm": 0.4459923207759857, + "learning_rate": 0.0004381443298969072, + "loss": 0.3865, + "step": 20174 + }, + { + "epoch": 1.1272173645836243, + "grad_norm": 0.444762647151947, + "learning_rate": 0.0004381163155535635, + "loss": 0.4062, + "step": 20175 + }, + { + "epoch": 1.1272732351873063, + "grad_norm": 0.6483334302902222, + "learning_rate": 0.00043808830121021963, + "loss": 0.4616, + "step": 20176 + }, + { + "epoch": 1.127329105790988, + "grad_norm": 0.44525039196014404, + "learning_rate": 0.00043806028686687584, + "loss": 0.3841, + "step": 20177 + }, + { + "epoch": 1.12738497639467, + "grad_norm": 1.5476775169372559, + "learning_rate": 0.00043803227252353204, + "loss": 0.4043, + "step": 20178 + }, + { + "epoch": 1.127440846998352, + "grad_norm": 0.336083322763443, + "learning_rate": 0.00043800425818018825, + "loss": 0.3238, + "step": 20179 + }, + { + "epoch": 1.1274967176020336, + "grad_norm": 0.7387925386428833, + "learning_rate": 0.00043797624383684446, + "loss": 0.5157, + "step": 20180 + }, + { + "epoch": 1.1275525882057156, + "grad_norm": 0.3204903304576874, + "learning_rate": 0.00043794822949350066, + "loss": 0.4322, + "step": 20181 + }, + { + "epoch": 1.1276084588093975, + "grad_norm": 0.49290749430656433, + "learning_rate": 0.0004379202151501569, + "loss": 0.481, + "step": 20182 + }, + { + "epoch": 1.1276643294130793, + "grad_norm": 0.41995590925216675, + "learning_rate": 0.0004378922008068131, + "loss": 0.4404, + "step": 20183 + }, + { + "epoch": 1.1277202000167612, + "grad_norm": 0.4278847277164459, + "learning_rate": 0.00043786418646346933, + "loss": 0.4277, + "step": 20184 + }, + { + "epoch": 1.1277760706204432, + "grad_norm": 0.3842882513999939, + "learning_rate": 0.0004378361721201255, + "loss": 0.4332, + "step": 20185 + }, + { + "epoch": 1.127831941224125, + "grad_norm": 0.8698646426200867, + "learning_rate": 0.00043780815777678175, + "loss": 0.4017, + "step": 20186 + }, + { + "epoch": 1.1278878118278068, + "grad_norm": 0.4830312132835388, + "learning_rate": 0.0004377801434334379, + "loss": 0.4328, + "step": 20187 + }, + { + "epoch": 1.1279436824314888, + "grad_norm": 0.4358850121498108, + "learning_rate": 0.00043775212909009416, + "loss": 0.4367, + "step": 20188 + }, + { + "epoch": 1.1279995530351705, + "grad_norm": 2.3276658058166504, + "learning_rate": 0.0004377241147467503, + "loss": 0.4835, + "step": 20189 + }, + { + "epoch": 1.1280554236388525, + "grad_norm": 1.3835389614105225, + "learning_rate": 0.00043769610040340657, + "loss": 0.4449, + "step": 20190 + }, + { + "epoch": 1.1281112942425342, + "grad_norm": 0.4616181552410126, + "learning_rate": 0.0004376680860600628, + "loss": 0.34, + "step": 20191 + }, + { + "epoch": 1.1281671648462162, + "grad_norm": 0.4229688048362732, + "learning_rate": 0.000437640071716719, + "loss": 0.347, + "step": 20192 + }, + { + "epoch": 1.128223035449898, + "grad_norm": 0.49801090359687805, + "learning_rate": 0.0004376120573733752, + "loss": 0.3345, + "step": 20193 + }, + { + "epoch": 1.1282789060535798, + "grad_norm": 0.4398432672023773, + "learning_rate": 0.0004375840430300314, + "loss": 0.436, + "step": 20194 + }, + { + "epoch": 1.1283347766572618, + "grad_norm": 0.3930342495441437, + "learning_rate": 0.0004375560286866876, + "loss": 0.3787, + "step": 20195 + }, + { + "epoch": 1.1283906472609437, + "grad_norm": 0.5603844523429871, + "learning_rate": 0.0004375280143433438, + "loss": 0.4605, + "step": 20196 + }, + { + "epoch": 1.1284465178646255, + "grad_norm": 0.5162261128425598, + "learning_rate": 0.0004375, + "loss": 0.4912, + "step": 20197 + }, + { + "epoch": 1.1285023884683074, + "grad_norm": 4.1075758934021, + "learning_rate": 0.0004374719856566562, + "loss": 0.4899, + "step": 20198 + }, + { + "epoch": 1.1285582590719894, + "grad_norm": 0.5793423056602478, + "learning_rate": 0.0004374439713133124, + "loss": 0.4064, + "step": 20199 + }, + { + "epoch": 1.128614129675671, + "grad_norm": 0.47728127241134644, + "learning_rate": 0.0004374159569699686, + "loss": 0.4596, + "step": 20200 + }, + { + "epoch": 1.128670000279353, + "grad_norm": 0.47217175364494324, + "learning_rate": 0.00043738794262662483, + "loss": 0.3322, + "step": 20201 + }, + { + "epoch": 1.1287258708830348, + "grad_norm": 0.625697135925293, + "learning_rate": 0.00043735992828328104, + "loss": 0.3601, + "step": 20202 + }, + { + "epoch": 1.1287817414867167, + "grad_norm": 0.6087594032287598, + "learning_rate": 0.00043733191393993724, + "loss": 0.4384, + "step": 20203 + }, + { + "epoch": 1.1288376120903987, + "grad_norm": 0.6659859418869019, + "learning_rate": 0.00043730389959659345, + "loss": 0.3764, + "step": 20204 + }, + { + "epoch": 1.1288934826940804, + "grad_norm": 0.5835936069488525, + "learning_rate": 0.0004372758852532497, + "loss": 0.6626, + "step": 20205 + }, + { + "epoch": 1.1289493532977624, + "grad_norm": 2.4032580852508545, + "learning_rate": 0.00043724787090990586, + "loss": 0.4809, + "step": 20206 + }, + { + "epoch": 1.1290052239014443, + "grad_norm": 0.39018839597702026, + "learning_rate": 0.0004372198565665621, + "loss": 0.4052, + "step": 20207 + }, + { + "epoch": 1.129061094505126, + "grad_norm": 0.5010324716567993, + "learning_rate": 0.0004371918422232183, + "loss": 0.4623, + "step": 20208 + }, + { + "epoch": 1.129116965108808, + "grad_norm": 0.3385714590549469, + "learning_rate": 0.00043716382787987453, + "loss": 0.3867, + "step": 20209 + }, + { + "epoch": 1.12917283571249, + "grad_norm": 0.4571993350982666, + "learning_rate": 0.0004371358135365307, + "loss": 0.437, + "step": 20210 + }, + { + "epoch": 1.1292287063161717, + "grad_norm": 0.47834813594818115, + "learning_rate": 0.00043710779919318694, + "loss": 0.4191, + "step": 20211 + }, + { + "epoch": 1.1292845769198536, + "grad_norm": 0.3952212631702423, + "learning_rate": 0.0004370797848498431, + "loss": 0.3488, + "step": 20212 + }, + { + "epoch": 1.1293404475235356, + "grad_norm": 0.41578876972198486, + "learning_rate": 0.00043705177050649936, + "loss": 0.423, + "step": 20213 + }, + { + "epoch": 1.1293963181272173, + "grad_norm": 0.7042384147644043, + "learning_rate": 0.00043702375616315556, + "loss": 0.3359, + "step": 20214 + }, + { + "epoch": 1.1294521887308993, + "grad_norm": 0.5852886438369751, + "learning_rate": 0.00043699574181981177, + "loss": 0.4878, + "step": 20215 + }, + { + "epoch": 1.1295080593345812, + "grad_norm": 0.4415552318096161, + "learning_rate": 0.000436967727476468, + "loss": 0.3791, + "step": 20216 + }, + { + "epoch": 1.129563929938263, + "grad_norm": 0.5413569808006287, + "learning_rate": 0.0004369397131331242, + "loss": 0.3678, + "step": 20217 + }, + { + "epoch": 1.1296198005419449, + "grad_norm": 0.3833702802658081, + "learning_rate": 0.0004369116987897804, + "loss": 0.3679, + "step": 20218 + }, + { + "epoch": 1.1296756711456268, + "grad_norm": 0.42483600974082947, + "learning_rate": 0.00043688368444643654, + "loss": 0.3749, + "step": 20219 + }, + { + "epoch": 1.1297315417493086, + "grad_norm": 0.41900384426116943, + "learning_rate": 0.0004368556701030928, + "loss": 0.4221, + "step": 20220 + }, + { + "epoch": 1.1297874123529905, + "grad_norm": 0.3020046055316925, + "learning_rate": 0.000436827655759749, + "loss": 0.3983, + "step": 20221 + }, + { + "epoch": 1.1298432829566725, + "grad_norm": 0.5248331427574158, + "learning_rate": 0.0004367996414164052, + "loss": 0.4453, + "step": 20222 + }, + { + "epoch": 1.1298991535603542, + "grad_norm": 1.4825390577316284, + "learning_rate": 0.0004367716270730614, + "loss": 0.4016, + "step": 20223 + }, + { + "epoch": 1.1299550241640361, + "grad_norm": 0.5738599300384521, + "learning_rate": 0.0004367436127297176, + "loss": 0.3996, + "step": 20224 + }, + { + "epoch": 1.1300108947677179, + "grad_norm": 0.4383203983306885, + "learning_rate": 0.0004367155983863738, + "loss": 0.4442, + "step": 20225 + }, + { + "epoch": 1.1300667653713998, + "grad_norm": 0.7375996112823486, + "learning_rate": 0.00043668758404303003, + "loss": 0.5104, + "step": 20226 + }, + { + "epoch": 1.1301226359750818, + "grad_norm": 0.42466095089912415, + "learning_rate": 0.00043665956969968624, + "loss": 0.4494, + "step": 20227 + }, + { + "epoch": 1.1301785065787635, + "grad_norm": 0.42374876141548157, + "learning_rate": 0.00043663155535634244, + "loss": 0.5369, + "step": 20228 + }, + { + "epoch": 1.1302343771824455, + "grad_norm": 0.4740217626094818, + "learning_rate": 0.00043660354101299865, + "loss": 0.4948, + "step": 20229 + }, + { + "epoch": 1.1302902477861274, + "grad_norm": 0.5311890840530396, + "learning_rate": 0.0004365755266696549, + "loss": 0.53, + "step": 20230 + }, + { + "epoch": 1.1303461183898091, + "grad_norm": 0.9761452674865723, + "learning_rate": 0.00043654751232631106, + "loss": 0.4107, + "step": 20231 + }, + { + "epoch": 1.130401988993491, + "grad_norm": 0.7974960207939148, + "learning_rate": 0.0004365194979829673, + "loss": 0.4788, + "step": 20232 + }, + { + "epoch": 1.130457859597173, + "grad_norm": 0.9113526940345764, + "learning_rate": 0.00043649148363962347, + "loss": 0.5523, + "step": 20233 + }, + { + "epoch": 1.1305137302008548, + "grad_norm": 0.7450759410858154, + "learning_rate": 0.00043646346929627973, + "loss": 0.4529, + "step": 20234 + }, + { + "epoch": 1.1305696008045367, + "grad_norm": 0.40587303042411804, + "learning_rate": 0.0004364354549529359, + "loss": 0.3666, + "step": 20235 + }, + { + "epoch": 1.1306254714082185, + "grad_norm": 14.641036987304688, + "learning_rate": 0.00043640744060959214, + "loss": 0.4366, + "step": 20236 + }, + { + "epoch": 1.1306813420119004, + "grad_norm": 0.6041406393051147, + "learning_rate": 0.00043637942626624835, + "loss": 0.4068, + "step": 20237 + }, + { + "epoch": 1.1307372126155824, + "grad_norm": 0.39400455355644226, + "learning_rate": 0.00043635141192290456, + "loss": 0.2667, + "step": 20238 + }, + { + "epoch": 1.130793083219264, + "grad_norm": 0.7192378640174866, + "learning_rate": 0.00043632339757956076, + "loss": 0.4009, + "step": 20239 + }, + { + "epoch": 1.130848953822946, + "grad_norm": 0.48670870065689087, + "learning_rate": 0.0004362953832362169, + "loss": 0.4372, + "step": 20240 + }, + { + "epoch": 1.130904824426628, + "grad_norm": 0.6627904772758484, + "learning_rate": 0.0004362673688928732, + "loss": 0.4921, + "step": 20241 + }, + { + "epoch": 1.1309606950303097, + "grad_norm": 0.4957704246044159, + "learning_rate": 0.0004362393545495293, + "loss": 0.4011, + "step": 20242 + }, + { + "epoch": 1.1310165656339917, + "grad_norm": 0.4111786484718323, + "learning_rate": 0.0004362113402061856, + "loss": 0.3783, + "step": 20243 + }, + { + "epoch": 1.1310724362376736, + "grad_norm": 3.057539224624634, + "learning_rate": 0.00043618332586284174, + "loss": 0.4861, + "step": 20244 + }, + { + "epoch": 1.1311283068413553, + "grad_norm": 0.41411465406417847, + "learning_rate": 0.000436155311519498, + "loss": 0.3828, + "step": 20245 + }, + { + "epoch": 1.1311841774450373, + "grad_norm": 0.3174881637096405, + "learning_rate": 0.0004361272971761542, + "loss": 0.3745, + "step": 20246 + }, + { + "epoch": 1.1312400480487192, + "grad_norm": 0.412020206451416, + "learning_rate": 0.0004360992828328104, + "loss": 0.4181, + "step": 20247 + }, + { + "epoch": 1.131295918652401, + "grad_norm": 0.4328691065311432, + "learning_rate": 0.0004360712684894666, + "loss": 0.5361, + "step": 20248 + }, + { + "epoch": 1.131351789256083, + "grad_norm": 0.6218148469924927, + "learning_rate": 0.0004360432541461228, + "loss": 0.5947, + "step": 20249 + }, + { + "epoch": 1.1314076598597649, + "grad_norm": 0.7595844864845276, + "learning_rate": 0.000436015239802779, + "loss": 0.4398, + "step": 20250 + }, + { + "epoch": 1.1314635304634466, + "grad_norm": 0.5683853030204773, + "learning_rate": 0.00043598722545943523, + "loss": 0.461, + "step": 20251 + }, + { + "epoch": 1.1315194010671286, + "grad_norm": 1.5145710706710815, + "learning_rate": 0.00043595921111609144, + "loss": 0.4305, + "step": 20252 + }, + { + "epoch": 1.1315752716708105, + "grad_norm": 0.4300115704536438, + "learning_rate": 0.0004359311967727477, + "loss": 0.4671, + "step": 20253 + }, + { + "epoch": 1.1316311422744922, + "grad_norm": 0.4081111252307892, + "learning_rate": 0.00043590318242940385, + "loss": 0.4159, + "step": 20254 + }, + { + "epoch": 1.1316870128781742, + "grad_norm": 0.5081965923309326, + "learning_rate": 0.0004358751680860601, + "loss": 0.4399, + "step": 20255 + }, + { + "epoch": 1.1317428834818561, + "grad_norm": 0.4378863275051117, + "learning_rate": 0.00043584715374271626, + "loss": 0.4804, + "step": 20256 + }, + { + "epoch": 1.1317987540855379, + "grad_norm": 0.31742608547210693, + "learning_rate": 0.0004358191393993725, + "loss": 0.3593, + "step": 20257 + }, + { + "epoch": 1.1318546246892198, + "grad_norm": 0.5497108697891235, + "learning_rate": 0.00043579112505602867, + "loss": 0.445, + "step": 20258 + }, + { + "epoch": 1.1319104952929016, + "grad_norm": 0.3920937776565552, + "learning_rate": 0.00043576311071268493, + "loss": 0.3796, + "step": 20259 + }, + { + "epoch": 1.1319663658965835, + "grad_norm": 0.6654509902000427, + "learning_rate": 0.0004357350963693411, + "loss": 0.513, + "step": 20260 + }, + { + "epoch": 1.1320222365002655, + "grad_norm": 0.4562523365020752, + "learning_rate": 0.0004357070820259973, + "loss": 0.4272, + "step": 20261 + }, + { + "epoch": 1.1320781071039472, + "grad_norm": 0.2954956591129303, + "learning_rate": 0.00043567906768265355, + "loss": 0.2822, + "step": 20262 + }, + { + "epoch": 1.1321339777076291, + "grad_norm": 0.4238959848880768, + "learning_rate": 0.0004356510533393097, + "loss": 0.656, + "step": 20263 + }, + { + "epoch": 1.132189848311311, + "grad_norm": 0.7676238417625427, + "learning_rate": 0.00043562303899596596, + "loss": 0.3457, + "step": 20264 + }, + { + "epoch": 1.1322457189149928, + "grad_norm": 2.0430006980895996, + "learning_rate": 0.0004355950246526221, + "loss": 0.4964, + "step": 20265 + }, + { + "epoch": 1.1323015895186748, + "grad_norm": 0.4429648220539093, + "learning_rate": 0.0004355670103092784, + "loss": 0.4478, + "step": 20266 + }, + { + "epoch": 1.1323574601223567, + "grad_norm": 0.5286803245544434, + "learning_rate": 0.0004355389959659345, + "loss": 0.388, + "step": 20267 + }, + { + "epoch": 1.1324133307260384, + "grad_norm": 0.3572436273097992, + "learning_rate": 0.0004355109816225908, + "loss": 0.4392, + "step": 20268 + }, + { + "epoch": 1.1324692013297204, + "grad_norm": 0.8275236487388611, + "learning_rate": 0.000435482967279247, + "loss": 0.4106, + "step": 20269 + }, + { + "epoch": 1.1325250719334021, + "grad_norm": 0.38108009099960327, + "learning_rate": 0.0004354549529359032, + "loss": 0.4197, + "step": 20270 + }, + { + "epoch": 1.132580942537084, + "grad_norm": 0.7513809204101562, + "learning_rate": 0.0004354269385925594, + "loss": 0.5089, + "step": 20271 + }, + { + "epoch": 1.132636813140766, + "grad_norm": 0.41280707716941833, + "learning_rate": 0.0004353989242492156, + "loss": 0.3796, + "step": 20272 + }, + { + "epoch": 1.1326926837444478, + "grad_norm": 0.4312283992767334, + "learning_rate": 0.0004353709099058718, + "loss": 0.3909, + "step": 20273 + }, + { + "epoch": 1.1327485543481297, + "grad_norm": 0.36874279379844666, + "learning_rate": 0.000435342895562528, + "loss": 0.3667, + "step": 20274 + }, + { + "epoch": 1.1328044249518117, + "grad_norm": 0.40565186738967896, + "learning_rate": 0.0004353148812191842, + "loss": 0.3591, + "step": 20275 + }, + { + "epoch": 1.1328602955554934, + "grad_norm": 0.46471142768859863, + "learning_rate": 0.0004352868668758405, + "loss": 0.4543, + "step": 20276 + }, + { + "epoch": 1.1329161661591753, + "grad_norm": 0.560067355632782, + "learning_rate": 0.00043525885253249664, + "loss": 0.4233, + "step": 20277 + }, + { + "epoch": 1.1329720367628573, + "grad_norm": 0.3617836534976959, + "learning_rate": 0.0004352308381891529, + "loss": 0.3776, + "step": 20278 + }, + { + "epoch": 1.133027907366539, + "grad_norm": 0.30421361327171326, + "learning_rate": 0.00043520282384580905, + "loss": 0.2916, + "step": 20279 + }, + { + "epoch": 1.133083777970221, + "grad_norm": 0.5143460631370544, + "learning_rate": 0.0004351748095024653, + "loss": 0.4141, + "step": 20280 + }, + { + "epoch": 1.133139648573903, + "grad_norm": 3.121969223022461, + "learning_rate": 0.00043514679515912146, + "loss": 0.4127, + "step": 20281 + }, + { + "epoch": 1.1331955191775847, + "grad_norm": 0.3243098258972168, + "learning_rate": 0.00043511878081577767, + "loss": 0.3482, + "step": 20282 + }, + { + "epoch": 1.1332513897812666, + "grad_norm": 0.6707220077514648, + "learning_rate": 0.00043509076647243387, + "loss": 0.5178, + "step": 20283 + }, + { + "epoch": 1.1333072603849486, + "grad_norm": 0.5334377288818359, + "learning_rate": 0.0004350627521290901, + "loss": 0.3822, + "step": 20284 + }, + { + "epoch": 1.1333631309886303, + "grad_norm": 0.5402542948722839, + "learning_rate": 0.00043503473778574634, + "loss": 0.4158, + "step": 20285 + }, + { + "epoch": 1.1334190015923122, + "grad_norm": 0.877448558807373, + "learning_rate": 0.0004350067234424025, + "loss": 0.436, + "step": 20286 + }, + { + "epoch": 1.1334748721959942, + "grad_norm": 2.734869956970215, + "learning_rate": 0.00043497870909905875, + "loss": 0.4259, + "step": 20287 + }, + { + "epoch": 1.133530742799676, + "grad_norm": 0.5388036966323853, + "learning_rate": 0.0004349506947557149, + "loss": 0.5476, + "step": 20288 + }, + { + "epoch": 1.1335866134033579, + "grad_norm": 1.694144368171692, + "learning_rate": 0.00043492268041237116, + "loss": 0.5043, + "step": 20289 + }, + { + "epoch": 1.1336424840070398, + "grad_norm": 0.5879721641540527, + "learning_rate": 0.0004348946660690273, + "loss": 0.528, + "step": 20290 + }, + { + "epoch": 1.1336983546107215, + "grad_norm": 2.2570083141326904, + "learning_rate": 0.00043486665172568357, + "loss": 0.3049, + "step": 20291 + }, + { + "epoch": 1.1337542252144035, + "grad_norm": 0.3488616645336151, + "learning_rate": 0.0004348386373823398, + "loss": 0.3593, + "step": 20292 + }, + { + "epoch": 1.1338100958180852, + "grad_norm": 0.7918991446495056, + "learning_rate": 0.000434810623038996, + "loss": 0.4197, + "step": 20293 + }, + { + "epoch": 1.1338659664217672, + "grad_norm": 0.5704822540283203, + "learning_rate": 0.0004347826086956522, + "loss": 0.3408, + "step": 20294 + }, + { + "epoch": 1.1339218370254491, + "grad_norm": 1.1701263189315796, + "learning_rate": 0.0004347545943523084, + "loss": 0.3467, + "step": 20295 + }, + { + "epoch": 1.1339777076291309, + "grad_norm": 0.48633742332458496, + "learning_rate": 0.0004347265800089646, + "loss": 0.3425, + "step": 20296 + }, + { + "epoch": 1.1340335782328128, + "grad_norm": 0.8251427412033081, + "learning_rate": 0.0004346985656656208, + "loss": 0.3949, + "step": 20297 + }, + { + "epoch": 1.1340894488364948, + "grad_norm": 0.43474870920181274, + "learning_rate": 0.000434670551322277, + "loss": 0.4697, + "step": 20298 + }, + { + "epoch": 1.1341453194401765, + "grad_norm": 0.3737003803253174, + "learning_rate": 0.0004346425369789332, + "loss": 0.4277, + "step": 20299 + }, + { + "epoch": 1.1342011900438584, + "grad_norm": 1.8247487545013428, + "learning_rate": 0.0004346145226355894, + "loss": 0.4599, + "step": 20300 + }, + { + "epoch": 1.1342570606475404, + "grad_norm": 0.4165326654911041, + "learning_rate": 0.0004345865082922457, + "loss": 0.3282, + "step": 20301 + }, + { + "epoch": 1.1343129312512221, + "grad_norm": 0.9513731598854065, + "learning_rate": 0.00043455849394890184, + "loss": 0.3515, + "step": 20302 + }, + { + "epoch": 1.134368801854904, + "grad_norm": 1.063672423362732, + "learning_rate": 0.0004345304796055581, + "loss": 0.4604, + "step": 20303 + }, + { + "epoch": 1.1344246724585858, + "grad_norm": 1.015222430229187, + "learning_rate": 0.00043450246526221425, + "loss": 0.353, + "step": 20304 + }, + { + "epoch": 1.1344805430622678, + "grad_norm": 0.5367241501808167, + "learning_rate": 0.00043447445091887045, + "loss": 0.4731, + "step": 20305 + }, + { + "epoch": 1.1345364136659497, + "grad_norm": 0.45107945799827576, + "learning_rate": 0.00043444643657552666, + "loss": 0.488, + "step": 20306 + }, + { + "epoch": 1.1345922842696314, + "grad_norm": 0.4827367663383484, + "learning_rate": 0.00043441842223218287, + "loss": 0.3835, + "step": 20307 + }, + { + "epoch": 1.1346481548733134, + "grad_norm": 0.5115412473678589, + "learning_rate": 0.0004343904078888391, + "loss": 0.3681, + "step": 20308 + }, + { + "epoch": 1.1347040254769953, + "grad_norm": 0.6533808708190918, + "learning_rate": 0.0004343623935454953, + "loss": 0.5297, + "step": 20309 + }, + { + "epoch": 1.134759896080677, + "grad_norm": 1.0113577842712402, + "learning_rate": 0.00043433437920215154, + "loss": 0.328, + "step": 20310 + }, + { + "epoch": 1.134815766684359, + "grad_norm": 0.672483503818512, + "learning_rate": 0.0004343063648588077, + "loss": 0.4384, + "step": 20311 + }, + { + "epoch": 1.134871637288041, + "grad_norm": 0.3548274636268616, + "learning_rate": 0.00043427835051546395, + "loss": 0.3553, + "step": 20312 + }, + { + "epoch": 1.1349275078917227, + "grad_norm": 3.2015154361724854, + "learning_rate": 0.0004342503361721201, + "loss": 0.4488, + "step": 20313 + }, + { + "epoch": 1.1349833784954046, + "grad_norm": 0.6661390066146851, + "learning_rate": 0.00043422232182877636, + "loss": 0.4912, + "step": 20314 + }, + { + "epoch": 1.1350392490990866, + "grad_norm": 3.59792423248291, + "learning_rate": 0.0004341943074854325, + "loss": 0.5113, + "step": 20315 + }, + { + "epoch": 1.1350951197027683, + "grad_norm": 0.6911478638648987, + "learning_rate": 0.00043416629314208877, + "loss": 0.4938, + "step": 20316 + }, + { + "epoch": 1.1351509903064503, + "grad_norm": 0.697304368019104, + "learning_rate": 0.000434138278798745, + "loss": 0.4495, + "step": 20317 + }, + { + "epoch": 1.1352068609101322, + "grad_norm": 0.3663732409477234, + "learning_rate": 0.0004341102644554012, + "loss": 0.4372, + "step": 20318 + }, + { + "epoch": 1.135262731513814, + "grad_norm": 0.3473217189311981, + "learning_rate": 0.0004340822501120574, + "loss": 0.4483, + "step": 20319 + }, + { + "epoch": 1.135318602117496, + "grad_norm": 0.4119468331336975, + "learning_rate": 0.0004340542357687136, + "loss": 0.3608, + "step": 20320 + }, + { + "epoch": 1.1353744727211779, + "grad_norm": 0.7618096470832825, + "learning_rate": 0.0004340262214253698, + "loss": 0.4544, + "step": 20321 + }, + { + "epoch": 1.1354303433248596, + "grad_norm": 0.7409054040908813, + "learning_rate": 0.000433998207082026, + "loss": 0.4114, + "step": 20322 + }, + { + "epoch": 1.1354862139285415, + "grad_norm": 0.6367514133453369, + "learning_rate": 0.0004339701927386822, + "loss": 0.4332, + "step": 20323 + }, + { + "epoch": 1.1355420845322235, + "grad_norm": 0.8392884731292725, + "learning_rate": 0.0004339421783953385, + "loss": 0.5496, + "step": 20324 + }, + { + "epoch": 1.1355979551359052, + "grad_norm": 0.34370169043540955, + "learning_rate": 0.0004339141640519946, + "loss": 0.3481, + "step": 20325 + }, + { + "epoch": 1.1356538257395872, + "grad_norm": 0.5430455803871155, + "learning_rate": 0.00043388614970865083, + "loss": 0.4828, + "step": 20326 + }, + { + "epoch": 1.135709696343269, + "grad_norm": 0.7328382134437561, + "learning_rate": 0.00043385813536530704, + "loss": 0.4597, + "step": 20327 + }, + { + "epoch": 1.1357655669469509, + "grad_norm": 0.9107128977775574, + "learning_rate": 0.00043383012102196324, + "loss": 0.3652, + "step": 20328 + }, + { + "epoch": 1.1358214375506328, + "grad_norm": 0.47317057847976685, + "learning_rate": 0.00043380210667861945, + "loss": 0.4547, + "step": 20329 + }, + { + "epoch": 1.1358773081543145, + "grad_norm": 0.5106876492500305, + "learning_rate": 0.00043377409233527565, + "loss": 0.5663, + "step": 20330 + }, + { + "epoch": 1.1359331787579965, + "grad_norm": 4.9836106300354, + "learning_rate": 0.0004337460779919319, + "loss": 0.3401, + "step": 20331 + }, + { + "epoch": 1.1359890493616784, + "grad_norm": 0.8687230944633484, + "learning_rate": 0.00043371806364858807, + "loss": 0.4378, + "step": 20332 + }, + { + "epoch": 1.1360449199653602, + "grad_norm": 0.593262791633606, + "learning_rate": 0.0004336900493052443, + "loss": 0.3868, + "step": 20333 + }, + { + "epoch": 1.1361007905690421, + "grad_norm": 0.6652164459228516, + "learning_rate": 0.0004336620349619005, + "loss": 0.4011, + "step": 20334 + }, + { + "epoch": 1.1361566611727238, + "grad_norm": Infinity, + "learning_rate": 0.0004336620349619005, + "loss": 0.4021, + "step": 20335 + }, + { + "epoch": 1.1362125317764058, + "grad_norm": 0.44208306074142456, + "learning_rate": 0.00043363402061855674, + "loss": 0.5953, + "step": 20336 + }, + { + "epoch": 1.1362684023800877, + "grad_norm": 0.42944639921188354, + "learning_rate": 0.0004336060062752129, + "loss": 0.4029, + "step": 20337 + }, + { + "epoch": 1.1363242729837695, + "grad_norm": 0.4440453052520752, + "learning_rate": 0.00043357799193186915, + "loss": 0.5828, + "step": 20338 + }, + { + "epoch": 1.1363801435874514, + "grad_norm": 0.5610814094543457, + "learning_rate": 0.0004335499775885253, + "loss": 0.3977, + "step": 20339 + }, + { + "epoch": 1.1364360141911334, + "grad_norm": 1.2080084085464478, + "learning_rate": 0.00043352196324518156, + "loss": 0.3847, + "step": 20340 + }, + { + "epoch": 1.136491884794815, + "grad_norm": 0.5369206070899963, + "learning_rate": 0.00043349394890183777, + "loss": 0.4059, + "step": 20341 + }, + { + "epoch": 1.136547755398497, + "grad_norm": 0.5673915147781372, + "learning_rate": 0.00043346593455849397, + "loss": 0.5873, + "step": 20342 + }, + { + "epoch": 1.136603626002179, + "grad_norm": 0.4301750957965851, + "learning_rate": 0.0004334379202151502, + "loss": 0.4476, + "step": 20343 + }, + { + "epoch": 1.1366594966058607, + "grad_norm": 0.5145168304443359, + "learning_rate": 0.0004334099058718064, + "loss": 0.365, + "step": 20344 + }, + { + "epoch": 1.1367153672095427, + "grad_norm": 0.45338886976242065, + "learning_rate": 0.0004333818915284626, + "loss": 0.4431, + "step": 20345 + }, + { + "epoch": 1.1367712378132246, + "grad_norm": 0.46534526348114014, + "learning_rate": 0.0004333538771851188, + "loss": 0.3616, + "step": 20346 + }, + { + "epoch": 1.1368271084169064, + "grad_norm": 0.40755024552345276, + "learning_rate": 0.000433325862841775, + "loss": 0.4236, + "step": 20347 + }, + { + "epoch": 1.1368829790205883, + "grad_norm": 0.5130160450935364, + "learning_rate": 0.0004332978484984312, + "loss": 0.3915, + "step": 20348 + }, + { + "epoch": 1.1369388496242703, + "grad_norm": 0.4126884937286377, + "learning_rate": 0.0004332698341550874, + "loss": 0.3506, + "step": 20349 + }, + { + "epoch": 1.136994720227952, + "grad_norm": 0.4282427728176117, + "learning_rate": 0.0004332418198117436, + "loss": 0.3845, + "step": 20350 + }, + { + "epoch": 1.137050590831634, + "grad_norm": 0.6142197251319885, + "learning_rate": 0.0004332138054683998, + "loss": 0.3785, + "step": 20351 + }, + { + "epoch": 1.137106461435316, + "grad_norm": 0.4584487974643707, + "learning_rate": 0.00043318579112505603, + "loss": 0.4553, + "step": 20352 + }, + { + "epoch": 1.1371623320389976, + "grad_norm": 0.40613603591918945, + "learning_rate": 0.00043315777678171224, + "loss": 0.3747, + "step": 20353 + }, + { + "epoch": 1.1372182026426796, + "grad_norm": 0.38534390926361084, + "learning_rate": 0.00043312976243836844, + "loss": 0.4694, + "step": 20354 + }, + { + "epoch": 1.1372740732463615, + "grad_norm": 0.6163746118545532, + "learning_rate": 0.00043310174809502465, + "loss": 0.4982, + "step": 20355 + }, + { + "epoch": 1.1373299438500433, + "grad_norm": 0.9836499094963074, + "learning_rate": 0.00043307373375168085, + "loss": 0.3416, + "step": 20356 + }, + { + "epoch": 1.1373858144537252, + "grad_norm": 0.48339346051216125, + "learning_rate": 0.0004330457194083371, + "loss": 0.3821, + "step": 20357 + }, + { + "epoch": 1.137441685057407, + "grad_norm": 0.46573883295059204, + "learning_rate": 0.00043301770506499326, + "loss": 0.4098, + "step": 20358 + }, + { + "epoch": 1.137497555661089, + "grad_norm": 0.5694423317909241, + "learning_rate": 0.0004329896907216495, + "loss": 0.4828, + "step": 20359 + }, + { + "epoch": 1.1375534262647709, + "grad_norm": 0.8146705627441406, + "learning_rate": 0.0004329616763783057, + "loss": 0.5285, + "step": 20360 + }, + { + "epoch": 1.1376092968684526, + "grad_norm": 0.700524091720581, + "learning_rate": 0.00043293366203496194, + "loss": 0.43, + "step": 20361 + }, + { + "epoch": 1.1376651674721345, + "grad_norm": 1.1555020809173584, + "learning_rate": 0.0004329056476916181, + "loss": 0.4757, + "step": 20362 + }, + { + "epoch": 1.1377210380758165, + "grad_norm": 0.39428094029426575, + "learning_rate": 0.00043287763334827435, + "loss": 0.4883, + "step": 20363 + }, + { + "epoch": 1.1377769086794982, + "grad_norm": 1.0125209093093872, + "learning_rate": 0.00043284961900493055, + "loss": 0.4839, + "step": 20364 + }, + { + "epoch": 1.1378327792831802, + "grad_norm": 0.37441831827163696, + "learning_rate": 0.00043282160466158676, + "loss": 0.4407, + "step": 20365 + }, + { + "epoch": 1.1378886498868621, + "grad_norm": 0.47823554277420044, + "learning_rate": 0.00043279359031824297, + "loss": 0.453, + "step": 20366 + }, + { + "epoch": 1.1379445204905438, + "grad_norm": 0.3534151613712311, + "learning_rate": 0.00043276557597489917, + "loss": 0.4044, + "step": 20367 + }, + { + "epoch": 1.1380003910942258, + "grad_norm": 0.4590812027454376, + "learning_rate": 0.0004327375616315554, + "loss": 0.4723, + "step": 20368 + }, + { + "epoch": 1.1380562616979075, + "grad_norm": 0.9099534153938293, + "learning_rate": 0.00043270954728821153, + "loss": 0.3393, + "step": 20369 + }, + { + "epoch": 1.1381121323015895, + "grad_norm": 0.47644779086112976, + "learning_rate": 0.0004326815329448678, + "loss": 0.4889, + "step": 20370 + }, + { + "epoch": 1.1381680029052714, + "grad_norm": 0.781937837600708, + "learning_rate": 0.00043265351860152394, + "loss": 0.5463, + "step": 20371 + }, + { + "epoch": 1.1382238735089532, + "grad_norm": 0.2793029248714447, + "learning_rate": 0.0004326255042581802, + "loss": 0.3338, + "step": 20372 + }, + { + "epoch": 1.138279744112635, + "grad_norm": 1.310092568397522, + "learning_rate": 0.0004325974899148364, + "loss": 0.6548, + "step": 20373 + }, + { + "epoch": 1.138335614716317, + "grad_norm": 0.5403621196746826, + "learning_rate": 0.0004325694755714926, + "loss": 0.4843, + "step": 20374 + }, + { + "epoch": 1.1383914853199988, + "grad_norm": 0.6605207324028015, + "learning_rate": 0.0004325414612281488, + "loss": 0.3544, + "step": 20375 + }, + { + "epoch": 1.1384473559236807, + "grad_norm": 0.47628745436668396, + "learning_rate": 0.000432513446884805, + "loss": 0.3499, + "step": 20376 + }, + { + "epoch": 1.1385032265273627, + "grad_norm": 0.6930079460144043, + "learning_rate": 0.00043248543254146123, + "loss": 0.3758, + "step": 20377 + }, + { + "epoch": 1.1385590971310444, + "grad_norm": 0.9409862756729126, + "learning_rate": 0.00043245741819811744, + "loss": 0.4612, + "step": 20378 + }, + { + "epoch": 1.1386149677347264, + "grad_norm": 0.4799898564815521, + "learning_rate": 0.00043242940385477364, + "loss": 0.4289, + "step": 20379 + }, + { + "epoch": 1.1386708383384083, + "grad_norm": 0.7940284609794617, + "learning_rate": 0.0004324013895114299, + "loss": 0.5422, + "step": 20380 + }, + { + "epoch": 1.13872670894209, + "grad_norm": 2.949397325515747, + "learning_rate": 0.00043237337516808605, + "loss": 0.4504, + "step": 20381 + }, + { + "epoch": 1.138782579545772, + "grad_norm": 0.5605247616767883, + "learning_rate": 0.0004323453608247423, + "loss": 0.3946, + "step": 20382 + }, + { + "epoch": 1.138838450149454, + "grad_norm": 0.4171755015850067, + "learning_rate": 0.00043231734648139846, + "loss": 0.3486, + "step": 20383 + }, + { + "epoch": 1.1388943207531357, + "grad_norm": 0.5275369882583618, + "learning_rate": 0.0004322893321380547, + "loss": 0.4188, + "step": 20384 + }, + { + "epoch": 1.1389501913568176, + "grad_norm": 0.6801758408546448, + "learning_rate": 0.0004322613177947109, + "loss": 0.4414, + "step": 20385 + }, + { + "epoch": 1.1390060619604996, + "grad_norm": 1.5852351188659668, + "learning_rate": 0.00043223330345136714, + "loss": 0.4871, + "step": 20386 + }, + { + "epoch": 1.1390619325641813, + "grad_norm": 1.1836655139923096, + "learning_rate": 0.0004322052891080233, + "loss": 0.5012, + "step": 20387 + }, + { + "epoch": 1.1391178031678633, + "grad_norm": 0.6547019481658936, + "learning_rate": 0.00043217727476467955, + "loss": 0.423, + "step": 20388 + }, + { + "epoch": 1.1391736737715452, + "grad_norm": 0.5539771914482117, + "learning_rate": 0.00043214926042133575, + "loss": 0.4998, + "step": 20389 + }, + { + "epoch": 1.139229544375227, + "grad_norm": 0.47164416313171387, + "learning_rate": 0.0004321212460779919, + "loss": 0.4179, + "step": 20390 + }, + { + "epoch": 1.139285414978909, + "grad_norm": 0.3972627818584442, + "learning_rate": 0.00043209323173464817, + "loss": 0.4723, + "step": 20391 + }, + { + "epoch": 1.1393412855825906, + "grad_norm": 2.376716375350952, + "learning_rate": 0.0004320652173913043, + "loss": 0.421, + "step": 20392 + }, + { + "epoch": 1.1393971561862726, + "grad_norm": 0.4211837947368622, + "learning_rate": 0.0004320372030479606, + "loss": 0.4445, + "step": 20393 + }, + { + "epoch": 1.1394530267899545, + "grad_norm": 0.6440092325210571, + "learning_rate": 0.00043200918870461673, + "loss": 0.429, + "step": 20394 + }, + { + "epoch": 1.1395088973936363, + "grad_norm": 2.081003189086914, + "learning_rate": 0.000431981174361273, + "loss": 0.4044, + "step": 20395 + }, + { + "epoch": 1.1395647679973182, + "grad_norm": 1.258732557296753, + "learning_rate": 0.0004319531600179292, + "loss": 0.4131, + "step": 20396 + }, + { + "epoch": 1.1396206386010002, + "grad_norm": 0.46998581290245056, + "learning_rate": 0.0004319251456745854, + "loss": 0.3728, + "step": 20397 + }, + { + "epoch": 1.1396765092046819, + "grad_norm": 0.5122700929641724, + "learning_rate": 0.0004318971313312416, + "loss": 0.4701, + "step": 20398 + }, + { + "epoch": 1.1397323798083638, + "grad_norm": 0.500299870967865, + "learning_rate": 0.0004318691169878978, + "loss": 0.4823, + "step": 20399 + }, + { + "epoch": 1.1397882504120458, + "grad_norm": 0.3489965498447418, + "learning_rate": 0.000431841102644554, + "loss": 0.4316, + "step": 20400 + }, + { + "epoch": 1.1398441210157275, + "grad_norm": 0.8636832237243652, + "learning_rate": 0.0004318130883012102, + "loss": 0.4077, + "step": 20401 + }, + { + "epoch": 1.1398999916194095, + "grad_norm": 0.32977449893951416, + "learning_rate": 0.00043178507395786643, + "loss": 0.3312, + "step": 20402 + }, + { + "epoch": 1.1399558622230912, + "grad_norm": 0.39501407742500305, + "learning_rate": 0.0004317570596145227, + "loss": 0.3438, + "step": 20403 + }, + { + "epoch": 1.1400117328267731, + "grad_norm": 0.9587835073471069, + "learning_rate": 0.00043172904527117884, + "loss": 0.4355, + "step": 20404 + }, + { + "epoch": 1.140067603430455, + "grad_norm": 0.5135812163352966, + "learning_rate": 0.0004317010309278351, + "loss": 0.4794, + "step": 20405 + }, + { + "epoch": 1.1401234740341368, + "grad_norm": 0.5185942053794861, + "learning_rate": 0.00043167301658449125, + "loss": 0.3992, + "step": 20406 + }, + { + "epoch": 1.1401793446378188, + "grad_norm": 0.359264612197876, + "learning_rate": 0.0004316450022411475, + "loss": 0.4196, + "step": 20407 + }, + { + "epoch": 1.1402352152415007, + "grad_norm": 0.4672488570213318, + "learning_rate": 0.00043161698789780366, + "loss": 0.3315, + "step": 20408 + }, + { + "epoch": 1.1402910858451825, + "grad_norm": 0.4901663064956665, + "learning_rate": 0.0004315889735544599, + "loss": 0.3482, + "step": 20409 + }, + { + "epoch": 1.1403469564488644, + "grad_norm": 0.4467635154724121, + "learning_rate": 0.0004315609592111161, + "loss": 0.3662, + "step": 20410 + }, + { + "epoch": 1.1404028270525464, + "grad_norm": 0.5868239998817444, + "learning_rate": 0.0004315329448677723, + "loss": 0.4553, + "step": 20411 + }, + { + "epoch": 1.140458697656228, + "grad_norm": 0.7057517170906067, + "learning_rate": 0.00043150493052442854, + "loss": 0.4576, + "step": 20412 + }, + { + "epoch": 1.14051456825991, + "grad_norm": 0.46243739128112793, + "learning_rate": 0.0004314769161810847, + "loss": 0.3787, + "step": 20413 + }, + { + "epoch": 1.140570438863592, + "grad_norm": 0.44251713156700134, + "learning_rate": 0.00043144890183774095, + "loss": 0.5557, + "step": 20414 + }, + { + "epoch": 1.1406263094672737, + "grad_norm": 0.609992504119873, + "learning_rate": 0.0004314208874943971, + "loss": 0.3747, + "step": 20415 + }, + { + "epoch": 1.1406821800709557, + "grad_norm": 0.5317761898040771, + "learning_rate": 0.00043139287315105336, + "loss": 0.4641, + "step": 20416 + }, + { + "epoch": 1.1407380506746376, + "grad_norm": 0.5691670179367065, + "learning_rate": 0.0004313648588077095, + "loss": 0.4428, + "step": 20417 + }, + { + "epoch": 1.1407939212783194, + "grad_norm": 2.36458158493042, + "learning_rate": 0.0004313368444643658, + "loss": 0.5505, + "step": 20418 + }, + { + "epoch": 1.1408497918820013, + "grad_norm": 0.48227253556251526, + "learning_rate": 0.000431308830121022, + "loss": 0.4062, + "step": 20419 + }, + { + "epoch": 1.1409056624856833, + "grad_norm": 0.3949969410896301, + "learning_rate": 0.0004312808157776782, + "loss": 0.3787, + "step": 20420 + }, + { + "epoch": 1.140961533089365, + "grad_norm": 0.8941529989242554, + "learning_rate": 0.0004312528014343344, + "loss": 0.465, + "step": 20421 + }, + { + "epoch": 1.141017403693047, + "grad_norm": 2.4682776927948, + "learning_rate": 0.0004312247870909906, + "loss": 0.4618, + "step": 20422 + }, + { + "epoch": 1.141073274296729, + "grad_norm": 0.38751164078712463, + "learning_rate": 0.0004311967727476468, + "loss": 0.4336, + "step": 20423 + }, + { + "epoch": 1.1411291449004106, + "grad_norm": 0.5992921590805054, + "learning_rate": 0.000431168758404303, + "loss": 0.3287, + "step": 20424 + }, + { + "epoch": 1.1411850155040926, + "grad_norm": 0.40413007140159607, + "learning_rate": 0.0004311407440609592, + "loss": 0.4747, + "step": 20425 + }, + { + "epoch": 1.1412408861077743, + "grad_norm": 0.43655145168304443, + "learning_rate": 0.0004311127297176154, + "loss": 0.4033, + "step": 20426 + }, + { + "epoch": 1.1412967567114563, + "grad_norm": 2.0601868629455566, + "learning_rate": 0.00043108471537427163, + "loss": 0.4975, + "step": 20427 + }, + { + "epoch": 1.1413526273151382, + "grad_norm": 0.6562598347663879, + "learning_rate": 0.0004310567010309279, + "loss": 0.4979, + "step": 20428 + }, + { + "epoch": 1.14140849791882, + "grad_norm": 0.5567091107368469, + "learning_rate": 0.00043102868668758404, + "loss": 0.434, + "step": 20429 + }, + { + "epoch": 1.1414643685225019, + "grad_norm": 2.2100353240966797, + "learning_rate": 0.0004310006723442403, + "loss": 0.5148, + "step": 20430 + }, + { + "epoch": 1.1415202391261838, + "grad_norm": 0.32187768816947937, + "learning_rate": 0.00043097265800089645, + "loss": 0.368, + "step": 20431 + }, + { + "epoch": 1.1415761097298656, + "grad_norm": 0.343256413936615, + "learning_rate": 0.00043094464365755266, + "loss": 0.3515, + "step": 20432 + }, + { + "epoch": 1.1416319803335475, + "grad_norm": 0.4573245346546173, + "learning_rate": 0.00043091662931420886, + "loss": 0.375, + "step": 20433 + }, + { + "epoch": 1.1416878509372295, + "grad_norm": 0.7324321866035461, + "learning_rate": 0.00043088861497086507, + "loss": 0.5987, + "step": 20434 + }, + { + "epoch": 1.1417437215409112, + "grad_norm": 0.47522982954978943, + "learning_rate": 0.00043086060062752133, + "loss": 0.4111, + "step": 20435 + }, + { + "epoch": 1.1417995921445931, + "grad_norm": 0.6223275065422058, + "learning_rate": 0.0004308325862841775, + "loss": 0.5809, + "step": 20436 + }, + { + "epoch": 1.1418554627482749, + "grad_norm": 0.5661481022834778, + "learning_rate": 0.00043080457194083374, + "loss": 0.5252, + "step": 20437 + }, + { + "epoch": 1.1419113333519568, + "grad_norm": 1.0930447578430176, + "learning_rate": 0.0004307765575974899, + "loss": 0.4597, + "step": 20438 + }, + { + "epoch": 1.1419672039556388, + "grad_norm": 0.7637152075767517, + "learning_rate": 0.00043074854325414615, + "loss": 0.5016, + "step": 20439 + }, + { + "epoch": 1.1420230745593205, + "grad_norm": 0.5007243156433105, + "learning_rate": 0.0004307205289108023, + "loss": 0.4903, + "step": 20440 + }, + { + "epoch": 1.1420789451630025, + "grad_norm": 1.7573068141937256, + "learning_rate": 0.00043069251456745856, + "loss": 0.4455, + "step": 20441 + }, + { + "epoch": 1.1421348157666844, + "grad_norm": 0.5969480276107788, + "learning_rate": 0.0004306645002241147, + "loss": 0.4382, + "step": 20442 + }, + { + "epoch": 1.1421906863703661, + "grad_norm": 0.8849652409553528, + "learning_rate": 0.000430636485880771, + "loss": 0.4974, + "step": 20443 + }, + { + "epoch": 1.142246556974048, + "grad_norm": 0.45523157715797424, + "learning_rate": 0.0004306084715374272, + "loss": 0.4909, + "step": 20444 + }, + { + "epoch": 1.14230242757773, + "grad_norm": 0.45901358127593994, + "learning_rate": 0.0004305804571940834, + "loss": 0.4114, + "step": 20445 + }, + { + "epoch": 1.1423582981814118, + "grad_norm": 0.5203260779380798, + "learning_rate": 0.0004305524428507396, + "loss": 0.3839, + "step": 20446 + }, + { + "epoch": 1.1424141687850937, + "grad_norm": 2.2447052001953125, + "learning_rate": 0.0004305244285073958, + "loss": 0.5165, + "step": 20447 + }, + { + "epoch": 1.1424700393887757, + "grad_norm": 0.4821569323539734, + "learning_rate": 0.000430496414164052, + "loss": 0.3908, + "step": 20448 + }, + { + "epoch": 1.1425259099924574, + "grad_norm": 0.4378868043422699, + "learning_rate": 0.0004304683998207082, + "loss": 0.4717, + "step": 20449 + }, + { + "epoch": 1.1425817805961394, + "grad_norm": 0.39448973536491394, + "learning_rate": 0.0004304403854773644, + "loss": 0.5173, + "step": 20450 + }, + { + "epoch": 1.1426376511998213, + "grad_norm": 0.8526505827903748, + "learning_rate": 0.0004304123711340207, + "loss": 0.3725, + "step": 20451 + }, + { + "epoch": 1.142693521803503, + "grad_norm": 0.9043100476264954, + "learning_rate": 0.00043038435679067683, + "loss": 0.4125, + "step": 20452 + }, + { + "epoch": 1.142749392407185, + "grad_norm": 1.1458415985107422, + "learning_rate": 0.00043035634244733303, + "loss": 0.3365, + "step": 20453 + }, + { + "epoch": 1.142805263010867, + "grad_norm": 0.4513207674026489, + "learning_rate": 0.00043032832810398924, + "loss": 0.3524, + "step": 20454 + }, + { + "epoch": 1.1428611336145487, + "grad_norm": 0.6103252172470093, + "learning_rate": 0.00043030031376064545, + "loss": 0.4813, + "step": 20455 + }, + { + "epoch": 1.1429170042182306, + "grad_norm": 0.7696529626846313, + "learning_rate": 0.00043027229941730165, + "loss": 0.4407, + "step": 20456 + }, + { + "epoch": 1.1429728748219126, + "grad_norm": 0.4778670370578766, + "learning_rate": 0.00043024428507395786, + "loss": 0.4597, + "step": 20457 + }, + { + "epoch": 1.1430287454255943, + "grad_norm": 0.43743181228637695, + "learning_rate": 0.00043021627073061406, + "loss": 0.375, + "step": 20458 + }, + { + "epoch": 1.1430846160292762, + "grad_norm": 0.4759598672389984, + "learning_rate": 0.00043018825638727027, + "loss": 0.5053, + "step": 20459 + }, + { + "epoch": 1.143140486632958, + "grad_norm": 2.241217851638794, + "learning_rate": 0.00043016024204392653, + "loss": 0.4647, + "step": 20460 + }, + { + "epoch": 1.14319635723664, + "grad_norm": 0.4209653437137604, + "learning_rate": 0.0004301322277005827, + "loss": 0.4466, + "step": 20461 + }, + { + "epoch": 1.1432522278403219, + "grad_norm": 0.34176701307296753, + "learning_rate": 0.00043010421335723894, + "loss": 0.423, + "step": 20462 + }, + { + "epoch": 1.1433080984440036, + "grad_norm": 0.7379328608512878, + "learning_rate": 0.0004300761990138951, + "loss": 0.4418, + "step": 20463 + }, + { + "epoch": 1.1433639690476856, + "grad_norm": 0.373388409614563, + "learning_rate": 0.00043004818467055135, + "loss": 0.3564, + "step": 20464 + }, + { + "epoch": 1.1434198396513675, + "grad_norm": 0.5008513927459717, + "learning_rate": 0.0004300201703272075, + "loss": 0.3748, + "step": 20465 + }, + { + "epoch": 1.1434757102550492, + "grad_norm": 0.755798876285553, + "learning_rate": 0.00042999215598386376, + "loss": 0.5035, + "step": 20466 + }, + { + "epoch": 1.1435315808587312, + "grad_norm": 0.6020072102546692, + "learning_rate": 0.00042996414164051997, + "loss": 0.392, + "step": 20467 + }, + { + "epoch": 1.1435874514624131, + "grad_norm": 0.39553967118263245, + "learning_rate": 0.0004299361272971762, + "loss": 0.3464, + "step": 20468 + }, + { + "epoch": 1.1436433220660949, + "grad_norm": 0.7238013744354248, + "learning_rate": 0.0004299081129538324, + "loss": 0.4764, + "step": 20469 + }, + { + "epoch": 1.1436991926697768, + "grad_norm": 0.49231207370758057, + "learning_rate": 0.0004298800986104886, + "loss": 0.3932, + "step": 20470 + }, + { + "epoch": 1.1437550632734585, + "grad_norm": 1.968328833580017, + "learning_rate": 0.0004298520842671448, + "loss": 0.4002, + "step": 20471 + }, + { + "epoch": 1.1438109338771405, + "grad_norm": 1.3144546747207642, + "learning_rate": 0.000429824069923801, + "loss": 0.4897, + "step": 20472 + }, + { + "epoch": 1.1438668044808225, + "grad_norm": 0.8512692451477051, + "learning_rate": 0.0004297960555804572, + "loss": 0.4258, + "step": 20473 + }, + { + "epoch": 1.1439226750845042, + "grad_norm": 0.48370498418807983, + "learning_rate": 0.00042976804123711346, + "loss": 0.3236, + "step": 20474 + }, + { + "epoch": 1.1439785456881861, + "grad_norm": 0.5749701857566833, + "learning_rate": 0.0004297400268937696, + "loss": 0.3061, + "step": 20475 + }, + { + "epoch": 1.144034416291868, + "grad_norm": 0.8240038752555847, + "learning_rate": 0.0004297120125504258, + "loss": 0.4956, + "step": 20476 + }, + { + "epoch": 1.1440902868955498, + "grad_norm": 0.5337473154067993, + "learning_rate": 0.00042968399820708203, + "loss": 0.4858, + "step": 20477 + }, + { + "epoch": 1.1441461574992318, + "grad_norm": 0.4157274067401886, + "learning_rate": 0.00042965598386373823, + "loss": 0.3995, + "step": 20478 + }, + { + "epoch": 1.1442020281029137, + "grad_norm": 0.5040929913520813, + "learning_rate": 0.00042962796952039444, + "loss": 0.4774, + "step": 20479 + }, + { + "epoch": 1.1442578987065954, + "grad_norm": 0.944706916809082, + "learning_rate": 0.00042959995517705065, + "loss": 0.4668, + "step": 20480 + }, + { + "epoch": 1.1443137693102774, + "grad_norm": 0.49325016140937805, + "learning_rate": 0.00042957194083370685, + "loss": 0.5081, + "step": 20481 + }, + { + "epoch": 1.1443696399139593, + "grad_norm": 2.5957415103912354, + "learning_rate": 0.00042954392649036306, + "loss": 0.4026, + "step": 20482 + }, + { + "epoch": 1.144425510517641, + "grad_norm": 0.38936495780944824, + "learning_rate": 0.0004295159121470193, + "loss": 0.3368, + "step": 20483 + }, + { + "epoch": 1.144481381121323, + "grad_norm": 0.475053608417511, + "learning_rate": 0.00042948789780367547, + "loss": 0.4478, + "step": 20484 + }, + { + "epoch": 1.144537251725005, + "grad_norm": 0.6938390135765076, + "learning_rate": 0.00042945988346033173, + "loss": 0.5389, + "step": 20485 + }, + { + "epoch": 1.1445931223286867, + "grad_norm": 0.4481906592845917, + "learning_rate": 0.0004294318691169879, + "loss": 0.4537, + "step": 20486 + }, + { + "epoch": 1.1446489929323687, + "grad_norm": 0.6446820497512817, + "learning_rate": 0.00042940385477364414, + "loss": 0.4954, + "step": 20487 + }, + { + "epoch": 1.1447048635360506, + "grad_norm": 1.0902436971664429, + "learning_rate": 0.0004293758404303003, + "loss": 0.5892, + "step": 20488 + }, + { + "epoch": 1.1447607341397323, + "grad_norm": 0.78903728723526, + "learning_rate": 0.00042934782608695655, + "loss": 0.4804, + "step": 20489 + }, + { + "epoch": 1.1448166047434143, + "grad_norm": 0.5603873133659363, + "learning_rate": 0.00042931981174361276, + "loss": 0.4226, + "step": 20490 + }, + { + "epoch": 1.1448724753470962, + "grad_norm": 0.6780089735984802, + "learning_rate": 0.00042929179740026896, + "loss": 0.4179, + "step": 20491 + }, + { + "epoch": 1.144928345950778, + "grad_norm": 5.301609516143799, + "learning_rate": 0.00042926378305692517, + "loss": 0.4379, + "step": 20492 + }, + { + "epoch": 1.14498421655446, + "grad_norm": 1.843172550201416, + "learning_rate": 0.0004292357687135814, + "loss": 0.3705, + "step": 20493 + }, + { + "epoch": 1.1450400871581417, + "grad_norm": 1.6984986066818237, + "learning_rate": 0.0004292077543702376, + "loss": 0.4894, + "step": 20494 + }, + { + "epoch": 1.1450959577618236, + "grad_norm": 0.42500609159469604, + "learning_rate": 0.0004291797400268938, + "loss": 0.462, + "step": 20495 + }, + { + "epoch": 1.1451518283655056, + "grad_norm": 0.5046170949935913, + "learning_rate": 0.00042915172568355, + "loss": 0.3666, + "step": 20496 + }, + { + "epoch": 1.1452076989691873, + "grad_norm": 0.38249441981315613, + "learning_rate": 0.00042912371134020614, + "loss": 0.4446, + "step": 20497 + }, + { + "epoch": 1.1452635695728692, + "grad_norm": 0.40185171365737915, + "learning_rate": 0.0004290956969968624, + "loss": 0.5335, + "step": 20498 + }, + { + "epoch": 1.1453194401765512, + "grad_norm": 0.5250332951545715, + "learning_rate": 0.0004290676826535186, + "loss": 0.4642, + "step": 20499 + }, + { + "epoch": 1.145375310780233, + "grad_norm": 0.4024147689342499, + "learning_rate": 0.0004290396683101748, + "loss": 0.4212, + "step": 20500 + }, + { + "epoch": 1.145375310780233, + "eval_cer": 0.08730210685956816, + "eval_loss": 0.32656076550483704, + "eval_runtime": 56.1079, + "eval_samples_per_second": 80.88, + "eval_steps_per_second": 5.062, + "eval_wer": 0.34487863467500907, + "step": 20500 + }, + { + "epoch": 1.1454311813839149, + "grad_norm": 0.6354258060455322, + "learning_rate": 0.000429011653966831, + "loss": 0.5448, + "step": 20501 + }, + { + "epoch": 1.1454870519875968, + "grad_norm": 0.8274503350257874, + "learning_rate": 0.00042898363962348723, + "loss": 0.4167, + "step": 20502 + }, + { + "epoch": 1.1455429225912785, + "grad_norm": 0.6744612455368042, + "learning_rate": 0.00042895562528014343, + "loss": 0.385, + "step": 20503 + }, + { + "epoch": 1.1455987931949605, + "grad_norm": 0.38898006081581116, + "learning_rate": 0.00042892761093679964, + "loss": 0.395, + "step": 20504 + }, + { + "epoch": 1.1456546637986422, + "grad_norm": 1.7351113557815552, + "learning_rate": 0.00042889959659345584, + "loss": 0.3112, + "step": 20505 + }, + { + "epoch": 1.1457105344023242, + "grad_norm": 0.48407119512557983, + "learning_rate": 0.0004288715822501121, + "loss": 0.39, + "step": 20506 + }, + { + "epoch": 1.1457664050060061, + "grad_norm": 0.6022807955741882, + "learning_rate": 0.00042884356790676826, + "loss": 0.4006, + "step": 20507 + }, + { + "epoch": 1.1458222756096879, + "grad_norm": 0.8687308430671692, + "learning_rate": 0.0004288155535634245, + "loss": 0.4968, + "step": 20508 + }, + { + "epoch": 1.1458781462133698, + "grad_norm": 0.4185049533843994, + "learning_rate": 0.00042878753922008067, + "loss": 0.4519, + "step": 20509 + }, + { + "epoch": 1.1459340168170518, + "grad_norm": 0.7282737493515015, + "learning_rate": 0.00042875952487673693, + "loss": 0.3652, + "step": 20510 + }, + { + "epoch": 1.1459898874207335, + "grad_norm": 2.708588123321533, + "learning_rate": 0.0004287315105333931, + "loss": 0.41, + "step": 20511 + }, + { + "epoch": 1.1460457580244154, + "grad_norm": 0.7072773575782776, + "learning_rate": 0.00042870349619004934, + "loss": 0.3954, + "step": 20512 + }, + { + "epoch": 1.1461016286280974, + "grad_norm": 0.4102955758571625, + "learning_rate": 0.0004286754818467055, + "loss": 0.4906, + "step": 20513 + }, + { + "epoch": 1.1461574992317791, + "grad_norm": 0.4554625153541565, + "learning_rate": 0.00042864746750336175, + "loss": 0.4493, + "step": 20514 + }, + { + "epoch": 1.146213369835461, + "grad_norm": 2.132985830307007, + "learning_rate": 0.00042861945316001796, + "loss": 0.3576, + "step": 20515 + }, + { + "epoch": 1.146269240439143, + "grad_norm": 0.45237353444099426, + "learning_rate": 0.00042859143881667416, + "loss": 0.5481, + "step": 20516 + }, + { + "epoch": 1.1463251110428248, + "grad_norm": 0.5661503672599792, + "learning_rate": 0.00042856342447333037, + "loss": 0.4333, + "step": 20517 + }, + { + "epoch": 1.1463809816465067, + "grad_norm": 0.6642976403236389, + "learning_rate": 0.0004285354101299865, + "loss": 0.4458, + "step": 20518 + }, + { + "epoch": 1.1464368522501887, + "grad_norm": 1.0390591621398926, + "learning_rate": 0.0004285073957866428, + "loss": 0.4528, + "step": 20519 + }, + { + "epoch": 1.1464927228538704, + "grad_norm": 0.5346958041191101, + "learning_rate": 0.00042847938144329893, + "loss": 0.5211, + "step": 20520 + }, + { + "epoch": 1.1465485934575523, + "grad_norm": 0.5834360122680664, + "learning_rate": 0.0004284513670999552, + "loss": 0.4908, + "step": 20521 + }, + { + "epoch": 1.1466044640612343, + "grad_norm": 0.3788934051990509, + "learning_rate": 0.0004284233527566114, + "loss": 0.418, + "step": 20522 + }, + { + "epoch": 1.146660334664916, + "grad_norm": 0.4654427766799927, + "learning_rate": 0.0004283953384132676, + "loss": 0.3929, + "step": 20523 + }, + { + "epoch": 1.146716205268598, + "grad_norm": 9.58620548248291, + "learning_rate": 0.0004283673240699238, + "loss": 0.3268, + "step": 20524 + }, + { + "epoch": 1.14677207587228, + "grad_norm": 1.1155493259429932, + "learning_rate": 0.00042833930972658, + "loss": 0.378, + "step": 20525 + }, + { + "epoch": 1.1468279464759616, + "grad_norm": 0.9072237610816956, + "learning_rate": 0.0004283112953832362, + "loss": 0.4294, + "step": 20526 + }, + { + "epoch": 1.1468838170796436, + "grad_norm": 0.7875970005989075, + "learning_rate": 0.0004282832810398924, + "loss": 0.415, + "step": 20527 + }, + { + "epoch": 1.1469396876833253, + "grad_norm": 0.334419846534729, + "learning_rate": 0.00042825526669654863, + "loss": 0.3636, + "step": 20528 + }, + { + "epoch": 1.1469955582870073, + "grad_norm": 0.5470633506774902, + "learning_rate": 0.0004282272523532049, + "loss": 0.5432, + "step": 20529 + }, + { + "epoch": 1.1470514288906892, + "grad_norm": 0.43276447057724, + "learning_rate": 0.00042819923800986104, + "loss": 0.4037, + "step": 20530 + }, + { + "epoch": 1.147107299494371, + "grad_norm": 3.1238811016082764, + "learning_rate": 0.0004281712236665173, + "loss": 0.4868, + "step": 20531 + }, + { + "epoch": 1.147163170098053, + "grad_norm": 0.5478940010070801, + "learning_rate": 0.00042814320932317346, + "loss": 0.3193, + "step": 20532 + }, + { + "epoch": 1.1472190407017349, + "grad_norm": 0.4722549021244049, + "learning_rate": 0.0004281151949798297, + "loss": 0.4197, + "step": 20533 + }, + { + "epoch": 1.1472749113054166, + "grad_norm": 0.8621640205383301, + "learning_rate": 0.00042808718063648587, + "loss": 0.3979, + "step": 20534 + }, + { + "epoch": 1.1473307819090985, + "grad_norm": 0.4819306433200836, + "learning_rate": 0.00042805916629314213, + "loss": 0.4401, + "step": 20535 + }, + { + "epoch": 1.1473866525127805, + "grad_norm": 2.9893667697906494, + "learning_rate": 0.0004280311519497983, + "loss": 0.4017, + "step": 20536 + }, + { + "epoch": 1.1474425231164622, + "grad_norm": 0.3493970036506653, + "learning_rate": 0.00042800313760645454, + "loss": 0.3364, + "step": 20537 + }, + { + "epoch": 1.1474983937201442, + "grad_norm": 0.661090612411499, + "learning_rate": 0.00042797512326311075, + "loss": 0.5903, + "step": 20538 + }, + { + "epoch": 1.147554264323826, + "grad_norm": 0.6855655312538147, + "learning_rate": 0.0004279471089197669, + "loss": 0.4426, + "step": 20539 + }, + { + "epoch": 1.1476101349275079, + "grad_norm": 0.3625918924808502, + "learning_rate": 0.00042791909457642316, + "loss": 0.5181, + "step": 20540 + }, + { + "epoch": 1.1476660055311898, + "grad_norm": 0.4531007707118988, + "learning_rate": 0.0004278910802330793, + "loss": 0.4188, + "step": 20541 + }, + { + "epoch": 1.1477218761348715, + "grad_norm": 0.5085087418556213, + "learning_rate": 0.00042786306588973557, + "loss": 0.4654, + "step": 20542 + }, + { + "epoch": 1.1477777467385535, + "grad_norm": 0.35702717304229736, + "learning_rate": 0.0004278350515463917, + "loss": 0.3919, + "step": 20543 + }, + { + "epoch": 1.1478336173422354, + "grad_norm": 0.65211421251297, + "learning_rate": 0.000427807037203048, + "loss": 0.4131, + "step": 20544 + }, + { + "epoch": 1.1478894879459172, + "grad_norm": 10.794947624206543, + "learning_rate": 0.0004277790228597042, + "loss": 0.4584, + "step": 20545 + }, + { + "epoch": 1.1479453585495991, + "grad_norm": 0.5607660412788391, + "learning_rate": 0.0004277510085163604, + "loss": 0.488, + "step": 20546 + }, + { + "epoch": 1.148001229153281, + "grad_norm": 0.4878135323524475, + "learning_rate": 0.0004277229941730166, + "loss": 0.414, + "step": 20547 + }, + { + "epoch": 1.1480570997569628, + "grad_norm": 0.3859521448612213, + "learning_rate": 0.0004276949798296728, + "loss": 0.4478, + "step": 20548 + }, + { + "epoch": 1.1481129703606447, + "grad_norm": 1.2694286108016968, + "learning_rate": 0.000427666965486329, + "loss": 0.4648, + "step": 20549 + }, + { + "epoch": 1.1481688409643267, + "grad_norm": 0.5278862118721008, + "learning_rate": 0.0004276389511429852, + "loss": 0.4528, + "step": 20550 + }, + { + "epoch": 1.1482247115680084, + "grad_norm": 0.6556491255760193, + "learning_rate": 0.0004276109367996414, + "loss": 0.6156, + "step": 20551 + }, + { + "epoch": 1.1482805821716904, + "grad_norm": 0.4723840355873108, + "learning_rate": 0.0004275829224562976, + "loss": 0.3502, + "step": 20552 + }, + { + "epoch": 1.1483364527753723, + "grad_norm": 0.4646940231323242, + "learning_rate": 0.00042755490811295383, + "loss": 0.4022, + "step": 20553 + }, + { + "epoch": 1.148392323379054, + "grad_norm": 0.5677382349967957, + "learning_rate": 0.0004275268937696101, + "loss": 0.4399, + "step": 20554 + }, + { + "epoch": 1.148448193982736, + "grad_norm": 0.47657275199890137, + "learning_rate": 0.00042749887942626624, + "loss": 0.3253, + "step": 20555 + }, + { + "epoch": 1.148504064586418, + "grad_norm": 0.3647094964981079, + "learning_rate": 0.0004274708650829225, + "loss": 0.4449, + "step": 20556 + }, + { + "epoch": 1.1485599351900997, + "grad_norm": 1.1287418603897095, + "learning_rate": 0.00042744285073957866, + "loss": 0.3082, + "step": 20557 + }, + { + "epoch": 1.1486158057937816, + "grad_norm": 0.472523033618927, + "learning_rate": 0.0004274148363962349, + "loss": 0.4861, + "step": 20558 + }, + { + "epoch": 1.1486716763974636, + "grad_norm": 0.36564135551452637, + "learning_rate": 0.00042738682205289107, + "loss": 0.414, + "step": 20559 + }, + { + "epoch": 1.1487275470011453, + "grad_norm": 0.7337106466293335, + "learning_rate": 0.0004273588077095473, + "loss": 0.4723, + "step": 20560 + }, + { + "epoch": 1.1487834176048273, + "grad_norm": 9.382475852966309, + "learning_rate": 0.00042733079336620353, + "loss": 0.4967, + "step": 20561 + }, + { + "epoch": 1.148839288208509, + "grad_norm": 0.6335256099700928, + "learning_rate": 0.0004273027790228597, + "loss": 0.4135, + "step": 20562 + }, + { + "epoch": 1.148895158812191, + "grad_norm": 0.4596165418624878, + "learning_rate": 0.00042727476467951594, + "loss": 0.3815, + "step": 20563 + }, + { + "epoch": 1.148951029415873, + "grad_norm": 0.47558167576789856, + "learning_rate": 0.0004272467503361721, + "loss": 0.3742, + "step": 20564 + }, + { + "epoch": 1.1490069000195546, + "grad_norm": 0.45829054713249207, + "learning_rate": 0.00042721873599282836, + "loss": 0.4272, + "step": 20565 + }, + { + "epoch": 1.1490627706232366, + "grad_norm": 0.5803989768028259, + "learning_rate": 0.0004271907216494845, + "loss": 0.6161, + "step": 20566 + }, + { + "epoch": 1.1491186412269185, + "grad_norm": 0.4638248383998871, + "learning_rate": 0.00042716270730614077, + "loss": 0.4135, + "step": 20567 + }, + { + "epoch": 1.1491745118306003, + "grad_norm": 0.7335249185562134, + "learning_rate": 0.0004271346929627969, + "loss": 0.4968, + "step": 20568 + }, + { + "epoch": 1.1492303824342822, + "grad_norm": 0.9847994446754456, + "learning_rate": 0.0004271066786194532, + "loss": 0.3534, + "step": 20569 + }, + { + "epoch": 1.1492862530379642, + "grad_norm": 0.7245907783508301, + "learning_rate": 0.0004270786642761094, + "loss": 0.4727, + "step": 20570 + }, + { + "epoch": 1.149342123641646, + "grad_norm": 0.42467159032821655, + "learning_rate": 0.0004270506499327656, + "loss": 0.4494, + "step": 20571 + }, + { + "epoch": 1.1493979942453278, + "grad_norm": 0.4570463001728058, + "learning_rate": 0.0004270226355894218, + "loss": 0.6393, + "step": 20572 + }, + { + "epoch": 1.1494538648490096, + "grad_norm": 0.7190275192260742, + "learning_rate": 0.000426994621246078, + "loss": 0.3223, + "step": 20573 + }, + { + "epoch": 1.1495097354526915, + "grad_norm": 0.40043044090270996, + "learning_rate": 0.0004269666069027342, + "loss": 0.3894, + "step": 20574 + }, + { + "epoch": 1.1495656060563735, + "grad_norm": 0.467826247215271, + "learning_rate": 0.0004269385925593904, + "loss": 0.4104, + "step": 20575 + }, + { + "epoch": 1.1496214766600552, + "grad_norm": 0.4919075667858124, + "learning_rate": 0.0004269105782160466, + "loss": 0.5121, + "step": 20576 + }, + { + "epoch": 1.1496773472637372, + "grad_norm": 0.5776292085647583, + "learning_rate": 0.0004268825638727029, + "loss": 0.3932, + "step": 20577 + }, + { + "epoch": 1.149733217867419, + "grad_norm": 0.7403066754341125, + "learning_rate": 0.00042685454952935903, + "loss": 0.4426, + "step": 20578 + }, + { + "epoch": 1.1497890884711008, + "grad_norm": 0.42079514265060425, + "learning_rate": 0.0004268265351860153, + "loss": 0.396, + "step": 20579 + }, + { + "epoch": 1.1498449590747828, + "grad_norm": 0.5782553553581238, + "learning_rate": 0.00042679852084267144, + "loss": 0.4791, + "step": 20580 + }, + { + "epoch": 1.1499008296784647, + "grad_norm": 0.43431055545806885, + "learning_rate": 0.00042677050649932765, + "loss": 0.5607, + "step": 20581 + }, + { + "epoch": 1.1499567002821465, + "grad_norm": 0.3606863021850586, + "learning_rate": 0.00042674249215598386, + "loss": 0.3359, + "step": 20582 + }, + { + "epoch": 1.1500125708858284, + "grad_norm": 0.427318811416626, + "learning_rate": 0.00042671447781264006, + "loss": 0.5564, + "step": 20583 + }, + { + "epoch": 1.1500684414895104, + "grad_norm": 0.6921379566192627, + "learning_rate": 0.00042668646346929627, + "loss": 0.3743, + "step": 20584 + }, + { + "epoch": 1.150124312093192, + "grad_norm": 0.45985519886016846, + "learning_rate": 0.00042665844912595247, + "loss": 0.4027, + "step": 20585 + }, + { + "epoch": 1.150180182696874, + "grad_norm": 2.219175338745117, + "learning_rate": 0.00042663043478260873, + "loss": 0.4546, + "step": 20586 + }, + { + "epoch": 1.150236053300556, + "grad_norm": 0.8333242535591125, + "learning_rate": 0.0004266024204392649, + "loss": 0.4595, + "step": 20587 + }, + { + "epoch": 1.1502919239042377, + "grad_norm": 0.3999834358692169, + "learning_rate": 0.00042657440609592114, + "loss": 0.4778, + "step": 20588 + }, + { + "epoch": 1.1503477945079197, + "grad_norm": 0.9359585046768188, + "learning_rate": 0.0004265463917525773, + "loss": 0.4081, + "step": 20589 + }, + { + "epoch": 1.1504036651116016, + "grad_norm": 2.444596529006958, + "learning_rate": 0.00042651837740923356, + "loss": 0.4277, + "step": 20590 + }, + { + "epoch": 1.1504595357152834, + "grad_norm": 6.18245267868042, + "learning_rate": 0.0004264903630658897, + "loss": 0.473, + "step": 20591 + }, + { + "epoch": 1.1505154063189653, + "grad_norm": 0.3552677035331726, + "learning_rate": 0.00042646234872254597, + "loss": 0.3291, + "step": 20592 + }, + { + "epoch": 1.1505712769226473, + "grad_norm": 0.3999452590942383, + "learning_rate": 0.0004264343343792022, + "loss": 0.3424, + "step": 20593 + }, + { + "epoch": 1.150627147526329, + "grad_norm": 0.3637092411518097, + "learning_rate": 0.0004264063200358584, + "loss": 0.4377, + "step": 20594 + }, + { + "epoch": 1.150683018130011, + "grad_norm": 0.5734565258026123, + "learning_rate": 0.0004263783056925146, + "loss": 0.4284, + "step": 20595 + }, + { + "epoch": 1.1507388887336927, + "grad_norm": 3.0243678092956543, + "learning_rate": 0.0004263502913491708, + "loss": 0.3917, + "step": 20596 + }, + { + "epoch": 1.1507947593373746, + "grad_norm": 1.3070906400680542, + "learning_rate": 0.000426322277005827, + "loss": 0.4812, + "step": 20597 + }, + { + "epoch": 1.1508506299410566, + "grad_norm": 3.2046782970428467, + "learning_rate": 0.0004262942626624832, + "loss": 0.3023, + "step": 20598 + }, + { + "epoch": 1.1509065005447383, + "grad_norm": 0.5574734807014465, + "learning_rate": 0.0004262662483191394, + "loss": 0.4082, + "step": 20599 + }, + { + "epoch": 1.1509623711484203, + "grad_norm": 0.5327267646789551, + "learning_rate": 0.00042623823397579567, + "loss": 0.5354, + "step": 20600 + }, + { + "epoch": 1.1510182417521022, + "grad_norm": 0.4255421459674835, + "learning_rate": 0.0004262102196324518, + "loss": 0.4241, + "step": 20601 + }, + { + "epoch": 1.151074112355784, + "grad_norm": 0.47421857714653015, + "learning_rate": 0.000426182205289108, + "loss": 0.4635, + "step": 20602 + }, + { + "epoch": 1.151129982959466, + "grad_norm": 0.5547838807106018, + "learning_rate": 0.00042615419094576423, + "loss": 0.4228, + "step": 20603 + }, + { + "epoch": 1.1511858535631478, + "grad_norm": 3.3710014820098877, + "learning_rate": 0.00042612617660242044, + "loss": 0.3884, + "step": 20604 + }, + { + "epoch": 1.1512417241668296, + "grad_norm": 1.0053952932357788, + "learning_rate": 0.00042609816225907664, + "loss": 0.4801, + "step": 20605 + }, + { + "epoch": 1.1512975947705115, + "grad_norm": 0.5057304501533508, + "learning_rate": 0.00042607014791573285, + "loss": 0.4451, + "step": 20606 + }, + { + "epoch": 1.1513534653741933, + "grad_norm": 0.33185556530952454, + "learning_rate": 0.00042604213357238905, + "loss": 0.4265, + "step": 20607 + }, + { + "epoch": 1.1514093359778752, + "grad_norm": 0.5056342482566833, + "learning_rate": 0.00042601411922904526, + "loss": 0.4173, + "step": 20608 + }, + { + "epoch": 1.1514652065815572, + "grad_norm": 0.4987640678882599, + "learning_rate": 0.0004259861048857015, + "loss": 0.3806, + "step": 20609 + }, + { + "epoch": 1.1515210771852389, + "grad_norm": 0.6150398850440979, + "learning_rate": 0.00042595809054235767, + "loss": 0.4745, + "step": 20610 + }, + { + "epoch": 1.1515769477889208, + "grad_norm": 0.36036503314971924, + "learning_rate": 0.00042593007619901393, + "loss": 0.2796, + "step": 20611 + }, + { + "epoch": 1.1516328183926028, + "grad_norm": 0.36705923080444336, + "learning_rate": 0.0004259020618556701, + "loss": 0.4228, + "step": 20612 + }, + { + "epoch": 1.1516886889962845, + "grad_norm": 0.42649534344673157, + "learning_rate": 0.00042587404751232634, + "loss": 0.4745, + "step": 20613 + }, + { + "epoch": 1.1517445595999665, + "grad_norm": 0.4371563792228699, + "learning_rate": 0.0004258460331689825, + "loss": 0.4337, + "step": 20614 + }, + { + "epoch": 1.1518004302036484, + "grad_norm": 0.7786169648170471, + "learning_rate": 0.00042581801882563876, + "loss": 0.4729, + "step": 20615 + }, + { + "epoch": 1.1518563008073301, + "grad_norm": 0.6214587688446045, + "learning_rate": 0.00042579000448229496, + "loss": 0.4644, + "step": 20616 + }, + { + "epoch": 1.151912171411012, + "grad_norm": 0.5841861367225647, + "learning_rate": 0.00042576199013895117, + "loss": 0.3771, + "step": 20617 + }, + { + "epoch": 1.151968042014694, + "grad_norm": 0.5064770579338074, + "learning_rate": 0.0004257339757956074, + "loss": 0.362, + "step": 20618 + }, + { + "epoch": 1.1520239126183758, + "grad_norm": 0.3840835392475128, + "learning_rate": 0.0004257059614522636, + "loss": 0.3898, + "step": 20619 + }, + { + "epoch": 1.1520797832220577, + "grad_norm": 0.4683740735054016, + "learning_rate": 0.0004256779471089198, + "loss": 0.4531, + "step": 20620 + }, + { + "epoch": 1.1521356538257397, + "grad_norm": 0.4226134717464447, + "learning_rate": 0.000425649932765576, + "loss": 0.3367, + "step": 20621 + }, + { + "epoch": 1.1521915244294214, + "grad_norm": 2.8405680656433105, + "learning_rate": 0.0004256219184222322, + "loss": 0.4183, + "step": 20622 + }, + { + "epoch": 1.1522473950331034, + "grad_norm": 0.4135917127132416, + "learning_rate": 0.00042559390407888835, + "loss": 0.3695, + "step": 20623 + }, + { + "epoch": 1.1523032656367853, + "grad_norm": 0.47850629687309265, + "learning_rate": 0.0004255658897355446, + "loss": 0.5713, + "step": 20624 + }, + { + "epoch": 1.152359136240467, + "grad_norm": 0.36854448914527893, + "learning_rate": 0.0004255378753922008, + "loss": 0.3549, + "step": 20625 + }, + { + "epoch": 1.152415006844149, + "grad_norm": 0.39779940247535706, + "learning_rate": 0.000425509861048857, + "loss": 0.4107, + "step": 20626 + }, + { + "epoch": 1.152470877447831, + "grad_norm": 0.40212276577949524, + "learning_rate": 0.0004254818467055132, + "loss": 0.355, + "step": 20627 + }, + { + "epoch": 1.1525267480515127, + "grad_norm": 0.5031728148460388, + "learning_rate": 0.00042545383236216943, + "loss": 0.3901, + "step": 20628 + }, + { + "epoch": 1.1525826186551946, + "grad_norm": 2.1520004272460938, + "learning_rate": 0.00042542581801882564, + "loss": 0.4903, + "step": 20629 + }, + { + "epoch": 1.1526384892588764, + "grad_norm": 0.49818217754364014, + "learning_rate": 0.00042539780367548184, + "loss": 0.3584, + "step": 20630 + }, + { + "epoch": 1.1526943598625583, + "grad_norm": 0.5126453638076782, + "learning_rate": 0.00042536978933213805, + "loss": 0.3988, + "step": 20631 + }, + { + "epoch": 1.1527502304662403, + "grad_norm": 1.083182454109192, + "learning_rate": 0.0004253417749887943, + "loss": 0.4964, + "step": 20632 + }, + { + "epoch": 1.152806101069922, + "grad_norm": 0.5107222199440002, + "learning_rate": 0.00042531376064545046, + "loss": 0.4892, + "step": 20633 + }, + { + "epoch": 1.152861971673604, + "grad_norm": 0.8960114121437073, + "learning_rate": 0.0004252857463021067, + "loss": 0.4788, + "step": 20634 + }, + { + "epoch": 1.1529178422772859, + "grad_norm": 0.43065309524536133, + "learning_rate": 0.00042525773195876287, + "loss": 0.4346, + "step": 20635 + }, + { + "epoch": 1.1529737128809676, + "grad_norm": 2.4121463298797607, + "learning_rate": 0.00042522971761541913, + "loss": 0.3869, + "step": 20636 + }, + { + "epoch": 1.1530295834846496, + "grad_norm": 0.5178459286689758, + "learning_rate": 0.0004252017032720753, + "loss": 0.4148, + "step": 20637 + }, + { + "epoch": 1.1530854540883315, + "grad_norm": 4.95328426361084, + "learning_rate": 0.00042517368892873154, + "loss": 0.5429, + "step": 20638 + }, + { + "epoch": 1.1531413246920132, + "grad_norm": 0.6648011803627014, + "learning_rate": 0.0004251456745853877, + "loss": 0.4566, + "step": 20639 + }, + { + "epoch": 1.1531971952956952, + "grad_norm": 0.502936065196991, + "learning_rate": 0.00042511766024204396, + "loss": 0.4566, + "step": 20640 + }, + { + "epoch": 1.153253065899377, + "grad_norm": 1.0489327907562256, + "learning_rate": 0.00042508964589870016, + "loss": 0.4261, + "step": 20641 + }, + { + "epoch": 1.1533089365030589, + "grad_norm": 0.8499877452850342, + "learning_rate": 0.00042506163155535637, + "loss": 0.4446, + "step": 20642 + }, + { + "epoch": 1.1533648071067408, + "grad_norm": 0.3425532579421997, + "learning_rate": 0.00042503361721201257, + "loss": 0.3072, + "step": 20643 + }, + { + "epoch": 1.1534206777104226, + "grad_norm": 4.370369911193848, + "learning_rate": 0.0004250056028686688, + "loss": 0.3987, + "step": 20644 + }, + { + "epoch": 1.1534765483141045, + "grad_norm": 0.46765902638435364, + "learning_rate": 0.000424977588525325, + "loss": 0.4049, + "step": 20645 + }, + { + "epoch": 1.1535324189177865, + "grad_norm": 0.3483951985836029, + "learning_rate": 0.00042494957418198114, + "loss": 0.3478, + "step": 20646 + }, + { + "epoch": 1.1535882895214682, + "grad_norm": 0.8164330720901489, + "learning_rate": 0.0004249215598386374, + "loss": 0.5316, + "step": 20647 + }, + { + "epoch": 1.1536441601251501, + "grad_norm": 0.4574241638183594, + "learning_rate": 0.0004248935454952936, + "loss": 0.5075, + "step": 20648 + }, + { + "epoch": 1.153700030728832, + "grad_norm": 0.7814725637435913, + "learning_rate": 0.0004248655311519498, + "loss": 0.5858, + "step": 20649 + }, + { + "epoch": 1.1537559013325138, + "grad_norm": 0.46084287762641907, + "learning_rate": 0.000424837516808606, + "loss": 0.5514, + "step": 20650 + }, + { + "epoch": 1.1538117719361958, + "grad_norm": 1.2474102973937988, + "learning_rate": 0.0004248095024652622, + "loss": 0.533, + "step": 20651 + }, + { + "epoch": 1.1538676425398777, + "grad_norm": 0.7082433104515076, + "learning_rate": 0.0004247814881219184, + "loss": 0.48, + "step": 20652 + }, + { + "epoch": 1.1539235131435595, + "grad_norm": 1.9532262086868286, + "learning_rate": 0.00042475347377857463, + "loss": 0.4223, + "step": 20653 + }, + { + "epoch": 1.1539793837472414, + "grad_norm": 0.4188125729560852, + "learning_rate": 0.00042472545943523084, + "loss": 0.4032, + "step": 20654 + }, + { + "epoch": 1.1540352543509234, + "grad_norm": 1.0623873472213745, + "learning_rate": 0.00042469744509188704, + "loss": 0.4448, + "step": 20655 + }, + { + "epoch": 1.154091124954605, + "grad_norm": 2.390270471572876, + "learning_rate": 0.00042466943074854325, + "loss": 0.4795, + "step": 20656 + }, + { + "epoch": 1.154146995558287, + "grad_norm": 1.3559377193450928, + "learning_rate": 0.0004246414164051995, + "loss": 0.414, + "step": 20657 + }, + { + "epoch": 1.154202866161969, + "grad_norm": 0.4999708831310272, + "learning_rate": 0.00042461340206185566, + "loss": 0.4192, + "step": 20658 + }, + { + "epoch": 1.1542587367656507, + "grad_norm": 0.48563891649246216, + "learning_rate": 0.0004245853877185119, + "loss": 0.4716, + "step": 20659 + }, + { + "epoch": 1.1543146073693327, + "grad_norm": 0.5057837963104248, + "learning_rate": 0.00042455737337516807, + "loss": 0.402, + "step": 20660 + }, + { + "epoch": 1.1543704779730146, + "grad_norm": 0.3206183910369873, + "learning_rate": 0.00042452935903182433, + "loss": 0.3797, + "step": 20661 + }, + { + "epoch": 1.1544263485766963, + "grad_norm": 0.4696154296398163, + "learning_rate": 0.0004245013446884805, + "loss": 0.4804, + "step": 20662 + }, + { + "epoch": 1.1544822191803783, + "grad_norm": 0.3316584527492523, + "learning_rate": 0.00042447333034513674, + "loss": 0.4321, + "step": 20663 + }, + { + "epoch": 1.15453808978406, + "grad_norm": 0.589804470539093, + "learning_rate": 0.00042444531600179295, + "loss": 0.4646, + "step": 20664 + }, + { + "epoch": 1.154593960387742, + "grad_norm": 0.5501276254653931, + "learning_rate": 0.00042441730165844915, + "loss": 0.382, + "step": 20665 + }, + { + "epoch": 1.154649830991424, + "grad_norm": 0.39172884821891785, + "learning_rate": 0.00042438928731510536, + "loss": 0.4801, + "step": 20666 + }, + { + "epoch": 1.1547057015951057, + "grad_norm": 0.3931137025356293, + "learning_rate": 0.0004243612729717615, + "loss": 0.5804, + "step": 20667 + }, + { + "epoch": 1.1547615721987876, + "grad_norm": 0.43218597769737244, + "learning_rate": 0.00042433325862841777, + "loss": 0.5416, + "step": 20668 + }, + { + "epoch": 1.1548174428024696, + "grad_norm": 0.39650076627731323, + "learning_rate": 0.0004243052442850739, + "loss": 0.416, + "step": 20669 + }, + { + "epoch": 1.1548733134061513, + "grad_norm": 0.5833979249000549, + "learning_rate": 0.0004242772299417302, + "loss": 0.4381, + "step": 20670 + }, + { + "epoch": 1.1549291840098332, + "grad_norm": 0.4886675179004669, + "learning_rate": 0.00042424921559838634, + "loss": 0.5622, + "step": 20671 + }, + { + "epoch": 1.1549850546135152, + "grad_norm": 0.36559465527534485, + "learning_rate": 0.0004242212012550426, + "loss": 0.4625, + "step": 20672 + }, + { + "epoch": 1.155040925217197, + "grad_norm": 0.8295004963874817, + "learning_rate": 0.0004241931869116988, + "loss": 0.3423, + "step": 20673 + }, + { + "epoch": 1.1550967958208789, + "grad_norm": 1.3600749969482422, + "learning_rate": 0.000424165172568355, + "loss": 0.4792, + "step": 20674 + }, + { + "epoch": 1.1551526664245606, + "grad_norm": 0.6916358470916748, + "learning_rate": 0.0004241371582250112, + "loss": 0.419, + "step": 20675 + }, + { + "epoch": 1.1552085370282426, + "grad_norm": 1.167070746421814, + "learning_rate": 0.0004241091438816674, + "loss": 0.424, + "step": 20676 + }, + { + "epoch": 1.1552644076319245, + "grad_norm": 0.44102269411087036, + "learning_rate": 0.0004240811295383236, + "loss": 0.4749, + "step": 20677 + }, + { + "epoch": 1.1553202782356062, + "grad_norm": 1.4750887155532837, + "learning_rate": 0.00042405311519497983, + "loss": 0.3917, + "step": 20678 + }, + { + "epoch": 1.1553761488392882, + "grad_norm": 0.4688643515110016, + "learning_rate": 0.00042402510085163604, + "loss": 0.4929, + "step": 20679 + }, + { + "epoch": 1.1554320194429701, + "grad_norm": 0.6403281688690186, + "learning_rate": 0.0004239970865082923, + "loss": 0.413, + "step": 20680 + }, + { + "epoch": 1.1554878900466519, + "grad_norm": 0.3764358460903168, + "learning_rate": 0.00042396907216494845, + "loss": 0.3752, + "step": 20681 + }, + { + "epoch": 1.1555437606503338, + "grad_norm": 0.5447361469268799, + "learning_rate": 0.0004239410578216047, + "loss": 0.5683, + "step": 20682 + }, + { + "epoch": 1.1555996312540158, + "grad_norm": 0.8225564956665039, + "learning_rate": 0.00042391304347826086, + "loss": 0.4358, + "step": 20683 + }, + { + "epoch": 1.1556555018576975, + "grad_norm": 0.45116621255874634, + "learning_rate": 0.0004238850291349171, + "loss": 0.3653, + "step": 20684 + }, + { + "epoch": 1.1557113724613794, + "grad_norm": 1.6224507093429565, + "learning_rate": 0.00042385701479157327, + "loss": 0.5421, + "step": 20685 + }, + { + "epoch": 1.1557672430650614, + "grad_norm": 0.5492835640907288, + "learning_rate": 0.00042382900044822953, + "loss": 0.4727, + "step": 20686 + }, + { + "epoch": 1.1558231136687431, + "grad_norm": 0.4245157539844513, + "learning_rate": 0.00042380098610488574, + "loss": 0.4487, + "step": 20687 + }, + { + "epoch": 1.155878984272425, + "grad_norm": 0.5740664601325989, + "learning_rate": 0.0004237729717615419, + "loss": 0.2729, + "step": 20688 + }, + { + "epoch": 1.155934854876107, + "grad_norm": 0.6451188921928406, + "learning_rate": 0.00042374495741819815, + "loss": 0.3522, + "step": 20689 + }, + { + "epoch": 1.1559907254797888, + "grad_norm": 0.7771868705749512, + "learning_rate": 0.0004237169430748543, + "loss": 0.4155, + "step": 20690 + }, + { + "epoch": 1.1560465960834707, + "grad_norm": 0.43820640444755554, + "learning_rate": 0.00042368892873151056, + "loss": 0.4015, + "step": 20691 + }, + { + "epoch": 1.1561024666871527, + "grad_norm": 1.7667236328125, + "learning_rate": 0.0004236609143881667, + "loss": 0.5545, + "step": 20692 + }, + { + "epoch": 1.1561583372908344, + "grad_norm": 0.40907013416290283, + "learning_rate": 0.00042363290004482297, + "loss": 0.5018, + "step": 20693 + }, + { + "epoch": 1.1562142078945163, + "grad_norm": 0.9809287190437317, + "learning_rate": 0.0004236048857014791, + "loss": 0.4831, + "step": 20694 + }, + { + "epoch": 1.1562700784981983, + "grad_norm": 0.7496708631515503, + "learning_rate": 0.0004235768713581354, + "loss": 0.5066, + "step": 20695 + }, + { + "epoch": 1.15632594910188, + "grad_norm": 1.0318022966384888, + "learning_rate": 0.0004235488570147916, + "loss": 0.3834, + "step": 20696 + }, + { + "epoch": 1.156381819705562, + "grad_norm": 0.7169497609138489, + "learning_rate": 0.0004235208426714478, + "loss": 0.5029, + "step": 20697 + }, + { + "epoch": 1.1564376903092437, + "grad_norm": 0.5065441131591797, + "learning_rate": 0.000423492828328104, + "loss": 0.508, + "step": 20698 + }, + { + "epoch": 1.1564935609129257, + "grad_norm": 0.3843490481376648, + "learning_rate": 0.0004234648139847602, + "loss": 0.3396, + "step": 20699 + }, + { + "epoch": 1.1565494315166076, + "grad_norm": 0.7110924124717712, + "learning_rate": 0.0004234367996414164, + "loss": 0.5805, + "step": 20700 + }, + { + "epoch": 1.1566053021202893, + "grad_norm": 0.47780048847198486, + "learning_rate": 0.0004234087852980726, + "loss": 0.4163, + "step": 20701 + }, + { + "epoch": 1.1566611727239713, + "grad_norm": 1.265553593635559, + "learning_rate": 0.0004233807709547288, + "loss": 0.4772, + "step": 20702 + }, + { + "epoch": 1.1567170433276532, + "grad_norm": 0.5036099553108215, + "learning_rate": 0.0004233527566113851, + "loss": 0.4993, + "step": 20703 + }, + { + "epoch": 1.156772913931335, + "grad_norm": 0.41352134943008423, + "learning_rate": 0.00042332474226804124, + "loss": 0.4644, + "step": 20704 + }, + { + "epoch": 1.156828784535017, + "grad_norm": 1.2280915975570679, + "learning_rate": 0.0004232967279246975, + "loss": 0.4146, + "step": 20705 + }, + { + "epoch": 1.1568846551386989, + "grad_norm": 0.4748990535736084, + "learning_rate": 0.00042326871358135365, + "loss": 0.4253, + "step": 20706 + }, + { + "epoch": 1.1569405257423806, + "grad_norm": 1.003219723701477, + "learning_rate": 0.0004232406992380099, + "loss": 0.3971, + "step": 20707 + }, + { + "epoch": 1.1569963963460626, + "grad_norm": 0.4461818337440491, + "learning_rate": 0.00042321268489466606, + "loss": 0.457, + "step": 20708 + }, + { + "epoch": 1.1570522669497443, + "grad_norm": 0.5697447657585144, + "learning_rate": 0.00042318467055132226, + "loss": 0.4594, + "step": 20709 + }, + { + "epoch": 1.1571081375534262, + "grad_norm": 0.492354154586792, + "learning_rate": 0.00042315665620797847, + "loss": 0.3342, + "step": 20710 + }, + { + "epoch": 1.1571640081571082, + "grad_norm": 0.4515548050403595, + "learning_rate": 0.0004231286418646347, + "loss": 0.4672, + "step": 20711 + }, + { + "epoch": 1.15721987876079, + "grad_norm": 0.49982771277427673, + "learning_rate": 0.00042310062752129094, + "loss": 0.4646, + "step": 20712 + }, + { + "epoch": 1.1572757493644719, + "grad_norm": 0.7837000489234924, + "learning_rate": 0.0004230726131779471, + "loss": 0.4238, + "step": 20713 + }, + { + "epoch": 1.1573316199681538, + "grad_norm": 0.3699081838130951, + "learning_rate": 0.00042304459883460335, + "loss": 0.4017, + "step": 20714 + }, + { + "epoch": 1.1573874905718355, + "grad_norm": 0.6186193227767944, + "learning_rate": 0.0004230165844912595, + "loss": 0.4809, + "step": 20715 + }, + { + "epoch": 1.1574433611755175, + "grad_norm": 1.7799900770187378, + "learning_rate": 0.00042298857014791576, + "loss": 0.4494, + "step": 20716 + }, + { + "epoch": 1.1574992317791994, + "grad_norm": 0.42213767766952515, + "learning_rate": 0.0004229605558045719, + "loss": 0.3542, + "step": 20717 + }, + { + "epoch": 1.1575551023828812, + "grad_norm": 0.44648149609565735, + "learning_rate": 0.00042293254146122817, + "loss": 0.4478, + "step": 20718 + }, + { + "epoch": 1.1576109729865631, + "grad_norm": 0.37672507762908936, + "learning_rate": 0.0004229045271178844, + "loss": 0.4741, + "step": 20719 + }, + { + "epoch": 1.157666843590245, + "grad_norm": 0.5919115543365479, + "learning_rate": 0.0004228765127745406, + "loss": 0.402, + "step": 20720 + }, + { + "epoch": 1.1577227141939268, + "grad_norm": 1.2035694122314453, + "learning_rate": 0.0004228484984311968, + "loss": 0.4242, + "step": 20721 + }, + { + "epoch": 1.1577785847976088, + "grad_norm": 0.5906327366828918, + "learning_rate": 0.000422820484087853, + "loss": 0.4729, + "step": 20722 + }, + { + "epoch": 1.1578344554012907, + "grad_norm": 0.5088474750518799, + "learning_rate": 0.0004227924697445092, + "loss": 0.4519, + "step": 20723 + }, + { + "epoch": 1.1578903260049724, + "grad_norm": 0.7282567024230957, + "learning_rate": 0.0004227644554011654, + "loss": 0.5343, + "step": 20724 + }, + { + "epoch": 1.1579461966086544, + "grad_norm": 0.42200952768325806, + "learning_rate": 0.0004227364410578216, + "loss": 0.4169, + "step": 20725 + }, + { + "epoch": 1.1580020672123363, + "grad_norm": 0.4575338661670685, + "learning_rate": 0.0004227084267144778, + "loss": 0.5474, + "step": 20726 + }, + { + "epoch": 1.158057937816018, + "grad_norm": 0.38230976462364197, + "learning_rate": 0.000422680412371134, + "loss": 0.4121, + "step": 20727 + }, + { + "epoch": 1.1581138084197, + "grad_norm": 0.49246734380722046, + "learning_rate": 0.0004226523980277903, + "loss": 0.4624, + "step": 20728 + }, + { + "epoch": 1.158169679023382, + "grad_norm": 0.8844696283340454, + "learning_rate": 0.00042262438368444644, + "loss": 0.574, + "step": 20729 + }, + { + "epoch": 1.1582255496270637, + "grad_norm": 5.51059103012085, + "learning_rate": 0.00042259636934110264, + "loss": 0.4228, + "step": 20730 + }, + { + "epoch": 1.1582814202307457, + "grad_norm": 0.5678775906562805, + "learning_rate": 0.00042256835499775885, + "loss": 0.4593, + "step": 20731 + }, + { + "epoch": 1.1583372908344274, + "grad_norm": 0.547385036945343, + "learning_rate": 0.00042254034065441505, + "loss": 0.4754, + "step": 20732 + }, + { + "epoch": 1.1583931614381093, + "grad_norm": 0.5175642967224121, + "learning_rate": 0.00042251232631107126, + "loss": 0.3609, + "step": 20733 + }, + { + "epoch": 1.1584490320417913, + "grad_norm": 0.4762882888317108, + "learning_rate": 0.00042248431196772746, + "loss": 0.5385, + "step": 20734 + }, + { + "epoch": 1.158504902645473, + "grad_norm": 1.2463847398757935, + "learning_rate": 0.0004224562976243837, + "loss": 0.4883, + "step": 20735 + }, + { + "epoch": 1.158560773249155, + "grad_norm": 0.40370139479637146, + "learning_rate": 0.0004224282832810399, + "loss": 0.5141, + "step": 20736 + }, + { + "epoch": 1.158616643852837, + "grad_norm": 0.31559205055236816, + "learning_rate": 0.00042240026893769614, + "loss": 0.3997, + "step": 20737 + }, + { + "epoch": 1.1586725144565186, + "grad_norm": 0.40364524722099304, + "learning_rate": 0.0004223722545943523, + "loss": 0.3668, + "step": 20738 + }, + { + "epoch": 1.1587283850602006, + "grad_norm": 0.5192253589630127, + "learning_rate": 0.00042234424025100855, + "loss": 0.3727, + "step": 20739 + }, + { + "epoch": 1.1587842556638825, + "grad_norm": 0.3146277666091919, + "learning_rate": 0.0004223162259076647, + "loss": 0.3843, + "step": 20740 + }, + { + "epoch": 1.1588401262675643, + "grad_norm": 0.7322179675102234, + "learning_rate": 0.00042228821156432096, + "loss": 0.4516, + "step": 20741 + }, + { + "epoch": 1.1588959968712462, + "grad_norm": 0.4595886170864105, + "learning_rate": 0.0004222601972209771, + "loss": 0.4959, + "step": 20742 + }, + { + "epoch": 1.158951867474928, + "grad_norm": 0.42870816588401794, + "learning_rate": 0.00042223218287763337, + "loss": 0.3445, + "step": 20743 + }, + { + "epoch": 1.15900773807861, + "grad_norm": 0.3775968551635742, + "learning_rate": 0.0004222041685342896, + "loss": 0.3995, + "step": 20744 + }, + { + "epoch": 1.1590636086822919, + "grad_norm": 0.6125504970550537, + "learning_rate": 0.0004221761541909458, + "loss": 0.3996, + "step": 20745 + }, + { + "epoch": 1.1591194792859736, + "grad_norm": 0.41319888830184937, + "learning_rate": 0.000422148139847602, + "loss": 0.3662, + "step": 20746 + }, + { + "epoch": 1.1591753498896555, + "grad_norm": 0.3611396849155426, + "learning_rate": 0.0004221201255042582, + "loss": 0.4288, + "step": 20747 + }, + { + "epoch": 1.1592312204933375, + "grad_norm": 0.3642005920410156, + "learning_rate": 0.0004220921111609144, + "loss": 0.449, + "step": 20748 + }, + { + "epoch": 1.1592870910970192, + "grad_norm": 0.41246575117111206, + "learning_rate": 0.0004220640968175706, + "loss": 0.4255, + "step": 20749 + }, + { + "epoch": 1.1593429617007012, + "grad_norm": 0.34570518136024475, + "learning_rate": 0.0004220360824742268, + "loss": 0.3822, + "step": 20750 + }, + { + "epoch": 1.1593988323043831, + "grad_norm": 0.6184771060943604, + "learning_rate": 0.000422008068130883, + "loss": 0.4138, + "step": 20751 + }, + { + "epoch": 1.1594547029080648, + "grad_norm": 0.6864798665046692, + "learning_rate": 0.0004219800537875392, + "loss": 0.4805, + "step": 20752 + }, + { + "epoch": 1.1595105735117468, + "grad_norm": 0.5587151050567627, + "learning_rate": 0.00042195203944419543, + "loss": 0.4378, + "step": 20753 + }, + { + "epoch": 1.1595664441154288, + "grad_norm": 0.5914813280105591, + "learning_rate": 0.00042192402510085164, + "loss": 0.4715, + "step": 20754 + }, + { + "epoch": 1.1596223147191105, + "grad_norm": 2.549452066421509, + "learning_rate": 0.00042189601075750784, + "loss": 0.5032, + "step": 20755 + }, + { + "epoch": 1.1596781853227924, + "grad_norm": 0.43182384967803955, + "learning_rate": 0.00042186799641416405, + "loss": 0.3451, + "step": 20756 + }, + { + "epoch": 1.1597340559264744, + "grad_norm": 0.5013192296028137, + "learning_rate": 0.00042183998207082025, + "loss": 0.4832, + "step": 20757 + }, + { + "epoch": 1.1597899265301561, + "grad_norm": 0.5790650248527527, + "learning_rate": 0.0004218119677274765, + "loss": 0.4796, + "step": 20758 + }, + { + "epoch": 1.159845797133838, + "grad_norm": 0.3794909715652466, + "learning_rate": 0.00042178395338413266, + "loss": 0.4494, + "step": 20759 + }, + { + "epoch": 1.15990166773752, + "grad_norm": 0.4153733253479004, + "learning_rate": 0.0004217559390407889, + "loss": 0.4488, + "step": 20760 + }, + { + "epoch": 1.1599575383412017, + "grad_norm": 0.42221319675445557, + "learning_rate": 0.0004217279246974451, + "loss": 0.4721, + "step": 20761 + }, + { + "epoch": 1.1600134089448837, + "grad_norm": 0.46664130687713623, + "learning_rate": 0.00042169991035410134, + "loss": 0.3884, + "step": 20762 + }, + { + "epoch": 1.1600692795485656, + "grad_norm": 0.45028916001319885, + "learning_rate": 0.0004216718960107575, + "loss": 0.3842, + "step": 20763 + }, + { + "epoch": 1.1601251501522474, + "grad_norm": 0.5644640326499939, + "learning_rate": 0.00042164388166741375, + "loss": 0.5757, + "step": 20764 + }, + { + "epoch": 1.1601810207559293, + "grad_norm": 0.792530357837677, + "learning_rate": 0.0004216158673240699, + "loss": 0.4807, + "step": 20765 + }, + { + "epoch": 1.160236891359611, + "grad_norm": 0.4336341321468353, + "learning_rate": 0.00042158785298072616, + "loss": 0.3901, + "step": 20766 + }, + { + "epoch": 1.160292761963293, + "grad_norm": 0.6042793393135071, + "learning_rate": 0.00042155983863738236, + "loss": 0.4237, + "step": 20767 + }, + { + "epoch": 1.160348632566975, + "grad_norm": 0.41389790177345276, + "learning_rate": 0.00042153182429403857, + "loss": 0.2937, + "step": 20768 + }, + { + "epoch": 1.1604045031706567, + "grad_norm": 0.38278475403785706, + "learning_rate": 0.0004215038099506948, + "loss": 0.5087, + "step": 20769 + }, + { + "epoch": 1.1604603737743386, + "grad_norm": 1.7464804649353027, + "learning_rate": 0.000421475795607351, + "loss": 0.4398, + "step": 20770 + }, + { + "epoch": 1.1605162443780206, + "grad_norm": 0.5278064608573914, + "learning_rate": 0.0004214477812640072, + "loss": 0.5187, + "step": 20771 + }, + { + "epoch": 1.1605721149817023, + "grad_norm": 0.579298734664917, + "learning_rate": 0.00042141976692066334, + "loss": 0.4233, + "step": 20772 + }, + { + "epoch": 1.1606279855853843, + "grad_norm": 1.5956724882125854, + "learning_rate": 0.0004213917525773196, + "loss": 0.3886, + "step": 20773 + }, + { + "epoch": 1.160683856189066, + "grad_norm": 1.4982813596725464, + "learning_rate": 0.0004213637382339758, + "loss": 0.3892, + "step": 20774 + }, + { + "epoch": 1.160739726792748, + "grad_norm": 0.5616388916969299, + "learning_rate": 0.000421335723890632, + "loss": 0.5485, + "step": 20775 + }, + { + "epoch": 1.16079559739643, + "grad_norm": 1.090511441230774, + "learning_rate": 0.0004213077095472882, + "loss": 0.4076, + "step": 20776 + }, + { + "epoch": 1.1608514680001116, + "grad_norm": 0.36427435278892517, + "learning_rate": 0.0004212796952039444, + "loss": 0.4658, + "step": 20777 + }, + { + "epoch": 1.1609073386037936, + "grad_norm": 0.5029274225234985, + "learning_rate": 0.00042125168086060063, + "loss": 0.5659, + "step": 20778 + }, + { + "epoch": 1.1609632092074755, + "grad_norm": 0.4948616325855255, + "learning_rate": 0.00042122366651725683, + "loss": 0.449, + "step": 20779 + }, + { + "epoch": 1.1610190798111573, + "grad_norm": 1.7187049388885498, + "learning_rate": 0.00042119565217391304, + "loss": 0.3977, + "step": 20780 + }, + { + "epoch": 1.1610749504148392, + "grad_norm": 0.5496429204940796, + "learning_rate": 0.00042116763783056925, + "loss": 0.3168, + "step": 20781 + }, + { + "epoch": 1.1611308210185212, + "grad_norm": 0.40350040793418884, + "learning_rate": 0.00042113962348722545, + "loss": 0.4015, + "step": 20782 + }, + { + "epoch": 1.161186691622203, + "grad_norm": 0.4099721908569336, + "learning_rate": 0.0004211116091438817, + "loss": 0.4179, + "step": 20783 + }, + { + "epoch": 1.1612425622258848, + "grad_norm": 0.447115033864975, + "learning_rate": 0.00042108359480053786, + "loss": 0.425, + "step": 20784 + }, + { + "epoch": 1.1612984328295668, + "grad_norm": 0.3867757320404053, + "learning_rate": 0.0004210555804571941, + "loss": 0.4824, + "step": 20785 + }, + { + "epoch": 1.1613543034332485, + "grad_norm": 0.8540741205215454, + "learning_rate": 0.0004210275661138503, + "loss": 0.382, + "step": 20786 + }, + { + "epoch": 1.1614101740369305, + "grad_norm": 0.39895182847976685, + "learning_rate": 0.00042099955177050654, + "loss": 0.3918, + "step": 20787 + }, + { + "epoch": 1.1614660446406124, + "grad_norm": 0.34172430634498596, + "learning_rate": 0.0004209715374271627, + "loss": 0.4427, + "step": 20788 + }, + { + "epoch": 1.1615219152442942, + "grad_norm": 4.137818813323975, + "learning_rate": 0.00042094352308381895, + "loss": 0.6364, + "step": 20789 + }, + { + "epoch": 1.161577785847976, + "grad_norm": 0.631015419960022, + "learning_rate": 0.00042091550874047515, + "loss": 0.4009, + "step": 20790 + }, + { + "epoch": 1.161633656451658, + "grad_norm": 1.8745111227035522, + "learning_rate": 0.00042088749439713136, + "loss": 0.4111, + "step": 20791 + }, + { + "epoch": 1.1616895270553398, + "grad_norm": 0.7809178829193115, + "learning_rate": 0.00042085948005378756, + "loss": 0.4182, + "step": 20792 + }, + { + "epoch": 1.1617453976590217, + "grad_norm": 0.5435146689414978, + "learning_rate": 0.0004208314657104437, + "loss": 0.4554, + "step": 20793 + }, + { + "epoch": 1.1618012682627037, + "grad_norm": 0.5920735001564026, + "learning_rate": 0.0004208034513671, + "loss": 0.4868, + "step": 20794 + }, + { + "epoch": 1.1618571388663854, + "grad_norm": 0.3459606170654297, + "learning_rate": 0.00042077543702375613, + "loss": 0.3576, + "step": 20795 + }, + { + "epoch": 1.1619130094700674, + "grad_norm": 1.1702375411987305, + "learning_rate": 0.0004207474226804124, + "loss": 0.5906, + "step": 20796 + }, + { + "epoch": 1.1619688800737493, + "grad_norm": 4.218400478363037, + "learning_rate": 0.00042071940833706854, + "loss": 0.3451, + "step": 20797 + }, + { + "epoch": 1.162024750677431, + "grad_norm": 1.1880065202713013, + "learning_rate": 0.0004206913939937248, + "loss": 0.4537, + "step": 20798 + }, + { + "epoch": 1.162080621281113, + "grad_norm": 0.6565853357315063, + "learning_rate": 0.000420663379650381, + "loss": 0.4667, + "step": 20799 + }, + { + "epoch": 1.1621364918847947, + "grad_norm": 0.4321233928203583, + "learning_rate": 0.0004206353653070372, + "loss": 0.3353, + "step": 20800 + }, + { + "epoch": 1.1621923624884767, + "grad_norm": 0.33756133913993835, + "learning_rate": 0.0004206073509636934, + "loss": 0.3369, + "step": 20801 + }, + { + "epoch": 1.1622482330921586, + "grad_norm": 0.36357560753822327, + "learning_rate": 0.0004205793366203496, + "loss": 0.4106, + "step": 20802 + }, + { + "epoch": 1.1623041036958404, + "grad_norm": 0.5828847289085388, + "learning_rate": 0.00042055132227700583, + "loss": 0.4829, + "step": 20803 + }, + { + "epoch": 1.1623599742995223, + "grad_norm": 0.30819013714790344, + "learning_rate": 0.00042052330793366203, + "loss": 0.341, + "step": 20804 + }, + { + "epoch": 1.1624158449032043, + "grad_norm": 0.4753556549549103, + "learning_rate": 0.00042049529359031824, + "loss": 0.5542, + "step": 20805 + }, + { + "epoch": 1.162471715506886, + "grad_norm": 1.2714931964874268, + "learning_rate": 0.0004204672792469745, + "loss": 0.4455, + "step": 20806 + }, + { + "epoch": 1.162527586110568, + "grad_norm": 0.34421107172966003, + "learning_rate": 0.00042043926490363065, + "loss": 0.3715, + "step": 20807 + }, + { + "epoch": 1.1625834567142497, + "grad_norm": 0.5650119781494141, + "learning_rate": 0.0004204112505602869, + "loss": 0.5957, + "step": 20808 + }, + { + "epoch": 1.1626393273179316, + "grad_norm": 2.8605799674987793, + "learning_rate": 0.00042038323621694306, + "loss": 0.5199, + "step": 20809 + }, + { + "epoch": 1.1626951979216136, + "grad_norm": 0.45128875970840454, + "learning_rate": 0.0004203552218735993, + "loss": 0.3244, + "step": 20810 + }, + { + "epoch": 1.1627510685252953, + "grad_norm": 0.4296343922615051, + "learning_rate": 0.0004203272075302555, + "loss": 0.4221, + "step": 20811 + }, + { + "epoch": 1.1628069391289773, + "grad_norm": 0.4030792713165283, + "learning_rate": 0.00042029919318691173, + "loss": 0.4091, + "step": 20812 + }, + { + "epoch": 1.1628628097326592, + "grad_norm": 1.4344799518585205, + "learning_rate": 0.00042027117884356794, + "loss": 0.4266, + "step": 20813 + }, + { + "epoch": 1.162918680336341, + "grad_norm": 0.5077324509620667, + "learning_rate": 0.00042024316450022415, + "loss": 0.378, + "step": 20814 + }, + { + "epoch": 1.1629745509400229, + "grad_norm": 0.3744187653064728, + "learning_rate": 0.00042021515015688035, + "loss": 0.4121, + "step": 20815 + }, + { + "epoch": 1.1630304215437048, + "grad_norm": 2.24889874458313, + "learning_rate": 0.0004201871358135365, + "loss": 0.4994, + "step": 20816 + }, + { + "epoch": 1.1630862921473866, + "grad_norm": 0.5369593501091003, + "learning_rate": 0.00042015912147019276, + "loss": 0.5267, + "step": 20817 + }, + { + "epoch": 1.1631421627510685, + "grad_norm": 0.5651959776878357, + "learning_rate": 0.0004201311071268489, + "loss": 0.43, + "step": 20818 + }, + { + "epoch": 1.1631980333547505, + "grad_norm": 0.344714492559433, + "learning_rate": 0.0004201030927835052, + "loss": 0.3755, + "step": 20819 + }, + { + "epoch": 1.1632539039584322, + "grad_norm": 0.54056715965271, + "learning_rate": 0.00042007507844016133, + "loss": 0.4764, + "step": 20820 + }, + { + "epoch": 1.1633097745621142, + "grad_norm": 0.3365499973297119, + "learning_rate": 0.0004200470640968176, + "loss": 0.378, + "step": 20821 + }, + { + "epoch": 1.163365645165796, + "grad_norm": 0.43799713253974915, + "learning_rate": 0.0004200190497534738, + "loss": 0.4062, + "step": 20822 + }, + { + "epoch": 1.1634215157694778, + "grad_norm": 0.5295700430870056, + "learning_rate": 0.00041999103541013, + "loss": 0.4761, + "step": 20823 + }, + { + "epoch": 1.1634773863731598, + "grad_norm": 0.5793594121932983, + "learning_rate": 0.0004199630210667862, + "loss": 0.6959, + "step": 20824 + }, + { + "epoch": 1.1635332569768417, + "grad_norm": 0.3703233003616333, + "learning_rate": 0.0004199350067234424, + "loss": 0.4141, + "step": 20825 + }, + { + "epoch": 1.1635891275805235, + "grad_norm": 0.543444037437439, + "learning_rate": 0.0004199069923800986, + "loss": 0.5059, + "step": 20826 + }, + { + "epoch": 1.1636449981842054, + "grad_norm": 0.3411475121974945, + "learning_rate": 0.0004198789780367548, + "loss": 0.354, + "step": 20827 + }, + { + "epoch": 1.1637008687878874, + "grad_norm": 2.3099560737609863, + "learning_rate": 0.00041985096369341103, + "loss": 0.43, + "step": 20828 + }, + { + "epoch": 1.163756739391569, + "grad_norm": 2.109989881515503, + "learning_rate": 0.0004198229493500673, + "loss": 0.4181, + "step": 20829 + }, + { + "epoch": 1.163812609995251, + "grad_norm": 0.4622809886932373, + "learning_rate": 0.00041979493500672344, + "loss": 0.5294, + "step": 20830 + }, + { + "epoch": 1.1638684805989328, + "grad_norm": 0.5098580121994019, + "learning_rate": 0.0004197669206633797, + "loss": 0.4237, + "step": 20831 + }, + { + "epoch": 1.1639243512026147, + "grad_norm": 0.38552913069725037, + "learning_rate": 0.00041973890632003585, + "loss": 0.4802, + "step": 20832 + }, + { + "epoch": 1.1639802218062967, + "grad_norm": 0.44642582535743713, + "learning_rate": 0.0004197108919766921, + "loss": 0.4102, + "step": 20833 + }, + { + "epoch": 1.1640360924099784, + "grad_norm": 0.3959483504295349, + "learning_rate": 0.00041968287763334826, + "loss": 0.4197, + "step": 20834 + }, + { + "epoch": 1.1640919630136604, + "grad_norm": 0.4431978166103363, + "learning_rate": 0.0004196548632900045, + "loss": 0.4292, + "step": 20835 + }, + { + "epoch": 1.1641478336173423, + "grad_norm": 0.43161967396736145, + "learning_rate": 0.0004196268489466607, + "loss": 0.5113, + "step": 20836 + }, + { + "epoch": 1.164203704221024, + "grad_norm": 0.49541011452674866, + "learning_rate": 0.0004195988346033169, + "loss": 0.3959, + "step": 20837 + }, + { + "epoch": 1.164259574824706, + "grad_norm": 0.43946942687034607, + "learning_rate": 0.00041957082025997314, + "loss": 0.4403, + "step": 20838 + }, + { + "epoch": 1.164315445428388, + "grad_norm": 0.5001556277275085, + "learning_rate": 0.0004195428059166293, + "loss": 0.4732, + "step": 20839 + }, + { + "epoch": 1.1643713160320697, + "grad_norm": 0.34555888175964355, + "learning_rate": 0.00041951479157328555, + "loss": 0.4057, + "step": 20840 + }, + { + "epoch": 1.1644271866357516, + "grad_norm": 0.36191776394844055, + "learning_rate": 0.0004194867772299417, + "loss": 0.3379, + "step": 20841 + }, + { + "epoch": 1.1644830572394334, + "grad_norm": 3.1381990909576416, + "learning_rate": 0.00041945876288659796, + "loss": 0.4221, + "step": 20842 + }, + { + "epoch": 1.1645389278431153, + "grad_norm": 0.5201879143714905, + "learning_rate": 0.0004194307485432541, + "loss": 0.4632, + "step": 20843 + }, + { + "epoch": 1.1645947984467973, + "grad_norm": 0.43290677666664124, + "learning_rate": 0.0004194027341999104, + "loss": 0.4949, + "step": 20844 + }, + { + "epoch": 1.164650669050479, + "grad_norm": 1.471893310546875, + "learning_rate": 0.0004193747198565666, + "loss": 0.4824, + "step": 20845 + }, + { + "epoch": 1.164706539654161, + "grad_norm": 0.7570842504501343, + "learning_rate": 0.0004193467055132228, + "loss": 0.4639, + "step": 20846 + }, + { + "epoch": 1.1647624102578429, + "grad_norm": 2.8146610260009766, + "learning_rate": 0.000419318691169879, + "loss": 0.4207, + "step": 20847 + }, + { + "epoch": 1.1648182808615246, + "grad_norm": 0.43162500858306885, + "learning_rate": 0.0004192906768265352, + "loss": 0.4416, + "step": 20848 + }, + { + "epoch": 1.1648741514652066, + "grad_norm": 0.4567854106426239, + "learning_rate": 0.0004192626624831914, + "loss": 0.5218, + "step": 20849 + }, + { + "epoch": 1.1649300220688885, + "grad_norm": 0.3996514678001404, + "learning_rate": 0.0004192346481398476, + "loss": 0.3146, + "step": 20850 + }, + { + "epoch": 1.1649858926725702, + "grad_norm": 0.4616847336292267, + "learning_rate": 0.0004192066337965038, + "loss": 0.4794, + "step": 20851 + }, + { + "epoch": 1.1650417632762522, + "grad_norm": 0.6287479996681213, + "learning_rate": 0.00041917861945316, + "loss": 0.4547, + "step": 20852 + }, + { + "epoch": 1.1650976338799341, + "grad_norm": 0.39289212226867676, + "learning_rate": 0.00041915060510981623, + "loss": 0.4231, + "step": 20853 + }, + { + "epoch": 1.1651535044836159, + "grad_norm": 0.38536718487739563, + "learning_rate": 0.0004191225907664725, + "loss": 0.3763, + "step": 20854 + }, + { + "epoch": 1.1652093750872978, + "grad_norm": 0.572665810585022, + "learning_rate": 0.00041909457642312864, + "loss": 0.4581, + "step": 20855 + }, + { + "epoch": 1.1652652456909798, + "grad_norm": 0.37772780656814575, + "learning_rate": 0.0004190665620797849, + "loss": 0.3228, + "step": 20856 + }, + { + "epoch": 1.1653211162946615, + "grad_norm": 0.34642383456230164, + "learning_rate": 0.00041903854773644105, + "loss": 0.3918, + "step": 20857 + }, + { + "epoch": 1.1653769868983435, + "grad_norm": 1.029593825340271, + "learning_rate": 0.00041901053339309726, + "loss": 0.3791, + "step": 20858 + }, + { + "epoch": 1.1654328575020254, + "grad_norm": 0.5496186017990112, + "learning_rate": 0.00041898251904975346, + "loss": 0.4609, + "step": 20859 + }, + { + "epoch": 1.1654887281057071, + "grad_norm": 0.34662240743637085, + "learning_rate": 0.00041895450470640967, + "loss": 0.3399, + "step": 20860 + }, + { + "epoch": 1.165544598709389, + "grad_norm": 0.5085844397544861, + "learning_rate": 0.00041892649036306593, + "loss": 0.4078, + "step": 20861 + }, + { + "epoch": 1.165600469313071, + "grad_norm": 0.4033202826976776, + "learning_rate": 0.0004188984760197221, + "loss": 0.4033, + "step": 20862 + }, + { + "epoch": 1.1656563399167528, + "grad_norm": 1.5661920309066772, + "learning_rate": 0.00041887046167637834, + "loss": 0.5883, + "step": 20863 + }, + { + "epoch": 1.1657122105204347, + "grad_norm": 0.5242140889167786, + "learning_rate": 0.0004188424473330345, + "loss": 0.3761, + "step": 20864 + }, + { + "epoch": 1.1657680811241165, + "grad_norm": 0.32173842191696167, + "learning_rate": 0.00041881443298969075, + "loss": 0.3613, + "step": 20865 + }, + { + "epoch": 1.1658239517277984, + "grad_norm": 0.5041488409042358, + "learning_rate": 0.0004187864186463469, + "loss": 0.4277, + "step": 20866 + }, + { + "epoch": 1.1658798223314804, + "grad_norm": 0.4079604744911194, + "learning_rate": 0.00041875840430300316, + "loss": 0.5361, + "step": 20867 + }, + { + "epoch": 1.165935692935162, + "grad_norm": 0.3749579191207886, + "learning_rate": 0.0004187303899596593, + "loss": 0.3152, + "step": 20868 + }, + { + "epoch": 1.165991563538844, + "grad_norm": 0.3338612914085388, + "learning_rate": 0.0004187023756163156, + "loss": 0.3837, + "step": 20869 + }, + { + "epoch": 1.166047434142526, + "grad_norm": 0.3639768958091736, + "learning_rate": 0.0004186743612729718, + "loss": 0.3572, + "step": 20870 + }, + { + "epoch": 1.1661033047462077, + "grad_norm": 0.36424338817596436, + "learning_rate": 0.000418646346929628, + "loss": 0.4291, + "step": 20871 + }, + { + "epoch": 1.1661591753498897, + "grad_norm": 0.6470553278923035, + "learning_rate": 0.0004186183325862842, + "loss": 0.4112, + "step": 20872 + }, + { + "epoch": 1.1662150459535716, + "grad_norm": 0.5773619413375854, + "learning_rate": 0.0004185903182429404, + "loss": 0.3326, + "step": 20873 + }, + { + "epoch": 1.1662709165572533, + "grad_norm": 0.38473889231681824, + "learning_rate": 0.0004185623038995966, + "loss": 0.3877, + "step": 20874 + }, + { + "epoch": 1.1663267871609353, + "grad_norm": 0.3924236595630646, + "learning_rate": 0.0004185342895562528, + "loss": 0.4219, + "step": 20875 + }, + { + "epoch": 1.166382657764617, + "grad_norm": 0.44573453068733215, + "learning_rate": 0.000418506275212909, + "loss": 0.4748, + "step": 20876 + }, + { + "epoch": 1.166438528368299, + "grad_norm": 0.4679346978664398, + "learning_rate": 0.0004184782608695653, + "loss": 0.4217, + "step": 20877 + }, + { + "epoch": 1.166494398971981, + "grad_norm": 0.4875365197658539, + "learning_rate": 0.00041845024652622143, + "loss": 0.3858, + "step": 20878 + }, + { + "epoch": 1.1665502695756627, + "grad_norm": 0.9753456115722656, + "learning_rate": 0.00041842223218287763, + "loss": 0.4587, + "step": 20879 + }, + { + "epoch": 1.1666061401793446, + "grad_norm": 0.5865544676780701, + "learning_rate": 0.00041839421783953384, + "loss": 0.3656, + "step": 20880 + }, + { + "epoch": 1.1666620107830266, + "grad_norm": 0.4055899381637573, + "learning_rate": 0.00041836620349619004, + "loss": 0.384, + "step": 20881 + }, + { + "epoch": 1.1667178813867083, + "grad_norm": 0.29809293150901794, + "learning_rate": 0.00041833818915284625, + "loss": 0.3143, + "step": 20882 + }, + { + "epoch": 1.1667737519903902, + "grad_norm": 2.1312031745910645, + "learning_rate": 0.00041831017480950246, + "loss": 0.4499, + "step": 20883 + }, + { + "epoch": 1.1668296225940722, + "grad_norm": 0.48832938075065613, + "learning_rate": 0.0004182821604661587, + "loss": 0.436, + "step": 20884 + }, + { + "epoch": 1.166885493197754, + "grad_norm": 0.37263813614845276, + "learning_rate": 0.00041825414612281487, + "loss": 0.534, + "step": 20885 + }, + { + "epoch": 1.1669413638014359, + "grad_norm": 0.5336165428161621, + "learning_rate": 0.00041822613177947113, + "loss": 0.6123, + "step": 20886 + }, + { + "epoch": 1.1669972344051178, + "grad_norm": 0.7255441546440125, + "learning_rate": 0.0004181981174361273, + "loss": 0.6275, + "step": 20887 + }, + { + "epoch": 1.1670531050087996, + "grad_norm": 1.3660138845443726, + "learning_rate": 0.00041817010309278354, + "loss": 0.4421, + "step": 20888 + }, + { + "epoch": 1.1671089756124815, + "grad_norm": 0.5568463206291199, + "learning_rate": 0.0004181420887494397, + "loss": 0.4023, + "step": 20889 + }, + { + "epoch": 1.1671648462161635, + "grad_norm": 1.3047116994857788, + "learning_rate": 0.00041811407440609595, + "loss": 0.3671, + "step": 20890 + }, + { + "epoch": 1.1672207168198452, + "grad_norm": 0.44206511974334717, + "learning_rate": 0.0004180860600627521, + "loss": 0.3837, + "step": 20891 + }, + { + "epoch": 1.1672765874235271, + "grad_norm": 1.1840434074401855, + "learning_rate": 0.00041805804571940836, + "loss": 0.3894, + "step": 20892 + }, + { + "epoch": 1.167332458027209, + "grad_norm": 0.6642120480537415, + "learning_rate": 0.00041803003137606457, + "loss": 0.3126, + "step": 20893 + }, + { + "epoch": 1.1673883286308908, + "grad_norm": 1.7632179260253906, + "learning_rate": 0.0004180020170327208, + "loss": 0.3966, + "step": 20894 + }, + { + "epoch": 1.1674441992345728, + "grad_norm": 0.4146164059638977, + "learning_rate": 0.000417974002689377, + "loss": 0.4244, + "step": 20895 + }, + { + "epoch": 1.1675000698382547, + "grad_norm": 0.8305696249008179, + "learning_rate": 0.0004179459883460332, + "loss": 0.454, + "step": 20896 + }, + { + "epoch": 1.1675559404419364, + "grad_norm": 0.7836677432060242, + "learning_rate": 0.0004179179740026894, + "loss": 0.4291, + "step": 20897 + }, + { + "epoch": 1.1676118110456184, + "grad_norm": 0.8145248889923096, + "learning_rate": 0.0004178899596593456, + "loss": 0.3919, + "step": 20898 + }, + { + "epoch": 1.1676676816493001, + "grad_norm": 0.4881545901298523, + "learning_rate": 0.0004178619453160018, + "loss": 0.4199, + "step": 20899 + }, + { + "epoch": 1.167723552252982, + "grad_norm": 0.3498864471912384, + "learning_rate": 0.000417833930972658, + "loss": 0.5492, + "step": 20900 + }, + { + "epoch": 1.167779422856664, + "grad_norm": 0.33714261651039124, + "learning_rate": 0.0004178059166293142, + "loss": 0.3885, + "step": 20901 + }, + { + "epoch": 1.1678352934603458, + "grad_norm": 0.4431227147579193, + "learning_rate": 0.0004177779022859704, + "loss": 0.3954, + "step": 20902 + }, + { + "epoch": 1.1678911640640277, + "grad_norm": 0.5165025591850281, + "learning_rate": 0.0004177498879426266, + "loss": 0.5195, + "step": 20903 + }, + { + "epoch": 1.1679470346677097, + "grad_norm": 1.524924635887146, + "learning_rate": 0.00041772187359928283, + "loss": 0.4049, + "step": 20904 + }, + { + "epoch": 1.1680029052713914, + "grad_norm": 0.5299085974693298, + "learning_rate": 0.00041769385925593904, + "loss": 0.4512, + "step": 20905 + }, + { + "epoch": 1.1680587758750733, + "grad_norm": 0.4551352262496948, + "learning_rate": 0.00041766584491259524, + "loss": 0.4047, + "step": 20906 + }, + { + "epoch": 1.1681146464787553, + "grad_norm": 0.34013140201568604, + "learning_rate": 0.00041763783056925145, + "loss": 0.3373, + "step": 20907 + }, + { + "epoch": 1.168170517082437, + "grad_norm": 0.8094983100891113, + "learning_rate": 0.00041760981622590766, + "loss": 0.3384, + "step": 20908 + }, + { + "epoch": 1.168226387686119, + "grad_norm": 0.909254252910614, + "learning_rate": 0.0004175818018825639, + "loss": 0.6067, + "step": 20909 + }, + { + "epoch": 1.1682822582898007, + "grad_norm": 0.48185020685195923, + "learning_rate": 0.00041755378753922007, + "loss": 0.3556, + "step": 20910 + }, + { + "epoch": 1.1683381288934827, + "grad_norm": 0.5074177980422974, + "learning_rate": 0.00041752577319587633, + "loss": 0.4104, + "step": 20911 + }, + { + "epoch": 1.1683939994971646, + "grad_norm": 0.473606675863266, + "learning_rate": 0.0004174977588525325, + "loss": 0.4351, + "step": 20912 + }, + { + "epoch": 1.1684498701008463, + "grad_norm": 0.5752854943275452, + "learning_rate": 0.00041746974450918874, + "loss": 0.4246, + "step": 20913 + }, + { + "epoch": 1.1685057407045283, + "grad_norm": 0.6673492789268494, + "learning_rate": 0.0004174417301658449, + "loss": 0.4752, + "step": 20914 + }, + { + "epoch": 1.1685616113082102, + "grad_norm": 1.006258249282837, + "learning_rate": 0.00041741371582250115, + "loss": 0.3081, + "step": 20915 + }, + { + "epoch": 1.168617481911892, + "grad_norm": 0.5450448393821716, + "learning_rate": 0.00041738570147915736, + "loss": 0.4001, + "step": 20916 + }, + { + "epoch": 1.168673352515574, + "grad_norm": 0.33110079169273376, + "learning_rate": 0.00041735768713581356, + "loss": 0.3991, + "step": 20917 + }, + { + "epoch": 1.1687292231192559, + "grad_norm": 0.5840045809745789, + "learning_rate": 0.00041732967279246977, + "loss": 0.6241, + "step": 20918 + }, + { + "epoch": 1.1687850937229376, + "grad_norm": 1.1450505256652832, + "learning_rate": 0.000417301658449126, + "loss": 0.4271, + "step": 20919 + }, + { + "epoch": 1.1688409643266195, + "grad_norm": 0.45537737011909485, + "learning_rate": 0.0004172736441057822, + "loss": 0.3455, + "step": 20920 + }, + { + "epoch": 1.1688968349303015, + "grad_norm": 0.3958534300327301, + "learning_rate": 0.00041724562976243833, + "loss": 0.3207, + "step": 20921 + }, + { + "epoch": 1.1689527055339832, + "grad_norm": 0.3922674059867859, + "learning_rate": 0.0004172176154190946, + "loss": 0.395, + "step": 20922 + }, + { + "epoch": 1.1690085761376652, + "grad_norm": 0.4707679748535156, + "learning_rate": 0.00041718960107575074, + "loss": 0.5172, + "step": 20923 + }, + { + "epoch": 1.1690644467413471, + "grad_norm": 0.48344358801841736, + "learning_rate": 0.000417161586732407, + "loss": 0.468, + "step": 20924 + }, + { + "epoch": 1.1691203173450289, + "grad_norm": 1.2754100561141968, + "learning_rate": 0.0004171335723890632, + "loss": 0.4354, + "step": 20925 + }, + { + "epoch": 1.1691761879487108, + "grad_norm": 0.7674077749252319, + "learning_rate": 0.0004171055580457194, + "loss": 0.457, + "step": 20926 + }, + { + "epoch": 1.1692320585523928, + "grad_norm": 0.444220632314682, + "learning_rate": 0.0004170775437023756, + "loss": 0.5125, + "step": 20927 + }, + { + "epoch": 1.1692879291560745, + "grad_norm": 0.913804829120636, + "learning_rate": 0.0004170495293590318, + "loss": 0.6981, + "step": 20928 + }, + { + "epoch": 1.1693437997597564, + "grad_norm": 1.192964792251587, + "learning_rate": 0.00041702151501568803, + "loss": 0.4877, + "step": 20929 + }, + { + "epoch": 1.1693996703634384, + "grad_norm": 1.2667412757873535, + "learning_rate": 0.00041699350067234424, + "loss": 0.4577, + "step": 20930 + }, + { + "epoch": 1.1694555409671201, + "grad_norm": 0.4244190752506256, + "learning_rate": 0.00041696548632900044, + "loss": 0.5164, + "step": 20931 + }, + { + "epoch": 1.169511411570802, + "grad_norm": 0.6122274994850159, + "learning_rate": 0.0004169374719856567, + "loss": 0.6678, + "step": 20932 + }, + { + "epoch": 1.1695672821744838, + "grad_norm": 0.5745131969451904, + "learning_rate": 0.00041690945764231286, + "loss": 0.3497, + "step": 20933 + }, + { + "epoch": 1.1696231527781658, + "grad_norm": 1.6979825496673584, + "learning_rate": 0.0004168814432989691, + "loss": 0.4391, + "step": 20934 + }, + { + "epoch": 1.1696790233818477, + "grad_norm": 0.36939409375190735, + "learning_rate": 0.00041685342895562527, + "loss": 0.3562, + "step": 20935 + }, + { + "epoch": 1.1697348939855294, + "grad_norm": 0.3616001307964325, + "learning_rate": 0.00041682541461228153, + "loss": 0.4192, + "step": 20936 + }, + { + "epoch": 1.1697907645892114, + "grad_norm": 0.9177143573760986, + "learning_rate": 0.0004167974002689377, + "loss": 0.369, + "step": 20937 + }, + { + "epoch": 1.1698466351928933, + "grad_norm": 0.441985547542572, + "learning_rate": 0.00041676938592559394, + "loss": 0.3996, + "step": 20938 + }, + { + "epoch": 1.169902505796575, + "grad_norm": 0.35396796464920044, + "learning_rate": 0.0004167413715822501, + "loss": 0.378, + "step": 20939 + }, + { + "epoch": 1.169958376400257, + "grad_norm": 0.3953338861465454, + "learning_rate": 0.00041671335723890635, + "loss": 0.3976, + "step": 20940 + }, + { + "epoch": 1.170014247003939, + "grad_norm": 0.9806121587753296, + "learning_rate": 0.00041668534289556256, + "loss": 0.5072, + "step": 20941 + }, + { + "epoch": 1.1700701176076207, + "grad_norm": 0.3748386800289154, + "learning_rate": 0.0004166573285522187, + "loss": 0.4714, + "step": 20942 + }, + { + "epoch": 1.1701259882113026, + "grad_norm": 0.5787975192070007, + "learning_rate": 0.00041662931420887497, + "loss": 0.4059, + "step": 20943 + }, + { + "epoch": 1.1701818588149844, + "grad_norm": 0.9557050466537476, + "learning_rate": 0.0004166012998655311, + "loss": 0.4918, + "step": 20944 + }, + { + "epoch": 1.1702377294186663, + "grad_norm": 0.6039550304412842, + "learning_rate": 0.0004165732855221874, + "loss": 0.421, + "step": 20945 + }, + { + "epoch": 1.1702936000223483, + "grad_norm": 1.6680552959442139, + "learning_rate": 0.00041654527117884353, + "loss": 0.3637, + "step": 20946 + }, + { + "epoch": 1.17034947062603, + "grad_norm": 0.6607884764671326, + "learning_rate": 0.0004165172568354998, + "loss": 0.4344, + "step": 20947 + }, + { + "epoch": 1.170405341229712, + "grad_norm": 0.5056108236312866, + "learning_rate": 0.000416489242492156, + "loss": 0.3829, + "step": 20948 + }, + { + "epoch": 1.170461211833394, + "grad_norm": 0.48238107562065125, + "learning_rate": 0.0004164612281488122, + "loss": 0.4661, + "step": 20949 + }, + { + "epoch": 1.1705170824370756, + "grad_norm": 0.6004096269607544, + "learning_rate": 0.0004164332138054684, + "loss": 0.4726, + "step": 20950 + }, + { + "epoch": 1.1705729530407576, + "grad_norm": 0.6912716031074524, + "learning_rate": 0.0004164051994621246, + "loss": 0.4377, + "step": 20951 + }, + { + "epoch": 1.1706288236444395, + "grad_norm": 0.5216689705848694, + "learning_rate": 0.0004163771851187808, + "loss": 0.4374, + "step": 20952 + }, + { + "epoch": 1.1706846942481213, + "grad_norm": 0.2882271409034729, + "learning_rate": 0.000416349170775437, + "loss": 0.3557, + "step": 20953 + }, + { + "epoch": 1.1707405648518032, + "grad_norm": 0.9510630369186401, + "learning_rate": 0.00041632115643209323, + "loss": 0.3813, + "step": 20954 + }, + { + "epoch": 1.1707964354554852, + "grad_norm": 0.7968623042106628, + "learning_rate": 0.0004162931420887495, + "loss": 0.3914, + "step": 20955 + }, + { + "epoch": 1.170852306059167, + "grad_norm": 0.38662877678871155, + "learning_rate": 0.00041626512774540564, + "loss": 0.3717, + "step": 20956 + }, + { + "epoch": 1.1709081766628489, + "grad_norm": 0.5155742168426514, + "learning_rate": 0.0004162371134020619, + "loss": 0.4449, + "step": 20957 + }, + { + "epoch": 1.1709640472665308, + "grad_norm": 0.6989040970802307, + "learning_rate": 0.00041620909905871806, + "loss": 0.4318, + "step": 20958 + }, + { + "epoch": 1.1710199178702125, + "grad_norm": 1.8501482009887695, + "learning_rate": 0.0004161810847153743, + "loss": 0.3635, + "step": 20959 + }, + { + "epoch": 1.1710757884738945, + "grad_norm": 0.4445251524448395, + "learning_rate": 0.00041615307037203047, + "loss": 0.3252, + "step": 20960 + }, + { + "epoch": 1.1711316590775764, + "grad_norm": 0.3902353346347809, + "learning_rate": 0.0004161250560286867, + "loss": 0.3727, + "step": 20961 + }, + { + "epoch": 1.1711875296812582, + "grad_norm": 0.5890637636184692, + "learning_rate": 0.0004160970416853429, + "loss": 0.3977, + "step": 20962 + }, + { + "epoch": 1.1712434002849401, + "grad_norm": 0.6632817387580872, + "learning_rate": 0.0004160690273419991, + "loss": 0.4679, + "step": 20963 + }, + { + "epoch": 1.171299270888622, + "grad_norm": 0.4736965298652649, + "learning_rate": 0.00041604101299865534, + "loss": 0.4712, + "step": 20964 + }, + { + "epoch": 1.1713551414923038, + "grad_norm": 1.1225155591964722, + "learning_rate": 0.0004160129986553115, + "loss": 0.3434, + "step": 20965 + }, + { + "epoch": 1.1714110120959857, + "grad_norm": 0.3322979509830475, + "learning_rate": 0.00041598498431196776, + "loss": 0.4219, + "step": 20966 + }, + { + "epoch": 1.1714668826996675, + "grad_norm": 0.7051735520362854, + "learning_rate": 0.0004159569699686239, + "loss": 0.4605, + "step": 20967 + }, + { + "epoch": 1.1715227533033494, + "grad_norm": 0.4983975291252136, + "learning_rate": 0.00041592895562528017, + "loss": 0.4088, + "step": 20968 + }, + { + "epoch": 1.1715786239070314, + "grad_norm": 1.4409910440444946, + "learning_rate": 0.0004159009412819363, + "loss": 0.4276, + "step": 20969 + }, + { + "epoch": 1.171634494510713, + "grad_norm": 0.40704649686813354, + "learning_rate": 0.0004158729269385926, + "loss": 0.4693, + "step": 20970 + }, + { + "epoch": 1.171690365114395, + "grad_norm": 0.41881540417671204, + "learning_rate": 0.0004158449125952488, + "loss": 0.5277, + "step": 20971 + }, + { + "epoch": 1.171746235718077, + "grad_norm": 0.7873241901397705, + "learning_rate": 0.000415816898251905, + "loss": 0.4296, + "step": 20972 + }, + { + "epoch": 1.1718021063217587, + "grad_norm": 1.4150946140289307, + "learning_rate": 0.0004157888839085612, + "loss": 0.721, + "step": 20973 + }, + { + "epoch": 1.1718579769254407, + "grad_norm": 0.5701517462730408, + "learning_rate": 0.0004157608695652174, + "loss": 0.6868, + "step": 20974 + }, + { + "epoch": 1.1719138475291226, + "grad_norm": 1.325251579284668, + "learning_rate": 0.0004157328552218736, + "loss": 0.4037, + "step": 20975 + }, + { + "epoch": 1.1719697181328044, + "grad_norm": 0.5655409693717957, + "learning_rate": 0.0004157048408785298, + "loss": 0.4513, + "step": 20976 + }, + { + "epoch": 1.1720255887364863, + "grad_norm": 1.8675804138183594, + "learning_rate": 0.000415676826535186, + "loss": 0.4627, + "step": 20977 + }, + { + "epoch": 1.172081459340168, + "grad_norm": 1.7151024341583252, + "learning_rate": 0.0004156488121918422, + "loss": 0.5259, + "step": 20978 + }, + { + "epoch": 1.17213732994385, + "grad_norm": 0.3808020353317261, + "learning_rate": 0.00041562079784849843, + "loss": 0.4525, + "step": 20979 + }, + { + "epoch": 1.172193200547532, + "grad_norm": 0.5010273456573486, + "learning_rate": 0.0004155927835051547, + "loss": 0.3747, + "step": 20980 + }, + { + "epoch": 1.1722490711512137, + "grad_norm": 0.3718937635421753, + "learning_rate": 0.00041556476916181084, + "loss": 0.332, + "step": 20981 + }, + { + "epoch": 1.1723049417548956, + "grad_norm": 0.34148839116096497, + "learning_rate": 0.0004155367548184671, + "loss": 0.3967, + "step": 20982 + }, + { + "epoch": 1.1723608123585776, + "grad_norm": 0.5877685546875, + "learning_rate": 0.00041550874047512325, + "loss": 0.5145, + "step": 20983 + }, + { + "epoch": 1.1724166829622593, + "grad_norm": 0.6465590000152588, + "learning_rate": 0.0004154807261317795, + "loss": 0.5653, + "step": 20984 + }, + { + "epoch": 1.1724725535659413, + "grad_norm": 1.0524135828018188, + "learning_rate": 0.00041545271178843567, + "loss": 0.4204, + "step": 20985 + }, + { + "epoch": 1.1725284241696232, + "grad_norm": 0.423750638961792, + "learning_rate": 0.00041542469744509187, + "loss": 0.4628, + "step": 20986 + }, + { + "epoch": 1.172584294773305, + "grad_norm": 0.5666519999504089, + "learning_rate": 0.00041539668310174813, + "loss": 0.51, + "step": 20987 + }, + { + "epoch": 1.172640165376987, + "grad_norm": 0.5450125932693481, + "learning_rate": 0.0004153686687584043, + "loss": 0.468, + "step": 20988 + }, + { + "epoch": 1.1726960359806688, + "grad_norm": 0.3017868995666504, + "learning_rate": 0.00041534065441506054, + "loss": 0.3657, + "step": 20989 + }, + { + "epoch": 1.1727519065843506, + "grad_norm": 0.4011910557746887, + "learning_rate": 0.0004153126400717167, + "loss": 0.4163, + "step": 20990 + }, + { + "epoch": 1.1728077771880325, + "grad_norm": 0.4581397771835327, + "learning_rate": 0.00041528462572837296, + "loss": 0.5333, + "step": 20991 + }, + { + "epoch": 1.1728636477917145, + "grad_norm": 0.7185795903205872, + "learning_rate": 0.0004152566113850291, + "loss": 0.4211, + "step": 20992 + }, + { + "epoch": 1.1729195183953962, + "grad_norm": 1.4021614789962769, + "learning_rate": 0.00041522859704168537, + "loss": 0.5976, + "step": 20993 + }, + { + "epoch": 1.1729753889990782, + "grad_norm": 0.5829489231109619, + "learning_rate": 0.0004152005826983415, + "loss": 0.5197, + "step": 20994 + }, + { + "epoch": 1.1730312596027601, + "grad_norm": 0.38423392176628113, + "learning_rate": 0.0004151725683549978, + "loss": 0.3514, + "step": 20995 + }, + { + "epoch": 1.1730871302064418, + "grad_norm": 0.8088028430938721, + "learning_rate": 0.000415144554011654, + "loss": 0.4929, + "step": 20996 + }, + { + "epoch": 1.1731430008101238, + "grad_norm": 0.3372558653354645, + "learning_rate": 0.0004151165396683102, + "loss": 0.3966, + "step": 20997 + }, + { + "epoch": 1.1731988714138057, + "grad_norm": 0.4188452661037445, + "learning_rate": 0.0004150885253249664, + "loss": 0.4091, + "step": 20998 + }, + { + "epoch": 1.1732547420174875, + "grad_norm": 0.386389821767807, + "learning_rate": 0.0004150605109816226, + "loss": 0.3766, + "step": 20999 + }, + { + "epoch": 1.1733106126211694, + "grad_norm": 1.9423644542694092, + "learning_rate": 0.0004150324966382788, + "loss": 0.4532, + "step": 21000 + }, + { + "epoch": 1.1733106126211694, + "eval_cer": 0.08676202633847228, + "eval_loss": 0.3263513147830963, + "eval_runtime": 56.0878, + "eval_samples_per_second": 80.909, + "eval_steps_per_second": 5.063, + "eval_wer": 0.3442641267003715, + "step": 21000 + }, + { + "epoch": 1.1733664832248512, + "grad_norm": 0.4116101861000061, + "learning_rate": 0.000415004482294935, + "loss": 0.4176, + "step": 21001 + }, + { + "epoch": 1.173422353828533, + "grad_norm": 0.46282145380973816, + "learning_rate": 0.0004149764679515912, + "loss": 0.4148, + "step": 21002 + }, + { + "epoch": 1.173478224432215, + "grad_norm": 0.721747636795044, + "learning_rate": 0.0004149484536082475, + "loss": 0.4696, + "step": 21003 + }, + { + "epoch": 1.1735340950358968, + "grad_norm": 0.4077196419239044, + "learning_rate": 0.00041492043926490363, + "loss": 0.3764, + "step": 21004 + }, + { + "epoch": 1.1735899656395787, + "grad_norm": 2.356966495513916, + "learning_rate": 0.0004148924249215599, + "loss": 0.4212, + "step": 21005 + }, + { + "epoch": 1.1736458362432607, + "grad_norm": 0.3053162395954132, + "learning_rate": 0.00041486441057821604, + "loss": 0.3157, + "step": 21006 + }, + { + "epoch": 1.1737017068469424, + "grad_norm": 0.41734546422958374, + "learning_rate": 0.00041483639623487225, + "loss": 0.5528, + "step": 21007 + }, + { + "epoch": 1.1737575774506244, + "grad_norm": 0.37933266162872314, + "learning_rate": 0.00041480838189152845, + "loss": 0.3795, + "step": 21008 + }, + { + "epoch": 1.1738134480543063, + "grad_norm": 0.6062071919441223, + "learning_rate": 0.00041478036754818466, + "loss": 0.4832, + "step": 21009 + }, + { + "epoch": 1.173869318657988, + "grad_norm": 0.3543606102466583, + "learning_rate": 0.0004147523532048409, + "loss": 0.4286, + "step": 21010 + }, + { + "epoch": 1.17392518926167, + "grad_norm": 1.1411341428756714, + "learning_rate": 0.00041472433886149707, + "loss": 0.4234, + "step": 21011 + }, + { + "epoch": 1.1739810598653517, + "grad_norm": 0.4244077503681183, + "learning_rate": 0.00041469632451815333, + "loss": 0.4449, + "step": 21012 + }, + { + "epoch": 1.1740369304690337, + "grad_norm": 0.36268746852874756, + "learning_rate": 0.0004146683101748095, + "loss": 0.3839, + "step": 21013 + }, + { + "epoch": 1.1740928010727156, + "grad_norm": 0.48378613591194153, + "learning_rate": 0.00041464029583146574, + "loss": 0.4256, + "step": 21014 + }, + { + "epoch": 1.1741486716763974, + "grad_norm": 0.3160778284072876, + "learning_rate": 0.0004146122814881219, + "loss": 0.375, + "step": 21015 + }, + { + "epoch": 1.1742045422800793, + "grad_norm": 0.45634791254997253, + "learning_rate": 0.00041458426714477815, + "loss": 0.3017, + "step": 21016 + }, + { + "epoch": 1.1742604128837613, + "grad_norm": 0.37480485439300537, + "learning_rate": 0.0004145562528014343, + "loss": 0.4059, + "step": 21017 + }, + { + "epoch": 1.174316283487443, + "grad_norm": 0.4308887720108032, + "learning_rate": 0.00041452823845809057, + "loss": 0.398, + "step": 21018 + }, + { + "epoch": 1.174372154091125, + "grad_norm": 0.4243583381175995, + "learning_rate": 0.00041450022411474677, + "loss": 0.4748, + "step": 21019 + }, + { + "epoch": 1.174428024694807, + "grad_norm": 0.34599217772483826, + "learning_rate": 0.000414472209771403, + "loss": 0.3231, + "step": 21020 + }, + { + "epoch": 1.1744838952984886, + "grad_norm": 0.3507658839225769, + "learning_rate": 0.0004144441954280592, + "loss": 0.4497, + "step": 21021 + }, + { + "epoch": 1.1745397659021706, + "grad_norm": 0.409562885761261, + "learning_rate": 0.0004144161810847154, + "loss": 0.3624, + "step": 21022 + }, + { + "epoch": 1.1745956365058525, + "grad_norm": 0.7308661341667175, + "learning_rate": 0.0004143881667413716, + "loss": 0.3887, + "step": 21023 + }, + { + "epoch": 1.1746515071095343, + "grad_norm": 1.665435791015625, + "learning_rate": 0.0004143601523980278, + "loss": 0.3706, + "step": 21024 + }, + { + "epoch": 1.1747073777132162, + "grad_norm": 0.37854501605033875, + "learning_rate": 0.000414332138054684, + "loss": 0.463, + "step": 21025 + }, + { + "epoch": 1.1747632483168982, + "grad_norm": 0.49657538533210754, + "learning_rate": 0.00041430412371134027, + "loss": 0.4269, + "step": 21026 + }, + { + "epoch": 1.1748191189205799, + "grad_norm": 0.4403928220272064, + "learning_rate": 0.0004142761093679964, + "loss": 0.4779, + "step": 21027 + }, + { + "epoch": 1.1748749895242618, + "grad_norm": 0.37407106161117554, + "learning_rate": 0.0004142480950246526, + "loss": 0.3392, + "step": 21028 + }, + { + "epoch": 1.1749308601279438, + "grad_norm": 0.466636061668396, + "learning_rate": 0.00041422008068130883, + "loss": 0.3339, + "step": 21029 + }, + { + "epoch": 1.1749867307316255, + "grad_norm": 0.43736761808395386, + "learning_rate": 0.00041419206633796504, + "loss": 0.4598, + "step": 21030 + }, + { + "epoch": 1.1750426013353075, + "grad_norm": 0.5083590745925903, + "learning_rate": 0.00041416405199462124, + "loss": 0.5194, + "step": 21031 + }, + { + "epoch": 1.1750984719389894, + "grad_norm": 0.8163910508155823, + "learning_rate": 0.00041413603765127745, + "loss": 0.468, + "step": 21032 + }, + { + "epoch": 1.1751543425426711, + "grad_norm": 0.4659527540206909, + "learning_rate": 0.00041410802330793365, + "loss": 0.4087, + "step": 21033 + }, + { + "epoch": 1.175210213146353, + "grad_norm": 0.4124884307384491, + "learning_rate": 0.00041408000896458986, + "loss": 0.4467, + "step": 21034 + }, + { + "epoch": 1.1752660837500348, + "grad_norm": 0.6270312666893005, + "learning_rate": 0.0004140519946212461, + "loss": 0.3781, + "step": 21035 + }, + { + "epoch": 1.1753219543537168, + "grad_norm": 0.6443387866020203, + "learning_rate": 0.00041402398027790227, + "loss": 0.5847, + "step": 21036 + }, + { + "epoch": 1.1753778249573987, + "grad_norm": 0.4103495478630066, + "learning_rate": 0.00041399596593455853, + "loss": 0.3904, + "step": 21037 + }, + { + "epoch": 1.1754336955610805, + "grad_norm": 0.9441747665405273, + "learning_rate": 0.0004139679515912147, + "loss": 0.3934, + "step": 21038 + }, + { + "epoch": 1.1754895661647624, + "grad_norm": 0.32512691617012024, + "learning_rate": 0.00041393993724787094, + "loss": 0.4228, + "step": 21039 + }, + { + "epoch": 1.1755454367684444, + "grad_norm": 0.7858197093009949, + "learning_rate": 0.0004139119229045271, + "loss": 0.4418, + "step": 21040 + }, + { + "epoch": 1.175601307372126, + "grad_norm": 0.4379258155822754, + "learning_rate": 0.00041388390856118335, + "loss": 0.4226, + "step": 21041 + }, + { + "epoch": 1.175657177975808, + "grad_norm": 0.49117860198020935, + "learning_rate": 0.00041385589421783956, + "loss": 0.4736, + "step": 21042 + }, + { + "epoch": 1.17571304857949, + "grad_norm": 0.5344110727310181, + "learning_rate": 0.00041382787987449577, + "loss": 0.4163, + "step": 21043 + }, + { + "epoch": 1.1757689191831717, + "grad_norm": 0.3326652944087982, + "learning_rate": 0.00041379986553115197, + "loss": 0.3314, + "step": 21044 + }, + { + "epoch": 1.1758247897868537, + "grad_norm": 0.4426407516002655, + "learning_rate": 0.0004137718511878082, + "loss": 0.3674, + "step": 21045 + }, + { + "epoch": 1.1758806603905354, + "grad_norm": 0.5659340620040894, + "learning_rate": 0.0004137438368444644, + "loss": 0.4699, + "step": 21046 + }, + { + "epoch": 1.1759365309942174, + "grad_norm": 1.4919838905334473, + "learning_rate": 0.0004137158225011206, + "loss": 0.4741, + "step": 21047 + }, + { + "epoch": 1.1759924015978993, + "grad_norm": 0.4591537117958069, + "learning_rate": 0.0004136878081577768, + "loss": 0.3426, + "step": 21048 + }, + { + "epoch": 1.176048272201581, + "grad_norm": 0.5882066488265991, + "learning_rate": 0.00041365979381443295, + "loss": 0.382, + "step": 21049 + }, + { + "epoch": 1.176104142805263, + "grad_norm": 0.40349259972572327, + "learning_rate": 0.0004136317794710892, + "loss": 0.4139, + "step": 21050 + }, + { + "epoch": 1.176160013408945, + "grad_norm": 0.39566347002983093, + "learning_rate": 0.0004136037651277454, + "loss": 0.3707, + "step": 21051 + }, + { + "epoch": 1.1762158840126267, + "grad_norm": 2.6094813346862793, + "learning_rate": 0.0004135757507844016, + "loss": 0.387, + "step": 21052 + }, + { + "epoch": 1.1762717546163086, + "grad_norm": 0.5529335141181946, + "learning_rate": 0.0004135477364410578, + "loss": 0.5347, + "step": 21053 + }, + { + "epoch": 1.1763276252199906, + "grad_norm": 0.8971514105796814, + "learning_rate": 0.00041351972209771403, + "loss": 0.4493, + "step": 21054 + }, + { + "epoch": 1.1763834958236723, + "grad_norm": 0.47371241450309753, + "learning_rate": 0.00041349170775437024, + "loss": 0.5437, + "step": 21055 + }, + { + "epoch": 1.1764393664273542, + "grad_norm": 1.265568494796753, + "learning_rate": 0.00041346369341102644, + "loss": 0.4325, + "step": 21056 + }, + { + "epoch": 1.1764952370310362, + "grad_norm": 0.3908389210700989, + "learning_rate": 0.00041343567906768265, + "loss": 0.4473, + "step": 21057 + }, + { + "epoch": 1.176551107634718, + "grad_norm": 0.6488955616950989, + "learning_rate": 0.0004134076647243389, + "loss": 0.4199, + "step": 21058 + }, + { + "epoch": 1.1766069782383999, + "grad_norm": 0.426641047000885, + "learning_rate": 0.00041337965038099506, + "loss": 0.4384, + "step": 21059 + }, + { + "epoch": 1.1766628488420818, + "grad_norm": 0.5512905716896057, + "learning_rate": 0.0004133516360376513, + "loss": 0.4871, + "step": 21060 + }, + { + "epoch": 1.1767187194457636, + "grad_norm": 0.38511550426483154, + "learning_rate": 0.00041332362169430747, + "loss": 0.402, + "step": 21061 + }, + { + "epoch": 1.1767745900494455, + "grad_norm": 4.504671573638916, + "learning_rate": 0.00041329560735096373, + "loss": 0.4819, + "step": 21062 + }, + { + "epoch": 1.1768304606531275, + "grad_norm": 0.6879104375839233, + "learning_rate": 0.0004132675930076199, + "loss": 0.4627, + "step": 21063 + }, + { + "epoch": 1.1768863312568092, + "grad_norm": 0.5197402238845825, + "learning_rate": 0.00041323957866427614, + "loss": 0.5645, + "step": 21064 + }, + { + "epoch": 1.1769422018604911, + "grad_norm": 0.36264267563819885, + "learning_rate": 0.0004132115643209323, + "loss": 0.3658, + "step": 21065 + }, + { + "epoch": 1.176998072464173, + "grad_norm": 0.6212230324745178, + "learning_rate": 0.00041318354997758855, + "loss": 0.4546, + "step": 21066 + }, + { + "epoch": 1.1770539430678548, + "grad_norm": 0.34389805793762207, + "learning_rate": 0.00041315553563424476, + "loss": 0.4373, + "step": 21067 + }, + { + "epoch": 1.1771098136715368, + "grad_norm": 0.8442031741142273, + "learning_rate": 0.00041312752129090097, + "loss": 0.4681, + "step": 21068 + }, + { + "epoch": 1.1771656842752185, + "grad_norm": 0.3647095561027527, + "learning_rate": 0.00041309950694755717, + "loss": 0.427, + "step": 21069 + }, + { + "epoch": 1.1772215548789005, + "grad_norm": 1.2599656581878662, + "learning_rate": 0.0004130714926042133, + "loss": 0.4298, + "step": 21070 + }, + { + "epoch": 1.1772774254825824, + "grad_norm": 0.35133230686187744, + "learning_rate": 0.0004130434782608696, + "loss": 0.4342, + "step": 21071 + }, + { + "epoch": 1.1773332960862641, + "grad_norm": 1.438297986984253, + "learning_rate": 0.00041301546391752573, + "loss": 0.4886, + "step": 21072 + }, + { + "epoch": 1.177389166689946, + "grad_norm": 0.9779869318008423, + "learning_rate": 0.000412987449574182, + "loss": 0.5477, + "step": 21073 + }, + { + "epoch": 1.177445037293628, + "grad_norm": 0.38519906997680664, + "learning_rate": 0.0004129594352308382, + "loss": 0.57, + "step": 21074 + }, + { + "epoch": 1.1775009078973098, + "grad_norm": 0.3895972669124603, + "learning_rate": 0.0004129314208874944, + "loss": 0.442, + "step": 21075 + }, + { + "epoch": 1.1775567785009917, + "grad_norm": 1.4724907875061035, + "learning_rate": 0.0004129034065441506, + "loss": 0.423, + "step": 21076 + }, + { + "epoch": 1.1776126491046737, + "grad_norm": 0.49910229444503784, + "learning_rate": 0.0004128753922008068, + "loss": 0.6968, + "step": 21077 + }, + { + "epoch": 1.1776685197083554, + "grad_norm": 0.38375893235206604, + "learning_rate": 0.000412847377857463, + "loss": 0.4153, + "step": 21078 + }, + { + "epoch": 1.1777243903120374, + "grad_norm": 0.5524494647979736, + "learning_rate": 0.00041281936351411923, + "loss": 0.4738, + "step": 21079 + }, + { + "epoch": 1.177780260915719, + "grad_norm": 0.3931167721748352, + "learning_rate": 0.00041279134917077544, + "loss": 0.4441, + "step": 21080 + }, + { + "epoch": 1.177836131519401, + "grad_norm": 0.43682950735092163, + "learning_rate": 0.0004127633348274317, + "loss": 0.6724, + "step": 21081 + }, + { + "epoch": 1.177892002123083, + "grad_norm": 0.3174096643924713, + "learning_rate": 0.00041273532048408785, + "loss": 0.3549, + "step": 21082 + }, + { + "epoch": 1.1779478727267647, + "grad_norm": 0.4600127339363098, + "learning_rate": 0.0004127073061407441, + "loss": 0.5535, + "step": 21083 + }, + { + "epoch": 1.1780037433304467, + "grad_norm": 0.36205777525901794, + "learning_rate": 0.00041267929179740026, + "loss": 0.3718, + "step": 21084 + }, + { + "epoch": 1.1780596139341286, + "grad_norm": 0.5744758248329163, + "learning_rate": 0.0004126512774540565, + "loss": 0.3591, + "step": 21085 + }, + { + "epoch": 1.1781154845378103, + "grad_norm": 0.5685778856277466, + "learning_rate": 0.00041262326311071267, + "loss": 0.4758, + "step": 21086 + }, + { + "epoch": 1.1781713551414923, + "grad_norm": 0.42402902245521545, + "learning_rate": 0.00041259524876736893, + "loss": 0.5007, + "step": 21087 + }, + { + "epoch": 1.1782272257451742, + "grad_norm": 0.7443286776542664, + "learning_rate": 0.0004125672344240251, + "loss": 0.4421, + "step": 21088 + }, + { + "epoch": 1.178283096348856, + "grad_norm": 0.4900764226913452, + "learning_rate": 0.00041253922008068134, + "loss": 0.3999, + "step": 21089 + }, + { + "epoch": 1.178338966952538, + "grad_norm": 0.43612587451934814, + "learning_rate": 0.00041251120573733755, + "loss": 0.4289, + "step": 21090 + }, + { + "epoch": 1.1783948375562199, + "grad_norm": 0.38370487093925476, + "learning_rate": 0.0004124831913939937, + "loss": 0.4247, + "step": 21091 + }, + { + "epoch": 1.1784507081599016, + "grad_norm": 0.5446457266807556, + "learning_rate": 0.00041245517705064996, + "loss": 0.4203, + "step": 21092 + }, + { + "epoch": 1.1785065787635836, + "grad_norm": 0.5094758868217468, + "learning_rate": 0.0004124271627073061, + "loss": 0.5424, + "step": 21093 + }, + { + "epoch": 1.1785624493672655, + "grad_norm": 0.7764329314231873, + "learning_rate": 0.00041239914836396237, + "loss": 0.3826, + "step": 21094 + }, + { + "epoch": 1.1786183199709472, + "grad_norm": 0.4039691686630249, + "learning_rate": 0.0004123711340206185, + "loss": 0.3563, + "step": 21095 + }, + { + "epoch": 1.1786741905746292, + "grad_norm": 1.2114931344985962, + "learning_rate": 0.0004123431196772748, + "loss": 0.4853, + "step": 21096 + }, + { + "epoch": 1.1787300611783111, + "grad_norm": 0.48257124423980713, + "learning_rate": 0.000412315105333931, + "loss": 0.5156, + "step": 21097 + }, + { + "epoch": 1.1787859317819929, + "grad_norm": 0.41901084780693054, + "learning_rate": 0.0004122870909905872, + "loss": 0.3128, + "step": 21098 + }, + { + "epoch": 1.1788418023856748, + "grad_norm": 0.49611184000968933, + "learning_rate": 0.0004122590766472434, + "loss": 0.5235, + "step": 21099 + }, + { + "epoch": 1.1788976729893568, + "grad_norm": 0.5395388603210449, + "learning_rate": 0.0004122310623038996, + "loss": 0.3478, + "step": 21100 + }, + { + "epoch": 1.1789535435930385, + "grad_norm": 0.7097464203834534, + "learning_rate": 0.0004122030479605558, + "loss": 0.525, + "step": 21101 + }, + { + "epoch": 1.1790094141967205, + "grad_norm": 0.6533705592155457, + "learning_rate": 0.000412175033617212, + "loss": 0.4385, + "step": 21102 + }, + { + "epoch": 1.1790652848004022, + "grad_norm": 0.38403192162513733, + "learning_rate": 0.0004121470192738682, + "loss": 0.5279, + "step": 21103 + }, + { + "epoch": 1.1791211554040841, + "grad_norm": 0.46363890171051025, + "learning_rate": 0.00041211900493052443, + "loss": 0.3989, + "step": 21104 + }, + { + "epoch": 1.179177026007766, + "grad_norm": 0.873522937297821, + "learning_rate": 0.00041209099058718064, + "loss": 0.368, + "step": 21105 + }, + { + "epoch": 1.1792328966114478, + "grad_norm": 0.4265742003917694, + "learning_rate": 0.0004120629762438369, + "loss": 0.5371, + "step": 21106 + }, + { + "epoch": 1.1792887672151298, + "grad_norm": 0.36898669600486755, + "learning_rate": 0.00041203496190049305, + "loss": 0.3031, + "step": 21107 + }, + { + "epoch": 1.1793446378188117, + "grad_norm": 0.3947764039039612, + "learning_rate": 0.0004120069475571493, + "loss": 0.3857, + "step": 21108 + }, + { + "epoch": 1.1794005084224934, + "grad_norm": 0.4106178879737854, + "learning_rate": 0.00041197893321380546, + "loss": 0.4069, + "step": 21109 + }, + { + "epoch": 1.1794563790261754, + "grad_norm": 0.4608002007007599, + "learning_rate": 0.0004119509188704617, + "loss": 0.3774, + "step": 21110 + }, + { + "epoch": 1.1795122496298573, + "grad_norm": 1.0271695852279663, + "learning_rate": 0.00041192290452711787, + "loss": 0.4746, + "step": 21111 + }, + { + "epoch": 1.179568120233539, + "grad_norm": 0.5484672784805298, + "learning_rate": 0.0004118948901837741, + "loss": 0.3387, + "step": 21112 + }, + { + "epoch": 1.179623990837221, + "grad_norm": 0.49107545614242554, + "learning_rate": 0.00041186687584043034, + "loss": 0.4579, + "step": 21113 + }, + { + "epoch": 1.1796798614409028, + "grad_norm": 0.42124927043914795, + "learning_rate": 0.0004118388614970865, + "loss": 0.4582, + "step": 21114 + }, + { + "epoch": 1.1797357320445847, + "grad_norm": 1.0690102577209473, + "learning_rate": 0.00041181084715374275, + "loss": 0.5212, + "step": 21115 + }, + { + "epoch": 1.1797916026482667, + "grad_norm": 0.49727416038513184, + "learning_rate": 0.0004117828328103989, + "loss": 0.5656, + "step": 21116 + }, + { + "epoch": 1.1798474732519484, + "grad_norm": 0.4060685336589813, + "learning_rate": 0.00041175481846705516, + "loss": 0.4366, + "step": 21117 + }, + { + "epoch": 1.1799033438556303, + "grad_norm": 0.5009297132492065, + "learning_rate": 0.0004117268041237113, + "loss": 0.4695, + "step": 21118 + }, + { + "epoch": 1.1799592144593123, + "grad_norm": 0.5368232131004333, + "learning_rate": 0.00041169878978036757, + "loss": 0.4844, + "step": 21119 + }, + { + "epoch": 1.180015085062994, + "grad_norm": 0.4437754452228546, + "learning_rate": 0.0004116707754370237, + "loss": 0.4187, + "step": 21120 + }, + { + "epoch": 1.180070955666676, + "grad_norm": 0.6504024267196655, + "learning_rate": 0.00041164276109368, + "loss": 0.3707, + "step": 21121 + }, + { + "epoch": 1.180126826270358, + "grad_norm": 0.5466588735580444, + "learning_rate": 0.0004116147467503362, + "loss": 0.4115, + "step": 21122 + }, + { + "epoch": 1.1801826968740396, + "grad_norm": 0.6516076922416687, + "learning_rate": 0.0004115867324069924, + "loss": 0.3633, + "step": 21123 + }, + { + "epoch": 1.1802385674777216, + "grad_norm": 0.46513885259628296, + "learning_rate": 0.0004115587180636486, + "loss": 0.4797, + "step": 21124 + }, + { + "epoch": 1.1802944380814036, + "grad_norm": 0.397909015417099, + "learning_rate": 0.0004115307037203048, + "loss": 0.4919, + "step": 21125 + }, + { + "epoch": 1.1803503086850853, + "grad_norm": 0.4465582072734833, + "learning_rate": 0.000411502689376961, + "loss": 0.5013, + "step": 21126 + }, + { + "epoch": 1.1804061792887672, + "grad_norm": 0.47273269295692444, + "learning_rate": 0.0004114746750336172, + "loss": 0.3818, + "step": 21127 + }, + { + "epoch": 1.1804620498924492, + "grad_norm": 0.5441597700119019, + "learning_rate": 0.0004114466606902734, + "loss": 0.483, + "step": 21128 + }, + { + "epoch": 1.180517920496131, + "grad_norm": 0.44469133019447327, + "learning_rate": 0.0004114186463469297, + "loss": 0.4638, + "step": 21129 + }, + { + "epoch": 1.1805737910998129, + "grad_norm": 0.37609848380088806, + "learning_rate": 0.00041139063200358583, + "loss": 0.4752, + "step": 21130 + }, + { + "epoch": 1.1806296617034948, + "grad_norm": 0.38735252618789673, + "learning_rate": 0.0004113626176602421, + "loss": 0.5217, + "step": 21131 + }, + { + "epoch": 1.1806855323071765, + "grad_norm": 0.4454619884490967, + "learning_rate": 0.00041133460331689825, + "loss": 0.4032, + "step": 21132 + }, + { + "epoch": 1.1807414029108585, + "grad_norm": 0.5060156583786011, + "learning_rate": 0.00041130658897355445, + "loss": 0.4192, + "step": 21133 + }, + { + "epoch": 1.1807972735145404, + "grad_norm": 0.45684677362442017, + "learning_rate": 0.00041127857463021066, + "loss": 0.4285, + "step": 21134 + }, + { + "epoch": 1.1808531441182222, + "grad_norm": 0.3573480248451233, + "learning_rate": 0.00041125056028686686, + "loss": 0.3585, + "step": 21135 + }, + { + "epoch": 1.1809090147219041, + "grad_norm": 0.5558732151985168, + "learning_rate": 0.00041122254594352307, + "loss": 0.4993, + "step": 21136 + }, + { + "epoch": 1.1809648853255859, + "grad_norm": 0.5300498008728027, + "learning_rate": 0.0004111945316001793, + "loss": 0.3407, + "step": 21137 + }, + { + "epoch": 1.1810207559292678, + "grad_norm": 1.4076348543167114, + "learning_rate": 0.00041116651725683554, + "loss": 0.3579, + "step": 21138 + }, + { + "epoch": 1.1810766265329498, + "grad_norm": 0.36906397342681885, + "learning_rate": 0.0004111385029134917, + "loss": 0.4398, + "step": 21139 + }, + { + "epoch": 1.1811324971366315, + "grad_norm": 0.5668591856956482, + "learning_rate": 0.00041111048857014795, + "loss": 0.4838, + "step": 21140 + }, + { + "epoch": 1.1811883677403134, + "grad_norm": 3.2791216373443604, + "learning_rate": 0.0004110824742268041, + "loss": 0.3796, + "step": 21141 + }, + { + "epoch": 1.1812442383439954, + "grad_norm": 0.5450183153152466, + "learning_rate": 0.00041105445988346036, + "loss": 0.4535, + "step": 21142 + }, + { + "epoch": 1.1813001089476771, + "grad_norm": 0.5707497596740723, + "learning_rate": 0.0004110264455401165, + "loss": 0.539, + "step": 21143 + }, + { + "epoch": 1.181355979551359, + "grad_norm": 0.41036897897720337, + "learning_rate": 0.00041099843119677277, + "loss": 0.4669, + "step": 21144 + }, + { + "epoch": 1.181411850155041, + "grad_norm": 0.436750590801239, + "learning_rate": 0.000410970416853429, + "loss": 0.5137, + "step": 21145 + }, + { + "epoch": 1.1814677207587228, + "grad_norm": 0.3587353825569153, + "learning_rate": 0.0004109424025100852, + "loss": 0.3764, + "step": 21146 + }, + { + "epoch": 1.1815235913624047, + "grad_norm": 0.4138387143611908, + "learning_rate": 0.0004109143881667414, + "loss": 0.4109, + "step": 21147 + }, + { + "epoch": 1.1815794619660864, + "grad_norm": 0.7289189696311951, + "learning_rate": 0.0004108863738233976, + "loss": 0.4495, + "step": 21148 + }, + { + "epoch": 1.1816353325697684, + "grad_norm": 0.5523490905761719, + "learning_rate": 0.0004108583594800538, + "loss": 0.5041, + "step": 21149 + }, + { + "epoch": 1.1816912031734503, + "grad_norm": 0.41928422451019287, + "learning_rate": 0.00041083034513671, + "loss": 0.3776, + "step": 21150 + }, + { + "epoch": 1.181747073777132, + "grad_norm": 0.397873193025589, + "learning_rate": 0.0004108023307933662, + "loss": 0.5069, + "step": 21151 + }, + { + "epoch": 1.181802944380814, + "grad_norm": 0.3889109194278717, + "learning_rate": 0.00041077431645002247, + "loss": 0.4007, + "step": 21152 + }, + { + "epoch": 1.181858814984496, + "grad_norm": 0.6203274726867676, + "learning_rate": 0.0004107463021066786, + "loss": 0.4949, + "step": 21153 + }, + { + "epoch": 1.1819146855881777, + "grad_norm": 0.46071481704711914, + "learning_rate": 0.0004107182877633349, + "loss": 0.5512, + "step": 21154 + }, + { + "epoch": 1.1819705561918596, + "grad_norm": 0.9957932233810425, + "learning_rate": 0.00041069027341999103, + "loss": 0.345, + "step": 21155 + }, + { + "epoch": 1.1820264267955416, + "grad_norm": 1.0422321557998657, + "learning_rate": 0.00041066225907664724, + "loss": 0.7465, + "step": 21156 + }, + { + "epoch": 1.1820822973992233, + "grad_norm": 0.4649060368537903, + "learning_rate": 0.00041063424473330345, + "loss": 0.4497, + "step": 21157 + }, + { + "epoch": 1.1821381680029053, + "grad_norm": 0.3471508026123047, + "learning_rate": 0.00041060623038995965, + "loss": 0.4177, + "step": 21158 + }, + { + "epoch": 1.1821940386065872, + "grad_norm": 0.45204344391822815, + "learning_rate": 0.00041057821604661586, + "loss": 0.4174, + "step": 21159 + }, + { + "epoch": 1.182249909210269, + "grad_norm": 0.5803166627883911, + "learning_rate": 0.00041055020170327206, + "loss": 0.4986, + "step": 21160 + }, + { + "epoch": 1.182305779813951, + "grad_norm": 0.526440441608429, + "learning_rate": 0.0004105221873599283, + "loss": 0.4579, + "step": 21161 + }, + { + "epoch": 1.1823616504176329, + "grad_norm": 0.3146607577800751, + "learning_rate": 0.0004104941730165845, + "loss": 0.3958, + "step": 21162 + }, + { + "epoch": 1.1824175210213146, + "grad_norm": 1.424686074256897, + "learning_rate": 0.00041046615867324074, + "loss": 0.339, + "step": 21163 + }, + { + "epoch": 1.1824733916249965, + "grad_norm": 0.6565869450569153, + "learning_rate": 0.0004104381443298969, + "loss": 0.4355, + "step": 21164 + }, + { + "epoch": 1.1825292622286785, + "grad_norm": 0.6585330367088318, + "learning_rate": 0.00041041012998655315, + "loss": 0.4864, + "step": 21165 + }, + { + "epoch": 1.1825851328323602, + "grad_norm": 0.5260930061340332, + "learning_rate": 0.0004103821156432093, + "loss": 0.4913, + "step": 21166 + }, + { + "epoch": 1.1826410034360422, + "grad_norm": 0.6218986511230469, + "learning_rate": 0.00041035410129986556, + "loss": 0.344, + "step": 21167 + }, + { + "epoch": 1.1826968740397241, + "grad_norm": 1.3200958967208862, + "learning_rate": 0.00041032608695652176, + "loss": 0.4372, + "step": 21168 + }, + { + "epoch": 1.1827527446434059, + "grad_norm": 0.5066361427307129, + "learning_rate": 0.00041029807261317797, + "loss": 0.3412, + "step": 21169 + }, + { + "epoch": 1.1828086152470878, + "grad_norm": 0.45777076482772827, + "learning_rate": 0.0004102700582698342, + "loss": 0.4626, + "step": 21170 + }, + { + "epoch": 1.1828644858507695, + "grad_norm": 0.45076876878738403, + "learning_rate": 0.0004102420439264904, + "loss": 0.3071, + "step": 21171 + }, + { + "epoch": 1.1829203564544515, + "grad_norm": 0.556447446346283, + "learning_rate": 0.0004102140295831466, + "loss": 0.403, + "step": 21172 + }, + { + "epoch": 1.1829762270581334, + "grad_norm": 0.41459524631500244, + "learning_rate": 0.0004101860152398028, + "loss": 0.3604, + "step": 21173 + }, + { + "epoch": 1.1830320976618152, + "grad_norm": 0.6461360454559326, + "learning_rate": 0.000410158000896459, + "loss": 0.456, + "step": 21174 + }, + { + "epoch": 1.1830879682654971, + "grad_norm": 0.45282265543937683, + "learning_rate": 0.0004101299865531152, + "loss": 0.3777, + "step": 21175 + }, + { + "epoch": 1.183143838869179, + "grad_norm": 0.407338410615921, + "learning_rate": 0.0004101019722097714, + "loss": 0.3402, + "step": 21176 + }, + { + "epoch": 1.1831997094728608, + "grad_norm": 0.571313738822937, + "learning_rate": 0.0004100739578664276, + "loss": 0.3516, + "step": 21177 + }, + { + "epoch": 1.1832555800765427, + "grad_norm": 0.4159110486507416, + "learning_rate": 0.0004100459435230838, + "loss": 0.4095, + "step": 21178 + }, + { + "epoch": 1.1833114506802247, + "grad_norm": 0.40227019786834717, + "learning_rate": 0.00041001792917974003, + "loss": 0.3622, + "step": 21179 + }, + { + "epoch": 1.1833673212839064, + "grad_norm": 0.40823855996131897, + "learning_rate": 0.00040998991483639623, + "loss": 0.3475, + "step": 21180 + }, + { + "epoch": 1.1834231918875884, + "grad_norm": 0.6175603866577148, + "learning_rate": 0.00040996190049305244, + "loss": 0.5381, + "step": 21181 + }, + { + "epoch": 1.18347906249127, + "grad_norm": 0.3992163836956024, + "learning_rate": 0.00040993388614970865, + "loss": 0.4269, + "step": 21182 + }, + { + "epoch": 1.183534933094952, + "grad_norm": 1.0534424781799316, + "learning_rate": 0.00040990587180636485, + "loss": 0.3459, + "step": 21183 + }, + { + "epoch": 1.183590803698634, + "grad_norm": 0.3208126127719879, + "learning_rate": 0.0004098778574630211, + "loss": 0.4072, + "step": 21184 + }, + { + "epoch": 1.1836466743023157, + "grad_norm": 1.1497232913970947, + "learning_rate": 0.00040984984311967726, + "loss": 0.5631, + "step": 21185 + }, + { + "epoch": 1.1837025449059977, + "grad_norm": 0.7141205668449402, + "learning_rate": 0.0004098218287763335, + "loss": 0.4252, + "step": 21186 + }, + { + "epoch": 1.1837584155096796, + "grad_norm": 0.47553175687789917, + "learning_rate": 0.0004097938144329897, + "loss": 0.4067, + "step": 21187 + }, + { + "epoch": 1.1838142861133614, + "grad_norm": 1.0070921182632446, + "learning_rate": 0.00040976580008964593, + "loss": 0.3693, + "step": 21188 + }, + { + "epoch": 1.1838701567170433, + "grad_norm": 0.8313623070716858, + "learning_rate": 0.0004097377857463021, + "loss": 0.3348, + "step": 21189 + }, + { + "epoch": 1.1839260273207253, + "grad_norm": 0.574149489402771, + "learning_rate": 0.00040970977140295835, + "loss": 0.4102, + "step": 21190 + }, + { + "epoch": 1.183981897924407, + "grad_norm": 0.5500195622444153, + "learning_rate": 0.0004096817570596145, + "loss": 0.4014, + "step": 21191 + }, + { + "epoch": 1.184037768528089, + "grad_norm": 0.6554253101348877, + "learning_rate": 0.00040965374271627076, + "loss": 0.4428, + "step": 21192 + }, + { + "epoch": 1.184093639131771, + "grad_norm": 3.4447851181030273, + "learning_rate": 0.00040962572837292696, + "loss": 0.3234, + "step": 21193 + }, + { + "epoch": 1.1841495097354526, + "grad_norm": 0.3698806166648865, + "learning_rate": 0.00040959771402958317, + "loss": 0.4314, + "step": 21194 + }, + { + "epoch": 1.1842053803391346, + "grad_norm": 0.30500131845474243, + "learning_rate": 0.0004095696996862394, + "loss": 0.323, + "step": 21195 + }, + { + "epoch": 1.1842612509428165, + "grad_norm": 0.5200318098068237, + "learning_rate": 0.0004095416853428956, + "loss": 0.4821, + "step": 21196 + }, + { + "epoch": 1.1843171215464983, + "grad_norm": 0.409484326839447, + "learning_rate": 0.0004095136709995518, + "loss": 0.4069, + "step": 21197 + }, + { + "epoch": 1.1843729921501802, + "grad_norm": 0.48632651567459106, + "learning_rate": 0.00040948565665620794, + "loss": 0.406, + "step": 21198 + }, + { + "epoch": 1.1844288627538622, + "grad_norm": 0.45600032806396484, + "learning_rate": 0.0004094576423128642, + "loss": 0.3578, + "step": 21199 + }, + { + "epoch": 1.184484733357544, + "grad_norm": 0.6244105696678162, + "learning_rate": 0.0004094296279695204, + "loss": 0.3562, + "step": 21200 + }, + { + "epoch": 1.1845406039612258, + "grad_norm": 0.643643856048584, + "learning_rate": 0.0004094016136261766, + "loss": 0.4347, + "step": 21201 + }, + { + "epoch": 1.1845964745649078, + "grad_norm": 0.37916797399520874, + "learning_rate": 0.0004093735992828328, + "loss": 0.3483, + "step": 21202 + }, + { + "epoch": 1.1846523451685895, + "grad_norm": 1.6599559783935547, + "learning_rate": 0.000409345584939489, + "loss": 0.4, + "step": 21203 + }, + { + "epoch": 1.1847082157722715, + "grad_norm": 0.5259612798690796, + "learning_rate": 0.00040931757059614523, + "loss": 0.4831, + "step": 21204 + }, + { + "epoch": 1.1847640863759532, + "grad_norm": 0.49186190962791443, + "learning_rate": 0.00040928955625280143, + "loss": 0.5581, + "step": 21205 + }, + { + "epoch": 1.1848199569796352, + "grad_norm": 0.5795938968658447, + "learning_rate": 0.00040926154190945764, + "loss": 0.5901, + "step": 21206 + }, + { + "epoch": 1.184875827583317, + "grad_norm": 1.1873283386230469, + "learning_rate": 0.0004092335275661139, + "loss": 0.4059, + "step": 21207 + }, + { + "epoch": 1.1849316981869988, + "grad_norm": 0.6022738218307495, + "learning_rate": 0.00040920551322277005, + "loss": 0.4135, + "step": 21208 + }, + { + "epoch": 1.1849875687906808, + "grad_norm": 0.5700743198394775, + "learning_rate": 0.0004091774988794263, + "loss": 0.5529, + "step": 21209 + }, + { + "epoch": 1.1850434393943627, + "grad_norm": 0.7739371657371521, + "learning_rate": 0.00040914948453608246, + "loss": 0.5094, + "step": 21210 + }, + { + "epoch": 1.1850993099980445, + "grad_norm": 0.4633810222148895, + "learning_rate": 0.0004091214701927387, + "loss": 0.5586, + "step": 21211 + }, + { + "epoch": 1.1851551806017264, + "grad_norm": 0.6235848069190979, + "learning_rate": 0.0004090934558493949, + "loss": 0.386, + "step": 21212 + }, + { + "epoch": 1.1852110512054082, + "grad_norm": 0.5357570052146912, + "learning_rate": 0.00040906544150605113, + "loss": 0.4162, + "step": 21213 + }, + { + "epoch": 1.18526692180909, + "grad_norm": 6.247970104217529, + "learning_rate": 0.0004090374271627073, + "loss": 0.4246, + "step": 21214 + }, + { + "epoch": 1.185322792412772, + "grad_norm": 0.3840339481830597, + "learning_rate": 0.00040900941281936355, + "loss": 0.4602, + "step": 21215 + }, + { + "epoch": 1.1853786630164538, + "grad_norm": 0.43228641152381897, + "learning_rate": 0.00040898139847601975, + "loss": 0.4095, + "step": 21216 + }, + { + "epoch": 1.1854345336201357, + "grad_norm": 0.5716155171394348, + "learning_rate": 0.00040895338413267596, + "loss": 0.3721, + "step": 21217 + }, + { + "epoch": 1.1854904042238177, + "grad_norm": 0.48812228441238403, + "learning_rate": 0.00040892536978933216, + "loss": 0.4239, + "step": 21218 + }, + { + "epoch": 1.1855462748274994, + "grad_norm": 0.7867795825004578, + "learning_rate": 0.0004088973554459883, + "loss": 0.3964, + "step": 21219 + }, + { + "epoch": 1.1856021454311814, + "grad_norm": 1.38527512550354, + "learning_rate": 0.0004088693411026446, + "loss": 0.4257, + "step": 21220 + }, + { + "epoch": 1.1856580160348633, + "grad_norm": 0.8453176617622375, + "learning_rate": 0.0004088413267593007, + "loss": 0.5635, + "step": 21221 + }, + { + "epoch": 1.185713886638545, + "grad_norm": 0.6625310182571411, + "learning_rate": 0.000408813312415957, + "loss": 0.4184, + "step": 21222 + }, + { + "epoch": 1.185769757242227, + "grad_norm": 0.35929128527641296, + "learning_rate": 0.0004087852980726132, + "loss": 0.3716, + "step": 21223 + }, + { + "epoch": 1.185825627845909, + "grad_norm": 1.7318400144577026, + "learning_rate": 0.0004087572837292694, + "loss": 0.4148, + "step": 21224 + }, + { + "epoch": 1.1858814984495907, + "grad_norm": 0.3533374071121216, + "learning_rate": 0.0004087292693859256, + "loss": 0.3145, + "step": 21225 + }, + { + "epoch": 1.1859373690532726, + "grad_norm": 0.5856642723083496, + "learning_rate": 0.0004087012550425818, + "loss": 0.3782, + "step": 21226 + }, + { + "epoch": 1.1859932396569546, + "grad_norm": 0.9528685808181763, + "learning_rate": 0.000408673240699238, + "loss": 0.5576, + "step": 21227 + }, + { + "epoch": 1.1860491102606363, + "grad_norm": 0.4543408453464508, + "learning_rate": 0.0004086452263558942, + "loss": 0.3667, + "step": 21228 + }, + { + "epoch": 1.1861049808643183, + "grad_norm": 0.4620549976825714, + "learning_rate": 0.00040861721201255043, + "loss": 0.4322, + "step": 21229 + }, + { + "epoch": 1.1861608514680002, + "grad_norm": 0.4289272725582123, + "learning_rate": 0.00040858919766920663, + "loss": 0.3856, + "step": 21230 + }, + { + "epoch": 1.186216722071682, + "grad_norm": 0.40406152606010437, + "learning_rate": 0.00040856118332586284, + "loss": 0.3356, + "step": 21231 + }, + { + "epoch": 1.186272592675364, + "grad_norm": 0.4737820625305176, + "learning_rate": 0.0004085331689825191, + "loss": 0.3518, + "step": 21232 + }, + { + "epoch": 1.1863284632790458, + "grad_norm": 0.5235486626625061, + "learning_rate": 0.00040850515463917525, + "loss": 0.4605, + "step": 21233 + }, + { + "epoch": 1.1863843338827276, + "grad_norm": 0.352274626493454, + "learning_rate": 0.0004084771402958315, + "loss": 0.3864, + "step": 21234 + }, + { + "epoch": 1.1864402044864095, + "grad_norm": 0.4834408760070801, + "learning_rate": 0.00040844912595248766, + "loss": 0.4836, + "step": 21235 + }, + { + "epoch": 1.1864960750900915, + "grad_norm": 0.4378686547279358, + "learning_rate": 0.0004084211116091439, + "loss": 0.2859, + "step": 21236 + }, + { + "epoch": 1.1865519456937732, + "grad_norm": 0.509317934513092, + "learning_rate": 0.0004083930972658001, + "loss": 0.5151, + "step": 21237 + }, + { + "epoch": 1.1866078162974552, + "grad_norm": 0.4203728139400482, + "learning_rate": 0.00040836508292245633, + "loss": 0.5133, + "step": 21238 + }, + { + "epoch": 1.1866636869011369, + "grad_norm": 1.2022720575332642, + "learning_rate": 0.00040833706857911254, + "loss": 0.4973, + "step": 21239 + }, + { + "epoch": 1.1867195575048188, + "grad_norm": 3.1479856967926025, + "learning_rate": 0.0004083090542357687, + "loss": 0.3476, + "step": 21240 + }, + { + "epoch": 1.1867754281085008, + "grad_norm": 0.7273097634315491, + "learning_rate": 0.00040828103989242495, + "loss": 0.4088, + "step": 21241 + }, + { + "epoch": 1.1868312987121825, + "grad_norm": 0.4102596640586853, + "learning_rate": 0.0004082530255490811, + "loss": 0.3809, + "step": 21242 + }, + { + "epoch": 1.1868871693158645, + "grad_norm": 0.38217398524284363, + "learning_rate": 0.00040822501120573736, + "loss": 0.433, + "step": 21243 + }, + { + "epoch": 1.1869430399195464, + "grad_norm": 0.8381626605987549, + "learning_rate": 0.0004081969968623935, + "loss": 0.3647, + "step": 21244 + }, + { + "epoch": 1.1869989105232281, + "grad_norm": 0.4480650722980499, + "learning_rate": 0.0004081689825190498, + "loss": 0.5355, + "step": 21245 + }, + { + "epoch": 1.18705478112691, + "grad_norm": 0.3644790053367615, + "learning_rate": 0.0004081409681757059, + "loss": 0.486, + "step": 21246 + }, + { + "epoch": 1.1871106517305918, + "grad_norm": 0.5613203644752502, + "learning_rate": 0.0004081129538323622, + "loss": 0.4392, + "step": 21247 + }, + { + "epoch": 1.1871665223342738, + "grad_norm": 0.46083977818489075, + "learning_rate": 0.0004080849394890184, + "loss": 0.4448, + "step": 21248 + }, + { + "epoch": 1.1872223929379557, + "grad_norm": 22.182035446166992, + "learning_rate": 0.0004080569251456746, + "loss": 0.453, + "step": 21249 + }, + { + "epoch": 1.1872782635416375, + "grad_norm": 0.4387044906616211, + "learning_rate": 0.0004080289108023308, + "loss": 0.5422, + "step": 21250 + }, + { + "epoch": 1.1873341341453194, + "grad_norm": 0.4700745642185211, + "learning_rate": 0.000408000896458987, + "loss": 0.4362, + "step": 21251 + }, + { + "epoch": 1.1873900047490014, + "grad_norm": 0.35312002897262573, + "learning_rate": 0.0004079728821156432, + "loss": 0.3396, + "step": 21252 + }, + { + "epoch": 1.187445875352683, + "grad_norm": 0.37464088201522827, + "learning_rate": 0.0004079448677722994, + "loss": 0.3292, + "step": 21253 + }, + { + "epoch": 1.187501745956365, + "grad_norm": 0.4446514844894409, + "learning_rate": 0.0004079168534289556, + "loss": 0.4011, + "step": 21254 + }, + { + "epoch": 1.187557616560047, + "grad_norm": 0.6048838496208191, + "learning_rate": 0.0004078888390856119, + "loss": 0.3714, + "step": 21255 + }, + { + "epoch": 1.1876134871637287, + "grad_norm": 0.5501270294189453, + "learning_rate": 0.00040786082474226804, + "loss": 0.4539, + "step": 21256 + }, + { + "epoch": 1.1876693577674107, + "grad_norm": 0.906524658203125, + "learning_rate": 0.0004078328103989243, + "loss": 0.4762, + "step": 21257 + }, + { + "epoch": 1.1877252283710926, + "grad_norm": 0.49433979392051697, + "learning_rate": 0.00040780479605558045, + "loss": 0.4101, + "step": 21258 + }, + { + "epoch": 1.1877810989747744, + "grad_norm": 0.4065938889980316, + "learning_rate": 0.0004077767817122367, + "loss": 0.4262, + "step": 21259 + }, + { + "epoch": 1.1878369695784563, + "grad_norm": 0.5190191268920898, + "learning_rate": 0.00040774876736889286, + "loss": 0.3579, + "step": 21260 + }, + { + "epoch": 1.1878928401821383, + "grad_norm": 0.3936119079589844, + "learning_rate": 0.00040772075302554907, + "loss": 0.4284, + "step": 21261 + }, + { + "epoch": 1.18794871078582, + "grad_norm": 0.519843339920044, + "learning_rate": 0.0004076927386822053, + "loss": 0.4129, + "step": 21262 + }, + { + "epoch": 1.188004581389502, + "grad_norm": 0.45611265301704407, + "learning_rate": 0.0004076647243388615, + "loss": 0.4186, + "step": 21263 + }, + { + "epoch": 1.1880604519931839, + "grad_norm": 0.5285108089447021, + "learning_rate": 0.00040763670999551774, + "loss": 0.3666, + "step": 21264 + }, + { + "epoch": 1.1881163225968656, + "grad_norm": 0.4490058422088623, + "learning_rate": 0.0004076086956521739, + "loss": 0.4247, + "step": 21265 + }, + { + "epoch": 1.1881721932005476, + "grad_norm": 0.3641122877597809, + "learning_rate": 0.00040758068130883015, + "loss": 0.3342, + "step": 21266 + }, + { + "epoch": 1.1882280638042295, + "grad_norm": 0.3952866196632385, + "learning_rate": 0.0004075526669654863, + "loss": 0.3941, + "step": 21267 + }, + { + "epoch": 1.1882839344079112, + "grad_norm": 0.5393572449684143, + "learning_rate": 0.00040752465262214256, + "loss": 0.4961, + "step": 21268 + }, + { + "epoch": 1.1883398050115932, + "grad_norm": 0.9350998997688293, + "learning_rate": 0.0004074966382787987, + "loss": 0.4423, + "step": 21269 + }, + { + "epoch": 1.188395675615275, + "grad_norm": 2.480207681655884, + "learning_rate": 0.000407468623935455, + "loss": 0.4458, + "step": 21270 + }, + { + "epoch": 1.1884515462189569, + "grad_norm": 0.4420393705368042, + "learning_rate": 0.0004074406095921112, + "loss": 0.5904, + "step": 21271 + }, + { + "epoch": 1.1885074168226388, + "grad_norm": 0.6498571038246155, + "learning_rate": 0.0004074125952487674, + "loss": 0.3989, + "step": 21272 + }, + { + "epoch": 1.1885632874263206, + "grad_norm": 0.7657992839813232, + "learning_rate": 0.0004073845809054236, + "loss": 0.3804, + "step": 21273 + }, + { + "epoch": 1.1886191580300025, + "grad_norm": 0.8250354528427124, + "learning_rate": 0.0004073565665620798, + "loss": 0.5582, + "step": 21274 + }, + { + "epoch": 1.1886750286336845, + "grad_norm": 0.417275071144104, + "learning_rate": 0.000407328552218736, + "loss": 0.4914, + "step": 21275 + }, + { + "epoch": 1.1887308992373662, + "grad_norm": 0.5108992457389832, + "learning_rate": 0.0004073005378753922, + "loss": 0.3955, + "step": 21276 + }, + { + "epoch": 1.1887867698410481, + "grad_norm": 0.49825337529182434, + "learning_rate": 0.0004072725235320484, + "loss": 0.5132, + "step": 21277 + }, + { + "epoch": 1.18884264044473, + "grad_norm": 1.3592010736465454, + "learning_rate": 0.0004072445091887047, + "loss": 0.3699, + "step": 21278 + }, + { + "epoch": 1.1888985110484118, + "grad_norm": 0.42320847511291504, + "learning_rate": 0.0004072164948453608, + "loss": 0.6176, + "step": 21279 + }, + { + "epoch": 1.1889543816520938, + "grad_norm": 2.7850985527038574, + "learning_rate": 0.0004071884805020171, + "loss": 0.3729, + "step": 21280 + }, + { + "epoch": 1.1890102522557755, + "grad_norm": 0.4374750852584839, + "learning_rate": 0.00040716046615867324, + "loss": 0.5862, + "step": 21281 + }, + { + "epoch": 1.1890661228594575, + "grad_norm": 1.1318914890289307, + "learning_rate": 0.00040713245181532944, + "loss": 0.5704, + "step": 21282 + }, + { + "epoch": 1.1891219934631394, + "grad_norm": 0.3032497763633728, + "learning_rate": 0.00040710443747198565, + "loss": 0.4389, + "step": 21283 + }, + { + "epoch": 1.1891778640668211, + "grad_norm": 0.46664300560951233, + "learning_rate": 0.00040707642312864186, + "loss": 0.3826, + "step": 21284 + }, + { + "epoch": 1.189233734670503, + "grad_norm": 0.7380884885787964, + "learning_rate": 0.00040704840878529806, + "loss": 0.483, + "step": 21285 + }, + { + "epoch": 1.189289605274185, + "grad_norm": 0.5542359948158264, + "learning_rate": 0.00040702039444195427, + "loss": 0.3732, + "step": 21286 + }, + { + "epoch": 1.1893454758778668, + "grad_norm": 0.602325975894928, + "learning_rate": 0.00040699238009861053, + "loss": 0.6395, + "step": 21287 + }, + { + "epoch": 1.1894013464815487, + "grad_norm": 0.39276134967803955, + "learning_rate": 0.0004069643657552667, + "loss": 0.3715, + "step": 21288 + }, + { + "epoch": 1.1894572170852307, + "grad_norm": 0.47140204906463623, + "learning_rate": 0.00040693635141192294, + "loss": 0.4946, + "step": 21289 + }, + { + "epoch": 1.1895130876889124, + "grad_norm": 0.385883629322052, + "learning_rate": 0.0004069083370685791, + "loss": 0.3513, + "step": 21290 + }, + { + "epoch": 1.1895689582925943, + "grad_norm": 0.6699520349502563, + "learning_rate": 0.00040688032272523535, + "loss": 0.5045, + "step": 21291 + }, + { + "epoch": 1.1896248288962763, + "grad_norm": 0.8147412538528442, + "learning_rate": 0.0004068523083818915, + "loss": 0.4364, + "step": 21292 + }, + { + "epoch": 1.189680699499958, + "grad_norm": 0.907586395740509, + "learning_rate": 0.00040682429403854776, + "loss": 0.4571, + "step": 21293 + }, + { + "epoch": 1.18973657010364, + "grad_norm": 1.388944387435913, + "learning_rate": 0.00040679627969520397, + "loss": 0.538, + "step": 21294 + }, + { + "epoch": 1.189792440707322, + "grad_norm": 0.39950835704803467, + "learning_rate": 0.0004067682653518602, + "loss": 0.3129, + "step": 21295 + }, + { + "epoch": 1.1898483113110037, + "grad_norm": 0.48571667075157166, + "learning_rate": 0.0004067402510085164, + "loss": 0.312, + "step": 21296 + }, + { + "epoch": 1.1899041819146856, + "grad_norm": 2.388258457183838, + "learning_rate": 0.0004067122366651726, + "loss": 0.3497, + "step": 21297 + }, + { + "epoch": 1.1899600525183676, + "grad_norm": 0.8776718378067017, + "learning_rate": 0.0004066842223218288, + "loss": 0.498, + "step": 21298 + }, + { + "epoch": 1.1900159231220493, + "grad_norm": 0.539547860622406, + "learning_rate": 0.000406656207978485, + "loss": 0.4168, + "step": 21299 + }, + { + "epoch": 1.1900717937257312, + "grad_norm": 0.33133426308631897, + "learning_rate": 0.0004066281936351412, + "loss": 0.3763, + "step": 21300 + }, + { + "epoch": 1.1901276643294132, + "grad_norm": 0.4033154845237732, + "learning_rate": 0.0004066001792917974, + "loss": 0.5165, + "step": 21301 + }, + { + "epoch": 1.190183534933095, + "grad_norm": 0.48263394832611084, + "learning_rate": 0.0004065721649484536, + "loss": 0.5568, + "step": 21302 + }, + { + "epoch": 1.1902394055367769, + "grad_norm": 1.4849607944488525, + "learning_rate": 0.0004065441506051098, + "loss": 0.6675, + "step": 21303 + }, + { + "epoch": 1.1902952761404586, + "grad_norm": 0.5213201642036438, + "learning_rate": 0.000406516136261766, + "loss": 0.3876, + "step": 21304 + }, + { + "epoch": 1.1903511467441406, + "grad_norm": 0.47904980182647705, + "learning_rate": 0.00040648812191842223, + "loss": 0.4461, + "step": 21305 + }, + { + "epoch": 1.1904070173478225, + "grad_norm": 0.47275856137275696, + "learning_rate": 0.00040646010757507844, + "loss": 0.4454, + "step": 21306 + }, + { + "epoch": 1.1904628879515042, + "grad_norm": 0.664704442024231, + "learning_rate": 0.00040643209323173464, + "loss": 0.4112, + "step": 21307 + }, + { + "epoch": 1.1905187585551862, + "grad_norm": 0.6802934408187866, + "learning_rate": 0.00040640407888839085, + "loss": 0.4299, + "step": 21308 + }, + { + "epoch": 1.1905746291588681, + "grad_norm": 0.5678035616874695, + "learning_rate": 0.00040637606454504706, + "loss": 0.4379, + "step": 21309 + }, + { + "epoch": 1.1906304997625499, + "grad_norm": 0.42434272170066833, + "learning_rate": 0.0004063480502017033, + "loss": 0.3428, + "step": 21310 + }, + { + "epoch": 1.1906863703662318, + "grad_norm": 0.695537805557251, + "learning_rate": 0.00040632003585835947, + "loss": 0.3916, + "step": 21311 + }, + { + "epoch": 1.1907422409699138, + "grad_norm": 8.605961799621582, + "learning_rate": 0.0004062920215150157, + "loss": 0.4104, + "step": 21312 + }, + { + "epoch": 1.1907981115735955, + "grad_norm": 1.4504082202911377, + "learning_rate": 0.0004062640071716719, + "loss": 0.4171, + "step": 21313 + }, + { + "epoch": 1.1908539821772774, + "grad_norm": 0.7516776919364929, + "learning_rate": 0.00040623599282832814, + "loss": 0.3959, + "step": 21314 + }, + { + "epoch": 1.1909098527809592, + "grad_norm": 0.5082627534866333, + "learning_rate": 0.0004062079784849843, + "loss": 0.431, + "step": 21315 + }, + { + "epoch": 1.1909657233846411, + "grad_norm": 0.5123956799507141, + "learning_rate": 0.00040617996414164055, + "loss": 0.377, + "step": 21316 + }, + { + "epoch": 1.191021593988323, + "grad_norm": 0.5828293561935425, + "learning_rate": 0.0004061519497982967, + "loss": 0.4176, + "step": 21317 + }, + { + "epoch": 1.1910774645920048, + "grad_norm": 0.3690257966518402, + "learning_rate": 0.00040612393545495296, + "loss": 0.3541, + "step": 21318 + }, + { + "epoch": 1.1911333351956868, + "grad_norm": 0.4626559317111969, + "learning_rate": 0.00040609592111160917, + "loss": 0.34, + "step": 21319 + }, + { + "epoch": 1.1911892057993687, + "grad_norm": 0.46131864190101624, + "learning_rate": 0.0004060679067682654, + "loss": 0.4272, + "step": 21320 + }, + { + "epoch": 1.1912450764030504, + "grad_norm": 0.4185774028301239, + "learning_rate": 0.0004060398924249216, + "loss": 0.4101, + "step": 21321 + }, + { + "epoch": 1.1913009470067324, + "grad_norm": 0.5478460788726807, + "learning_rate": 0.0004060118780815778, + "loss": 0.349, + "step": 21322 + }, + { + "epoch": 1.1913568176104143, + "grad_norm": 0.6294136643409729, + "learning_rate": 0.000405983863738234, + "loss": 0.3708, + "step": 21323 + }, + { + "epoch": 1.191412688214096, + "grad_norm": 0.737400233745575, + "learning_rate": 0.0004059558493948902, + "loss": 0.4534, + "step": 21324 + }, + { + "epoch": 1.191468558817778, + "grad_norm": 0.3892228901386261, + "learning_rate": 0.0004059278350515464, + "loss": 0.5141, + "step": 21325 + }, + { + "epoch": 1.19152442942146, + "grad_norm": 0.35934048891067505, + "learning_rate": 0.0004058998207082026, + "loss": 0.3995, + "step": 21326 + }, + { + "epoch": 1.1915803000251417, + "grad_norm": 0.5284218788146973, + "learning_rate": 0.0004058718063648588, + "loss": 0.4262, + "step": 21327 + }, + { + "epoch": 1.1916361706288237, + "grad_norm": 0.36740168929100037, + "learning_rate": 0.000405843792021515, + "loss": 0.3015, + "step": 21328 + }, + { + "epoch": 1.1916920412325056, + "grad_norm": 0.6551441550254822, + "learning_rate": 0.0004058157776781712, + "loss": 0.3762, + "step": 21329 + }, + { + "epoch": 1.1917479118361873, + "grad_norm": 0.5074504017829895, + "learning_rate": 0.00040578776333482743, + "loss": 0.4658, + "step": 21330 + }, + { + "epoch": 1.1918037824398693, + "grad_norm": 0.3335220515727997, + "learning_rate": 0.00040575974899148364, + "loss": 0.4566, + "step": 21331 + }, + { + "epoch": 1.1918596530435512, + "grad_norm": 0.43570417165756226, + "learning_rate": 0.00040573173464813984, + "loss": 0.421, + "step": 21332 + }, + { + "epoch": 1.191915523647233, + "grad_norm": 0.37865784764289856, + "learning_rate": 0.00040570372030479605, + "loss": 0.4023, + "step": 21333 + }, + { + "epoch": 1.191971394250915, + "grad_norm": 1.4099000692367554, + "learning_rate": 0.00040567570596145225, + "loss": 0.3342, + "step": 21334 + }, + { + "epoch": 1.1920272648545969, + "grad_norm": 0.3412969708442688, + "learning_rate": 0.0004056476916181085, + "loss": 0.3997, + "step": 21335 + }, + { + "epoch": 1.1920831354582786, + "grad_norm": 0.5008954405784607, + "learning_rate": 0.00040561967727476467, + "loss": 0.5002, + "step": 21336 + }, + { + "epoch": 1.1921390060619605, + "grad_norm": 0.37767207622528076, + "learning_rate": 0.0004055916629314209, + "loss": 0.4682, + "step": 21337 + }, + { + "epoch": 1.1921948766656423, + "grad_norm": 0.362146258354187, + "learning_rate": 0.0004055636485880771, + "loss": 0.3596, + "step": 21338 + }, + { + "epoch": 1.1922507472693242, + "grad_norm": 0.5462859869003296, + "learning_rate": 0.00040553563424473334, + "loss": 0.5072, + "step": 21339 + }, + { + "epoch": 1.1923066178730062, + "grad_norm": 1.4310919046401978, + "learning_rate": 0.0004055076199013895, + "loss": 0.6074, + "step": 21340 + }, + { + "epoch": 1.192362488476688, + "grad_norm": 0.4074028730392456, + "learning_rate": 0.00040547960555804575, + "loss": 0.3769, + "step": 21341 + }, + { + "epoch": 1.1924183590803699, + "grad_norm": 0.40007373690605164, + "learning_rate": 0.00040545159121470196, + "loss": 0.3663, + "step": 21342 + }, + { + "epoch": 1.1924742296840518, + "grad_norm": 0.5663937330245972, + "learning_rate": 0.00040542357687135816, + "loss": 0.5088, + "step": 21343 + }, + { + "epoch": 1.1925301002877335, + "grad_norm": 0.5127149820327759, + "learning_rate": 0.00040539556252801437, + "loss": 0.449, + "step": 21344 + }, + { + "epoch": 1.1925859708914155, + "grad_norm": 0.7708170413970947, + "learning_rate": 0.0004053675481846706, + "loss": 0.4053, + "step": 21345 + }, + { + "epoch": 1.1926418414950974, + "grad_norm": 0.8127928972244263, + "learning_rate": 0.0004053395338413268, + "loss": 0.4423, + "step": 21346 + }, + { + "epoch": 1.1926977120987792, + "grad_norm": 1.258310079574585, + "learning_rate": 0.00040531151949798293, + "loss": 0.5498, + "step": 21347 + }, + { + "epoch": 1.1927535827024611, + "grad_norm": 0.4431467056274414, + "learning_rate": 0.0004052835051546392, + "loss": 0.3685, + "step": 21348 + }, + { + "epoch": 1.1928094533061429, + "grad_norm": 1.1937798261642456, + "learning_rate": 0.00040525549081129534, + "loss": 0.4572, + "step": 21349 + }, + { + "epoch": 1.1928653239098248, + "grad_norm": 0.3697749972343445, + "learning_rate": 0.0004052274764679516, + "loss": 0.3692, + "step": 21350 + }, + { + "epoch": 1.1929211945135068, + "grad_norm": 1.011499285697937, + "learning_rate": 0.0004051994621246078, + "loss": 0.4698, + "step": 21351 + }, + { + "epoch": 1.1929770651171885, + "grad_norm": 0.349971204996109, + "learning_rate": 0.000405171447781264, + "loss": 0.442, + "step": 21352 + }, + { + "epoch": 1.1930329357208704, + "grad_norm": 0.3478279113769531, + "learning_rate": 0.0004051434334379202, + "loss": 0.459, + "step": 21353 + }, + { + "epoch": 1.1930888063245524, + "grad_norm": 0.40230193734169006, + "learning_rate": 0.0004051154190945764, + "loss": 0.4769, + "step": 21354 + }, + { + "epoch": 1.1931446769282341, + "grad_norm": 0.5612246990203857, + "learning_rate": 0.00040508740475123263, + "loss": 0.4719, + "step": 21355 + }, + { + "epoch": 1.193200547531916, + "grad_norm": 0.6178476214408875, + "learning_rate": 0.00040505939040788884, + "loss": 0.4489, + "step": 21356 + }, + { + "epoch": 1.193256418135598, + "grad_norm": 0.33357906341552734, + "learning_rate": 0.00040503137606454504, + "loss": 0.3224, + "step": 21357 + }, + { + "epoch": 1.1933122887392797, + "grad_norm": 0.5455013513565063, + "learning_rate": 0.0004050033617212013, + "loss": 0.4678, + "step": 21358 + }, + { + "epoch": 1.1933681593429617, + "grad_norm": 0.4588977098464966, + "learning_rate": 0.00040497534737785745, + "loss": 0.472, + "step": 21359 + }, + { + "epoch": 1.1934240299466436, + "grad_norm": 0.7069182991981506, + "learning_rate": 0.0004049473330345137, + "loss": 0.385, + "step": 21360 + }, + { + "epoch": 1.1934799005503254, + "grad_norm": 0.9217914342880249, + "learning_rate": 0.00040491931869116987, + "loss": 0.6028, + "step": 21361 + }, + { + "epoch": 1.1935357711540073, + "grad_norm": 1.2642642259597778, + "learning_rate": 0.0004048913043478261, + "loss": 0.4289, + "step": 21362 + }, + { + "epoch": 1.1935916417576893, + "grad_norm": 1.149126410484314, + "learning_rate": 0.0004048632900044823, + "loss": 0.4243, + "step": 21363 + }, + { + "epoch": 1.193647512361371, + "grad_norm": 0.3420371413230896, + "learning_rate": 0.00040483527566113854, + "loss": 0.3711, + "step": 21364 + }, + { + "epoch": 1.193703382965053, + "grad_norm": 0.6746293306350708, + "learning_rate": 0.00040480726131779474, + "loss": 0.4216, + "step": 21365 + }, + { + "epoch": 1.193759253568735, + "grad_norm": 0.45280200242996216, + "learning_rate": 0.00040477924697445095, + "loss": 0.3871, + "step": 21366 + }, + { + "epoch": 1.1938151241724166, + "grad_norm": 0.5310050249099731, + "learning_rate": 0.00040475123263110716, + "loss": 0.4949, + "step": 21367 + }, + { + "epoch": 1.1938709947760986, + "grad_norm": 0.4678191542625427, + "learning_rate": 0.0004047232182877633, + "loss": 0.3831, + "step": 21368 + }, + { + "epoch": 1.1939268653797805, + "grad_norm": 0.47518232464790344, + "learning_rate": 0.00040469520394441957, + "loss": 0.5182, + "step": 21369 + }, + { + "epoch": 1.1939827359834623, + "grad_norm": 0.5091590285301208, + "learning_rate": 0.0004046671896010757, + "loss": 0.4574, + "step": 21370 + }, + { + "epoch": 1.1940386065871442, + "grad_norm": 0.4199393093585968, + "learning_rate": 0.000404639175257732, + "loss": 0.5174, + "step": 21371 + }, + { + "epoch": 1.194094477190826, + "grad_norm": 0.6327903270721436, + "learning_rate": 0.00040461116091438813, + "loss": 0.3819, + "step": 21372 + }, + { + "epoch": 1.194150347794508, + "grad_norm": 0.4231743812561035, + "learning_rate": 0.0004045831465710444, + "loss": 0.4621, + "step": 21373 + }, + { + "epoch": 1.1942062183981899, + "grad_norm": 0.4753187596797943, + "learning_rate": 0.0004045551322277006, + "loss": 0.4429, + "step": 21374 + }, + { + "epoch": 1.1942620890018716, + "grad_norm": 0.4416128396987915, + "learning_rate": 0.0004045271178843568, + "loss": 0.5054, + "step": 21375 + }, + { + "epoch": 1.1943179596055535, + "grad_norm": 0.3350101113319397, + "learning_rate": 0.000404499103541013, + "loss": 0.459, + "step": 21376 + }, + { + "epoch": 1.1943738302092355, + "grad_norm": 0.5640031695365906, + "learning_rate": 0.0004044710891976692, + "loss": 0.4227, + "step": 21377 + }, + { + "epoch": 1.1944297008129172, + "grad_norm": 0.41889259219169617, + "learning_rate": 0.0004044430748543254, + "loss": 0.4918, + "step": 21378 + }, + { + "epoch": 1.1944855714165992, + "grad_norm": 0.6273728013038635, + "learning_rate": 0.0004044150605109816, + "loss": 0.4287, + "step": 21379 + }, + { + "epoch": 1.1945414420202811, + "grad_norm": 0.5254083871841431, + "learning_rate": 0.00040438704616763783, + "loss": 0.7139, + "step": 21380 + }, + { + "epoch": 1.1945973126239628, + "grad_norm": 0.3634721636772156, + "learning_rate": 0.0004043590318242941, + "loss": 0.428, + "step": 21381 + }, + { + "epoch": 1.1946531832276448, + "grad_norm": 0.45925626158714294, + "learning_rate": 0.00040433101748095024, + "loss": 0.5086, + "step": 21382 + }, + { + "epoch": 1.1947090538313265, + "grad_norm": 0.5679476261138916, + "learning_rate": 0.0004043030031376065, + "loss": 0.3449, + "step": 21383 + }, + { + "epoch": 1.1947649244350085, + "grad_norm": 0.39481863379478455, + "learning_rate": 0.00040427498879426265, + "loss": 0.3587, + "step": 21384 + }, + { + "epoch": 1.1948207950386904, + "grad_norm": 0.5487396121025085, + "learning_rate": 0.0004042469744509189, + "loss": 0.516, + "step": 21385 + }, + { + "epoch": 1.1948766656423722, + "grad_norm": 0.3954319953918457, + "learning_rate": 0.00040421896010757507, + "loss": 0.4118, + "step": 21386 + }, + { + "epoch": 1.1949325362460541, + "grad_norm": 0.3736436367034912, + "learning_rate": 0.0004041909457642313, + "loss": 0.4459, + "step": 21387 + }, + { + "epoch": 1.194988406849736, + "grad_norm": 3.2229199409484863, + "learning_rate": 0.0004041629314208875, + "loss": 0.4866, + "step": 21388 + }, + { + "epoch": 1.1950442774534178, + "grad_norm": 0.6585288643836975, + "learning_rate": 0.0004041349170775437, + "loss": 0.5075, + "step": 21389 + }, + { + "epoch": 1.1951001480570997, + "grad_norm": 0.64420485496521, + "learning_rate": 0.00040410690273419994, + "loss": 0.395, + "step": 21390 + }, + { + "epoch": 1.1951560186607817, + "grad_norm": 0.556313157081604, + "learning_rate": 0.0004040788883908561, + "loss": 0.4137, + "step": 21391 + }, + { + "epoch": 1.1952118892644634, + "grad_norm": 0.44109874963760376, + "learning_rate": 0.00040405087404751235, + "loss": 0.3807, + "step": 21392 + }, + { + "epoch": 1.1952677598681454, + "grad_norm": 0.4753761291503906, + "learning_rate": 0.0004040228597041685, + "loss": 0.4538, + "step": 21393 + }, + { + "epoch": 1.1953236304718273, + "grad_norm": 0.3326793909072876, + "learning_rate": 0.00040399484536082477, + "loss": 0.3574, + "step": 21394 + }, + { + "epoch": 1.195379501075509, + "grad_norm": 0.29670366644859314, + "learning_rate": 0.0004039668310174809, + "loss": 0.2871, + "step": 21395 + }, + { + "epoch": 1.195435371679191, + "grad_norm": 0.3802247643470764, + "learning_rate": 0.0004039388166741372, + "loss": 0.563, + "step": 21396 + }, + { + "epoch": 1.195491242282873, + "grad_norm": 1.5096145868301392, + "learning_rate": 0.0004039108023307934, + "loss": 0.6109, + "step": 21397 + }, + { + "epoch": 1.1955471128865547, + "grad_norm": 0.41475358605384827, + "learning_rate": 0.0004038827879874496, + "loss": 0.3874, + "step": 21398 + }, + { + "epoch": 1.1956029834902366, + "grad_norm": 0.38279634714126587, + "learning_rate": 0.0004038547736441058, + "loss": 0.3048, + "step": 21399 + }, + { + "epoch": 1.1956588540939186, + "grad_norm": 1.8059622049331665, + "learning_rate": 0.000403826759300762, + "loss": 0.4087, + "step": 21400 + }, + { + "epoch": 1.1957147246976003, + "grad_norm": 5.257320880889893, + "learning_rate": 0.0004037987449574182, + "loss": 0.3328, + "step": 21401 + }, + { + "epoch": 1.1957705953012823, + "grad_norm": 0.5199947357177734, + "learning_rate": 0.0004037707306140744, + "loss": 0.432, + "step": 21402 + }, + { + "epoch": 1.1958264659049642, + "grad_norm": 0.32498306035995483, + "learning_rate": 0.0004037427162707306, + "loss": 0.3394, + "step": 21403 + }, + { + "epoch": 1.195882336508646, + "grad_norm": 0.7680503129959106, + "learning_rate": 0.0004037147019273868, + "loss": 0.4248, + "step": 21404 + }, + { + "epoch": 1.195938207112328, + "grad_norm": 0.9237014055252075, + "learning_rate": 0.00040368668758404303, + "loss": 0.3859, + "step": 21405 + }, + { + "epoch": 1.1959940777160096, + "grad_norm": 0.43229711055755615, + "learning_rate": 0.0004036586732406993, + "loss": 0.4573, + "step": 21406 + }, + { + "epoch": 1.1960499483196916, + "grad_norm": 0.5608473420143127, + "learning_rate": 0.00040363065889735544, + "loss": 0.5248, + "step": 21407 + }, + { + "epoch": 1.1961058189233735, + "grad_norm": 3.745220422744751, + "learning_rate": 0.0004036026445540117, + "loss": 0.4911, + "step": 21408 + }, + { + "epoch": 1.1961616895270553, + "grad_norm": 0.3788805603981018, + "learning_rate": 0.00040357463021066785, + "loss": 0.485, + "step": 21409 + }, + { + "epoch": 1.1962175601307372, + "grad_norm": 0.6164228916168213, + "learning_rate": 0.00040354661586732406, + "loss": 0.5364, + "step": 21410 + }, + { + "epoch": 1.1962734307344192, + "grad_norm": 0.4191749095916748, + "learning_rate": 0.00040351860152398027, + "loss": 0.3651, + "step": 21411 + }, + { + "epoch": 1.196329301338101, + "grad_norm": 0.4662926495075226, + "learning_rate": 0.00040349058718063647, + "loss": 0.2945, + "step": 21412 + }, + { + "epoch": 1.1963851719417828, + "grad_norm": 0.47060105204582214, + "learning_rate": 0.00040346257283729273, + "loss": 0.3852, + "step": 21413 + }, + { + "epoch": 1.1964410425454648, + "grad_norm": 0.48430588841438293, + "learning_rate": 0.0004034345584939489, + "loss": 0.4374, + "step": 21414 + }, + { + "epoch": 1.1964969131491465, + "grad_norm": 0.5512615442276001, + "learning_rate": 0.00040340654415060514, + "loss": 0.479, + "step": 21415 + }, + { + "epoch": 1.1965527837528285, + "grad_norm": 0.3313872218132019, + "learning_rate": 0.0004033785298072613, + "loss": 0.5038, + "step": 21416 + }, + { + "epoch": 1.1966086543565102, + "grad_norm": 0.4652712941169739, + "learning_rate": 0.00040335051546391755, + "loss": 0.4835, + "step": 21417 + }, + { + "epoch": 1.1966645249601922, + "grad_norm": 0.4012984335422516, + "learning_rate": 0.0004033225011205737, + "loss": 0.3467, + "step": 21418 + }, + { + "epoch": 1.196720395563874, + "grad_norm": 0.8244191408157349, + "learning_rate": 0.00040329448677722997, + "loss": 0.7424, + "step": 21419 + }, + { + "epoch": 1.1967762661675558, + "grad_norm": 1.3520435094833374, + "learning_rate": 0.0004032664724338861, + "loss": 0.424, + "step": 21420 + }, + { + "epoch": 1.1968321367712378, + "grad_norm": 0.8857206106185913, + "learning_rate": 0.0004032384580905424, + "loss": 0.4291, + "step": 21421 + }, + { + "epoch": 1.1968880073749197, + "grad_norm": 0.37849220633506775, + "learning_rate": 0.0004032104437471986, + "loss": 0.4308, + "step": 21422 + }, + { + "epoch": 1.1969438779786015, + "grad_norm": 0.3199082314968109, + "learning_rate": 0.0004031824294038548, + "loss": 0.3659, + "step": 21423 + }, + { + "epoch": 1.1969997485822834, + "grad_norm": 0.3770596981048584, + "learning_rate": 0.000403154415060511, + "loss": 0.3771, + "step": 21424 + }, + { + "epoch": 1.1970556191859654, + "grad_norm": 0.5033682584762573, + "learning_rate": 0.0004031264007171672, + "loss": 0.4291, + "step": 21425 + }, + { + "epoch": 1.197111489789647, + "grad_norm": 0.33647286891937256, + "learning_rate": 0.0004030983863738234, + "loss": 0.3692, + "step": 21426 + }, + { + "epoch": 1.197167360393329, + "grad_norm": 0.5038193464279175, + "learning_rate": 0.0004030703720304796, + "loss": 0.5468, + "step": 21427 + }, + { + "epoch": 1.197223230997011, + "grad_norm": 0.3421250283718109, + "learning_rate": 0.0004030423576871358, + "loss": 0.3398, + "step": 21428 + }, + { + "epoch": 1.1972791016006927, + "grad_norm": 0.33884039521217346, + "learning_rate": 0.0004030143433437921, + "loss": 0.3968, + "step": 21429 + }, + { + "epoch": 1.1973349722043747, + "grad_norm": 0.6084532737731934, + "learning_rate": 0.00040298632900044823, + "loss": 0.3714, + "step": 21430 + }, + { + "epoch": 1.1973908428080566, + "grad_norm": 0.45607882738113403, + "learning_rate": 0.00040295831465710444, + "loss": 0.3833, + "step": 21431 + }, + { + "epoch": 1.1974467134117384, + "grad_norm": 0.5152944922447205, + "learning_rate": 0.00040293030031376064, + "loss": 0.4742, + "step": 21432 + }, + { + "epoch": 1.1975025840154203, + "grad_norm": 0.2924326956272125, + "learning_rate": 0.00040290228597041685, + "loss": 0.3345, + "step": 21433 + }, + { + "epoch": 1.1975584546191023, + "grad_norm": 0.4562065005302429, + "learning_rate": 0.00040287427162707305, + "loss": 0.4222, + "step": 21434 + }, + { + "epoch": 1.197614325222784, + "grad_norm": 0.5442754030227661, + "learning_rate": 0.00040284625728372926, + "loss": 0.4108, + "step": 21435 + }, + { + "epoch": 1.197670195826466, + "grad_norm": 2.778737783432007, + "learning_rate": 0.0004028182429403855, + "loss": 0.4562, + "step": 21436 + }, + { + "epoch": 1.197726066430148, + "grad_norm": 0.6510493159294128, + "learning_rate": 0.00040279022859704167, + "loss": 0.4251, + "step": 21437 + }, + { + "epoch": 1.1977819370338296, + "grad_norm": 0.391571968793869, + "learning_rate": 0.00040276221425369793, + "loss": 0.427, + "step": 21438 + }, + { + "epoch": 1.1978378076375116, + "grad_norm": 0.4636944532394409, + "learning_rate": 0.0004027341999103541, + "loss": 0.4196, + "step": 21439 + }, + { + "epoch": 1.1978936782411933, + "grad_norm": 0.3461504280567169, + "learning_rate": 0.00040270618556701034, + "loss": 0.5002, + "step": 21440 + }, + { + "epoch": 1.1979495488448753, + "grad_norm": 0.8985955715179443, + "learning_rate": 0.0004026781712236665, + "loss": 0.5167, + "step": 21441 + }, + { + "epoch": 1.1980054194485572, + "grad_norm": 0.34976667165756226, + "learning_rate": 0.00040265015688032275, + "loss": 0.4428, + "step": 21442 + }, + { + "epoch": 1.198061290052239, + "grad_norm": 0.6014752984046936, + "learning_rate": 0.0004026221425369789, + "loss": 0.4794, + "step": 21443 + }, + { + "epoch": 1.1981171606559209, + "grad_norm": 2.7713146209716797, + "learning_rate": 0.00040259412819363517, + "loss": 0.4208, + "step": 21444 + }, + { + "epoch": 1.1981730312596028, + "grad_norm": 0.5871027708053589, + "learning_rate": 0.00040256611385029137, + "loss": 0.4186, + "step": 21445 + }, + { + "epoch": 1.1982289018632846, + "grad_norm": 0.4223265051841736, + "learning_rate": 0.0004025380995069476, + "loss": 0.5079, + "step": 21446 + }, + { + "epoch": 1.1982847724669665, + "grad_norm": 0.37781113386154175, + "learning_rate": 0.0004025100851636038, + "loss": 0.5285, + "step": 21447 + }, + { + "epoch": 1.1983406430706485, + "grad_norm": 0.364205002784729, + "learning_rate": 0.00040248207082026, + "loss": 0.4464, + "step": 21448 + }, + { + "epoch": 1.1983965136743302, + "grad_norm": 0.47674980759620667, + "learning_rate": 0.0004024540564769162, + "loss": 0.6034, + "step": 21449 + }, + { + "epoch": 1.1984523842780122, + "grad_norm": 0.3675402104854584, + "learning_rate": 0.0004024260421335724, + "loss": 0.4362, + "step": 21450 + }, + { + "epoch": 1.1985082548816939, + "grad_norm": 0.37128445506095886, + "learning_rate": 0.0004023980277902286, + "loss": 0.3692, + "step": 21451 + }, + { + "epoch": 1.1985641254853758, + "grad_norm": 0.6837619543075562, + "learning_rate": 0.0004023700134468848, + "loss": 0.5053, + "step": 21452 + }, + { + "epoch": 1.1986199960890578, + "grad_norm": 0.6525677442550659, + "learning_rate": 0.000402341999103541, + "loss": 0.5481, + "step": 21453 + }, + { + "epoch": 1.1986758666927395, + "grad_norm": 0.44276392459869385, + "learning_rate": 0.0004023139847601972, + "loss": 0.3606, + "step": 21454 + }, + { + "epoch": 1.1987317372964215, + "grad_norm": 0.5392950773239136, + "learning_rate": 0.00040228597041685343, + "loss": 0.428, + "step": 21455 + }, + { + "epoch": 1.1987876079001034, + "grad_norm": 0.42962801456451416, + "learning_rate": 0.00040225795607350964, + "loss": 0.3616, + "step": 21456 + }, + { + "epoch": 1.1988434785037851, + "grad_norm": 0.34076565504074097, + "learning_rate": 0.00040222994173016584, + "loss": 0.4135, + "step": 21457 + }, + { + "epoch": 1.198899349107467, + "grad_norm": 0.3690205216407776, + "learning_rate": 0.00040220192738682205, + "loss": 0.4595, + "step": 21458 + }, + { + "epoch": 1.198955219711149, + "grad_norm": 0.37179452180862427, + "learning_rate": 0.00040217391304347825, + "loss": 0.4179, + "step": 21459 + }, + { + "epoch": 1.1990110903148308, + "grad_norm": 0.37223708629608154, + "learning_rate": 0.00040214589870013446, + "loss": 0.4551, + "step": 21460 + }, + { + "epoch": 1.1990669609185127, + "grad_norm": 0.4488182067871094, + "learning_rate": 0.0004021178843567907, + "loss": 0.4147, + "step": 21461 + }, + { + "epoch": 1.1991228315221947, + "grad_norm": 0.7281560301780701, + "learning_rate": 0.00040208987001344687, + "loss": 0.4039, + "step": 21462 + }, + { + "epoch": 1.1991787021258764, + "grad_norm": 0.5471133589744568, + "learning_rate": 0.00040206185567010313, + "loss": 0.7284, + "step": 21463 + }, + { + "epoch": 1.1992345727295584, + "grad_norm": 0.46017634868621826, + "learning_rate": 0.0004020338413267593, + "loss": 0.4724, + "step": 21464 + }, + { + "epoch": 1.1992904433332403, + "grad_norm": 0.6952259540557861, + "learning_rate": 0.00040200582698341554, + "loss": 0.4366, + "step": 21465 + }, + { + "epoch": 1.199346313936922, + "grad_norm": 0.8417477011680603, + "learning_rate": 0.0004019778126400717, + "loss": 0.3428, + "step": 21466 + }, + { + "epoch": 1.199402184540604, + "grad_norm": 1.9849334955215454, + "learning_rate": 0.00040194979829672795, + "loss": 0.7762, + "step": 21467 + }, + { + "epoch": 1.199458055144286, + "grad_norm": 0.6315877437591553, + "learning_rate": 0.00040192178395338416, + "loss": 0.4642, + "step": 21468 + }, + { + "epoch": 1.1995139257479677, + "grad_norm": 0.5186702013015747, + "learning_rate": 0.00040189376961004037, + "loss": 0.3953, + "step": 21469 + }, + { + "epoch": 1.1995697963516496, + "grad_norm": 0.43993762135505676, + "learning_rate": 0.00040186575526669657, + "loss": 0.3571, + "step": 21470 + }, + { + "epoch": 1.1996256669553316, + "grad_norm": 0.5644364953041077, + "learning_rate": 0.0004018377409233528, + "loss": 0.3676, + "step": 21471 + }, + { + "epoch": 1.1996815375590133, + "grad_norm": 0.45813456177711487, + "learning_rate": 0.000401809726580009, + "loss": 0.4522, + "step": 21472 + }, + { + "epoch": 1.1997374081626953, + "grad_norm": 0.6959822177886963, + "learning_rate": 0.00040178171223666513, + "loss": 0.4436, + "step": 21473 + }, + { + "epoch": 1.199793278766377, + "grad_norm": 0.6931326985359192, + "learning_rate": 0.0004017536978933214, + "loss": 0.4617, + "step": 21474 + }, + { + "epoch": 1.199849149370059, + "grad_norm": 0.5528848171234131, + "learning_rate": 0.00040172568354997755, + "loss": 0.3496, + "step": 21475 + }, + { + "epoch": 1.1999050199737409, + "grad_norm": 0.32486557960510254, + "learning_rate": 0.0004016976692066338, + "loss": 0.4227, + "step": 21476 + }, + { + "epoch": 1.1999608905774226, + "grad_norm": 1.603967308998108, + "learning_rate": 0.00040166965486329, + "loss": 0.4164, + "step": 21477 + }, + { + "epoch": 1.2000167611811046, + "grad_norm": 0.7085089087486267, + "learning_rate": 0.0004016416405199462, + "loss": 0.7214, + "step": 21478 + }, + { + "epoch": 1.2000726317847865, + "grad_norm": 0.3846448063850403, + "learning_rate": 0.0004016136261766024, + "loss": 0.4852, + "step": 21479 + }, + { + "epoch": 1.2001285023884682, + "grad_norm": 0.4160142242908478, + "learning_rate": 0.00040158561183325863, + "loss": 0.4155, + "step": 21480 + }, + { + "epoch": 1.2001843729921502, + "grad_norm": 0.4866863787174225, + "learning_rate": 0.00040155759748991483, + "loss": 0.3661, + "step": 21481 + }, + { + "epoch": 1.2002402435958321, + "grad_norm": 0.5483714938163757, + "learning_rate": 0.00040152958314657104, + "loss": 0.471, + "step": 21482 + }, + { + "epoch": 1.2002961141995139, + "grad_norm": 0.4257173538208008, + "learning_rate": 0.00040150156880322725, + "loss": 0.421, + "step": 21483 + }, + { + "epoch": 1.2003519848031958, + "grad_norm": 0.35383835434913635, + "learning_rate": 0.0004014735544598835, + "loss": 0.4151, + "step": 21484 + }, + { + "epoch": 1.2004078554068776, + "grad_norm": 0.4159003794193268, + "learning_rate": 0.00040144554011653966, + "loss": 0.4575, + "step": 21485 + }, + { + "epoch": 1.2004637260105595, + "grad_norm": 0.34293133020401, + "learning_rate": 0.0004014175257731959, + "loss": 0.3997, + "step": 21486 + }, + { + "epoch": 1.2005195966142415, + "grad_norm": 0.46744897961616516, + "learning_rate": 0.00040138951142985207, + "loss": 0.3546, + "step": 21487 + }, + { + "epoch": 1.2005754672179232, + "grad_norm": 0.7021288871765137, + "learning_rate": 0.00040136149708650833, + "loss": 0.3251, + "step": 21488 + }, + { + "epoch": 1.2006313378216051, + "grad_norm": 0.5722562670707703, + "learning_rate": 0.0004013334827431645, + "loss": 0.3516, + "step": 21489 + }, + { + "epoch": 1.200687208425287, + "grad_norm": 0.4708857536315918, + "learning_rate": 0.00040130546839982074, + "loss": 0.3771, + "step": 21490 + }, + { + "epoch": 1.2007430790289688, + "grad_norm": 0.367046058177948, + "learning_rate": 0.00040127745405647695, + "loss": 0.3892, + "step": 21491 + }, + { + "epoch": 1.2007989496326508, + "grad_norm": 0.3746107220649719, + "learning_rate": 0.00040124943971313315, + "loss": 0.4034, + "step": 21492 + }, + { + "epoch": 1.2008548202363327, + "grad_norm": 0.49138540029525757, + "learning_rate": 0.00040122142536978936, + "loss": 0.3966, + "step": 21493 + }, + { + "epoch": 1.2009106908400144, + "grad_norm": 0.4655064344406128, + "learning_rate": 0.00040119341102644556, + "loss": 0.4938, + "step": 21494 + }, + { + "epoch": 1.2009665614436964, + "grad_norm": 0.3701217472553253, + "learning_rate": 0.00040116539668310177, + "loss": 0.3772, + "step": 21495 + }, + { + "epoch": 1.2010224320473784, + "grad_norm": 0.43462511897087097, + "learning_rate": 0.0004011373823397579, + "loss": 0.3836, + "step": 21496 + }, + { + "epoch": 1.20107830265106, + "grad_norm": 0.3761542737483978, + "learning_rate": 0.0004011093679964142, + "loss": 0.4478, + "step": 21497 + }, + { + "epoch": 1.201134173254742, + "grad_norm": 0.41024067997932434, + "learning_rate": 0.00040108135365307033, + "loss": 0.4117, + "step": 21498 + }, + { + "epoch": 1.201190043858424, + "grad_norm": 0.38044729828834534, + "learning_rate": 0.0004010533393097266, + "loss": 0.4919, + "step": 21499 + }, + { + "epoch": 1.2012459144621057, + "grad_norm": 1.994361162185669, + "learning_rate": 0.0004010253249663828, + "loss": 0.5725, + "step": 21500 + }, + { + "epoch": 1.2012459144621057, + "eval_cer": 0.08568732065507949, + "eval_loss": 0.3264184296131134, + "eval_runtime": 56.1801, + "eval_samples_per_second": 80.776, + "eval_steps_per_second": 5.055, + "eval_wer": 0.33934806290327085, + "step": 21500 + }, + { + "epoch": 1.2013017850657877, + "grad_norm": 0.47625845670700073, + "learning_rate": 0.000400997310623039, + "loss": 0.4163, + "step": 21501 + }, + { + "epoch": 1.2013576556694696, + "grad_norm": 0.8015896677970886, + "learning_rate": 0.0004009692962796952, + "loss": 0.3892, + "step": 21502 + }, + { + "epoch": 1.2014135262731513, + "grad_norm": 0.3729545474052429, + "learning_rate": 0.0004009412819363514, + "loss": 0.3485, + "step": 21503 + }, + { + "epoch": 1.2014693968768333, + "grad_norm": 0.39010292291641235, + "learning_rate": 0.0004009132675930076, + "loss": 0.4047, + "step": 21504 + }, + { + "epoch": 1.2015252674805152, + "grad_norm": 0.4437546730041504, + "learning_rate": 0.00040088525324966383, + "loss": 0.4003, + "step": 21505 + }, + { + "epoch": 1.201581138084197, + "grad_norm": 0.4621172845363617, + "learning_rate": 0.00040085723890632003, + "loss": 0.4033, + "step": 21506 + }, + { + "epoch": 1.201637008687879, + "grad_norm": 0.3736850619316101, + "learning_rate": 0.0004008292245629763, + "loss": 0.4281, + "step": 21507 + }, + { + "epoch": 1.2016928792915607, + "grad_norm": 0.49688059091567993, + "learning_rate": 0.00040080121021963245, + "loss": 0.4177, + "step": 21508 + }, + { + "epoch": 1.2017487498952426, + "grad_norm": 0.42732372879981995, + "learning_rate": 0.0004007731958762887, + "loss": 0.5059, + "step": 21509 + }, + { + "epoch": 1.2018046204989246, + "grad_norm": 0.35351043939590454, + "learning_rate": 0.00040074518153294486, + "loss": 0.3747, + "step": 21510 + }, + { + "epoch": 1.2018604911026063, + "grad_norm": 0.45051950216293335, + "learning_rate": 0.0004007171671896011, + "loss": 0.3537, + "step": 21511 + }, + { + "epoch": 1.2019163617062882, + "grad_norm": 0.3840484917163849, + "learning_rate": 0.00040068915284625727, + "loss": 0.3782, + "step": 21512 + }, + { + "epoch": 1.2019722323099702, + "grad_norm": 0.4602525532245636, + "learning_rate": 0.00040066113850291353, + "loss": 0.4705, + "step": 21513 + }, + { + "epoch": 1.202028102913652, + "grad_norm": 0.3842848241329193, + "learning_rate": 0.0004006331241595697, + "loss": 0.3865, + "step": 21514 + }, + { + "epoch": 1.2020839735173339, + "grad_norm": 0.41128841042518616, + "learning_rate": 0.00040060510981622594, + "loss": 0.4388, + "step": 21515 + }, + { + "epoch": 1.2021398441210158, + "grad_norm": 0.562713623046875, + "learning_rate": 0.00040057709547288215, + "loss": 0.59, + "step": 21516 + }, + { + "epoch": 1.2021957147246976, + "grad_norm": 6.747410774230957, + "learning_rate": 0.0004005490811295383, + "loss": 0.6255, + "step": 21517 + }, + { + "epoch": 1.2022515853283795, + "grad_norm": 0.5308858156204224, + "learning_rate": 0.00040052106678619456, + "loss": 0.3649, + "step": 21518 + }, + { + "epoch": 1.2023074559320612, + "grad_norm": 0.39798206090927124, + "learning_rate": 0.0004004930524428507, + "loss": 0.403, + "step": 21519 + }, + { + "epoch": 1.2023633265357432, + "grad_norm": 0.6960659027099609, + "learning_rate": 0.00040046503809950697, + "loss": 0.4296, + "step": 21520 + }, + { + "epoch": 1.2024191971394251, + "grad_norm": 0.8804600238800049, + "learning_rate": 0.0004004370237561631, + "loss": 0.4456, + "step": 21521 + }, + { + "epoch": 1.2024750677431069, + "grad_norm": 0.41844838857650757, + "learning_rate": 0.0004004090094128194, + "loss": 0.3471, + "step": 21522 + }, + { + "epoch": 1.2025309383467888, + "grad_norm": 2.2466442584991455, + "learning_rate": 0.0004003809950694756, + "loss": 0.3748, + "step": 21523 + }, + { + "epoch": 1.2025868089504708, + "grad_norm": 0.3886549174785614, + "learning_rate": 0.0004003529807261318, + "loss": 0.4037, + "step": 21524 + }, + { + "epoch": 1.2026426795541525, + "grad_norm": 0.3839401304721832, + "learning_rate": 0.000400324966382788, + "loss": 0.359, + "step": 21525 + }, + { + "epoch": 1.2026985501578344, + "grad_norm": 0.8218753337860107, + "learning_rate": 0.0004002969520394442, + "loss": 0.4959, + "step": 21526 + }, + { + "epoch": 1.2027544207615164, + "grad_norm": 0.4441520571708679, + "learning_rate": 0.0004002689376961004, + "loss": 0.3668, + "step": 21527 + }, + { + "epoch": 1.2028102913651981, + "grad_norm": 1.510513186454773, + "learning_rate": 0.0004002409233527566, + "loss": 0.4812, + "step": 21528 + }, + { + "epoch": 1.20286616196888, + "grad_norm": 0.4773935377597809, + "learning_rate": 0.0004002129090094128, + "loss": 0.43, + "step": 21529 + }, + { + "epoch": 1.202922032572562, + "grad_norm": 0.3920912444591522, + "learning_rate": 0.00040018489466606903, + "loss": 0.4039, + "step": 21530 + }, + { + "epoch": 1.2029779031762438, + "grad_norm": 0.5823320746421814, + "learning_rate": 0.00040015688032272523, + "loss": 0.389, + "step": 21531 + }, + { + "epoch": 1.2030337737799257, + "grad_norm": 0.49722981452941895, + "learning_rate": 0.0004001288659793815, + "loss": 0.4534, + "step": 21532 + }, + { + "epoch": 1.2030896443836077, + "grad_norm": 2.018977165222168, + "learning_rate": 0.00040010085163603765, + "loss": 0.3517, + "step": 21533 + }, + { + "epoch": 1.2031455149872894, + "grad_norm": 0.5902232527732849, + "learning_rate": 0.0004000728372926939, + "loss": 0.4748, + "step": 21534 + }, + { + "epoch": 1.2032013855909713, + "grad_norm": 0.4720207452774048, + "learning_rate": 0.00040004482294935006, + "loss": 0.3653, + "step": 21535 + }, + { + "epoch": 1.2032572561946533, + "grad_norm": 0.4094485640525818, + "learning_rate": 0.0004000168086060063, + "loss": 0.4815, + "step": 21536 + }, + { + "epoch": 1.203313126798335, + "grad_norm": 0.5844366550445557, + "learning_rate": 0.00039998879426266247, + "loss": 0.4412, + "step": 21537 + }, + { + "epoch": 1.203368997402017, + "grad_norm": 0.6237966418266296, + "learning_rate": 0.0003999607799193187, + "loss": 0.4539, + "step": 21538 + }, + { + "epoch": 1.203424868005699, + "grad_norm": 0.8236005902290344, + "learning_rate": 0.00039993276557597493, + "loss": 0.4412, + "step": 21539 + }, + { + "epoch": 1.2034807386093807, + "grad_norm": 0.8514248132705688, + "learning_rate": 0.0003999047512326311, + "loss": 0.5026, + "step": 21540 + }, + { + "epoch": 1.2035366092130626, + "grad_norm": 0.3849639892578125, + "learning_rate": 0.00039987673688928735, + "loss": 0.3795, + "step": 21541 + }, + { + "epoch": 1.2035924798167443, + "grad_norm": 0.5346600413322449, + "learning_rate": 0.0003998487225459435, + "loss": 0.4223, + "step": 21542 + }, + { + "epoch": 1.2036483504204263, + "grad_norm": 0.5769097208976746, + "learning_rate": 0.00039982070820259976, + "loss": 0.4208, + "step": 21543 + }, + { + "epoch": 1.2037042210241082, + "grad_norm": 0.34941428899765015, + "learning_rate": 0.0003997926938592559, + "loss": 0.3689, + "step": 21544 + }, + { + "epoch": 1.20376009162779, + "grad_norm": 0.6228439807891846, + "learning_rate": 0.00039976467951591217, + "loss": 0.49, + "step": 21545 + }, + { + "epoch": 1.203815962231472, + "grad_norm": 0.922775387763977, + "learning_rate": 0.0003997366651725683, + "loss": 0.4217, + "step": 21546 + }, + { + "epoch": 1.2038718328351539, + "grad_norm": 0.5753631591796875, + "learning_rate": 0.0003997086508292246, + "loss": 0.5344, + "step": 21547 + }, + { + "epoch": 1.2039277034388356, + "grad_norm": 0.3059348464012146, + "learning_rate": 0.0003996806364858808, + "loss": 0.336, + "step": 21548 + }, + { + "epoch": 1.2039835740425175, + "grad_norm": 4.483749866485596, + "learning_rate": 0.000399652622142537, + "loss": 0.4514, + "step": 21549 + }, + { + "epoch": 1.2040394446461995, + "grad_norm": 0.42004063725471497, + "learning_rate": 0.0003996246077991932, + "loss": 0.4331, + "step": 21550 + }, + { + "epoch": 1.2040953152498812, + "grad_norm": 0.7768537402153015, + "learning_rate": 0.0003995965934558494, + "loss": 0.476, + "step": 21551 + }, + { + "epoch": 1.2041511858535632, + "grad_norm": 0.6286733150482178, + "learning_rate": 0.0003995685791125056, + "loss": 0.4932, + "step": 21552 + }, + { + "epoch": 1.204207056457245, + "grad_norm": 0.45300501585006714, + "learning_rate": 0.0003995405647691618, + "loss": 0.5008, + "step": 21553 + }, + { + "epoch": 1.2042629270609269, + "grad_norm": 0.43664419651031494, + "learning_rate": 0.000399512550425818, + "loss": 0.3827, + "step": 21554 + }, + { + "epoch": 1.2043187976646088, + "grad_norm": 0.42830878496170044, + "learning_rate": 0.0003994845360824743, + "loss": 0.4036, + "step": 21555 + }, + { + "epoch": 1.2043746682682905, + "grad_norm": 1.5465083122253418, + "learning_rate": 0.00039945652173913043, + "loss": 0.4128, + "step": 21556 + }, + { + "epoch": 1.2044305388719725, + "grad_norm": 0.3796955943107605, + "learning_rate": 0.0003994285073957867, + "loss": 0.3919, + "step": 21557 + }, + { + "epoch": 1.2044864094756544, + "grad_norm": 0.4201398193836212, + "learning_rate": 0.00039940049305244285, + "loss": 0.444, + "step": 21558 + }, + { + "epoch": 1.2045422800793362, + "grad_norm": 0.6139762997627258, + "learning_rate": 0.00039937247870909905, + "loss": 0.4231, + "step": 21559 + }, + { + "epoch": 1.2045981506830181, + "grad_norm": 0.34903109073638916, + "learning_rate": 0.00039934446436575526, + "loss": 0.3405, + "step": 21560 + }, + { + "epoch": 1.2046540212867, + "grad_norm": 0.6452236771583557, + "learning_rate": 0.00039931645002241146, + "loss": 0.3474, + "step": 21561 + }, + { + "epoch": 1.2047098918903818, + "grad_norm": 0.3806743919849396, + "learning_rate": 0.0003992884356790677, + "loss": 0.4443, + "step": 21562 + }, + { + "epoch": 1.2047657624940638, + "grad_norm": 0.4278491735458374, + "learning_rate": 0.0003992604213357239, + "loss": 0.3904, + "step": 21563 + }, + { + "epoch": 1.2048216330977457, + "grad_norm": 1.814373254776001, + "learning_rate": 0.00039923240699238013, + "loss": 0.4347, + "step": 21564 + }, + { + "epoch": 1.2048775037014274, + "grad_norm": 0.4204850196838379, + "learning_rate": 0.0003992043926490363, + "loss": 0.3896, + "step": 21565 + }, + { + "epoch": 1.2049333743051094, + "grad_norm": 0.4882332384586334, + "learning_rate": 0.00039917637830569255, + "loss": 0.5241, + "step": 21566 + }, + { + "epoch": 1.2049892449087913, + "grad_norm": 0.4039786756038666, + "learning_rate": 0.0003991483639623487, + "loss": 0.416, + "step": 21567 + }, + { + "epoch": 1.205045115512473, + "grad_norm": 0.3706180453300476, + "learning_rate": 0.00039912034961900496, + "loss": 0.3139, + "step": 21568 + }, + { + "epoch": 1.205100986116155, + "grad_norm": 1.0054551362991333, + "learning_rate": 0.0003990923352756611, + "loss": 0.3587, + "step": 21569 + }, + { + "epoch": 1.205156856719837, + "grad_norm": 0.3731605112552643, + "learning_rate": 0.00039906432093231737, + "loss": 0.3846, + "step": 21570 + }, + { + "epoch": 1.2052127273235187, + "grad_norm": 0.44739142060279846, + "learning_rate": 0.0003990363065889736, + "loss": 0.5589, + "step": 21571 + }, + { + "epoch": 1.2052685979272006, + "grad_norm": 0.41846051812171936, + "learning_rate": 0.0003990082922456298, + "loss": 0.4556, + "step": 21572 + }, + { + "epoch": 1.2053244685308826, + "grad_norm": 0.413831502199173, + "learning_rate": 0.000398980277902286, + "loss": 0.4673, + "step": 21573 + }, + { + "epoch": 1.2053803391345643, + "grad_norm": 0.4175083041191101, + "learning_rate": 0.0003989522635589422, + "loss": 0.3075, + "step": 21574 + }, + { + "epoch": 1.2054362097382463, + "grad_norm": 0.6315475702285767, + "learning_rate": 0.0003989242492155984, + "loss": 0.415, + "step": 21575 + }, + { + "epoch": 1.205492080341928, + "grad_norm": 0.5263885855674744, + "learning_rate": 0.0003988962348722546, + "loss": 0.3603, + "step": 21576 + }, + { + "epoch": 1.20554795094561, + "grad_norm": 0.5101985335350037, + "learning_rate": 0.0003988682205289108, + "loss": 0.3308, + "step": 21577 + }, + { + "epoch": 1.205603821549292, + "grad_norm": 0.539709210395813, + "learning_rate": 0.00039884020618556707, + "loss": 0.4184, + "step": 21578 + }, + { + "epoch": 1.2056596921529736, + "grad_norm": 0.37249842286109924, + "learning_rate": 0.0003988121918422232, + "loss": 0.3805, + "step": 21579 + }, + { + "epoch": 1.2057155627566556, + "grad_norm": 0.692095160484314, + "learning_rate": 0.00039878417749887943, + "loss": 0.4077, + "step": 21580 + }, + { + "epoch": 1.2057714333603375, + "grad_norm": 0.4884965419769287, + "learning_rate": 0.00039875616315553563, + "loss": 0.3786, + "step": 21581 + }, + { + "epoch": 1.2058273039640193, + "grad_norm": 0.6780127286911011, + "learning_rate": 0.00039872814881219184, + "loss": 0.4794, + "step": 21582 + }, + { + "epoch": 1.2058831745677012, + "grad_norm": 0.46858808398246765, + "learning_rate": 0.00039870013446884804, + "loss": 0.3909, + "step": 21583 + }, + { + "epoch": 1.2059390451713832, + "grad_norm": 0.4734932482242584, + "learning_rate": 0.00039867212012550425, + "loss": 0.5019, + "step": 21584 + }, + { + "epoch": 1.205994915775065, + "grad_norm": 0.5211474895477295, + "learning_rate": 0.00039864410578216046, + "loss": 0.4355, + "step": 21585 + }, + { + "epoch": 1.2060507863787469, + "grad_norm": 1.1425402164459229, + "learning_rate": 0.00039861609143881666, + "loss": 0.5992, + "step": 21586 + }, + { + "epoch": 1.2061066569824286, + "grad_norm": 0.6003851890563965, + "learning_rate": 0.0003985880770954729, + "loss": 0.5373, + "step": 21587 + }, + { + "epoch": 1.2061625275861105, + "grad_norm": 0.4438078999519348, + "learning_rate": 0.0003985600627521291, + "loss": 0.4722, + "step": 21588 + }, + { + "epoch": 1.2062183981897925, + "grad_norm": 0.9972470998764038, + "learning_rate": 0.00039853204840878533, + "loss": 0.5134, + "step": 21589 + }, + { + "epoch": 1.2062742687934742, + "grad_norm": 0.6115269660949707, + "learning_rate": 0.0003985040340654415, + "loss": 0.439, + "step": 21590 + }, + { + "epoch": 1.2063301393971562, + "grad_norm": 0.6973984241485596, + "learning_rate": 0.00039847601972209775, + "loss": 0.389, + "step": 21591 + }, + { + "epoch": 1.2063860100008381, + "grad_norm": 0.45814400911331177, + "learning_rate": 0.0003984480053787539, + "loss": 0.3374, + "step": 21592 + }, + { + "epoch": 1.2064418806045198, + "grad_norm": 1.0681427717208862, + "learning_rate": 0.00039841999103541016, + "loss": 0.4489, + "step": 21593 + }, + { + "epoch": 1.2064977512082018, + "grad_norm": 0.621804416179657, + "learning_rate": 0.00039839197669206636, + "loss": 0.3543, + "step": 21594 + }, + { + "epoch": 1.2065536218118837, + "grad_norm": 0.5867434740066528, + "learning_rate": 0.00039836396234872257, + "loss": 0.441, + "step": 21595 + }, + { + "epoch": 1.2066094924155655, + "grad_norm": 0.3256969451904297, + "learning_rate": 0.0003983359480053788, + "loss": 0.3485, + "step": 21596 + }, + { + "epoch": 1.2066653630192474, + "grad_norm": 0.48101121187210083, + "learning_rate": 0.000398307933662035, + "loss": 0.4385, + "step": 21597 + }, + { + "epoch": 1.2067212336229294, + "grad_norm": 0.6577916741371155, + "learning_rate": 0.0003982799193186912, + "loss": 0.4724, + "step": 21598 + }, + { + "epoch": 1.206777104226611, + "grad_norm": 0.48731115460395813, + "learning_rate": 0.0003982519049753474, + "loss": 0.3821, + "step": 21599 + }, + { + "epoch": 1.206832974830293, + "grad_norm": 0.8771786093711853, + "learning_rate": 0.0003982238906320036, + "loss": 0.4912, + "step": 21600 + }, + { + "epoch": 1.206888845433975, + "grad_norm": 0.7074530720710754, + "learning_rate": 0.00039819587628865975, + "loss": 0.408, + "step": 21601 + }, + { + "epoch": 1.2069447160376567, + "grad_norm": 0.3967683017253876, + "learning_rate": 0.000398167861945316, + "loss": 0.4058, + "step": 21602 + }, + { + "epoch": 1.2070005866413387, + "grad_norm": 1.1660258769989014, + "learning_rate": 0.0003981398476019722, + "loss": 0.6182, + "step": 21603 + }, + { + "epoch": 1.2070564572450206, + "grad_norm": 0.39022311568260193, + "learning_rate": 0.0003981118332586284, + "loss": 0.4726, + "step": 21604 + }, + { + "epoch": 1.2071123278487024, + "grad_norm": 0.40434950590133667, + "learning_rate": 0.0003980838189152846, + "loss": 0.367, + "step": 21605 + }, + { + "epoch": 1.2071681984523843, + "grad_norm": 0.758584201335907, + "learning_rate": 0.00039805580457194083, + "loss": 0.4449, + "step": 21606 + }, + { + "epoch": 1.2072240690560663, + "grad_norm": 0.4580763578414917, + "learning_rate": 0.00039802779022859704, + "loss": 0.3968, + "step": 21607 + }, + { + "epoch": 1.207279939659748, + "grad_norm": 0.894011378288269, + "learning_rate": 0.00039799977588525324, + "loss": 0.4661, + "step": 21608 + }, + { + "epoch": 1.20733581026343, + "grad_norm": 0.4893732964992523, + "learning_rate": 0.00039797176154190945, + "loss": 0.5979, + "step": 21609 + }, + { + "epoch": 1.2073916808671117, + "grad_norm": 0.41947251558303833, + "learning_rate": 0.0003979437471985657, + "loss": 0.5692, + "step": 21610 + }, + { + "epoch": 1.2074475514707936, + "grad_norm": 0.3776715099811554, + "learning_rate": 0.00039791573285522186, + "loss": 0.4446, + "step": 21611 + }, + { + "epoch": 1.2075034220744756, + "grad_norm": 0.4979943633079529, + "learning_rate": 0.0003978877185118781, + "loss": 0.3982, + "step": 21612 + }, + { + "epoch": 1.2075592926781573, + "grad_norm": 0.8407187461853027, + "learning_rate": 0.0003978597041685343, + "loss": 0.5081, + "step": 21613 + }, + { + "epoch": 1.2076151632818393, + "grad_norm": 0.41821739077568054, + "learning_rate": 0.00039783168982519053, + "loss": 0.3618, + "step": 21614 + }, + { + "epoch": 1.2076710338855212, + "grad_norm": 0.4500957429409027, + "learning_rate": 0.0003978036754818467, + "loss": 0.5904, + "step": 21615 + }, + { + "epoch": 1.207726904489203, + "grad_norm": 0.3127511143684387, + "learning_rate": 0.00039777566113850295, + "loss": 0.4151, + "step": 21616 + }, + { + "epoch": 1.207782775092885, + "grad_norm": 0.48544996976852417, + "learning_rate": 0.0003977476467951591, + "loss": 0.4208, + "step": 21617 + }, + { + "epoch": 1.2078386456965668, + "grad_norm": 0.3648468852043152, + "learning_rate": 0.00039771963245181536, + "loss": 0.4241, + "step": 21618 + }, + { + "epoch": 1.2078945163002486, + "grad_norm": 1.4501152038574219, + "learning_rate": 0.00039769161810847156, + "loss": 0.3725, + "step": 21619 + }, + { + "epoch": 1.2079503869039305, + "grad_norm": 0.343585342168808, + "learning_rate": 0.00039766360376512777, + "loss": 0.4253, + "step": 21620 + }, + { + "epoch": 1.2080062575076123, + "grad_norm": 2.254070281982422, + "learning_rate": 0.000397635589421784, + "loss": 0.4866, + "step": 21621 + }, + { + "epoch": 1.2080621281112942, + "grad_norm": 7.608532905578613, + "learning_rate": 0.0003976075750784401, + "loss": 0.4148, + "step": 21622 + }, + { + "epoch": 1.2081179987149762, + "grad_norm": 0.4442855417728424, + "learning_rate": 0.0003975795607350964, + "loss": 0.4253, + "step": 21623 + }, + { + "epoch": 1.208173869318658, + "grad_norm": 7.6061272621154785, + "learning_rate": 0.00039755154639175254, + "loss": 0.4194, + "step": 21624 + }, + { + "epoch": 1.2082297399223398, + "grad_norm": 0.4114547371864319, + "learning_rate": 0.0003975235320484088, + "loss": 0.4292, + "step": 21625 + }, + { + "epoch": 1.2082856105260218, + "grad_norm": 0.4195103943347931, + "learning_rate": 0.000397495517705065, + "loss": 0.2737, + "step": 21626 + }, + { + "epoch": 1.2083414811297035, + "grad_norm": 0.7487592101097107, + "learning_rate": 0.0003974675033617212, + "loss": 0.6669, + "step": 21627 + }, + { + "epoch": 1.2083973517333855, + "grad_norm": 0.5150639414787292, + "learning_rate": 0.0003974394890183774, + "loss": 0.5141, + "step": 21628 + }, + { + "epoch": 1.2084532223370674, + "grad_norm": 0.3895978331565857, + "learning_rate": 0.0003974114746750336, + "loss": 0.3738, + "step": 21629 + }, + { + "epoch": 1.2085090929407492, + "grad_norm": 0.3447321951389313, + "learning_rate": 0.0003973834603316898, + "loss": 0.4436, + "step": 21630 + }, + { + "epoch": 1.208564963544431, + "grad_norm": 1.4119763374328613, + "learning_rate": 0.00039735544598834603, + "loss": 0.4407, + "step": 21631 + }, + { + "epoch": 1.208620834148113, + "grad_norm": 0.40854713320732117, + "learning_rate": 0.00039732743164500224, + "loss": 0.5441, + "step": 21632 + }, + { + "epoch": 1.2086767047517948, + "grad_norm": 0.3964885175228119, + "learning_rate": 0.0003972994173016585, + "loss": 0.3854, + "step": 21633 + }, + { + "epoch": 1.2087325753554767, + "grad_norm": 0.4294934570789337, + "learning_rate": 0.00039727140295831465, + "loss": 0.4551, + "step": 21634 + }, + { + "epoch": 1.2087884459591587, + "grad_norm": 0.516025722026825, + "learning_rate": 0.0003972433886149709, + "loss": 0.4257, + "step": 21635 + }, + { + "epoch": 1.2088443165628404, + "grad_norm": 0.477682888507843, + "learning_rate": 0.00039721537427162706, + "loss": 0.6991, + "step": 21636 + }, + { + "epoch": 1.2089001871665224, + "grad_norm": 0.3706193268299103, + "learning_rate": 0.0003971873599282833, + "loss": 0.426, + "step": 21637 + }, + { + "epoch": 1.2089560577702043, + "grad_norm": 0.4007098078727722, + "learning_rate": 0.0003971593455849395, + "loss": 0.4453, + "step": 21638 + }, + { + "epoch": 1.209011928373886, + "grad_norm": 0.38510119915008545, + "learning_rate": 0.00039713133124159573, + "loss": 0.3008, + "step": 21639 + }, + { + "epoch": 1.209067798977568, + "grad_norm": 0.3716702461242676, + "learning_rate": 0.0003971033168982519, + "loss": 0.3795, + "step": 21640 + }, + { + "epoch": 1.20912366958125, + "grad_norm": 0.4667091965675354, + "learning_rate": 0.00039707530255490814, + "loss": 0.4191, + "step": 21641 + }, + { + "epoch": 1.2091795401849317, + "grad_norm": 0.3308575451374054, + "learning_rate": 0.00039704728821156435, + "loss": 0.3968, + "step": 21642 + }, + { + "epoch": 1.2092354107886136, + "grad_norm": 0.385418176651001, + "learning_rate": 0.0003970192738682205, + "loss": 0.5436, + "step": 21643 + }, + { + "epoch": 1.2092912813922954, + "grad_norm": 0.3967883586883545, + "learning_rate": 0.00039699125952487676, + "loss": 0.388, + "step": 21644 + }, + { + "epoch": 1.2093471519959773, + "grad_norm": 0.40781131386756897, + "learning_rate": 0.0003969632451815329, + "loss": 0.3857, + "step": 21645 + }, + { + "epoch": 1.2094030225996593, + "grad_norm": 0.35294386744499207, + "learning_rate": 0.0003969352308381892, + "loss": 0.3637, + "step": 21646 + }, + { + "epoch": 1.209458893203341, + "grad_norm": 1.4403133392333984, + "learning_rate": 0.0003969072164948453, + "loss": 0.3237, + "step": 21647 + }, + { + "epoch": 1.209514763807023, + "grad_norm": 1.019602656364441, + "learning_rate": 0.0003968792021515016, + "loss": 0.4667, + "step": 21648 + }, + { + "epoch": 1.209570634410705, + "grad_norm": 0.4305775463581085, + "learning_rate": 0.0003968511878081578, + "loss": 0.3887, + "step": 21649 + }, + { + "epoch": 1.2096265050143866, + "grad_norm": 1.463646650314331, + "learning_rate": 0.000396823173464814, + "loss": 0.4654, + "step": 21650 + }, + { + "epoch": 1.2096823756180686, + "grad_norm": 0.42081528902053833, + "learning_rate": 0.0003967951591214702, + "loss": 0.4228, + "step": 21651 + }, + { + "epoch": 1.2097382462217503, + "grad_norm": 0.5957900881767273, + "learning_rate": 0.0003967671447781264, + "loss": 0.4848, + "step": 21652 + }, + { + "epoch": 1.2097941168254323, + "grad_norm": 0.48647916316986084, + "learning_rate": 0.0003967391304347826, + "loss": 0.4083, + "step": 21653 + }, + { + "epoch": 1.2098499874291142, + "grad_norm": 0.359414666891098, + "learning_rate": 0.0003967111160914388, + "loss": 0.3168, + "step": 21654 + }, + { + "epoch": 1.209905858032796, + "grad_norm": 0.5486290454864502, + "learning_rate": 0.000396683101748095, + "loss": 0.3981, + "step": 21655 + }, + { + "epoch": 1.2099617286364779, + "grad_norm": 0.5243586897850037, + "learning_rate": 0.00039665508740475123, + "loss": 0.3551, + "step": 21656 + }, + { + "epoch": 1.2100175992401598, + "grad_norm": 1.6905564069747925, + "learning_rate": 0.00039662707306140744, + "loss": 0.5861, + "step": 21657 + }, + { + "epoch": 1.2100734698438416, + "grad_norm": 0.4881708323955536, + "learning_rate": 0.0003965990587180637, + "loss": 0.3942, + "step": 21658 + }, + { + "epoch": 1.2101293404475235, + "grad_norm": 0.43971604108810425, + "learning_rate": 0.00039657104437471985, + "loss": 0.4136, + "step": 21659 + }, + { + "epoch": 1.2101852110512055, + "grad_norm": 0.44318363070487976, + "learning_rate": 0.0003965430300313761, + "loss": 0.4839, + "step": 21660 + }, + { + "epoch": 1.2102410816548872, + "grad_norm": 0.39993155002593994, + "learning_rate": 0.00039651501568803226, + "loss": 0.3599, + "step": 21661 + }, + { + "epoch": 1.2102969522585691, + "grad_norm": 0.4240773320198059, + "learning_rate": 0.0003964870013446885, + "loss": 0.4406, + "step": 21662 + }, + { + "epoch": 1.210352822862251, + "grad_norm": 0.7957979440689087, + "learning_rate": 0.00039645898700134467, + "loss": 0.3956, + "step": 21663 + }, + { + "epoch": 1.2104086934659328, + "grad_norm": 0.6829054951667786, + "learning_rate": 0.00039643097265800093, + "loss": 0.4211, + "step": 21664 + }, + { + "epoch": 1.2104645640696148, + "grad_norm": 1.157313585281372, + "learning_rate": 0.00039640295831465714, + "loss": 0.4259, + "step": 21665 + }, + { + "epoch": 1.2105204346732967, + "grad_norm": 0.5519260764122009, + "learning_rate": 0.0003963749439713133, + "loss": 0.4182, + "step": 21666 + }, + { + "epoch": 1.2105763052769785, + "grad_norm": 1.9977655410766602, + "learning_rate": 0.00039634692962796955, + "loss": 0.5335, + "step": 21667 + }, + { + "epoch": 1.2106321758806604, + "grad_norm": 0.5255705118179321, + "learning_rate": 0.0003963189152846257, + "loss": 0.4106, + "step": 21668 + }, + { + "epoch": 1.2106880464843424, + "grad_norm": 0.6461685299873352, + "learning_rate": 0.00039629090094128196, + "loss": 0.5323, + "step": 21669 + }, + { + "epoch": 1.210743917088024, + "grad_norm": 0.38813653588294983, + "learning_rate": 0.0003962628865979381, + "loss": 0.3763, + "step": 21670 + }, + { + "epoch": 1.210799787691706, + "grad_norm": 0.5178031921386719, + "learning_rate": 0.0003962348722545944, + "loss": 0.38, + "step": 21671 + }, + { + "epoch": 1.210855658295388, + "grad_norm": 1.1521176099777222, + "learning_rate": 0.0003962068579112505, + "loss": 0.5001, + "step": 21672 + }, + { + "epoch": 1.2109115288990697, + "grad_norm": 0.3842429518699646, + "learning_rate": 0.0003961788435679068, + "loss": 0.3672, + "step": 21673 + }, + { + "epoch": 1.2109673995027517, + "grad_norm": 1.5254346132278442, + "learning_rate": 0.000396150829224563, + "loss": 0.6227, + "step": 21674 + }, + { + "epoch": 1.2110232701064336, + "grad_norm": 1.739448070526123, + "learning_rate": 0.0003961228148812192, + "loss": 0.4527, + "step": 21675 + }, + { + "epoch": 1.2110791407101154, + "grad_norm": 0.832627534866333, + "learning_rate": 0.0003960948005378754, + "loss": 0.5041, + "step": 21676 + }, + { + "epoch": 1.2111350113137973, + "grad_norm": 0.41034606099128723, + "learning_rate": 0.0003960667861945316, + "loss": 0.3861, + "step": 21677 + }, + { + "epoch": 1.211190881917479, + "grad_norm": 0.9859232306480408, + "learning_rate": 0.0003960387718511878, + "loss": 0.424, + "step": 21678 + }, + { + "epoch": 1.211246752521161, + "grad_norm": 0.5198342800140381, + "learning_rate": 0.000396010757507844, + "loss": 0.5104, + "step": 21679 + }, + { + "epoch": 1.211302623124843, + "grad_norm": 3.823765993118286, + "learning_rate": 0.0003959827431645002, + "loss": 0.3905, + "step": 21680 + }, + { + "epoch": 1.2113584937285247, + "grad_norm": 2.163594961166382, + "learning_rate": 0.0003959547288211565, + "loss": 0.542, + "step": 21681 + }, + { + "epoch": 1.2114143643322066, + "grad_norm": 0.49924588203430176, + "learning_rate": 0.00039592671447781264, + "loss": 0.2964, + "step": 21682 + }, + { + "epoch": 1.2114702349358886, + "grad_norm": 0.5575711727142334, + "learning_rate": 0.0003958987001344689, + "loss": 0.2947, + "step": 21683 + }, + { + "epoch": 1.2115261055395703, + "grad_norm": 0.3407621383666992, + "learning_rate": 0.00039587068579112505, + "loss": 0.3539, + "step": 21684 + }, + { + "epoch": 1.2115819761432522, + "grad_norm": 0.4081416428089142, + "learning_rate": 0.0003958426714477813, + "loss": 0.4129, + "step": 21685 + }, + { + "epoch": 1.211637846746934, + "grad_norm": 0.4464820623397827, + "learning_rate": 0.00039581465710443746, + "loss": 0.4185, + "step": 21686 + }, + { + "epoch": 1.211693717350616, + "grad_norm": 0.40370994806289673, + "learning_rate": 0.00039578664276109367, + "loss": 0.4407, + "step": 21687 + }, + { + "epoch": 1.2117495879542979, + "grad_norm": 0.353770911693573, + "learning_rate": 0.0003957586284177499, + "loss": 0.3083, + "step": 21688 + }, + { + "epoch": 1.2118054585579796, + "grad_norm": 0.9552428126335144, + "learning_rate": 0.0003957306140744061, + "loss": 0.4553, + "step": 21689 + }, + { + "epoch": 1.2118613291616616, + "grad_norm": 1.58132803440094, + "learning_rate": 0.00039570259973106234, + "loss": 0.5378, + "step": 21690 + }, + { + "epoch": 1.2119171997653435, + "grad_norm": 0.390926718711853, + "learning_rate": 0.0003956745853877185, + "loss": 0.3862, + "step": 21691 + }, + { + "epoch": 1.2119730703690252, + "grad_norm": 0.5612441897392273, + "learning_rate": 0.00039564657104437475, + "loss": 0.5418, + "step": 21692 + }, + { + "epoch": 1.2120289409727072, + "grad_norm": 1.4157379865646362, + "learning_rate": 0.0003956185567010309, + "loss": 0.5144, + "step": 21693 + }, + { + "epoch": 1.2120848115763891, + "grad_norm": 0.3649585247039795, + "learning_rate": 0.00039559054235768716, + "loss": 0.3644, + "step": 21694 + }, + { + "epoch": 1.2121406821800709, + "grad_norm": 0.7483549118041992, + "learning_rate": 0.0003955625280143433, + "loss": 0.45, + "step": 21695 + }, + { + "epoch": 1.2121965527837528, + "grad_norm": 0.49825525283813477, + "learning_rate": 0.0003955345136709996, + "loss": 0.3484, + "step": 21696 + }, + { + "epoch": 1.2122524233874348, + "grad_norm": 0.46685731410980225, + "learning_rate": 0.0003955064993276558, + "loss": 0.4457, + "step": 21697 + }, + { + "epoch": 1.2123082939911165, + "grad_norm": 3.518693208694458, + "learning_rate": 0.000395478484984312, + "loss": 0.4562, + "step": 21698 + }, + { + "epoch": 1.2123641645947985, + "grad_norm": 0.4617132544517517, + "learning_rate": 0.0003954504706409682, + "loss": 0.455, + "step": 21699 + }, + { + "epoch": 1.2124200351984804, + "grad_norm": 0.6674181818962097, + "learning_rate": 0.0003954224562976244, + "loss": 0.4319, + "step": 21700 + }, + { + "epoch": 1.2124759058021621, + "grad_norm": 0.4075428247451782, + "learning_rate": 0.0003953944419542806, + "loss": 0.4275, + "step": 21701 + }, + { + "epoch": 1.212531776405844, + "grad_norm": 0.773970365524292, + "learning_rate": 0.0003953664276109368, + "loss": 0.4564, + "step": 21702 + }, + { + "epoch": 1.212587647009526, + "grad_norm": 0.8073118925094604, + "learning_rate": 0.000395338413267593, + "loss": 0.3575, + "step": 21703 + }, + { + "epoch": 1.2126435176132078, + "grad_norm": 0.6293278336524963, + "learning_rate": 0.0003953103989242493, + "loss": 0.4323, + "step": 21704 + }, + { + "epoch": 1.2126993882168897, + "grad_norm": 0.45682501792907715, + "learning_rate": 0.0003952823845809054, + "loss": 0.4065, + "step": 21705 + }, + { + "epoch": 1.2127552588205717, + "grad_norm": 0.5485930442810059, + "learning_rate": 0.0003952543702375617, + "loss": 0.5345, + "step": 21706 + }, + { + "epoch": 1.2128111294242534, + "grad_norm": 0.3583813011646271, + "learning_rate": 0.00039522635589421784, + "loss": 0.3953, + "step": 21707 + }, + { + "epoch": 1.2128670000279353, + "grad_norm": 0.4102644622325897, + "learning_rate": 0.00039519834155087404, + "loss": 0.3827, + "step": 21708 + }, + { + "epoch": 1.212922870631617, + "grad_norm": 0.33833619952201843, + "learning_rate": 0.00039517032720753025, + "loss": 0.415, + "step": 21709 + }, + { + "epoch": 1.212978741235299, + "grad_norm": 0.38612401485443115, + "learning_rate": 0.00039514231286418645, + "loss": 0.3534, + "step": 21710 + }, + { + "epoch": 1.213034611838981, + "grad_norm": 0.36602938175201416, + "learning_rate": 0.00039511429852084266, + "loss": 0.3154, + "step": 21711 + }, + { + "epoch": 1.2130904824426627, + "grad_norm": 0.510066032409668, + "learning_rate": 0.00039508628417749887, + "loss": 0.4429, + "step": 21712 + }, + { + "epoch": 1.2131463530463447, + "grad_norm": 0.626248836517334, + "learning_rate": 0.0003950582698341551, + "loss": 0.3615, + "step": 21713 + }, + { + "epoch": 1.2132022236500266, + "grad_norm": 0.4390048682689667, + "learning_rate": 0.0003950302554908113, + "loss": 0.4932, + "step": 21714 + }, + { + "epoch": 1.2132580942537083, + "grad_norm": 0.6874333024024963, + "learning_rate": 0.00039500224114746754, + "loss": 0.3215, + "step": 21715 + }, + { + "epoch": 1.2133139648573903, + "grad_norm": 0.42703142762184143, + "learning_rate": 0.0003949742268041237, + "loss": 0.3252, + "step": 21716 + }, + { + "epoch": 1.2133698354610722, + "grad_norm": 0.7478651404380798, + "learning_rate": 0.00039494621246077995, + "loss": 0.5545, + "step": 21717 + }, + { + "epoch": 1.213425706064754, + "grad_norm": 0.720202624797821, + "learning_rate": 0.0003949181981174361, + "loss": 0.4056, + "step": 21718 + }, + { + "epoch": 1.213481576668436, + "grad_norm": 0.9805912971496582, + "learning_rate": 0.00039489018377409236, + "loss": 0.4991, + "step": 21719 + }, + { + "epoch": 1.2135374472721177, + "grad_norm": 0.4338807165622711, + "learning_rate": 0.00039486216943074857, + "loss": 0.3901, + "step": 21720 + }, + { + "epoch": 1.2135933178757996, + "grad_norm": 0.37030571699142456, + "learning_rate": 0.00039483415508740477, + "loss": 0.3745, + "step": 21721 + }, + { + "epoch": 1.2136491884794816, + "grad_norm": 0.7881125807762146, + "learning_rate": 0.000394806140744061, + "loss": 0.4673, + "step": 21722 + }, + { + "epoch": 1.2137050590831633, + "grad_norm": 0.3654315173625946, + "learning_rate": 0.0003947781264007172, + "loss": 0.4548, + "step": 21723 + }, + { + "epoch": 1.2137609296868452, + "grad_norm": 0.39163827896118164, + "learning_rate": 0.0003947501120573734, + "loss": 0.3755, + "step": 21724 + }, + { + "epoch": 1.2138168002905272, + "grad_norm": 0.3292686939239502, + "learning_rate": 0.0003947220977140296, + "loss": 0.3294, + "step": 21725 + }, + { + "epoch": 1.213872670894209, + "grad_norm": 1.443267583847046, + "learning_rate": 0.0003946940833706858, + "loss": 0.4973, + "step": 21726 + }, + { + "epoch": 1.2139285414978909, + "grad_norm": 0.4499897062778473, + "learning_rate": 0.000394666069027342, + "loss": 0.4667, + "step": 21727 + }, + { + "epoch": 1.2139844121015728, + "grad_norm": 1.2440288066864014, + "learning_rate": 0.0003946380546839982, + "loss": 0.4506, + "step": 21728 + }, + { + "epoch": 1.2140402827052545, + "grad_norm": 0.7247431874275208, + "learning_rate": 0.0003946100403406544, + "loss": 0.4417, + "step": 21729 + }, + { + "epoch": 1.2140961533089365, + "grad_norm": 0.5722503066062927, + "learning_rate": 0.0003945820259973106, + "loss": 0.3675, + "step": 21730 + }, + { + "epoch": 1.2141520239126184, + "grad_norm": 0.396140456199646, + "learning_rate": 0.00039455401165396683, + "loss": 0.4291, + "step": 21731 + }, + { + "epoch": 1.2142078945163002, + "grad_norm": 0.42983588576316833, + "learning_rate": 0.00039452599731062304, + "loss": 0.4274, + "step": 21732 + }, + { + "epoch": 1.2142637651199821, + "grad_norm": 0.4343034029006958, + "learning_rate": 0.00039449798296727924, + "loss": 0.5471, + "step": 21733 + }, + { + "epoch": 1.214319635723664, + "grad_norm": 0.32187965512275696, + "learning_rate": 0.00039446996862393545, + "loss": 0.3834, + "step": 21734 + }, + { + "epoch": 1.2143755063273458, + "grad_norm": 0.5059075951576233, + "learning_rate": 0.00039444195428059165, + "loss": 0.4699, + "step": 21735 + }, + { + "epoch": 1.2144313769310278, + "grad_norm": 0.5340994000434875, + "learning_rate": 0.0003944139399372479, + "loss": 0.5601, + "step": 21736 + }, + { + "epoch": 1.2144872475347097, + "grad_norm": 0.3768419623374939, + "learning_rate": 0.00039438592559390407, + "loss": 0.324, + "step": 21737 + }, + { + "epoch": 1.2145431181383914, + "grad_norm": 0.360627144575119, + "learning_rate": 0.0003943579112505603, + "loss": 0.3558, + "step": 21738 + }, + { + "epoch": 1.2145989887420734, + "grad_norm": 0.8937025666236877, + "learning_rate": 0.0003943298969072165, + "loss": 0.3749, + "step": 21739 + }, + { + "epoch": 1.2146548593457553, + "grad_norm": 0.8202126026153564, + "learning_rate": 0.00039430188256387274, + "loss": 0.4795, + "step": 21740 + }, + { + "epoch": 1.214710729949437, + "grad_norm": 0.5425358414649963, + "learning_rate": 0.0003942738682205289, + "loss": 0.3758, + "step": 21741 + }, + { + "epoch": 1.214766600553119, + "grad_norm": 0.7426279187202454, + "learning_rate": 0.00039424585387718515, + "loss": 0.5383, + "step": 21742 + }, + { + "epoch": 1.2148224711568008, + "grad_norm": 0.5692573189735413, + "learning_rate": 0.0003942178395338413, + "loss": 0.482, + "step": 21743 + }, + { + "epoch": 1.2148783417604827, + "grad_norm": 0.4715287387371063, + "learning_rate": 0.00039418982519049756, + "loss": 0.4059, + "step": 21744 + }, + { + "epoch": 1.2149342123641647, + "grad_norm": 0.40609481930732727, + "learning_rate": 0.00039416181084715377, + "loss": 0.5014, + "step": 21745 + }, + { + "epoch": 1.2149900829678464, + "grad_norm": 0.41076964139938354, + "learning_rate": 0.00039413379650380997, + "loss": 0.5475, + "step": 21746 + }, + { + "epoch": 1.2150459535715283, + "grad_norm": 0.5387173891067505, + "learning_rate": 0.0003941057821604662, + "loss": 0.4368, + "step": 21747 + }, + { + "epoch": 1.2151018241752103, + "grad_norm": 0.3469725251197815, + "learning_rate": 0.0003940777678171224, + "loss": 0.432, + "step": 21748 + }, + { + "epoch": 1.215157694778892, + "grad_norm": 0.9808183908462524, + "learning_rate": 0.0003940497534737786, + "loss": 0.3758, + "step": 21749 + }, + { + "epoch": 1.215213565382574, + "grad_norm": 0.3836555480957031, + "learning_rate": 0.00039402173913043474, + "loss": 0.4709, + "step": 21750 + }, + { + "epoch": 1.215269435986256, + "grad_norm": 0.7030276656150818, + "learning_rate": 0.000393993724787091, + "loss": 0.3586, + "step": 21751 + }, + { + "epoch": 1.2153253065899376, + "grad_norm": 1.1104055643081665, + "learning_rate": 0.0003939657104437472, + "loss": 0.3696, + "step": 21752 + }, + { + "epoch": 1.2153811771936196, + "grad_norm": 0.7099254131317139, + "learning_rate": 0.0003939376961004034, + "loss": 0.4714, + "step": 21753 + }, + { + "epoch": 1.2154370477973013, + "grad_norm": 0.3901706039905548, + "learning_rate": 0.0003939096817570596, + "loss": 0.423, + "step": 21754 + }, + { + "epoch": 1.2154929184009833, + "grad_norm": 0.45917725563049316, + "learning_rate": 0.0003938816674137158, + "loss": 0.5664, + "step": 21755 + }, + { + "epoch": 1.2155487890046652, + "grad_norm": 0.5689057111740112, + "learning_rate": 0.00039385365307037203, + "loss": 0.449, + "step": 21756 + }, + { + "epoch": 1.215604659608347, + "grad_norm": 0.7241020202636719, + "learning_rate": 0.00039382563872702824, + "loss": 0.4471, + "step": 21757 + }, + { + "epoch": 1.215660530212029, + "grad_norm": 0.48595479130744934, + "learning_rate": 0.00039379762438368444, + "loss": 0.4678, + "step": 21758 + }, + { + "epoch": 1.2157164008157109, + "grad_norm": 0.3558432459831238, + "learning_rate": 0.0003937696100403407, + "loss": 0.3556, + "step": 21759 + }, + { + "epoch": 1.2157722714193926, + "grad_norm": 0.6230709552764893, + "learning_rate": 0.00039374159569699685, + "loss": 0.5618, + "step": 21760 + }, + { + "epoch": 1.2158281420230745, + "grad_norm": 0.37911057472229004, + "learning_rate": 0.0003937135813536531, + "loss": 0.3348, + "step": 21761 + }, + { + "epoch": 1.2158840126267565, + "grad_norm": 0.5563651919364929, + "learning_rate": 0.00039368556701030927, + "loss": 0.4357, + "step": 21762 + }, + { + "epoch": 1.2159398832304382, + "grad_norm": 0.6686147451400757, + "learning_rate": 0.0003936575526669655, + "loss": 0.5035, + "step": 21763 + }, + { + "epoch": 1.2159957538341202, + "grad_norm": 0.6569532752037048, + "learning_rate": 0.0003936295383236217, + "loss": 0.5263, + "step": 21764 + }, + { + "epoch": 1.2160516244378021, + "grad_norm": 0.3757701516151428, + "learning_rate": 0.00039360152398027794, + "loss": 0.3603, + "step": 21765 + }, + { + "epoch": 1.2161074950414839, + "grad_norm": 0.7377208471298218, + "learning_rate": 0.0003935735096369341, + "loss": 0.4018, + "step": 21766 + }, + { + "epoch": 1.2161633656451658, + "grad_norm": 0.4327174723148346, + "learning_rate": 0.00039354549529359035, + "loss": 0.4692, + "step": 21767 + }, + { + "epoch": 1.2162192362488478, + "grad_norm": 0.6532443165779114, + "learning_rate": 0.00039351748095024655, + "loss": 0.5052, + "step": 21768 + }, + { + "epoch": 1.2162751068525295, + "grad_norm": 0.3339000344276428, + "learning_rate": 0.00039348946660690276, + "loss": 0.3958, + "step": 21769 + }, + { + "epoch": 1.2163309774562114, + "grad_norm": 0.6143476963043213, + "learning_rate": 0.00039346145226355897, + "loss": 0.5329, + "step": 21770 + }, + { + "epoch": 1.2163868480598934, + "grad_norm": 0.5013025403022766, + "learning_rate": 0.0003934334379202151, + "loss": 0.4845, + "step": 21771 + }, + { + "epoch": 1.2164427186635751, + "grad_norm": 0.4338642656803131, + "learning_rate": 0.0003934054235768714, + "loss": 0.3298, + "step": 21772 + }, + { + "epoch": 1.216498589267257, + "grad_norm": 0.37501707673072815, + "learning_rate": 0.00039337740923352753, + "loss": 0.3765, + "step": 21773 + }, + { + "epoch": 1.216554459870939, + "grad_norm": 0.4513912498950958, + "learning_rate": 0.0003933493948901838, + "loss": 0.3738, + "step": 21774 + }, + { + "epoch": 1.2166103304746207, + "grad_norm": 0.6516028642654419, + "learning_rate": 0.00039332138054684, + "loss": 0.399, + "step": 21775 + }, + { + "epoch": 1.2166662010783027, + "grad_norm": 0.5054274201393127, + "learning_rate": 0.0003932933662034962, + "loss": 0.481, + "step": 21776 + }, + { + "epoch": 1.2167220716819844, + "grad_norm": 0.47329333424568176, + "learning_rate": 0.0003932653518601524, + "loss": 0.4378, + "step": 21777 + }, + { + "epoch": 1.2167779422856664, + "grad_norm": 0.6748711466789246, + "learning_rate": 0.0003932373375168086, + "loss": 0.4836, + "step": 21778 + }, + { + "epoch": 1.2168338128893483, + "grad_norm": 0.7961514592170715, + "learning_rate": 0.0003932093231734648, + "loss": 0.4093, + "step": 21779 + }, + { + "epoch": 1.21688968349303, + "grad_norm": 0.36265578866004944, + "learning_rate": 0.000393181308830121, + "loss": 0.5552, + "step": 21780 + }, + { + "epoch": 1.216945554096712, + "grad_norm": 3.3280439376831055, + "learning_rate": 0.00039315329448677723, + "loss": 0.4588, + "step": 21781 + }, + { + "epoch": 1.217001424700394, + "grad_norm": 0.49034401774406433, + "learning_rate": 0.00039312528014343344, + "loss": 0.3819, + "step": 21782 + }, + { + "epoch": 1.2170572953040757, + "grad_norm": 0.522057056427002, + "learning_rate": 0.00039309726580008964, + "loss": 0.5954, + "step": 21783 + }, + { + "epoch": 1.2171131659077576, + "grad_norm": 0.5839349031448364, + "learning_rate": 0.0003930692514567459, + "loss": 0.4962, + "step": 21784 + }, + { + "epoch": 1.2171690365114396, + "grad_norm": 0.556492030620575, + "learning_rate": 0.00039304123711340205, + "loss": 0.447, + "step": 21785 + }, + { + "epoch": 1.2172249071151213, + "grad_norm": 0.6528459787368774, + "learning_rate": 0.0003930132227700583, + "loss": 0.3741, + "step": 21786 + }, + { + "epoch": 1.2172807777188033, + "grad_norm": 0.4331824481487274, + "learning_rate": 0.00039298520842671446, + "loss": 0.3873, + "step": 21787 + }, + { + "epoch": 1.217336648322485, + "grad_norm": 0.44903767108917236, + "learning_rate": 0.0003929571940833707, + "loss": 0.4262, + "step": 21788 + }, + { + "epoch": 1.217392518926167, + "grad_norm": 0.37415945529937744, + "learning_rate": 0.0003929291797400269, + "loss": 0.3316, + "step": 21789 + }, + { + "epoch": 1.217448389529849, + "grad_norm": 0.6525360345840454, + "learning_rate": 0.00039290116539668314, + "loss": 0.3973, + "step": 21790 + }, + { + "epoch": 1.2175042601335306, + "grad_norm": 0.3534714877605438, + "learning_rate": 0.00039287315105333934, + "loss": 0.3772, + "step": 21791 + }, + { + "epoch": 1.2175601307372126, + "grad_norm": 0.367005854845047, + "learning_rate": 0.0003928451367099955, + "loss": 0.3607, + "step": 21792 + }, + { + "epoch": 1.2176160013408945, + "grad_norm": 2.701216459274292, + "learning_rate": 0.00039281712236665175, + "loss": 0.4675, + "step": 21793 + }, + { + "epoch": 1.2176718719445763, + "grad_norm": 0.3021065294742584, + "learning_rate": 0.0003927891080233079, + "loss": 0.2975, + "step": 21794 + }, + { + "epoch": 1.2177277425482582, + "grad_norm": 0.6308786273002625, + "learning_rate": 0.00039276109367996417, + "loss": 0.5192, + "step": 21795 + }, + { + "epoch": 1.2177836131519402, + "grad_norm": 0.4449220299720764, + "learning_rate": 0.0003927330793366203, + "loss": 0.5481, + "step": 21796 + }, + { + "epoch": 1.217839483755622, + "grad_norm": 0.6427714228630066, + "learning_rate": 0.0003927050649932766, + "loss": 0.3151, + "step": 21797 + }, + { + "epoch": 1.2178953543593038, + "grad_norm": 1.8664333820343018, + "learning_rate": 0.00039267705064993273, + "loss": 0.4753, + "step": 21798 + }, + { + "epoch": 1.2179512249629858, + "grad_norm": 0.39346545934677124, + "learning_rate": 0.000392649036306589, + "loss": 0.3856, + "step": 21799 + }, + { + "epoch": 1.2180070955666675, + "grad_norm": 0.5228044390678406, + "learning_rate": 0.0003926210219632452, + "loss": 0.3912, + "step": 21800 + }, + { + "epoch": 1.2180629661703495, + "grad_norm": 0.3601877987384796, + "learning_rate": 0.0003925930076199014, + "loss": 0.4181, + "step": 21801 + }, + { + "epoch": 1.2181188367740314, + "grad_norm": 0.6187717318534851, + "learning_rate": 0.0003925649932765576, + "loss": 0.5403, + "step": 21802 + }, + { + "epoch": 1.2181747073777132, + "grad_norm": 0.6482531428337097, + "learning_rate": 0.0003925369789332138, + "loss": 0.4306, + "step": 21803 + }, + { + "epoch": 1.2182305779813951, + "grad_norm": 0.39071545004844666, + "learning_rate": 0.00039250896458987, + "loss": 0.3084, + "step": 21804 + }, + { + "epoch": 1.218286448585077, + "grad_norm": 0.4309264123439789, + "learning_rate": 0.0003924809502465262, + "loss": 0.4009, + "step": 21805 + }, + { + "epoch": 1.2183423191887588, + "grad_norm": 0.8949998021125793, + "learning_rate": 0.00039245293590318243, + "loss": 0.4555, + "step": 21806 + }, + { + "epoch": 1.2183981897924407, + "grad_norm": 0.37118205428123474, + "learning_rate": 0.0003924249215598387, + "loss": 0.4161, + "step": 21807 + }, + { + "epoch": 1.2184540603961227, + "grad_norm": 0.4389185905456543, + "learning_rate": 0.00039239690721649484, + "loss": 0.4184, + "step": 21808 + }, + { + "epoch": 1.2185099309998044, + "grad_norm": 0.3687610626220703, + "learning_rate": 0.0003923688928731511, + "loss": 0.3876, + "step": 21809 + }, + { + "epoch": 1.2185658016034864, + "grad_norm": 0.5985124111175537, + "learning_rate": 0.00039234087852980725, + "loss": 0.4794, + "step": 21810 + }, + { + "epoch": 1.218621672207168, + "grad_norm": 0.3734775483608246, + "learning_rate": 0.0003923128641864635, + "loss": 0.3951, + "step": 21811 + }, + { + "epoch": 1.21867754281085, + "grad_norm": 0.5083323121070862, + "learning_rate": 0.00039228484984311966, + "loss": 0.5211, + "step": 21812 + }, + { + "epoch": 1.218733413414532, + "grad_norm": 0.4522364139556885, + "learning_rate": 0.00039225683549977587, + "loss": 0.4679, + "step": 21813 + }, + { + "epoch": 1.2187892840182137, + "grad_norm": 0.34698325395584106, + "learning_rate": 0.0003922288211564321, + "loss": 0.255, + "step": 21814 + }, + { + "epoch": 1.2188451546218957, + "grad_norm": 0.44008392095565796, + "learning_rate": 0.0003922008068130883, + "loss": 0.3537, + "step": 21815 + }, + { + "epoch": 1.2189010252255776, + "grad_norm": 0.4086516201496124, + "learning_rate": 0.00039217279246974454, + "loss": 0.4886, + "step": 21816 + }, + { + "epoch": 1.2189568958292594, + "grad_norm": 0.41411083936691284, + "learning_rate": 0.0003921447781264007, + "loss": 0.3968, + "step": 21817 + }, + { + "epoch": 1.2190127664329413, + "grad_norm": 2.232368230819702, + "learning_rate": 0.00039211676378305695, + "loss": 0.5671, + "step": 21818 + }, + { + "epoch": 1.2190686370366233, + "grad_norm": 0.4682779610157013, + "learning_rate": 0.0003920887494397131, + "loss": 0.3199, + "step": 21819 + }, + { + "epoch": 1.219124507640305, + "grad_norm": 0.5762607455253601, + "learning_rate": 0.00039206073509636937, + "loss": 0.4303, + "step": 21820 + }, + { + "epoch": 1.219180378243987, + "grad_norm": 0.7907471060752869, + "learning_rate": 0.0003920327207530255, + "loss": 0.5571, + "step": 21821 + }, + { + "epoch": 1.2192362488476687, + "grad_norm": 0.41996192932128906, + "learning_rate": 0.0003920047064096818, + "loss": 0.521, + "step": 21822 + }, + { + "epoch": 1.2192921194513506, + "grad_norm": 1.7585126161575317, + "learning_rate": 0.000391976692066338, + "loss": 0.4161, + "step": 21823 + }, + { + "epoch": 1.2193479900550326, + "grad_norm": 0.4663344919681549, + "learning_rate": 0.0003919486777229942, + "loss": 0.4298, + "step": 21824 + }, + { + "epoch": 1.2194038606587143, + "grad_norm": 0.7805770635604858, + "learning_rate": 0.0003919206633796504, + "loss": 0.51, + "step": 21825 + }, + { + "epoch": 1.2194597312623963, + "grad_norm": 0.33804070949554443, + "learning_rate": 0.0003918926490363066, + "loss": 0.4385, + "step": 21826 + }, + { + "epoch": 1.2195156018660782, + "grad_norm": 0.31663820147514343, + "learning_rate": 0.0003918646346929628, + "loss": 0.2983, + "step": 21827 + }, + { + "epoch": 1.21957147246976, + "grad_norm": 0.5477887392044067, + "learning_rate": 0.000391836620349619, + "loss": 0.5514, + "step": 21828 + }, + { + "epoch": 1.219627343073442, + "grad_norm": 0.6690928339958191, + "learning_rate": 0.0003918086060062752, + "loss": 0.4269, + "step": 21829 + }, + { + "epoch": 1.2196832136771238, + "grad_norm": 8.210844993591309, + "learning_rate": 0.0003917805916629315, + "loss": 0.4406, + "step": 21830 + }, + { + "epoch": 1.2197390842808056, + "grad_norm": 1.0440722703933716, + "learning_rate": 0.00039175257731958763, + "loss": 0.5064, + "step": 21831 + }, + { + "epoch": 1.2197949548844875, + "grad_norm": 0.5533505082130432, + "learning_rate": 0.0003917245629762439, + "loss": 0.3837, + "step": 21832 + }, + { + "epoch": 1.2198508254881695, + "grad_norm": 0.6232450008392334, + "learning_rate": 0.00039169654863290004, + "loss": 0.531, + "step": 21833 + }, + { + "epoch": 1.2199066960918512, + "grad_norm": 0.381740003824234, + "learning_rate": 0.0003916685342895563, + "loss": 0.3449, + "step": 21834 + }, + { + "epoch": 1.2199625666955332, + "grad_norm": 0.48397907614707947, + "learning_rate": 0.00039164051994621245, + "loss": 0.3886, + "step": 21835 + }, + { + "epoch": 1.220018437299215, + "grad_norm": 0.37273919582366943, + "learning_rate": 0.00039161250560286866, + "loss": 0.3633, + "step": 21836 + }, + { + "epoch": 1.2200743079028968, + "grad_norm": 0.4283891022205353, + "learning_rate": 0.00039158449125952486, + "loss": 0.4037, + "step": 21837 + }, + { + "epoch": 1.2201301785065788, + "grad_norm": 0.48107197880744934, + "learning_rate": 0.00039155647691618107, + "loss": 0.4001, + "step": 21838 + }, + { + "epoch": 1.2201860491102607, + "grad_norm": 0.38939139246940613, + "learning_rate": 0.00039152846257283733, + "loss": 0.4515, + "step": 21839 + }, + { + "epoch": 1.2202419197139425, + "grad_norm": 0.6342381834983826, + "learning_rate": 0.0003915004482294935, + "loss": 0.3232, + "step": 21840 + }, + { + "epoch": 1.2202977903176244, + "grad_norm": 0.37972262501716614, + "learning_rate": 0.00039147243388614974, + "loss": 0.4393, + "step": 21841 + }, + { + "epoch": 1.2203536609213064, + "grad_norm": 3.580918550491333, + "learning_rate": 0.0003914444195428059, + "loss": 0.4468, + "step": 21842 + }, + { + "epoch": 1.220409531524988, + "grad_norm": 0.5229895114898682, + "learning_rate": 0.00039141640519946215, + "loss": 0.37, + "step": 21843 + }, + { + "epoch": 1.22046540212867, + "grad_norm": 0.4962216913700104, + "learning_rate": 0.0003913883908561183, + "loss": 0.5448, + "step": 21844 + }, + { + "epoch": 1.2205212727323518, + "grad_norm": 0.44370752573013306, + "learning_rate": 0.00039136037651277456, + "loss": 0.4243, + "step": 21845 + }, + { + "epoch": 1.2205771433360337, + "grad_norm": 0.44998666644096375, + "learning_rate": 0.00039133236216943077, + "loss": 0.4211, + "step": 21846 + }, + { + "epoch": 1.2206330139397157, + "grad_norm": 0.3515540659427643, + "learning_rate": 0.000391304347826087, + "loss": 0.38, + "step": 21847 + }, + { + "epoch": 1.2206888845433974, + "grad_norm": 0.8735324740409851, + "learning_rate": 0.0003912763334827432, + "loss": 0.3267, + "step": 21848 + }, + { + "epoch": 1.2207447551470794, + "grad_norm": 0.5807090997695923, + "learning_rate": 0.0003912483191393994, + "loss": 0.4402, + "step": 21849 + }, + { + "epoch": 1.2208006257507613, + "grad_norm": 0.49914848804473877, + "learning_rate": 0.0003912203047960556, + "loss": 0.4, + "step": 21850 + }, + { + "epoch": 1.220856496354443, + "grad_norm": 0.4042775332927704, + "learning_rate": 0.0003911922904527118, + "loss": 0.42, + "step": 21851 + }, + { + "epoch": 1.220912366958125, + "grad_norm": 0.5668222904205322, + "learning_rate": 0.000391164276109368, + "loss": 0.5495, + "step": 21852 + }, + { + "epoch": 1.220968237561807, + "grad_norm": 0.7227033972740173, + "learning_rate": 0.0003911362617660242, + "loss": 0.3906, + "step": 21853 + }, + { + "epoch": 1.2210241081654887, + "grad_norm": 0.5231893062591553, + "learning_rate": 0.0003911082474226804, + "loss": 0.465, + "step": 21854 + }, + { + "epoch": 1.2210799787691706, + "grad_norm": 0.3861871361732483, + "learning_rate": 0.0003910802330793367, + "loss": 0.4336, + "step": 21855 + }, + { + "epoch": 1.2211358493728524, + "grad_norm": 1.2740650177001953, + "learning_rate": 0.00039105221873599283, + "loss": 0.371, + "step": 21856 + }, + { + "epoch": 1.2211917199765343, + "grad_norm": 0.3873489797115326, + "learning_rate": 0.00039102420439264903, + "loss": 0.3884, + "step": 21857 + }, + { + "epoch": 1.2212475905802163, + "grad_norm": 0.39378583431243896, + "learning_rate": 0.00039099619004930524, + "loss": 0.4119, + "step": 21858 + }, + { + "epoch": 1.221303461183898, + "grad_norm": 0.37109944224357605, + "learning_rate": 0.00039096817570596145, + "loss": 0.3019, + "step": 21859 + }, + { + "epoch": 1.22135933178758, + "grad_norm": 0.7130315899848938, + "learning_rate": 0.00039094016136261765, + "loss": 0.489, + "step": 21860 + }, + { + "epoch": 1.221415202391262, + "grad_norm": 0.48330435156822205, + "learning_rate": 0.00039091214701927386, + "loss": 0.4349, + "step": 21861 + }, + { + "epoch": 1.2214710729949436, + "grad_norm": 0.44642481207847595, + "learning_rate": 0.0003908841326759301, + "loss": 0.3891, + "step": 21862 + }, + { + "epoch": 1.2215269435986256, + "grad_norm": 1.535696268081665, + "learning_rate": 0.00039085611833258627, + "loss": 0.4754, + "step": 21863 + }, + { + "epoch": 1.2215828142023075, + "grad_norm": 0.4336716830730438, + "learning_rate": 0.00039082810398924253, + "loss": 0.418, + "step": 21864 + }, + { + "epoch": 1.2216386848059893, + "grad_norm": 0.473528653383255, + "learning_rate": 0.0003908000896458987, + "loss": 0.4847, + "step": 21865 + }, + { + "epoch": 1.2216945554096712, + "grad_norm": 2.9657299518585205, + "learning_rate": 0.00039077207530255494, + "loss": 0.4479, + "step": 21866 + }, + { + "epoch": 1.2217504260133532, + "grad_norm": 0.4881056845188141, + "learning_rate": 0.0003907440609592111, + "loss": 0.4043, + "step": 21867 + }, + { + "epoch": 1.2218062966170349, + "grad_norm": 0.5301072001457214, + "learning_rate": 0.00039071604661586735, + "loss": 0.5902, + "step": 21868 + }, + { + "epoch": 1.2218621672207168, + "grad_norm": 0.3536742329597473, + "learning_rate": 0.0003906880322725235, + "loss": 0.3868, + "step": 21869 + }, + { + "epoch": 1.2219180378243988, + "grad_norm": 0.37932661175727844, + "learning_rate": 0.00039066001792917976, + "loss": 0.3491, + "step": 21870 + }, + { + "epoch": 1.2219739084280805, + "grad_norm": 1.5939180850982666, + "learning_rate": 0.00039063200358583597, + "loss": 0.5238, + "step": 21871 + }, + { + "epoch": 1.2220297790317625, + "grad_norm": 0.40941447019577026, + "learning_rate": 0.0003906039892424922, + "loss": 0.2761, + "step": 21872 + }, + { + "epoch": 1.2220856496354444, + "grad_norm": 0.4078024625778198, + "learning_rate": 0.0003905759748991484, + "loss": 0.402, + "step": 21873 + }, + { + "epoch": 1.2221415202391261, + "grad_norm": 0.5004361271858215, + "learning_rate": 0.0003905479605558046, + "loss": 0.4149, + "step": 21874 + }, + { + "epoch": 1.222197390842808, + "grad_norm": 0.3842844069004059, + "learning_rate": 0.0003905199462124608, + "loss": 0.3844, + "step": 21875 + }, + { + "epoch": 1.22225326144649, + "grad_norm": 0.8864195942878723, + "learning_rate": 0.000390491931869117, + "loss": 0.6921, + "step": 21876 + }, + { + "epoch": 1.2223091320501718, + "grad_norm": 0.5959190726280212, + "learning_rate": 0.0003904639175257732, + "loss": 0.4075, + "step": 21877 + }, + { + "epoch": 1.2223650026538537, + "grad_norm": 0.3810768127441406, + "learning_rate": 0.0003904359031824294, + "loss": 0.3323, + "step": 21878 + }, + { + "epoch": 1.2224208732575355, + "grad_norm": 0.4467350244522095, + "learning_rate": 0.0003904078888390856, + "loss": 0.489, + "step": 21879 + }, + { + "epoch": 1.2224767438612174, + "grad_norm": 0.4286898374557495, + "learning_rate": 0.0003903798744957418, + "loss": 0.4357, + "step": 21880 + }, + { + "epoch": 1.2225326144648994, + "grad_norm": 0.3426765501499176, + "learning_rate": 0.00039035186015239803, + "loss": 0.3575, + "step": 21881 + }, + { + "epoch": 1.222588485068581, + "grad_norm": 0.6049132347106934, + "learning_rate": 0.00039032384580905423, + "loss": 0.4361, + "step": 21882 + }, + { + "epoch": 1.222644355672263, + "grad_norm": 0.3592977523803711, + "learning_rate": 0.00039029583146571044, + "loss": 0.4179, + "step": 21883 + }, + { + "epoch": 1.222700226275945, + "grad_norm": 1.136107325553894, + "learning_rate": 0.00039026781712236665, + "loss": 0.426, + "step": 21884 + }, + { + "epoch": 1.2227560968796267, + "grad_norm": 0.36687418818473816, + "learning_rate": 0.0003902398027790229, + "loss": 0.4157, + "step": 21885 + }, + { + "epoch": 1.2228119674833087, + "grad_norm": 0.7252906560897827, + "learning_rate": 0.00039021178843567906, + "loss": 0.5435, + "step": 21886 + }, + { + "epoch": 1.2228678380869906, + "grad_norm": 0.4148891270160675, + "learning_rate": 0.0003901837740923353, + "loss": 0.4299, + "step": 21887 + }, + { + "epoch": 1.2229237086906724, + "grad_norm": 0.3942616283893585, + "learning_rate": 0.00039015575974899147, + "loss": 0.4372, + "step": 21888 + }, + { + "epoch": 1.2229795792943543, + "grad_norm": 0.6880713105201721, + "learning_rate": 0.00039012774540564773, + "loss": 0.4338, + "step": 21889 + }, + { + "epoch": 1.223035449898036, + "grad_norm": 0.5936171412467957, + "learning_rate": 0.0003900997310623039, + "loss": 0.4491, + "step": 21890 + }, + { + "epoch": 1.223091320501718, + "grad_norm": 1.4683047533035278, + "learning_rate": 0.00039007171671896014, + "loss": 0.5006, + "step": 21891 + }, + { + "epoch": 1.2231471911054, + "grad_norm": 0.40185579657554626, + "learning_rate": 0.0003900437023756163, + "loss": 0.3231, + "step": 21892 + }, + { + "epoch": 1.2232030617090817, + "grad_norm": 0.36461010575294495, + "learning_rate": 0.00039001568803227255, + "loss": 0.3803, + "step": 21893 + }, + { + "epoch": 1.2232589323127636, + "grad_norm": 0.755522608757019, + "learning_rate": 0.00038998767368892876, + "loss": 0.4526, + "step": 21894 + }, + { + "epoch": 1.2233148029164456, + "grad_norm": 0.5007203817367554, + "learning_rate": 0.00038995965934558496, + "loss": 0.4061, + "step": 21895 + }, + { + "epoch": 1.2233706735201273, + "grad_norm": 0.4221416413784027, + "learning_rate": 0.00038993164500224117, + "loss": 0.3589, + "step": 21896 + }, + { + "epoch": 1.2234265441238092, + "grad_norm": 0.4226832091808319, + "learning_rate": 0.0003899036306588974, + "loss": 0.465, + "step": 21897 + }, + { + "epoch": 1.2234824147274912, + "grad_norm": 0.43770140409469604, + "learning_rate": 0.0003898756163155536, + "loss": 0.4023, + "step": 21898 + }, + { + "epoch": 1.223538285331173, + "grad_norm": 0.4443731904029846, + "learning_rate": 0.00038984760197220973, + "loss": 0.3553, + "step": 21899 + }, + { + "epoch": 1.2235941559348549, + "grad_norm": 0.5148442387580872, + "learning_rate": 0.000389819587628866, + "loss": 0.4884, + "step": 21900 + }, + { + "epoch": 1.2236500265385368, + "grad_norm": 0.7836220264434814, + "learning_rate": 0.0003897915732855222, + "loss": 0.4704, + "step": 21901 + }, + { + "epoch": 1.2237058971422186, + "grad_norm": 0.45599767565727234, + "learning_rate": 0.0003897635589421784, + "loss": 0.3033, + "step": 21902 + }, + { + "epoch": 1.2237617677459005, + "grad_norm": 0.5581499934196472, + "learning_rate": 0.0003897355445988346, + "loss": 0.6616, + "step": 21903 + }, + { + "epoch": 1.2238176383495825, + "grad_norm": 0.38346531987190247, + "learning_rate": 0.0003897075302554908, + "loss": 0.4145, + "step": 21904 + }, + { + "epoch": 1.2238735089532642, + "grad_norm": 0.38181692361831665, + "learning_rate": 0.000389679515912147, + "loss": 0.3835, + "step": 21905 + }, + { + "epoch": 1.2239293795569461, + "grad_norm": 0.35479313135147095, + "learning_rate": 0.00038965150156880323, + "loss": 0.4377, + "step": 21906 + }, + { + "epoch": 1.223985250160628, + "grad_norm": 0.7376660108566284, + "learning_rate": 0.00038962348722545943, + "loss": 0.3722, + "step": 21907 + }, + { + "epoch": 1.2240411207643098, + "grad_norm": 0.5771657824516296, + "learning_rate": 0.00038959547288211564, + "loss": 0.4141, + "step": 21908 + }, + { + "epoch": 1.2240969913679918, + "grad_norm": 4.460803985595703, + "learning_rate": 0.00038956745853877185, + "loss": 0.5123, + "step": 21909 + }, + { + "epoch": 1.2241528619716737, + "grad_norm": 0.544114887714386, + "learning_rate": 0.0003895394441954281, + "loss": 0.363, + "step": 21910 + }, + { + "epoch": 1.2242087325753555, + "grad_norm": 0.5640119314193726, + "learning_rate": 0.00038951142985208426, + "loss": 0.5356, + "step": 21911 + }, + { + "epoch": 1.2242646031790374, + "grad_norm": 0.3696463704109192, + "learning_rate": 0.0003894834155087405, + "loss": 0.384, + "step": 21912 + }, + { + "epoch": 1.2243204737827191, + "grad_norm": 0.46427276730537415, + "learning_rate": 0.00038945540116539667, + "loss": 0.4678, + "step": 21913 + }, + { + "epoch": 1.224376344386401, + "grad_norm": 0.7411181926727295, + "learning_rate": 0.00038942738682205293, + "loss": 0.393, + "step": 21914 + }, + { + "epoch": 1.224432214990083, + "grad_norm": 0.5524126291275024, + "learning_rate": 0.0003893993724787091, + "loss": 0.3758, + "step": 21915 + }, + { + "epoch": 1.2244880855937648, + "grad_norm": 0.7233712673187256, + "learning_rate": 0.00038937135813536534, + "loss": 0.3905, + "step": 21916 + }, + { + "epoch": 1.2245439561974467, + "grad_norm": 0.8385598063468933, + "learning_rate": 0.00038934334379202155, + "loss": 0.4798, + "step": 21917 + }, + { + "epoch": 1.2245998268011287, + "grad_norm": 0.40635430812835693, + "learning_rate": 0.00038931532944867775, + "loss": 0.3909, + "step": 21918 + }, + { + "epoch": 1.2246556974048104, + "grad_norm": 0.37370091676712036, + "learning_rate": 0.00038928731510533396, + "loss": 0.4893, + "step": 21919 + }, + { + "epoch": 1.2247115680084923, + "grad_norm": 1.1287177801132202, + "learning_rate": 0.0003892593007619901, + "loss": 0.3384, + "step": 21920 + }, + { + "epoch": 1.2247674386121743, + "grad_norm": 1.1984639167785645, + "learning_rate": 0.00038923128641864637, + "loss": 0.5144, + "step": 21921 + }, + { + "epoch": 1.224823309215856, + "grad_norm": 0.576853334903717, + "learning_rate": 0.0003892032720753025, + "loss": 0.3652, + "step": 21922 + }, + { + "epoch": 1.224879179819538, + "grad_norm": 0.4328376054763794, + "learning_rate": 0.0003891752577319588, + "loss": 0.4192, + "step": 21923 + }, + { + "epoch": 1.2249350504232197, + "grad_norm": 0.5542796850204468, + "learning_rate": 0.00038914724338861493, + "loss": 0.6967, + "step": 21924 + }, + { + "epoch": 1.2249909210269017, + "grad_norm": 0.422422856092453, + "learning_rate": 0.0003891192290452712, + "loss": 0.3986, + "step": 21925 + }, + { + "epoch": 1.2250467916305836, + "grad_norm": 0.5440177917480469, + "learning_rate": 0.0003890912147019274, + "loss": 0.5272, + "step": 21926 + }, + { + "epoch": 1.2251026622342653, + "grad_norm": 0.4508422315120697, + "learning_rate": 0.0003890632003585836, + "loss": 0.3855, + "step": 21927 + }, + { + "epoch": 1.2251585328379473, + "grad_norm": 0.8525639772415161, + "learning_rate": 0.0003890351860152398, + "loss": 0.4498, + "step": 21928 + }, + { + "epoch": 1.2252144034416292, + "grad_norm": 0.4113635718822479, + "learning_rate": 0.000389007171671896, + "loss": 0.4755, + "step": 21929 + }, + { + "epoch": 1.225270274045311, + "grad_norm": 0.9447039365768433, + "learning_rate": 0.0003889791573285522, + "loss": 0.3655, + "step": 21930 + }, + { + "epoch": 1.225326144648993, + "grad_norm": 0.33538153767585754, + "learning_rate": 0.00038895114298520843, + "loss": 0.3478, + "step": 21931 + }, + { + "epoch": 1.2253820152526749, + "grad_norm": 0.6714051961898804, + "learning_rate": 0.00038892312864186463, + "loss": 0.5153, + "step": 21932 + }, + { + "epoch": 1.2254378858563566, + "grad_norm": 0.30235806107521057, + "learning_rate": 0.0003888951142985209, + "loss": 0.3883, + "step": 21933 + }, + { + "epoch": 1.2254937564600386, + "grad_norm": 0.662337601184845, + "learning_rate": 0.00038886709995517705, + "loss": 0.3165, + "step": 21934 + }, + { + "epoch": 1.2255496270637205, + "grad_norm": 0.42141208052635193, + "learning_rate": 0.0003888390856118333, + "loss": 0.3702, + "step": 21935 + }, + { + "epoch": 1.2256054976674022, + "grad_norm": 0.37790966033935547, + "learning_rate": 0.00038881107126848946, + "loss": 0.433, + "step": 21936 + }, + { + "epoch": 1.2256613682710842, + "grad_norm": 0.5203299522399902, + "learning_rate": 0.0003887830569251457, + "loss": 0.405, + "step": 21937 + }, + { + "epoch": 1.2257172388747661, + "grad_norm": 0.4604302942752838, + "learning_rate": 0.00038875504258180187, + "loss": 0.3897, + "step": 21938 + }, + { + "epoch": 1.2257731094784479, + "grad_norm": 0.7884395718574524, + "learning_rate": 0.00038872702823845813, + "loss": 0.3933, + "step": 21939 + }, + { + "epoch": 1.2258289800821298, + "grad_norm": 0.4376145899295807, + "learning_rate": 0.0003886990138951143, + "loss": 0.4705, + "step": 21940 + }, + { + "epoch": 1.2258848506858118, + "grad_norm": 0.5935254096984863, + "learning_rate": 0.0003886709995517705, + "loss": 0.5142, + "step": 21941 + }, + { + "epoch": 1.2259407212894935, + "grad_norm": 0.3698137104511261, + "learning_rate": 0.00038864298520842675, + "loss": 0.4195, + "step": 21942 + }, + { + "epoch": 1.2259965918931754, + "grad_norm": 0.5349197387695312, + "learning_rate": 0.0003886149708650829, + "loss": 0.5103, + "step": 21943 + }, + { + "epoch": 1.2260524624968574, + "grad_norm": 0.7464731931686401, + "learning_rate": 0.00038858695652173916, + "loss": 0.4879, + "step": 21944 + }, + { + "epoch": 1.2261083331005391, + "grad_norm": 0.3783627450466156, + "learning_rate": 0.0003885589421783953, + "loss": 0.5562, + "step": 21945 + }, + { + "epoch": 1.226164203704221, + "grad_norm": 2.8459248542785645, + "learning_rate": 0.00038853092783505157, + "loss": 0.3769, + "step": 21946 + }, + { + "epoch": 1.2262200743079028, + "grad_norm": 8.124322891235352, + "learning_rate": 0.0003885029134917077, + "loss": 0.4847, + "step": 21947 + }, + { + "epoch": 1.2262759449115848, + "grad_norm": 0.45102906227111816, + "learning_rate": 0.000388474899148364, + "loss": 0.4598, + "step": 21948 + }, + { + "epoch": 1.2263318155152667, + "grad_norm": 0.41896045207977295, + "learning_rate": 0.0003884468848050202, + "loss": 0.3734, + "step": 21949 + }, + { + "epoch": 1.2263876861189484, + "grad_norm": 0.37562254071235657, + "learning_rate": 0.0003884188704616764, + "loss": 0.4067, + "step": 21950 + }, + { + "epoch": 1.2264435567226304, + "grad_norm": 1.8190633058547974, + "learning_rate": 0.0003883908561183326, + "loss": 0.5818, + "step": 21951 + }, + { + "epoch": 1.2264994273263123, + "grad_norm": 0.38195744156837463, + "learning_rate": 0.0003883628417749888, + "loss": 0.3805, + "step": 21952 + }, + { + "epoch": 1.226555297929994, + "grad_norm": 0.3362424671649933, + "learning_rate": 0.000388334827431645, + "loss": 0.3663, + "step": 21953 + }, + { + "epoch": 1.226611168533676, + "grad_norm": 0.927706778049469, + "learning_rate": 0.0003883068130883012, + "loss": 0.4766, + "step": 21954 + }, + { + "epoch": 1.226667039137358, + "grad_norm": 0.4888676106929779, + "learning_rate": 0.0003882787987449574, + "loss": 0.4455, + "step": 21955 + }, + { + "epoch": 1.2267229097410397, + "grad_norm": 0.391945481300354, + "learning_rate": 0.0003882507844016137, + "loss": 0.3944, + "step": 21956 + }, + { + "epoch": 1.2267787803447217, + "grad_norm": 1.153382658958435, + "learning_rate": 0.00038822277005826983, + "loss": 0.5933, + "step": 21957 + }, + { + "epoch": 1.2268346509484034, + "grad_norm": 0.8311321139335632, + "learning_rate": 0.0003881947557149261, + "loss": 0.5328, + "step": 21958 + }, + { + "epoch": 1.2268905215520853, + "grad_norm": 0.4703986942768097, + "learning_rate": 0.00038816674137158224, + "loss": 0.3864, + "step": 21959 + }, + { + "epoch": 1.2269463921557673, + "grad_norm": 0.5322163701057434, + "learning_rate": 0.0003881387270282385, + "loss": 0.4625, + "step": 21960 + }, + { + "epoch": 1.227002262759449, + "grad_norm": 1.4926271438598633, + "learning_rate": 0.00038811071268489466, + "loss": 0.5441, + "step": 21961 + }, + { + "epoch": 1.227058133363131, + "grad_norm": 0.3756577670574188, + "learning_rate": 0.00038808269834155086, + "loss": 0.4169, + "step": 21962 + }, + { + "epoch": 1.227114003966813, + "grad_norm": 0.5565135478973389, + "learning_rate": 0.00038805468399820707, + "loss": 0.4135, + "step": 21963 + }, + { + "epoch": 1.2271698745704946, + "grad_norm": 0.5935763120651245, + "learning_rate": 0.0003880266696548633, + "loss": 0.4849, + "step": 21964 + }, + { + "epoch": 1.2272257451741766, + "grad_norm": 0.4999317526817322, + "learning_rate": 0.00038799865531151953, + "loss": 0.5902, + "step": 21965 + }, + { + "epoch": 1.2272816157778585, + "grad_norm": 0.5753591656684875, + "learning_rate": 0.0003879706409681757, + "loss": 0.4367, + "step": 21966 + }, + { + "epoch": 1.2273374863815403, + "grad_norm": 0.4415311813354492, + "learning_rate": 0.00038794262662483195, + "loss": 0.3695, + "step": 21967 + }, + { + "epoch": 1.2273933569852222, + "grad_norm": 0.4534989595413208, + "learning_rate": 0.0003879146122814881, + "loss": 0.3761, + "step": 21968 + }, + { + "epoch": 1.2274492275889042, + "grad_norm": 0.3017331063747406, + "learning_rate": 0.00038788659793814436, + "loss": 0.4346, + "step": 21969 + }, + { + "epoch": 1.227505098192586, + "grad_norm": 0.6508709788322449, + "learning_rate": 0.0003878585835948005, + "loss": 0.503, + "step": 21970 + }, + { + "epoch": 1.2275609687962679, + "grad_norm": 0.443988561630249, + "learning_rate": 0.00038783056925145677, + "loss": 0.4825, + "step": 21971 + }, + { + "epoch": 1.2276168393999498, + "grad_norm": 0.47649016976356506, + "learning_rate": 0.000387802554908113, + "loss": 0.5084, + "step": 21972 + }, + { + "epoch": 1.2276727100036315, + "grad_norm": 0.44840413331985474, + "learning_rate": 0.0003877745405647692, + "loss": 0.5107, + "step": 21973 + }, + { + "epoch": 1.2277285806073135, + "grad_norm": 0.48610174655914307, + "learning_rate": 0.0003877465262214254, + "loss": 0.353, + "step": 21974 + }, + { + "epoch": 1.2277844512109954, + "grad_norm": 0.391142874956131, + "learning_rate": 0.0003877185118780816, + "loss": 0.4487, + "step": 21975 + }, + { + "epoch": 1.2278403218146772, + "grad_norm": 0.6183385252952576, + "learning_rate": 0.0003876904975347378, + "loss": 0.4665, + "step": 21976 + }, + { + "epoch": 1.2278961924183591, + "grad_norm": 0.69745934009552, + "learning_rate": 0.000387662483191394, + "loss": 0.3771, + "step": 21977 + }, + { + "epoch": 1.227952063022041, + "grad_norm": 4.799753665924072, + "learning_rate": 0.0003876344688480502, + "loss": 0.5418, + "step": 21978 + }, + { + "epoch": 1.2280079336257228, + "grad_norm": 0.45456886291503906, + "learning_rate": 0.0003876064545047064, + "loss": 0.5392, + "step": 21979 + }, + { + "epoch": 1.2280638042294048, + "grad_norm": 0.48963654041290283, + "learning_rate": 0.0003875784401613626, + "loss": 0.5206, + "step": 21980 + }, + { + "epoch": 1.2281196748330865, + "grad_norm": 0.48504847288131714, + "learning_rate": 0.0003875504258180189, + "loss": 0.329, + "step": 21981 + }, + { + "epoch": 1.2281755454367684, + "grad_norm": 0.42875707149505615, + "learning_rate": 0.00038752241147467503, + "loss": 0.4065, + "step": 21982 + }, + { + "epoch": 1.2282314160404504, + "grad_norm": 0.41376492381095886, + "learning_rate": 0.00038749439713133124, + "loss": 0.5368, + "step": 21983 + }, + { + "epoch": 1.2282872866441321, + "grad_norm": 0.720114529132843, + "learning_rate": 0.00038746638278798744, + "loss": 0.3436, + "step": 21984 + }, + { + "epoch": 1.228343157247814, + "grad_norm": 0.7360184788703918, + "learning_rate": 0.00038743836844464365, + "loss": 0.5507, + "step": 21985 + }, + { + "epoch": 1.228399027851496, + "grad_norm": 0.462643027305603, + "learning_rate": 0.00038741035410129986, + "loss": 0.5197, + "step": 21986 + }, + { + "epoch": 1.2284548984551777, + "grad_norm": 0.3806595206260681, + "learning_rate": 0.00038738233975795606, + "loss": 0.3347, + "step": 21987 + }, + { + "epoch": 1.2285107690588597, + "grad_norm": 0.6937063336372375, + "learning_rate": 0.0003873543254146123, + "loss": 0.4819, + "step": 21988 + }, + { + "epoch": 1.2285666396625416, + "grad_norm": 0.7092167139053345, + "learning_rate": 0.0003873263110712685, + "loss": 0.5591, + "step": 21989 + }, + { + "epoch": 1.2286225102662234, + "grad_norm": 0.49718597531318665, + "learning_rate": 0.00038729829672792473, + "loss": 0.418, + "step": 21990 + }, + { + "epoch": 1.2286783808699053, + "grad_norm": 0.38610920310020447, + "learning_rate": 0.0003872702823845809, + "loss": 0.4328, + "step": 21991 + }, + { + "epoch": 1.228734251473587, + "grad_norm": 0.511288583278656, + "learning_rate": 0.00038724226804123714, + "loss": 0.3855, + "step": 21992 + }, + { + "epoch": 1.228790122077269, + "grad_norm": 21.665550231933594, + "learning_rate": 0.0003872142536978933, + "loss": 0.433, + "step": 21993 + }, + { + "epoch": 1.228845992680951, + "grad_norm": 0.3687036633491516, + "learning_rate": 0.00038718623935454956, + "loss": 0.3465, + "step": 21994 + }, + { + "epoch": 1.2289018632846327, + "grad_norm": 0.3611290454864502, + "learning_rate": 0.0003871582250112057, + "loss": 0.4856, + "step": 21995 + }, + { + "epoch": 1.2289577338883146, + "grad_norm": 2.734959363937378, + "learning_rate": 0.00038713021066786197, + "loss": 0.4785, + "step": 21996 + }, + { + "epoch": 1.2290136044919966, + "grad_norm": 0.5323562026023865, + "learning_rate": 0.0003871021963245182, + "loss": 0.5199, + "step": 21997 + }, + { + "epoch": 1.2290694750956783, + "grad_norm": 0.389276385307312, + "learning_rate": 0.0003870741819811744, + "loss": 0.4352, + "step": 21998 + }, + { + "epoch": 1.2291253456993603, + "grad_norm": 0.7936498522758484, + "learning_rate": 0.0003870461676378306, + "loss": 0.5109, + "step": 21999 + }, + { + "epoch": 1.2291812163030422, + "grad_norm": 0.3487698435783386, + "learning_rate": 0.0003870181532944868, + "loss": 0.4016, + "step": 22000 + }, + { + "epoch": 1.2291812163030422, + "eval_cer": 0.0860964725650006, + "eval_loss": 0.32487329840660095, + "eval_runtime": 56.4561, + "eval_samples_per_second": 80.381, + "eval_steps_per_second": 5.03, + "eval_wer": 0.33976704561325105, + "step": 22000 + }, + { + "epoch": 1.229237086906724, + "grad_norm": 0.7832374572753906, + "learning_rate": 0.000386990138951143, + "loss": 0.5513, + "step": 22001 + }, + { + "epoch": 1.229292957510406, + "grad_norm": 1.7278186082839966, + "learning_rate": 0.0003869621246077992, + "loss": 0.369, + "step": 22002 + }, + { + "epoch": 1.2293488281140879, + "grad_norm": 0.7937616109848022, + "learning_rate": 0.0003869341102644554, + "loss": 0.4919, + "step": 22003 + }, + { + "epoch": 1.2294046987177696, + "grad_norm": 0.38919904828071594, + "learning_rate": 0.00038690609592111167, + "loss": 0.4229, + "step": 22004 + }, + { + "epoch": 1.2294605693214515, + "grad_norm": 0.3334086835384369, + "learning_rate": 0.0003868780815777678, + "loss": 0.3849, + "step": 22005 + }, + { + "epoch": 1.2295164399251335, + "grad_norm": 0.7115708589553833, + "learning_rate": 0.000386850067234424, + "loss": 0.3891, + "step": 22006 + }, + { + "epoch": 1.2295723105288152, + "grad_norm": 0.3865632712841034, + "learning_rate": 0.00038682205289108023, + "loss": 0.4452, + "step": 22007 + }, + { + "epoch": 1.2296281811324972, + "grad_norm": 0.3922528028488159, + "learning_rate": 0.00038679403854773644, + "loss": 0.3568, + "step": 22008 + }, + { + "epoch": 1.2296840517361791, + "grad_norm": 0.7845605611801147, + "learning_rate": 0.00038676602420439264, + "loss": 0.547, + "step": 22009 + }, + { + "epoch": 1.2297399223398608, + "grad_norm": 0.3541022539138794, + "learning_rate": 0.00038673800986104885, + "loss": 0.4285, + "step": 22010 + }, + { + "epoch": 1.2297957929435428, + "grad_norm": 0.49020111560821533, + "learning_rate": 0.00038670999551770506, + "loss": 0.4589, + "step": 22011 + }, + { + "epoch": 1.2298516635472247, + "grad_norm": 0.46779024600982666, + "learning_rate": 0.00038668198117436126, + "loss": 0.3691, + "step": 22012 + }, + { + "epoch": 1.2299075341509065, + "grad_norm": 0.49968963861465454, + "learning_rate": 0.0003866539668310175, + "loss": 0.3438, + "step": 22013 + }, + { + "epoch": 1.2299634047545884, + "grad_norm": 0.4281384348869324, + "learning_rate": 0.00038662595248767367, + "loss": 0.3571, + "step": 22014 + }, + { + "epoch": 1.2300192753582702, + "grad_norm": 0.7402196526527405, + "learning_rate": 0.00038659793814432993, + "loss": 0.4343, + "step": 22015 + }, + { + "epoch": 1.230075145961952, + "grad_norm": 0.6166602969169617, + "learning_rate": 0.0003865699238009861, + "loss": 0.6279, + "step": 22016 + }, + { + "epoch": 1.230131016565634, + "grad_norm": 0.6243363618850708, + "learning_rate": 0.00038654190945764234, + "loss": 0.442, + "step": 22017 + }, + { + "epoch": 1.2301868871693158, + "grad_norm": 0.28398558497428894, + "learning_rate": 0.0003865138951142985, + "loss": 0.3535, + "step": 22018 + }, + { + "epoch": 1.2302427577729977, + "grad_norm": 0.5465614199638367, + "learning_rate": 0.00038648588077095476, + "loss": 0.4901, + "step": 22019 + }, + { + "epoch": 1.2302986283766797, + "grad_norm": 0.42301544547080994, + "learning_rate": 0.00038645786642761096, + "loss": 0.389, + "step": 22020 + }, + { + "epoch": 1.2303544989803614, + "grad_norm": 0.4260500967502594, + "learning_rate": 0.00038642985208426717, + "loss": 0.3903, + "step": 22021 + }, + { + "epoch": 1.2304103695840434, + "grad_norm": 1.21914541721344, + "learning_rate": 0.0003864018377409234, + "loss": 0.3688, + "step": 22022 + }, + { + "epoch": 1.2304662401877253, + "grad_norm": 0.35371720790863037, + "learning_rate": 0.0003863738233975796, + "loss": 0.4448, + "step": 22023 + }, + { + "epoch": 1.230522110791407, + "grad_norm": 0.514683187007904, + "learning_rate": 0.0003863458090542358, + "loss": 0.4605, + "step": 22024 + }, + { + "epoch": 1.230577981395089, + "grad_norm": 0.39141368865966797, + "learning_rate": 0.000386317794710892, + "loss": 0.4151, + "step": 22025 + }, + { + "epoch": 1.2306338519987707, + "grad_norm": 0.6120151877403259, + "learning_rate": 0.0003862897803675482, + "loss": 0.4315, + "step": 22026 + }, + { + "epoch": 1.2306897226024527, + "grad_norm": 0.6838048696517944, + "learning_rate": 0.00038626176602420435, + "loss": 0.4946, + "step": 22027 + }, + { + "epoch": 1.2307455932061346, + "grad_norm": 0.6194482445716858, + "learning_rate": 0.0003862337516808606, + "loss": 0.3892, + "step": 22028 + }, + { + "epoch": 1.2308014638098164, + "grad_norm": 0.838614821434021, + "learning_rate": 0.0003862057373375168, + "loss": 0.4635, + "step": 22029 + }, + { + "epoch": 1.2308573344134983, + "grad_norm": 0.3676556348800659, + "learning_rate": 0.000386177722994173, + "loss": 0.4059, + "step": 22030 + }, + { + "epoch": 1.2309132050171803, + "grad_norm": 0.3776608109474182, + "learning_rate": 0.0003861497086508292, + "loss": 0.4184, + "step": 22031 + }, + { + "epoch": 1.230969075620862, + "grad_norm": 0.3974507451057434, + "learning_rate": 0.00038612169430748543, + "loss": 0.4038, + "step": 22032 + }, + { + "epoch": 1.231024946224544, + "grad_norm": 0.3687041699886322, + "learning_rate": 0.00038609367996414164, + "loss": 0.4523, + "step": 22033 + }, + { + "epoch": 1.231080816828226, + "grad_norm": 0.583185076713562, + "learning_rate": 0.00038606566562079784, + "loss": 0.4451, + "step": 22034 + }, + { + "epoch": 1.2311366874319076, + "grad_norm": 0.4797806143760681, + "learning_rate": 0.00038603765127745405, + "loss": 0.4741, + "step": 22035 + }, + { + "epoch": 1.2311925580355896, + "grad_norm": 0.36524349451065063, + "learning_rate": 0.0003860096369341103, + "loss": 0.4031, + "step": 22036 + }, + { + "epoch": 1.2312484286392715, + "grad_norm": 0.3986961543560028, + "learning_rate": 0.00038598162259076646, + "loss": 0.3452, + "step": 22037 + }, + { + "epoch": 1.2313042992429533, + "grad_norm": 0.49360889196395874, + "learning_rate": 0.0003859536082474227, + "loss": 0.4854, + "step": 22038 + }, + { + "epoch": 1.2313601698466352, + "grad_norm": 0.9238951206207275, + "learning_rate": 0.00038592559390407887, + "loss": 0.4854, + "step": 22039 + }, + { + "epoch": 1.2314160404503172, + "grad_norm": 1.1380177736282349, + "learning_rate": 0.00038589757956073513, + "loss": 0.4625, + "step": 22040 + }, + { + "epoch": 1.231471911053999, + "grad_norm": 1.1678961515426636, + "learning_rate": 0.0003858695652173913, + "loss": 0.3319, + "step": 22041 + }, + { + "epoch": 1.2315277816576808, + "grad_norm": 0.49395516514778137, + "learning_rate": 0.00038584155087404754, + "loss": 0.4372, + "step": 22042 + }, + { + "epoch": 1.2315836522613628, + "grad_norm": 0.37726572155952454, + "learning_rate": 0.00038581353653070375, + "loss": 0.4598, + "step": 22043 + }, + { + "epoch": 1.2316395228650445, + "grad_norm": 0.36770644783973694, + "learning_rate": 0.00038578552218735996, + "loss": 0.3733, + "step": 22044 + }, + { + "epoch": 1.2316953934687265, + "grad_norm": 0.4988788366317749, + "learning_rate": 0.00038575750784401616, + "loss": 0.4402, + "step": 22045 + }, + { + "epoch": 1.2317512640724084, + "grad_norm": 0.952750027179718, + "learning_rate": 0.00038572949350067237, + "loss": 0.4777, + "step": 22046 + }, + { + "epoch": 1.2318071346760902, + "grad_norm": 0.7564629316329956, + "learning_rate": 0.0003857014791573286, + "loss": 0.6312, + "step": 22047 + }, + { + "epoch": 1.231863005279772, + "grad_norm": 0.3832717537879944, + "learning_rate": 0.0003856734648139847, + "loss": 0.4799, + "step": 22048 + }, + { + "epoch": 1.2319188758834538, + "grad_norm": 0.5122115015983582, + "learning_rate": 0.000385645450470641, + "loss": 0.4163, + "step": 22049 + }, + { + "epoch": 1.2319747464871358, + "grad_norm": 1.2702395915985107, + "learning_rate": 0.00038561743612729714, + "loss": 0.5285, + "step": 22050 + }, + { + "epoch": 1.2320306170908177, + "grad_norm": 0.44712427258491516, + "learning_rate": 0.0003855894217839534, + "loss": 0.568, + "step": 22051 + }, + { + "epoch": 1.2320864876944995, + "grad_norm": 0.7351887822151184, + "learning_rate": 0.0003855614074406096, + "loss": 0.4111, + "step": 22052 + }, + { + "epoch": 1.2321423582981814, + "grad_norm": 0.47593939304351807, + "learning_rate": 0.0003855333930972658, + "loss": 0.4946, + "step": 22053 + }, + { + "epoch": 1.2321982289018634, + "grad_norm": 0.5342929363250732, + "learning_rate": 0.000385505378753922, + "loss": 0.3186, + "step": 22054 + }, + { + "epoch": 1.232254099505545, + "grad_norm": 0.49450555443763733, + "learning_rate": 0.0003854773644105782, + "loss": 0.3796, + "step": 22055 + }, + { + "epoch": 1.232309970109227, + "grad_norm": 0.40453997254371643, + "learning_rate": 0.0003854493500672344, + "loss": 0.3482, + "step": 22056 + }, + { + "epoch": 1.232365840712909, + "grad_norm": 0.4038919508457184, + "learning_rate": 0.00038542133572389063, + "loss": 0.415, + "step": 22057 + }, + { + "epoch": 1.2324217113165907, + "grad_norm": 0.9050509333610535, + "learning_rate": 0.00038539332138054684, + "loss": 0.4197, + "step": 22058 + }, + { + "epoch": 1.2324775819202727, + "grad_norm": 0.4734930694103241, + "learning_rate": 0.0003853653070372031, + "loss": 0.4878, + "step": 22059 + }, + { + "epoch": 1.2325334525239544, + "grad_norm": 0.36822670698165894, + "learning_rate": 0.00038533729269385925, + "loss": 0.4077, + "step": 22060 + }, + { + "epoch": 1.2325893231276364, + "grad_norm": 0.4197281002998352, + "learning_rate": 0.0003853092783505155, + "loss": 0.4307, + "step": 22061 + }, + { + "epoch": 1.2326451937313183, + "grad_norm": 0.4032486081123352, + "learning_rate": 0.00038528126400717166, + "loss": 0.4402, + "step": 22062 + }, + { + "epoch": 1.232701064335, + "grad_norm": 0.4172937273979187, + "learning_rate": 0.0003852532496638279, + "loss": 0.4217, + "step": 22063 + }, + { + "epoch": 1.232756934938682, + "grad_norm": 0.39025938510894775, + "learning_rate": 0.00038522523532048407, + "loss": 0.414, + "step": 22064 + }, + { + "epoch": 1.232812805542364, + "grad_norm": 0.7541998624801636, + "learning_rate": 0.00038519722097714033, + "loss": 0.4023, + "step": 22065 + }, + { + "epoch": 1.2328686761460457, + "grad_norm": 0.44969066977500916, + "learning_rate": 0.0003851692066337965, + "loss": 0.4192, + "step": 22066 + }, + { + "epoch": 1.2329245467497276, + "grad_norm": 1.1896955966949463, + "learning_rate": 0.00038514119229045274, + "loss": 0.4963, + "step": 22067 + }, + { + "epoch": 1.2329804173534096, + "grad_norm": 0.3711759150028229, + "learning_rate": 0.00038511317794710895, + "loss": 0.3859, + "step": 22068 + }, + { + "epoch": 1.2330362879570913, + "grad_norm": 2.0626697540283203, + "learning_rate": 0.0003850851636037651, + "loss": 0.4708, + "step": 22069 + }, + { + "epoch": 1.2330921585607733, + "grad_norm": 0.3469131290912628, + "learning_rate": 0.00038505714926042136, + "loss": 0.4378, + "step": 22070 + }, + { + "epoch": 1.2331480291644552, + "grad_norm": 0.8353829979896545, + "learning_rate": 0.0003850291349170775, + "loss": 0.3451, + "step": 22071 + }, + { + "epoch": 1.233203899768137, + "grad_norm": 1.5091123580932617, + "learning_rate": 0.00038500112057373377, + "loss": 0.3807, + "step": 22072 + }, + { + "epoch": 1.2332597703718189, + "grad_norm": 4.465595722198486, + "learning_rate": 0.0003849731062303899, + "loss": 0.4988, + "step": 22073 + }, + { + "epoch": 1.2333156409755008, + "grad_norm": 0.4359217584133148, + "learning_rate": 0.0003849450918870462, + "loss": 0.3789, + "step": 22074 + }, + { + "epoch": 1.2333715115791826, + "grad_norm": 0.465413898229599, + "learning_rate": 0.0003849170775437024, + "loss": 0.4598, + "step": 22075 + }, + { + "epoch": 1.2334273821828645, + "grad_norm": 0.36445316672325134, + "learning_rate": 0.0003848890632003586, + "loss": 0.4519, + "step": 22076 + }, + { + "epoch": 1.2334832527865465, + "grad_norm": 0.40972334146499634, + "learning_rate": 0.0003848610488570148, + "loss": 0.4458, + "step": 22077 + }, + { + "epoch": 1.2335391233902282, + "grad_norm": 0.46612170338630676, + "learning_rate": 0.000384833034513671, + "loss": 0.3959, + "step": 22078 + }, + { + "epoch": 1.2335949939939101, + "grad_norm": 0.3292100429534912, + "learning_rate": 0.0003848050201703272, + "loss": 0.3262, + "step": 22079 + }, + { + "epoch": 1.233650864597592, + "grad_norm": 0.9177613854408264, + "learning_rate": 0.0003847770058269834, + "loss": 0.3229, + "step": 22080 + }, + { + "epoch": 1.2337067352012738, + "grad_norm": 0.6356449723243713, + "learning_rate": 0.0003847489914836396, + "loss": 0.4248, + "step": 22081 + }, + { + "epoch": 1.2337626058049558, + "grad_norm": 0.43858957290649414, + "learning_rate": 0.00038472097714029583, + "loss": 0.4498, + "step": 22082 + }, + { + "epoch": 1.2338184764086375, + "grad_norm": 0.615395724773407, + "learning_rate": 0.00038469296279695204, + "loss": 0.4195, + "step": 22083 + }, + { + "epoch": 1.2338743470123195, + "grad_norm": 0.4587251842021942, + "learning_rate": 0.0003846649484536083, + "loss": 0.4751, + "step": 22084 + }, + { + "epoch": 1.2339302176160014, + "grad_norm": 0.5527089834213257, + "learning_rate": 0.00038463693411026445, + "loss": 0.3476, + "step": 22085 + }, + { + "epoch": 1.2339860882196831, + "grad_norm": 0.4734629988670349, + "learning_rate": 0.0003846089197669207, + "loss": 0.3693, + "step": 22086 + }, + { + "epoch": 1.234041958823365, + "grad_norm": 0.4863887131214142, + "learning_rate": 0.00038458090542357686, + "loss": 0.4043, + "step": 22087 + }, + { + "epoch": 1.234097829427047, + "grad_norm": 0.5031855702400208, + "learning_rate": 0.0003845528910802331, + "loss": 0.596, + "step": 22088 + }, + { + "epoch": 1.2341537000307288, + "grad_norm": 7.170724391937256, + "learning_rate": 0.00038452487673688927, + "loss": 0.4269, + "step": 22089 + }, + { + "epoch": 1.2342095706344107, + "grad_norm": 0.9633040428161621, + "learning_rate": 0.0003844968623935455, + "loss": 0.5332, + "step": 22090 + }, + { + "epoch": 1.2342654412380925, + "grad_norm": 0.5291377902030945, + "learning_rate": 0.00038446884805020174, + "loss": 0.4149, + "step": 22091 + }, + { + "epoch": 1.2343213118417744, + "grad_norm": 0.3807668387889862, + "learning_rate": 0.0003844408337068579, + "loss": 0.4423, + "step": 22092 + }, + { + "epoch": 1.2343771824454564, + "grad_norm": 0.35690411925315857, + "learning_rate": 0.00038441281936351415, + "loss": 0.373, + "step": 22093 + }, + { + "epoch": 1.234433053049138, + "grad_norm": 0.5055349469184875, + "learning_rate": 0.0003843848050201703, + "loss": 0.4591, + "step": 22094 + }, + { + "epoch": 1.23448892365282, + "grad_norm": 0.8165109157562256, + "learning_rate": 0.00038435679067682656, + "loss": 0.6159, + "step": 22095 + }, + { + "epoch": 1.234544794256502, + "grad_norm": 0.4793422222137451, + "learning_rate": 0.0003843287763334827, + "loss": 0.4157, + "step": 22096 + }, + { + "epoch": 1.2346006648601837, + "grad_norm": 0.4225730299949646, + "learning_rate": 0.00038430076199013897, + "loss": 0.3626, + "step": 22097 + }, + { + "epoch": 1.2346565354638657, + "grad_norm": 0.322427362203598, + "learning_rate": 0.0003842727476467951, + "loss": 0.4219, + "step": 22098 + }, + { + "epoch": 1.2347124060675476, + "grad_norm": 3.6834537982940674, + "learning_rate": 0.0003842447333034514, + "loss": 0.5735, + "step": 22099 + }, + { + "epoch": 1.2347682766712293, + "grad_norm": 0.4435715973377228, + "learning_rate": 0.0003842167189601076, + "loss": 0.5268, + "step": 22100 + }, + { + "epoch": 1.2348241472749113, + "grad_norm": 0.7339180111885071, + "learning_rate": 0.0003841887046167638, + "loss": 0.4457, + "step": 22101 + }, + { + "epoch": 1.2348800178785933, + "grad_norm": 0.5640562772750854, + "learning_rate": 0.00038416069027342, + "loss": 0.3899, + "step": 22102 + }, + { + "epoch": 1.234935888482275, + "grad_norm": 0.8491385579109192, + "learning_rate": 0.0003841326759300762, + "loss": 0.3928, + "step": 22103 + }, + { + "epoch": 1.234991759085957, + "grad_norm": 0.55189049243927, + "learning_rate": 0.0003841046615867324, + "loss": 0.4801, + "step": 22104 + }, + { + "epoch": 1.2350476296896389, + "grad_norm": 1.1085104942321777, + "learning_rate": 0.0003840766472433886, + "loss": 0.3638, + "step": 22105 + }, + { + "epoch": 1.2351035002933206, + "grad_norm": 2.608584403991699, + "learning_rate": 0.0003840486329000448, + "loss": 0.4443, + "step": 22106 + }, + { + "epoch": 1.2351593708970026, + "grad_norm": 1.1412060260772705, + "learning_rate": 0.0003840206185567011, + "loss": 0.4317, + "step": 22107 + }, + { + "epoch": 1.2352152415006845, + "grad_norm": 1.4748257398605347, + "learning_rate": 0.00038399260421335724, + "loss": 0.4025, + "step": 22108 + }, + { + "epoch": 1.2352711121043662, + "grad_norm": 0.41426244378089905, + "learning_rate": 0.0003839645898700135, + "loss": 0.4413, + "step": 22109 + }, + { + "epoch": 1.2353269827080482, + "grad_norm": 0.5039001703262329, + "learning_rate": 0.00038393657552666965, + "loss": 0.4474, + "step": 22110 + }, + { + "epoch": 1.2353828533117301, + "grad_norm": 0.4787699282169342, + "learning_rate": 0.00038390856118332585, + "loss": 0.404, + "step": 22111 + }, + { + "epoch": 1.2354387239154119, + "grad_norm": 1.0470391511917114, + "learning_rate": 0.00038388054683998206, + "loss": 0.3855, + "step": 22112 + }, + { + "epoch": 1.2354945945190938, + "grad_norm": 0.3264579176902771, + "learning_rate": 0.00038385253249663827, + "loss": 0.3735, + "step": 22113 + }, + { + "epoch": 1.2355504651227758, + "grad_norm": 0.8396238684654236, + "learning_rate": 0.0003838245181532945, + "loss": 0.7827, + "step": 22114 + }, + { + "epoch": 1.2356063357264575, + "grad_norm": 0.7366436123847961, + "learning_rate": 0.0003837965038099507, + "loss": 0.5144, + "step": 22115 + }, + { + "epoch": 1.2356622063301395, + "grad_norm": 1.4443719387054443, + "learning_rate": 0.00038376848946660694, + "loss": 0.4255, + "step": 22116 + }, + { + "epoch": 1.2357180769338212, + "grad_norm": 0.8236305713653564, + "learning_rate": 0.0003837404751232631, + "loss": 0.4515, + "step": 22117 + }, + { + "epoch": 1.2357739475375031, + "grad_norm": 0.4317024052143097, + "learning_rate": 0.00038371246077991935, + "loss": 0.2978, + "step": 22118 + }, + { + "epoch": 1.235829818141185, + "grad_norm": 20.0855712890625, + "learning_rate": 0.0003836844464365755, + "loss": 0.4134, + "step": 22119 + }, + { + "epoch": 1.2358856887448668, + "grad_norm": 1.8305392265319824, + "learning_rate": 0.00038365643209323176, + "loss": 0.5683, + "step": 22120 + }, + { + "epoch": 1.2359415593485488, + "grad_norm": 0.38242781162261963, + "learning_rate": 0.0003836284177498879, + "loss": 0.3881, + "step": 22121 + }, + { + "epoch": 1.2359974299522307, + "grad_norm": 0.5439473390579224, + "learning_rate": 0.00038360040340654417, + "loss": 0.4288, + "step": 22122 + }, + { + "epoch": 1.2360533005559124, + "grad_norm": 0.610670268535614, + "learning_rate": 0.0003835723890632004, + "loss": 0.3966, + "step": 22123 + }, + { + "epoch": 1.2361091711595944, + "grad_norm": 0.6913561820983887, + "learning_rate": 0.0003835443747198566, + "loss": 0.3822, + "step": 22124 + }, + { + "epoch": 1.2361650417632761, + "grad_norm": 0.49412989616394043, + "learning_rate": 0.0003835163603765128, + "loss": 0.4131, + "step": 22125 + }, + { + "epoch": 1.236220912366958, + "grad_norm": 0.4156755208969116, + "learning_rate": 0.000383488346033169, + "loss": 0.3751, + "step": 22126 + }, + { + "epoch": 1.23627678297064, + "grad_norm": 0.3556377589702606, + "learning_rate": 0.0003834603316898252, + "loss": 0.3629, + "step": 22127 + }, + { + "epoch": 1.2363326535743218, + "grad_norm": 0.5613542795181274, + "learning_rate": 0.0003834323173464814, + "loss": 0.3522, + "step": 22128 + }, + { + "epoch": 1.2363885241780037, + "grad_norm": 1.3232223987579346, + "learning_rate": 0.0003834043030031376, + "loss": 0.5346, + "step": 22129 + }, + { + "epoch": 1.2364443947816857, + "grad_norm": 0.4431587755680084, + "learning_rate": 0.00038337628865979387, + "loss": 0.3615, + "step": 22130 + }, + { + "epoch": 1.2365002653853674, + "grad_norm": 0.36514315009117126, + "learning_rate": 0.00038334827431645, + "loss": 0.38, + "step": 22131 + }, + { + "epoch": 1.2365561359890493, + "grad_norm": 0.43991604447364807, + "learning_rate": 0.00038332025997310623, + "loss": 0.4872, + "step": 22132 + }, + { + "epoch": 1.2366120065927313, + "grad_norm": 0.3329859673976898, + "learning_rate": 0.00038329224562976244, + "loss": 0.3609, + "step": 22133 + }, + { + "epoch": 1.236667877196413, + "grad_norm": 0.9053539037704468, + "learning_rate": 0.00038326423128641864, + "loss": 0.4661, + "step": 22134 + }, + { + "epoch": 1.236723747800095, + "grad_norm": 0.5121911764144897, + "learning_rate": 0.00038323621694307485, + "loss": 0.4543, + "step": 22135 + }, + { + "epoch": 1.236779618403777, + "grad_norm": 0.4803566038608551, + "learning_rate": 0.00038320820259973105, + "loss": 0.4662, + "step": 22136 + }, + { + "epoch": 1.2368354890074587, + "grad_norm": 0.344549298286438, + "learning_rate": 0.00038318018825638726, + "loss": 0.3508, + "step": 22137 + }, + { + "epoch": 1.2368913596111406, + "grad_norm": 0.37081900238990784, + "learning_rate": 0.00038315217391304347, + "loss": 0.4256, + "step": 22138 + }, + { + "epoch": 1.2369472302148226, + "grad_norm": 0.8075159788131714, + "learning_rate": 0.0003831241595696997, + "loss": 0.3826, + "step": 22139 + }, + { + "epoch": 1.2370031008185043, + "grad_norm": 0.5195875763893127, + "learning_rate": 0.0003830961452263559, + "loss": 0.4309, + "step": 22140 + }, + { + "epoch": 1.2370589714221862, + "grad_norm": 0.726192831993103, + "learning_rate": 0.00038306813088301214, + "loss": 0.5012, + "step": 22141 + }, + { + "epoch": 1.2371148420258682, + "grad_norm": 0.43199217319488525, + "learning_rate": 0.0003830401165396683, + "loss": 0.319, + "step": 22142 + }, + { + "epoch": 1.23717071262955, + "grad_norm": 0.3762364387512207, + "learning_rate": 0.00038301210219632455, + "loss": 0.3968, + "step": 22143 + }, + { + "epoch": 1.2372265832332319, + "grad_norm": 0.4929472506046295, + "learning_rate": 0.0003829840878529807, + "loss": 0.3932, + "step": 22144 + }, + { + "epoch": 1.2372824538369138, + "grad_norm": 0.4044296443462372, + "learning_rate": 0.00038295607350963696, + "loss": 0.3369, + "step": 22145 + }, + { + "epoch": 1.2373383244405955, + "grad_norm": 0.43806010484695435, + "learning_rate": 0.00038292805916629317, + "loss": 0.481, + "step": 22146 + }, + { + "epoch": 1.2373941950442775, + "grad_norm": 2.760568857192993, + "learning_rate": 0.00038290004482294937, + "loss": 0.3636, + "step": 22147 + }, + { + "epoch": 1.2374500656479592, + "grad_norm": 0.5466210246086121, + "learning_rate": 0.0003828720304796056, + "loss": 0.443, + "step": 22148 + }, + { + "epoch": 1.2375059362516412, + "grad_norm": 0.46739524602890015, + "learning_rate": 0.0003828440161362618, + "loss": 0.3808, + "step": 22149 + }, + { + "epoch": 1.2375618068553231, + "grad_norm": 0.7085709571838379, + "learning_rate": 0.000382816001792918, + "loss": 0.4199, + "step": 22150 + }, + { + "epoch": 1.2376176774590049, + "grad_norm": 0.7449663877487183, + "learning_rate": 0.0003827879874495742, + "loss": 0.5437, + "step": 22151 + }, + { + "epoch": 1.2376735480626868, + "grad_norm": 0.5222344398498535, + "learning_rate": 0.0003827599731062304, + "loss": 0.3605, + "step": 22152 + }, + { + "epoch": 1.2377294186663688, + "grad_norm": 0.5835716128349304, + "learning_rate": 0.00038273195876288655, + "loss": 0.4176, + "step": 22153 + }, + { + "epoch": 1.2377852892700505, + "grad_norm": 0.4987730383872986, + "learning_rate": 0.0003827039444195428, + "loss": 0.4459, + "step": 22154 + }, + { + "epoch": 1.2378411598737324, + "grad_norm": 0.3153283894062042, + "learning_rate": 0.000382675930076199, + "loss": 0.3454, + "step": 22155 + }, + { + "epoch": 1.2378970304774144, + "grad_norm": 0.9783908724784851, + "learning_rate": 0.0003826479157328552, + "loss": 0.4653, + "step": 22156 + }, + { + "epoch": 1.2379529010810961, + "grad_norm": 0.5315763354301453, + "learning_rate": 0.00038261990138951143, + "loss": 0.5369, + "step": 22157 + }, + { + "epoch": 1.238008771684778, + "grad_norm": 0.5541585683822632, + "learning_rate": 0.00038259188704616764, + "loss": 0.462, + "step": 22158 + }, + { + "epoch": 1.2380646422884598, + "grad_norm": 2.163034439086914, + "learning_rate": 0.00038256387270282384, + "loss": 0.4116, + "step": 22159 + }, + { + "epoch": 1.2381205128921418, + "grad_norm": 0.46736761927604675, + "learning_rate": 0.00038253585835948005, + "loss": 0.3878, + "step": 22160 + }, + { + "epoch": 1.2381763834958237, + "grad_norm": 0.35733386874198914, + "learning_rate": 0.00038250784401613625, + "loss": 0.395, + "step": 22161 + }, + { + "epoch": 1.2382322540995054, + "grad_norm": 0.4237576127052307, + "learning_rate": 0.0003824798296727925, + "loss": 0.37, + "step": 22162 + }, + { + "epoch": 1.2382881247031874, + "grad_norm": 0.38146939873695374, + "learning_rate": 0.00038245181532944866, + "loss": 0.3834, + "step": 22163 + }, + { + "epoch": 1.2383439953068693, + "grad_norm": 1.085280418395996, + "learning_rate": 0.0003824238009861049, + "loss": 0.4508, + "step": 22164 + }, + { + "epoch": 1.238399865910551, + "grad_norm": 0.4025571644306183, + "learning_rate": 0.0003823957866427611, + "loss": 0.3803, + "step": 22165 + }, + { + "epoch": 1.238455736514233, + "grad_norm": 0.6787862777709961, + "learning_rate": 0.00038236777229941734, + "loss": 0.474, + "step": 22166 + }, + { + "epoch": 1.238511607117915, + "grad_norm": 0.49670740962028503, + "learning_rate": 0.0003823397579560735, + "loss": 0.4428, + "step": 22167 + }, + { + "epoch": 1.2385674777215967, + "grad_norm": 0.43930745124816895, + "learning_rate": 0.00038231174361272975, + "loss": 0.471, + "step": 22168 + }, + { + "epoch": 1.2386233483252787, + "grad_norm": 0.46799877285957336, + "learning_rate": 0.00038228372926938595, + "loss": 0.4798, + "step": 22169 + }, + { + "epoch": 1.2386792189289606, + "grad_norm": 0.4903738498687744, + "learning_rate": 0.00038225571492604216, + "loss": 0.4364, + "step": 22170 + }, + { + "epoch": 1.2387350895326423, + "grad_norm": 1.1268013715744019, + "learning_rate": 0.00038222770058269837, + "loss": 0.3921, + "step": 22171 + }, + { + "epoch": 1.2387909601363243, + "grad_norm": 1.9064656496047974, + "learning_rate": 0.00038219968623935457, + "loss": 0.493, + "step": 22172 + }, + { + "epoch": 1.2388468307400062, + "grad_norm": 0.37023162841796875, + "learning_rate": 0.0003821716718960108, + "loss": 0.3485, + "step": 22173 + }, + { + "epoch": 1.238902701343688, + "grad_norm": 0.5183319449424744, + "learning_rate": 0.000382143657552667, + "loss": 0.4328, + "step": 22174 + }, + { + "epoch": 1.23895857194737, + "grad_norm": 0.7660924196243286, + "learning_rate": 0.0003821156432093232, + "loss": 0.474, + "step": 22175 + }, + { + "epoch": 1.2390144425510519, + "grad_norm": 0.41471219062805176, + "learning_rate": 0.00038208762886597934, + "loss": 0.5203, + "step": 22176 + }, + { + "epoch": 1.2390703131547336, + "grad_norm": 0.35721442103385925, + "learning_rate": 0.0003820596145226356, + "loss": 0.3614, + "step": 22177 + }, + { + "epoch": 1.2391261837584155, + "grad_norm": 3.9719550609588623, + "learning_rate": 0.0003820316001792918, + "loss": 0.5257, + "step": 22178 + }, + { + "epoch": 1.2391820543620975, + "grad_norm": 0.5735817551612854, + "learning_rate": 0.000382003585835948, + "loss": 0.4369, + "step": 22179 + }, + { + "epoch": 1.2392379249657792, + "grad_norm": 0.39598214626312256, + "learning_rate": 0.0003819755714926042, + "loss": 0.3994, + "step": 22180 + }, + { + "epoch": 1.2392937955694612, + "grad_norm": 1.105025053024292, + "learning_rate": 0.0003819475571492604, + "loss": 0.8979, + "step": 22181 + }, + { + "epoch": 1.239349666173143, + "grad_norm": 0.37233054637908936, + "learning_rate": 0.00038191954280591663, + "loss": 0.4096, + "step": 22182 + }, + { + "epoch": 1.2394055367768249, + "grad_norm": 0.4781186282634735, + "learning_rate": 0.00038189152846257284, + "loss": 0.4139, + "step": 22183 + }, + { + "epoch": 1.2394614073805068, + "grad_norm": 0.5105757713317871, + "learning_rate": 0.00038186351411922904, + "loss": 0.6914, + "step": 22184 + }, + { + "epoch": 1.2395172779841885, + "grad_norm": 0.4273495376110077, + "learning_rate": 0.0003818354997758853, + "loss": 0.3671, + "step": 22185 + }, + { + "epoch": 1.2395731485878705, + "grad_norm": 0.5067193508148193, + "learning_rate": 0.00038180748543254145, + "loss": 0.4864, + "step": 22186 + }, + { + "epoch": 1.2396290191915524, + "grad_norm": 0.3471386730670929, + "learning_rate": 0.0003817794710891977, + "loss": 0.33, + "step": 22187 + }, + { + "epoch": 1.2396848897952342, + "grad_norm": 0.33355921506881714, + "learning_rate": 0.00038175145674585386, + "loss": 0.3827, + "step": 22188 + }, + { + "epoch": 1.2397407603989161, + "grad_norm": 0.43913644552230835, + "learning_rate": 0.0003817234424025101, + "loss": 0.4047, + "step": 22189 + }, + { + "epoch": 1.239796631002598, + "grad_norm": 0.43768179416656494, + "learning_rate": 0.0003816954280591663, + "loss": 0.3754, + "step": 22190 + }, + { + "epoch": 1.2398525016062798, + "grad_norm": 1.3574825525283813, + "learning_rate": 0.00038166741371582254, + "loss": 0.6369, + "step": 22191 + }, + { + "epoch": 1.2399083722099618, + "grad_norm": 0.5751194357872009, + "learning_rate": 0.0003816393993724787, + "loss": 0.4548, + "step": 22192 + }, + { + "epoch": 1.2399642428136435, + "grad_norm": 0.3140955865383148, + "learning_rate": 0.00038161138502913495, + "loss": 0.3236, + "step": 22193 + }, + { + "epoch": 1.2400201134173254, + "grad_norm": 0.9700238704681396, + "learning_rate": 0.00038158337068579115, + "loss": 0.4459, + "step": 22194 + }, + { + "epoch": 1.2400759840210074, + "grad_norm": 1.6099621057510376, + "learning_rate": 0.00038155535634244736, + "loss": 0.4354, + "step": 22195 + }, + { + "epoch": 1.2401318546246891, + "grad_norm": 1.7534199953079224, + "learning_rate": 0.00038152734199910356, + "loss": 0.4333, + "step": 22196 + }, + { + "epoch": 1.240187725228371, + "grad_norm": 0.6659896969795227, + "learning_rate": 0.0003814993276557597, + "loss": 0.485, + "step": 22197 + }, + { + "epoch": 1.240243595832053, + "grad_norm": 0.7331928014755249, + "learning_rate": 0.000381471313312416, + "loss": 0.3924, + "step": 22198 + }, + { + "epoch": 1.2402994664357347, + "grad_norm": 0.9988560080528259, + "learning_rate": 0.00038144329896907213, + "loss": 0.4125, + "step": 22199 + }, + { + "epoch": 1.2403553370394167, + "grad_norm": 0.652360200881958, + "learning_rate": 0.0003814152846257284, + "loss": 0.3905, + "step": 22200 + }, + { + "epoch": 1.2404112076430986, + "grad_norm": 0.3241461515426636, + "learning_rate": 0.0003813872702823846, + "loss": 0.3571, + "step": 22201 + }, + { + "epoch": 1.2404670782467804, + "grad_norm": 1.5724233388900757, + "learning_rate": 0.0003813592559390408, + "loss": 0.4186, + "step": 22202 + }, + { + "epoch": 1.2405229488504623, + "grad_norm": 0.5505042672157288, + "learning_rate": 0.000381331241595697, + "loss": 0.4454, + "step": 22203 + }, + { + "epoch": 1.2405788194541443, + "grad_norm": 7.416508674621582, + "learning_rate": 0.0003813032272523532, + "loss": 0.3867, + "step": 22204 + }, + { + "epoch": 1.240634690057826, + "grad_norm": 0.5139902830123901, + "learning_rate": 0.0003812752129090094, + "loss": 0.3765, + "step": 22205 + }, + { + "epoch": 1.240690560661508, + "grad_norm": 0.46588394045829773, + "learning_rate": 0.0003812471985656656, + "loss": 0.4013, + "step": 22206 + }, + { + "epoch": 1.24074643126519, + "grad_norm": 2.2968881130218506, + "learning_rate": 0.00038121918422232183, + "loss": 0.4047, + "step": 22207 + }, + { + "epoch": 1.2408023018688716, + "grad_norm": 0.42349088191986084, + "learning_rate": 0.00038119116987897803, + "loss": 0.4761, + "step": 22208 + }, + { + "epoch": 1.2408581724725536, + "grad_norm": 0.42675670981407166, + "learning_rate": 0.00038116315553563424, + "loss": 0.4099, + "step": 22209 + }, + { + "epoch": 1.2409140430762355, + "grad_norm": 0.39291220903396606, + "learning_rate": 0.0003811351411922905, + "loss": 0.3609, + "step": 22210 + }, + { + "epoch": 1.2409699136799173, + "grad_norm": 0.5601966977119446, + "learning_rate": 0.00038110712684894665, + "loss": 0.3914, + "step": 22211 + }, + { + "epoch": 1.2410257842835992, + "grad_norm": 0.36858227849006653, + "learning_rate": 0.0003810791125056029, + "loss": 0.3751, + "step": 22212 + }, + { + "epoch": 1.2410816548872812, + "grad_norm": 1.2626174688339233, + "learning_rate": 0.00038105109816225906, + "loss": 0.4482, + "step": 22213 + }, + { + "epoch": 1.241137525490963, + "grad_norm": 0.8042632341384888, + "learning_rate": 0.0003810230838189153, + "loss": 0.4045, + "step": 22214 + }, + { + "epoch": 1.2411933960946449, + "grad_norm": 0.40652915835380554, + "learning_rate": 0.0003809950694755715, + "loss": 0.4207, + "step": 22215 + }, + { + "epoch": 1.2412492666983266, + "grad_norm": 0.5462132692337036, + "learning_rate": 0.00038096705513222774, + "loss": 0.4174, + "step": 22216 + }, + { + "epoch": 1.2413051373020085, + "grad_norm": 0.4265103042125702, + "learning_rate": 0.00038093904078888394, + "loss": 0.4282, + "step": 22217 + }, + { + "epoch": 1.2413610079056905, + "grad_norm": 0.41260454058647156, + "learning_rate": 0.0003809110264455401, + "loss": 0.3671, + "step": 22218 + }, + { + "epoch": 1.2414168785093722, + "grad_norm": 0.40576663613319397, + "learning_rate": 0.00038088301210219635, + "loss": 0.3692, + "step": 22219 + }, + { + "epoch": 1.2414727491130542, + "grad_norm": 0.4526228904724121, + "learning_rate": 0.0003808549977588525, + "loss": 0.4179, + "step": 22220 + }, + { + "epoch": 1.2415286197167361, + "grad_norm": 0.8006122708320618, + "learning_rate": 0.00038082698341550876, + "loss": 0.4189, + "step": 22221 + }, + { + "epoch": 1.2415844903204178, + "grad_norm": 0.3902728259563446, + "learning_rate": 0.0003807989690721649, + "loss": 0.521, + "step": 22222 + }, + { + "epoch": 1.2416403609240998, + "grad_norm": 0.9971689581871033, + "learning_rate": 0.0003807709547288212, + "loss": 0.4428, + "step": 22223 + }, + { + "epoch": 1.2416962315277817, + "grad_norm": 0.9970661997795105, + "learning_rate": 0.00038074294038547733, + "loss": 0.4869, + "step": 22224 + }, + { + "epoch": 1.2417521021314635, + "grad_norm": 0.5467401742935181, + "learning_rate": 0.0003807149260421336, + "loss": 0.4429, + "step": 22225 + }, + { + "epoch": 1.2418079727351454, + "grad_norm": 0.6356617212295532, + "learning_rate": 0.0003806869116987898, + "loss": 0.4963, + "step": 22226 + }, + { + "epoch": 1.2418638433388272, + "grad_norm": 1.5432418584823608, + "learning_rate": 0.000380658897355446, + "loss": 0.41, + "step": 22227 + }, + { + "epoch": 1.241919713942509, + "grad_norm": 0.4003041684627533, + "learning_rate": 0.0003806308830121022, + "loss": 0.3804, + "step": 22228 + }, + { + "epoch": 1.241975584546191, + "grad_norm": 0.5486892461776733, + "learning_rate": 0.0003806028686687584, + "loss": 0.4338, + "step": 22229 + }, + { + "epoch": 1.2420314551498728, + "grad_norm": 0.88789963722229, + "learning_rate": 0.0003805748543254146, + "loss": 0.3602, + "step": 22230 + }, + { + "epoch": 1.2420873257535547, + "grad_norm": 0.359703004360199, + "learning_rate": 0.0003805468399820708, + "loss": 0.3512, + "step": 22231 + }, + { + "epoch": 1.2421431963572367, + "grad_norm": 0.45967134833335876, + "learning_rate": 0.00038051882563872703, + "loss": 0.4489, + "step": 22232 + }, + { + "epoch": 1.2421990669609184, + "grad_norm": 1.3535155057907104, + "learning_rate": 0.0003804908112953833, + "loss": 0.4566, + "step": 22233 + }, + { + "epoch": 1.2422549375646004, + "grad_norm": 0.618141770362854, + "learning_rate": 0.00038046279695203944, + "loss": 0.409, + "step": 22234 + }, + { + "epoch": 1.2423108081682823, + "grad_norm": 0.7595838904380798, + "learning_rate": 0.0003804347826086957, + "loss": 0.4957, + "step": 22235 + }, + { + "epoch": 1.242366678771964, + "grad_norm": 0.36247995495796204, + "learning_rate": 0.00038040676826535185, + "loss": 0.3937, + "step": 22236 + }, + { + "epoch": 1.242422549375646, + "grad_norm": 0.39464959502220154, + "learning_rate": 0.0003803787539220081, + "loss": 0.3804, + "step": 22237 + }, + { + "epoch": 1.242478419979328, + "grad_norm": 0.3614933490753174, + "learning_rate": 0.00038035073957866426, + "loss": 0.4394, + "step": 22238 + }, + { + "epoch": 1.2425342905830097, + "grad_norm": 2.5543885231018066, + "learning_rate": 0.00038032272523532047, + "loss": 0.5421, + "step": 22239 + }, + { + "epoch": 1.2425901611866916, + "grad_norm": 0.4391956627368927, + "learning_rate": 0.00038029471089197673, + "loss": 0.3495, + "step": 22240 + }, + { + "epoch": 1.2426460317903736, + "grad_norm": 0.7509691119194031, + "learning_rate": 0.0003802666965486329, + "loss": 0.3658, + "step": 22241 + }, + { + "epoch": 1.2427019023940553, + "grad_norm": 0.7995715737342834, + "learning_rate": 0.00038023868220528914, + "loss": 0.4226, + "step": 22242 + }, + { + "epoch": 1.2427577729977373, + "grad_norm": 0.621570885181427, + "learning_rate": 0.0003802106678619453, + "loss": 0.5208, + "step": 22243 + }, + { + "epoch": 1.2428136436014192, + "grad_norm": 0.537453830242157, + "learning_rate": 0.00038018265351860155, + "loss": 0.3882, + "step": 22244 + }, + { + "epoch": 1.242869514205101, + "grad_norm": 5.0931715965271, + "learning_rate": 0.0003801546391752577, + "loss": 0.499, + "step": 22245 + }, + { + "epoch": 1.242925384808783, + "grad_norm": 0.48329398036003113, + "learning_rate": 0.00038012662483191396, + "loss": 0.4055, + "step": 22246 + }, + { + "epoch": 1.2429812554124648, + "grad_norm": 0.8483046889305115, + "learning_rate": 0.0003800986104885701, + "loss": 0.4179, + "step": 22247 + }, + { + "epoch": 1.2430371260161466, + "grad_norm": 1.3760179281234741, + "learning_rate": 0.0003800705961452264, + "loss": 0.4117, + "step": 22248 + }, + { + "epoch": 1.2430929966198285, + "grad_norm": 0.6310622692108154, + "learning_rate": 0.0003800425818018826, + "loss": 0.4797, + "step": 22249 + }, + { + "epoch": 1.2431488672235103, + "grad_norm": 0.5627763271331787, + "learning_rate": 0.0003800145674585388, + "loss": 0.4136, + "step": 22250 + }, + { + "epoch": 1.2432047378271922, + "grad_norm": 0.7118086218833923, + "learning_rate": 0.000379986553115195, + "loss": 0.5765, + "step": 22251 + }, + { + "epoch": 1.2432606084308742, + "grad_norm": 0.413181334733963, + "learning_rate": 0.0003799585387718512, + "loss": 0.3988, + "step": 22252 + }, + { + "epoch": 1.2433164790345559, + "grad_norm": 0.5951038599014282, + "learning_rate": 0.0003799305244285074, + "loss": 0.512, + "step": 22253 + }, + { + "epoch": 1.2433723496382378, + "grad_norm": 3.2922003269195557, + "learning_rate": 0.0003799025100851636, + "loss": 0.666, + "step": 22254 + }, + { + "epoch": 1.2434282202419198, + "grad_norm": 2.9616754055023193, + "learning_rate": 0.0003798744957418198, + "loss": 0.5312, + "step": 22255 + }, + { + "epoch": 1.2434840908456015, + "grad_norm": 0.7243424654006958, + "learning_rate": 0.0003798464813984761, + "loss": 0.4769, + "step": 22256 + }, + { + "epoch": 1.2435399614492835, + "grad_norm": 0.4163322150707245, + "learning_rate": 0.00037981846705513223, + "loss": 0.4281, + "step": 22257 + }, + { + "epoch": 1.2435958320529654, + "grad_norm": 0.6142958402633667, + "learning_rate": 0.0003797904527117885, + "loss": 0.3851, + "step": 22258 + }, + { + "epoch": 1.2436517026566472, + "grad_norm": 0.5269081592559814, + "learning_rate": 0.00037976243836844464, + "loss": 0.5835, + "step": 22259 + }, + { + "epoch": 1.243707573260329, + "grad_norm": 0.4491977393627167, + "learning_rate": 0.00037973442402510085, + "loss": 0.4515, + "step": 22260 + }, + { + "epoch": 1.2437634438640108, + "grad_norm": 0.5265170931816101, + "learning_rate": 0.00037970640968175705, + "loss": 0.5195, + "step": 22261 + }, + { + "epoch": 1.2438193144676928, + "grad_norm": 0.670263409614563, + "learning_rate": 0.00037967839533841326, + "loss": 0.536, + "step": 22262 + }, + { + "epoch": 1.2438751850713747, + "grad_norm": 0.5775216221809387, + "learning_rate": 0.00037965038099506946, + "loss": 0.4321, + "step": 22263 + }, + { + "epoch": 1.2439310556750565, + "grad_norm": 0.6732906699180603, + "learning_rate": 0.00037962236665172567, + "loss": 0.6163, + "step": 22264 + }, + { + "epoch": 1.2439869262787384, + "grad_norm": 0.44938376545906067, + "learning_rate": 0.00037959435230838193, + "loss": 0.4855, + "step": 22265 + }, + { + "epoch": 1.2440427968824204, + "grad_norm": 0.4198455810546875, + "learning_rate": 0.0003795663379650381, + "loss": 0.3743, + "step": 22266 + }, + { + "epoch": 1.244098667486102, + "grad_norm": 0.7637399435043335, + "learning_rate": 0.00037953832362169434, + "loss": 0.3988, + "step": 22267 + }, + { + "epoch": 1.244154538089784, + "grad_norm": 0.3389941155910492, + "learning_rate": 0.0003795103092783505, + "loss": 0.4842, + "step": 22268 + }, + { + "epoch": 1.244210408693466, + "grad_norm": 0.4376116096973419, + "learning_rate": 0.00037948229493500675, + "loss": 0.3433, + "step": 22269 + }, + { + "epoch": 1.2442662792971477, + "grad_norm": 0.9643962979316711, + "learning_rate": 0.0003794542805916629, + "loss": 0.4248, + "step": 22270 + }, + { + "epoch": 1.2443221499008297, + "grad_norm": 2.5233943462371826, + "learning_rate": 0.00037942626624831916, + "loss": 0.4329, + "step": 22271 + }, + { + "epoch": 1.2443780205045116, + "grad_norm": 0.4018675982952118, + "learning_rate": 0.00037939825190497537, + "loss": 0.4128, + "step": 22272 + }, + { + "epoch": 1.2444338911081934, + "grad_norm": 0.8601824641227722, + "learning_rate": 0.0003793702375616316, + "loss": 0.4448, + "step": 22273 + }, + { + "epoch": 1.2444897617118753, + "grad_norm": 0.3547903001308441, + "learning_rate": 0.0003793422232182878, + "loss": 0.4047, + "step": 22274 + }, + { + "epoch": 1.2445456323155573, + "grad_norm": 0.35034510493278503, + "learning_rate": 0.000379314208874944, + "loss": 0.4128, + "step": 22275 + }, + { + "epoch": 1.244601502919239, + "grad_norm": 0.472555935382843, + "learning_rate": 0.0003792861945316002, + "loss": 0.4242, + "step": 22276 + }, + { + "epoch": 1.244657373522921, + "grad_norm": 0.671947717666626, + "learning_rate": 0.0003792581801882564, + "loss": 0.4916, + "step": 22277 + }, + { + "epoch": 1.244713244126603, + "grad_norm": 0.45328348875045776, + "learning_rate": 0.0003792301658449126, + "loss": 0.4906, + "step": 22278 + }, + { + "epoch": 1.2447691147302846, + "grad_norm": 0.42913758754730225, + "learning_rate": 0.0003792021515015688, + "loss": 0.3791, + "step": 22279 + }, + { + "epoch": 1.2448249853339666, + "grad_norm": 2.416449785232544, + "learning_rate": 0.000379174137158225, + "loss": 0.4487, + "step": 22280 + }, + { + "epoch": 1.2448808559376485, + "grad_norm": 0.7528881430625916, + "learning_rate": 0.0003791461228148812, + "loss": 0.498, + "step": 22281 + }, + { + "epoch": 1.2449367265413303, + "grad_norm": 5.271004676818848, + "learning_rate": 0.00037911810847153743, + "loss": 0.2796, + "step": 22282 + }, + { + "epoch": 1.2449925971450122, + "grad_norm": 0.7692973613739014, + "learning_rate": 0.00037909009412819363, + "loss": 0.531, + "step": 22283 + }, + { + "epoch": 1.245048467748694, + "grad_norm": 0.41921335458755493, + "learning_rate": 0.00037906207978484984, + "loss": 0.4619, + "step": 22284 + }, + { + "epoch": 1.2451043383523759, + "grad_norm": 0.46546563506126404, + "learning_rate": 0.00037903406544150605, + "loss": 0.4458, + "step": 22285 + }, + { + "epoch": 1.2451602089560578, + "grad_norm": 0.4347539246082306, + "learning_rate": 0.00037900605109816225, + "loss": 0.4928, + "step": 22286 + }, + { + "epoch": 1.2452160795597396, + "grad_norm": 0.335897296667099, + "learning_rate": 0.00037897803675481846, + "loss": 0.3566, + "step": 22287 + }, + { + "epoch": 1.2452719501634215, + "grad_norm": 12.626581192016602, + "learning_rate": 0.0003789500224114747, + "loss": 0.3925, + "step": 22288 + }, + { + "epoch": 1.2453278207671035, + "grad_norm": 0.5169591903686523, + "learning_rate": 0.00037892200806813087, + "loss": 0.3886, + "step": 22289 + }, + { + "epoch": 1.2453836913707852, + "grad_norm": 0.41169852018356323, + "learning_rate": 0.00037889399372478713, + "loss": 0.4102, + "step": 22290 + }, + { + "epoch": 1.2454395619744671, + "grad_norm": 1.8617851734161377, + "learning_rate": 0.0003788659793814433, + "loss": 0.3723, + "step": 22291 + }, + { + "epoch": 1.245495432578149, + "grad_norm": 0.3990767002105713, + "learning_rate": 0.00037883796503809954, + "loss": 0.492, + "step": 22292 + }, + { + "epoch": 1.2455513031818308, + "grad_norm": 0.8348760008811951, + "learning_rate": 0.0003788099506947557, + "loss": 0.4203, + "step": 22293 + }, + { + "epoch": 1.2456071737855128, + "grad_norm": 1.102249264717102, + "learning_rate": 0.00037878193635141195, + "loss": 0.5339, + "step": 22294 + }, + { + "epoch": 1.2456630443891945, + "grad_norm": 0.4700538218021393, + "learning_rate": 0.0003787539220080681, + "loss": 0.4611, + "step": 22295 + }, + { + "epoch": 1.2457189149928765, + "grad_norm": 0.5563593506813049, + "learning_rate": 0.00037872590766472436, + "loss": 0.439, + "step": 22296 + }, + { + "epoch": 1.2457747855965584, + "grad_norm": 0.4659672677516937, + "learning_rate": 0.00037869789332138057, + "loss": 0.4849, + "step": 22297 + }, + { + "epoch": 1.2458306562002401, + "grad_norm": 0.478538453578949, + "learning_rate": 0.0003786698789780368, + "loss": 0.4214, + "step": 22298 + }, + { + "epoch": 1.245886526803922, + "grad_norm": 0.29376929998397827, + "learning_rate": 0.000378641864634693, + "loss": 0.3747, + "step": 22299 + }, + { + "epoch": 1.245942397407604, + "grad_norm": 0.35298553109169006, + "learning_rate": 0.0003786138502913492, + "loss": 0.3484, + "step": 22300 + }, + { + "epoch": 1.2459982680112858, + "grad_norm": 0.3595268428325653, + "learning_rate": 0.0003785858359480054, + "loss": 0.4589, + "step": 22301 + }, + { + "epoch": 1.2460541386149677, + "grad_norm": 3.60465407371521, + "learning_rate": 0.00037855782160466154, + "loss": 0.4617, + "step": 22302 + }, + { + "epoch": 1.2461100092186497, + "grad_norm": 0.4145047962665558, + "learning_rate": 0.0003785298072613178, + "loss": 0.3997, + "step": 22303 + }, + { + "epoch": 1.2461658798223314, + "grad_norm": 0.5480155944824219, + "learning_rate": 0.000378501792917974, + "loss": 0.4978, + "step": 22304 + }, + { + "epoch": 1.2462217504260134, + "grad_norm": 0.4074835181236267, + "learning_rate": 0.0003784737785746302, + "loss": 0.327, + "step": 22305 + }, + { + "epoch": 1.2462776210296953, + "grad_norm": 0.4701080620288849, + "learning_rate": 0.0003784457642312864, + "loss": 0.5201, + "step": 22306 + }, + { + "epoch": 1.246333491633377, + "grad_norm": 2.3354907035827637, + "learning_rate": 0.00037841774988794263, + "loss": 0.338, + "step": 22307 + }, + { + "epoch": 1.246389362237059, + "grad_norm": 0.36453717947006226, + "learning_rate": 0.00037838973554459883, + "loss": 0.365, + "step": 22308 + }, + { + "epoch": 1.246445232840741, + "grad_norm": 0.4555460512638092, + "learning_rate": 0.00037836172120125504, + "loss": 0.3128, + "step": 22309 + }, + { + "epoch": 1.2465011034444227, + "grad_norm": 0.37031981348991394, + "learning_rate": 0.00037833370685791124, + "loss": 0.371, + "step": 22310 + }, + { + "epoch": 1.2465569740481046, + "grad_norm": 0.34020766615867615, + "learning_rate": 0.0003783056925145675, + "loss": 0.3356, + "step": 22311 + }, + { + "epoch": 1.2466128446517866, + "grad_norm": 0.38890284299850464, + "learning_rate": 0.00037827767817122366, + "loss": 0.408, + "step": 22312 + }, + { + "epoch": 1.2466687152554683, + "grad_norm": 0.5291041135787964, + "learning_rate": 0.0003782496638278799, + "loss": 0.4517, + "step": 22313 + }, + { + "epoch": 1.2467245858591502, + "grad_norm": 0.5618576407432556, + "learning_rate": 0.00037822164948453607, + "loss": 0.5528, + "step": 22314 + }, + { + "epoch": 1.2467804564628322, + "grad_norm": 0.4928412139415741, + "learning_rate": 0.00037819363514119233, + "loss": 0.4844, + "step": 22315 + }, + { + "epoch": 1.246836327066514, + "grad_norm": 1.9147447347640991, + "learning_rate": 0.0003781656207978485, + "loss": 0.3965, + "step": 22316 + }, + { + "epoch": 1.2468921976701959, + "grad_norm": 1.2902641296386719, + "learning_rate": 0.00037813760645450474, + "loss": 0.4592, + "step": 22317 + }, + { + "epoch": 1.2469480682738776, + "grad_norm": 1.7565152645111084, + "learning_rate": 0.0003781095921111609, + "loss": 0.4605, + "step": 22318 + }, + { + "epoch": 1.2470039388775596, + "grad_norm": 0.41859036684036255, + "learning_rate": 0.00037808157776781715, + "loss": 0.4046, + "step": 22319 + }, + { + "epoch": 1.2470598094812415, + "grad_norm": 0.5694624781608582, + "learning_rate": 0.00037805356342447336, + "loss": 0.4546, + "step": 22320 + }, + { + "epoch": 1.2471156800849232, + "grad_norm": 0.36664727330207825, + "learning_rate": 0.00037802554908112956, + "loss": 0.3837, + "step": 22321 + }, + { + "epoch": 1.2471715506886052, + "grad_norm": 0.5350744128227234, + "learning_rate": 0.00037799753473778577, + "loss": 0.4558, + "step": 22322 + }, + { + "epoch": 1.2472274212922871, + "grad_norm": 0.36135169863700867, + "learning_rate": 0.0003779695203944419, + "loss": 0.4627, + "step": 22323 + }, + { + "epoch": 1.2472832918959689, + "grad_norm": 0.5096461772918701, + "learning_rate": 0.0003779415060510982, + "loss": 0.4221, + "step": 22324 + }, + { + "epoch": 1.2473391624996508, + "grad_norm": 0.42341727018356323, + "learning_rate": 0.00037791349170775433, + "loss": 0.4315, + "step": 22325 + }, + { + "epoch": 1.2473950331033328, + "grad_norm": 0.35235530138015747, + "learning_rate": 0.0003778854773644106, + "loss": 0.4212, + "step": 22326 + }, + { + "epoch": 1.2474509037070145, + "grad_norm": 0.36862942576408386, + "learning_rate": 0.0003778574630210668, + "loss": 0.3532, + "step": 22327 + }, + { + "epoch": 1.2475067743106965, + "grad_norm": 0.41185110807418823, + "learning_rate": 0.000377829448677723, + "loss": 0.3818, + "step": 22328 + }, + { + "epoch": 1.2475626449143782, + "grad_norm": 0.4343967139720917, + "learning_rate": 0.0003778014343343792, + "loss": 0.3833, + "step": 22329 + }, + { + "epoch": 1.2476185155180601, + "grad_norm": 0.32535508275032043, + "learning_rate": 0.0003777734199910354, + "loss": 0.3265, + "step": 22330 + }, + { + "epoch": 1.247674386121742, + "grad_norm": 4.67064905166626, + "learning_rate": 0.0003777454056476916, + "loss": 0.5032, + "step": 22331 + }, + { + "epoch": 1.2477302567254238, + "grad_norm": 0.6687915921211243, + "learning_rate": 0.0003777173913043478, + "loss": 0.3806, + "step": 22332 + }, + { + "epoch": 1.2477861273291058, + "grad_norm": 0.3375919461250305, + "learning_rate": 0.00037768937696100403, + "loss": 0.4337, + "step": 22333 + }, + { + "epoch": 1.2478419979327877, + "grad_norm": 2.587874412536621, + "learning_rate": 0.00037766136261766024, + "loss": 0.3858, + "step": 22334 + }, + { + "epoch": 1.2478978685364694, + "grad_norm": 1.708436369895935, + "learning_rate": 0.00037763334827431644, + "loss": 0.4103, + "step": 22335 + }, + { + "epoch": 1.2479537391401514, + "grad_norm": 0.36097604036331177, + "learning_rate": 0.0003776053339309727, + "loss": 0.3854, + "step": 22336 + }, + { + "epoch": 1.2480096097438333, + "grad_norm": 0.6181383728981018, + "learning_rate": 0.00037757731958762886, + "loss": 0.3201, + "step": 22337 + }, + { + "epoch": 1.248065480347515, + "grad_norm": 0.5885568261146545, + "learning_rate": 0.0003775493052442851, + "loss": 0.4169, + "step": 22338 + }, + { + "epoch": 1.248121350951197, + "grad_norm": 0.8478989601135254, + "learning_rate": 0.00037752129090094127, + "loss": 0.4088, + "step": 22339 + }, + { + "epoch": 1.248177221554879, + "grad_norm": 0.6324607729911804, + "learning_rate": 0.00037749327655759753, + "loss": 0.4749, + "step": 22340 + }, + { + "epoch": 1.2482330921585607, + "grad_norm": 0.3617061674594879, + "learning_rate": 0.0003774652622142537, + "loss": 0.3171, + "step": 22341 + }, + { + "epoch": 1.2482889627622427, + "grad_norm": 0.37902477383613586, + "learning_rate": 0.00037743724787090994, + "loss": 0.4799, + "step": 22342 + }, + { + "epoch": 1.2483448333659246, + "grad_norm": 0.60105961561203, + "learning_rate": 0.00037740923352756615, + "loss": 0.3065, + "step": 22343 + }, + { + "epoch": 1.2484007039696063, + "grad_norm": 0.7204610109329224, + "learning_rate": 0.00037738121918422235, + "loss": 0.4714, + "step": 22344 + }, + { + "epoch": 1.2484565745732883, + "grad_norm": 0.4284088909626007, + "learning_rate": 0.00037735320484087856, + "loss": 0.3839, + "step": 22345 + }, + { + "epoch": 1.2485124451769702, + "grad_norm": 0.42751193046569824, + "learning_rate": 0.0003773251904975347, + "loss": 0.3874, + "step": 22346 + }, + { + "epoch": 1.248568315780652, + "grad_norm": 1.0439791679382324, + "learning_rate": 0.00037729717615419097, + "loss": 0.436, + "step": 22347 + }, + { + "epoch": 1.248624186384334, + "grad_norm": 2.562744617462158, + "learning_rate": 0.0003772691618108471, + "loss": 0.3921, + "step": 22348 + }, + { + "epoch": 1.2486800569880159, + "grad_norm": 0.3041156530380249, + "learning_rate": 0.0003772411474675034, + "loss": 0.3259, + "step": 22349 + }, + { + "epoch": 1.2487359275916976, + "grad_norm": 0.35184845328330994, + "learning_rate": 0.00037721313312415953, + "loss": 0.3851, + "step": 22350 + }, + { + "epoch": 1.2487917981953796, + "grad_norm": 1.128727912902832, + "learning_rate": 0.0003771851187808158, + "loss": 0.3301, + "step": 22351 + }, + { + "epoch": 1.2488476687990613, + "grad_norm": 0.530863344669342, + "learning_rate": 0.000377157104437472, + "loss": 0.4767, + "step": 22352 + }, + { + "epoch": 1.2489035394027432, + "grad_norm": 0.49333205819129944, + "learning_rate": 0.0003771290900941282, + "loss": 0.4822, + "step": 22353 + }, + { + "epoch": 1.2489594100064252, + "grad_norm": 0.6739469170570374, + "learning_rate": 0.0003771010757507844, + "loss": 0.3325, + "step": 22354 + }, + { + "epoch": 1.249015280610107, + "grad_norm": 0.5146006941795349, + "learning_rate": 0.0003770730614074406, + "loss": 0.5663, + "step": 22355 + }, + { + "epoch": 1.2490711512137889, + "grad_norm": 0.4454844892024994, + "learning_rate": 0.0003770450470640968, + "loss": 0.4634, + "step": 22356 + }, + { + "epoch": 1.2491270218174708, + "grad_norm": 0.7796899080276489, + "learning_rate": 0.000377017032720753, + "loss": 0.4367, + "step": 22357 + }, + { + "epoch": 1.2491828924211525, + "grad_norm": 0.4695679545402527, + "learning_rate": 0.00037698901837740923, + "loss": 0.4745, + "step": 22358 + }, + { + "epoch": 1.2492387630248345, + "grad_norm": 0.35611432790756226, + "learning_rate": 0.0003769610040340655, + "loss": 0.3809, + "step": 22359 + }, + { + "epoch": 1.2492946336285164, + "grad_norm": 0.7483975291252136, + "learning_rate": 0.00037693298969072164, + "loss": 0.5619, + "step": 22360 + }, + { + "epoch": 1.2493505042321982, + "grad_norm": 0.526287317276001, + "learning_rate": 0.0003769049753473779, + "loss": 0.442, + "step": 22361 + }, + { + "epoch": 1.2494063748358801, + "grad_norm": 0.4652131199836731, + "learning_rate": 0.00037687696100403406, + "loss": 0.4421, + "step": 22362 + }, + { + "epoch": 1.2494622454395619, + "grad_norm": 0.5279948711395264, + "learning_rate": 0.0003768489466606903, + "loss": 0.4504, + "step": 22363 + }, + { + "epoch": 1.2495181160432438, + "grad_norm": 0.7509207129478455, + "learning_rate": 0.00037682093231734647, + "loss": 0.4889, + "step": 22364 + }, + { + "epoch": 1.2495739866469258, + "grad_norm": 0.3887670040130615, + "learning_rate": 0.00037679291797400273, + "loss": 0.3969, + "step": 22365 + }, + { + "epoch": 1.2496298572506075, + "grad_norm": 2.1809237003326416, + "learning_rate": 0.00037676490363065893, + "loss": 0.4522, + "step": 22366 + }, + { + "epoch": 1.2496857278542894, + "grad_norm": 0.43216344714164734, + "learning_rate": 0.0003767368892873151, + "loss": 0.402, + "step": 22367 + }, + { + "epoch": 1.2497415984579714, + "grad_norm": 0.7593470215797424, + "learning_rate": 0.00037670887494397134, + "loss": 0.3291, + "step": 22368 + }, + { + "epoch": 1.2497974690616531, + "grad_norm": 0.5682711005210876, + "learning_rate": 0.0003766808606006275, + "loss": 0.5145, + "step": 22369 + }, + { + "epoch": 1.249853339665335, + "grad_norm": 0.5042622089385986, + "learning_rate": 0.00037665284625728376, + "loss": 0.4257, + "step": 22370 + }, + { + "epoch": 1.249909210269017, + "grad_norm": 0.9222066402435303, + "learning_rate": 0.0003766248319139399, + "loss": 0.4331, + "step": 22371 + }, + { + "epoch": 1.2499650808726988, + "grad_norm": 0.5928090810775757, + "learning_rate": 0.00037659681757059617, + "loss": 0.5868, + "step": 22372 + }, + { + "epoch": 1.2500209514763807, + "grad_norm": 1.6790715456008911, + "learning_rate": 0.0003765688032272523, + "loss": 0.3503, + "step": 22373 + }, + { + "epoch": 1.2500768220800627, + "grad_norm": 0.7727295160293579, + "learning_rate": 0.0003765407888839086, + "loss": 0.486, + "step": 22374 + }, + { + "epoch": 1.2501326926837444, + "grad_norm": 0.6548869013786316, + "learning_rate": 0.0003765127745405648, + "loss": 0.4845, + "step": 22375 + }, + { + "epoch": 1.2501885632874263, + "grad_norm": 1.9419338703155518, + "learning_rate": 0.000376484760197221, + "loss": 0.3771, + "step": 22376 + }, + { + "epoch": 1.2502444338911083, + "grad_norm": 1.0709879398345947, + "learning_rate": 0.0003764567458538772, + "loss": 0.4156, + "step": 22377 + }, + { + "epoch": 1.25030030449479, + "grad_norm": 0.6076272130012512, + "learning_rate": 0.0003764287315105334, + "loss": 0.5472, + "step": 22378 + }, + { + "epoch": 1.250356175098472, + "grad_norm": 0.37780600786209106, + "learning_rate": 0.0003764007171671896, + "loss": 0.4126, + "step": 22379 + }, + { + "epoch": 1.250412045702154, + "grad_norm": 1.4258743524551392, + "learning_rate": 0.0003763727028238458, + "loss": 0.5018, + "step": 22380 + }, + { + "epoch": 1.2504679163058356, + "grad_norm": 0.6520093679428101, + "learning_rate": 0.000376344688480502, + "loss": 0.3449, + "step": 22381 + }, + { + "epoch": 1.2505237869095176, + "grad_norm": 0.48096081614494324, + "learning_rate": 0.0003763166741371583, + "loss": 0.4509, + "step": 22382 + }, + { + "epoch": 1.2505796575131995, + "grad_norm": 0.39711013436317444, + "learning_rate": 0.00037628865979381443, + "loss": 0.3509, + "step": 22383 + }, + { + "epoch": 1.2506355281168813, + "grad_norm": 0.3120546340942383, + "learning_rate": 0.0003762606454504707, + "loss": 0.3039, + "step": 22384 + }, + { + "epoch": 1.2506913987205632, + "grad_norm": 0.5022658109664917, + "learning_rate": 0.00037623263110712684, + "loss": 0.3868, + "step": 22385 + }, + { + "epoch": 1.2507472693242452, + "grad_norm": 1.1117961406707764, + "learning_rate": 0.0003762046167637831, + "loss": 0.3448, + "step": 22386 + }, + { + "epoch": 1.250803139927927, + "grad_norm": 0.8109502196311951, + "learning_rate": 0.00037617660242043926, + "loss": 0.4407, + "step": 22387 + }, + { + "epoch": 1.2508590105316089, + "grad_norm": 0.5181779861450195, + "learning_rate": 0.00037614858807709546, + "loss": 0.3912, + "step": 22388 + }, + { + "epoch": 1.2509148811352906, + "grad_norm": 0.4389181435108185, + "learning_rate": 0.00037612057373375167, + "loss": 0.4705, + "step": 22389 + }, + { + "epoch": 1.2509707517389725, + "grad_norm": 0.42970865964889526, + "learning_rate": 0.00037609255939040787, + "loss": 0.4431, + "step": 22390 + }, + { + "epoch": 1.2510266223426545, + "grad_norm": 0.8180570006370544, + "learning_rate": 0.00037606454504706413, + "loss": 0.4026, + "step": 22391 + }, + { + "epoch": 1.2510824929463362, + "grad_norm": 0.7845239043235779, + "learning_rate": 0.0003760365307037203, + "loss": 0.4917, + "step": 22392 + }, + { + "epoch": 1.2511383635500182, + "grad_norm": 0.5171070694923401, + "learning_rate": 0.00037600851636037654, + "loss": 0.3732, + "step": 22393 + }, + { + "epoch": 1.2511942341537, + "grad_norm": 0.6202595829963684, + "learning_rate": 0.0003759805020170327, + "loss": 0.3878, + "step": 22394 + }, + { + "epoch": 1.2512501047573819, + "grad_norm": 0.36031240224838257, + "learning_rate": 0.00037595248767368896, + "loss": 0.2623, + "step": 22395 + }, + { + "epoch": 1.2513059753610638, + "grad_norm": 0.5361998677253723, + "learning_rate": 0.0003759244733303451, + "loss": 0.4793, + "step": 22396 + }, + { + "epoch": 1.2513618459647455, + "grad_norm": 0.4070245027542114, + "learning_rate": 0.00037589645898700137, + "loss": 0.3945, + "step": 22397 + }, + { + "epoch": 1.2514177165684275, + "grad_norm": 0.46556150913238525, + "learning_rate": 0.0003758684446436576, + "loss": 0.5507, + "step": 22398 + }, + { + "epoch": 1.2514735871721094, + "grad_norm": 0.40525344014167786, + "learning_rate": 0.0003758404303003138, + "loss": 0.3813, + "step": 22399 + }, + { + "epoch": 1.2515294577757912, + "grad_norm": 0.4730057120323181, + "learning_rate": 0.00037581241595697, + "loss": 0.5241, + "step": 22400 + }, + { + "epoch": 1.2515853283794731, + "grad_norm": 0.3505377471446991, + "learning_rate": 0.0003757844016136262, + "loss": 0.4155, + "step": 22401 + }, + { + "epoch": 1.251641198983155, + "grad_norm": 2.056258201599121, + "learning_rate": 0.0003757563872702824, + "loss": 0.3433, + "step": 22402 + }, + { + "epoch": 1.2516970695868368, + "grad_norm": 0.4214402437210083, + "learning_rate": 0.0003757283729269386, + "loss": 0.4537, + "step": 22403 + }, + { + "epoch": 1.2517529401905187, + "grad_norm": 0.4588784873485565, + "learning_rate": 0.0003757003585835948, + "loss": 0.4923, + "step": 22404 + }, + { + "epoch": 1.2518088107942007, + "grad_norm": 0.5002533793449402, + "learning_rate": 0.000375672344240251, + "loss": 0.4823, + "step": 22405 + }, + { + "epoch": 1.2518646813978824, + "grad_norm": 0.5099477767944336, + "learning_rate": 0.0003756443298969072, + "loss": 0.4476, + "step": 22406 + }, + { + "epoch": 1.2519205520015644, + "grad_norm": 0.9376460313796997, + "learning_rate": 0.0003756163155535635, + "loss": 0.5323, + "step": 22407 + }, + { + "epoch": 1.2519764226052463, + "grad_norm": 0.7048667073249817, + "learning_rate": 0.00037558830121021963, + "loss": 0.428, + "step": 22408 + }, + { + "epoch": 1.252032293208928, + "grad_norm": 0.6812004446983337, + "learning_rate": 0.00037556028686687584, + "loss": 0.4673, + "step": 22409 + }, + { + "epoch": 1.25208816381261, + "grad_norm": 0.5055246353149414, + "learning_rate": 0.00037553227252353204, + "loss": 0.329, + "step": 22410 + }, + { + "epoch": 1.252144034416292, + "grad_norm": 0.38088008761405945, + "learning_rate": 0.00037550425818018825, + "loss": 0.3957, + "step": 22411 + }, + { + "epoch": 1.2521999050199737, + "grad_norm": 0.4376618564128876, + "learning_rate": 0.00037547624383684445, + "loss": 0.4885, + "step": 22412 + }, + { + "epoch": 1.2522557756236556, + "grad_norm": 0.4015871286392212, + "learning_rate": 0.00037544822949350066, + "loss": 0.4132, + "step": 22413 + }, + { + "epoch": 1.2523116462273376, + "grad_norm": 0.4802173674106598, + "learning_rate": 0.0003754202151501569, + "loss": 0.405, + "step": 22414 + }, + { + "epoch": 1.2523675168310193, + "grad_norm": 1.2092593908309937, + "learning_rate": 0.00037539220080681307, + "loss": 0.3886, + "step": 22415 + }, + { + "epoch": 1.2524233874347013, + "grad_norm": 0.5395183563232422, + "learning_rate": 0.00037536418646346933, + "loss": 0.4433, + "step": 22416 + }, + { + "epoch": 1.2524792580383832, + "grad_norm": 0.505708634853363, + "learning_rate": 0.0003753361721201255, + "loss": 0.4575, + "step": 22417 + }, + { + "epoch": 1.252535128642065, + "grad_norm": 0.3667711615562439, + "learning_rate": 0.00037530815777678174, + "loss": 0.4203, + "step": 22418 + }, + { + "epoch": 1.252590999245747, + "grad_norm": 0.9159713983535767, + "learning_rate": 0.0003752801434334379, + "loss": 0.4444, + "step": 22419 + }, + { + "epoch": 1.2526468698494289, + "grad_norm": 0.9385999441146851, + "learning_rate": 0.00037525212909009416, + "loss": 0.4039, + "step": 22420 + }, + { + "epoch": 1.2527027404531106, + "grad_norm": 0.4741370975971222, + "learning_rate": 0.0003752241147467503, + "loss": 0.3885, + "step": 22421 + }, + { + "epoch": 1.2527586110567925, + "grad_norm": 1.2207626104354858, + "learning_rate": 0.00037519610040340657, + "loss": 0.704, + "step": 22422 + }, + { + "epoch": 1.2528144816604743, + "grad_norm": 1.357728362083435, + "learning_rate": 0.0003751680860600628, + "loss": 0.3819, + "step": 22423 + }, + { + "epoch": 1.2528703522641562, + "grad_norm": 0.7167433500289917, + "learning_rate": 0.000375140071716719, + "loss": 0.438, + "step": 22424 + }, + { + "epoch": 1.2529262228678382, + "grad_norm": 0.5324569940567017, + "learning_rate": 0.0003751120573733752, + "loss": 0.3999, + "step": 22425 + }, + { + "epoch": 1.25298209347152, + "grad_norm": 0.5178864598274231, + "learning_rate": 0.0003750840430300314, + "loss": 0.3129, + "step": 22426 + }, + { + "epoch": 1.2530379640752018, + "grad_norm": 0.683005690574646, + "learning_rate": 0.0003750560286866876, + "loss": 0.5293, + "step": 22427 + }, + { + "epoch": 1.2530938346788836, + "grad_norm": 2.4840962886810303, + "learning_rate": 0.0003750280143433438, + "loss": 0.5285, + "step": 22428 + }, + { + "epoch": 1.2531497052825655, + "grad_norm": 0.5532412528991699, + "learning_rate": 0.000375, + "loss": 0.4501, + "step": 22429 + }, + { + "epoch": 1.2532055758862475, + "grad_norm": 0.6051845550537109, + "learning_rate": 0.0003749719856566562, + "loss": 0.4398, + "step": 22430 + }, + { + "epoch": 1.2532614464899292, + "grad_norm": 0.37130042910575867, + "learning_rate": 0.0003749439713133124, + "loss": 0.4456, + "step": 22431 + }, + { + "epoch": 1.2533173170936112, + "grad_norm": 0.35893866419792175, + "learning_rate": 0.0003749159569699686, + "loss": 0.3545, + "step": 22432 + }, + { + "epoch": 1.2533731876972931, + "grad_norm": 0.4050882160663605, + "learning_rate": 0.00037488794262662483, + "loss": 0.463, + "step": 22433 + }, + { + "epoch": 1.2534290583009748, + "grad_norm": 0.5107850432395935, + "learning_rate": 0.00037485992828328104, + "loss": 0.4045, + "step": 22434 + }, + { + "epoch": 1.2534849289046568, + "grad_norm": 0.4215306043624878, + "learning_rate": 0.00037483191393993724, + "loss": 0.3634, + "step": 22435 + }, + { + "epoch": 1.2535407995083387, + "grad_norm": 0.49630022048950195, + "learning_rate": 0.00037480389959659345, + "loss": 0.4488, + "step": 22436 + }, + { + "epoch": 1.2535966701120205, + "grad_norm": 0.50335693359375, + "learning_rate": 0.0003747758852532497, + "loss": 0.3792, + "step": 22437 + }, + { + "epoch": 1.2536525407157024, + "grad_norm": 0.6083126068115234, + "learning_rate": 0.00037474787090990586, + "loss": 0.4446, + "step": 22438 + }, + { + "epoch": 1.2537084113193844, + "grad_norm": 0.8255182504653931, + "learning_rate": 0.0003747198565665621, + "loss": 0.3534, + "step": 22439 + }, + { + "epoch": 1.253764281923066, + "grad_norm": 0.7408893704414368, + "learning_rate": 0.00037469184222321827, + "loss": 0.4114, + "step": 22440 + }, + { + "epoch": 1.253820152526748, + "grad_norm": 0.46986210346221924, + "learning_rate": 0.00037466382787987453, + "loss": 0.3999, + "step": 22441 + }, + { + "epoch": 1.25387602313043, + "grad_norm": 0.4491029679775238, + "learning_rate": 0.0003746358135365307, + "loss": 0.4373, + "step": 22442 + }, + { + "epoch": 1.2539318937341117, + "grad_norm": 0.7146137952804565, + "learning_rate": 0.00037460779919318694, + "loss": 0.4591, + "step": 22443 + }, + { + "epoch": 1.2539877643377937, + "grad_norm": 0.5770229697227478, + "learning_rate": 0.0003745797848498431, + "loss": 0.3891, + "step": 22444 + }, + { + "epoch": 1.2540436349414756, + "grad_norm": 3.0682430267333984, + "learning_rate": 0.00037455177050649936, + "loss": 0.4121, + "step": 22445 + }, + { + "epoch": 1.2540995055451574, + "grad_norm": 0.4323097765445709, + "learning_rate": 0.00037452375616315556, + "loss": 0.366, + "step": 22446 + }, + { + "epoch": 1.2541553761488393, + "grad_norm": 0.5697695016860962, + "learning_rate": 0.00037449574181981177, + "loss": 0.5215, + "step": 22447 + }, + { + "epoch": 1.2542112467525213, + "grad_norm": 0.3198217451572418, + "learning_rate": 0.00037446772747646797, + "loss": 0.3642, + "step": 22448 + }, + { + "epoch": 1.254267117356203, + "grad_norm": 0.4991450905799866, + "learning_rate": 0.0003744397131331242, + "loss": 0.4688, + "step": 22449 + }, + { + "epoch": 1.254322987959885, + "grad_norm": 1.9839564561843872, + "learning_rate": 0.0003744116987897804, + "loss": 0.4157, + "step": 22450 + }, + { + "epoch": 1.254378858563567, + "grad_norm": 0.45878440141677856, + "learning_rate": 0.00037438368444643654, + "loss": 0.4306, + "step": 22451 + }, + { + "epoch": 1.2544347291672486, + "grad_norm": 0.6043434739112854, + "learning_rate": 0.0003743556701030928, + "loss": 0.4314, + "step": 22452 + }, + { + "epoch": 1.2544905997709306, + "grad_norm": 0.4261232912540436, + "learning_rate": 0.000374327655759749, + "loss": 0.4642, + "step": 22453 + }, + { + "epoch": 1.2545464703746125, + "grad_norm": 0.334146112203598, + "learning_rate": 0.0003742996414164052, + "loss": 0.3482, + "step": 22454 + }, + { + "epoch": 1.2546023409782943, + "grad_norm": 0.3852083683013916, + "learning_rate": 0.0003742716270730614, + "loss": 0.5017, + "step": 22455 + }, + { + "epoch": 1.2546582115819762, + "grad_norm": 0.4703032374382019, + "learning_rate": 0.0003742436127297176, + "loss": 0.5226, + "step": 22456 + }, + { + "epoch": 1.254714082185658, + "grad_norm": 0.43411970138549805, + "learning_rate": 0.0003742155983863738, + "loss": 0.3752, + "step": 22457 + }, + { + "epoch": 1.25476995278934, + "grad_norm": 0.49465787410736084, + "learning_rate": 0.00037418758404303003, + "loss": 0.3589, + "step": 22458 + }, + { + "epoch": 1.2548258233930218, + "grad_norm": 0.4484471082687378, + "learning_rate": 0.00037415956969968624, + "loss": 0.387, + "step": 22459 + }, + { + "epoch": 1.2548816939967036, + "grad_norm": 0.4299778938293457, + "learning_rate": 0.00037413155535634244, + "loss": 0.4264, + "step": 22460 + }, + { + "epoch": 1.2549375646003855, + "grad_norm": 0.5885958075523376, + "learning_rate": 0.00037410354101299865, + "loss": 0.4863, + "step": 22461 + }, + { + "epoch": 1.2549934352040673, + "grad_norm": 0.5245383381843567, + "learning_rate": 0.0003740755266696549, + "loss": 0.5125, + "step": 22462 + }, + { + "epoch": 1.2550493058077492, + "grad_norm": 0.5639734268188477, + "learning_rate": 0.00037404751232631106, + "loss": 0.4253, + "step": 22463 + }, + { + "epoch": 1.2551051764114312, + "grad_norm": 0.635902464389801, + "learning_rate": 0.0003740194979829673, + "loss": 0.4374, + "step": 22464 + }, + { + "epoch": 1.2551610470151129, + "grad_norm": 0.44458943605422974, + "learning_rate": 0.00037399148363962347, + "loss": 0.3991, + "step": 22465 + }, + { + "epoch": 1.2552169176187948, + "grad_norm": 0.3949022889137268, + "learning_rate": 0.00037396346929627973, + "loss": 0.3522, + "step": 22466 + }, + { + "epoch": 1.2552727882224768, + "grad_norm": 0.5285905003547668, + "learning_rate": 0.0003739354549529359, + "loss": 0.4785, + "step": 22467 + }, + { + "epoch": 1.2553286588261585, + "grad_norm": 0.4128982126712799, + "learning_rate": 0.00037390744060959214, + "loss": 0.5815, + "step": 22468 + }, + { + "epoch": 1.2553845294298405, + "grad_norm": 0.35645443201065063, + "learning_rate": 0.00037387942626624835, + "loss": 0.3892, + "step": 22469 + }, + { + "epoch": 1.2554404000335224, + "grad_norm": 0.5057388544082642, + "learning_rate": 0.00037385141192290455, + "loss": 0.342, + "step": 22470 + }, + { + "epoch": 1.2554962706372041, + "grad_norm": 0.5581644773483276, + "learning_rate": 0.00037382339757956076, + "loss": 0.4823, + "step": 22471 + }, + { + "epoch": 1.255552141240886, + "grad_norm": 0.7794705629348755, + "learning_rate": 0.0003737953832362169, + "loss": 0.392, + "step": 22472 + }, + { + "epoch": 1.255608011844568, + "grad_norm": 0.33841952681541443, + "learning_rate": 0.00037376736889287317, + "loss": 0.3456, + "step": 22473 + }, + { + "epoch": 1.2556638824482498, + "grad_norm": 0.35464364290237427, + "learning_rate": 0.0003737393545495293, + "loss": 0.3965, + "step": 22474 + }, + { + "epoch": 1.2557197530519317, + "grad_norm": 0.6756699681282043, + "learning_rate": 0.0003737113402061856, + "loss": 0.4087, + "step": 22475 + }, + { + "epoch": 1.2557756236556137, + "grad_norm": 0.404360830783844, + "learning_rate": 0.00037368332586284174, + "loss": 0.4477, + "step": 22476 + }, + { + "epoch": 1.2558314942592954, + "grad_norm": 0.447662889957428, + "learning_rate": 0.000373655311519498, + "loss": 0.4896, + "step": 22477 + }, + { + "epoch": 1.2558873648629774, + "grad_norm": 0.40657907724380493, + "learning_rate": 0.0003736272971761542, + "loss": 0.3694, + "step": 22478 + }, + { + "epoch": 1.2559432354666593, + "grad_norm": 0.456844687461853, + "learning_rate": 0.0003735992828328104, + "loss": 0.3725, + "step": 22479 + }, + { + "epoch": 1.255999106070341, + "grad_norm": 0.583537220954895, + "learning_rate": 0.0003735712684894666, + "loss": 0.6197, + "step": 22480 + }, + { + "epoch": 1.256054976674023, + "grad_norm": 0.4022231101989746, + "learning_rate": 0.0003735432541461228, + "loss": 0.4612, + "step": 22481 + }, + { + "epoch": 1.256110847277705, + "grad_norm": 0.5165391564369202, + "learning_rate": 0.000373515239802779, + "loss": 0.4612, + "step": 22482 + }, + { + "epoch": 1.2561667178813867, + "grad_norm": 0.5637795329093933, + "learning_rate": 0.00037348722545943523, + "loss": 0.4458, + "step": 22483 + }, + { + "epoch": 1.2562225884850686, + "grad_norm": 0.3809378445148468, + "learning_rate": 0.00037345921111609144, + "loss": 0.4223, + "step": 22484 + }, + { + "epoch": 1.2562784590887506, + "grad_norm": 0.5353775024414062, + "learning_rate": 0.0003734311967727477, + "loss": 0.4472, + "step": 22485 + }, + { + "epoch": 1.2563343296924323, + "grad_norm": 0.7497721314430237, + "learning_rate": 0.00037340318242940385, + "loss": 0.4554, + "step": 22486 + }, + { + "epoch": 1.2563902002961143, + "grad_norm": 0.5310592651367188, + "learning_rate": 0.0003733751680860601, + "loss": 0.3622, + "step": 22487 + }, + { + "epoch": 1.2564460708997962, + "grad_norm": 0.45614078640937805, + "learning_rate": 0.00037334715374271626, + "loss": 0.5494, + "step": 22488 + }, + { + "epoch": 1.256501941503478, + "grad_norm": 0.3777497708797455, + "learning_rate": 0.0003733191393993725, + "loss": 0.4129, + "step": 22489 + }, + { + "epoch": 1.2565578121071599, + "grad_norm": 0.5255526900291443, + "learning_rate": 0.00037329112505602867, + "loss": 0.4416, + "step": 22490 + }, + { + "epoch": 1.2566136827108416, + "grad_norm": 0.4099194407463074, + "learning_rate": 0.00037326311071268493, + "loss": 0.4554, + "step": 22491 + }, + { + "epoch": 1.2566695533145236, + "grad_norm": 0.5259906053543091, + "learning_rate": 0.0003732350963693411, + "loss": 0.426, + "step": 22492 + }, + { + "epoch": 1.2567254239182053, + "grad_norm": 0.6627949476242065, + "learning_rate": 0.0003732070820259973, + "loss": 0.485, + "step": 22493 + }, + { + "epoch": 1.2567812945218872, + "grad_norm": 3.9346063137054443, + "learning_rate": 0.00037317906768265355, + "loss": 0.7424, + "step": 22494 + }, + { + "epoch": 1.2568371651255692, + "grad_norm": 0.4953204393386841, + "learning_rate": 0.0003731510533393097, + "loss": 0.4393, + "step": 22495 + }, + { + "epoch": 1.256893035729251, + "grad_norm": 0.8178825974464417, + "learning_rate": 0.00037312303899596596, + "loss": 0.5413, + "step": 22496 + }, + { + "epoch": 1.2569489063329329, + "grad_norm": 1.538346290588379, + "learning_rate": 0.0003730950246526221, + "loss": 0.4586, + "step": 22497 + }, + { + "epoch": 1.2570047769366148, + "grad_norm": 0.5124504566192627, + "learning_rate": 0.00037306701030927837, + "loss": 0.4884, + "step": 22498 + }, + { + "epoch": 1.2570606475402966, + "grad_norm": 0.41794002056121826, + "learning_rate": 0.0003730389959659345, + "loss": 0.5174, + "step": 22499 + }, + { + "epoch": 1.2571165181439785, + "grad_norm": 0.5320823192596436, + "learning_rate": 0.0003730109816225908, + "loss": 0.4479, + "step": 22500 + }, + { + "epoch": 1.2571165181439785, + "eval_cer": 0.08753668728792292, + "eval_loss": 0.3243005871772766, + "eval_runtime": 56.3787, + "eval_samples_per_second": 80.491, + "eval_steps_per_second": 5.037, + "eval_wer": 0.3519454763833412, + "step": 22500 + }, + { + "epoch": 1.2571723887476605, + "grad_norm": 0.5802021622657776, + "learning_rate": 0.000372982967279247, + "loss": 0.4077, + "step": 22501 + }, + { + "epoch": 1.2572282593513422, + "grad_norm": 0.36161428689956665, + "learning_rate": 0.0003729549529359032, + "loss": 0.3824, + "step": 22502 + }, + { + "epoch": 1.2572841299550241, + "grad_norm": 1.4083143472671509, + "learning_rate": 0.0003729269385925594, + "loss": 0.5107, + "step": 22503 + }, + { + "epoch": 1.257340000558706, + "grad_norm": 0.401215136051178, + "learning_rate": 0.0003728989242492156, + "loss": 0.4086, + "step": 22504 + }, + { + "epoch": 1.2573958711623878, + "grad_norm": 0.425324410200119, + "learning_rate": 0.0003728709099058718, + "loss": 0.4106, + "step": 22505 + }, + { + "epoch": 1.2574517417660698, + "grad_norm": 0.5277112126350403, + "learning_rate": 0.000372842895562528, + "loss": 0.4375, + "step": 22506 + }, + { + "epoch": 1.2575076123697517, + "grad_norm": 1.5795817375183105, + "learning_rate": 0.0003728148812191842, + "loss": 0.6641, + "step": 22507 + }, + { + "epoch": 1.2575634829734335, + "grad_norm": 0.9531577825546265, + "learning_rate": 0.0003727868668758405, + "loss": 0.4553, + "step": 22508 + }, + { + "epoch": 1.2576193535771154, + "grad_norm": 0.574181079864502, + "learning_rate": 0.00037275885253249664, + "loss": 0.3925, + "step": 22509 + }, + { + "epoch": 1.2576752241807974, + "grad_norm": 0.44868093729019165, + "learning_rate": 0.0003727308381891529, + "loss": 0.6424, + "step": 22510 + }, + { + "epoch": 1.257731094784479, + "grad_norm": 0.4035612642765045, + "learning_rate": 0.00037270282384580905, + "loss": 0.4171, + "step": 22511 + }, + { + "epoch": 1.257786965388161, + "grad_norm": 0.3639061152935028, + "learning_rate": 0.0003726748095024653, + "loss": 0.341, + "step": 22512 + }, + { + "epoch": 1.257842835991843, + "grad_norm": 0.5307701230049133, + "learning_rate": 0.00037264679515912146, + "loss": 0.401, + "step": 22513 + }, + { + "epoch": 1.2578987065955247, + "grad_norm": 0.3727248013019562, + "learning_rate": 0.00037261878081577766, + "loss": 0.4974, + "step": 22514 + }, + { + "epoch": 1.2579545771992067, + "grad_norm": 2.7209088802337646, + "learning_rate": 0.00037259076647243387, + "loss": 0.3311, + "step": 22515 + }, + { + "epoch": 1.2580104478028886, + "grad_norm": 0.8306828141212463, + "learning_rate": 0.0003725627521290901, + "loss": 0.4709, + "step": 22516 + }, + { + "epoch": 1.2580663184065703, + "grad_norm": 0.5786606669425964, + "learning_rate": 0.00037253473778574634, + "loss": 0.56, + "step": 22517 + }, + { + "epoch": 1.2581221890102523, + "grad_norm": 0.7911596894264221, + "learning_rate": 0.0003725067234424025, + "loss": 0.5308, + "step": 22518 + }, + { + "epoch": 1.2581780596139343, + "grad_norm": 0.7039480209350586, + "learning_rate": 0.00037247870909905875, + "loss": 0.4342, + "step": 22519 + }, + { + "epoch": 1.258233930217616, + "grad_norm": 0.49513086676597595, + "learning_rate": 0.0003724506947557149, + "loss": 0.3735, + "step": 22520 + }, + { + "epoch": 1.258289800821298, + "grad_norm": 3.4181017875671387, + "learning_rate": 0.00037242268041237116, + "loss": 0.53, + "step": 22521 + }, + { + "epoch": 1.2583456714249799, + "grad_norm": 0.6502416133880615, + "learning_rate": 0.0003723946660690273, + "loss": 0.4789, + "step": 22522 + }, + { + "epoch": 1.2584015420286616, + "grad_norm": 0.47867655754089355, + "learning_rate": 0.00037236665172568357, + "loss": 0.3094, + "step": 22523 + }, + { + "epoch": 1.2584574126323436, + "grad_norm": 1.037257432937622, + "learning_rate": 0.0003723386373823398, + "loss": 0.4119, + "step": 22524 + }, + { + "epoch": 1.2585132832360253, + "grad_norm": 0.5666090250015259, + "learning_rate": 0.000372310623038996, + "loss": 0.3936, + "step": 22525 + }, + { + "epoch": 1.2585691538397072, + "grad_norm": 0.4216514229774475, + "learning_rate": 0.0003722826086956522, + "loss": 0.4117, + "step": 22526 + }, + { + "epoch": 1.258625024443389, + "grad_norm": 0.7706602811813354, + "learning_rate": 0.0003722545943523084, + "loss": 0.4718, + "step": 22527 + }, + { + "epoch": 1.258680895047071, + "grad_norm": 1.6483956575393677, + "learning_rate": 0.0003722265800089646, + "loss": 0.4988, + "step": 22528 + }, + { + "epoch": 1.2587367656507529, + "grad_norm": 0.3889979124069214, + "learning_rate": 0.0003721985656656208, + "loss": 0.4405, + "step": 22529 + }, + { + "epoch": 1.2587926362544346, + "grad_norm": 1.1102033853530884, + "learning_rate": 0.000372170551322277, + "loss": 0.4504, + "step": 22530 + }, + { + "epoch": 1.2588485068581166, + "grad_norm": 0.3588782548904419, + "learning_rate": 0.0003721425369789332, + "loss": 0.3284, + "step": 22531 + }, + { + "epoch": 1.2589043774617985, + "grad_norm": 0.4394785761833191, + "learning_rate": 0.0003721145226355894, + "loss": 0.353, + "step": 22532 + }, + { + "epoch": 1.2589602480654802, + "grad_norm": 0.9448420405387878, + "learning_rate": 0.0003720865082922457, + "loss": 0.5137, + "step": 22533 + }, + { + "epoch": 1.2590161186691622, + "grad_norm": 1.3025652170181274, + "learning_rate": 0.00037205849394890184, + "loss": 0.5551, + "step": 22534 + }, + { + "epoch": 1.2590719892728441, + "grad_norm": 0.6043887734413147, + "learning_rate": 0.0003720304796055581, + "loss": 0.3934, + "step": 22535 + }, + { + "epoch": 1.2591278598765259, + "grad_norm": 0.47628819942474365, + "learning_rate": 0.00037200246526221425, + "loss": 0.4272, + "step": 22536 + }, + { + "epoch": 1.2591837304802078, + "grad_norm": 0.6069056987762451, + "learning_rate": 0.00037197445091887045, + "loss": 0.504, + "step": 22537 + }, + { + "epoch": 1.2592396010838898, + "grad_norm": 0.3447617292404175, + "learning_rate": 0.00037194643657552666, + "loss": 0.5162, + "step": 22538 + }, + { + "epoch": 1.2592954716875715, + "grad_norm": 0.35751670598983765, + "learning_rate": 0.00037191842223218286, + "loss": 0.3681, + "step": 22539 + }, + { + "epoch": 1.2593513422912535, + "grad_norm": 0.5541776418685913, + "learning_rate": 0.0003718904078888391, + "loss": 0.386, + "step": 22540 + }, + { + "epoch": 1.2594072128949354, + "grad_norm": 0.47219473123550415, + "learning_rate": 0.0003718623935454953, + "loss": 0.4728, + "step": 22541 + }, + { + "epoch": 1.2594630834986171, + "grad_norm": 0.4149962067604065, + "learning_rate": 0.00037183437920215154, + "loss": 0.3382, + "step": 22542 + }, + { + "epoch": 1.259518954102299, + "grad_norm": 0.3887511193752289, + "learning_rate": 0.0003718063648588077, + "loss": 0.3612, + "step": 22543 + }, + { + "epoch": 1.259574824705981, + "grad_norm": 0.5005051493644714, + "learning_rate": 0.00037177835051546395, + "loss": 0.5067, + "step": 22544 + }, + { + "epoch": 1.2596306953096628, + "grad_norm": 0.5066700577735901, + "learning_rate": 0.0003717503361721201, + "loss": 0.5603, + "step": 22545 + }, + { + "epoch": 1.2596865659133447, + "grad_norm": 1.0324211120605469, + "learning_rate": 0.00037172232182877636, + "loss": 0.4541, + "step": 22546 + }, + { + "epoch": 1.2597424365170267, + "grad_norm": 0.5915161371231079, + "learning_rate": 0.0003716943074854325, + "loss": 0.5249, + "step": 22547 + }, + { + "epoch": 1.2597983071207084, + "grad_norm": 0.9875571727752686, + "learning_rate": 0.00037166629314208877, + "loss": 0.3568, + "step": 22548 + }, + { + "epoch": 1.2598541777243903, + "grad_norm": 0.330582857131958, + "learning_rate": 0.000371638278798745, + "loss": 0.4356, + "step": 22549 + }, + { + "epoch": 1.2599100483280723, + "grad_norm": 0.46805813908576965, + "learning_rate": 0.0003716102644554012, + "loss": 0.4695, + "step": 22550 + }, + { + "epoch": 1.259965918931754, + "grad_norm": 0.6079339385032654, + "learning_rate": 0.0003715822501120574, + "loss": 0.3928, + "step": 22551 + }, + { + "epoch": 1.260021789535436, + "grad_norm": 0.5090405941009521, + "learning_rate": 0.0003715542357687136, + "loss": 0.3895, + "step": 22552 + }, + { + "epoch": 1.260077660139118, + "grad_norm": 0.37442702054977417, + "learning_rate": 0.0003715262214253698, + "loss": 0.3845, + "step": 22553 + }, + { + "epoch": 1.2601335307427997, + "grad_norm": 0.5094758868217468, + "learning_rate": 0.000371498207082026, + "loss": 0.4046, + "step": 22554 + }, + { + "epoch": 1.2601894013464816, + "grad_norm": 0.7414910793304443, + "learning_rate": 0.0003714701927386822, + "loss": 0.4096, + "step": 22555 + }, + { + "epoch": 1.2602452719501636, + "grad_norm": 3.904399871826172, + "learning_rate": 0.00037144217839533847, + "loss": 0.4021, + "step": 22556 + }, + { + "epoch": 1.2603011425538453, + "grad_norm": 0.5285754799842834, + "learning_rate": 0.0003714141640519946, + "loss": 0.3786, + "step": 22557 + }, + { + "epoch": 1.2603570131575272, + "grad_norm": 0.3128261864185333, + "learning_rate": 0.00037138614970865083, + "loss": 0.3082, + "step": 22558 + }, + { + "epoch": 1.260412883761209, + "grad_norm": 0.4690313935279846, + "learning_rate": 0.00037135813536530703, + "loss": 0.4628, + "step": 22559 + }, + { + "epoch": 1.260468754364891, + "grad_norm": 0.47490590810775757, + "learning_rate": 0.00037133012102196324, + "loss": 0.4046, + "step": 22560 + }, + { + "epoch": 1.2605246249685726, + "grad_norm": 0.5724405646324158, + "learning_rate": 0.00037130210667861945, + "loss": 0.4464, + "step": 22561 + }, + { + "epoch": 1.2605804955722546, + "grad_norm": 0.5209167003631592, + "learning_rate": 0.00037127409233527565, + "loss": 0.3952, + "step": 22562 + }, + { + "epoch": 1.2606363661759366, + "grad_norm": 0.35911476612091064, + "learning_rate": 0.0003712460779919319, + "loss": 0.33, + "step": 22563 + }, + { + "epoch": 1.2606922367796183, + "grad_norm": 0.5583000779151917, + "learning_rate": 0.00037121806364858806, + "loss": 0.3861, + "step": 22564 + }, + { + "epoch": 1.2607481073833002, + "grad_norm": 0.4037957489490509, + "learning_rate": 0.0003711900493052443, + "loss": 0.4075, + "step": 22565 + }, + { + "epoch": 1.2608039779869822, + "grad_norm": 0.33433833718299866, + "learning_rate": 0.0003711620349619005, + "loss": 0.4499, + "step": 22566 + }, + { + "epoch": 1.260859848590664, + "grad_norm": 0.44187840819358826, + "learning_rate": 0.00037113402061855674, + "loss": 0.3992, + "step": 22567 + }, + { + "epoch": 1.2609157191943459, + "grad_norm": 0.4182825982570648, + "learning_rate": 0.0003711060062752129, + "loss": 0.3951, + "step": 22568 + }, + { + "epoch": 1.2609715897980278, + "grad_norm": 0.47042086720466614, + "learning_rate": 0.00037107799193186915, + "loss": 0.3816, + "step": 22569 + }, + { + "epoch": 1.2610274604017095, + "grad_norm": 0.852584183216095, + "learning_rate": 0.0003710499775885253, + "loss": 0.6296, + "step": 22570 + }, + { + "epoch": 1.2610833310053915, + "grad_norm": 0.5029460191726685, + "learning_rate": 0.00037102196324518156, + "loss": 0.4192, + "step": 22571 + }, + { + "epoch": 1.2611392016090734, + "grad_norm": 0.3609471023082733, + "learning_rate": 0.00037099394890183776, + "loss": 0.4548, + "step": 22572 + }, + { + "epoch": 1.2611950722127552, + "grad_norm": 0.8067110180854797, + "learning_rate": 0.00037096593455849397, + "loss": 0.4043, + "step": 22573 + }, + { + "epoch": 1.2612509428164371, + "grad_norm": 0.35192257165908813, + "learning_rate": 0.0003709379202151502, + "loss": 0.3034, + "step": 22574 + }, + { + "epoch": 1.261306813420119, + "grad_norm": 0.48422324657440186, + "learning_rate": 0.0003709099058718064, + "loss": 0.4062, + "step": 22575 + }, + { + "epoch": 1.2613626840238008, + "grad_norm": 0.822407603263855, + "learning_rate": 0.0003708818915284626, + "loss": 0.4989, + "step": 22576 + }, + { + "epoch": 1.2614185546274828, + "grad_norm": 0.6904353499412537, + "learning_rate": 0.0003708538771851188, + "loss": 0.5439, + "step": 22577 + }, + { + "epoch": 1.2614744252311647, + "grad_norm": 0.37183821201324463, + "learning_rate": 0.000370825862841775, + "loss": 0.3578, + "step": 22578 + }, + { + "epoch": 1.2615302958348464, + "grad_norm": 0.6183751225471497, + "learning_rate": 0.0003707978484984312, + "loss": 0.5234, + "step": 22579 + }, + { + "epoch": 1.2615861664385284, + "grad_norm": 0.2638550400733948, + "learning_rate": 0.0003707698341550874, + "loss": 0.2639, + "step": 22580 + }, + { + "epoch": 1.2616420370422103, + "grad_norm": 0.5109890699386597, + "learning_rate": 0.0003707418198117436, + "loss": 0.4569, + "step": 22581 + }, + { + "epoch": 1.261697907645892, + "grad_norm": 1.687898874282837, + "learning_rate": 0.0003707138054683998, + "loss": 0.4297, + "step": 22582 + }, + { + "epoch": 1.261753778249574, + "grad_norm": 0.43842563033103943, + "learning_rate": 0.00037068579112505603, + "loss": 0.4014, + "step": 22583 + }, + { + "epoch": 1.261809648853256, + "grad_norm": 2.918959140777588, + "learning_rate": 0.00037065777678171223, + "loss": 0.3879, + "step": 22584 + }, + { + "epoch": 1.2618655194569377, + "grad_norm": 1.3529930114746094, + "learning_rate": 0.00037062976243836844, + "loss": 0.4481, + "step": 22585 + }, + { + "epoch": 1.2619213900606197, + "grad_norm": 0.5728101134300232, + "learning_rate": 0.00037060174809502465, + "loss": 0.4691, + "step": 22586 + }, + { + "epoch": 1.2619772606643016, + "grad_norm": 0.45233115553855896, + "learning_rate": 0.00037057373375168085, + "loss": 0.4401, + "step": 22587 + }, + { + "epoch": 1.2620331312679833, + "grad_norm": 0.3886379301548004, + "learning_rate": 0.0003705457194083371, + "loss": 0.3827, + "step": 22588 + }, + { + "epoch": 1.2620890018716653, + "grad_norm": 0.34922873973846436, + "learning_rate": 0.00037051770506499326, + "loss": 0.3852, + "step": 22589 + }, + { + "epoch": 1.2621448724753472, + "grad_norm": 0.48267748951911926, + "learning_rate": 0.0003704896907216495, + "loss": 0.5474, + "step": 22590 + }, + { + "epoch": 1.262200743079029, + "grad_norm": 1.150854468345642, + "learning_rate": 0.0003704616763783057, + "loss": 0.4453, + "step": 22591 + }, + { + "epoch": 1.262256613682711, + "grad_norm": 0.5684496760368347, + "learning_rate": 0.00037043366203496194, + "loss": 0.3935, + "step": 22592 + }, + { + "epoch": 1.2623124842863926, + "grad_norm": 0.390376478433609, + "learning_rate": 0.0003704056476916181, + "loss": 0.3831, + "step": 22593 + }, + { + "epoch": 1.2623683548900746, + "grad_norm": 0.5266631841659546, + "learning_rate": 0.00037037763334827435, + "loss": 0.3817, + "step": 22594 + }, + { + "epoch": 1.2624242254937563, + "grad_norm": 0.4812847971916199, + "learning_rate": 0.00037034961900493055, + "loss": 0.4478, + "step": 22595 + }, + { + "epoch": 1.2624800960974383, + "grad_norm": 3.396860361099243, + "learning_rate": 0.00037032160466158676, + "loss": 0.3901, + "step": 22596 + }, + { + "epoch": 1.2625359667011202, + "grad_norm": 0.5402068495750427, + "learning_rate": 0.00037029359031824296, + "loss": 0.3527, + "step": 22597 + }, + { + "epoch": 1.262591837304802, + "grad_norm": 0.3931836783885956, + "learning_rate": 0.00037026557597489917, + "loss": 0.4337, + "step": 22598 + }, + { + "epoch": 1.262647707908484, + "grad_norm": 0.6832761764526367, + "learning_rate": 0.0003702375616315554, + "loss": 0.4039, + "step": 22599 + }, + { + "epoch": 1.2627035785121659, + "grad_norm": 0.4289974272251129, + "learning_rate": 0.00037020954728821153, + "loss": 0.3454, + "step": 22600 + }, + { + "epoch": 1.2627594491158476, + "grad_norm": 0.7890150547027588, + "learning_rate": 0.0003701815329448678, + "loss": 0.4186, + "step": 22601 + }, + { + "epoch": 1.2628153197195295, + "grad_norm": 0.3604927659034729, + "learning_rate": 0.00037015351860152394, + "loss": 0.4279, + "step": 22602 + }, + { + "epoch": 1.2628711903232115, + "grad_norm": 0.3957960903644562, + "learning_rate": 0.0003701255042581802, + "loss": 0.4567, + "step": 22603 + }, + { + "epoch": 1.2629270609268932, + "grad_norm": 0.35362693667411804, + "learning_rate": 0.0003700974899148364, + "loss": 0.3428, + "step": 22604 + }, + { + "epoch": 1.2629829315305752, + "grad_norm": 0.42199522256851196, + "learning_rate": 0.0003700694755714926, + "loss": 0.4348, + "step": 22605 + }, + { + "epoch": 1.2630388021342571, + "grad_norm": 0.37590116262435913, + "learning_rate": 0.0003700414612281488, + "loss": 0.5218, + "step": 22606 + }, + { + "epoch": 1.2630946727379389, + "grad_norm": 0.46405041217803955, + "learning_rate": 0.000370013446884805, + "loss": 0.4203, + "step": 22607 + }, + { + "epoch": 1.2631505433416208, + "grad_norm": 0.38377073407173157, + "learning_rate": 0.00036998543254146123, + "loss": 0.3721, + "step": 22608 + }, + { + "epoch": 1.2632064139453028, + "grad_norm": 0.7001894116401672, + "learning_rate": 0.00036995741819811743, + "loss": 0.5083, + "step": 22609 + }, + { + "epoch": 1.2632622845489845, + "grad_norm": 0.7390238046646118, + "learning_rate": 0.00036992940385477364, + "loss": 0.6112, + "step": 22610 + }, + { + "epoch": 1.2633181551526664, + "grad_norm": 0.48358577489852905, + "learning_rate": 0.0003699013895114299, + "loss": 0.4413, + "step": 22611 + }, + { + "epoch": 1.2633740257563484, + "grad_norm": 0.6571111679077148, + "learning_rate": 0.00036987337516808605, + "loss": 0.3757, + "step": 22612 + }, + { + "epoch": 1.2634298963600301, + "grad_norm": 0.44452258944511414, + "learning_rate": 0.0003698453608247423, + "loss": 0.5647, + "step": 22613 + }, + { + "epoch": 1.263485766963712, + "grad_norm": 0.4447081685066223, + "learning_rate": 0.00036981734648139846, + "loss": 0.2796, + "step": 22614 + }, + { + "epoch": 1.263541637567394, + "grad_norm": 0.9426806569099426, + "learning_rate": 0.0003697893321380547, + "loss": 0.4162, + "step": 22615 + }, + { + "epoch": 1.2635975081710757, + "grad_norm": 0.39632725715637207, + "learning_rate": 0.0003697613177947109, + "loss": 0.4356, + "step": 22616 + }, + { + "epoch": 1.2636533787747577, + "grad_norm": 0.70098876953125, + "learning_rate": 0.00036973330345136713, + "loss": 0.3835, + "step": 22617 + }, + { + "epoch": 1.2637092493784396, + "grad_norm": 0.3655398190021515, + "learning_rate": 0.0003697052891080233, + "loss": 0.411, + "step": 22618 + }, + { + "epoch": 1.2637651199821214, + "grad_norm": 0.564582884311676, + "learning_rate": 0.00036967727476467955, + "loss": 0.5234, + "step": 22619 + }, + { + "epoch": 1.2638209905858033, + "grad_norm": 0.3777754604816437, + "learning_rate": 0.00036964926042133575, + "loss": 0.4685, + "step": 22620 + }, + { + "epoch": 1.2638768611894853, + "grad_norm": 0.38292911648750305, + "learning_rate": 0.0003696212460779919, + "loss": 0.3408, + "step": 22621 + }, + { + "epoch": 1.263932731793167, + "grad_norm": 0.6350862979888916, + "learning_rate": 0.00036959323173464816, + "loss": 0.3932, + "step": 22622 + }, + { + "epoch": 1.263988602396849, + "grad_norm": 0.39728906750679016, + "learning_rate": 0.0003695652173913043, + "loss": 0.4909, + "step": 22623 + }, + { + "epoch": 1.2640444730005307, + "grad_norm": 0.41136133670806885, + "learning_rate": 0.0003695372030479606, + "loss": 0.3736, + "step": 22624 + }, + { + "epoch": 1.2641003436042126, + "grad_norm": 0.42424601316452026, + "learning_rate": 0.0003695091887046167, + "loss": 0.3703, + "step": 22625 + }, + { + "epoch": 1.2641562142078946, + "grad_norm": 0.47298935055732727, + "learning_rate": 0.000369481174361273, + "loss": 0.3889, + "step": 22626 + }, + { + "epoch": 1.2642120848115763, + "grad_norm": 0.3226607143878937, + "learning_rate": 0.0003694531600179292, + "loss": 0.4244, + "step": 22627 + }, + { + "epoch": 1.2642679554152583, + "grad_norm": 0.8527957201004028, + "learning_rate": 0.0003694251456745854, + "loss": 0.3649, + "step": 22628 + }, + { + "epoch": 1.26432382601894, + "grad_norm": 0.4560803174972534, + "learning_rate": 0.0003693971313312416, + "loss": 0.3802, + "step": 22629 + }, + { + "epoch": 1.264379696622622, + "grad_norm": 0.8546123504638672, + "learning_rate": 0.0003693691169878978, + "loss": 0.5547, + "step": 22630 + }, + { + "epoch": 1.264435567226304, + "grad_norm": 0.40657252073287964, + "learning_rate": 0.000369341102644554, + "loss": 0.4802, + "step": 22631 + }, + { + "epoch": 1.2644914378299856, + "grad_norm": 8.057674407958984, + "learning_rate": 0.0003693130883012102, + "loss": 0.5066, + "step": 22632 + }, + { + "epoch": 1.2645473084336676, + "grad_norm": 0.35192155838012695, + "learning_rate": 0.00036928507395786643, + "loss": 0.3925, + "step": 22633 + }, + { + "epoch": 1.2646031790373495, + "grad_norm": 0.4613438546657562, + "learning_rate": 0.0003692570596145227, + "loss": 0.3444, + "step": 22634 + }, + { + "epoch": 1.2646590496410313, + "grad_norm": 1.193873405456543, + "learning_rate": 0.00036922904527117884, + "loss": 0.4132, + "step": 22635 + }, + { + "epoch": 1.2647149202447132, + "grad_norm": 0.4901255667209625, + "learning_rate": 0.0003692010309278351, + "loss": 0.4278, + "step": 22636 + }, + { + "epoch": 1.2647707908483952, + "grad_norm": 0.4096038043498993, + "learning_rate": 0.00036917301658449125, + "loss": 0.4548, + "step": 22637 + }, + { + "epoch": 1.264826661452077, + "grad_norm": 0.4461634159088135, + "learning_rate": 0.0003691450022411475, + "loss": 0.4621, + "step": 22638 + }, + { + "epoch": 1.2648825320557588, + "grad_norm": 0.47403502464294434, + "learning_rate": 0.00036911698789780366, + "loss": 0.3722, + "step": 22639 + }, + { + "epoch": 1.2649384026594408, + "grad_norm": 0.473911315202713, + "learning_rate": 0.0003690889735544599, + "loss": 0.5205, + "step": 22640 + }, + { + "epoch": 1.2649942732631225, + "grad_norm": 0.4359968900680542, + "learning_rate": 0.0003690609592111161, + "loss": 0.4878, + "step": 22641 + }, + { + "epoch": 1.2650501438668045, + "grad_norm": 0.3755466639995575, + "learning_rate": 0.0003690329448677723, + "loss": 0.3937, + "step": 22642 + }, + { + "epoch": 1.2651060144704864, + "grad_norm": 0.4002552032470703, + "learning_rate": 0.00036900493052442854, + "loss": 0.3363, + "step": 22643 + }, + { + "epoch": 1.2651618850741682, + "grad_norm": 1.8784281015396118, + "learning_rate": 0.0003689769161810847, + "loss": 0.4238, + "step": 22644 + }, + { + "epoch": 1.26521775567785, + "grad_norm": 0.494704008102417, + "learning_rate": 0.00036894890183774095, + "loss": 0.4661, + "step": 22645 + }, + { + "epoch": 1.265273626281532, + "grad_norm": 0.6944200992584229, + "learning_rate": 0.0003689208874943971, + "loss": 0.4396, + "step": 22646 + }, + { + "epoch": 1.2653294968852138, + "grad_norm": 0.7300136089324951, + "learning_rate": 0.00036889287315105336, + "loss": 0.3548, + "step": 22647 + }, + { + "epoch": 1.2653853674888957, + "grad_norm": 0.37495091557502747, + "learning_rate": 0.0003688648588077095, + "loss": 0.3366, + "step": 22648 + }, + { + "epoch": 1.2654412380925777, + "grad_norm": 0.4405169188976288, + "learning_rate": 0.0003688368444643658, + "loss": 0.5619, + "step": 22649 + }, + { + "epoch": 1.2654971086962594, + "grad_norm": 0.9203236103057861, + "learning_rate": 0.000368808830121022, + "loss": 0.4619, + "step": 22650 + }, + { + "epoch": 1.2655529792999414, + "grad_norm": 0.681865930557251, + "learning_rate": 0.0003687808157776782, + "loss": 0.4014, + "step": 22651 + }, + { + "epoch": 1.2656088499036233, + "grad_norm": 0.346930593252182, + "learning_rate": 0.0003687528014343344, + "loss": 0.3705, + "step": 22652 + }, + { + "epoch": 1.265664720507305, + "grad_norm": 0.6175767779350281, + "learning_rate": 0.0003687247870909906, + "loss": 0.4017, + "step": 22653 + }, + { + "epoch": 1.265720591110987, + "grad_norm": 0.4682302176952362, + "learning_rate": 0.0003686967727476468, + "loss": 0.4372, + "step": 22654 + }, + { + "epoch": 1.265776461714669, + "grad_norm": 0.39394620060920715, + "learning_rate": 0.000368668758404303, + "loss": 0.3893, + "step": 22655 + }, + { + "epoch": 1.2658323323183507, + "grad_norm": 0.40305376052856445, + "learning_rate": 0.0003686407440609592, + "loss": 0.4714, + "step": 22656 + }, + { + "epoch": 1.2658882029220326, + "grad_norm": 0.433015376329422, + "learning_rate": 0.0003686127297176154, + "loss": 0.4416, + "step": 22657 + }, + { + "epoch": 1.2659440735257144, + "grad_norm": 0.397549033164978, + "learning_rate": 0.00036858471537427163, + "loss": 0.3068, + "step": 22658 + }, + { + "epoch": 1.2659999441293963, + "grad_norm": 0.7279560565948486, + "learning_rate": 0.0003685567010309279, + "loss": 0.6128, + "step": 22659 + }, + { + "epoch": 1.2660558147330783, + "grad_norm": 0.438360333442688, + "learning_rate": 0.00036852868668758404, + "loss": 0.4384, + "step": 22660 + }, + { + "epoch": 1.26611168533676, + "grad_norm": 0.528027355670929, + "learning_rate": 0.0003685006723442403, + "loss": 0.54, + "step": 22661 + }, + { + "epoch": 1.266167555940442, + "grad_norm": 0.45643481612205505, + "learning_rate": 0.00036847265800089645, + "loss": 0.5423, + "step": 22662 + }, + { + "epoch": 1.2662234265441237, + "grad_norm": 0.6138466000556946, + "learning_rate": 0.00036844464365755266, + "loss": 0.3692, + "step": 22663 + }, + { + "epoch": 1.2662792971478056, + "grad_norm": 0.485612154006958, + "learning_rate": 0.00036841662931420886, + "loss": 0.347, + "step": 22664 + }, + { + "epoch": 1.2663351677514876, + "grad_norm": 0.5340859293937683, + "learning_rate": 0.00036838861497086507, + "loss": 0.3938, + "step": 22665 + }, + { + "epoch": 1.2663910383551693, + "grad_norm": 2.031508445739746, + "learning_rate": 0.00036836060062752133, + "loss": 0.3881, + "step": 22666 + }, + { + "epoch": 1.2664469089588513, + "grad_norm": 0.3201711177825928, + "learning_rate": 0.0003683325862841775, + "loss": 0.3629, + "step": 22667 + }, + { + "epoch": 1.2665027795625332, + "grad_norm": 0.5515170097351074, + "learning_rate": 0.00036830457194083374, + "loss": 0.4838, + "step": 22668 + }, + { + "epoch": 1.266558650166215, + "grad_norm": 0.6220821738243103, + "learning_rate": 0.0003682765575974899, + "loss": 0.6062, + "step": 22669 + }, + { + "epoch": 1.266614520769897, + "grad_norm": 0.4443313181400299, + "learning_rate": 0.00036824854325414615, + "loss": 0.4021, + "step": 22670 + }, + { + "epoch": 1.2666703913735788, + "grad_norm": 0.3347076177597046, + "learning_rate": 0.0003682205289108023, + "loss": 0.3967, + "step": 22671 + }, + { + "epoch": 1.2667262619772606, + "grad_norm": 0.45791250467300415, + "learning_rate": 0.00036819251456745856, + "loss": 0.4478, + "step": 22672 + }, + { + "epoch": 1.2667821325809425, + "grad_norm": 0.599399209022522, + "learning_rate": 0.0003681645002241147, + "loss": 0.5332, + "step": 22673 + }, + { + "epoch": 1.2668380031846245, + "grad_norm": 0.47833317518234253, + "learning_rate": 0.000368136485880771, + "loss": 0.4043, + "step": 22674 + }, + { + "epoch": 1.2668938737883062, + "grad_norm": 0.41777825355529785, + "learning_rate": 0.0003681084715374272, + "loss": 0.5245, + "step": 22675 + }, + { + "epoch": 1.2669497443919882, + "grad_norm": 4.794032096862793, + "learning_rate": 0.0003680804571940834, + "loss": 0.3908, + "step": 22676 + }, + { + "epoch": 1.26700561499567, + "grad_norm": 0.5474116206169128, + "learning_rate": 0.0003680524428507396, + "loss": 0.4102, + "step": 22677 + }, + { + "epoch": 1.2670614855993518, + "grad_norm": 0.8248022794723511, + "learning_rate": 0.0003680244285073958, + "loss": 0.397, + "step": 22678 + }, + { + "epoch": 1.2671173562030338, + "grad_norm": 0.43728142976760864, + "learning_rate": 0.000367996414164052, + "loss": 0.3698, + "step": 22679 + }, + { + "epoch": 1.2671732268067157, + "grad_norm": 1.9377825260162354, + "learning_rate": 0.0003679683998207082, + "loss": 0.5615, + "step": 22680 + }, + { + "epoch": 1.2672290974103975, + "grad_norm": 0.42992180585861206, + "learning_rate": 0.0003679403854773644, + "loss": 0.5196, + "step": 22681 + }, + { + "epoch": 1.2672849680140794, + "grad_norm": 0.4016419053077698, + "learning_rate": 0.0003679123711340207, + "loss": 0.3631, + "step": 22682 + }, + { + "epoch": 1.2673408386177614, + "grad_norm": 0.7323240041732788, + "learning_rate": 0.0003678843567906768, + "loss": 0.5352, + "step": 22683 + }, + { + "epoch": 1.267396709221443, + "grad_norm": 0.4914844036102295, + "learning_rate": 0.00036785634244733303, + "loss": 0.3825, + "step": 22684 + }, + { + "epoch": 1.267452579825125, + "grad_norm": 0.583889901638031, + "learning_rate": 0.00036782832810398924, + "loss": 0.4649, + "step": 22685 + }, + { + "epoch": 1.267508450428807, + "grad_norm": 0.3386187255382538, + "learning_rate": 0.00036780031376064544, + "loss": 0.3649, + "step": 22686 + }, + { + "epoch": 1.2675643210324887, + "grad_norm": 0.3480525314807892, + "learning_rate": 0.00036777229941730165, + "loss": 0.4988, + "step": 22687 + }, + { + "epoch": 1.2676201916361707, + "grad_norm": 6.332148551940918, + "learning_rate": 0.00036774428507395786, + "loss": 0.4637, + "step": 22688 + }, + { + "epoch": 1.2676760622398526, + "grad_norm": 0.750042736530304, + "learning_rate": 0.00036771627073061406, + "loss": 0.408, + "step": 22689 + }, + { + "epoch": 1.2677319328435344, + "grad_norm": 0.5804164409637451, + "learning_rate": 0.00036768825638727027, + "loss": 0.4888, + "step": 22690 + }, + { + "epoch": 1.2677878034472163, + "grad_norm": 0.42330676317214966, + "learning_rate": 0.00036766024204392653, + "loss": 0.3835, + "step": 22691 + }, + { + "epoch": 1.267843674050898, + "grad_norm": 0.42583051323890686, + "learning_rate": 0.0003676322277005827, + "loss": 0.4373, + "step": 22692 + }, + { + "epoch": 1.26789954465458, + "grad_norm": 0.43243902921676636, + "learning_rate": 0.00036760421335723894, + "loss": 0.4546, + "step": 22693 + }, + { + "epoch": 1.267955415258262, + "grad_norm": 0.3359048664569855, + "learning_rate": 0.0003675761990138951, + "loss": 0.4588, + "step": 22694 + }, + { + "epoch": 1.2680112858619437, + "grad_norm": 0.34100860357284546, + "learning_rate": 0.00036754818467055135, + "loss": 0.3985, + "step": 22695 + }, + { + "epoch": 1.2680671564656256, + "grad_norm": 0.5589975118637085, + "learning_rate": 0.0003675201703272075, + "loss": 0.6012, + "step": 22696 + }, + { + "epoch": 1.2681230270693074, + "grad_norm": 1.9564964771270752, + "learning_rate": 0.00036749215598386376, + "loss": 0.4321, + "step": 22697 + }, + { + "epoch": 1.2681788976729893, + "grad_norm": 1.7460817098617554, + "learning_rate": 0.00036746414164051997, + "loss": 0.3684, + "step": 22698 + }, + { + "epoch": 1.2682347682766713, + "grad_norm": 0.8719371557235718, + "learning_rate": 0.0003674361272971762, + "loss": 0.4926, + "step": 22699 + }, + { + "epoch": 1.268290638880353, + "grad_norm": 0.5297956466674805, + "learning_rate": 0.0003674081129538324, + "loss": 0.5815, + "step": 22700 + }, + { + "epoch": 1.268346509484035, + "grad_norm": 0.37122493982315063, + "learning_rate": 0.0003673800986104886, + "loss": 0.3489, + "step": 22701 + }, + { + "epoch": 1.2684023800877169, + "grad_norm": 1.1089903116226196, + "learning_rate": 0.0003673520842671448, + "loss": 0.4785, + "step": 22702 + }, + { + "epoch": 1.2684582506913986, + "grad_norm": 0.4829234778881073, + "learning_rate": 0.000367324069923801, + "loss": 0.413, + "step": 22703 + }, + { + "epoch": 1.2685141212950806, + "grad_norm": 0.3774591088294983, + "learning_rate": 0.0003672960555804572, + "loss": 0.4528, + "step": 22704 + }, + { + "epoch": 1.2685699918987625, + "grad_norm": 0.35643208026885986, + "learning_rate": 0.00036726804123711346, + "loss": 0.3483, + "step": 22705 + }, + { + "epoch": 1.2686258625024442, + "grad_norm": 0.6106137633323669, + "learning_rate": 0.0003672400268937696, + "loss": 0.3944, + "step": 22706 + }, + { + "epoch": 1.2686817331061262, + "grad_norm": 0.5135478377342224, + "learning_rate": 0.0003672120125504258, + "loss": 0.3521, + "step": 22707 + }, + { + "epoch": 1.2687376037098081, + "grad_norm": 0.5085130333900452, + "learning_rate": 0.000367183998207082, + "loss": 0.5181, + "step": 22708 + }, + { + "epoch": 1.2687934743134899, + "grad_norm": 0.33775150775909424, + "learning_rate": 0.00036715598386373823, + "loss": 0.433, + "step": 22709 + }, + { + "epoch": 1.2688493449171718, + "grad_norm": 0.4415935277938843, + "learning_rate": 0.00036712796952039444, + "loss": 0.4995, + "step": 22710 + }, + { + "epoch": 1.2689052155208538, + "grad_norm": 0.42596298456192017, + "learning_rate": 0.00036709995517705064, + "loss": 0.4317, + "step": 22711 + }, + { + "epoch": 1.2689610861245355, + "grad_norm": 0.6285608410835266, + "learning_rate": 0.00036707194083370685, + "loss": 0.5743, + "step": 22712 + }, + { + "epoch": 1.2690169567282175, + "grad_norm": 1.3441320657730103, + "learning_rate": 0.00036704392649036306, + "loss": 0.3639, + "step": 22713 + }, + { + "epoch": 1.2690728273318994, + "grad_norm": 0.5348745584487915, + "learning_rate": 0.0003670159121470193, + "loss": 0.433, + "step": 22714 + }, + { + "epoch": 1.2691286979355811, + "grad_norm": 0.489104300737381, + "learning_rate": 0.00036698789780367547, + "loss": 0.3773, + "step": 22715 + }, + { + "epoch": 1.269184568539263, + "grad_norm": 1.2556734085083008, + "learning_rate": 0.00036695988346033173, + "loss": 0.5042, + "step": 22716 + }, + { + "epoch": 1.269240439142945, + "grad_norm": 0.7395796775817871, + "learning_rate": 0.0003669318691169879, + "loss": 0.4965, + "step": 22717 + }, + { + "epoch": 1.2692963097466268, + "grad_norm": 1.3102895021438599, + "learning_rate": 0.00036690385477364414, + "loss": 0.3283, + "step": 22718 + }, + { + "epoch": 1.2693521803503087, + "grad_norm": 0.3887738585472107, + "learning_rate": 0.0003668758404303003, + "loss": 0.3496, + "step": 22719 + }, + { + "epoch": 1.2694080509539907, + "grad_norm": 0.6881412267684937, + "learning_rate": 0.00036684782608695655, + "loss": 0.4376, + "step": 22720 + }, + { + "epoch": 1.2694639215576724, + "grad_norm": 0.46281591057777405, + "learning_rate": 0.00036681981174361276, + "loss": 0.4757, + "step": 22721 + }, + { + "epoch": 1.2695197921613544, + "grad_norm": 0.3707222044467926, + "learning_rate": 0.00036679179740026896, + "loss": 0.4191, + "step": 22722 + }, + { + "epoch": 1.2695756627650363, + "grad_norm": 1.2471442222595215, + "learning_rate": 0.00036676378305692517, + "loss": 0.3281, + "step": 22723 + }, + { + "epoch": 1.269631533368718, + "grad_norm": 0.634489893913269, + "learning_rate": 0.0003667357687135814, + "loss": 0.4373, + "step": 22724 + }, + { + "epoch": 1.2696874039724, + "grad_norm": 0.44983139634132385, + "learning_rate": 0.0003667077543702376, + "loss": 0.3747, + "step": 22725 + }, + { + "epoch": 1.2697432745760817, + "grad_norm": 0.4391081631183624, + "learning_rate": 0.0003666797400268938, + "loss": 0.4183, + "step": 22726 + }, + { + "epoch": 1.2697991451797637, + "grad_norm": 0.6746876239776611, + "learning_rate": 0.00036665172568355, + "loss": 0.5612, + "step": 22727 + }, + { + "epoch": 1.2698550157834456, + "grad_norm": 0.5539418458938599, + "learning_rate": 0.00036662371134020614, + "loss": 0.3972, + "step": 22728 + }, + { + "epoch": 1.2699108863871273, + "grad_norm": 0.3526007831096649, + "learning_rate": 0.0003665956969968624, + "loss": 0.4865, + "step": 22729 + }, + { + "epoch": 1.2699667569908093, + "grad_norm": 0.5482349395751953, + "learning_rate": 0.0003665676826535186, + "loss": 0.5475, + "step": 22730 + }, + { + "epoch": 1.270022627594491, + "grad_norm": 1.5584754943847656, + "learning_rate": 0.0003665396683101748, + "loss": 0.4099, + "step": 22731 + }, + { + "epoch": 1.270078498198173, + "grad_norm": 0.37571364641189575, + "learning_rate": 0.000366511653966831, + "loss": 0.3402, + "step": 22732 + }, + { + "epoch": 1.270134368801855, + "grad_norm": 0.5885977745056152, + "learning_rate": 0.0003664836396234872, + "loss": 0.3541, + "step": 22733 + }, + { + "epoch": 1.2701902394055367, + "grad_norm": 0.42615941166877747, + "learning_rate": 0.00036645562528014343, + "loss": 0.3639, + "step": 22734 + }, + { + "epoch": 1.2702461100092186, + "grad_norm": 0.2828490436077118, + "learning_rate": 0.00036642761093679964, + "loss": 0.3724, + "step": 22735 + }, + { + "epoch": 1.2703019806129006, + "grad_norm": 0.427658349275589, + "learning_rate": 0.00036639959659345584, + "loss": 0.4403, + "step": 22736 + }, + { + "epoch": 1.2703578512165823, + "grad_norm": 0.37338244915008545, + "learning_rate": 0.0003663715822501121, + "loss": 0.4287, + "step": 22737 + }, + { + "epoch": 1.2704137218202642, + "grad_norm": 0.3581800162792206, + "learning_rate": 0.00036634356790676826, + "loss": 0.5239, + "step": 22738 + }, + { + "epoch": 1.2704695924239462, + "grad_norm": 0.8234225511550903, + "learning_rate": 0.0003663155535634245, + "loss": 0.5981, + "step": 22739 + }, + { + "epoch": 1.270525463027628, + "grad_norm": 0.3835955560207367, + "learning_rate": 0.00036628753922008067, + "loss": 0.3383, + "step": 22740 + }, + { + "epoch": 1.2705813336313099, + "grad_norm": 0.7153425216674805, + "learning_rate": 0.0003662595248767369, + "loss": 0.6186, + "step": 22741 + }, + { + "epoch": 1.2706372042349918, + "grad_norm": 0.6371944546699524, + "learning_rate": 0.0003662315105333931, + "loss": 0.3877, + "step": 22742 + }, + { + "epoch": 1.2706930748386736, + "grad_norm": 0.4671350121498108, + "learning_rate": 0.00036620349619004934, + "loss": 0.4748, + "step": 22743 + }, + { + "epoch": 1.2707489454423555, + "grad_norm": 0.7062814831733704, + "learning_rate": 0.0003661754818467055, + "loss": 0.3208, + "step": 22744 + }, + { + "epoch": 1.2708048160460375, + "grad_norm": 0.530273973941803, + "learning_rate": 0.00036614746750336175, + "loss": 0.4917, + "step": 22745 + }, + { + "epoch": 1.2708606866497192, + "grad_norm": 0.5804493427276611, + "learning_rate": 0.00036611945316001796, + "loss": 0.4565, + "step": 22746 + }, + { + "epoch": 1.2709165572534011, + "grad_norm": 2.0665152072906494, + "learning_rate": 0.00036609143881667416, + "loss": 0.4188, + "step": 22747 + }, + { + "epoch": 1.270972427857083, + "grad_norm": 0.4250573515892029, + "learning_rate": 0.00036606342447333037, + "loss": 0.37, + "step": 22748 + }, + { + "epoch": 1.2710282984607648, + "grad_norm": 0.4313550293445587, + "learning_rate": 0.0003660354101299865, + "loss": 0.4509, + "step": 22749 + }, + { + "epoch": 1.2710841690644468, + "grad_norm": 0.3104036748409271, + "learning_rate": 0.0003660073957866428, + "loss": 0.3565, + "step": 22750 + }, + { + "epoch": 1.2711400396681287, + "grad_norm": 0.43279698491096497, + "learning_rate": 0.00036597938144329893, + "loss": 0.3448, + "step": 22751 + }, + { + "epoch": 1.2711959102718104, + "grad_norm": 0.33111846446990967, + "learning_rate": 0.0003659513670999552, + "loss": 0.3942, + "step": 22752 + }, + { + "epoch": 1.2712517808754924, + "grad_norm": 0.7926313877105713, + "learning_rate": 0.0003659233527566114, + "loss": 0.4659, + "step": 22753 + }, + { + "epoch": 1.2713076514791743, + "grad_norm": 0.5256655216217041, + "learning_rate": 0.0003658953384132676, + "loss": 0.44, + "step": 22754 + }, + { + "epoch": 1.271363522082856, + "grad_norm": 0.4217732846736908, + "learning_rate": 0.0003658673240699238, + "loss": 0.5728, + "step": 22755 + }, + { + "epoch": 1.271419392686538, + "grad_norm": 0.4280329942703247, + "learning_rate": 0.00036583930972658, + "loss": 0.434, + "step": 22756 + }, + { + "epoch": 1.27147526329022, + "grad_norm": 0.5392220616340637, + "learning_rate": 0.0003658112953832362, + "loss": 0.4067, + "step": 22757 + }, + { + "epoch": 1.2715311338939017, + "grad_norm": 0.3500911295413971, + "learning_rate": 0.0003657832810398924, + "loss": 0.3734, + "step": 22758 + }, + { + "epoch": 1.2715870044975837, + "grad_norm": 0.4214332103729248, + "learning_rate": 0.00036575526669654863, + "loss": 0.4297, + "step": 22759 + }, + { + "epoch": 1.2716428751012654, + "grad_norm": 0.353361576795578, + "learning_rate": 0.0003657272523532049, + "loss": 0.3421, + "step": 22760 + }, + { + "epoch": 1.2716987457049473, + "grad_norm": 0.4650281071662903, + "learning_rate": 0.00036569923800986104, + "loss": 0.4085, + "step": 22761 + }, + { + "epoch": 1.2717546163086293, + "grad_norm": 0.4737824499607086, + "learning_rate": 0.0003656712236665173, + "loss": 0.4689, + "step": 22762 + }, + { + "epoch": 1.271810486912311, + "grad_norm": 0.5304277539253235, + "learning_rate": 0.00036564320932317345, + "loss": 0.3177, + "step": 22763 + }, + { + "epoch": 1.271866357515993, + "grad_norm": 0.7993812561035156, + "learning_rate": 0.0003656151949798297, + "loss": 0.4455, + "step": 22764 + }, + { + "epoch": 1.2719222281196747, + "grad_norm": 0.36563605070114136, + "learning_rate": 0.00036558718063648587, + "loss": 0.3525, + "step": 22765 + }, + { + "epoch": 1.2719780987233567, + "grad_norm": 1.1552432775497437, + "learning_rate": 0.0003655591662931421, + "loss": 0.5425, + "step": 22766 + }, + { + "epoch": 1.2720339693270386, + "grad_norm": 0.4758748710155487, + "learning_rate": 0.0003655311519497983, + "loss": 0.4243, + "step": 22767 + }, + { + "epoch": 1.2720898399307203, + "grad_norm": 0.40175044536590576, + "learning_rate": 0.00036550313760645454, + "loss": 0.3926, + "step": 22768 + }, + { + "epoch": 1.2721457105344023, + "grad_norm": 0.44260919094085693, + "learning_rate": 0.00036547512326311074, + "loss": 0.4193, + "step": 22769 + }, + { + "epoch": 1.2722015811380842, + "grad_norm": 0.4271306097507477, + "learning_rate": 0.0003654471089197669, + "loss": 0.3775, + "step": 22770 + }, + { + "epoch": 1.272257451741766, + "grad_norm": 2.8890438079833984, + "learning_rate": 0.00036541909457642316, + "loss": 0.5381, + "step": 22771 + }, + { + "epoch": 1.272313322345448, + "grad_norm": 0.6127537488937378, + "learning_rate": 0.0003653910802330793, + "loss": 0.7028, + "step": 22772 + }, + { + "epoch": 1.2723691929491299, + "grad_norm": 0.45295262336730957, + "learning_rate": 0.00036536306588973557, + "loss": 0.4893, + "step": 22773 + }, + { + "epoch": 1.2724250635528116, + "grad_norm": 0.49266254901885986, + "learning_rate": 0.0003653350515463917, + "loss": 0.4733, + "step": 22774 + }, + { + "epoch": 1.2724809341564935, + "grad_norm": 0.5203219652175903, + "learning_rate": 0.000365307037203048, + "loss": 0.4058, + "step": 22775 + }, + { + "epoch": 1.2725368047601755, + "grad_norm": 0.37119391560554504, + "learning_rate": 0.0003652790228597042, + "loss": 0.3984, + "step": 22776 + }, + { + "epoch": 1.2725926753638572, + "grad_norm": 0.6792534589767456, + "learning_rate": 0.0003652510085163604, + "loss": 0.713, + "step": 22777 + }, + { + "epoch": 1.2726485459675392, + "grad_norm": 0.46333184838294983, + "learning_rate": 0.0003652229941730166, + "loss": 0.381, + "step": 22778 + }, + { + "epoch": 1.2727044165712211, + "grad_norm": 0.518741250038147, + "learning_rate": 0.0003651949798296728, + "loss": 0.5731, + "step": 22779 + }, + { + "epoch": 1.2727602871749029, + "grad_norm": 0.3699832856655121, + "learning_rate": 0.000365166965486329, + "loss": 0.4655, + "step": 22780 + }, + { + "epoch": 1.2728161577785848, + "grad_norm": 0.6998206377029419, + "learning_rate": 0.0003651389511429852, + "loss": 0.4224, + "step": 22781 + }, + { + "epoch": 1.2728720283822668, + "grad_norm": 0.39450469613075256, + "learning_rate": 0.0003651109367996414, + "loss": 0.4067, + "step": 22782 + }, + { + "epoch": 1.2729278989859485, + "grad_norm": 0.5956903696060181, + "learning_rate": 0.0003650829224562976, + "loss": 0.3601, + "step": 22783 + }, + { + "epoch": 1.2729837695896304, + "grad_norm": 0.4839878976345062, + "learning_rate": 0.00036505490811295383, + "loss": 0.4426, + "step": 22784 + }, + { + "epoch": 1.2730396401933124, + "grad_norm": 0.43798011541366577, + "learning_rate": 0.0003650268937696101, + "loss": 0.4529, + "step": 22785 + }, + { + "epoch": 1.2730955107969941, + "grad_norm": 0.4083445370197296, + "learning_rate": 0.00036499887942626624, + "loss": 0.3116, + "step": 22786 + }, + { + "epoch": 1.273151381400676, + "grad_norm": 0.36424142122268677, + "learning_rate": 0.0003649708650829225, + "loss": 0.415, + "step": 22787 + }, + { + "epoch": 1.273207252004358, + "grad_norm": 0.48015356063842773, + "learning_rate": 0.00036494285073957865, + "loss": 0.3826, + "step": 22788 + }, + { + "epoch": 1.2732631226080398, + "grad_norm": 0.5101886987686157, + "learning_rate": 0.0003649148363962349, + "loss": 0.3269, + "step": 22789 + }, + { + "epoch": 1.2733189932117217, + "grad_norm": 0.44838643074035645, + "learning_rate": 0.00036488682205289107, + "loss": 0.4463, + "step": 22790 + }, + { + "epoch": 1.2733748638154037, + "grad_norm": 1.2206228971481323, + "learning_rate": 0.00036485880770954727, + "loss": 0.4924, + "step": 22791 + }, + { + "epoch": 1.2734307344190854, + "grad_norm": 0.45822882652282715, + "learning_rate": 0.00036483079336620353, + "loss": 0.3805, + "step": 22792 + }, + { + "epoch": 1.2734866050227673, + "grad_norm": 0.4206635057926178, + "learning_rate": 0.0003648027790228597, + "loss": 0.424, + "step": 22793 + }, + { + "epoch": 1.273542475626449, + "grad_norm": 0.7071791887283325, + "learning_rate": 0.00036477476467951594, + "loss": 0.4598, + "step": 22794 + }, + { + "epoch": 1.273598346230131, + "grad_norm": 0.3163171112537384, + "learning_rate": 0.0003647467503361721, + "loss": 0.41, + "step": 22795 + }, + { + "epoch": 1.273654216833813, + "grad_norm": 1.4973394870758057, + "learning_rate": 0.00036471873599282836, + "loss": 0.3525, + "step": 22796 + }, + { + "epoch": 1.2737100874374947, + "grad_norm": 1.0626578330993652, + "learning_rate": 0.0003646907216494845, + "loss": 0.7156, + "step": 22797 + }, + { + "epoch": 1.2737659580411766, + "grad_norm": 0.9857659935951233, + "learning_rate": 0.00036466270730614077, + "loss": 0.5303, + "step": 22798 + }, + { + "epoch": 1.2738218286448584, + "grad_norm": 0.462295264005661, + "learning_rate": 0.0003646346929627969, + "loss": 0.4199, + "step": 22799 + }, + { + "epoch": 1.2738776992485403, + "grad_norm": 0.4952392578125, + "learning_rate": 0.0003646066786194532, + "loss": 0.4199, + "step": 22800 + }, + { + "epoch": 1.2739335698522223, + "grad_norm": 0.3521595299243927, + "learning_rate": 0.0003645786642761094, + "loss": 0.3988, + "step": 22801 + }, + { + "epoch": 1.273989440455904, + "grad_norm": 0.4139104187488556, + "learning_rate": 0.0003645506499327656, + "loss": 0.3796, + "step": 22802 + }, + { + "epoch": 1.274045311059586, + "grad_norm": 0.3258519768714905, + "learning_rate": 0.0003645226355894218, + "loss": 0.4315, + "step": 22803 + }, + { + "epoch": 1.274101181663268, + "grad_norm": Infinity, + "learning_rate": 0.0003645226355894218, + "loss": 0.3368, + "step": 22804 + }, + { + "epoch": 1.2741570522669496, + "grad_norm": 0.4501104950904846, + "learning_rate": 0.000364494621246078, + "loss": 0.3598, + "step": 22805 + }, + { + "epoch": 1.2742129228706316, + "grad_norm": 0.532156229019165, + "learning_rate": 0.0003644666069027342, + "loss": 0.3651, + "step": 22806 + }, + { + "epoch": 1.2742687934743135, + "grad_norm": 0.3544784486293793, + "learning_rate": 0.0003644385925593904, + "loss": 0.3445, + "step": 22807 + }, + { + "epoch": 1.2743246640779953, + "grad_norm": 0.49451392889022827, + "learning_rate": 0.0003644105782160466, + "loss": 0.5138, + "step": 22808 + }, + { + "epoch": 1.2743805346816772, + "grad_norm": 0.5886214375495911, + "learning_rate": 0.0003643825638727029, + "loss": 0.3983, + "step": 22809 + }, + { + "epoch": 1.2744364052853592, + "grad_norm": 0.3175130784511566, + "learning_rate": 0.00036435454952935903, + "loss": 0.4224, + "step": 22810 + }, + { + "epoch": 1.274492275889041, + "grad_norm": 1.1436595916748047, + "learning_rate": 0.0003643265351860153, + "loss": 0.3729, + "step": 22811 + }, + { + "epoch": 1.2745481464927229, + "grad_norm": 0.7195921540260315, + "learning_rate": 0.00036429852084267144, + "loss": 0.3727, + "step": 22812 + }, + { + "epoch": 1.2746040170964048, + "grad_norm": 0.5495458841323853, + "learning_rate": 0.00036427050649932765, + "loss": 0.5556, + "step": 22813 + }, + { + "epoch": 1.2746598877000865, + "grad_norm": 0.5382723212242126, + "learning_rate": 0.00036424249215598385, + "loss": 0.6886, + "step": 22814 + }, + { + "epoch": 1.2747157583037685, + "grad_norm": 1.5814590454101562, + "learning_rate": 0.00036421447781264006, + "loss": 0.4457, + "step": 22815 + }, + { + "epoch": 1.2747716289074504, + "grad_norm": 0.32218337059020996, + "learning_rate": 0.00036418646346929627, + "loss": 0.4565, + "step": 22816 + }, + { + "epoch": 1.2748274995111322, + "grad_norm": 0.4090750217437744, + "learning_rate": 0.00036415844912595247, + "loss": 0.3062, + "step": 22817 + }, + { + "epoch": 1.2748833701148141, + "grad_norm": 0.5075331926345825, + "learning_rate": 0.00036413043478260873, + "loss": 0.4083, + "step": 22818 + }, + { + "epoch": 1.274939240718496, + "grad_norm": 0.6882632970809937, + "learning_rate": 0.0003641024204392649, + "loss": 0.4786, + "step": 22819 + }, + { + "epoch": 1.2749951113221778, + "grad_norm": 0.6669955849647522, + "learning_rate": 0.00036407440609592114, + "loss": 0.4296, + "step": 22820 + }, + { + "epoch": 1.2750509819258597, + "grad_norm": 0.5129563808441162, + "learning_rate": 0.0003640463917525773, + "loss": 0.4688, + "step": 22821 + }, + { + "epoch": 1.2751068525295417, + "grad_norm": 0.5263575315475464, + "learning_rate": 0.00036401837740923355, + "loss": 0.4382, + "step": 22822 + }, + { + "epoch": 1.2751627231332234, + "grad_norm": 2.5272090435028076, + "learning_rate": 0.0003639903630658897, + "loss": 0.3888, + "step": 22823 + }, + { + "epoch": 1.2752185937369054, + "grad_norm": 0.37081536650657654, + "learning_rate": 0.00036396234872254597, + "loss": 0.4184, + "step": 22824 + }, + { + "epoch": 1.2752744643405873, + "grad_norm": 0.8294228315353394, + "learning_rate": 0.00036393433437920217, + "loss": 0.4225, + "step": 22825 + }, + { + "epoch": 1.275330334944269, + "grad_norm": 0.34471219778060913, + "learning_rate": 0.0003639063200358584, + "loss": 0.2858, + "step": 22826 + }, + { + "epoch": 1.275386205547951, + "grad_norm": 0.868983268737793, + "learning_rate": 0.0003638783056925146, + "loss": 0.4456, + "step": 22827 + }, + { + "epoch": 1.2754420761516327, + "grad_norm": 0.6281024813652039, + "learning_rate": 0.0003638502913491708, + "loss": 0.4543, + "step": 22828 + }, + { + "epoch": 1.2754979467553147, + "grad_norm": 0.4224196672439575, + "learning_rate": 0.000363822277005827, + "loss": 0.4491, + "step": 22829 + }, + { + "epoch": 1.2755538173589966, + "grad_norm": 0.4466993510723114, + "learning_rate": 0.0003637942626624832, + "loss": 0.4972, + "step": 22830 + }, + { + "epoch": 1.2756096879626784, + "grad_norm": 3.5049827098846436, + "learning_rate": 0.0003637662483191394, + "loss": 0.5074, + "step": 22831 + }, + { + "epoch": 1.2756655585663603, + "grad_norm": 0.46206772327423096, + "learning_rate": 0.00036373823397579567, + "loss": 0.3736, + "step": 22832 + }, + { + "epoch": 1.275721429170042, + "grad_norm": 0.4143012762069702, + "learning_rate": 0.0003637102196324518, + "loss": 0.3793, + "step": 22833 + }, + { + "epoch": 1.275777299773724, + "grad_norm": 0.4378529191017151, + "learning_rate": 0.000363682205289108, + "loss": 0.3821, + "step": 22834 + }, + { + "epoch": 1.275833170377406, + "grad_norm": 0.42659464478492737, + "learning_rate": 0.00036365419094576423, + "loss": 0.3308, + "step": 22835 + }, + { + "epoch": 1.2758890409810877, + "grad_norm": 0.4786287844181061, + "learning_rate": 0.00036362617660242044, + "loss": 0.4192, + "step": 22836 + }, + { + "epoch": 1.2759449115847696, + "grad_norm": 0.5788267850875854, + "learning_rate": 0.00036359816225907664, + "loss": 0.3284, + "step": 22837 + }, + { + "epoch": 1.2760007821884516, + "grad_norm": 0.42928868532180786, + "learning_rate": 0.00036357014791573285, + "loss": 0.3153, + "step": 22838 + }, + { + "epoch": 1.2760566527921333, + "grad_norm": 2.4183273315429688, + "learning_rate": 0.00036354213357238905, + "loss": 0.4956, + "step": 22839 + }, + { + "epoch": 1.2761125233958153, + "grad_norm": 0.44712671637535095, + "learning_rate": 0.00036351411922904526, + "loss": 0.4536, + "step": 22840 + }, + { + "epoch": 1.2761683939994972, + "grad_norm": 0.40309712290763855, + "learning_rate": 0.0003634861048857015, + "loss": 0.4167, + "step": 22841 + }, + { + "epoch": 1.276224264603179, + "grad_norm": 0.8747986555099487, + "learning_rate": 0.00036345809054235767, + "loss": 0.3757, + "step": 22842 + }, + { + "epoch": 1.276280135206861, + "grad_norm": 1.1710498332977295, + "learning_rate": 0.00036343007619901393, + "loss": 0.4487, + "step": 22843 + }, + { + "epoch": 1.2763360058105429, + "grad_norm": 0.5748889446258545, + "learning_rate": 0.0003634020618556701, + "loss": 0.5239, + "step": 22844 + }, + { + "epoch": 1.2763918764142246, + "grad_norm": 0.39938288927078247, + "learning_rate": 0.00036337404751232634, + "loss": 0.5349, + "step": 22845 + }, + { + "epoch": 1.2764477470179065, + "grad_norm": 0.4949954152107239, + "learning_rate": 0.0003633460331689825, + "loss": 0.363, + "step": 22846 + }, + { + "epoch": 1.2765036176215885, + "grad_norm": 0.5126270651817322, + "learning_rate": 0.00036331801882563875, + "loss": 0.4775, + "step": 22847 + }, + { + "epoch": 1.2765594882252702, + "grad_norm": 0.4578453302383423, + "learning_rate": 0.00036329000448229496, + "loss": 0.4985, + "step": 22848 + }, + { + "epoch": 1.2766153588289522, + "grad_norm": 0.44908177852630615, + "learning_rate": 0.00036326199013895117, + "loss": 0.4239, + "step": 22849 + }, + { + "epoch": 1.2766712294326341, + "grad_norm": 0.5613842606544495, + "learning_rate": 0.00036323397579560737, + "loss": 0.4686, + "step": 22850 + }, + { + "epoch": 1.2767271000363158, + "grad_norm": 0.33682048320770264, + "learning_rate": 0.0003632059614522636, + "loss": 0.3124, + "step": 22851 + }, + { + "epoch": 1.2767829706399978, + "grad_norm": 0.4980594515800476, + "learning_rate": 0.0003631779471089198, + "loss": 0.3646, + "step": 22852 + }, + { + "epoch": 1.2768388412436797, + "grad_norm": 0.4326109290122986, + "learning_rate": 0.000363149932765576, + "loss": 0.3737, + "step": 22853 + }, + { + "epoch": 1.2768947118473615, + "grad_norm": 0.3588540256023407, + "learning_rate": 0.0003631219184222322, + "loss": 0.3783, + "step": 22854 + }, + { + "epoch": 1.2769505824510434, + "grad_norm": 0.3565175235271454, + "learning_rate": 0.00036309390407888835, + "loss": 0.4386, + "step": 22855 + }, + { + "epoch": 1.2770064530547254, + "grad_norm": 0.473531037569046, + "learning_rate": 0.0003630658897355446, + "loss": 0.2905, + "step": 22856 + }, + { + "epoch": 1.277062323658407, + "grad_norm": 0.40934568643569946, + "learning_rate": 0.0003630378753922008, + "loss": 0.3156, + "step": 22857 + }, + { + "epoch": 1.277118194262089, + "grad_norm": 0.41967833042144775, + "learning_rate": 0.000363009861048857, + "loss": 0.4558, + "step": 22858 + }, + { + "epoch": 1.277174064865771, + "grad_norm": 0.47910410165786743, + "learning_rate": 0.0003629818467055132, + "loss": 0.384, + "step": 22859 + }, + { + "epoch": 1.2772299354694527, + "grad_norm": 0.47813963890075684, + "learning_rate": 0.00036295383236216943, + "loss": 0.3466, + "step": 22860 + }, + { + "epoch": 1.2772858060731347, + "grad_norm": 0.34057682752609253, + "learning_rate": 0.00036292581801882564, + "loss": 0.3901, + "step": 22861 + }, + { + "epoch": 1.2773416766768164, + "grad_norm": 0.5030060410499573, + "learning_rate": 0.00036289780367548184, + "loss": 0.4006, + "step": 22862 + }, + { + "epoch": 1.2773975472804984, + "grad_norm": 0.8282269835472107, + "learning_rate": 0.00036286978933213805, + "loss": 0.3353, + "step": 22863 + }, + { + "epoch": 1.2774534178841803, + "grad_norm": 0.3078173100948334, + "learning_rate": 0.0003628417749887943, + "loss": 0.3109, + "step": 22864 + }, + { + "epoch": 1.277509288487862, + "grad_norm": 8.146272659301758, + "learning_rate": 0.00036281376064545046, + "loss": 0.393, + "step": 22865 + }, + { + "epoch": 1.277565159091544, + "grad_norm": 0.5241509079933167, + "learning_rate": 0.0003627857463021067, + "loss": 0.4842, + "step": 22866 + }, + { + "epoch": 1.2776210296952257, + "grad_norm": 1.2938892841339111, + "learning_rate": 0.00036275773195876287, + "loss": 0.4185, + "step": 22867 + }, + { + "epoch": 1.2776769002989077, + "grad_norm": 0.39422985911369324, + "learning_rate": 0.00036272971761541913, + "loss": 0.4018, + "step": 22868 + }, + { + "epoch": 1.2777327709025896, + "grad_norm": 0.7882826924324036, + "learning_rate": 0.0003627017032720753, + "loss": 0.5614, + "step": 22869 + }, + { + "epoch": 1.2777886415062714, + "grad_norm": 0.6569625735282898, + "learning_rate": 0.00036267368892873154, + "loss": 0.362, + "step": 22870 + }, + { + "epoch": 1.2778445121099533, + "grad_norm": 0.4699888825416565, + "learning_rate": 0.0003626456745853877, + "loss": 0.4699, + "step": 22871 + }, + { + "epoch": 1.2779003827136353, + "grad_norm": 0.5151633024215698, + "learning_rate": 0.00036261766024204395, + "loss": 0.3792, + "step": 22872 + }, + { + "epoch": 1.277956253317317, + "grad_norm": 0.6927143931388855, + "learning_rate": 0.00036258964589870016, + "loss": 0.4831, + "step": 22873 + }, + { + "epoch": 1.278012123920999, + "grad_norm": 1.1190656423568726, + "learning_rate": 0.00036256163155535637, + "loss": 0.5537, + "step": 22874 + }, + { + "epoch": 1.278067994524681, + "grad_norm": 4.553806304931641, + "learning_rate": 0.00036253361721201257, + "loss": 0.3856, + "step": 22875 + }, + { + "epoch": 1.2781238651283626, + "grad_norm": 1.9341027736663818, + "learning_rate": 0.0003625056028686688, + "loss": 0.3969, + "step": 22876 + }, + { + "epoch": 1.2781797357320446, + "grad_norm": 1.4958651065826416, + "learning_rate": 0.000362477588525325, + "loss": 0.4086, + "step": 22877 + }, + { + "epoch": 1.2782356063357265, + "grad_norm": 0.5544141530990601, + "learning_rate": 0.00036244957418198113, + "loss": 0.5369, + "step": 22878 + }, + { + "epoch": 1.2782914769394083, + "grad_norm": 0.3837625980377197, + "learning_rate": 0.0003624215598386374, + "loss": 0.4432, + "step": 22879 + }, + { + "epoch": 1.2783473475430902, + "grad_norm": 1.1959116458892822, + "learning_rate": 0.0003623935454952936, + "loss": 0.4703, + "step": 22880 + }, + { + "epoch": 1.2784032181467722, + "grad_norm": 0.423955500125885, + "learning_rate": 0.0003623655311519498, + "loss": 0.3087, + "step": 22881 + }, + { + "epoch": 1.2784590887504539, + "grad_norm": 0.45973727107048035, + "learning_rate": 0.000362337516808606, + "loss": 0.4136, + "step": 22882 + }, + { + "epoch": 1.2785149593541358, + "grad_norm": 0.4074322581291199, + "learning_rate": 0.0003623095024652622, + "loss": 0.4312, + "step": 22883 + }, + { + "epoch": 1.2785708299578178, + "grad_norm": 0.966677188873291, + "learning_rate": 0.0003622814881219184, + "loss": 0.425, + "step": 22884 + }, + { + "epoch": 1.2786267005614995, + "grad_norm": 0.3929910361766815, + "learning_rate": 0.00036225347377857463, + "loss": 0.3465, + "step": 22885 + }, + { + "epoch": 1.2786825711651815, + "grad_norm": 0.6964865922927856, + "learning_rate": 0.00036222545943523084, + "loss": 0.4876, + "step": 22886 + }, + { + "epoch": 1.2787384417688634, + "grad_norm": 0.27374210953712463, + "learning_rate": 0.00036219744509188704, + "loss": 0.3384, + "step": 22887 + }, + { + "epoch": 1.2787943123725451, + "grad_norm": 0.3725210130214691, + "learning_rate": 0.00036216943074854325, + "loss": 0.3239, + "step": 22888 + }, + { + "epoch": 1.278850182976227, + "grad_norm": 0.4208071231842041, + "learning_rate": 0.0003621414164051995, + "loss": 0.317, + "step": 22889 + }, + { + "epoch": 1.278906053579909, + "grad_norm": 0.6098574995994568, + "learning_rate": 0.00036211340206185566, + "loss": 0.5835, + "step": 22890 + }, + { + "epoch": 1.2789619241835908, + "grad_norm": 1.183098316192627, + "learning_rate": 0.0003620853877185119, + "loss": 0.5175, + "step": 22891 + }, + { + "epoch": 1.2790177947872727, + "grad_norm": 0.35798341035842896, + "learning_rate": 0.00036205737337516807, + "loss": 0.4051, + "step": 22892 + }, + { + "epoch": 1.2790736653909547, + "grad_norm": 0.40371617674827576, + "learning_rate": 0.00036202935903182433, + "loss": 0.4314, + "step": 22893 + }, + { + "epoch": 1.2791295359946364, + "grad_norm": 0.6733956933021545, + "learning_rate": 0.0003620013446884805, + "loss": 0.578, + "step": 22894 + }, + { + "epoch": 1.2791854065983184, + "grad_norm": 0.7281233072280884, + "learning_rate": 0.00036197333034513674, + "loss": 0.4177, + "step": 22895 + }, + { + "epoch": 1.279241277202, + "grad_norm": 0.36107710003852844, + "learning_rate": 0.00036194531600179295, + "loss": 0.4535, + "step": 22896 + }, + { + "epoch": 1.279297147805682, + "grad_norm": 0.5872259140014648, + "learning_rate": 0.00036191730165844915, + "loss": 0.5266, + "step": 22897 + }, + { + "epoch": 1.279353018409364, + "grad_norm": 0.39316296577453613, + "learning_rate": 0.00036188928731510536, + "loss": 0.4815, + "step": 22898 + }, + { + "epoch": 1.2794088890130457, + "grad_norm": 0.3728954493999481, + "learning_rate": 0.0003618612729717615, + "loss": 0.3966, + "step": 22899 + }, + { + "epoch": 1.2794647596167277, + "grad_norm": 0.4714077413082123, + "learning_rate": 0.00036183325862841777, + "loss": 0.3941, + "step": 22900 + }, + { + "epoch": 1.2795206302204094, + "grad_norm": 0.5449408292770386, + "learning_rate": 0.0003618052442850739, + "loss": 0.4389, + "step": 22901 + }, + { + "epoch": 1.2795765008240914, + "grad_norm": 0.5441756844520569, + "learning_rate": 0.0003617772299417302, + "loss": 0.4478, + "step": 22902 + }, + { + "epoch": 1.2796323714277733, + "grad_norm": 0.5068823099136353, + "learning_rate": 0.00036174921559838633, + "loss": 0.5652, + "step": 22903 + }, + { + "epoch": 1.279688242031455, + "grad_norm": 0.536864161491394, + "learning_rate": 0.0003617212012550426, + "loss": 0.4605, + "step": 22904 + }, + { + "epoch": 1.279744112635137, + "grad_norm": 0.32756635546684265, + "learning_rate": 0.0003616931869116988, + "loss": 0.3862, + "step": 22905 + }, + { + "epoch": 1.279799983238819, + "grad_norm": 0.39653781056404114, + "learning_rate": 0.000361665172568355, + "loss": 0.413, + "step": 22906 + }, + { + "epoch": 1.2798558538425007, + "grad_norm": 0.30865588784217834, + "learning_rate": 0.0003616371582250112, + "loss": 0.3001, + "step": 22907 + }, + { + "epoch": 1.2799117244461826, + "grad_norm": 1.1913204193115234, + "learning_rate": 0.0003616091438816674, + "loss": 0.3808, + "step": 22908 + }, + { + "epoch": 1.2799675950498646, + "grad_norm": 0.6734180450439453, + "learning_rate": 0.0003615811295383236, + "loss": 0.3881, + "step": 22909 + }, + { + "epoch": 1.2800234656535463, + "grad_norm": 0.37707018852233887, + "learning_rate": 0.00036155311519497983, + "loss": 0.3783, + "step": 22910 + }, + { + "epoch": 1.2800793362572283, + "grad_norm": 0.41479209065437317, + "learning_rate": 0.00036152510085163603, + "loss": 0.4139, + "step": 22911 + }, + { + "epoch": 1.2801352068609102, + "grad_norm": 0.4097507894039154, + "learning_rate": 0.0003614970865082923, + "loss": 0.4529, + "step": 22912 + }, + { + "epoch": 1.280191077464592, + "grad_norm": 0.45703649520874023, + "learning_rate": 0.00036146907216494845, + "loss": 0.4278, + "step": 22913 + }, + { + "epoch": 1.2802469480682739, + "grad_norm": 0.4147685170173645, + "learning_rate": 0.0003614410578216047, + "loss": 0.4619, + "step": 22914 + }, + { + "epoch": 1.2803028186719558, + "grad_norm": 0.3799510598182678, + "learning_rate": 0.00036141304347826086, + "loss": 0.3381, + "step": 22915 + }, + { + "epoch": 1.2803586892756376, + "grad_norm": 0.3922449052333832, + "learning_rate": 0.0003613850291349171, + "loss": 0.4263, + "step": 22916 + }, + { + "epoch": 1.2804145598793195, + "grad_norm": 0.8964294791221619, + "learning_rate": 0.00036135701479157327, + "loss": 0.4378, + "step": 22917 + }, + { + "epoch": 1.2804704304830015, + "grad_norm": 0.3352363407611847, + "learning_rate": 0.00036132900044822953, + "loss": 0.4298, + "step": 22918 + }, + { + "epoch": 1.2805263010866832, + "grad_norm": 0.4727666676044464, + "learning_rate": 0.00036130098610488574, + "loss": 0.3983, + "step": 22919 + }, + { + "epoch": 1.2805821716903651, + "grad_norm": 0.5771908164024353, + "learning_rate": 0.0003612729717615419, + "loss": 0.3954, + "step": 22920 + }, + { + "epoch": 1.280638042294047, + "grad_norm": 0.3465748727321625, + "learning_rate": 0.00036124495741819815, + "loss": 0.3231, + "step": 22921 + }, + { + "epoch": 1.2806939128977288, + "grad_norm": 0.4009021818637848, + "learning_rate": 0.0003612169430748543, + "loss": 0.3905, + "step": 22922 + }, + { + "epoch": 1.2807497835014108, + "grad_norm": 0.3646935224533081, + "learning_rate": 0.00036118892873151056, + "loss": 0.4115, + "step": 22923 + }, + { + "epoch": 1.2808056541050927, + "grad_norm": 0.44627293944358826, + "learning_rate": 0.0003611609143881667, + "loss": 0.5009, + "step": 22924 + }, + { + "epoch": 1.2808615247087745, + "grad_norm": 0.424024373292923, + "learning_rate": 0.00036113290004482297, + "loss": 0.4003, + "step": 22925 + }, + { + "epoch": 1.2809173953124564, + "grad_norm": 0.39636996388435364, + "learning_rate": 0.0003611048857014791, + "loss": 0.4191, + "step": 22926 + }, + { + "epoch": 1.2809732659161384, + "grad_norm": 2.072721004486084, + "learning_rate": 0.0003610768713581354, + "loss": 0.3856, + "step": 22927 + }, + { + "epoch": 1.28102913651982, + "grad_norm": 0.43775948882102966, + "learning_rate": 0.0003610488570147916, + "loss": 0.3587, + "step": 22928 + }, + { + "epoch": 1.281085007123502, + "grad_norm": 0.7359970808029175, + "learning_rate": 0.0003610208426714478, + "loss": 0.5087, + "step": 22929 + }, + { + "epoch": 1.2811408777271838, + "grad_norm": 0.40103355050086975, + "learning_rate": 0.000360992828328104, + "loss": 0.5469, + "step": 22930 + }, + { + "epoch": 1.2811967483308657, + "grad_norm": 0.5053516626358032, + "learning_rate": 0.0003609648139847602, + "loss": 0.5494, + "step": 22931 + }, + { + "epoch": 1.2812526189345474, + "grad_norm": 0.4920208752155304, + "learning_rate": 0.0003609367996414164, + "loss": 0.4374, + "step": 22932 + }, + { + "epoch": 1.2813084895382294, + "grad_norm": 1.9453294277191162, + "learning_rate": 0.0003609087852980726, + "loss": 0.4378, + "step": 22933 + }, + { + "epoch": 1.2813643601419114, + "grad_norm": 0.47329792380332947, + "learning_rate": 0.0003608807709547288, + "loss": 0.3703, + "step": 22934 + }, + { + "epoch": 1.281420230745593, + "grad_norm": 0.8505758047103882, + "learning_rate": 0.0003608527566113851, + "loss": 0.4811, + "step": 22935 + }, + { + "epoch": 1.281476101349275, + "grad_norm": 0.5575684309005737, + "learning_rate": 0.00036082474226804123, + "loss": 0.4029, + "step": 22936 + }, + { + "epoch": 1.281531971952957, + "grad_norm": 0.4711216688156128, + "learning_rate": 0.0003607967279246975, + "loss": 0.4286, + "step": 22937 + }, + { + "epoch": 1.2815878425566387, + "grad_norm": 0.4192838668823242, + "learning_rate": 0.00036076871358135365, + "loss": 0.3855, + "step": 22938 + }, + { + "epoch": 1.2816437131603207, + "grad_norm": 0.4708634316921234, + "learning_rate": 0.0003607406992380099, + "loss": 0.3921, + "step": 22939 + }, + { + "epoch": 1.2816995837640026, + "grad_norm": 0.5153896808624268, + "learning_rate": 0.00036071268489466606, + "loss": 0.4475, + "step": 22940 + }, + { + "epoch": 1.2817554543676843, + "grad_norm": 0.3758596181869507, + "learning_rate": 0.00036068467055132226, + "loss": 0.4432, + "step": 22941 + }, + { + "epoch": 1.2818113249713663, + "grad_norm": 7.463012218475342, + "learning_rate": 0.00036065665620797847, + "loss": 0.3491, + "step": 22942 + }, + { + "epoch": 1.2818671955750482, + "grad_norm": 1.014026403427124, + "learning_rate": 0.0003606286418646347, + "loss": 0.4772, + "step": 22943 + }, + { + "epoch": 1.28192306617873, + "grad_norm": 0.45451489090919495, + "learning_rate": 0.00036060062752129094, + "loss": 0.4922, + "step": 22944 + }, + { + "epoch": 1.281978936782412, + "grad_norm": 0.990219235420227, + "learning_rate": 0.0003605726131779471, + "loss": 0.4023, + "step": 22945 + }, + { + "epoch": 1.2820348073860939, + "grad_norm": 1.125373125076294, + "learning_rate": 0.00036054459883460335, + "loss": 0.5038, + "step": 22946 + }, + { + "epoch": 1.2820906779897756, + "grad_norm": 0.49549588561058044, + "learning_rate": 0.0003605165844912595, + "loss": 0.3453, + "step": 22947 + }, + { + "epoch": 1.2821465485934576, + "grad_norm": 0.6584512591362, + "learning_rate": 0.00036048857014791576, + "loss": 0.3222, + "step": 22948 + }, + { + "epoch": 1.2822024191971395, + "grad_norm": 0.46356913447380066, + "learning_rate": 0.0003604605558045719, + "loss": 0.4889, + "step": 22949 + }, + { + "epoch": 1.2822582898008212, + "grad_norm": 0.577289879322052, + "learning_rate": 0.00036043254146122817, + "loss": 0.3591, + "step": 22950 + }, + { + "epoch": 1.2823141604045032, + "grad_norm": 0.40623170137405396, + "learning_rate": 0.0003604045271178844, + "loss": 0.3764, + "step": 22951 + }, + { + "epoch": 1.2823700310081851, + "grad_norm": 0.4693724811077118, + "learning_rate": 0.0003603765127745406, + "loss": 0.3788, + "step": 22952 + }, + { + "epoch": 1.2824259016118669, + "grad_norm": 0.4486573338508606, + "learning_rate": 0.0003603484984311968, + "loss": 0.4571, + "step": 22953 + }, + { + "epoch": 1.2824817722155488, + "grad_norm": 0.5096392035484314, + "learning_rate": 0.000360320484087853, + "loss": 0.5556, + "step": 22954 + }, + { + "epoch": 1.2825376428192308, + "grad_norm": 0.5890207290649414, + "learning_rate": 0.0003602924697445092, + "loss": 0.5523, + "step": 22955 + }, + { + "epoch": 1.2825935134229125, + "grad_norm": 0.7578632831573486, + "learning_rate": 0.0003602644554011654, + "loss": 0.3972, + "step": 22956 + }, + { + "epoch": 1.2826493840265945, + "grad_norm": 0.5250567197799683, + "learning_rate": 0.0003602364410578216, + "loss": 0.494, + "step": 22957 + }, + { + "epoch": 1.2827052546302764, + "grad_norm": 0.6536781191825867, + "learning_rate": 0.0003602084267144778, + "loss": 0.6117, + "step": 22958 + }, + { + "epoch": 1.2827611252339581, + "grad_norm": 0.4629703760147095, + "learning_rate": 0.000360180412371134, + "loss": 0.4557, + "step": 22959 + }, + { + "epoch": 1.28281699583764, + "grad_norm": 0.393717885017395, + "learning_rate": 0.0003601523980277903, + "loss": 0.3761, + "step": 22960 + }, + { + "epoch": 1.282872866441322, + "grad_norm": 0.43894869089126587, + "learning_rate": 0.00036012438368444643, + "loss": 0.3665, + "step": 22961 + }, + { + "epoch": 1.2829287370450038, + "grad_norm": 0.5668302178382874, + "learning_rate": 0.00036009636934110264, + "loss": 0.3839, + "step": 22962 + }, + { + "epoch": 1.2829846076486857, + "grad_norm": 0.6826529502868652, + "learning_rate": 0.00036006835499775885, + "loss": 0.4587, + "step": 22963 + }, + { + "epoch": 1.2830404782523674, + "grad_norm": 0.3918973505496979, + "learning_rate": 0.00036004034065441505, + "loss": 0.3417, + "step": 22964 + }, + { + "epoch": 1.2830963488560494, + "grad_norm": 0.43852823972702026, + "learning_rate": 0.00036001232631107126, + "loss": 0.4468, + "step": 22965 + }, + { + "epoch": 1.2831522194597311, + "grad_norm": 0.45075172185897827, + "learning_rate": 0.00035998431196772746, + "loss": 0.405, + "step": 22966 + }, + { + "epoch": 1.283208090063413, + "grad_norm": 0.3649675250053406, + "learning_rate": 0.0003599562976243837, + "loss": 0.4324, + "step": 22967 + }, + { + "epoch": 1.283263960667095, + "grad_norm": 0.5760963559150696, + "learning_rate": 0.0003599282832810399, + "loss": 0.4347, + "step": 22968 + }, + { + "epoch": 1.2833198312707768, + "grad_norm": 0.46042895317077637, + "learning_rate": 0.00035990026893769613, + "loss": 0.5303, + "step": 22969 + }, + { + "epoch": 1.2833757018744587, + "grad_norm": 0.9572412371635437, + "learning_rate": 0.0003598722545943523, + "loss": 0.7174, + "step": 22970 + }, + { + "epoch": 1.2834315724781407, + "grad_norm": 0.546775758266449, + "learning_rate": 0.00035984424025100855, + "loss": 0.4261, + "step": 22971 + }, + { + "epoch": 1.2834874430818224, + "grad_norm": 0.3529961109161377, + "learning_rate": 0.0003598162259076647, + "loss": 0.4857, + "step": 22972 + }, + { + "epoch": 1.2835433136855043, + "grad_norm": 0.9587299823760986, + "learning_rate": 0.00035978821156432096, + "loss": 0.4046, + "step": 22973 + }, + { + "epoch": 1.2835991842891863, + "grad_norm": 0.5276785492897034, + "learning_rate": 0.0003597601972209771, + "loss": 0.4362, + "step": 22974 + }, + { + "epoch": 1.283655054892868, + "grad_norm": 0.6729061007499695, + "learning_rate": 0.00035973218287763337, + "loss": 0.4333, + "step": 22975 + }, + { + "epoch": 1.28371092549655, + "grad_norm": 0.9796398878097534, + "learning_rate": 0.0003597041685342896, + "loss": 0.424, + "step": 22976 + }, + { + "epoch": 1.283766796100232, + "grad_norm": 0.4914032220840454, + "learning_rate": 0.0003596761541909458, + "loss": 0.4594, + "step": 22977 + }, + { + "epoch": 1.2838226667039137, + "grad_norm": 0.3497810363769531, + "learning_rate": 0.000359648139847602, + "loss": 0.4513, + "step": 22978 + }, + { + "epoch": 1.2838785373075956, + "grad_norm": 0.55415278673172, + "learning_rate": 0.0003596201255042582, + "loss": 0.4738, + "step": 22979 + }, + { + "epoch": 1.2839344079112776, + "grad_norm": 0.7781448364257812, + "learning_rate": 0.0003595921111609144, + "loss": 0.53, + "step": 22980 + }, + { + "epoch": 1.2839902785149593, + "grad_norm": 0.5305830240249634, + "learning_rate": 0.0003595640968175706, + "loss": 0.4287, + "step": 22981 + }, + { + "epoch": 1.2840461491186412, + "grad_norm": 0.389861136674881, + "learning_rate": 0.0003595360824742268, + "loss": 0.391, + "step": 22982 + }, + { + "epoch": 1.2841020197223232, + "grad_norm": 0.43160179257392883, + "learning_rate": 0.000359508068130883, + "loss": 0.3338, + "step": 22983 + }, + { + "epoch": 1.284157890326005, + "grad_norm": 0.3809400200843811, + "learning_rate": 0.0003594800537875392, + "loss": 0.4391, + "step": 22984 + }, + { + "epoch": 1.2842137609296869, + "grad_norm": 0.5789852738380432, + "learning_rate": 0.00035945203944419543, + "loss": 0.4714, + "step": 22985 + }, + { + "epoch": 1.2842696315333688, + "grad_norm": 0.6951218843460083, + "learning_rate": 0.00035942402510085163, + "loss": 0.3767, + "step": 22986 + }, + { + "epoch": 1.2843255021370505, + "grad_norm": 0.5226820707321167, + "learning_rate": 0.00035939601075750784, + "loss": 0.42, + "step": 22987 + }, + { + "epoch": 1.2843813727407325, + "grad_norm": 0.44359713792800903, + "learning_rate": 0.00035936799641416405, + "loss": 0.4459, + "step": 22988 + }, + { + "epoch": 1.2844372433444144, + "grad_norm": 0.35391196608543396, + "learning_rate": 0.00035933998207082025, + "loss": 0.3015, + "step": 22989 + }, + { + "epoch": 1.2844931139480962, + "grad_norm": 0.43604540824890137, + "learning_rate": 0.0003593119677274765, + "loss": 0.3975, + "step": 22990 + }, + { + "epoch": 1.2845489845517781, + "grad_norm": 1.9722318649291992, + "learning_rate": 0.00035928395338413266, + "loss": 0.6036, + "step": 22991 + }, + { + "epoch": 1.28460485515546, + "grad_norm": 0.3947241008281708, + "learning_rate": 0.0003592559390407889, + "loss": 0.3881, + "step": 22992 + }, + { + "epoch": 1.2846607257591418, + "grad_norm": 0.35213130712509155, + "learning_rate": 0.0003592279246974451, + "loss": 0.459, + "step": 22993 + }, + { + "epoch": 1.2847165963628238, + "grad_norm": 0.4469945728778839, + "learning_rate": 0.00035919991035410133, + "loss": 0.3962, + "step": 22994 + }, + { + "epoch": 1.2847724669665057, + "grad_norm": 2.356968641281128, + "learning_rate": 0.0003591718960107575, + "loss": 0.4577, + "step": 22995 + }, + { + "epoch": 1.2848283375701874, + "grad_norm": 0.3564368188381195, + "learning_rate": 0.00035914388166741375, + "loss": 0.3842, + "step": 22996 + }, + { + "epoch": 1.2848842081738694, + "grad_norm": 0.42884746193885803, + "learning_rate": 0.0003591158673240699, + "loss": 0.4605, + "step": 22997 + }, + { + "epoch": 1.2849400787775511, + "grad_norm": 1.463221788406372, + "learning_rate": 0.00035908785298072616, + "loss": 0.4992, + "step": 22998 + }, + { + "epoch": 1.284995949381233, + "grad_norm": 0.5381494760513306, + "learning_rate": 0.00035905983863738236, + "loss": 0.4326, + "step": 22999 + }, + { + "epoch": 1.2850518199849148, + "grad_norm": 0.425192654132843, + "learning_rate": 0.00035903182429403857, + "loss": 0.3502, + "step": 23000 + }, + { + "epoch": 1.2850518199849148, + "eval_cer": 0.08668019595648806, + "eval_loss": 0.3252941966056824, + "eval_runtime": 56.1347, + "eval_samples_per_second": 80.841, + "eval_steps_per_second": 5.059, + "eval_wer": 0.34630317588894166, + "step": 23000 + }, + { + "epoch": 1.2851076905885968, + "grad_norm": 0.4819425642490387, + "learning_rate": 0.0003590038099506948, + "loss": 0.4153, + "step": 23001 + }, + { + "epoch": 1.2851635611922787, + "grad_norm": 0.6484516859054565, + "learning_rate": 0.000358975795607351, + "loss": 0.4181, + "step": 23002 + }, + { + "epoch": 1.2852194317959604, + "grad_norm": 1.3630108833312988, + "learning_rate": 0.0003589477812640072, + "loss": 0.4137, + "step": 23003 + }, + { + "epoch": 1.2852753023996424, + "grad_norm": 0.5045841336250305, + "learning_rate": 0.00035891976692066334, + "loss": 0.3388, + "step": 23004 + }, + { + "epoch": 1.2853311730033243, + "grad_norm": 1.1644790172576904, + "learning_rate": 0.0003588917525773196, + "loss": 0.4074, + "step": 23005 + }, + { + "epoch": 1.285387043607006, + "grad_norm": 0.5454564690589905, + "learning_rate": 0.0003588637382339758, + "loss": 0.3623, + "step": 23006 + }, + { + "epoch": 1.285442914210688, + "grad_norm": 1.9226794242858887, + "learning_rate": 0.000358835723890632, + "loss": 0.4026, + "step": 23007 + }, + { + "epoch": 1.28549878481437, + "grad_norm": 0.34287068247795105, + "learning_rate": 0.0003588077095472882, + "loss": 0.4162, + "step": 23008 + }, + { + "epoch": 1.2855546554180517, + "grad_norm": 0.7114571928977966, + "learning_rate": 0.0003587796952039444, + "loss": 0.374, + "step": 23009 + }, + { + "epoch": 1.2856105260217336, + "grad_norm": 0.4381263256072998, + "learning_rate": 0.00035875168086060063, + "loss": 0.3794, + "step": 23010 + }, + { + "epoch": 1.2856663966254156, + "grad_norm": 0.3431432843208313, + "learning_rate": 0.00035872366651725683, + "loss": 0.3042, + "step": 23011 + }, + { + "epoch": 1.2857222672290973, + "grad_norm": 0.4632206857204437, + "learning_rate": 0.00035869565217391304, + "loss": 0.4928, + "step": 23012 + }, + { + "epoch": 1.2857781378327793, + "grad_norm": 1.818110704421997, + "learning_rate": 0.00035866763783056924, + "loss": 0.4062, + "step": 23013 + }, + { + "epoch": 1.2858340084364612, + "grad_norm": 2.261815309524536, + "learning_rate": 0.00035863962348722545, + "loss": 0.3881, + "step": 23014 + }, + { + "epoch": 1.285889879040143, + "grad_norm": 0.37469515204429626, + "learning_rate": 0.0003586116091438817, + "loss": 0.4367, + "step": 23015 + }, + { + "epoch": 1.285945749643825, + "grad_norm": 0.6317456364631653, + "learning_rate": 0.00035858359480053786, + "loss": 0.6243, + "step": 23016 + }, + { + "epoch": 1.2860016202475069, + "grad_norm": 1.465313196182251, + "learning_rate": 0.0003585555804571941, + "loss": 0.3868, + "step": 23017 + }, + { + "epoch": 1.2860574908511886, + "grad_norm": 0.322623074054718, + "learning_rate": 0.0003585275661138503, + "loss": 0.3027, + "step": 23018 + }, + { + "epoch": 1.2861133614548705, + "grad_norm": 0.4687741994857788, + "learning_rate": 0.00035849955177050653, + "loss": 0.6284, + "step": 23019 + }, + { + "epoch": 1.2861692320585525, + "grad_norm": 0.3998739421367645, + "learning_rate": 0.0003584715374271627, + "loss": 0.5045, + "step": 23020 + }, + { + "epoch": 1.2862251026622342, + "grad_norm": 0.41404494643211365, + "learning_rate": 0.00035844352308381895, + "loss": 0.4386, + "step": 23021 + }, + { + "epoch": 1.2862809732659162, + "grad_norm": 0.8908786177635193, + "learning_rate": 0.00035841550874047515, + "loss": 0.3896, + "step": 23022 + }, + { + "epoch": 1.2863368438695981, + "grad_norm": 0.8986330628395081, + "learning_rate": 0.00035838749439713136, + "loss": 0.43, + "step": 23023 + }, + { + "epoch": 1.2863927144732799, + "grad_norm": 0.5863031148910522, + "learning_rate": 0.00035835948005378756, + "loss": 0.5025, + "step": 23024 + }, + { + "epoch": 1.2864485850769618, + "grad_norm": 0.6159592270851135, + "learning_rate": 0.0003583314657104437, + "loss": 0.4712, + "step": 23025 + }, + { + "epoch": 1.2865044556806438, + "grad_norm": 1.7602349519729614, + "learning_rate": 0.0003583034513671, + "loss": 0.4929, + "step": 23026 + }, + { + "epoch": 1.2865603262843255, + "grad_norm": 2.146481513977051, + "learning_rate": 0.0003582754370237561, + "loss": 0.4189, + "step": 23027 + }, + { + "epoch": 1.2866161968880074, + "grad_norm": 0.34621232748031616, + "learning_rate": 0.0003582474226804124, + "loss": 0.3928, + "step": 23028 + }, + { + "epoch": 1.2866720674916894, + "grad_norm": 12.673985481262207, + "learning_rate": 0.00035821940833706854, + "loss": 0.4293, + "step": 23029 + }, + { + "epoch": 1.2867279380953711, + "grad_norm": 0.40274500846862793, + "learning_rate": 0.0003581913939937248, + "loss": 0.538, + "step": 23030 + }, + { + "epoch": 1.286783808699053, + "grad_norm": 0.42598047852516174, + "learning_rate": 0.000358163379650381, + "loss": 0.3911, + "step": 23031 + }, + { + "epoch": 1.2868396793027348, + "grad_norm": 0.5978063344955444, + "learning_rate": 0.0003581353653070372, + "loss": 0.338, + "step": 23032 + }, + { + "epoch": 1.2868955499064167, + "grad_norm": 0.40683767199516296, + "learning_rate": 0.0003581073509636934, + "loss": 0.3825, + "step": 23033 + }, + { + "epoch": 1.2869514205100985, + "grad_norm": 0.4001394510269165, + "learning_rate": 0.0003580793366203496, + "loss": 0.4067, + "step": 23034 + }, + { + "epoch": 1.2870072911137804, + "grad_norm": 0.5037068724632263, + "learning_rate": 0.00035805132227700583, + "loss": 0.459, + "step": 23035 + }, + { + "epoch": 1.2870631617174624, + "grad_norm": 0.4092312753200531, + "learning_rate": 0.00035802330793366203, + "loss": 0.4353, + "step": 23036 + }, + { + "epoch": 1.287119032321144, + "grad_norm": 1.0850716829299927, + "learning_rate": 0.00035799529359031824, + "loss": 0.3263, + "step": 23037 + }, + { + "epoch": 1.287174902924826, + "grad_norm": 0.3273458182811737, + "learning_rate": 0.0003579672792469745, + "loss": 0.426, + "step": 23038 + }, + { + "epoch": 1.287230773528508, + "grad_norm": 0.3510645627975464, + "learning_rate": 0.00035793926490363065, + "loss": 0.3659, + "step": 23039 + }, + { + "epoch": 1.2872866441321897, + "grad_norm": 0.5230023860931396, + "learning_rate": 0.0003579112505602869, + "loss": 0.4272, + "step": 23040 + }, + { + "epoch": 1.2873425147358717, + "grad_norm": 1.747653841972351, + "learning_rate": 0.00035788323621694306, + "loss": 0.5205, + "step": 23041 + }, + { + "epoch": 1.2873983853395536, + "grad_norm": 0.7223813533782959, + "learning_rate": 0.0003578552218735993, + "loss": 0.5074, + "step": 23042 + }, + { + "epoch": 1.2874542559432354, + "grad_norm": 0.9321101307868958, + "learning_rate": 0.0003578272075302555, + "loss": 0.4746, + "step": 23043 + }, + { + "epoch": 1.2875101265469173, + "grad_norm": 0.539239227771759, + "learning_rate": 0.00035779919318691173, + "loss": 0.442, + "step": 23044 + }, + { + "epoch": 1.2875659971505993, + "grad_norm": 0.6300764679908752, + "learning_rate": 0.00035777117884356794, + "loss": 0.4743, + "step": 23045 + }, + { + "epoch": 1.287621867754281, + "grad_norm": 0.3568425178527832, + "learning_rate": 0.00035774316450022415, + "loss": 0.3785, + "step": 23046 + }, + { + "epoch": 1.287677738357963, + "grad_norm": 0.39426496624946594, + "learning_rate": 0.00035771515015688035, + "loss": 0.5151, + "step": 23047 + }, + { + "epoch": 1.287733608961645, + "grad_norm": 1.2311253547668457, + "learning_rate": 0.0003576871358135365, + "loss": 0.4038, + "step": 23048 + }, + { + "epoch": 1.2877894795653266, + "grad_norm": 0.4609341621398926, + "learning_rate": 0.00035765912147019276, + "loss": 0.3175, + "step": 23049 + }, + { + "epoch": 1.2878453501690086, + "grad_norm": 0.7065279483795166, + "learning_rate": 0.0003576311071268489, + "loss": 0.3605, + "step": 23050 + }, + { + "epoch": 1.2879012207726905, + "grad_norm": 0.5047543048858643, + "learning_rate": 0.0003576030927835052, + "loss": 0.3772, + "step": 23051 + }, + { + "epoch": 1.2879570913763723, + "grad_norm": 0.4664541482925415, + "learning_rate": 0.0003575750784401613, + "loss": 0.5722, + "step": 23052 + }, + { + "epoch": 1.2880129619800542, + "grad_norm": 0.5888304114341736, + "learning_rate": 0.0003575470640968176, + "loss": 0.4238, + "step": 23053 + }, + { + "epoch": 1.2880688325837362, + "grad_norm": 0.7972829341888428, + "learning_rate": 0.0003575190497534738, + "loss": 0.4099, + "step": 23054 + }, + { + "epoch": 1.288124703187418, + "grad_norm": 0.610550045967102, + "learning_rate": 0.00035749103541013, + "loss": 0.6544, + "step": 23055 + }, + { + "epoch": 1.2881805737910998, + "grad_norm": 0.4781971871852875, + "learning_rate": 0.0003574630210667862, + "loss": 0.443, + "step": 23056 + }, + { + "epoch": 1.2882364443947818, + "grad_norm": 0.34081128239631653, + "learning_rate": 0.0003574350067234424, + "loss": 0.3835, + "step": 23057 + }, + { + "epoch": 1.2882923149984635, + "grad_norm": 0.539320707321167, + "learning_rate": 0.0003574069923800986, + "loss": 0.3871, + "step": 23058 + }, + { + "epoch": 1.2883481856021455, + "grad_norm": 0.6793875694274902, + "learning_rate": 0.0003573789780367548, + "loss": 0.6569, + "step": 23059 + }, + { + "epoch": 1.2884040562058274, + "grad_norm": 0.42065057158470154, + "learning_rate": 0.000357350963693411, + "loss": 0.6447, + "step": 23060 + }, + { + "epoch": 1.2884599268095092, + "grad_norm": 0.4125975966453552, + "learning_rate": 0.0003573229493500673, + "loss": 0.3824, + "step": 23061 + }, + { + "epoch": 1.288515797413191, + "grad_norm": 0.541318953037262, + "learning_rate": 0.00035729493500672344, + "loss": 0.4118, + "step": 23062 + }, + { + "epoch": 1.2885716680168728, + "grad_norm": 0.44438448548316956, + "learning_rate": 0.0003572669206633797, + "loss": 0.4943, + "step": 23063 + }, + { + "epoch": 1.2886275386205548, + "grad_norm": 0.5515915155410767, + "learning_rate": 0.00035723890632003585, + "loss": 0.457, + "step": 23064 + }, + { + "epoch": 1.2886834092242367, + "grad_norm": 0.7269308567047119, + "learning_rate": 0.0003572108919766921, + "loss": 0.5614, + "step": 23065 + }, + { + "epoch": 1.2887392798279185, + "grad_norm": 0.47067710757255554, + "learning_rate": 0.00035718287763334826, + "loss": 0.4512, + "step": 23066 + }, + { + "epoch": 1.2887951504316004, + "grad_norm": 0.5844859480857849, + "learning_rate": 0.0003571548632900045, + "loss": 0.4875, + "step": 23067 + }, + { + "epoch": 1.2888510210352822, + "grad_norm": 0.48336780071258545, + "learning_rate": 0.0003571268489466607, + "loss": 0.447, + "step": 23068 + }, + { + "epoch": 1.288906891638964, + "grad_norm": 0.47372961044311523, + "learning_rate": 0.0003570988346033169, + "loss": 0.3642, + "step": 23069 + }, + { + "epoch": 1.288962762242646, + "grad_norm": 1.6165481805801392, + "learning_rate": 0.00035707082025997314, + "loss": 0.6212, + "step": 23070 + }, + { + "epoch": 1.2890186328463278, + "grad_norm": 0.3669467866420746, + "learning_rate": 0.0003570428059166293, + "loss": 0.4019, + "step": 23071 + }, + { + "epoch": 1.2890745034500097, + "grad_norm": 0.6499282121658325, + "learning_rate": 0.00035701479157328555, + "loss": 0.439, + "step": 23072 + }, + { + "epoch": 1.2891303740536917, + "grad_norm": 0.48386669158935547, + "learning_rate": 0.0003569867772299417, + "loss": 0.3333, + "step": 23073 + }, + { + "epoch": 1.2891862446573734, + "grad_norm": 0.4446118175983429, + "learning_rate": 0.00035695876288659796, + "loss": 0.4128, + "step": 23074 + }, + { + "epoch": 1.2892421152610554, + "grad_norm": 0.4326138496398926, + "learning_rate": 0.0003569307485432541, + "loss": 0.4419, + "step": 23075 + }, + { + "epoch": 1.2892979858647373, + "grad_norm": 0.5244967937469482, + "learning_rate": 0.0003569027341999104, + "loss": 0.5102, + "step": 23076 + }, + { + "epoch": 1.289353856468419, + "grad_norm": 0.6697872877120972, + "learning_rate": 0.0003568747198565666, + "loss": 0.4321, + "step": 23077 + }, + { + "epoch": 1.289409727072101, + "grad_norm": 0.618821918964386, + "learning_rate": 0.0003568467055132228, + "loss": 0.3336, + "step": 23078 + }, + { + "epoch": 1.289465597675783, + "grad_norm": 0.539172887802124, + "learning_rate": 0.000356818691169879, + "loss": 0.4035, + "step": 23079 + }, + { + "epoch": 1.2895214682794647, + "grad_norm": 2.247129201889038, + "learning_rate": 0.0003567906768265352, + "loss": 0.3594, + "step": 23080 + }, + { + "epoch": 1.2895773388831466, + "grad_norm": 0.6245310306549072, + "learning_rate": 0.0003567626624831914, + "loss": 0.3908, + "step": 23081 + }, + { + "epoch": 1.2896332094868286, + "grad_norm": 1.383185863494873, + "learning_rate": 0.0003567346481398476, + "loss": 0.4305, + "step": 23082 + }, + { + "epoch": 1.2896890800905103, + "grad_norm": 0.6771489381790161, + "learning_rate": 0.0003567066337965038, + "loss": 0.4307, + "step": 23083 + }, + { + "epoch": 1.2897449506941923, + "grad_norm": 0.8194113969802856, + "learning_rate": 0.00035667861945316, + "loss": 0.4276, + "step": 23084 + }, + { + "epoch": 1.2898008212978742, + "grad_norm": 0.9687812924385071, + "learning_rate": 0.0003566506051098162, + "loss": 0.3933, + "step": 23085 + }, + { + "epoch": 1.289856691901556, + "grad_norm": 0.6105136275291443, + "learning_rate": 0.0003566225907664725, + "loss": 0.5755, + "step": 23086 + }, + { + "epoch": 1.289912562505238, + "grad_norm": 0.4947996139526367, + "learning_rate": 0.00035659457642312864, + "loss": 0.4626, + "step": 23087 + }, + { + "epoch": 1.2899684331089198, + "grad_norm": 0.43575364351272583, + "learning_rate": 0.0003565665620797849, + "loss": 0.4763, + "step": 23088 + }, + { + "epoch": 1.2900243037126016, + "grad_norm": 0.41856786608695984, + "learning_rate": 0.00035653854773644105, + "loss": 0.4833, + "step": 23089 + }, + { + "epoch": 1.2900801743162835, + "grad_norm": 0.3766491115093231, + "learning_rate": 0.00035651053339309726, + "loss": 0.477, + "step": 23090 + }, + { + "epoch": 1.2901360449199655, + "grad_norm": 0.3377707004547119, + "learning_rate": 0.00035648251904975346, + "loss": 0.4229, + "step": 23091 + }, + { + "epoch": 1.2901919155236472, + "grad_norm": 0.4713912904262543, + "learning_rate": 0.00035645450470640967, + "loss": 0.4116, + "step": 23092 + }, + { + "epoch": 1.2902477861273292, + "grad_norm": 0.7103260159492493, + "learning_rate": 0.0003564264903630659, + "loss": 0.4046, + "step": 23093 + }, + { + "epoch": 1.290303656731011, + "grad_norm": 0.4933820962905884, + "learning_rate": 0.0003563984760197221, + "loss": 0.4711, + "step": 23094 + }, + { + "epoch": 1.2903595273346928, + "grad_norm": 0.4605441689491272, + "learning_rate": 0.00035637046167637834, + "loss": 0.4638, + "step": 23095 + }, + { + "epoch": 1.2904153979383748, + "grad_norm": 0.3449331521987915, + "learning_rate": 0.0003563424473330345, + "loss": 0.3619, + "step": 23096 + }, + { + "epoch": 1.2904712685420565, + "grad_norm": 0.4575003683567047, + "learning_rate": 0.00035631443298969075, + "loss": 0.4737, + "step": 23097 + }, + { + "epoch": 1.2905271391457385, + "grad_norm": 0.5318312644958496, + "learning_rate": 0.0003562864186463469, + "loss": 0.4354, + "step": 23098 + }, + { + "epoch": 1.2905830097494204, + "grad_norm": 0.5086705088615417, + "learning_rate": 0.00035625840430300316, + "loss": 0.425, + "step": 23099 + }, + { + "epoch": 1.2906388803531021, + "grad_norm": 0.4708501696586609, + "learning_rate": 0.0003562303899596593, + "loss": 0.3928, + "step": 23100 + }, + { + "epoch": 1.290694750956784, + "grad_norm": 0.6724531650543213, + "learning_rate": 0.0003562023756163156, + "loss": 0.3729, + "step": 23101 + }, + { + "epoch": 1.2907506215604658, + "grad_norm": 0.5694445967674255, + "learning_rate": 0.0003561743612729718, + "loss": 0.4251, + "step": 23102 + }, + { + "epoch": 1.2908064921641478, + "grad_norm": 0.713492751121521, + "learning_rate": 0.000356146346929628, + "loss": 0.6653, + "step": 23103 + }, + { + "epoch": 1.2908623627678297, + "grad_norm": 0.5420005321502686, + "learning_rate": 0.0003561183325862842, + "loss": 0.4243, + "step": 23104 + }, + { + "epoch": 1.2909182333715115, + "grad_norm": 0.45595982670783997, + "learning_rate": 0.0003560903182429404, + "loss": 0.3692, + "step": 23105 + }, + { + "epoch": 1.2909741039751934, + "grad_norm": 0.4193773865699768, + "learning_rate": 0.0003560623038995966, + "loss": 0.388, + "step": 23106 + }, + { + "epoch": 1.2910299745788754, + "grad_norm": 0.8126273155212402, + "learning_rate": 0.0003560342895562528, + "loss": 0.4034, + "step": 23107 + }, + { + "epoch": 1.291085845182557, + "grad_norm": 2.4831559658050537, + "learning_rate": 0.000356006275212909, + "loss": 0.3715, + "step": 23108 + }, + { + "epoch": 1.291141715786239, + "grad_norm": 0.9919785261154175, + "learning_rate": 0.0003559782608695653, + "loss": 0.4939, + "step": 23109 + }, + { + "epoch": 1.291197586389921, + "grad_norm": 0.4920808970928192, + "learning_rate": 0.0003559502465262214, + "loss": 0.4172, + "step": 23110 + }, + { + "epoch": 1.2912534569936027, + "grad_norm": 0.7681438326835632, + "learning_rate": 0.00035592223218287763, + "loss": 0.5248, + "step": 23111 + }, + { + "epoch": 1.2913093275972847, + "grad_norm": 0.3199984133243561, + "learning_rate": 0.00035589421783953384, + "loss": 0.3632, + "step": 23112 + }, + { + "epoch": 1.2913651982009666, + "grad_norm": 0.726560115814209, + "learning_rate": 0.00035586620349619004, + "loss": 0.5564, + "step": 23113 + }, + { + "epoch": 1.2914210688046484, + "grad_norm": 0.35271507501602173, + "learning_rate": 0.00035583818915284625, + "loss": 0.3949, + "step": 23114 + }, + { + "epoch": 1.2914769394083303, + "grad_norm": 0.3599671721458435, + "learning_rate": 0.00035581017480950245, + "loss": 0.4238, + "step": 23115 + }, + { + "epoch": 1.2915328100120123, + "grad_norm": 0.4480501711368561, + "learning_rate": 0.0003557821604661587, + "loss": 0.4134, + "step": 23116 + }, + { + "epoch": 1.291588680615694, + "grad_norm": 0.7704495787620544, + "learning_rate": 0.00035575414612281487, + "loss": 0.4021, + "step": 23117 + }, + { + "epoch": 1.291644551219376, + "grad_norm": 0.481635183095932, + "learning_rate": 0.0003557261317794711, + "loss": 0.3369, + "step": 23118 + }, + { + "epoch": 1.2917004218230579, + "grad_norm": 0.8766515851020813, + "learning_rate": 0.0003556981174361273, + "loss": 0.6764, + "step": 23119 + }, + { + "epoch": 1.2917562924267396, + "grad_norm": 0.49407103657722473, + "learning_rate": 0.00035567010309278354, + "loss": 0.3948, + "step": 23120 + }, + { + "epoch": 1.2918121630304216, + "grad_norm": 0.5386486053466797, + "learning_rate": 0.0003556420887494397, + "loss": 0.4156, + "step": 23121 + }, + { + "epoch": 1.2918680336341035, + "grad_norm": 0.5620720982551575, + "learning_rate": 0.00035561407440609595, + "loss": 0.5062, + "step": 23122 + }, + { + "epoch": 1.2919239042377852, + "grad_norm": 0.4731769263744354, + "learning_rate": 0.0003555860600627521, + "loss": 0.3793, + "step": 23123 + }, + { + "epoch": 1.2919797748414672, + "grad_norm": 9.283329010009766, + "learning_rate": 0.00035555804571940836, + "loss": 0.4641, + "step": 23124 + }, + { + "epoch": 1.2920356454451492, + "grad_norm": 0.38244524598121643, + "learning_rate": 0.00035553003137606457, + "loss": 0.4634, + "step": 23125 + }, + { + "epoch": 1.2920915160488309, + "grad_norm": 0.4314992129802704, + "learning_rate": 0.0003555020170327208, + "loss": 0.3224, + "step": 23126 + }, + { + "epoch": 1.2921473866525128, + "grad_norm": 1.039000391960144, + "learning_rate": 0.000355474002689377, + "loss": 0.4054, + "step": 23127 + }, + { + "epoch": 1.2922032572561948, + "grad_norm": 1.290700078010559, + "learning_rate": 0.0003554459883460332, + "loss": 0.3765, + "step": 23128 + }, + { + "epoch": 1.2922591278598765, + "grad_norm": 0.48200294375419617, + "learning_rate": 0.0003554179740026894, + "loss": 0.393, + "step": 23129 + }, + { + "epoch": 1.2923149984635585, + "grad_norm": 5.538875102996826, + "learning_rate": 0.0003553899596593456, + "loss": 0.4937, + "step": 23130 + }, + { + "epoch": 1.2923708690672402, + "grad_norm": 0.3698146939277649, + "learning_rate": 0.0003553619453160018, + "loss": 0.3296, + "step": 23131 + }, + { + "epoch": 1.2924267396709221, + "grad_norm": 0.46782779693603516, + "learning_rate": 0.000355333930972658, + "loss": 0.449, + "step": 23132 + }, + { + "epoch": 1.292482610274604, + "grad_norm": 0.41036975383758545, + "learning_rate": 0.0003553059166293142, + "loss": 0.4095, + "step": 23133 + }, + { + "epoch": 1.2925384808782858, + "grad_norm": 0.3133053779602051, + "learning_rate": 0.0003552779022859704, + "loss": 0.3391, + "step": 23134 + }, + { + "epoch": 1.2925943514819678, + "grad_norm": 2.796971559524536, + "learning_rate": 0.0003552498879426266, + "loss": 0.3976, + "step": 23135 + }, + { + "epoch": 1.2926502220856495, + "grad_norm": 0.5360254049301147, + "learning_rate": 0.00035522187359928283, + "loss": 0.4966, + "step": 23136 + }, + { + "epoch": 1.2927060926893315, + "grad_norm": 0.4420730173587799, + "learning_rate": 0.00035519385925593904, + "loss": 0.3552, + "step": 23137 + }, + { + "epoch": 1.2927619632930134, + "grad_norm": 1.405976414680481, + "learning_rate": 0.00035516584491259524, + "loss": 0.4385, + "step": 23138 + }, + { + "epoch": 1.2928178338966951, + "grad_norm": 1.168595790863037, + "learning_rate": 0.00035513783056925145, + "loss": 0.4589, + "step": 23139 + }, + { + "epoch": 1.292873704500377, + "grad_norm": 0.47112229466438293, + "learning_rate": 0.00035510981622590765, + "loss": 0.368, + "step": 23140 + }, + { + "epoch": 1.292929575104059, + "grad_norm": 0.46784985065460205, + "learning_rate": 0.0003550818018825639, + "loss": 0.5169, + "step": 23141 + }, + { + "epoch": 1.2929854457077408, + "grad_norm": 0.502106249332428, + "learning_rate": 0.00035505378753922007, + "loss": 0.4449, + "step": 23142 + }, + { + "epoch": 1.2930413163114227, + "grad_norm": 0.4208340048789978, + "learning_rate": 0.0003550257731958763, + "loss": 0.3281, + "step": 23143 + }, + { + "epoch": 1.2930971869151047, + "grad_norm": 0.32905805110931396, + "learning_rate": 0.0003549977588525325, + "loss": 0.3446, + "step": 23144 + }, + { + "epoch": 1.2931530575187864, + "grad_norm": 0.6764785647392273, + "learning_rate": 0.00035496974450918874, + "loss": 0.4029, + "step": 23145 + }, + { + "epoch": 1.2932089281224683, + "grad_norm": 0.501689076423645, + "learning_rate": 0.0003549417301658449, + "loss": 0.5031, + "step": 23146 + }, + { + "epoch": 1.2932647987261503, + "grad_norm": 0.43465685844421387, + "learning_rate": 0.00035491371582250115, + "loss": 0.4209, + "step": 23147 + }, + { + "epoch": 1.293320669329832, + "grad_norm": 0.5635067224502563, + "learning_rate": 0.00035488570147915736, + "loss": 0.6199, + "step": 23148 + }, + { + "epoch": 1.293376539933514, + "grad_norm": 0.40618962049484253, + "learning_rate": 0.00035485768713581356, + "loss": 0.4049, + "step": 23149 + }, + { + "epoch": 1.293432410537196, + "grad_norm": 5.6145853996276855, + "learning_rate": 0.00035482967279246977, + "loss": 0.4004, + "step": 23150 + }, + { + "epoch": 1.2934882811408777, + "grad_norm": 0.3984795808792114, + "learning_rate": 0.00035480165844912597, + "loss": 0.4454, + "step": 23151 + }, + { + "epoch": 1.2935441517445596, + "grad_norm": 2.4297304153442383, + "learning_rate": 0.0003547736441057822, + "loss": 0.463, + "step": 23152 + }, + { + "epoch": 1.2936000223482416, + "grad_norm": 0.5755323767662048, + "learning_rate": 0.00035474562976243833, + "loss": 0.4062, + "step": 23153 + }, + { + "epoch": 1.2936558929519233, + "grad_norm": 0.32849884033203125, + "learning_rate": 0.0003547176154190946, + "loss": 0.3191, + "step": 23154 + }, + { + "epoch": 1.2937117635556052, + "grad_norm": 0.4215359091758728, + "learning_rate": 0.00035468960107575074, + "loss": 0.3321, + "step": 23155 + }, + { + "epoch": 1.2937676341592872, + "grad_norm": 0.48262619972229004, + "learning_rate": 0.000354661586732407, + "loss": 0.3733, + "step": 23156 + }, + { + "epoch": 1.293823504762969, + "grad_norm": 0.3368386924266815, + "learning_rate": 0.0003546335723890632, + "loss": 0.3884, + "step": 23157 + }, + { + "epoch": 1.2938793753666509, + "grad_norm": 0.3866458833217621, + "learning_rate": 0.0003546055580457194, + "loss": 0.4011, + "step": 23158 + }, + { + "epoch": 1.2939352459703328, + "grad_norm": 0.34830617904663086, + "learning_rate": 0.0003545775437023756, + "loss": 0.3227, + "step": 23159 + }, + { + "epoch": 1.2939911165740146, + "grad_norm": 0.5047380924224854, + "learning_rate": 0.0003545495293590318, + "loss": 0.5739, + "step": 23160 + }, + { + "epoch": 1.2940469871776965, + "grad_norm": 1.2399473190307617, + "learning_rate": 0.00035452151501568803, + "loss": 0.3562, + "step": 23161 + }, + { + "epoch": 1.2941028577813785, + "grad_norm": 0.9459379315376282, + "learning_rate": 0.00035449350067234424, + "loss": 0.3605, + "step": 23162 + }, + { + "epoch": 1.2941587283850602, + "grad_norm": 0.5703078508377075, + "learning_rate": 0.00035446548632900044, + "loss": 0.4788, + "step": 23163 + }, + { + "epoch": 1.2942145989887421, + "grad_norm": 0.37790876626968384, + "learning_rate": 0.0003544374719856567, + "loss": 0.3355, + "step": 23164 + }, + { + "epoch": 1.2942704695924239, + "grad_norm": 0.5025569796562195, + "learning_rate": 0.00035440945764231285, + "loss": 0.4815, + "step": 23165 + }, + { + "epoch": 1.2943263401961058, + "grad_norm": 0.4827951490879059, + "learning_rate": 0.0003543814432989691, + "loss": 0.4425, + "step": 23166 + }, + { + "epoch": 1.2943822107997878, + "grad_norm": 0.4027691185474396, + "learning_rate": 0.00035435342895562527, + "loss": 0.3516, + "step": 23167 + }, + { + "epoch": 1.2944380814034695, + "grad_norm": 0.3557083010673523, + "learning_rate": 0.0003543254146122815, + "loss": 0.3315, + "step": 23168 + }, + { + "epoch": 1.2944939520071514, + "grad_norm": 0.40028005838394165, + "learning_rate": 0.0003542974002689377, + "loss": 0.3514, + "step": 23169 + }, + { + "epoch": 1.2945498226108332, + "grad_norm": 0.4609617590904236, + "learning_rate": 0.00035426938592559394, + "loss": 0.4253, + "step": 23170 + }, + { + "epoch": 1.2946056932145151, + "grad_norm": 0.5744693875312805, + "learning_rate": 0.0003542413715822501, + "loss": 0.4704, + "step": 23171 + }, + { + "epoch": 1.294661563818197, + "grad_norm": 0.39907899498939514, + "learning_rate": 0.00035421335723890635, + "loss": 0.3711, + "step": 23172 + }, + { + "epoch": 1.2947174344218788, + "grad_norm": 0.4979357123374939, + "learning_rate": 0.00035418534289556255, + "loss": 0.4568, + "step": 23173 + }, + { + "epoch": 1.2947733050255608, + "grad_norm": 0.7419961094856262, + "learning_rate": 0.0003541573285522187, + "loss": 0.5009, + "step": 23174 + }, + { + "epoch": 1.2948291756292427, + "grad_norm": 1.0896360874176025, + "learning_rate": 0.00035412931420887497, + "loss": 0.4408, + "step": 23175 + }, + { + "epoch": 1.2948850462329244, + "grad_norm": 0.4767206013202667, + "learning_rate": 0.0003541012998655311, + "loss": 0.4047, + "step": 23176 + }, + { + "epoch": 1.2949409168366064, + "grad_norm": 0.41150835156440735, + "learning_rate": 0.0003540732855221874, + "loss": 0.4377, + "step": 23177 + }, + { + "epoch": 1.2949967874402883, + "grad_norm": 0.5455384850502014, + "learning_rate": 0.00035404527117884353, + "loss": 0.4679, + "step": 23178 + }, + { + "epoch": 1.29505265804397, + "grad_norm": 1.1818792819976807, + "learning_rate": 0.0003540172568354998, + "loss": 0.4656, + "step": 23179 + }, + { + "epoch": 1.295108528647652, + "grad_norm": 0.6187611222267151, + "learning_rate": 0.000353989242492156, + "loss": 0.3762, + "step": 23180 + }, + { + "epoch": 1.295164399251334, + "grad_norm": 0.5570297241210938, + "learning_rate": 0.0003539612281488122, + "loss": 0.4377, + "step": 23181 + }, + { + "epoch": 1.2952202698550157, + "grad_norm": 0.9963793158531189, + "learning_rate": 0.0003539332138054684, + "loss": 0.3862, + "step": 23182 + }, + { + "epoch": 1.2952761404586977, + "grad_norm": 0.34317824244499207, + "learning_rate": 0.0003539051994621246, + "loss": 0.3813, + "step": 23183 + }, + { + "epoch": 1.2953320110623796, + "grad_norm": 0.8140710592269897, + "learning_rate": 0.0003538771851187808, + "loss": 0.458, + "step": 23184 + }, + { + "epoch": 1.2953878816660613, + "grad_norm": 0.4671080708503723, + "learning_rate": 0.000353849170775437, + "loss": 0.3245, + "step": 23185 + }, + { + "epoch": 1.2954437522697433, + "grad_norm": 0.9361037611961365, + "learning_rate": 0.00035382115643209323, + "loss": 0.3523, + "step": 23186 + }, + { + "epoch": 1.2954996228734252, + "grad_norm": 0.640473484992981, + "learning_rate": 0.0003537931420887495, + "loss": 0.3665, + "step": 23187 + }, + { + "epoch": 1.295555493477107, + "grad_norm": 1.6778714656829834, + "learning_rate": 0.00035376512774540564, + "loss": 0.3824, + "step": 23188 + }, + { + "epoch": 1.295611364080789, + "grad_norm": 1.4423441886901855, + "learning_rate": 0.0003537371134020619, + "loss": 0.3808, + "step": 23189 + }, + { + "epoch": 1.2956672346844709, + "grad_norm": 0.43032360076904297, + "learning_rate": 0.00035370909905871805, + "loss": 0.355, + "step": 23190 + }, + { + "epoch": 1.2957231052881526, + "grad_norm": 0.5641056299209595, + "learning_rate": 0.0003536810847153743, + "loss": 0.3324, + "step": 23191 + }, + { + "epoch": 1.2957789758918346, + "grad_norm": 0.35325145721435547, + "learning_rate": 0.00035365307037203047, + "loss": 0.3351, + "step": 23192 + }, + { + "epoch": 1.2958348464955165, + "grad_norm": 0.5255812406539917, + "learning_rate": 0.0003536250560286867, + "loss": 0.3587, + "step": 23193 + }, + { + "epoch": 1.2958907170991982, + "grad_norm": 0.3802383244037628, + "learning_rate": 0.0003535970416853429, + "loss": 0.3598, + "step": 23194 + }, + { + "epoch": 1.2959465877028802, + "grad_norm": 0.9851107001304626, + "learning_rate": 0.0003535690273419991, + "loss": 0.6021, + "step": 23195 + }, + { + "epoch": 1.2960024583065621, + "grad_norm": 3.2204973697662354, + "learning_rate": 0.00035354101299865534, + "loss": 0.3693, + "step": 23196 + }, + { + "epoch": 1.2960583289102439, + "grad_norm": 0.794070303440094, + "learning_rate": 0.0003535129986553115, + "loss": 0.4132, + "step": 23197 + }, + { + "epoch": 1.2961141995139258, + "grad_norm": 2.3544716835021973, + "learning_rate": 0.00035348498431196775, + "loss": 0.3825, + "step": 23198 + }, + { + "epoch": 1.2961700701176075, + "grad_norm": 2.1471574306488037, + "learning_rate": 0.0003534569699686239, + "loss": 0.3855, + "step": 23199 + }, + { + "epoch": 1.2962259407212895, + "grad_norm": 0.4185112714767456, + "learning_rate": 0.00035342895562528017, + "loss": 0.3977, + "step": 23200 + }, + { + "epoch": 1.2962818113249714, + "grad_norm": 0.3288794457912445, + "learning_rate": 0.0003534009412819363, + "loss": 0.3275, + "step": 23201 + }, + { + "epoch": 1.2963376819286532, + "grad_norm": 0.37465837597846985, + "learning_rate": 0.0003533729269385926, + "loss": 0.3913, + "step": 23202 + }, + { + "epoch": 1.2963935525323351, + "grad_norm": 1.1794997453689575, + "learning_rate": 0.0003533449125952488, + "loss": 0.4342, + "step": 23203 + }, + { + "epoch": 1.2964494231360169, + "grad_norm": 0.5842931270599365, + "learning_rate": 0.000353316898251905, + "loss": 0.5238, + "step": 23204 + }, + { + "epoch": 1.2965052937396988, + "grad_norm": 0.4003884792327881, + "learning_rate": 0.0003532888839085612, + "loss": 0.3128, + "step": 23205 + }, + { + "epoch": 1.2965611643433808, + "grad_norm": 1.1909399032592773, + "learning_rate": 0.0003532608695652174, + "loss": 0.4014, + "step": 23206 + }, + { + "epoch": 1.2966170349470625, + "grad_norm": 2.355034351348877, + "learning_rate": 0.0003532328552218736, + "loss": 0.4245, + "step": 23207 + }, + { + "epoch": 1.2966729055507444, + "grad_norm": 0.3448690176010132, + "learning_rate": 0.0003532048408785298, + "loss": 0.3184, + "step": 23208 + }, + { + "epoch": 1.2967287761544264, + "grad_norm": 0.4501776099205017, + "learning_rate": 0.000353176826535186, + "loss": 0.4845, + "step": 23209 + }, + { + "epoch": 1.2967846467581081, + "grad_norm": 0.5888832807540894, + "learning_rate": 0.0003531488121918422, + "loss": 0.4265, + "step": 23210 + }, + { + "epoch": 1.29684051736179, + "grad_norm": 0.3871147930622101, + "learning_rate": 0.00035312079784849843, + "loss": 0.3788, + "step": 23211 + }, + { + "epoch": 1.296896387965472, + "grad_norm": 1.8482367992401123, + "learning_rate": 0.0003530927835051547, + "loss": 0.5599, + "step": 23212 + }, + { + "epoch": 1.2969522585691537, + "grad_norm": 0.46024754643440247, + "learning_rate": 0.00035306476916181084, + "loss": 0.4124, + "step": 23213 + }, + { + "epoch": 1.2970081291728357, + "grad_norm": 0.46032506227493286, + "learning_rate": 0.0003530367548184671, + "loss": 0.4074, + "step": 23214 + }, + { + "epoch": 1.2970639997765177, + "grad_norm": 0.34192395210266113, + "learning_rate": 0.00035300874047512325, + "loss": 0.3298, + "step": 23215 + }, + { + "epoch": 1.2971198703801994, + "grad_norm": 0.5871394872665405, + "learning_rate": 0.0003529807261317795, + "loss": 0.3906, + "step": 23216 + }, + { + "epoch": 1.2971757409838813, + "grad_norm": 0.5673811435699463, + "learning_rate": 0.00035295271178843566, + "loss": 0.6143, + "step": 23217 + }, + { + "epoch": 1.2972316115875633, + "grad_norm": 0.49545028805732727, + "learning_rate": 0.00035292469744509187, + "loss": 0.3273, + "step": 23218 + }, + { + "epoch": 1.297287482191245, + "grad_norm": 0.45388004183769226, + "learning_rate": 0.00035289668310174813, + "loss": 0.4517, + "step": 23219 + }, + { + "epoch": 1.297343352794927, + "grad_norm": 0.45631325244903564, + "learning_rate": 0.0003528686687584043, + "loss": 0.3716, + "step": 23220 + }, + { + "epoch": 1.297399223398609, + "grad_norm": 0.5525321364402771, + "learning_rate": 0.00035284065441506054, + "loss": 0.3987, + "step": 23221 + }, + { + "epoch": 1.2974550940022906, + "grad_norm": 0.4376016855239868, + "learning_rate": 0.0003528126400717167, + "loss": 0.4203, + "step": 23222 + }, + { + "epoch": 1.2975109646059726, + "grad_norm": 0.4481053948402405, + "learning_rate": 0.00035278462572837295, + "loss": 0.3553, + "step": 23223 + }, + { + "epoch": 1.2975668352096545, + "grad_norm": 0.5899950265884399, + "learning_rate": 0.0003527566113850291, + "loss": 0.4384, + "step": 23224 + }, + { + "epoch": 1.2976227058133363, + "grad_norm": 0.5776848196983337, + "learning_rate": 0.00035272859704168537, + "loss": 0.391, + "step": 23225 + }, + { + "epoch": 1.2976785764170182, + "grad_norm": 0.9760509133338928, + "learning_rate": 0.0003527005826983415, + "loss": 0.7461, + "step": 23226 + }, + { + "epoch": 1.2977344470207002, + "grad_norm": 0.48772111535072327, + "learning_rate": 0.0003526725683549978, + "loss": 0.4348, + "step": 23227 + }, + { + "epoch": 1.297790317624382, + "grad_norm": 0.6050601601600647, + "learning_rate": 0.000352644554011654, + "loss": 0.4152, + "step": 23228 + }, + { + "epoch": 1.2978461882280639, + "grad_norm": 0.5773379802703857, + "learning_rate": 0.0003526165396683102, + "loss": 0.6627, + "step": 23229 + }, + { + "epoch": 1.2979020588317458, + "grad_norm": 1.2637819051742554, + "learning_rate": 0.0003525885253249664, + "loss": 0.3805, + "step": 23230 + }, + { + "epoch": 1.2979579294354275, + "grad_norm": 0.36789849400520325, + "learning_rate": 0.0003525605109816226, + "loss": 0.379, + "step": 23231 + }, + { + "epoch": 1.2980138000391095, + "grad_norm": 1.228517770767212, + "learning_rate": 0.0003525324966382788, + "loss": 0.3998, + "step": 23232 + }, + { + "epoch": 1.2980696706427912, + "grad_norm": 0.4901246726512909, + "learning_rate": 0.000352504482294935, + "loss": 0.4528, + "step": 23233 + }, + { + "epoch": 1.2981255412464732, + "grad_norm": 1.201059103012085, + "learning_rate": 0.0003524764679515912, + "loss": 0.4024, + "step": 23234 + }, + { + "epoch": 1.2981814118501551, + "grad_norm": 0.3461498022079468, + "learning_rate": 0.0003524484536082475, + "loss": 0.4766, + "step": 23235 + }, + { + "epoch": 1.2982372824538368, + "grad_norm": 0.5885706543922424, + "learning_rate": 0.00035242043926490363, + "loss": 0.4115, + "step": 23236 + }, + { + "epoch": 1.2982931530575188, + "grad_norm": 0.5974566340446472, + "learning_rate": 0.0003523924249215599, + "loss": 0.4172, + "step": 23237 + }, + { + "epoch": 1.2983490236612005, + "grad_norm": 0.43050625920295715, + "learning_rate": 0.00035236441057821604, + "loss": 0.4209, + "step": 23238 + }, + { + "epoch": 1.2984048942648825, + "grad_norm": 0.35479655861854553, + "learning_rate": 0.00035233639623487225, + "loss": 0.3466, + "step": 23239 + }, + { + "epoch": 1.2984607648685644, + "grad_norm": 0.6782317757606506, + "learning_rate": 0.00035230838189152845, + "loss": 0.3929, + "step": 23240 + }, + { + "epoch": 1.2985166354722462, + "grad_norm": 0.5430116653442383, + "learning_rate": 0.00035228036754818466, + "loss": 0.381, + "step": 23241 + }, + { + "epoch": 1.2985725060759281, + "grad_norm": 1.8719778060913086, + "learning_rate": 0.0003522523532048409, + "loss": 0.4519, + "step": 23242 + }, + { + "epoch": 1.29862837667961, + "grad_norm": 0.929402768611908, + "learning_rate": 0.00035222433886149707, + "loss": 0.3879, + "step": 23243 + }, + { + "epoch": 1.2986842472832918, + "grad_norm": 0.4359492063522339, + "learning_rate": 0.00035219632451815333, + "loss": 0.3825, + "step": 23244 + }, + { + "epoch": 1.2987401178869737, + "grad_norm": 0.581925630569458, + "learning_rate": 0.0003521683101748095, + "loss": 0.3987, + "step": 23245 + }, + { + "epoch": 1.2987959884906557, + "grad_norm": 0.44612976908683777, + "learning_rate": 0.00035214029583146574, + "loss": 0.4513, + "step": 23246 + }, + { + "epoch": 1.2988518590943374, + "grad_norm": 0.41260048747062683, + "learning_rate": 0.0003521122814881219, + "loss": 0.3928, + "step": 23247 + }, + { + "epoch": 1.2989077296980194, + "grad_norm": 0.37949511408805847, + "learning_rate": 0.00035208426714477815, + "loss": 0.4805, + "step": 23248 + }, + { + "epoch": 1.2989636003017013, + "grad_norm": 1.403312087059021, + "learning_rate": 0.0003520562528014343, + "loss": 0.4372, + "step": 23249 + }, + { + "epoch": 1.299019470905383, + "grad_norm": 0.5618900060653687, + "learning_rate": 0.00035202823845809057, + "loss": 0.5751, + "step": 23250 + }, + { + "epoch": 1.299075341509065, + "grad_norm": 0.6078426241874695, + "learning_rate": 0.00035200022411474677, + "loss": 0.451, + "step": 23251 + }, + { + "epoch": 1.299131212112747, + "grad_norm": 2.5864150524139404, + "learning_rate": 0.000351972209771403, + "loss": 0.4828, + "step": 23252 + }, + { + "epoch": 1.2991870827164287, + "grad_norm": 0.32635724544525146, + "learning_rate": 0.0003519441954280592, + "loss": 0.3615, + "step": 23253 + }, + { + "epoch": 1.2992429533201106, + "grad_norm": 1.2745980024337769, + "learning_rate": 0.0003519161810847154, + "loss": 0.4268, + "step": 23254 + }, + { + "epoch": 1.2992988239237926, + "grad_norm": 1.1818221807479858, + "learning_rate": 0.0003518881667413716, + "loss": 0.3637, + "step": 23255 + }, + { + "epoch": 1.2993546945274743, + "grad_norm": 0.4415982961654663, + "learning_rate": 0.0003518601523980278, + "loss": 0.4968, + "step": 23256 + }, + { + "epoch": 1.2994105651311563, + "grad_norm": 8.18068790435791, + "learning_rate": 0.000351832138054684, + "loss": 0.4252, + "step": 23257 + }, + { + "epoch": 1.2994664357348382, + "grad_norm": 11.505196571350098, + "learning_rate": 0.00035180412371134027, + "loss": 0.3926, + "step": 23258 + }, + { + "epoch": 1.29952230633852, + "grad_norm": 0.3538796603679657, + "learning_rate": 0.0003517761093679964, + "loss": 0.4059, + "step": 23259 + }, + { + "epoch": 1.299578176942202, + "grad_norm": 0.4358343482017517, + "learning_rate": 0.0003517480950246526, + "loss": 0.3711, + "step": 23260 + }, + { + "epoch": 1.2996340475458839, + "grad_norm": 0.9733981490135193, + "learning_rate": 0.00035172008068130883, + "loss": 0.5541, + "step": 23261 + }, + { + "epoch": 1.2996899181495656, + "grad_norm": 0.7531125545501709, + "learning_rate": 0.00035169206633796504, + "loss": 0.4649, + "step": 23262 + }, + { + "epoch": 1.2997457887532475, + "grad_norm": 0.4589073061943054, + "learning_rate": 0.00035166405199462124, + "loss": 0.4309, + "step": 23263 + }, + { + "epoch": 1.2998016593569295, + "grad_norm": 2.474250555038452, + "learning_rate": 0.00035163603765127745, + "loss": 0.4389, + "step": 23264 + }, + { + "epoch": 1.2998575299606112, + "grad_norm": 0.6045207977294922, + "learning_rate": 0.00035160802330793365, + "loss": 0.3803, + "step": 23265 + }, + { + "epoch": 1.2999134005642932, + "grad_norm": 5.722507953643799, + "learning_rate": 0.00035158000896458986, + "loss": 0.4102, + "step": 23266 + }, + { + "epoch": 1.299969271167975, + "grad_norm": 0.3507864475250244, + "learning_rate": 0.0003515519946212461, + "loss": 0.3856, + "step": 23267 + }, + { + "epoch": 1.3000251417716568, + "grad_norm": 0.3437455892562866, + "learning_rate": 0.00035152398027790227, + "loss": 0.4205, + "step": 23268 + }, + { + "epoch": 1.3000810123753388, + "grad_norm": 0.5923049449920654, + "learning_rate": 0.00035149596593455853, + "loss": 0.5329, + "step": 23269 + }, + { + "epoch": 1.3001368829790205, + "grad_norm": 0.3233787715435028, + "learning_rate": 0.0003514679515912147, + "loss": 0.408, + "step": 23270 + }, + { + "epoch": 1.3001927535827025, + "grad_norm": 0.4363120198249817, + "learning_rate": 0.00035143993724787094, + "loss": 0.4567, + "step": 23271 + }, + { + "epoch": 1.3002486241863842, + "grad_norm": 0.6691205501556396, + "learning_rate": 0.0003514119229045271, + "loss": 0.4138, + "step": 23272 + }, + { + "epoch": 1.3003044947900662, + "grad_norm": 1.0730096101760864, + "learning_rate": 0.00035138390856118335, + "loss": 0.3361, + "step": 23273 + }, + { + "epoch": 1.300360365393748, + "grad_norm": 0.596891462802887, + "learning_rate": 0.00035135589421783956, + "loss": 0.5951, + "step": 23274 + }, + { + "epoch": 1.3004162359974298, + "grad_norm": 0.4300175905227661, + "learning_rate": 0.00035132787987449576, + "loss": 0.4057, + "step": 23275 + }, + { + "epoch": 1.3004721066011118, + "grad_norm": 2.8836381435394287, + "learning_rate": 0.00035129986553115197, + "loss": 0.3335, + "step": 23276 + }, + { + "epoch": 1.3005279772047937, + "grad_norm": 1.171670913696289, + "learning_rate": 0.0003512718511878082, + "loss": 0.3297, + "step": 23277 + }, + { + "epoch": 1.3005838478084755, + "grad_norm": 0.6238918304443359, + "learning_rate": 0.0003512438368444644, + "loss": 0.4194, + "step": 23278 + }, + { + "epoch": 1.3006397184121574, + "grad_norm": 0.317214697599411, + "learning_rate": 0.0003512158225011206, + "loss": 0.4066, + "step": 23279 + }, + { + "epoch": 1.3006955890158394, + "grad_norm": 1.919671893119812, + "learning_rate": 0.0003511878081577768, + "loss": 0.5513, + "step": 23280 + }, + { + "epoch": 1.300751459619521, + "grad_norm": 0.4900985360145569, + "learning_rate": 0.00035115979381443295, + "loss": 0.436, + "step": 23281 + }, + { + "epoch": 1.300807330223203, + "grad_norm": 0.3749198913574219, + "learning_rate": 0.0003511317794710892, + "loss": 0.3994, + "step": 23282 + }, + { + "epoch": 1.300863200826885, + "grad_norm": 0.7123638391494751, + "learning_rate": 0.0003511037651277454, + "loss": 0.4518, + "step": 23283 + }, + { + "epoch": 1.3009190714305667, + "grad_norm": 0.3349698781967163, + "learning_rate": 0.0003510757507844016, + "loss": 0.3576, + "step": 23284 + }, + { + "epoch": 1.3009749420342487, + "grad_norm": 0.5240136981010437, + "learning_rate": 0.0003510477364410578, + "loss": 0.4263, + "step": 23285 + }, + { + "epoch": 1.3010308126379306, + "grad_norm": 0.5248064398765564, + "learning_rate": 0.00035101972209771403, + "loss": 0.3876, + "step": 23286 + }, + { + "epoch": 1.3010866832416124, + "grad_norm": 0.411444753408432, + "learning_rate": 0.00035099170775437023, + "loss": 0.449, + "step": 23287 + }, + { + "epoch": 1.3011425538452943, + "grad_norm": 0.5235092639923096, + "learning_rate": 0.00035096369341102644, + "loss": 0.4839, + "step": 23288 + }, + { + "epoch": 1.3011984244489763, + "grad_norm": 0.30437466502189636, + "learning_rate": 0.00035093567906768265, + "loss": 0.3636, + "step": 23289 + }, + { + "epoch": 1.301254295052658, + "grad_norm": 4.002578258514404, + "learning_rate": 0.0003509076647243389, + "loss": 0.4495, + "step": 23290 + }, + { + "epoch": 1.30131016565634, + "grad_norm": 0.6235608458518982, + "learning_rate": 0.00035087965038099506, + "loss": 0.3766, + "step": 23291 + }, + { + "epoch": 1.301366036260022, + "grad_norm": 0.5298808813095093, + "learning_rate": 0.0003508516360376513, + "loss": 0.3862, + "step": 23292 + }, + { + "epoch": 1.3014219068637036, + "grad_norm": 0.4403875470161438, + "learning_rate": 0.00035082362169430747, + "loss": 0.5116, + "step": 23293 + }, + { + "epoch": 1.3014777774673856, + "grad_norm": 0.44336867332458496, + "learning_rate": 0.00035079560735096373, + "loss": 0.4272, + "step": 23294 + }, + { + "epoch": 1.3015336480710675, + "grad_norm": 0.41691508889198303, + "learning_rate": 0.0003507675930076199, + "loss": 0.4348, + "step": 23295 + }, + { + "epoch": 1.3015895186747493, + "grad_norm": 0.5355640053749084, + "learning_rate": 0.00035073957866427614, + "loss": 0.4806, + "step": 23296 + }, + { + "epoch": 1.3016453892784312, + "grad_norm": 0.4393343925476074, + "learning_rate": 0.0003507115643209323, + "loss": 0.5194, + "step": 23297 + }, + { + "epoch": 1.3017012598821132, + "grad_norm": 0.6580603718757629, + "learning_rate": 0.00035068354997758855, + "loss": 0.6549, + "step": 23298 + }, + { + "epoch": 1.3017571304857949, + "grad_norm": 0.41574355959892273, + "learning_rate": 0.00035065553563424476, + "loss": 0.5516, + "step": 23299 + }, + { + "epoch": 1.3018130010894768, + "grad_norm": 0.46065038442611694, + "learning_rate": 0.00035062752129090096, + "loss": 0.3419, + "step": 23300 + }, + { + "epoch": 1.3018688716931586, + "grad_norm": 0.9024022221565247, + "learning_rate": 0.00035059950694755717, + "loss": 0.533, + "step": 23301 + }, + { + "epoch": 1.3019247422968405, + "grad_norm": 0.4603220224380493, + "learning_rate": 0.0003505714926042133, + "loss": 0.3759, + "step": 23302 + }, + { + "epoch": 1.3019806129005225, + "grad_norm": 0.9483262896537781, + "learning_rate": 0.0003505434782608696, + "loss": 0.3953, + "step": 23303 + }, + { + "epoch": 1.3020364835042042, + "grad_norm": 0.37474891543388367, + "learning_rate": 0.00035051546391752573, + "loss": 0.3802, + "step": 23304 + }, + { + "epoch": 1.3020923541078862, + "grad_norm": 0.3814505338668823, + "learning_rate": 0.000350487449574182, + "loss": 0.5057, + "step": 23305 + }, + { + "epoch": 1.3021482247115679, + "grad_norm": 0.3941917419433594, + "learning_rate": 0.0003504594352308382, + "loss": 0.4056, + "step": 23306 + }, + { + "epoch": 1.3022040953152498, + "grad_norm": 0.49389225244522095, + "learning_rate": 0.0003504314208874944, + "loss": 0.3917, + "step": 23307 + }, + { + "epoch": 1.3022599659189318, + "grad_norm": 0.5500901341438293, + "learning_rate": 0.0003504034065441506, + "loss": 0.3556, + "step": 23308 + }, + { + "epoch": 1.3023158365226135, + "grad_norm": 0.4344927668571472, + "learning_rate": 0.0003503753922008068, + "loss": 0.4577, + "step": 23309 + }, + { + "epoch": 1.3023717071262955, + "grad_norm": 0.3695262670516968, + "learning_rate": 0.000350347377857463, + "loss": 0.4956, + "step": 23310 + }, + { + "epoch": 1.3024275777299774, + "grad_norm": 0.3933672308921814, + "learning_rate": 0.00035031936351411923, + "loss": 0.4095, + "step": 23311 + }, + { + "epoch": 1.3024834483336591, + "grad_norm": 0.4511883854866028, + "learning_rate": 0.00035029134917077543, + "loss": 0.464, + "step": 23312 + }, + { + "epoch": 1.302539318937341, + "grad_norm": 0.46228834986686707, + "learning_rate": 0.0003502633348274317, + "loss": 0.465, + "step": 23313 + }, + { + "epoch": 1.302595189541023, + "grad_norm": 0.7684696316719055, + "learning_rate": 0.00035023532048408785, + "loss": 0.539, + "step": 23314 + }, + { + "epoch": 1.3026510601447048, + "grad_norm": 0.38402462005615234, + "learning_rate": 0.0003502073061407441, + "loss": 0.3187, + "step": 23315 + }, + { + "epoch": 1.3027069307483867, + "grad_norm": 0.6657121181488037, + "learning_rate": 0.00035017929179740026, + "loss": 0.3465, + "step": 23316 + }, + { + "epoch": 1.3027628013520687, + "grad_norm": 0.5075316429138184, + "learning_rate": 0.0003501512774540565, + "loss": 0.357, + "step": 23317 + }, + { + "epoch": 1.3028186719557504, + "grad_norm": 0.8741670846939087, + "learning_rate": 0.00035012326311071267, + "loss": 0.31, + "step": 23318 + }, + { + "epoch": 1.3028745425594324, + "grad_norm": 0.41380810737609863, + "learning_rate": 0.00035009524876736893, + "loss": 0.3657, + "step": 23319 + }, + { + "epoch": 1.3029304131631143, + "grad_norm": 1.7030460834503174, + "learning_rate": 0.0003500672344240251, + "loss": 0.3918, + "step": 23320 + }, + { + "epoch": 1.302986283766796, + "grad_norm": 0.4567525386810303, + "learning_rate": 0.00035003922008068134, + "loss": 0.3648, + "step": 23321 + }, + { + "epoch": 1.303042154370478, + "grad_norm": 0.4302152991294861, + "learning_rate": 0.00035001120573733755, + "loss": 0.3267, + "step": 23322 + }, + { + "epoch": 1.30309802497416, + "grad_norm": 0.404590368270874, + "learning_rate": 0.0003499831913939937, + "loss": 0.3808, + "step": 23323 + }, + { + "epoch": 1.3031538955778417, + "grad_norm": 0.45887017250061035, + "learning_rate": 0.00034995517705064996, + "loss": 0.434, + "step": 23324 + }, + { + "epoch": 1.3032097661815236, + "grad_norm": 1.5765916109085083, + "learning_rate": 0.0003499271627073061, + "loss": 0.7287, + "step": 23325 + }, + { + "epoch": 1.3032656367852056, + "grad_norm": 0.4813688397407532, + "learning_rate": 0.00034989914836396237, + "loss": 0.4202, + "step": 23326 + }, + { + "epoch": 1.3033215073888873, + "grad_norm": 1.6180508136749268, + "learning_rate": 0.0003498711340206185, + "loss": 0.3786, + "step": 23327 + }, + { + "epoch": 1.3033773779925693, + "grad_norm": 1.277573585510254, + "learning_rate": 0.0003498431196772748, + "loss": 0.4623, + "step": 23328 + }, + { + "epoch": 1.3034332485962512, + "grad_norm": 0.6921570301055908, + "learning_rate": 0.000349815105333931, + "loss": 0.3599, + "step": 23329 + }, + { + "epoch": 1.303489119199933, + "grad_norm": 0.4676547944545746, + "learning_rate": 0.0003497870909905872, + "loss": 0.4354, + "step": 23330 + }, + { + "epoch": 1.3035449898036149, + "grad_norm": 0.8223618268966675, + "learning_rate": 0.0003497590766472434, + "loss": 0.4663, + "step": 23331 + }, + { + "epoch": 1.3036008604072968, + "grad_norm": 1.2097089290618896, + "learning_rate": 0.0003497310623038996, + "loss": 0.4129, + "step": 23332 + }, + { + "epoch": 1.3036567310109786, + "grad_norm": 0.5460992455482483, + "learning_rate": 0.0003497030479605558, + "loss": 0.5073, + "step": 23333 + }, + { + "epoch": 1.3037126016146605, + "grad_norm": 0.45090430974960327, + "learning_rate": 0.000349675033617212, + "loss": 0.4226, + "step": 23334 + }, + { + "epoch": 1.3037684722183422, + "grad_norm": 0.4467841684818268, + "learning_rate": 0.0003496470192738682, + "loss": 0.5399, + "step": 23335 + }, + { + "epoch": 1.3038243428220242, + "grad_norm": 0.3679758608341217, + "learning_rate": 0.00034961900493052443, + "loss": 0.4317, + "step": 23336 + }, + { + "epoch": 1.3038802134257061, + "grad_norm": 0.49549081921577454, + "learning_rate": 0.00034959099058718063, + "loss": 0.394, + "step": 23337 + }, + { + "epoch": 1.3039360840293879, + "grad_norm": 0.4481695592403412, + "learning_rate": 0.0003495629762438369, + "loss": 0.4661, + "step": 23338 + }, + { + "epoch": 1.3039919546330698, + "grad_norm": 0.37464845180511475, + "learning_rate": 0.00034953496190049305, + "loss": 0.5339, + "step": 23339 + }, + { + "epoch": 1.3040478252367516, + "grad_norm": 0.5136007070541382, + "learning_rate": 0.0003495069475571493, + "loss": 0.4254, + "step": 23340 + }, + { + "epoch": 1.3041036958404335, + "grad_norm": 0.37540891766548157, + "learning_rate": 0.00034947893321380546, + "loss": 0.4582, + "step": 23341 + }, + { + "epoch": 1.3041595664441155, + "grad_norm": 0.4333711564540863, + "learning_rate": 0.0003494509188704617, + "loss": 0.3433, + "step": 23342 + }, + { + "epoch": 1.3042154370477972, + "grad_norm": 0.4923591911792755, + "learning_rate": 0.00034942290452711787, + "loss": 0.5087, + "step": 23343 + }, + { + "epoch": 1.3042713076514791, + "grad_norm": 0.3864803612232208, + "learning_rate": 0.0003493948901837741, + "loss": 0.3836, + "step": 23344 + }, + { + "epoch": 1.304327178255161, + "grad_norm": 1.6553407907485962, + "learning_rate": 0.00034936687584043033, + "loss": 0.3222, + "step": 23345 + }, + { + "epoch": 1.3043830488588428, + "grad_norm": 0.48118168115615845, + "learning_rate": 0.0003493388614970865, + "loss": 0.4946, + "step": 23346 + }, + { + "epoch": 1.3044389194625248, + "grad_norm": 1.6341392993927002, + "learning_rate": 0.00034931084715374275, + "loss": 0.421, + "step": 23347 + }, + { + "epoch": 1.3044947900662067, + "grad_norm": 0.6232287287712097, + "learning_rate": 0.0003492828328103989, + "loss": 0.3996, + "step": 23348 + }, + { + "epoch": 1.3045506606698885, + "grad_norm": 5.11721658706665, + "learning_rate": 0.00034925481846705516, + "loss": 0.6534, + "step": 23349 + }, + { + "epoch": 1.3046065312735704, + "grad_norm": 0.4531579315662384, + "learning_rate": 0.0003492268041237113, + "loss": 0.4231, + "step": 23350 + }, + { + "epoch": 1.3046624018772524, + "grad_norm": 0.4260435402393341, + "learning_rate": 0.00034919878978036757, + "loss": 0.4156, + "step": 23351 + }, + { + "epoch": 1.304718272480934, + "grad_norm": 0.6362407803535461, + "learning_rate": 0.0003491707754370237, + "loss": 0.3875, + "step": 23352 + }, + { + "epoch": 1.304774143084616, + "grad_norm": 1.1108944416046143, + "learning_rate": 0.00034914276109368, + "loss": 0.3679, + "step": 23353 + }, + { + "epoch": 1.304830013688298, + "grad_norm": 1.10218346118927, + "learning_rate": 0.0003491147467503362, + "loss": 0.6377, + "step": 23354 + }, + { + "epoch": 1.3048858842919797, + "grad_norm": 0.4082014262676239, + "learning_rate": 0.0003490867324069924, + "loss": 0.4476, + "step": 23355 + }, + { + "epoch": 1.3049417548956617, + "grad_norm": 0.5438917279243469, + "learning_rate": 0.0003490587180636486, + "loss": 0.3195, + "step": 23356 + }, + { + "epoch": 1.3049976254993436, + "grad_norm": 0.4595499038696289, + "learning_rate": 0.0003490307037203048, + "loss": 0.431, + "step": 23357 + }, + { + "epoch": 1.3050534961030253, + "grad_norm": 0.34856218099594116, + "learning_rate": 0.000349002689376961, + "loss": 0.3846, + "step": 23358 + }, + { + "epoch": 1.3051093667067073, + "grad_norm": 0.624541699886322, + "learning_rate": 0.0003489746750336172, + "loss": 0.4684, + "step": 23359 + }, + { + "epoch": 1.3051652373103892, + "grad_norm": 0.5704278945922852, + "learning_rate": 0.0003489466606902734, + "loss": 0.3713, + "step": 23360 + }, + { + "epoch": 1.305221107914071, + "grad_norm": 0.4571402370929718, + "learning_rate": 0.0003489186463469297, + "loss": 0.4394, + "step": 23361 + }, + { + "epoch": 1.305276978517753, + "grad_norm": 0.7263664603233337, + "learning_rate": 0.00034889063200358583, + "loss": 0.4281, + "step": 23362 + }, + { + "epoch": 1.3053328491214349, + "grad_norm": 0.923302412033081, + "learning_rate": 0.0003488626176602421, + "loss": 0.5234, + "step": 23363 + }, + { + "epoch": 1.3053887197251166, + "grad_norm": 2.1097426414489746, + "learning_rate": 0.00034883460331689825, + "loss": 0.4197, + "step": 23364 + }, + { + "epoch": 1.3054445903287986, + "grad_norm": 0.37054529786109924, + "learning_rate": 0.00034880658897355445, + "loss": 0.4256, + "step": 23365 + }, + { + "epoch": 1.3055004609324805, + "grad_norm": 1.5219712257385254, + "learning_rate": 0.00034877857463021066, + "loss": 0.5167, + "step": 23366 + }, + { + "epoch": 1.3055563315361622, + "grad_norm": 1.2195183038711548, + "learning_rate": 0.00034875056028686686, + "loss": 0.3447, + "step": 23367 + }, + { + "epoch": 1.3056122021398442, + "grad_norm": 0.4939723610877991, + "learning_rate": 0.00034872254594352307, + "loss": 0.6458, + "step": 23368 + }, + { + "epoch": 1.305668072743526, + "grad_norm": 0.35337233543395996, + "learning_rate": 0.0003486945316001793, + "loss": 0.3543, + "step": 23369 + }, + { + "epoch": 1.3057239433472079, + "grad_norm": 1.859374761581421, + "learning_rate": 0.00034866651725683553, + "loss": 0.3953, + "step": 23370 + }, + { + "epoch": 1.3057798139508896, + "grad_norm": 0.43738606572151184, + "learning_rate": 0.0003486385029134917, + "loss": 0.3876, + "step": 23371 + }, + { + "epoch": 1.3058356845545716, + "grad_norm": 0.45651623606681824, + "learning_rate": 0.00034861048857014795, + "loss": 0.439, + "step": 23372 + }, + { + "epoch": 1.3058915551582535, + "grad_norm": 0.4183028042316437, + "learning_rate": 0.0003485824742268041, + "loss": 0.4387, + "step": 23373 + }, + { + "epoch": 1.3059474257619352, + "grad_norm": 0.5277854800224304, + "learning_rate": 0.00034855445988346036, + "loss": 0.4819, + "step": 23374 + }, + { + "epoch": 1.3060032963656172, + "grad_norm": 0.4763420820236206, + "learning_rate": 0.0003485264455401165, + "loss": 0.3834, + "step": 23375 + }, + { + "epoch": 1.3060591669692991, + "grad_norm": 0.4915890395641327, + "learning_rate": 0.00034849843119677277, + "loss": 0.447, + "step": 23376 + }, + { + "epoch": 1.3061150375729809, + "grad_norm": 1.7733360528945923, + "learning_rate": 0.000348470416853429, + "loss": 0.4437, + "step": 23377 + }, + { + "epoch": 1.3061709081766628, + "grad_norm": 0.5526987910270691, + "learning_rate": 0.0003484424025100852, + "loss": 0.3632, + "step": 23378 + }, + { + "epoch": 1.3062267787803448, + "grad_norm": 0.623612642288208, + "learning_rate": 0.0003484143881667414, + "loss": 0.3907, + "step": 23379 + }, + { + "epoch": 1.3062826493840265, + "grad_norm": 0.5144366025924683, + "learning_rate": 0.0003483863738233976, + "loss": 0.4564, + "step": 23380 + }, + { + "epoch": 1.3063385199877084, + "grad_norm": 0.5195882320404053, + "learning_rate": 0.0003483583594800538, + "loss": 0.3915, + "step": 23381 + }, + { + "epoch": 1.3063943905913904, + "grad_norm": 1.1427438259124756, + "learning_rate": 0.00034833034513671, + "loss": 0.4391, + "step": 23382 + }, + { + "epoch": 1.3064502611950721, + "grad_norm": 1.0403199195861816, + "learning_rate": 0.0003483023307933662, + "loss": 0.4873, + "step": 23383 + }, + { + "epoch": 1.306506131798754, + "grad_norm": 2.1303858757019043, + "learning_rate": 0.00034827431645002247, + "loss": 0.3701, + "step": 23384 + }, + { + "epoch": 1.306562002402436, + "grad_norm": 2.326528310775757, + "learning_rate": 0.0003482463021066786, + "loss": 0.3318, + "step": 23385 + }, + { + "epoch": 1.3066178730061178, + "grad_norm": 0.8641043305397034, + "learning_rate": 0.0003482182877633349, + "loss": 0.3486, + "step": 23386 + }, + { + "epoch": 1.3066737436097997, + "grad_norm": 0.515358567237854, + "learning_rate": 0.00034819027341999103, + "loss": 0.344, + "step": 23387 + }, + { + "epoch": 1.3067296142134817, + "grad_norm": 0.6092948913574219, + "learning_rate": 0.00034816225907664724, + "loss": 0.4572, + "step": 23388 + }, + { + "epoch": 1.3067854848171634, + "grad_norm": 1.737182855606079, + "learning_rate": 0.00034813424473330344, + "loss": 0.4217, + "step": 23389 + }, + { + "epoch": 1.3068413554208453, + "grad_norm": 0.5136735439300537, + "learning_rate": 0.00034810623038995965, + "loss": 0.4387, + "step": 23390 + }, + { + "epoch": 1.3068972260245273, + "grad_norm": 0.5833978652954102, + "learning_rate": 0.00034807821604661586, + "loss": 0.4698, + "step": 23391 + }, + { + "epoch": 1.306953096628209, + "grad_norm": 0.6573054790496826, + "learning_rate": 0.00034805020170327206, + "loss": 0.4674, + "step": 23392 + }, + { + "epoch": 1.307008967231891, + "grad_norm": 0.7358702421188354, + "learning_rate": 0.0003480221873599283, + "loss": 0.3108, + "step": 23393 + }, + { + "epoch": 1.307064837835573, + "grad_norm": 0.5160160064697266, + "learning_rate": 0.0003479941730165845, + "loss": 0.36, + "step": 23394 + }, + { + "epoch": 1.3071207084392547, + "grad_norm": 0.4788885712623596, + "learning_rate": 0.00034796615867324073, + "loss": 0.4034, + "step": 23395 + }, + { + "epoch": 1.3071765790429366, + "grad_norm": 0.3547261953353882, + "learning_rate": 0.0003479381443298969, + "loss": 0.4415, + "step": 23396 + }, + { + "epoch": 1.3072324496466186, + "grad_norm": 1.2191470861434937, + "learning_rate": 0.00034791012998655315, + "loss": 0.3845, + "step": 23397 + }, + { + "epoch": 1.3072883202503003, + "grad_norm": 0.5379496216773987, + "learning_rate": 0.0003478821156432093, + "loss": 0.5577, + "step": 23398 + }, + { + "epoch": 1.3073441908539822, + "grad_norm": 0.391899973154068, + "learning_rate": 0.00034785410129986556, + "loss": 0.3444, + "step": 23399 + }, + { + "epoch": 1.3074000614576642, + "grad_norm": 0.4175812900066376, + "learning_rate": 0.00034782608695652176, + "loss": 0.4134, + "step": 23400 + }, + { + "epoch": 1.307455932061346, + "grad_norm": 0.4697556793689728, + "learning_rate": 0.00034779807261317797, + "loss": 0.4022, + "step": 23401 + }, + { + "epoch": 1.3075118026650279, + "grad_norm": 2.536190986633301, + "learning_rate": 0.0003477700582698342, + "loss": 0.4674, + "step": 23402 + }, + { + "epoch": 1.3075676732687096, + "grad_norm": 0.44658878445625305, + "learning_rate": 0.0003477420439264904, + "loss": 0.408, + "step": 23403 + }, + { + "epoch": 1.3076235438723915, + "grad_norm": 0.6829010844230652, + "learning_rate": 0.0003477140295831466, + "loss": 0.358, + "step": 23404 + }, + { + "epoch": 1.3076794144760733, + "grad_norm": 0.3272370994091034, + "learning_rate": 0.0003476860152398028, + "loss": 0.4082, + "step": 23405 + }, + { + "epoch": 1.3077352850797552, + "grad_norm": 0.4040141999721527, + "learning_rate": 0.000347658000896459, + "loss": 0.4328, + "step": 23406 + }, + { + "epoch": 1.3077911556834372, + "grad_norm": 2.863860845565796, + "learning_rate": 0.0003476299865531152, + "loss": 0.3392, + "step": 23407 + }, + { + "epoch": 1.307847026287119, + "grad_norm": 1.3492611646652222, + "learning_rate": 0.0003476019722097714, + "loss": 0.6022, + "step": 23408 + }, + { + "epoch": 1.3079028968908009, + "grad_norm": 0.559354841709137, + "learning_rate": 0.0003475739578664276, + "loss": 0.6327, + "step": 23409 + }, + { + "epoch": 1.3079587674944828, + "grad_norm": 0.46902257204055786, + "learning_rate": 0.0003475459435230838, + "loss": 0.4671, + "step": 23410 + }, + { + "epoch": 1.3080146380981645, + "grad_norm": 0.347433865070343, + "learning_rate": 0.00034751792917974, + "loss": 0.3296, + "step": 23411 + }, + { + "epoch": 1.3080705087018465, + "grad_norm": 0.39806047081947327, + "learning_rate": 0.00034748991483639623, + "loss": 0.4279, + "step": 23412 + }, + { + "epoch": 1.3081263793055284, + "grad_norm": 0.6507840752601624, + "learning_rate": 0.00034746190049305244, + "loss": 0.4276, + "step": 23413 + }, + { + "epoch": 1.3081822499092102, + "grad_norm": 0.6112763285636902, + "learning_rate": 0.00034743388614970864, + "loss": 0.4651, + "step": 23414 + }, + { + "epoch": 1.3082381205128921, + "grad_norm": 0.4523104131221771, + "learning_rate": 0.00034740587180636485, + "loss": 0.3858, + "step": 23415 + }, + { + "epoch": 1.308293991116574, + "grad_norm": 0.6036708354949951, + "learning_rate": 0.0003473778574630211, + "loss": 0.4056, + "step": 23416 + }, + { + "epoch": 1.3083498617202558, + "grad_norm": 0.3790700137615204, + "learning_rate": 0.00034734984311967726, + "loss": 0.4148, + "step": 23417 + }, + { + "epoch": 1.3084057323239378, + "grad_norm": 0.4394712746143341, + "learning_rate": 0.0003473218287763335, + "loss": 0.4225, + "step": 23418 + }, + { + "epoch": 1.3084616029276197, + "grad_norm": 0.45203980803489685, + "learning_rate": 0.0003472938144329897, + "loss": 0.5725, + "step": 23419 + }, + { + "epoch": 1.3085174735313014, + "grad_norm": 0.4541444778442383, + "learning_rate": 0.00034726580008964593, + "loss": 0.342, + "step": 23420 + }, + { + "epoch": 1.3085733441349834, + "grad_norm": 0.9246968626976013, + "learning_rate": 0.0003472377857463021, + "loss": 0.3575, + "step": 23421 + }, + { + "epoch": 1.3086292147386653, + "grad_norm": 0.4083766043186188, + "learning_rate": 0.00034720977140295835, + "loss": 0.4689, + "step": 23422 + }, + { + "epoch": 1.308685085342347, + "grad_norm": 0.5005500912666321, + "learning_rate": 0.0003471817570596145, + "loss": 0.4559, + "step": 23423 + }, + { + "epoch": 1.308740955946029, + "grad_norm": 0.37983161211013794, + "learning_rate": 0.00034715374271627076, + "loss": 0.4089, + "step": 23424 + }, + { + "epoch": 1.308796826549711, + "grad_norm": 1.002068042755127, + "learning_rate": 0.00034712572837292696, + "loss": 0.4533, + "step": 23425 + }, + { + "epoch": 1.3088526971533927, + "grad_norm": 0.46272316575050354, + "learning_rate": 0.00034709771402958317, + "loss": 0.3509, + "step": 23426 + }, + { + "epoch": 1.3089085677570746, + "grad_norm": 0.4349181652069092, + "learning_rate": 0.0003470696996862394, + "loss": 0.5139, + "step": 23427 + }, + { + "epoch": 1.3089644383607566, + "grad_norm": 0.5109045505523682, + "learning_rate": 0.0003470416853428956, + "loss": 0.4759, + "step": 23428 + }, + { + "epoch": 1.3090203089644383, + "grad_norm": 2.024630308151245, + "learning_rate": 0.0003470136709995518, + "loss": 0.5005, + "step": 23429 + }, + { + "epoch": 1.3090761795681203, + "grad_norm": 0.37977609038352966, + "learning_rate": 0.00034698565665620794, + "loss": 0.3921, + "step": 23430 + }, + { + "epoch": 1.3091320501718022, + "grad_norm": 0.36681169271469116, + "learning_rate": 0.0003469576423128642, + "loss": 0.3508, + "step": 23431 + }, + { + "epoch": 1.309187920775484, + "grad_norm": 0.3598089814186096, + "learning_rate": 0.0003469296279695204, + "loss": 0.354, + "step": 23432 + }, + { + "epoch": 1.309243791379166, + "grad_norm": 0.7468915581703186, + "learning_rate": 0.0003469016136261766, + "loss": 0.4727, + "step": 23433 + }, + { + "epoch": 1.3092996619828479, + "grad_norm": 0.4188072681427002, + "learning_rate": 0.0003468735992828328, + "loss": 0.472, + "step": 23434 + }, + { + "epoch": 1.3093555325865296, + "grad_norm": 0.4200717508792877, + "learning_rate": 0.000346845584939489, + "loss": 0.4437, + "step": 23435 + }, + { + "epoch": 1.3094114031902115, + "grad_norm": 0.43880560994148254, + "learning_rate": 0.0003468175705961452, + "loss": 0.3967, + "step": 23436 + }, + { + "epoch": 1.3094672737938933, + "grad_norm": 0.3741229176521301, + "learning_rate": 0.00034678955625280143, + "loss": 0.3211, + "step": 23437 + }, + { + "epoch": 1.3095231443975752, + "grad_norm": 1.0856196880340576, + "learning_rate": 0.00034676154190945764, + "loss": 0.4939, + "step": 23438 + }, + { + "epoch": 1.309579015001257, + "grad_norm": 0.46875670552253723, + "learning_rate": 0.0003467335275661139, + "loss": 0.471, + "step": 23439 + }, + { + "epoch": 1.309634885604939, + "grad_norm": 0.5360708832740784, + "learning_rate": 0.00034670551322277005, + "loss": 0.4226, + "step": 23440 + }, + { + "epoch": 1.3096907562086209, + "grad_norm": 0.3418543040752411, + "learning_rate": 0.0003466774988794263, + "loss": 0.4328, + "step": 23441 + }, + { + "epoch": 1.3097466268123026, + "grad_norm": 0.4173596203327179, + "learning_rate": 0.00034664948453608246, + "loss": 0.4372, + "step": 23442 + }, + { + "epoch": 1.3098024974159845, + "grad_norm": 0.5218303203582764, + "learning_rate": 0.0003466214701927387, + "loss": 0.3847, + "step": 23443 + }, + { + "epoch": 1.3098583680196665, + "grad_norm": 9.132621765136719, + "learning_rate": 0.0003465934558493949, + "loss": 0.4094, + "step": 23444 + }, + { + "epoch": 1.3099142386233482, + "grad_norm": 0.6150526404380798, + "learning_rate": 0.00034656544150605113, + "loss": 0.3101, + "step": 23445 + }, + { + "epoch": 1.3099701092270302, + "grad_norm": 0.3916655480861664, + "learning_rate": 0.0003465374271627073, + "loss": 0.4015, + "step": 23446 + }, + { + "epoch": 1.3100259798307121, + "grad_norm": 0.3793591558933258, + "learning_rate": 0.00034650941281936354, + "loss": 0.4102, + "step": 23447 + }, + { + "epoch": 1.3100818504343938, + "grad_norm": 0.3127487897872925, + "learning_rate": 0.00034648139847601975, + "loss": 0.3687, + "step": 23448 + }, + { + "epoch": 1.3101377210380758, + "grad_norm": 1.0578207969665527, + "learning_rate": 0.00034645338413267596, + "loss": 0.4036, + "step": 23449 + }, + { + "epoch": 1.3101935916417577, + "grad_norm": 0.5785660147666931, + "learning_rate": 0.00034642536978933216, + "loss": 0.4545, + "step": 23450 + }, + { + "epoch": 1.3102494622454395, + "grad_norm": 0.7856819033622742, + "learning_rate": 0.0003463973554459883, + "loss": 0.6124, + "step": 23451 + }, + { + "epoch": 1.3103053328491214, + "grad_norm": 0.3886086940765381, + "learning_rate": 0.0003463693411026446, + "loss": 0.4239, + "step": 23452 + }, + { + "epoch": 1.3103612034528034, + "grad_norm": 1.050272822380066, + "learning_rate": 0.0003463413267593007, + "loss": 0.3908, + "step": 23453 + }, + { + "epoch": 1.310417074056485, + "grad_norm": 0.7755748629570007, + "learning_rate": 0.000346313312415957, + "loss": 0.3919, + "step": 23454 + }, + { + "epoch": 1.310472944660167, + "grad_norm": 0.5313194394111633, + "learning_rate": 0.0003462852980726132, + "loss": 0.5036, + "step": 23455 + }, + { + "epoch": 1.310528815263849, + "grad_norm": 0.37065356969833374, + "learning_rate": 0.0003462572837292694, + "loss": 0.2807, + "step": 23456 + }, + { + "epoch": 1.3105846858675307, + "grad_norm": 0.42370080947875977, + "learning_rate": 0.0003462292693859256, + "loss": 0.3829, + "step": 23457 + }, + { + "epoch": 1.3106405564712127, + "grad_norm": 0.4573573172092438, + "learning_rate": 0.0003462012550425818, + "loss": 0.4738, + "step": 23458 + }, + { + "epoch": 1.3106964270748946, + "grad_norm": 0.31823182106018066, + "learning_rate": 0.000346173240699238, + "loss": 0.3235, + "step": 23459 + }, + { + "epoch": 1.3107522976785764, + "grad_norm": 0.5220576524734497, + "learning_rate": 0.0003461452263558942, + "loss": 0.3822, + "step": 23460 + }, + { + "epoch": 1.3108081682822583, + "grad_norm": 0.5994536876678467, + "learning_rate": 0.0003461172120125504, + "loss": 0.4117, + "step": 23461 + }, + { + "epoch": 1.3108640388859403, + "grad_norm": 0.44433364272117615, + "learning_rate": 0.00034608919766920663, + "loss": 0.4713, + "step": 23462 + }, + { + "epoch": 1.310919909489622, + "grad_norm": 1.1041007041931152, + "learning_rate": 0.00034606118332586284, + "loss": 0.3171, + "step": 23463 + }, + { + "epoch": 1.310975780093304, + "grad_norm": 1.4434001445770264, + "learning_rate": 0.0003460331689825191, + "loss": 0.3669, + "step": 23464 + }, + { + "epoch": 1.311031650696986, + "grad_norm": 0.4607708752155304, + "learning_rate": 0.00034600515463917525, + "loss": 0.3975, + "step": 23465 + }, + { + "epoch": 1.3110875213006676, + "grad_norm": 0.907376766204834, + "learning_rate": 0.0003459771402958315, + "loss": 0.4628, + "step": 23466 + }, + { + "epoch": 1.3111433919043496, + "grad_norm": 0.5207427144050598, + "learning_rate": 0.00034594912595248766, + "loss": 0.3854, + "step": 23467 + }, + { + "epoch": 1.3111992625080315, + "grad_norm": 0.5380476713180542, + "learning_rate": 0.0003459211116091439, + "loss": 0.3818, + "step": 23468 + }, + { + "epoch": 1.3112551331117133, + "grad_norm": 0.7369285821914673, + "learning_rate": 0.00034589309726580007, + "loss": 0.4992, + "step": 23469 + }, + { + "epoch": 1.3113110037153952, + "grad_norm": 0.7006112933158875, + "learning_rate": 0.00034586508292245633, + "loss": 0.3755, + "step": 23470 + }, + { + "epoch": 1.311366874319077, + "grad_norm": 0.3975147604942322, + "learning_rate": 0.00034583706857911254, + "loss": 0.3632, + "step": 23471 + }, + { + "epoch": 1.311422744922759, + "grad_norm": 0.40082108974456787, + "learning_rate": 0.0003458090542357687, + "loss": 0.4048, + "step": 23472 + }, + { + "epoch": 1.3114786155264406, + "grad_norm": 2.8548154830932617, + "learning_rate": 0.00034578103989242495, + "loss": 0.4092, + "step": 23473 + }, + { + "epoch": 1.3115344861301226, + "grad_norm": 0.39776793122291565, + "learning_rate": 0.0003457530255490811, + "loss": 0.4461, + "step": 23474 + }, + { + "epoch": 1.3115903567338045, + "grad_norm": 1.2111907005310059, + "learning_rate": 0.00034572501120573736, + "loss": 0.42, + "step": 23475 + }, + { + "epoch": 1.3116462273374863, + "grad_norm": 0.9645395278930664, + "learning_rate": 0.0003456969968623935, + "loss": 0.5461, + "step": 23476 + }, + { + "epoch": 1.3117020979411682, + "grad_norm": 0.4397343397140503, + "learning_rate": 0.0003456689825190498, + "loss": 0.4543, + "step": 23477 + }, + { + "epoch": 1.3117579685448502, + "grad_norm": 0.47609153389930725, + "learning_rate": 0.0003456409681757059, + "loss": 0.4824, + "step": 23478 + }, + { + "epoch": 1.311813839148532, + "grad_norm": 0.7371615171432495, + "learning_rate": 0.0003456129538323622, + "loss": 0.6211, + "step": 23479 + }, + { + "epoch": 1.3118697097522138, + "grad_norm": 0.3916061520576477, + "learning_rate": 0.0003455849394890184, + "loss": 0.4703, + "step": 23480 + }, + { + "epoch": 1.3119255803558958, + "grad_norm": 0.4926336705684662, + "learning_rate": 0.0003455569251456746, + "loss": 0.566, + "step": 23481 + }, + { + "epoch": 1.3119814509595775, + "grad_norm": 0.5537511110305786, + "learning_rate": 0.0003455289108023308, + "loss": 0.3699, + "step": 23482 + }, + { + "epoch": 1.3120373215632595, + "grad_norm": 0.6688359379768372, + "learning_rate": 0.000345500896458987, + "loss": 0.446, + "step": 23483 + }, + { + "epoch": 1.3120931921669414, + "grad_norm": 0.4874840974807739, + "learning_rate": 0.0003454728821156432, + "loss": 0.4053, + "step": 23484 + }, + { + "epoch": 1.3121490627706232, + "grad_norm": 0.40187689661979675, + "learning_rate": 0.0003454448677722994, + "loss": 0.4129, + "step": 23485 + }, + { + "epoch": 1.312204933374305, + "grad_norm": 0.5295560956001282, + "learning_rate": 0.0003454168534289556, + "loss": 0.4225, + "step": 23486 + }, + { + "epoch": 1.312260803977987, + "grad_norm": 0.575935959815979, + "learning_rate": 0.0003453888390856119, + "loss": 0.4134, + "step": 23487 + }, + { + "epoch": 1.3123166745816688, + "grad_norm": 0.3821413516998291, + "learning_rate": 0.00034536082474226804, + "loss": 0.3447, + "step": 23488 + }, + { + "epoch": 1.3123725451853507, + "grad_norm": 0.619831383228302, + "learning_rate": 0.0003453328103989243, + "loss": 0.4958, + "step": 23489 + }, + { + "epoch": 1.3124284157890327, + "grad_norm": 0.5905696749687195, + "learning_rate": 0.00034530479605558045, + "loss": 0.5664, + "step": 23490 + }, + { + "epoch": 1.3124842863927144, + "grad_norm": 0.5200604200363159, + "learning_rate": 0.0003452767817122367, + "loss": 0.6507, + "step": 23491 + }, + { + "epoch": 1.3125401569963964, + "grad_norm": 3.786301612854004, + "learning_rate": 0.00034524876736889286, + "loss": 0.5122, + "step": 23492 + }, + { + "epoch": 1.3125960276000783, + "grad_norm": 1.5104137659072876, + "learning_rate": 0.00034522075302554907, + "loss": 0.4073, + "step": 23493 + }, + { + "epoch": 1.31265189820376, + "grad_norm": 0.682905375957489, + "learning_rate": 0.00034519273868220527, + "loss": 0.3131, + "step": 23494 + }, + { + "epoch": 1.312707768807442, + "grad_norm": 0.6666121482849121, + "learning_rate": 0.0003451647243388615, + "loss": 0.5191, + "step": 23495 + }, + { + "epoch": 1.312763639411124, + "grad_norm": 0.533260703086853, + "learning_rate": 0.00034513670999551774, + "loss": 0.4729, + "step": 23496 + }, + { + "epoch": 1.3128195100148057, + "grad_norm": 5.143322467803955, + "learning_rate": 0.0003451086956521739, + "loss": 0.5853, + "step": 23497 + }, + { + "epoch": 1.3128753806184876, + "grad_norm": 0.4476051926612854, + "learning_rate": 0.00034508068130883015, + "loss": 0.5444, + "step": 23498 + }, + { + "epoch": 1.3129312512221696, + "grad_norm": 0.41969385743141174, + "learning_rate": 0.0003450526669654863, + "loss": 0.3883, + "step": 23499 + }, + { + "epoch": 1.3129871218258513, + "grad_norm": 0.45721006393432617, + "learning_rate": 0.00034502465262214256, + "loss": 0.4566, + "step": 23500 + }, + { + "epoch": 1.3129871218258513, + "eval_cer": 0.08537636520353943, + "eval_loss": 0.3207121789455414, + "eval_runtime": 56.0805, + "eval_samples_per_second": 80.919, + "eval_steps_per_second": 5.064, + "eval_wer": 0.33870562274796795, + "step": 23500 + }, + { + "epoch": 1.3130429924295333, + "grad_norm": 0.3758026659488678, + "learning_rate": 0.0003449966382787987, + "loss": 0.2462, + "step": 23501 + }, + { + "epoch": 1.313098863033215, + "grad_norm": 0.5208552479743958, + "learning_rate": 0.00034496862393545497, + "loss": 0.3792, + "step": 23502 + }, + { + "epoch": 1.313154733636897, + "grad_norm": 1.0433789491653442, + "learning_rate": 0.0003449406095921112, + "loss": 0.4419, + "step": 23503 + }, + { + "epoch": 1.313210604240579, + "grad_norm": 0.466652512550354, + "learning_rate": 0.0003449125952487674, + "loss": 0.4342, + "step": 23504 + }, + { + "epoch": 1.3132664748442606, + "grad_norm": 0.4060458540916443, + "learning_rate": 0.0003448845809054236, + "loss": 0.4621, + "step": 23505 + }, + { + "epoch": 1.3133223454479426, + "grad_norm": 0.4168758988380432, + "learning_rate": 0.0003448565665620798, + "loss": 0.3503, + "step": 23506 + }, + { + "epoch": 1.3133782160516243, + "grad_norm": 0.5693914294242859, + "learning_rate": 0.000344828552218736, + "loss": 0.4682, + "step": 23507 + }, + { + "epoch": 1.3134340866553063, + "grad_norm": 0.41575300693511963, + "learning_rate": 0.0003448005378753922, + "loss": 0.412, + "step": 23508 + }, + { + "epoch": 1.3134899572589882, + "grad_norm": 1.681659460067749, + "learning_rate": 0.0003447725235320484, + "loss": 0.3946, + "step": 23509 + }, + { + "epoch": 1.31354582786267, + "grad_norm": 0.5335317850112915, + "learning_rate": 0.0003447445091887047, + "loss": 0.4858, + "step": 23510 + }, + { + "epoch": 1.3136016984663519, + "grad_norm": 0.36908823251724243, + "learning_rate": 0.0003447164948453608, + "loss": 0.331, + "step": 23511 + }, + { + "epoch": 1.3136575690700338, + "grad_norm": 0.8421762585639954, + "learning_rate": 0.0003446884805020171, + "loss": 0.4237, + "step": 23512 + }, + { + "epoch": 1.3137134396737156, + "grad_norm": 0.43445679545402527, + "learning_rate": 0.00034466046615867324, + "loss": 0.4311, + "step": 23513 + }, + { + "epoch": 1.3137693102773975, + "grad_norm": 0.44646209478378296, + "learning_rate": 0.00034463245181532944, + "loss": 0.4198, + "step": 23514 + }, + { + "epoch": 1.3138251808810795, + "grad_norm": 0.9650083780288696, + "learning_rate": 0.00034460443747198565, + "loss": 0.4907, + "step": 23515 + }, + { + "epoch": 1.3138810514847612, + "grad_norm": 1.9323903322219849, + "learning_rate": 0.00034457642312864185, + "loss": 0.411, + "step": 23516 + }, + { + "epoch": 1.3139369220884431, + "grad_norm": 0.7554362416267395, + "learning_rate": 0.00034454840878529806, + "loss": 0.4338, + "step": 23517 + }, + { + "epoch": 1.313992792692125, + "grad_norm": 0.44645553827285767, + "learning_rate": 0.00034452039444195427, + "loss": 0.378, + "step": 23518 + }, + { + "epoch": 1.3140486632958068, + "grad_norm": 0.3509562015533447, + "learning_rate": 0.0003444923800986105, + "loss": 0.4117, + "step": 23519 + }, + { + "epoch": 1.3141045338994888, + "grad_norm": 0.6100826263427734, + "learning_rate": 0.0003444643657552667, + "loss": 0.4313, + "step": 23520 + }, + { + "epoch": 1.3141604045031707, + "grad_norm": 0.5420549511909485, + "learning_rate": 0.00034443635141192294, + "loss": 0.3958, + "step": 23521 + }, + { + "epoch": 1.3142162751068525, + "grad_norm": 0.6072585582733154, + "learning_rate": 0.0003444083370685791, + "loss": 0.3783, + "step": 23522 + }, + { + "epoch": 1.3142721457105344, + "grad_norm": 0.5758639574050903, + "learning_rate": 0.00034438032272523535, + "loss": 0.4016, + "step": 23523 + }, + { + "epoch": 1.3143280163142164, + "grad_norm": 0.5676255226135254, + "learning_rate": 0.0003443523083818915, + "loss": 0.3948, + "step": 23524 + }, + { + "epoch": 1.314383886917898, + "grad_norm": 0.5443459749221802, + "learning_rate": 0.00034432429403854776, + "loss": 0.3117, + "step": 23525 + }, + { + "epoch": 1.31443975752158, + "grad_norm": 0.6542123556137085, + "learning_rate": 0.00034429627969520397, + "loss": 0.4078, + "step": 23526 + }, + { + "epoch": 1.314495628125262, + "grad_norm": 0.38695433735847473, + "learning_rate": 0.00034426826535186017, + "loss": 0.3987, + "step": 23527 + }, + { + "epoch": 1.3145514987289437, + "grad_norm": 0.44928720593452454, + "learning_rate": 0.0003442402510085164, + "loss": 0.4018, + "step": 23528 + }, + { + "epoch": 1.3146073693326257, + "grad_norm": 0.6084884405136108, + "learning_rate": 0.0003442122366651726, + "loss": 0.5442, + "step": 23529 + }, + { + "epoch": 1.3146632399363076, + "grad_norm": 0.4744155704975128, + "learning_rate": 0.0003441842223218288, + "loss": 0.5685, + "step": 23530 + }, + { + "epoch": 1.3147191105399894, + "grad_norm": 0.6516091823577881, + "learning_rate": 0.000344156207978485, + "loss": 0.457, + "step": 23531 + }, + { + "epoch": 1.3147749811436713, + "grad_norm": 0.53501296043396, + "learning_rate": 0.0003441281936351412, + "loss": 0.4438, + "step": 23532 + }, + { + "epoch": 1.3148308517473533, + "grad_norm": 0.4815616309642792, + "learning_rate": 0.0003441001792917974, + "loss": 0.3611, + "step": 23533 + }, + { + "epoch": 1.314886722351035, + "grad_norm": 0.6548426747322083, + "learning_rate": 0.0003440721649484536, + "loss": 0.4452, + "step": 23534 + }, + { + "epoch": 1.314942592954717, + "grad_norm": 0.5614412426948547, + "learning_rate": 0.0003440441506051098, + "loss": 0.5528, + "step": 23535 + }, + { + "epoch": 1.3149984635583987, + "grad_norm": 0.5969020128250122, + "learning_rate": 0.000344016136261766, + "loss": 0.496, + "step": 23536 + }, + { + "epoch": 1.3150543341620806, + "grad_norm": 0.6261134147644043, + "learning_rate": 0.00034398812191842223, + "loss": 0.388, + "step": 23537 + }, + { + "epoch": 1.3151102047657626, + "grad_norm": 0.36242353916168213, + "learning_rate": 0.00034396010757507844, + "loss": 0.4047, + "step": 23538 + }, + { + "epoch": 1.3151660753694443, + "grad_norm": 0.41708266735076904, + "learning_rate": 0.00034393209323173464, + "loss": 0.4169, + "step": 23539 + }, + { + "epoch": 1.3152219459731262, + "grad_norm": 0.3925190269947052, + "learning_rate": 0.00034390407888839085, + "loss": 0.5705, + "step": 23540 + }, + { + "epoch": 1.315277816576808, + "grad_norm": 1.4903907775878906, + "learning_rate": 0.00034387606454504705, + "loss": 0.3549, + "step": 23541 + }, + { + "epoch": 1.31533368718049, + "grad_norm": 0.42550787329673767, + "learning_rate": 0.0003438480502017033, + "loss": 0.3792, + "step": 23542 + }, + { + "epoch": 1.3153895577841719, + "grad_norm": 0.7983258962631226, + "learning_rate": 0.00034382003585835947, + "loss": 0.5151, + "step": 23543 + }, + { + "epoch": 1.3154454283878536, + "grad_norm": 0.8229614496231079, + "learning_rate": 0.0003437920215150157, + "loss": 0.4155, + "step": 23544 + }, + { + "epoch": 1.3155012989915356, + "grad_norm": 0.5831925272941589, + "learning_rate": 0.0003437640071716719, + "loss": 0.4727, + "step": 23545 + }, + { + "epoch": 1.3155571695952175, + "grad_norm": 0.5499426126480103, + "learning_rate": 0.00034373599282832814, + "loss": 0.5316, + "step": 23546 + }, + { + "epoch": 1.3156130401988992, + "grad_norm": 4.437705993652344, + "learning_rate": 0.0003437079784849843, + "loss": 0.4386, + "step": 23547 + }, + { + "epoch": 1.3156689108025812, + "grad_norm": 0.5308583974838257, + "learning_rate": 0.00034367996414164055, + "loss": 0.4283, + "step": 23548 + }, + { + "epoch": 1.3157247814062631, + "grad_norm": 0.4064851999282837, + "learning_rate": 0.0003436519497982967, + "loss": 0.435, + "step": 23549 + }, + { + "epoch": 1.3157806520099449, + "grad_norm": 0.4547351598739624, + "learning_rate": 0.00034362393545495296, + "loss": 0.6322, + "step": 23550 + }, + { + "epoch": 1.3158365226136268, + "grad_norm": 0.3699403703212738, + "learning_rate": 0.00034359592111160917, + "loss": 0.5151, + "step": 23551 + }, + { + "epoch": 1.3158923932173088, + "grad_norm": 0.49870046973228455, + "learning_rate": 0.00034356790676826537, + "loss": 0.4726, + "step": 23552 + }, + { + "epoch": 1.3159482638209905, + "grad_norm": 0.5270695090293884, + "learning_rate": 0.0003435398924249216, + "loss": 0.4442, + "step": 23553 + }, + { + "epoch": 1.3160041344246725, + "grad_norm": 0.3354946970939636, + "learning_rate": 0.0003435118780815778, + "loss": 0.3285, + "step": 23554 + }, + { + "epoch": 1.3160600050283544, + "grad_norm": 0.4447324573993683, + "learning_rate": 0.000343483863738234, + "loss": 0.5073, + "step": 23555 + }, + { + "epoch": 1.3161158756320361, + "grad_norm": 7.126904487609863, + "learning_rate": 0.0003434558493948902, + "loss": 0.4006, + "step": 23556 + }, + { + "epoch": 1.316171746235718, + "grad_norm": 0.4217229187488556, + "learning_rate": 0.0003434278350515464, + "loss": 0.4298, + "step": 23557 + }, + { + "epoch": 1.3162276168394, + "grad_norm": 1.299835205078125, + "learning_rate": 0.0003433998207082026, + "loss": 0.2892, + "step": 23558 + }, + { + "epoch": 1.3162834874430818, + "grad_norm": 2.078361749649048, + "learning_rate": 0.0003433718063648588, + "loss": 0.4126, + "step": 23559 + }, + { + "epoch": 1.3163393580467637, + "grad_norm": 0.5296849608421326, + "learning_rate": 0.000343343792021515, + "loss": 0.4225, + "step": 23560 + }, + { + "epoch": 1.3163952286504457, + "grad_norm": 0.4047715961933136, + "learning_rate": 0.0003433157776781712, + "loss": 0.4242, + "step": 23561 + }, + { + "epoch": 1.3164510992541274, + "grad_norm": 0.42523449659347534, + "learning_rate": 0.00034328776333482743, + "loss": 0.3647, + "step": 23562 + }, + { + "epoch": 1.3165069698578094, + "grad_norm": 0.6092323660850525, + "learning_rate": 0.00034325974899148364, + "loss": 0.4665, + "step": 23563 + }, + { + "epoch": 1.3165628404614913, + "grad_norm": 1.645725131034851, + "learning_rate": 0.00034323173464813984, + "loss": 0.3743, + "step": 23564 + }, + { + "epoch": 1.316618711065173, + "grad_norm": 0.43363380432128906, + "learning_rate": 0.00034320372030479605, + "loss": 0.3496, + "step": 23565 + }, + { + "epoch": 1.316674581668855, + "grad_norm": 0.8278103470802307, + "learning_rate": 0.00034317570596145225, + "loss": 0.4961, + "step": 23566 + }, + { + "epoch": 1.316730452272537, + "grad_norm": 2.9997591972351074, + "learning_rate": 0.0003431476916181085, + "loss": 0.4897, + "step": 23567 + }, + { + "epoch": 1.3167863228762187, + "grad_norm": 1.2639822959899902, + "learning_rate": 0.00034311967727476467, + "loss": 0.5205, + "step": 23568 + }, + { + "epoch": 1.3168421934799006, + "grad_norm": 0.456981360912323, + "learning_rate": 0.0003430916629314209, + "loss": 0.3685, + "step": 23569 + }, + { + "epoch": 1.3168980640835823, + "grad_norm": 0.527363121509552, + "learning_rate": 0.0003430636485880771, + "loss": 0.4055, + "step": 23570 + }, + { + "epoch": 1.3169539346872643, + "grad_norm": 0.6048545241355896, + "learning_rate": 0.00034303563424473334, + "loss": 0.6611, + "step": 23571 + }, + { + "epoch": 1.3170098052909462, + "grad_norm": 0.4473804235458374, + "learning_rate": 0.0003430076199013895, + "loss": 0.4029, + "step": 23572 + }, + { + "epoch": 1.317065675894628, + "grad_norm": 0.4991649389266968, + "learning_rate": 0.00034297960555804575, + "loss": 0.461, + "step": 23573 + }, + { + "epoch": 1.31712154649831, + "grad_norm": 0.5582190752029419, + "learning_rate": 0.00034295159121470195, + "loss": 0.4004, + "step": 23574 + }, + { + "epoch": 1.3171774171019917, + "grad_norm": 0.4616340398788452, + "learning_rate": 0.00034292357687135816, + "loss": 0.4201, + "step": 23575 + }, + { + "epoch": 1.3172332877056736, + "grad_norm": 0.5869447588920593, + "learning_rate": 0.00034289556252801437, + "loss": 0.4228, + "step": 23576 + }, + { + "epoch": 1.3172891583093556, + "grad_norm": 0.44037652015686035, + "learning_rate": 0.00034286754818467057, + "loss": 0.4037, + "step": 23577 + }, + { + "epoch": 1.3173450289130373, + "grad_norm": 0.3264438509941101, + "learning_rate": 0.0003428395338413268, + "loss": 0.329, + "step": 23578 + }, + { + "epoch": 1.3174008995167192, + "grad_norm": 0.3850671350955963, + "learning_rate": 0.00034281151949798293, + "loss": 0.3876, + "step": 23579 + }, + { + "epoch": 1.3174567701204012, + "grad_norm": 0.5215572714805603, + "learning_rate": 0.0003427835051546392, + "loss": 0.4039, + "step": 23580 + }, + { + "epoch": 1.317512640724083, + "grad_norm": 0.4167189598083496, + "learning_rate": 0.00034275549081129534, + "loss": 0.3987, + "step": 23581 + }, + { + "epoch": 1.3175685113277649, + "grad_norm": 0.3528139293193817, + "learning_rate": 0.0003427274764679516, + "loss": 0.4177, + "step": 23582 + }, + { + "epoch": 1.3176243819314468, + "grad_norm": 0.4976060390472412, + "learning_rate": 0.0003426994621246078, + "loss": 0.5084, + "step": 23583 + }, + { + "epoch": 1.3176802525351285, + "grad_norm": 0.5247829556465149, + "learning_rate": 0.000342671447781264, + "loss": 0.4926, + "step": 23584 + }, + { + "epoch": 1.3177361231388105, + "grad_norm": 0.5440949201583862, + "learning_rate": 0.0003426434334379202, + "loss": 0.3975, + "step": 23585 + }, + { + "epoch": 1.3177919937424925, + "grad_norm": 0.7016050815582275, + "learning_rate": 0.0003426154190945764, + "loss": 0.4126, + "step": 23586 + }, + { + "epoch": 1.3178478643461742, + "grad_norm": 0.42228344082832336, + "learning_rate": 0.00034258740475123263, + "loss": 0.3892, + "step": 23587 + }, + { + "epoch": 1.3179037349498561, + "grad_norm": 0.7191487550735474, + "learning_rate": 0.00034255939040788884, + "loss": 0.4729, + "step": 23588 + }, + { + "epoch": 1.317959605553538, + "grad_norm": 0.3287924826145172, + "learning_rate": 0.00034253137606454504, + "loss": 0.2671, + "step": 23589 + }, + { + "epoch": 1.3180154761572198, + "grad_norm": 0.4276678264141083, + "learning_rate": 0.0003425033617212013, + "loss": 0.4594, + "step": 23590 + }, + { + "epoch": 1.3180713467609018, + "grad_norm": 0.3954083025455475, + "learning_rate": 0.00034247534737785745, + "loss": 0.3785, + "step": 23591 + }, + { + "epoch": 1.3181272173645837, + "grad_norm": 0.6668315529823303, + "learning_rate": 0.0003424473330345137, + "loss": 0.5844, + "step": 23592 + }, + { + "epoch": 1.3181830879682654, + "grad_norm": 0.6298591494560242, + "learning_rate": 0.00034241931869116986, + "loss": 0.404, + "step": 23593 + }, + { + "epoch": 1.3182389585719474, + "grad_norm": 0.8551307916641235, + "learning_rate": 0.0003423913043478261, + "loss": 0.4211, + "step": 23594 + }, + { + "epoch": 1.3182948291756293, + "grad_norm": 0.39548906683921814, + "learning_rate": 0.0003423632900044823, + "loss": 0.3712, + "step": 23595 + }, + { + "epoch": 1.318350699779311, + "grad_norm": 0.7923518419265747, + "learning_rate": 0.00034233527566113854, + "loss": 0.5325, + "step": 23596 + }, + { + "epoch": 1.318406570382993, + "grad_norm": 0.3318021893501282, + "learning_rate": 0.00034230726131779474, + "loss": 0.3805, + "step": 23597 + }, + { + "epoch": 1.318462440986675, + "grad_norm": 1.2421467304229736, + "learning_rate": 0.00034227924697445095, + "loss": 0.5489, + "step": 23598 + }, + { + "epoch": 1.3185183115903567, + "grad_norm": 0.4162934124469757, + "learning_rate": 0.00034225123263110715, + "loss": 0.4584, + "step": 23599 + }, + { + "epoch": 1.3185741821940387, + "grad_norm": 0.44177907705307007, + "learning_rate": 0.0003422232182877633, + "loss": 0.4427, + "step": 23600 + }, + { + "epoch": 1.3186300527977206, + "grad_norm": 0.28752201795578003, + "learning_rate": 0.00034219520394441957, + "loss": 0.3287, + "step": 23601 + }, + { + "epoch": 1.3186859234014023, + "grad_norm": 0.4075329303741455, + "learning_rate": 0.0003421671896010757, + "loss": 0.4512, + "step": 23602 + }, + { + "epoch": 1.3187417940050843, + "grad_norm": 3.741101026535034, + "learning_rate": 0.000342139175257732, + "loss": 0.6237, + "step": 23603 + }, + { + "epoch": 1.318797664608766, + "grad_norm": 0.5826823711395264, + "learning_rate": 0.00034211116091438813, + "loss": 0.394, + "step": 23604 + }, + { + "epoch": 1.318853535212448, + "grad_norm": 0.7225218415260315, + "learning_rate": 0.0003420831465710444, + "loss": 0.3728, + "step": 23605 + }, + { + "epoch": 1.31890940581613, + "grad_norm": 0.5438814759254456, + "learning_rate": 0.0003420551322277006, + "loss": 0.4302, + "step": 23606 + }, + { + "epoch": 1.3189652764198116, + "grad_norm": 0.3892956078052521, + "learning_rate": 0.0003420271178843568, + "loss": 0.4478, + "step": 23607 + }, + { + "epoch": 1.3190211470234936, + "grad_norm": 0.6349735856056213, + "learning_rate": 0.000341999103541013, + "loss": 0.4161, + "step": 23608 + }, + { + "epoch": 1.3190770176271753, + "grad_norm": 0.3495505750179291, + "learning_rate": 0.0003419710891976692, + "loss": 0.3316, + "step": 23609 + }, + { + "epoch": 1.3191328882308573, + "grad_norm": 0.4075941741466522, + "learning_rate": 0.0003419430748543254, + "loss": 0.3919, + "step": 23610 + }, + { + "epoch": 1.3191887588345392, + "grad_norm": 0.5787622928619385, + "learning_rate": 0.0003419150605109816, + "loss": 0.4799, + "step": 23611 + }, + { + "epoch": 1.319244629438221, + "grad_norm": 0.4812084138393402, + "learning_rate": 0.00034188704616763783, + "loss": 0.4089, + "step": 23612 + }, + { + "epoch": 1.319300500041903, + "grad_norm": 0.41686397790908813, + "learning_rate": 0.0003418590318242941, + "loss": 0.4251, + "step": 23613 + }, + { + "epoch": 1.3193563706455849, + "grad_norm": 0.433151513338089, + "learning_rate": 0.00034183101748095024, + "loss": 0.3882, + "step": 23614 + }, + { + "epoch": 1.3194122412492666, + "grad_norm": 0.30525463819503784, + "learning_rate": 0.0003418030031376065, + "loss": 0.4363, + "step": 23615 + }, + { + "epoch": 1.3194681118529485, + "grad_norm": 0.3429461419582367, + "learning_rate": 0.00034177498879426265, + "loss": 0.3792, + "step": 23616 + }, + { + "epoch": 1.3195239824566305, + "grad_norm": 0.794349730014801, + "learning_rate": 0.0003417469744509189, + "loss": 0.4572, + "step": 23617 + }, + { + "epoch": 1.3195798530603122, + "grad_norm": 0.5123017430305481, + "learning_rate": 0.00034171896010757506, + "loss": 0.601, + "step": 23618 + }, + { + "epoch": 1.3196357236639942, + "grad_norm": 0.40440672636032104, + "learning_rate": 0.0003416909457642313, + "loss": 0.455, + "step": 23619 + }, + { + "epoch": 1.3196915942676761, + "grad_norm": 0.43828997015953064, + "learning_rate": 0.0003416629314208875, + "loss": 0.4098, + "step": 23620 + }, + { + "epoch": 1.3197474648713579, + "grad_norm": 0.6250551342964172, + "learning_rate": 0.0003416349170775437, + "loss": 0.473, + "step": 23621 + }, + { + "epoch": 1.3198033354750398, + "grad_norm": 1.2103159427642822, + "learning_rate": 0.00034160690273419994, + "loss": 0.5058, + "step": 23622 + }, + { + "epoch": 1.3198592060787218, + "grad_norm": 0.634890079498291, + "learning_rate": 0.0003415788883908561, + "loss": 0.5216, + "step": 23623 + }, + { + "epoch": 1.3199150766824035, + "grad_norm": 0.34890466928482056, + "learning_rate": 0.00034155087404751235, + "loss": 0.3071, + "step": 23624 + }, + { + "epoch": 1.3199709472860854, + "grad_norm": 0.7785325050354004, + "learning_rate": 0.0003415228597041685, + "loss": 0.4654, + "step": 23625 + }, + { + "epoch": 1.3200268178897674, + "grad_norm": 0.3444153070449829, + "learning_rate": 0.00034149484536082477, + "loss": 0.3331, + "step": 23626 + }, + { + "epoch": 1.3200826884934491, + "grad_norm": 0.4434806704521179, + "learning_rate": 0.0003414668310174809, + "loss": 0.6102, + "step": 23627 + }, + { + "epoch": 1.320138559097131, + "grad_norm": 0.36038872599601746, + "learning_rate": 0.0003414388166741372, + "loss": 0.3715, + "step": 23628 + }, + { + "epoch": 1.320194429700813, + "grad_norm": 0.6076656579971313, + "learning_rate": 0.0003414108023307934, + "loss": 0.5076, + "step": 23629 + }, + { + "epoch": 1.3202503003044948, + "grad_norm": 0.4611376225948334, + "learning_rate": 0.0003413827879874496, + "loss": 0.4005, + "step": 23630 + }, + { + "epoch": 1.3203061709081767, + "grad_norm": 7.565454959869385, + "learning_rate": 0.0003413547736441058, + "loss": 0.4051, + "step": 23631 + }, + { + "epoch": 1.3203620415118587, + "grad_norm": 0.5885747671127319, + "learning_rate": 0.000341326759300762, + "loss": 0.4513, + "step": 23632 + }, + { + "epoch": 1.3204179121155404, + "grad_norm": 3.759124517440796, + "learning_rate": 0.0003412987449574182, + "loss": 0.4347, + "step": 23633 + }, + { + "epoch": 1.3204737827192223, + "grad_norm": 0.711495578289032, + "learning_rate": 0.0003412707306140744, + "loss": 0.3839, + "step": 23634 + }, + { + "epoch": 1.3205296533229043, + "grad_norm": 0.7495306730270386, + "learning_rate": 0.0003412427162707306, + "loss": 0.3953, + "step": 23635 + }, + { + "epoch": 1.320585523926586, + "grad_norm": 2.244140625, + "learning_rate": 0.0003412147019273868, + "loss": 0.4029, + "step": 23636 + }, + { + "epoch": 1.320641394530268, + "grad_norm": 0.33000096678733826, + "learning_rate": 0.00034118668758404303, + "loss": 0.4269, + "step": 23637 + }, + { + "epoch": 1.3206972651339497, + "grad_norm": 0.49355843663215637, + "learning_rate": 0.0003411586732406993, + "loss": 0.3672, + "step": 23638 + }, + { + "epoch": 1.3207531357376316, + "grad_norm": 0.3391716182231903, + "learning_rate": 0.00034113065889735544, + "loss": 0.363, + "step": 23639 + }, + { + "epoch": 1.3208090063413136, + "grad_norm": 1.2276924848556519, + "learning_rate": 0.0003411026445540117, + "loss": 0.3744, + "step": 23640 + }, + { + "epoch": 1.3208648769449953, + "grad_norm": 0.48475953936576843, + "learning_rate": 0.00034107463021066785, + "loss": 0.4652, + "step": 23641 + }, + { + "epoch": 1.3209207475486773, + "grad_norm": 0.38536372780799866, + "learning_rate": 0.00034104661586732406, + "loss": 0.354, + "step": 23642 + }, + { + "epoch": 1.320976618152359, + "grad_norm": 1.0772712230682373, + "learning_rate": 0.00034101860152398026, + "loss": 0.4628, + "step": 23643 + }, + { + "epoch": 1.321032488756041, + "grad_norm": 0.5036483407020569, + "learning_rate": 0.00034099058718063647, + "loss": 0.4421, + "step": 23644 + }, + { + "epoch": 1.321088359359723, + "grad_norm": 0.43062624335289, + "learning_rate": 0.00034096257283729273, + "loss": 0.3613, + "step": 23645 + }, + { + "epoch": 1.3211442299634046, + "grad_norm": 0.47623303532600403, + "learning_rate": 0.0003409345584939489, + "loss": 0.4454, + "step": 23646 + }, + { + "epoch": 1.3212001005670866, + "grad_norm": 0.4240516424179077, + "learning_rate": 0.00034090654415060514, + "loss": 0.518, + "step": 23647 + }, + { + "epoch": 1.3212559711707685, + "grad_norm": 0.33697769045829773, + "learning_rate": 0.0003408785298072613, + "loss": 0.3369, + "step": 23648 + }, + { + "epoch": 1.3213118417744503, + "grad_norm": 4.921535491943359, + "learning_rate": 0.00034085051546391755, + "loss": 0.4289, + "step": 23649 + }, + { + "epoch": 1.3213677123781322, + "grad_norm": 0.8781418800354004, + "learning_rate": 0.0003408225011205737, + "loss": 0.4687, + "step": 23650 + }, + { + "epoch": 1.3214235829818142, + "grad_norm": 0.6379905343055725, + "learning_rate": 0.00034079448677722996, + "loss": 0.3894, + "step": 23651 + }, + { + "epoch": 1.321479453585496, + "grad_norm": 0.5977003574371338, + "learning_rate": 0.0003407664724338861, + "loss": 0.4371, + "step": 23652 + }, + { + "epoch": 1.3215353241891779, + "grad_norm": 1.7605507373809814, + "learning_rate": 0.0003407384580905424, + "loss": 0.3336, + "step": 23653 + }, + { + "epoch": 1.3215911947928598, + "grad_norm": 2.8999063968658447, + "learning_rate": 0.0003407104437471986, + "loss": 0.4708, + "step": 23654 + }, + { + "epoch": 1.3216470653965415, + "grad_norm": 0.4048449397087097, + "learning_rate": 0.0003406824294038548, + "loss": 0.5199, + "step": 23655 + }, + { + "epoch": 1.3217029360002235, + "grad_norm": 0.3568628132343292, + "learning_rate": 0.000340654415060511, + "loss": 0.3341, + "step": 23656 + }, + { + "epoch": 1.3217588066039054, + "grad_norm": 0.5233575701713562, + "learning_rate": 0.0003406264007171672, + "loss": 0.4949, + "step": 23657 + }, + { + "epoch": 1.3218146772075872, + "grad_norm": 0.6682443022727966, + "learning_rate": 0.0003405983863738234, + "loss": 0.4221, + "step": 23658 + }, + { + "epoch": 1.3218705478112691, + "grad_norm": 5.685514450073242, + "learning_rate": 0.0003405703720304796, + "loss": 0.6446, + "step": 23659 + }, + { + "epoch": 1.321926418414951, + "grad_norm": 0.668967604637146, + "learning_rate": 0.0003405423576871358, + "loss": 0.4123, + "step": 23660 + }, + { + "epoch": 1.3219822890186328, + "grad_norm": 0.5231613516807556, + "learning_rate": 0.0003405143433437921, + "loss": 0.432, + "step": 23661 + }, + { + "epoch": 1.3220381596223147, + "grad_norm": 0.5050115585327148, + "learning_rate": 0.00034048632900044823, + "loss": 0.4467, + "step": 23662 + }, + { + "epoch": 1.3220940302259967, + "grad_norm": 0.4238983392715454, + "learning_rate": 0.00034045831465710443, + "loss": 0.4717, + "step": 23663 + }, + { + "epoch": 1.3221499008296784, + "grad_norm": 0.32803571224212646, + "learning_rate": 0.00034043030031376064, + "loss": 0.2904, + "step": 23664 + }, + { + "epoch": 1.3222057714333604, + "grad_norm": 0.6409878134727478, + "learning_rate": 0.00034040228597041685, + "loss": 0.3894, + "step": 23665 + }, + { + "epoch": 1.3222616420370423, + "grad_norm": 0.39968976378440857, + "learning_rate": 0.00034037427162707305, + "loss": 0.4538, + "step": 23666 + }, + { + "epoch": 1.322317512640724, + "grad_norm": 1.119011640548706, + "learning_rate": 0.00034034625728372926, + "loss": 0.5454, + "step": 23667 + }, + { + "epoch": 1.322373383244406, + "grad_norm": 0.40956932306289673, + "learning_rate": 0.0003403182429403855, + "loss": 0.3792, + "step": 23668 + }, + { + "epoch": 1.322429253848088, + "grad_norm": 1.202910304069519, + "learning_rate": 0.00034029022859704167, + "loss": 0.3942, + "step": 23669 + }, + { + "epoch": 1.3224851244517697, + "grad_norm": 0.7594714164733887, + "learning_rate": 0.00034026221425369793, + "loss": 0.4855, + "step": 23670 + }, + { + "epoch": 1.3225409950554516, + "grad_norm": 0.7461695075035095, + "learning_rate": 0.0003402341999103541, + "loss": 0.4705, + "step": 23671 + }, + { + "epoch": 1.3225968656591334, + "grad_norm": 0.6346709132194519, + "learning_rate": 0.00034020618556701034, + "loss": 0.3685, + "step": 23672 + }, + { + "epoch": 1.3226527362628153, + "grad_norm": 0.3901936709880829, + "learning_rate": 0.0003401781712236665, + "loss": 0.4863, + "step": 23673 + }, + { + "epoch": 1.3227086068664973, + "grad_norm": 0.40705859661102295, + "learning_rate": 0.00034015015688032275, + "loss": 0.3789, + "step": 23674 + }, + { + "epoch": 1.322764477470179, + "grad_norm": 0.3566629886627197, + "learning_rate": 0.0003401221425369789, + "loss": 0.4327, + "step": 23675 + }, + { + "epoch": 1.322820348073861, + "grad_norm": 0.4028332829475403, + "learning_rate": 0.00034009412819363516, + "loss": 0.4309, + "step": 23676 + }, + { + "epoch": 1.3228762186775427, + "grad_norm": 0.3589525520801544, + "learning_rate": 0.00034006611385029137, + "loss": 0.3746, + "step": 23677 + }, + { + "epoch": 1.3229320892812246, + "grad_norm": 0.43733587861061096, + "learning_rate": 0.0003400380995069476, + "loss": 0.3147, + "step": 23678 + }, + { + "epoch": 1.3229879598849066, + "grad_norm": 0.4860137104988098, + "learning_rate": 0.0003400100851636038, + "loss": 0.3665, + "step": 23679 + }, + { + "epoch": 1.3230438304885883, + "grad_norm": 10.6921968460083, + "learning_rate": 0.00033998207082026, + "loss": 0.4895, + "step": 23680 + }, + { + "epoch": 1.3230997010922703, + "grad_norm": 0.48251381516456604, + "learning_rate": 0.0003399540564769162, + "loss": 0.5525, + "step": 23681 + }, + { + "epoch": 1.3231555716959522, + "grad_norm": 0.5078519582748413, + "learning_rate": 0.0003399260421335724, + "loss": 0.4298, + "step": 23682 + }, + { + "epoch": 1.323211442299634, + "grad_norm": 1.7563810348510742, + "learning_rate": 0.0003398980277902286, + "loss": 0.3696, + "step": 23683 + }, + { + "epoch": 1.323267312903316, + "grad_norm": 0.6586446166038513, + "learning_rate": 0.0003398700134468848, + "loss": 0.5209, + "step": 23684 + }, + { + "epoch": 1.3233231835069978, + "grad_norm": 0.5308419466018677, + "learning_rate": 0.000339841999103541, + "loss": 0.4796, + "step": 23685 + }, + { + "epoch": 1.3233790541106796, + "grad_norm": 0.36318719387054443, + "learning_rate": 0.0003398139847601972, + "loss": 0.4076, + "step": 23686 + }, + { + "epoch": 1.3234349247143615, + "grad_norm": 0.8732465505599976, + "learning_rate": 0.00033978597041685343, + "loss": 0.463, + "step": 23687 + }, + { + "epoch": 1.3234907953180435, + "grad_norm": 1.9823834896087646, + "learning_rate": 0.00033975795607350963, + "loss": 0.4978, + "step": 23688 + }, + { + "epoch": 1.3235466659217252, + "grad_norm": 0.41143447160720825, + "learning_rate": 0.00033972994173016584, + "loss": 0.3756, + "step": 23689 + }, + { + "epoch": 1.3236025365254072, + "grad_norm": 0.3460372984409332, + "learning_rate": 0.00033970192738682205, + "loss": 0.4032, + "step": 23690 + }, + { + "epoch": 1.323658407129089, + "grad_norm": 0.743229329586029, + "learning_rate": 0.00033967391304347825, + "loss": 0.4967, + "step": 23691 + }, + { + "epoch": 1.3237142777327708, + "grad_norm": 0.6544398069381714, + "learning_rate": 0.00033964589870013446, + "loss": 0.4114, + "step": 23692 + }, + { + "epoch": 1.3237701483364528, + "grad_norm": 0.4673772156238556, + "learning_rate": 0.0003396178843567907, + "loss": 0.3794, + "step": 23693 + }, + { + "epoch": 1.3238260189401347, + "grad_norm": 0.3888319432735443, + "learning_rate": 0.00033958987001344687, + "loss": 0.3825, + "step": 23694 + }, + { + "epoch": 1.3238818895438165, + "grad_norm": 0.4703053832054138, + "learning_rate": 0.00033956185567010313, + "loss": 0.5051, + "step": 23695 + }, + { + "epoch": 1.3239377601474984, + "grad_norm": 0.4253244698047638, + "learning_rate": 0.0003395338413267593, + "loss": 0.4093, + "step": 23696 + }, + { + "epoch": 1.3239936307511804, + "grad_norm": 0.4492633640766144, + "learning_rate": 0.00033950582698341554, + "loss": 0.4155, + "step": 23697 + }, + { + "epoch": 1.324049501354862, + "grad_norm": 0.5175361633300781, + "learning_rate": 0.0003394778126400717, + "loss": 0.3711, + "step": 23698 + }, + { + "epoch": 1.324105371958544, + "grad_norm": 0.6385105848312378, + "learning_rate": 0.00033944979829672795, + "loss": 0.4023, + "step": 23699 + }, + { + "epoch": 1.324161242562226, + "grad_norm": 0.48967647552490234, + "learning_rate": 0.00033942178395338416, + "loss": 0.5177, + "step": 23700 + }, + { + "epoch": 1.3242171131659077, + "grad_norm": 0.43169471621513367, + "learning_rate": 0.00033939376961004036, + "loss": 0.3273, + "step": 23701 + }, + { + "epoch": 1.3242729837695897, + "grad_norm": 0.5104267001152039, + "learning_rate": 0.00033936575526669657, + "loss": 0.4677, + "step": 23702 + }, + { + "epoch": 1.3243288543732716, + "grad_norm": 1.2873023748397827, + "learning_rate": 0.0003393377409233528, + "loss": 0.5368, + "step": 23703 + }, + { + "epoch": 1.3243847249769534, + "grad_norm": 0.346524715423584, + "learning_rate": 0.000339309726580009, + "loss": 0.3899, + "step": 23704 + }, + { + "epoch": 1.3244405955806353, + "grad_norm": 0.4862247407436371, + "learning_rate": 0.00033928171223666513, + "loss": 0.5089, + "step": 23705 + }, + { + "epoch": 1.324496466184317, + "grad_norm": 0.47750619053840637, + "learning_rate": 0.0003392536978933214, + "loss": 0.3983, + "step": 23706 + }, + { + "epoch": 1.324552336787999, + "grad_norm": 0.3553129732608795, + "learning_rate": 0.00033922568354997754, + "loss": 0.3478, + "step": 23707 + }, + { + "epoch": 1.324608207391681, + "grad_norm": 2.9279370307922363, + "learning_rate": 0.0003391976692066338, + "loss": 0.3637, + "step": 23708 + }, + { + "epoch": 1.3246640779953627, + "grad_norm": 0.96209716796875, + "learning_rate": 0.00033916965486329, + "loss": 0.7348, + "step": 23709 + }, + { + "epoch": 1.3247199485990446, + "grad_norm": 0.8014591336250305, + "learning_rate": 0.0003391416405199462, + "loss": 0.4366, + "step": 23710 + }, + { + "epoch": 1.3247758192027264, + "grad_norm": 0.31079208850860596, + "learning_rate": 0.0003391136261766024, + "loss": 0.3865, + "step": 23711 + }, + { + "epoch": 1.3248316898064083, + "grad_norm": 0.36179718375205994, + "learning_rate": 0.00033908561183325863, + "loss": 0.3617, + "step": 23712 + }, + { + "epoch": 1.3248875604100903, + "grad_norm": 0.8673931956291199, + "learning_rate": 0.00033905759748991483, + "loss": 0.4547, + "step": 23713 + }, + { + "epoch": 1.324943431013772, + "grad_norm": 0.4933382272720337, + "learning_rate": 0.00033902958314657104, + "loss": 0.3936, + "step": 23714 + }, + { + "epoch": 1.324999301617454, + "grad_norm": 0.4393457770347595, + "learning_rate": 0.00033900156880322725, + "loss": 0.4591, + "step": 23715 + }, + { + "epoch": 1.325055172221136, + "grad_norm": 0.46617335081100464, + "learning_rate": 0.0003389735544598835, + "loss": 0.4886, + "step": 23716 + }, + { + "epoch": 1.3251110428248176, + "grad_norm": 0.45171883702278137, + "learning_rate": 0.00033894554011653966, + "loss": 0.4733, + "step": 23717 + }, + { + "epoch": 1.3251669134284996, + "grad_norm": 0.4940762221813202, + "learning_rate": 0.0003389175257731959, + "loss": 0.5184, + "step": 23718 + }, + { + "epoch": 1.3252227840321815, + "grad_norm": 0.30718106031417847, + "learning_rate": 0.00033888951142985207, + "loss": 0.3046, + "step": 23719 + }, + { + "epoch": 1.3252786546358633, + "grad_norm": 0.8073253035545349, + "learning_rate": 0.00033886149708650833, + "loss": 0.3351, + "step": 23720 + }, + { + "epoch": 1.3253345252395452, + "grad_norm": 0.7107348442077637, + "learning_rate": 0.0003388334827431645, + "loss": 0.3617, + "step": 23721 + }, + { + "epoch": 1.3253903958432272, + "grad_norm": 0.46475228667259216, + "learning_rate": 0.00033880546839982074, + "loss": 0.334, + "step": 23722 + }, + { + "epoch": 1.3254462664469089, + "grad_norm": 0.6563297510147095, + "learning_rate": 0.00033877745405647695, + "loss": 0.5026, + "step": 23723 + }, + { + "epoch": 1.3255021370505908, + "grad_norm": 1.5969041585922241, + "learning_rate": 0.00033874943971313315, + "loss": 0.4689, + "step": 23724 + }, + { + "epoch": 1.3255580076542728, + "grad_norm": 0.6964733600616455, + "learning_rate": 0.00033872142536978936, + "loss": 0.5632, + "step": 23725 + }, + { + "epoch": 1.3256138782579545, + "grad_norm": 0.5347293615341187, + "learning_rate": 0.00033869341102644556, + "loss": 0.4518, + "step": 23726 + }, + { + "epoch": 1.3256697488616365, + "grad_norm": 0.6473166942596436, + "learning_rate": 0.00033866539668310177, + "loss": 0.4416, + "step": 23727 + }, + { + "epoch": 1.3257256194653184, + "grad_norm": 0.27681422233581543, + "learning_rate": 0.0003386373823397579, + "loss": 0.3244, + "step": 23728 + }, + { + "epoch": 1.3257814900690001, + "grad_norm": 0.5681509971618652, + "learning_rate": 0.0003386093679964142, + "loss": 0.4358, + "step": 23729 + }, + { + "epoch": 1.325837360672682, + "grad_norm": 0.5241336822509766, + "learning_rate": 0.00033858135365307033, + "loss": 0.4289, + "step": 23730 + }, + { + "epoch": 1.325893231276364, + "grad_norm": 1.0116266012191772, + "learning_rate": 0.0003385533393097266, + "loss": 0.4368, + "step": 23731 + }, + { + "epoch": 1.3259491018800458, + "grad_norm": 0.350147545337677, + "learning_rate": 0.0003385253249663828, + "loss": 0.4125, + "step": 23732 + }, + { + "epoch": 1.3260049724837277, + "grad_norm": 0.47547584772109985, + "learning_rate": 0.000338497310623039, + "loss": 0.4063, + "step": 23733 + }, + { + "epoch": 1.3260608430874097, + "grad_norm": 0.4203178882598877, + "learning_rate": 0.0003384692962796952, + "loss": 0.4199, + "step": 23734 + }, + { + "epoch": 1.3261167136910914, + "grad_norm": 0.612237274646759, + "learning_rate": 0.0003384412819363514, + "loss": 0.3818, + "step": 23735 + }, + { + "epoch": 1.3261725842947734, + "grad_norm": 0.3352597951889038, + "learning_rate": 0.0003384132675930076, + "loss": 0.3377, + "step": 23736 + }, + { + "epoch": 1.3262284548984553, + "grad_norm": 1.944063663482666, + "learning_rate": 0.00033838525324966383, + "loss": 0.4337, + "step": 23737 + }, + { + "epoch": 1.326284325502137, + "grad_norm": 0.4641205668449402, + "learning_rate": 0.00033835723890632003, + "loss": 0.5336, + "step": 23738 + }, + { + "epoch": 1.326340196105819, + "grad_norm": 2.928910493850708, + "learning_rate": 0.0003383292245629763, + "loss": 0.3763, + "step": 23739 + }, + { + "epoch": 1.3263960667095007, + "grad_norm": 0.5527085661888123, + "learning_rate": 0.00033830121021963244, + "loss": 0.4656, + "step": 23740 + }, + { + "epoch": 1.3264519373131827, + "grad_norm": 0.40862295031547546, + "learning_rate": 0.0003382731958762887, + "loss": 0.3698, + "step": 23741 + }, + { + "epoch": 1.3265078079168646, + "grad_norm": 0.3980526924133301, + "learning_rate": 0.00033824518153294486, + "loss": 0.3864, + "step": 23742 + }, + { + "epoch": 1.3265636785205464, + "grad_norm": 1.3795419931411743, + "learning_rate": 0.0003382171671896011, + "loss": 0.4644, + "step": 23743 + }, + { + "epoch": 1.3266195491242283, + "grad_norm": 0.720112144947052, + "learning_rate": 0.00033818915284625727, + "loss": 0.4139, + "step": 23744 + }, + { + "epoch": 1.32667541972791, + "grad_norm": 0.686043381690979, + "learning_rate": 0.00033816113850291353, + "loss": 0.4665, + "step": 23745 + }, + { + "epoch": 1.326731290331592, + "grad_norm": 0.6251727342605591, + "learning_rate": 0.0003381331241595697, + "loss": 0.4342, + "step": 23746 + }, + { + "epoch": 1.326787160935274, + "grad_norm": 0.7605039477348328, + "learning_rate": 0.00033810510981622594, + "loss": 0.4526, + "step": 23747 + }, + { + "epoch": 1.3268430315389557, + "grad_norm": 0.5772928595542908, + "learning_rate": 0.00033807709547288215, + "loss": 0.426, + "step": 23748 + }, + { + "epoch": 1.3268989021426376, + "grad_norm": 0.49635547399520874, + "learning_rate": 0.0003380490811295383, + "loss": 0.5862, + "step": 23749 + }, + { + "epoch": 1.3269547727463196, + "grad_norm": 0.4543931186199188, + "learning_rate": 0.00033802106678619456, + "loss": 0.4115, + "step": 23750 + }, + { + "epoch": 1.3270106433500013, + "grad_norm": 0.4984850585460663, + "learning_rate": 0.0003379930524428507, + "loss": 0.412, + "step": 23751 + }, + { + "epoch": 1.3270665139536832, + "grad_norm": 0.472870796918869, + "learning_rate": 0.00033796503809950697, + "loss": 0.463, + "step": 23752 + }, + { + "epoch": 1.3271223845573652, + "grad_norm": 0.4974968135356903, + "learning_rate": 0.0003379370237561631, + "loss": 0.4667, + "step": 23753 + }, + { + "epoch": 1.327178255161047, + "grad_norm": 0.4373999834060669, + "learning_rate": 0.0003379090094128194, + "loss": 0.3612, + "step": 23754 + }, + { + "epoch": 1.3272341257647289, + "grad_norm": 0.4713369607925415, + "learning_rate": 0.0003378809950694756, + "loss": 0.3915, + "step": 23755 + }, + { + "epoch": 1.3272899963684108, + "grad_norm": 0.7131215333938599, + "learning_rate": 0.0003378529807261318, + "loss": 0.3855, + "step": 23756 + }, + { + "epoch": 1.3273458669720926, + "grad_norm": 1.8658697605133057, + "learning_rate": 0.000337824966382788, + "loss": 0.4595, + "step": 23757 + }, + { + "epoch": 1.3274017375757745, + "grad_norm": 0.3171537518501282, + "learning_rate": 0.0003377969520394442, + "loss": 0.4168, + "step": 23758 + }, + { + "epoch": 1.3274576081794565, + "grad_norm": 0.6010354161262512, + "learning_rate": 0.0003377689376961004, + "loss": 0.5022, + "step": 23759 + }, + { + "epoch": 1.3275134787831382, + "grad_norm": 0.5116702914237976, + "learning_rate": 0.0003377409233527566, + "loss": 0.4165, + "step": 23760 + }, + { + "epoch": 1.3275693493868201, + "grad_norm": 0.6689817905426025, + "learning_rate": 0.0003377129090094128, + "loss": 0.3955, + "step": 23761 + }, + { + "epoch": 1.327625219990502, + "grad_norm": 0.7327821254730225, + "learning_rate": 0.000337684894666069, + "loss": 0.5668, + "step": 23762 + }, + { + "epoch": 1.3276810905941838, + "grad_norm": 0.7489685416221619, + "learning_rate": 0.00033765688032272523, + "loss": 0.4692, + "step": 23763 + }, + { + "epoch": 1.3277369611978658, + "grad_norm": 0.4186297655105591, + "learning_rate": 0.0003376288659793815, + "loss": 0.5084, + "step": 23764 + }, + { + "epoch": 1.3277928318015477, + "grad_norm": 0.39536789059638977, + "learning_rate": 0.00033760085163603764, + "loss": 0.4036, + "step": 23765 + }, + { + "epoch": 1.3278487024052295, + "grad_norm": 0.28467315435409546, + "learning_rate": 0.0003375728372926939, + "loss": 0.3175, + "step": 23766 + }, + { + "epoch": 1.3279045730089114, + "grad_norm": 0.4411020278930664, + "learning_rate": 0.00033754482294935006, + "loss": 0.432, + "step": 23767 + }, + { + "epoch": 1.3279604436125934, + "grad_norm": 0.44874507188796997, + "learning_rate": 0.0003375168086060063, + "loss": 0.3957, + "step": 23768 + }, + { + "epoch": 1.328016314216275, + "grad_norm": 0.47929778695106506, + "learning_rate": 0.00033748879426266247, + "loss": 0.3867, + "step": 23769 + }, + { + "epoch": 1.328072184819957, + "grad_norm": 0.4720273017883301, + "learning_rate": 0.0003374607799193187, + "loss": 0.4498, + "step": 23770 + }, + { + "epoch": 1.328128055423639, + "grad_norm": 0.7578483819961548, + "learning_rate": 0.00033743276557597493, + "loss": 0.4438, + "step": 23771 + }, + { + "epoch": 1.3281839260273207, + "grad_norm": 0.8227038979530334, + "learning_rate": 0.0003374047512326311, + "loss": 0.5773, + "step": 23772 + }, + { + "epoch": 1.3282397966310027, + "grad_norm": 0.590437650680542, + "learning_rate": 0.00033737673688928735, + "loss": 0.453, + "step": 23773 + }, + { + "epoch": 1.3282956672346844, + "grad_norm": 0.4821312129497528, + "learning_rate": 0.0003373487225459435, + "loss": 0.3969, + "step": 23774 + }, + { + "epoch": 1.3283515378383663, + "grad_norm": 0.40414074063301086, + "learning_rate": 0.00033732070820259976, + "loss": 0.3834, + "step": 23775 + }, + { + "epoch": 1.3284074084420483, + "grad_norm": 0.37725040316581726, + "learning_rate": 0.0003372926938592559, + "loss": 0.415, + "step": 23776 + }, + { + "epoch": 1.32846327904573, + "grad_norm": 0.47734546661376953, + "learning_rate": 0.00033726467951591217, + "loss": 0.5742, + "step": 23777 + }, + { + "epoch": 1.328519149649412, + "grad_norm": 0.48547959327697754, + "learning_rate": 0.0003372366651725683, + "loss": 0.3643, + "step": 23778 + }, + { + "epoch": 1.3285750202530937, + "grad_norm": 0.4639904797077179, + "learning_rate": 0.0003372086508292246, + "loss": 0.3874, + "step": 23779 + }, + { + "epoch": 1.3286308908567757, + "grad_norm": 0.4941656291484833, + "learning_rate": 0.0003371806364858808, + "loss": 0.3502, + "step": 23780 + }, + { + "epoch": 1.3286867614604576, + "grad_norm": 0.3892178237438202, + "learning_rate": 0.000337152622142537, + "loss": 0.4426, + "step": 23781 + }, + { + "epoch": 1.3287426320641393, + "grad_norm": 2.824342727661133, + "learning_rate": 0.0003371246077991932, + "loss": 0.3679, + "step": 23782 + }, + { + "epoch": 1.3287985026678213, + "grad_norm": 0.4226418733596802, + "learning_rate": 0.0003370965934558494, + "loss": 0.5342, + "step": 23783 + }, + { + "epoch": 1.3288543732715032, + "grad_norm": 0.6223406195640564, + "learning_rate": 0.0003370685791125056, + "loss": 0.536, + "step": 23784 + }, + { + "epoch": 1.328910243875185, + "grad_norm": 0.35209837555885315, + "learning_rate": 0.0003370405647691618, + "loss": 0.3488, + "step": 23785 + }, + { + "epoch": 1.328966114478867, + "grad_norm": 0.5457257032394409, + "learning_rate": 0.000337012550425818, + "loss": 0.3934, + "step": 23786 + }, + { + "epoch": 1.3290219850825489, + "grad_norm": 0.3860602378845215, + "learning_rate": 0.0003369845360824743, + "loss": 0.36, + "step": 23787 + }, + { + "epoch": 1.3290778556862306, + "grad_norm": 0.4797917306423187, + "learning_rate": 0.00033695652173913043, + "loss": 0.3336, + "step": 23788 + }, + { + "epoch": 1.3291337262899126, + "grad_norm": 1.1069077253341675, + "learning_rate": 0.0003369285073957867, + "loss": 0.3768, + "step": 23789 + }, + { + "epoch": 1.3291895968935945, + "grad_norm": 0.40592288970947266, + "learning_rate": 0.00033690049305244284, + "loss": 0.473, + "step": 23790 + }, + { + "epoch": 1.3292454674972762, + "grad_norm": 0.37931910157203674, + "learning_rate": 0.00033687247870909905, + "loss": 0.4231, + "step": 23791 + }, + { + "epoch": 1.3293013381009582, + "grad_norm": 3.3608503341674805, + "learning_rate": 0.00033684446436575526, + "loss": 0.5923, + "step": 23792 + }, + { + "epoch": 1.3293572087046401, + "grad_norm": 0.5477584600448608, + "learning_rate": 0.00033681645002241146, + "loss": 0.5192, + "step": 23793 + }, + { + "epoch": 1.3294130793083219, + "grad_norm": 0.43031641840934753, + "learning_rate": 0.0003367884356790677, + "loss": 0.4691, + "step": 23794 + }, + { + "epoch": 1.3294689499120038, + "grad_norm": 9.401269912719727, + "learning_rate": 0.0003367604213357239, + "loss": 0.4398, + "step": 23795 + }, + { + "epoch": 1.3295248205156858, + "grad_norm": 2.034733533859253, + "learning_rate": 0.00033673240699238013, + "loss": 0.4535, + "step": 23796 + }, + { + "epoch": 1.3295806911193675, + "grad_norm": 0.46519529819488525, + "learning_rate": 0.0003367043926490363, + "loss": 0.3247, + "step": 23797 + }, + { + "epoch": 1.3296365617230494, + "grad_norm": 0.45345285534858704, + "learning_rate": 0.00033667637830569254, + "loss": 0.3795, + "step": 23798 + }, + { + "epoch": 1.3296924323267314, + "grad_norm": 0.5647457242012024, + "learning_rate": 0.0003366483639623487, + "loss": 0.509, + "step": 23799 + }, + { + "epoch": 1.3297483029304131, + "grad_norm": 0.4917866885662079, + "learning_rate": 0.00033662034961900496, + "loss": 0.4168, + "step": 23800 + }, + { + "epoch": 1.329804173534095, + "grad_norm": 0.3912925124168396, + "learning_rate": 0.0003365923352756611, + "loss": 0.5145, + "step": 23801 + }, + { + "epoch": 1.329860044137777, + "grad_norm": 0.5386469960212708, + "learning_rate": 0.00033656432093231737, + "loss": 0.4618, + "step": 23802 + }, + { + "epoch": 1.3299159147414588, + "grad_norm": 0.3490935266017914, + "learning_rate": 0.0003365363065889736, + "loss": 0.4026, + "step": 23803 + }, + { + "epoch": 1.3299717853451407, + "grad_norm": 0.4973752498626709, + "learning_rate": 0.0003365082922456298, + "loss": 0.3878, + "step": 23804 + }, + { + "epoch": 1.3300276559488227, + "grad_norm": 0.5118416547775269, + "learning_rate": 0.000336480277902286, + "loss": 0.4194, + "step": 23805 + }, + { + "epoch": 1.3300835265525044, + "grad_norm": 0.41121795773506165, + "learning_rate": 0.0003364522635589422, + "loss": 0.4496, + "step": 23806 + }, + { + "epoch": 1.3301393971561863, + "grad_norm": 0.4151020050048828, + "learning_rate": 0.0003364242492155984, + "loss": 0.3923, + "step": 23807 + }, + { + "epoch": 1.330195267759868, + "grad_norm": 0.4525260925292969, + "learning_rate": 0.0003363962348722546, + "loss": 0.4448, + "step": 23808 + }, + { + "epoch": 1.33025113836355, + "grad_norm": 0.4224332869052887, + "learning_rate": 0.0003363682205289108, + "loss": 0.4931, + "step": 23809 + }, + { + "epoch": 1.3303070089672318, + "grad_norm": 0.31806090474128723, + "learning_rate": 0.00033634020618556707, + "loss": 0.3694, + "step": 23810 + }, + { + "epoch": 1.3303628795709137, + "grad_norm": 1.0334080457687378, + "learning_rate": 0.0003363121918422232, + "loss": 0.415, + "step": 23811 + }, + { + "epoch": 1.3304187501745957, + "grad_norm": 0.5941030383110046, + "learning_rate": 0.0003362841774988794, + "loss": 0.4241, + "step": 23812 + }, + { + "epoch": 1.3304746207782774, + "grad_norm": 0.5172542333602905, + "learning_rate": 0.00033625616315553563, + "loss": 0.4004, + "step": 23813 + }, + { + "epoch": 1.3305304913819593, + "grad_norm": 0.4096139073371887, + "learning_rate": 0.00033622814881219184, + "loss": 0.4269, + "step": 23814 + }, + { + "epoch": 1.3305863619856413, + "grad_norm": 0.4078289568424225, + "learning_rate": 0.00033620013446884804, + "loss": 0.3781, + "step": 23815 + }, + { + "epoch": 1.330642232589323, + "grad_norm": 0.3300056755542755, + "learning_rate": 0.00033617212012550425, + "loss": 0.3395, + "step": 23816 + }, + { + "epoch": 1.330698103193005, + "grad_norm": 0.6040382981300354, + "learning_rate": 0.00033614410578216046, + "loss": 0.3865, + "step": 23817 + }, + { + "epoch": 1.330753973796687, + "grad_norm": 0.6771998405456543, + "learning_rate": 0.00033611609143881666, + "loss": 0.328, + "step": 23818 + }, + { + "epoch": 1.3308098444003686, + "grad_norm": 0.39428776502609253, + "learning_rate": 0.0003360880770954729, + "loss": 0.4037, + "step": 23819 + }, + { + "epoch": 1.3308657150040506, + "grad_norm": 0.8698804974555969, + "learning_rate": 0.00033606006275212907, + "loss": 0.4385, + "step": 23820 + }, + { + "epoch": 1.3309215856077325, + "grad_norm": 1.108974814414978, + "learning_rate": 0.00033603204840878533, + "loss": 0.3924, + "step": 23821 + }, + { + "epoch": 1.3309774562114143, + "grad_norm": 0.360720694065094, + "learning_rate": 0.0003360040340654415, + "loss": 0.5392, + "step": 23822 + }, + { + "epoch": 1.3310333268150962, + "grad_norm": 0.3173142075538635, + "learning_rate": 0.00033597601972209774, + "loss": 0.4147, + "step": 23823 + }, + { + "epoch": 1.3310891974187782, + "grad_norm": 0.5126133561134338, + "learning_rate": 0.0003359480053787539, + "loss": 0.6237, + "step": 23824 + }, + { + "epoch": 1.33114506802246, + "grad_norm": 0.7322197556495667, + "learning_rate": 0.00033591999103541016, + "loss": 0.3371, + "step": 23825 + }, + { + "epoch": 1.3312009386261419, + "grad_norm": 0.3174044191837311, + "learning_rate": 0.00033589197669206636, + "loss": 0.3893, + "step": 23826 + }, + { + "epoch": 1.3312568092298238, + "grad_norm": 0.5738434791564941, + "learning_rate": 0.00033586396234872257, + "loss": 0.3919, + "step": 23827 + }, + { + "epoch": 1.3313126798335055, + "grad_norm": 1.1533018350601196, + "learning_rate": 0.0003358359480053788, + "loss": 0.5395, + "step": 23828 + }, + { + "epoch": 1.3313685504371875, + "grad_norm": 0.4033086895942688, + "learning_rate": 0.000335807933662035, + "loss": 0.4349, + "step": 23829 + }, + { + "epoch": 1.3314244210408694, + "grad_norm": 0.6028754115104675, + "learning_rate": 0.0003357799193186912, + "loss": 0.3159, + "step": 23830 + }, + { + "epoch": 1.3314802916445512, + "grad_norm": 0.6668214797973633, + "learning_rate": 0.0003357519049753474, + "loss": 0.4106, + "step": 23831 + }, + { + "epoch": 1.3315361622482331, + "grad_norm": 0.48326975107192993, + "learning_rate": 0.0003357238906320036, + "loss": 0.5403, + "step": 23832 + }, + { + "epoch": 1.331592032851915, + "grad_norm": 0.49557074904441833, + "learning_rate": 0.00033569587628865975, + "loss": 0.382, + "step": 23833 + }, + { + "epoch": 1.3316479034555968, + "grad_norm": 0.39908087253570557, + "learning_rate": 0.000335667861945316, + "loss": 0.2898, + "step": 23834 + }, + { + "epoch": 1.3317037740592788, + "grad_norm": 0.46634435653686523, + "learning_rate": 0.0003356398476019722, + "loss": 0.4757, + "step": 23835 + }, + { + "epoch": 1.3317596446629607, + "grad_norm": 0.39268437027931213, + "learning_rate": 0.0003356118332586284, + "loss": 0.4597, + "step": 23836 + }, + { + "epoch": 1.3318155152666424, + "grad_norm": 0.43147531151771545, + "learning_rate": 0.0003355838189152846, + "loss": 0.4905, + "step": 23837 + }, + { + "epoch": 1.3318713858703244, + "grad_norm": 0.6009288430213928, + "learning_rate": 0.00033555580457194083, + "loss": 0.4117, + "step": 23838 + }, + { + "epoch": 1.3319272564740063, + "grad_norm": 0.8124978542327881, + "learning_rate": 0.00033552779022859704, + "loss": 0.4016, + "step": 23839 + }, + { + "epoch": 1.331983127077688, + "grad_norm": 2.528296947479248, + "learning_rate": 0.00033549977588525324, + "loss": 0.3606, + "step": 23840 + }, + { + "epoch": 1.33203899768137, + "grad_norm": 2.198291778564453, + "learning_rate": 0.00033547176154190945, + "loss": 0.5061, + "step": 23841 + }, + { + "epoch": 1.3320948682850517, + "grad_norm": 0.30520644783973694, + "learning_rate": 0.0003354437471985657, + "loss": 0.3643, + "step": 23842 + }, + { + "epoch": 1.3321507388887337, + "grad_norm": 0.46163567900657654, + "learning_rate": 0.00033541573285522186, + "loss": 0.5002, + "step": 23843 + }, + { + "epoch": 1.3322066094924154, + "grad_norm": 0.3675830364227295, + "learning_rate": 0.0003353877185118781, + "loss": 0.416, + "step": 23844 + }, + { + "epoch": 1.3322624800960974, + "grad_norm": 1.7234365940093994, + "learning_rate": 0.00033535970416853427, + "loss": 0.5552, + "step": 23845 + }, + { + "epoch": 1.3323183506997793, + "grad_norm": 0.49691224098205566, + "learning_rate": 0.00033533168982519053, + "loss": 0.4566, + "step": 23846 + }, + { + "epoch": 1.332374221303461, + "grad_norm": 0.378010630607605, + "learning_rate": 0.0003353036754818467, + "loss": 0.4337, + "step": 23847 + }, + { + "epoch": 1.332430091907143, + "grad_norm": 0.4100237488746643, + "learning_rate": 0.00033527566113850294, + "loss": 0.5368, + "step": 23848 + }, + { + "epoch": 1.332485962510825, + "grad_norm": 0.45994672179222107, + "learning_rate": 0.0003352476467951591, + "loss": 0.5319, + "step": 23849 + }, + { + "epoch": 1.3325418331145067, + "grad_norm": 0.38129913806915283, + "learning_rate": 0.00033521963245181536, + "loss": 0.4588, + "step": 23850 + }, + { + "epoch": 1.3325977037181886, + "grad_norm": 0.40191739797592163, + "learning_rate": 0.00033519161810847156, + "loss": 0.3713, + "step": 23851 + }, + { + "epoch": 1.3326535743218706, + "grad_norm": 1.001427173614502, + "learning_rate": 0.00033516360376512777, + "loss": 0.4054, + "step": 23852 + }, + { + "epoch": 1.3327094449255523, + "grad_norm": 2.6200859546661377, + "learning_rate": 0.000335135589421784, + "loss": 0.3384, + "step": 23853 + }, + { + "epoch": 1.3327653155292343, + "grad_norm": 0.42654284834861755, + "learning_rate": 0.0003351075750784401, + "loss": 0.4962, + "step": 23854 + }, + { + "epoch": 1.3328211861329162, + "grad_norm": 0.7023627758026123, + "learning_rate": 0.0003350795607350964, + "loss": 0.7292, + "step": 23855 + }, + { + "epoch": 1.332877056736598, + "grad_norm": 0.4314132332801819, + "learning_rate": 0.00033505154639175254, + "loss": 0.4081, + "step": 23856 + }, + { + "epoch": 1.33293292734028, + "grad_norm": 0.3866397738456726, + "learning_rate": 0.0003350235320484088, + "loss": 0.472, + "step": 23857 + }, + { + "epoch": 1.3329887979439619, + "grad_norm": 0.5305342078208923, + "learning_rate": 0.000334995517705065, + "loss": 0.4226, + "step": 23858 + }, + { + "epoch": 1.3330446685476436, + "grad_norm": 0.3929923474788666, + "learning_rate": 0.0003349675033617212, + "loss": 0.5535, + "step": 23859 + }, + { + "epoch": 1.3331005391513255, + "grad_norm": 0.3817827105522156, + "learning_rate": 0.0003349394890183774, + "loss": 0.5319, + "step": 23860 + }, + { + "epoch": 1.3331564097550075, + "grad_norm": 2.7164275646209717, + "learning_rate": 0.0003349114746750336, + "loss": 0.3996, + "step": 23861 + }, + { + "epoch": 1.3332122803586892, + "grad_norm": 1.2627612352371216, + "learning_rate": 0.0003348834603316898, + "loss": 0.4675, + "step": 23862 + }, + { + "epoch": 1.3332681509623712, + "grad_norm": 0.7446377277374268, + "learning_rate": 0.00033485544598834603, + "loss": 0.3731, + "step": 23863 + }, + { + "epoch": 1.3333240215660531, + "grad_norm": 14.470941543579102, + "learning_rate": 0.00033482743164500224, + "loss": 0.611, + "step": 23864 + }, + { + "epoch": 1.3333798921697348, + "grad_norm": 0.3853253126144409, + "learning_rate": 0.0003347994173016585, + "loss": 0.4461, + "step": 23865 + }, + { + "epoch": 1.3334357627734168, + "grad_norm": 0.8284673094749451, + "learning_rate": 0.00033477140295831465, + "loss": 0.4125, + "step": 23866 + }, + { + "epoch": 1.3334916333770988, + "grad_norm": 0.4502894878387451, + "learning_rate": 0.0003347433886149709, + "loss": 0.4709, + "step": 23867 + }, + { + "epoch": 1.3335475039807805, + "grad_norm": 0.49213865399360657, + "learning_rate": 0.00033471537427162706, + "loss": 0.3345, + "step": 23868 + }, + { + "epoch": 1.3336033745844624, + "grad_norm": 0.549310564994812, + "learning_rate": 0.0003346873599282833, + "loss": 0.3761, + "step": 23869 + }, + { + "epoch": 1.3336592451881444, + "grad_norm": 0.4719569683074951, + "learning_rate": 0.00033465934558493947, + "loss": 0.4622, + "step": 23870 + }, + { + "epoch": 1.3337151157918261, + "grad_norm": 0.4062821567058563, + "learning_rate": 0.00033463133124159573, + "loss": 0.3596, + "step": 23871 + }, + { + "epoch": 1.333770986395508, + "grad_norm": 0.3721064031124115, + "learning_rate": 0.0003346033168982519, + "loss": 0.3322, + "step": 23872 + }, + { + "epoch": 1.33382685699919, + "grad_norm": 0.3098447620868683, + "learning_rate": 0.00033457530255490814, + "loss": 0.3168, + "step": 23873 + }, + { + "epoch": 1.3338827276028717, + "grad_norm": 0.4597738981246948, + "learning_rate": 0.00033454728821156435, + "loss": 0.4291, + "step": 23874 + }, + { + "epoch": 1.3339385982065537, + "grad_norm": 0.84902423620224, + "learning_rate": 0.0003345192738682205, + "loss": 0.3509, + "step": 23875 + }, + { + "epoch": 1.3339944688102354, + "grad_norm": 0.3375215530395508, + "learning_rate": 0.00033449125952487676, + "loss": 0.3433, + "step": 23876 + }, + { + "epoch": 1.3340503394139174, + "grad_norm": 0.5321007370948792, + "learning_rate": 0.0003344632451815329, + "loss": 0.339, + "step": 23877 + }, + { + "epoch": 1.334106210017599, + "grad_norm": 0.48192644119262695, + "learning_rate": 0.00033443523083818917, + "loss": 0.4763, + "step": 23878 + }, + { + "epoch": 1.334162080621281, + "grad_norm": 0.4229472875595093, + "learning_rate": 0.0003344072164948453, + "loss": 0.514, + "step": 23879 + }, + { + "epoch": 1.334217951224963, + "grad_norm": 0.429861456155777, + "learning_rate": 0.0003343792021515016, + "loss": 0.4235, + "step": 23880 + }, + { + "epoch": 1.3342738218286447, + "grad_norm": 0.3880572021007538, + "learning_rate": 0.0003343511878081578, + "loss": 0.3343, + "step": 23881 + }, + { + "epoch": 1.3343296924323267, + "grad_norm": 0.4389580190181732, + "learning_rate": 0.000334323173464814, + "loss": 0.4521, + "step": 23882 + }, + { + "epoch": 1.3343855630360086, + "grad_norm": 0.31863337755203247, + "learning_rate": 0.0003342951591214702, + "loss": 0.3313, + "step": 23883 + }, + { + "epoch": 1.3344414336396904, + "grad_norm": 0.689037024974823, + "learning_rate": 0.0003342671447781264, + "loss": 0.4503, + "step": 23884 + }, + { + "epoch": 1.3344973042433723, + "grad_norm": 0.41046079993247986, + "learning_rate": 0.0003342391304347826, + "loss": 0.4051, + "step": 23885 + }, + { + "epoch": 1.3345531748470543, + "grad_norm": 0.5940335988998413, + "learning_rate": 0.0003342111160914388, + "loss": 0.4778, + "step": 23886 + }, + { + "epoch": 1.334609045450736, + "grad_norm": 0.6991293430328369, + "learning_rate": 0.000334183101748095, + "loss": 0.4768, + "step": 23887 + }, + { + "epoch": 1.334664916054418, + "grad_norm": 0.4355214536190033, + "learning_rate": 0.00033415508740475123, + "loss": 0.4513, + "step": 23888 + }, + { + "epoch": 1.3347207866581, + "grad_norm": 0.42658400535583496, + "learning_rate": 0.00033412707306140744, + "loss": 0.4968, + "step": 23889 + }, + { + "epoch": 1.3347766572617816, + "grad_norm": 3.642376184463501, + "learning_rate": 0.0003340990587180637, + "loss": 0.3607, + "step": 23890 + }, + { + "epoch": 1.3348325278654636, + "grad_norm": 0.4272277355194092, + "learning_rate": 0.00033407104437471985, + "loss": 0.4114, + "step": 23891 + }, + { + "epoch": 1.3348883984691455, + "grad_norm": 0.39120742678642273, + "learning_rate": 0.0003340430300313761, + "loss": 0.3545, + "step": 23892 + }, + { + "epoch": 1.3349442690728273, + "grad_norm": 0.5397878885269165, + "learning_rate": 0.00033401501568803226, + "loss": 0.4217, + "step": 23893 + }, + { + "epoch": 1.3350001396765092, + "grad_norm": 0.3988956809043884, + "learning_rate": 0.0003339870013446885, + "loss": 0.4665, + "step": 23894 + }, + { + "epoch": 1.3350560102801912, + "grad_norm": 0.6379366517066956, + "learning_rate": 0.00033395898700134467, + "loss": 0.4304, + "step": 23895 + }, + { + "epoch": 1.335111880883873, + "grad_norm": 0.4335348606109619, + "learning_rate": 0.00033393097265800093, + "loss": 0.423, + "step": 23896 + }, + { + "epoch": 1.3351677514875548, + "grad_norm": 0.40940314531326294, + "learning_rate": 0.00033390295831465714, + "loss": 0.4268, + "step": 23897 + }, + { + "epoch": 1.3352236220912368, + "grad_norm": 0.5739796757698059, + "learning_rate": 0.0003338749439713133, + "loss": 0.5847, + "step": 23898 + }, + { + "epoch": 1.3352794926949185, + "grad_norm": 0.7935835123062134, + "learning_rate": 0.00033384692962796955, + "loss": 0.4539, + "step": 23899 + }, + { + "epoch": 1.3353353632986005, + "grad_norm": 3.825652599334717, + "learning_rate": 0.0003338189152846257, + "loss": 0.4287, + "step": 23900 + }, + { + "epoch": 1.3353912339022824, + "grad_norm": 0.4522132873535156, + "learning_rate": 0.00033379090094128196, + "loss": 0.4659, + "step": 23901 + }, + { + "epoch": 1.3354471045059642, + "grad_norm": 0.5011293292045593, + "learning_rate": 0.0003337628865979381, + "loss": 0.4335, + "step": 23902 + }, + { + "epoch": 1.335502975109646, + "grad_norm": 0.5191582441329956, + "learning_rate": 0.00033373487225459437, + "loss": 0.5184, + "step": 23903 + }, + { + "epoch": 1.335558845713328, + "grad_norm": 0.5818153619766235, + "learning_rate": 0.0003337068579112505, + "loss": 0.3602, + "step": 23904 + }, + { + "epoch": 1.3356147163170098, + "grad_norm": 1.8228679895401, + "learning_rate": 0.0003336788435679068, + "loss": 0.3275, + "step": 23905 + }, + { + "epoch": 1.3356705869206917, + "grad_norm": 0.41660550236701965, + "learning_rate": 0.000333650829224563, + "loss": 0.3796, + "step": 23906 + }, + { + "epoch": 1.3357264575243737, + "grad_norm": 1.0451536178588867, + "learning_rate": 0.0003336228148812192, + "loss": 0.4188, + "step": 23907 + }, + { + "epoch": 1.3357823281280554, + "grad_norm": 0.3572564423084259, + "learning_rate": 0.0003335948005378754, + "loss": 0.4703, + "step": 23908 + }, + { + "epoch": 1.3358381987317374, + "grad_norm": 0.4063127338886261, + "learning_rate": 0.0003335667861945316, + "loss": 0.3689, + "step": 23909 + }, + { + "epoch": 1.335894069335419, + "grad_norm": 0.3651787042617798, + "learning_rate": 0.0003335387718511878, + "loss": 0.4683, + "step": 23910 + }, + { + "epoch": 1.335949939939101, + "grad_norm": 0.35255298018455505, + "learning_rate": 0.000333510757507844, + "loss": 0.4102, + "step": 23911 + }, + { + "epoch": 1.3360058105427828, + "grad_norm": 0.41777917742729187, + "learning_rate": 0.0003334827431645002, + "loss": 0.4993, + "step": 23912 + }, + { + "epoch": 1.3360616811464647, + "grad_norm": 0.349961519241333, + "learning_rate": 0.0003334547288211565, + "loss": 0.3746, + "step": 23913 + }, + { + "epoch": 1.3361175517501467, + "grad_norm": 0.5918413400650024, + "learning_rate": 0.00033342671447781264, + "loss": 0.4242, + "step": 23914 + }, + { + "epoch": 1.3361734223538284, + "grad_norm": 0.6150894165039062, + "learning_rate": 0.0003333987001344689, + "loss": 0.3749, + "step": 23915 + }, + { + "epoch": 1.3362292929575104, + "grad_norm": 0.6249264478683472, + "learning_rate": 0.00033337068579112505, + "loss": 0.4314, + "step": 23916 + }, + { + "epoch": 1.3362851635611923, + "grad_norm": 0.47628751397132874, + "learning_rate": 0.0003333426714477813, + "loss": 0.2982, + "step": 23917 + }, + { + "epoch": 1.336341034164874, + "grad_norm": 0.35365965962409973, + "learning_rate": 0.00033331465710443746, + "loss": 0.4353, + "step": 23918 + }, + { + "epoch": 1.336396904768556, + "grad_norm": 0.7035937905311584, + "learning_rate": 0.00033328664276109367, + "loss": 0.6664, + "step": 23919 + }, + { + "epoch": 1.336452775372238, + "grad_norm": 0.4470439553260803, + "learning_rate": 0.0003332586284177499, + "loss": 0.4844, + "step": 23920 + }, + { + "epoch": 1.3365086459759197, + "grad_norm": 0.9344048500061035, + "learning_rate": 0.0003332306140744061, + "loss": 0.4432, + "step": 23921 + }, + { + "epoch": 1.3365645165796016, + "grad_norm": 2.8535714149475098, + "learning_rate": 0.00033320259973106234, + "loss": 0.343, + "step": 23922 + }, + { + "epoch": 1.3366203871832836, + "grad_norm": 0.3911079168319702, + "learning_rate": 0.0003331745853877185, + "loss": 0.5379, + "step": 23923 + }, + { + "epoch": 1.3366762577869653, + "grad_norm": 0.438228964805603, + "learning_rate": 0.00033314657104437475, + "loss": 0.3958, + "step": 23924 + }, + { + "epoch": 1.3367321283906473, + "grad_norm": 3.0790085792541504, + "learning_rate": 0.0003331185567010309, + "loss": 0.4011, + "step": 23925 + }, + { + "epoch": 1.3367879989943292, + "grad_norm": 0.8724040985107422, + "learning_rate": 0.00033309054235768716, + "loss": 0.4077, + "step": 23926 + }, + { + "epoch": 1.336843869598011, + "grad_norm": 0.31624653935432434, + "learning_rate": 0.0003330625280143433, + "loss": 0.3734, + "step": 23927 + }, + { + "epoch": 1.3368997402016929, + "grad_norm": 1.7090502977371216, + "learning_rate": 0.00033303451367099957, + "loss": 0.3525, + "step": 23928 + }, + { + "epoch": 1.3369556108053748, + "grad_norm": 0.44002005457878113, + "learning_rate": 0.0003330064993276558, + "loss": 0.4715, + "step": 23929 + }, + { + "epoch": 1.3370114814090566, + "grad_norm": 0.5365398526191711, + "learning_rate": 0.000332978484984312, + "loss": 0.4438, + "step": 23930 + }, + { + "epoch": 1.3370673520127385, + "grad_norm": 1.119419813156128, + "learning_rate": 0.0003329504706409682, + "loss": 0.4866, + "step": 23931 + }, + { + "epoch": 1.3371232226164205, + "grad_norm": 0.8813714385032654, + "learning_rate": 0.0003329224562976244, + "loss": 0.5015, + "step": 23932 + }, + { + "epoch": 1.3371790932201022, + "grad_norm": 0.8527490496635437, + "learning_rate": 0.0003328944419542806, + "loss": 0.4183, + "step": 23933 + }, + { + "epoch": 1.3372349638237842, + "grad_norm": 0.49694377183914185, + "learning_rate": 0.0003328664276109368, + "loss": 0.3761, + "step": 23934 + }, + { + "epoch": 1.337290834427466, + "grad_norm": 0.3948342800140381, + "learning_rate": 0.000332838413267593, + "loss": 0.4078, + "step": 23935 + }, + { + "epoch": 1.3373467050311478, + "grad_norm": 0.38980886340141296, + "learning_rate": 0.00033281039892424927, + "loss": 0.4995, + "step": 23936 + }, + { + "epoch": 1.3374025756348298, + "grad_norm": 0.3297223448753357, + "learning_rate": 0.0003327823845809054, + "loss": 0.4793, + "step": 23937 + }, + { + "epoch": 1.3374584462385117, + "grad_norm": 0.6306056380271912, + "learning_rate": 0.0003327543702375617, + "loss": 0.3877, + "step": 23938 + }, + { + "epoch": 1.3375143168421935, + "grad_norm": 0.4069092273712158, + "learning_rate": 0.00033272635589421784, + "loss": 0.4669, + "step": 23939 + }, + { + "epoch": 1.3375701874458754, + "grad_norm": 0.3328961431980133, + "learning_rate": 0.00033269834155087404, + "loss": 0.3923, + "step": 23940 + }, + { + "epoch": 1.3376260580495571, + "grad_norm": 0.5310647487640381, + "learning_rate": 0.00033267032720753025, + "loss": 0.3411, + "step": 23941 + }, + { + "epoch": 1.337681928653239, + "grad_norm": 0.5529271960258484, + "learning_rate": 0.00033264231286418645, + "loss": 0.4093, + "step": 23942 + }, + { + "epoch": 1.337737799256921, + "grad_norm": 0.5117146372795105, + "learning_rate": 0.00033261429852084266, + "loss": 0.5133, + "step": 23943 + }, + { + "epoch": 1.3377936698606028, + "grad_norm": 0.4439264237880707, + "learning_rate": 0.00033258628417749886, + "loss": 0.3999, + "step": 23944 + }, + { + "epoch": 1.3378495404642847, + "grad_norm": 0.6054041385650635, + "learning_rate": 0.0003325582698341551, + "loss": 0.5211, + "step": 23945 + }, + { + "epoch": 1.3379054110679665, + "grad_norm": 1.066917061805725, + "learning_rate": 0.0003325302554908113, + "loss": 0.3986, + "step": 23946 + }, + { + "epoch": 1.3379612816716484, + "grad_norm": 0.6248095631599426, + "learning_rate": 0.00033250224114746754, + "loss": 0.5348, + "step": 23947 + }, + { + "epoch": 1.3380171522753304, + "grad_norm": 0.45118245482444763, + "learning_rate": 0.0003324742268041237, + "loss": 0.3874, + "step": 23948 + }, + { + "epoch": 1.338073022879012, + "grad_norm": 0.324445903301239, + "learning_rate": 0.00033244621246077995, + "loss": 0.3469, + "step": 23949 + }, + { + "epoch": 1.338128893482694, + "grad_norm": 0.39078962802886963, + "learning_rate": 0.0003324181981174361, + "loss": 0.398, + "step": 23950 + }, + { + "epoch": 1.338184764086376, + "grad_norm": 0.4386056661605835, + "learning_rate": 0.00033239018377409236, + "loss": 0.5872, + "step": 23951 + }, + { + "epoch": 1.3382406346900577, + "grad_norm": 0.5958513021469116, + "learning_rate": 0.00033236216943074857, + "loss": 0.6835, + "step": 23952 + }, + { + "epoch": 1.3382965052937397, + "grad_norm": 0.41353484988212585, + "learning_rate": 0.00033233415508740477, + "loss": 0.4046, + "step": 23953 + }, + { + "epoch": 1.3383523758974216, + "grad_norm": 0.4241480827331543, + "learning_rate": 0.000332306140744061, + "loss": 0.306, + "step": 23954 + }, + { + "epoch": 1.3384082465011033, + "grad_norm": 0.40953585505485535, + "learning_rate": 0.0003322781264007172, + "loss": 0.4815, + "step": 23955 + }, + { + "epoch": 1.3384641171047853, + "grad_norm": 0.3774765133857727, + "learning_rate": 0.0003322501120573734, + "loss": 0.3892, + "step": 23956 + }, + { + "epoch": 1.3385199877084673, + "grad_norm": 0.4577121138572693, + "learning_rate": 0.0003322220977140296, + "loss": 0.5671, + "step": 23957 + }, + { + "epoch": 1.338575858312149, + "grad_norm": 0.42905816435813904, + "learning_rate": 0.0003321940833706858, + "loss": 0.453, + "step": 23958 + }, + { + "epoch": 1.338631728915831, + "grad_norm": 0.552696943283081, + "learning_rate": 0.000332166069027342, + "loss": 0.4876, + "step": 23959 + }, + { + "epoch": 1.3386875995195129, + "grad_norm": 0.5651734471321106, + "learning_rate": 0.0003321380546839982, + "loss": 0.4105, + "step": 23960 + }, + { + "epoch": 1.3387434701231946, + "grad_norm": 1.796733021736145, + "learning_rate": 0.0003321100403406544, + "loss": 0.4872, + "step": 23961 + }, + { + "epoch": 1.3387993407268766, + "grad_norm": 0.48724210262298584, + "learning_rate": 0.0003320820259973106, + "loss": 0.4054, + "step": 23962 + }, + { + "epoch": 1.3388552113305585, + "grad_norm": 3.8439786434173584, + "learning_rate": 0.00033205401165396683, + "loss": 0.4654, + "step": 23963 + }, + { + "epoch": 1.3389110819342402, + "grad_norm": 0.638763964176178, + "learning_rate": 0.00033202599731062304, + "loss": 0.4564, + "step": 23964 + }, + { + "epoch": 1.3389669525379222, + "grad_norm": 0.3369641602039337, + "learning_rate": 0.00033199798296727924, + "loss": 0.3371, + "step": 23965 + }, + { + "epoch": 1.3390228231416041, + "grad_norm": 4.832735061645508, + "learning_rate": 0.00033196996862393545, + "loss": 0.4313, + "step": 23966 + }, + { + "epoch": 1.3390786937452859, + "grad_norm": 0.4728822112083435, + "learning_rate": 0.00033194195428059165, + "loss": 0.5014, + "step": 23967 + }, + { + "epoch": 1.3391345643489678, + "grad_norm": 0.2975355088710785, + "learning_rate": 0.0003319139399372479, + "loss": 0.3813, + "step": 23968 + }, + { + "epoch": 1.3391904349526498, + "grad_norm": 0.6316016316413879, + "learning_rate": 0.00033188592559390406, + "loss": 0.3682, + "step": 23969 + }, + { + "epoch": 1.3392463055563315, + "grad_norm": 0.3969492018222809, + "learning_rate": 0.0003318579112505603, + "loss": 0.4038, + "step": 23970 + }, + { + "epoch": 1.3393021761600135, + "grad_norm": 0.5442119240760803, + "learning_rate": 0.0003318298969072165, + "loss": 0.6015, + "step": 23971 + }, + { + "epoch": 1.3393580467636954, + "grad_norm": 0.39547568559646606, + "learning_rate": 0.00033180188256387274, + "loss": 0.4252, + "step": 23972 + }, + { + "epoch": 1.3394139173673771, + "grad_norm": 0.46776655316352844, + "learning_rate": 0.0003317738682205289, + "loss": 0.4925, + "step": 23973 + }, + { + "epoch": 1.339469787971059, + "grad_norm": 0.577700138092041, + "learning_rate": 0.00033174585387718515, + "loss": 0.5316, + "step": 23974 + }, + { + "epoch": 1.3395256585747408, + "grad_norm": 0.346163272857666, + "learning_rate": 0.0003317178395338413, + "loss": 0.3728, + "step": 23975 + }, + { + "epoch": 1.3395815291784228, + "grad_norm": 0.4739697277545929, + "learning_rate": 0.00033168982519049756, + "loss": 0.3942, + "step": 23976 + }, + { + "epoch": 1.3396373997821047, + "grad_norm": 3.1744139194488525, + "learning_rate": 0.00033166181084715377, + "loss": 0.4108, + "step": 23977 + }, + { + "epoch": 1.3396932703857864, + "grad_norm": 1.763182282447815, + "learning_rate": 0.00033163379650380997, + "loss": 0.4722, + "step": 23978 + }, + { + "epoch": 1.3397491409894684, + "grad_norm": 0.3798116445541382, + "learning_rate": 0.0003316057821604662, + "loss": 0.4371, + "step": 23979 + }, + { + "epoch": 1.3398050115931501, + "grad_norm": 0.7259247899055481, + "learning_rate": 0.0003315777678171224, + "loss": 0.3715, + "step": 23980 + }, + { + "epoch": 1.339860882196832, + "grad_norm": 4.283868789672852, + "learning_rate": 0.0003315497534737786, + "loss": 0.4251, + "step": 23981 + }, + { + "epoch": 1.339916752800514, + "grad_norm": 0.3788101077079773, + "learning_rate": 0.00033152173913043474, + "loss": 0.3408, + "step": 23982 + }, + { + "epoch": 1.3399726234041958, + "grad_norm": 0.35214415192604065, + "learning_rate": 0.000331493724787091, + "loss": 0.4531, + "step": 23983 + }, + { + "epoch": 1.3400284940078777, + "grad_norm": 0.4298122823238373, + "learning_rate": 0.0003314657104437472, + "loss": 0.4437, + "step": 23984 + }, + { + "epoch": 1.3400843646115597, + "grad_norm": 0.4357004761695862, + "learning_rate": 0.0003314376961004034, + "loss": 0.4824, + "step": 23985 + }, + { + "epoch": 1.3401402352152414, + "grad_norm": 0.4048737585544586, + "learning_rate": 0.0003314096817570596, + "loss": 0.4846, + "step": 23986 + }, + { + "epoch": 1.3401961058189233, + "grad_norm": 0.42046496272087097, + "learning_rate": 0.0003313816674137158, + "loss": 0.5329, + "step": 23987 + }, + { + "epoch": 1.3402519764226053, + "grad_norm": 0.5722823143005371, + "learning_rate": 0.00033135365307037203, + "loss": 0.4171, + "step": 23988 + }, + { + "epoch": 1.340307847026287, + "grad_norm": 0.5299311876296997, + "learning_rate": 0.00033132563872702823, + "loss": 0.4209, + "step": 23989 + }, + { + "epoch": 1.340363717629969, + "grad_norm": 0.4778738021850586, + "learning_rate": 0.00033129762438368444, + "loss": 0.4721, + "step": 23990 + }, + { + "epoch": 1.340419588233651, + "grad_norm": 0.41203173995018005, + "learning_rate": 0.0003312696100403407, + "loss": 0.345, + "step": 23991 + }, + { + "epoch": 1.3404754588373327, + "grad_norm": 0.6948039531707764, + "learning_rate": 0.00033124159569699685, + "loss": 0.4792, + "step": 23992 + }, + { + "epoch": 1.3405313294410146, + "grad_norm": 0.6031925082206726, + "learning_rate": 0.0003312135813536531, + "loss": 0.4162, + "step": 23993 + }, + { + "epoch": 1.3405872000446966, + "grad_norm": 0.3519224524497986, + "learning_rate": 0.00033118556701030926, + "loss": 0.453, + "step": 23994 + }, + { + "epoch": 1.3406430706483783, + "grad_norm": 0.31841719150543213, + "learning_rate": 0.0003311575526669655, + "loss": 0.3437, + "step": 23995 + }, + { + "epoch": 1.3406989412520602, + "grad_norm": 0.42236074805259705, + "learning_rate": 0.0003311295383236217, + "loss": 0.4593, + "step": 23996 + }, + { + "epoch": 1.3407548118557422, + "grad_norm": 1.033592700958252, + "learning_rate": 0.00033110152398027794, + "loss": 0.3901, + "step": 23997 + }, + { + "epoch": 1.340810682459424, + "grad_norm": 0.5216073393821716, + "learning_rate": 0.0003310735096369341, + "loss": 0.4243, + "step": 23998 + }, + { + "epoch": 1.3408665530631059, + "grad_norm": 0.5262933969497681, + "learning_rate": 0.00033104549529359035, + "loss": 0.471, + "step": 23999 + }, + { + "epoch": 1.3409224236667878, + "grad_norm": 0.5546186566352844, + "learning_rate": 0.00033101748095024655, + "loss": 0.4414, + "step": 24000 + }, + { + "epoch": 1.3409224236667878, + "eval_cer": 0.08583461534265109, + "eval_loss": 0.3218265771865845, + "eval_runtime": 55.9602, + "eval_samples_per_second": 81.093, + "eval_steps_per_second": 5.075, + "eval_wer": 0.34309097511242703, + "step": 24000 + }, + { + "epoch": 1.3409782942704696, + "grad_norm": 0.6804379224777222, + "learning_rate": 0.00033098946660690276, + "loss": 0.4821, + "step": 24001 + }, + { + "epoch": 1.3410341648741515, + "grad_norm": 0.6108449101448059, + "learning_rate": 0.00033096145226355896, + "loss": 0.433, + "step": 24002 + }, + { + "epoch": 1.3410900354778335, + "grad_norm": 0.4614548981189728, + "learning_rate": 0.0003309334379202151, + "loss": 0.4326, + "step": 24003 + }, + { + "epoch": 1.3411459060815152, + "grad_norm": 0.5029597878456116, + "learning_rate": 0.0003309054235768714, + "loss": 0.4752, + "step": 24004 + }, + { + "epoch": 1.3412017766851971, + "grad_norm": 0.7062781453132629, + "learning_rate": 0.00033087740923352753, + "loss": 0.5194, + "step": 24005 + }, + { + "epoch": 1.341257647288879, + "grad_norm": 0.5434077382087708, + "learning_rate": 0.0003308493948901838, + "loss": 0.4281, + "step": 24006 + }, + { + "epoch": 1.3413135178925608, + "grad_norm": 0.45809781551361084, + "learning_rate": 0.00033082138054684, + "loss": 0.3557, + "step": 24007 + }, + { + "epoch": 1.3413693884962428, + "grad_norm": 0.4267924427986145, + "learning_rate": 0.0003307933662034962, + "loss": 0.5956, + "step": 24008 + }, + { + "epoch": 1.3414252590999245, + "grad_norm": 0.3666101396083832, + "learning_rate": 0.0003307653518601524, + "loss": 0.3738, + "step": 24009 + }, + { + "epoch": 1.3414811297036064, + "grad_norm": 1.382252812385559, + "learning_rate": 0.0003307373375168086, + "loss": 0.606, + "step": 24010 + }, + { + "epoch": 1.3415370003072884, + "grad_norm": 0.6301844120025635, + "learning_rate": 0.0003307093231734648, + "loss": 0.4741, + "step": 24011 + }, + { + "epoch": 1.3415928709109701, + "grad_norm": 0.4500781297683716, + "learning_rate": 0.000330681308830121, + "loss": 0.5086, + "step": 24012 + }, + { + "epoch": 1.341648741514652, + "grad_norm": 0.6220136880874634, + "learning_rate": 0.00033065329448677723, + "loss": 0.5367, + "step": 24013 + }, + { + "epoch": 1.3417046121183338, + "grad_norm": 0.4116290509700775, + "learning_rate": 0.00033062528014343343, + "loss": 0.4311, + "step": 24014 + }, + { + "epoch": 1.3417604827220158, + "grad_norm": 0.5392543077468872, + "learning_rate": 0.00033059726580008964, + "loss": 0.4192, + "step": 24015 + }, + { + "epoch": 1.3418163533256977, + "grad_norm": 0.3402247130870819, + "learning_rate": 0.0003305692514567459, + "loss": 0.406, + "step": 24016 + }, + { + "epoch": 1.3418722239293794, + "grad_norm": 0.5171025395393372, + "learning_rate": 0.00033054123711340205, + "loss": 0.4301, + "step": 24017 + }, + { + "epoch": 1.3419280945330614, + "grad_norm": 0.9486579298973083, + "learning_rate": 0.0003305132227700583, + "loss": 0.4847, + "step": 24018 + }, + { + "epoch": 1.3419839651367433, + "grad_norm": 0.4403744041919708, + "learning_rate": 0.00033048520842671446, + "loss": 0.5714, + "step": 24019 + }, + { + "epoch": 1.342039835740425, + "grad_norm": 0.47298166155815125, + "learning_rate": 0.0003304571940833707, + "loss": 0.3625, + "step": 24020 + }, + { + "epoch": 1.342095706344107, + "grad_norm": 0.32377156615257263, + "learning_rate": 0.0003304291797400269, + "loss": 0.3548, + "step": 24021 + }, + { + "epoch": 1.342151576947789, + "grad_norm": 0.48051658272743225, + "learning_rate": 0.00033040116539668314, + "loss": 0.3348, + "step": 24022 + }, + { + "epoch": 1.3422074475514707, + "grad_norm": 0.7361865043640137, + "learning_rate": 0.00033037315105333934, + "loss": 0.406, + "step": 24023 + }, + { + "epoch": 1.3422633181551527, + "grad_norm": 0.48605138063430786, + "learning_rate": 0.0003303451367099955, + "loss": 0.3427, + "step": 24024 + }, + { + "epoch": 1.3423191887588346, + "grad_norm": 0.6098130941390991, + "learning_rate": 0.00033031712236665175, + "loss": 0.4197, + "step": 24025 + }, + { + "epoch": 1.3423750593625163, + "grad_norm": 0.5747097730636597, + "learning_rate": 0.0003302891080233079, + "loss": 0.5043, + "step": 24026 + }, + { + "epoch": 1.3424309299661983, + "grad_norm": 0.3658466935157776, + "learning_rate": 0.00033026109367996416, + "loss": 0.3776, + "step": 24027 + }, + { + "epoch": 1.3424868005698802, + "grad_norm": 0.5213012099266052, + "learning_rate": 0.0003302330793366203, + "loss": 0.3247, + "step": 24028 + }, + { + "epoch": 1.342542671173562, + "grad_norm": 0.754045844078064, + "learning_rate": 0.0003302050649932766, + "loss": 0.7385, + "step": 24029 + }, + { + "epoch": 1.342598541777244, + "grad_norm": 0.4359830617904663, + "learning_rate": 0.00033017705064993273, + "loss": 0.4573, + "step": 24030 + }, + { + "epoch": 1.3426544123809259, + "grad_norm": 0.6199711561203003, + "learning_rate": 0.000330149036306589, + "loss": 0.3535, + "step": 24031 + }, + { + "epoch": 1.3427102829846076, + "grad_norm": 0.9197871685028076, + "learning_rate": 0.0003301210219632452, + "loss": 0.4864, + "step": 24032 + }, + { + "epoch": 1.3427661535882895, + "grad_norm": 0.3511781096458435, + "learning_rate": 0.0003300930076199014, + "loss": 0.4191, + "step": 24033 + }, + { + "epoch": 1.3428220241919715, + "grad_norm": 0.8055216670036316, + "learning_rate": 0.0003300649932765576, + "loss": 0.4133, + "step": 24034 + }, + { + "epoch": 1.3428778947956532, + "grad_norm": 0.6302999258041382, + "learning_rate": 0.0003300369789332138, + "loss": 0.4291, + "step": 24035 + }, + { + "epoch": 1.3429337653993352, + "grad_norm": 0.3414067029953003, + "learning_rate": 0.00033000896458987, + "loss": 0.3821, + "step": 24036 + }, + { + "epoch": 1.3429896360030171, + "grad_norm": 0.34605321288108826, + "learning_rate": 0.0003299809502465262, + "loss": 0.43, + "step": 24037 + }, + { + "epoch": 1.3430455066066989, + "grad_norm": 3.292076349258423, + "learning_rate": 0.00032995293590318243, + "loss": 0.4987, + "step": 24038 + }, + { + "epoch": 1.3431013772103808, + "grad_norm": 0.422026664018631, + "learning_rate": 0.0003299249215598387, + "loss": 0.4655, + "step": 24039 + }, + { + "epoch": 1.3431572478140628, + "grad_norm": 0.29493799805641174, + "learning_rate": 0.00032989690721649484, + "loss": 0.3409, + "step": 24040 + }, + { + "epoch": 1.3432131184177445, + "grad_norm": 0.6153509020805359, + "learning_rate": 0.0003298688928731511, + "loss": 0.5226, + "step": 24041 + }, + { + "epoch": 1.3432689890214264, + "grad_norm": 0.40690791606903076, + "learning_rate": 0.00032984087852980725, + "loss": 0.3897, + "step": 24042 + }, + { + "epoch": 1.3433248596251082, + "grad_norm": 0.4550330638885498, + "learning_rate": 0.0003298128641864635, + "loss": 0.5479, + "step": 24043 + }, + { + "epoch": 1.3433807302287901, + "grad_norm": 0.4270646572113037, + "learning_rate": 0.00032978484984311966, + "loss": 0.4963, + "step": 24044 + }, + { + "epoch": 1.343436600832472, + "grad_norm": 0.781332790851593, + "learning_rate": 0.00032975683549977587, + "loss": 0.4559, + "step": 24045 + }, + { + "epoch": 1.3434924714361538, + "grad_norm": 0.517257571220398, + "learning_rate": 0.0003297288211564321, + "loss": 0.4444, + "step": 24046 + }, + { + "epoch": 1.3435483420398358, + "grad_norm": 0.35603421926498413, + "learning_rate": 0.0003297008068130883, + "loss": 0.3783, + "step": 24047 + }, + { + "epoch": 1.3436042126435175, + "grad_norm": 0.5129023790359497, + "learning_rate": 0.00032967279246974454, + "loss": 0.5047, + "step": 24048 + }, + { + "epoch": 1.3436600832471994, + "grad_norm": 0.5617061853408813, + "learning_rate": 0.0003296447781264007, + "loss": 0.3978, + "step": 24049 + }, + { + "epoch": 1.3437159538508814, + "grad_norm": 0.5852189064025879, + "learning_rate": 0.00032961676378305695, + "loss": 0.4242, + "step": 24050 + }, + { + "epoch": 1.3437718244545631, + "grad_norm": 0.6436349153518677, + "learning_rate": 0.0003295887494397131, + "loss": 0.52, + "step": 24051 + }, + { + "epoch": 1.343827695058245, + "grad_norm": 0.4717715084552765, + "learning_rate": 0.00032956073509636936, + "loss": 0.484, + "step": 24052 + }, + { + "epoch": 1.343883565661927, + "grad_norm": 0.4077363610267639, + "learning_rate": 0.0003295327207530255, + "loss": 0.3109, + "step": 24053 + }, + { + "epoch": 1.3439394362656087, + "grad_norm": 0.7661545872688293, + "learning_rate": 0.0003295047064096818, + "loss": 0.4207, + "step": 24054 + }, + { + "epoch": 1.3439953068692907, + "grad_norm": 0.5423923134803772, + "learning_rate": 0.000329476692066338, + "loss": 0.5086, + "step": 24055 + }, + { + "epoch": 1.3440511774729726, + "grad_norm": 3.5000555515289307, + "learning_rate": 0.0003294486777229942, + "loss": 0.3722, + "step": 24056 + }, + { + "epoch": 1.3441070480766544, + "grad_norm": 2.8254024982452393, + "learning_rate": 0.0003294206633796504, + "loss": 0.5522, + "step": 24057 + }, + { + "epoch": 1.3441629186803363, + "grad_norm": 0.4250999987125397, + "learning_rate": 0.0003293926490363066, + "loss": 0.3861, + "step": 24058 + }, + { + "epoch": 1.3442187892840183, + "grad_norm": 1.2775375843048096, + "learning_rate": 0.0003293646346929628, + "loss": 0.3213, + "step": 24059 + }, + { + "epoch": 1.3442746598877, + "grad_norm": 0.7317556142807007, + "learning_rate": 0.000329336620349619, + "loss": 0.4556, + "step": 24060 + }, + { + "epoch": 1.344330530491382, + "grad_norm": 0.370909720659256, + "learning_rate": 0.0003293086060062752, + "loss": 0.3962, + "step": 24061 + }, + { + "epoch": 1.344386401095064, + "grad_norm": 0.44599461555480957, + "learning_rate": 0.0003292805916629315, + "loss": 0.418, + "step": 24062 + }, + { + "epoch": 1.3444422716987456, + "grad_norm": 0.4677540361881256, + "learning_rate": 0.00032925257731958763, + "loss": 0.4271, + "step": 24063 + }, + { + "epoch": 1.3444981423024276, + "grad_norm": 0.828528106212616, + "learning_rate": 0.0003292245629762439, + "loss": 0.3431, + "step": 24064 + }, + { + "epoch": 1.3445540129061095, + "grad_norm": 0.4184810519218445, + "learning_rate": 0.00032919654863290004, + "loss": 0.4025, + "step": 24065 + }, + { + "epoch": 1.3446098835097913, + "grad_norm": 0.7943224906921387, + "learning_rate": 0.0003291685342895563, + "loss": 0.4541, + "step": 24066 + }, + { + "epoch": 1.3446657541134732, + "grad_norm": 0.3603660464286804, + "learning_rate": 0.00032914051994621245, + "loss": 0.4508, + "step": 24067 + }, + { + "epoch": 1.3447216247171552, + "grad_norm": 0.38439294695854187, + "learning_rate": 0.00032911250560286866, + "loss": 0.3375, + "step": 24068 + }, + { + "epoch": 1.344777495320837, + "grad_norm": 0.3280714452266693, + "learning_rate": 0.00032908449125952486, + "loss": 0.4195, + "step": 24069 + }, + { + "epoch": 1.3448333659245189, + "grad_norm": 0.41851043701171875, + "learning_rate": 0.00032905647691618107, + "loss": 0.4591, + "step": 24070 + }, + { + "epoch": 1.3448892365282008, + "grad_norm": 0.40591245889663696, + "learning_rate": 0.00032902846257283733, + "loss": 0.484, + "step": 24071 + }, + { + "epoch": 1.3449451071318825, + "grad_norm": 0.390139102935791, + "learning_rate": 0.0003290004482294935, + "loss": 0.341, + "step": 24072 + }, + { + "epoch": 1.3450009777355645, + "grad_norm": 0.43859440088272095, + "learning_rate": 0.00032897243388614974, + "loss": 0.4427, + "step": 24073 + }, + { + "epoch": 1.3450568483392464, + "grad_norm": 0.40246498584747314, + "learning_rate": 0.0003289444195428059, + "loss": 0.4697, + "step": 24074 + }, + { + "epoch": 1.3451127189429282, + "grad_norm": 0.3920481204986572, + "learning_rate": 0.00032891640519946215, + "loss": 0.3774, + "step": 24075 + }, + { + "epoch": 1.3451685895466101, + "grad_norm": 0.49894461035728455, + "learning_rate": 0.0003288883908561183, + "loss": 0.4358, + "step": 24076 + }, + { + "epoch": 1.3452244601502918, + "grad_norm": 0.3771471083164215, + "learning_rate": 0.00032886037651277456, + "loss": 0.3648, + "step": 24077 + }, + { + "epoch": 1.3452803307539738, + "grad_norm": 0.44988444447517395, + "learning_rate": 0.00032883236216943077, + "loss": 0.3769, + "step": 24078 + }, + { + "epoch": 1.3453362013576557, + "grad_norm": 0.6123212575912476, + "learning_rate": 0.000328804347826087, + "loss": 0.3403, + "step": 24079 + }, + { + "epoch": 1.3453920719613375, + "grad_norm": 0.4056486189365387, + "learning_rate": 0.0003287763334827432, + "loss": 0.4368, + "step": 24080 + }, + { + "epoch": 1.3454479425650194, + "grad_norm": 0.7640455365180969, + "learning_rate": 0.0003287483191393994, + "loss": 0.4204, + "step": 24081 + }, + { + "epoch": 1.3455038131687012, + "grad_norm": 0.4993733763694763, + "learning_rate": 0.0003287203047960556, + "loss": 0.4173, + "step": 24082 + }, + { + "epoch": 1.345559683772383, + "grad_norm": 3.1745901107788086, + "learning_rate": 0.0003286922904527118, + "loss": 0.4405, + "step": 24083 + }, + { + "epoch": 1.345615554376065, + "grad_norm": 0.36312371492385864, + "learning_rate": 0.000328664276109368, + "loss": 0.4314, + "step": 24084 + }, + { + "epoch": 1.3456714249797468, + "grad_norm": 0.3165219724178314, + "learning_rate": 0.0003286362617660242, + "loss": 0.3792, + "step": 24085 + }, + { + "epoch": 1.3457272955834287, + "grad_norm": 0.4831896722316742, + "learning_rate": 0.0003286082474226804, + "loss": 0.3743, + "step": 24086 + }, + { + "epoch": 1.3457831661871107, + "grad_norm": 1.5481855869293213, + "learning_rate": 0.0003285802330793367, + "loss": 0.4843, + "step": 24087 + }, + { + "epoch": 1.3458390367907924, + "grad_norm": 0.5255795121192932, + "learning_rate": 0.00032855221873599283, + "loss": 0.5291, + "step": 24088 + }, + { + "epoch": 1.3458949073944744, + "grad_norm": 0.3503130376338959, + "learning_rate": 0.00032852420439264903, + "loss": 0.4238, + "step": 24089 + }, + { + "epoch": 1.3459507779981563, + "grad_norm": 0.4613272547721863, + "learning_rate": 0.00032849619004930524, + "loss": 0.4733, + "step": 24090 + }, + { + "epoch": 1.346006648601838, + "grad_norm": 0.3373156785964966, + "learning_rate": 0.00032846817570596144, + "loss": 0.3618, + "step": 24091 + }, + { + "epoch": 1.34606251920552, + "grad_norm": 0.38140544295310974, + "learning_rate": 0.00032844016136261765, + "loss": 0.4237, + "step": 24092 + }, + { + "epoch": 1.346118389809202, + "grad_norm": 0.4376666247844696, + "learning_rate": 0.00032841214701927386, + "loss": 0.3155, + "step": 24093 + }, + { + "epoch": 1.3461742604128837, + "grad_norm": 1.4560879468917847, + "learning_rate": 0.0003283841326759301, + "loss": 0.4329, + "step": 24094 + }, + { + "epoch": 1.3462301310165656, + "grad_norm": 0.4197195768356323, + "learning_rate": 0.00032835611833258627, + "loss": 0.3815, + "step": 24095 + }, + { + "epoch": 1.3462860016202476, + "grad_norm": 0.4503248929977417, + "learning_rate": 0.00032832810398924253, + "loss": 0.3855, + "step": 24096 + }, + { + "epoch": 1.3463418722239293, + "grad_norm": 0.5302558541297913, + "learning_rate": 0.0003283000896458987, + "loss": 0.3667, + "step": 24097 + }, + { + "epoch": 1.3463977428276113, + "grad_norm": 0.49930721521377563, + "learning_rate": 0.00032827207530255494, + "loss": 0.365, + "step": 24098 + }, + { + "epoch": 1.3464536134312932, + "grad_norm": 0.8954295516014099, + "learning_rate": 0.0003282440609592111, + "loss": 0.4672, + "step": 24099 + }, + { + "epoch": 1.346509484034975, + "grad_norm": 0.8907853960990906, + "learning_rate": 0.00032821604661586735, + "loss": 0.423, + "step": 24100 + }, + { + "epoch": 1.346565354638657, + "grad_norm": 0.8040705323219299, + "learning_rate": 0.0003281880322725235, + "loss": 0.4445, + "step": 24101 + }, + { + "epoch": 1.3466212252423388, + "grad_norm": 0.41484904289245605, + "learning_rate": 0.00032816001792917976, + "loss": 0.4123, + "step": 24102 + }, + { + "epoch": 1.3466770958460206, + "grad_norm": 0.7511779069900513, + "learning_rate": 0.00032813200358583597, + "loss": 0.3777, + "step": 24103 + }, + { + "epoch": 1.3467329664497025, + "grad_norm": 0.7043964862823486, + "learning_rate": 0.0003281039892424922, + "loss": 0.4013, + "step": 24104 + }, + { + "epoch": 1.3467888370533845, + "grad_norm": 0.4681253135204315, + "learning_rate": 0.0003280759748991484, + "loss": 0.4094, + "step": 24105 + }, + { + "epoch": 1.3468447076570662, + "grad_norm": 0.47609081864356995, + "learning_rate": 0.0003280479605558046, + "loss": 0.6868, + "step": 24106 + }, + { + "epoch": 1.3469005782607482, + "grad_norm": 0.7851690053939819, + "learning_rate": 0.0003280199462124608, + "loss": 0.5006, + "step": 24107 + }, + { + "epoch": 1.3469564488644301, + "grad_norm": 0.4762171506881714, + "learning_rate": 0.000327991931869117, + "loss": 0.4835, + "step": 24108 + }, + { + "epoch": 1.3470123194681118, + "grad_norm": 0.9922415614128113, + "learning_rate": 0.0003279639175257732, + "loss": 0.3439, + "step": 24109 + }, + { + "epoch": 1.3470681900717938, + "grad_norm": 5.273690223693848, + "learning_rate": 0.0003279359031824294, + "loss": 0.4016, + "step": 24110 + }, + { + "epoch": 1.3471240606754755, + "grad_norm": 0.5670871138572693, + "learning_rate": 0.0003279078888390856, + "loss": 0.3797, + "step": 24111 + }, + { + "epoch": 1.3471799312791575, + "grad_norm": 0.4637247920036316, + "learning_rate": 0.0003278798744957418, + "loss": 0.414, + "step": 24112 + }, + { + "epoch": 1.3472358018828394, + "grad_norm": 0.3492719531059265, + "learning_rate": 0.000327851860152398, + "loss": 0.441, + "step": 24113 + }, + { + "epoch": 1.3472916724865212, + "grad_norm": 0.6257534623146057, + "learning_rate": 0.00032782384580905423, + "loss": 0.4255, + "step": 24114 + }, + { + "epoch": 1.347347543090203, + "grad_norm": 0.3464243710041046, + "learning_rate": 0.00032779583146571044, + "loss": 0.3265, + "step": 24115 + }, + { + "epoch": 1.3474034136938848, + "grad_norm": 0.6989744901657104, + "learning_rate": 0.00032776781712236664, + "loss": 0.5497, + "step": 24116 + }, + { + "epoch": 1.3474592842975668, + "grad_norm": 0.42054513096809387, + "learning_rate": 0.0003277398027790229, + "loss": 0.4952, + "step": 24117 + }, + { + "epoch": 1.3475151549012487, + "grad_norm": 0.4940257966518402, + "learning_rate": 0.00032771178843567906, + "loss": 0.4763, + "step": 24118 + }, + { + "epoch": 1.3475710255049305, + "grad_norm": 0.6755890846252441, + "learning_rate": 0.0003276837740923353, + "loss": 0.3388, + "step": 24119 + }, + { + "epoch": 1.3476268961086124, + "grad_norm": 1.1169041395187378, + "learning_rate": 0.00032765575974899147, + "loss": 0.3862, + "step": 24120 + }, + { + "epoch": 1.3476827667122944, + "grad_norm": 0.44329094886779785, + "learning_rate": 0.00032762774540564773, + "loss": 0.4321, + "step": 24121 + }, + { + "epoch": 1.347738637315976, + "grad_norm": 0.4013907313346863, + "learning_rate": 0.0003275997310623039, + "loss": 0.2902, + "step": 24122 + }, + { + "epoch": 1.347794507919658, + "grad_norm": 0.3875655233860016, + "learning_rate": 0.00032757171671896014, + "loss": 0.4576, + "step": 24123 + }, + { + "epoch": 1.34785037852334, + "grad_norm": 0.4295107126235962, + "learning_rate": 0.0003275437023756163, + "loss": 0.4163, + "step": 24124 + }, + { + "epoch": 1.3479062491270217, + "grad_norm": 0.45567941665649414, + "learning_rate": 0.00032751568803227255, + "loss": 0.5005, + "step": 24125 + }, + { + "epoch": 1.3479621197307037, + "grad_norm": 0.5054987668991089, + "learning_rate": 0.00032748767368892876, + "loss": 0.3943, + "step": 24126 + }, + { + "epoch": 1.3480179903343856, + "grad_norm": 0.37284982204437256, + "learning_rate": 0.00032745965934558496, + "loss": 0.3571, + "step": 24127 + }, + { + "epoch": 1.3480738609380674, + "grad_norm": 0.3095344603061676, + "learning_rate": 0.00032743164500224117, + "loss": 0.3428, + "step": 24128 + }, + { + "epoch": 1.3481297315417493, + "grad_norm": 0.42661264538764954, + "learning_rate": 0.0003274036306588974, + "loss": 0.3466, + "step": 24129 + }, + { + "epoch": 1.3481856021454313, + "grad_norm": 0.4362087845802307, + "learning_rate": 0.0003273756163155536, + "loss": 0.4031, + "step": 24130 + }, + { + "epoch": 1.348241472749113, + "grad_norm": 0.8822104930877686, + "learning_rate": 0.00032734760197220973, + "loss": 0.4529, + "step": 24131 + }, + { + "epoch": 1.348297343352795, + "grad_norm": 0.4301079213619232, + "learning_rate": 0.000327319587628866, + "loss": 0.4412, + "step": 24132 + }, + { + "epoch": 1.348353213956477, + "grad_norm": 0.7093735337257385, + "learning_rate": 0.0003272915732855222, + "loss": 0.3773, + "step": 24133 + }, + { + "epoch": 1.3484090845601586, + "grad_norm": 0.7641628980636597, + "learning_rate": 0.0003272635589421784, + "loss": 0.5217, + "step": 24134 + }, + { + "epoch": 1.3484649551638406, + "grad_norm": 0.43361619114875793, + "learning_rate": 0.0003272355445988346, + "loss": 0.5112, + "step": 24135 + }, + { + "epoch": 1.3485208257675225, + "grad_norm": 8.33358097076416, + "learning_rate": 0.0003272075302554908, + "loss": 0.3829, + "step": 24136 + }, + { + "epoch": 1.3485766963712043, + "grad_norm": 0.3376998007297516, + "learning_rate": 0.000327179515912147, + "loss": 0.4076, + "step": 24137 + }, + { + "epoch": 1.3486325669748862, + "grad_norm": 0.39956969022750854, + "learning_rate": 0.0003271515015688032, + "loss": 0.3623, + "step": 24138 + }, + { + "epoch": 1.3486884375785682, + "grad_norm": 0.47680404782295227, + "learning_rate": 0.00032712348722545943, + "loss": 0.4064, + "step": 24139 + }, + { + "epoch": 1.3487443081822499, + "grad_norm": 1.892785906791687, + "learning_rate": 0.00032709547288211564, + "loss": 0.4663, + "step": 24140 + }, + { + "epoch": 1.3488001787859318, + "grad_norm": 0.7052782773971558, + "learning_rate": 0.00032706745853877184, + "loss": 0.3335, + "step": 24141 + }, + { + "epoch": 1.3488560493896138, + "grad_norm": 0.40630003809928894, + "learning_rate": 0.0003270394441954281, + "loss": 0.3918, + "step": 24142 + }, + { + "epoch": 1.3489119199932955, + "grad_norm": 0.3660355508327484, + "learning_rate": 0.00032701142985208426, + "loss": 0.3352, + "step": 24143 + }, + { + "epoch": 1.3489677905969775, + "grad_norm": 0.414669394493103, + "learning_rate": 0.0003269834155087405, + "loss": 0.3837, + "step": 24144 + }, + { + "epoch": 1.3490236612006592, + "grad_norm": 0.8310516476631165, + "learning_rate": 0.00032695540116539667, + "loss": 0.4935, + "step": 24145 + }, + { + "epoch": 1.3490795318043411, + "grad_norm": 0.5332468152046204, + "learning_rate": 0.00032692738682205293, + "loss": 0.5144, + "step": 24146 + }, + { + "epoch": 1.349135402408023, + "grad_norm": 0.34993669390678406, + "learning_rate": 0.0003268993724787091, + "loss": 0.3508, + "step": 24147 + }, + { + "epoch": 1.3491912730117048, + "grad_norm": 0.36711305379867554, + "learning_rate": 0.00032687135813536534, + "loss": 0.3819, + "step": 24148 + }, + { + "epoch": 1.3492471436153868, + "grad_norm": 0.822047233581543, + "learning_rate": 0.00032684334379202154, + "loss": 0.4214, + "step": 24149 + }, + { + "epoch": 1.3493030142190685, + "grad_norm": 0.6747734546661377, + "learning_rate": 0.00032681532944867775, + "loss": 0.3771, + "step": 24150 + }, + { + "epoch": 1.3493588848227505, + "grad_norm": 0.3941931128501892, + "learning_rate": 0.00032678731510533396, + "loss": 0.4865, + "step": 24151 + }, + { + "epoch": 1.3494147554264324, + "grad_norm": 0.3207758665084839, + "learning_rate": 0.0003267593007619901, + "loss": 0.3903, + "step": 24152 + }, + { + "epoch": 1.3494706260301141, + "grad_norm": 0.3484880030155182, + "learning_rate": 0.00032673128641864637, + "loss": 0.3737, + "step": 24153 + }, + { + "epoch": 1.349526496633796, + "grad_norm": 1.641538143157959, + "learning_rate": 0.0003267032720753025, + "loss": 0.3228, + "step": 24154 + }, + { + "epoch": 1.349582367237478, + "grad_norm": 0.44958609342575073, + "learning_rate": 0.0003266752577319588, + "loss": 0.5318, + "step": 24155 + }, + { + "epoch": 1.3496382378411598, + "grad_norm": 0.48251476883888245, + "learning_rate": 0.00032664724338861493, + "loss": 0.4127, + "step": 24156 + }, + { + "epoch": 1.3496941084448417, + "grad_norm": 0.5161843299865723, + "learning_rate": 0.0003266192290452712, + "loss": 0.3832, + "step": 24157 + }, + { + "epoch": 1.3497499790485237, + "grad_norm": 0.40255582332611084, + "learning_rate": 0.0003265912147019274, + "loss": 0.3457, + "step": 24158 + }, + { + "epoch": 1.3498058496522054, + "grad_norm": 0.5948872566223145, + "learning_rate": 0.0003265632003585836, + "loss": 0.5609, + "step": 24159 + }, + { + "epoch": 1.3498617202558874, + "grad_norm": 0.8607550263404846, + "learning_rate": 0.0003265351860152398, + "loss": 0.4793, + "step": 24160 + }, + { + "epoch": 1.3499175908595693, + "grad_norm": 2.080822467803955, + "learning_rate": 0.000326507171671896, + "loss": 0.4146, + "step": 24161 + }, + { + "epoch": 1.349973461463251, + "grad_norm": 0.4331766366958618, + "learning_rate": 0.0003264791573285522, + "loss": 0.4627, + "step": 24162 + }, + { + "epoch": 1.350029332066933, + "grad_norm": 0.7538106441497803, + "learning_rate": 0.0003264511429852084, + "loss": 0.4343, + "step": 24163 + }, + { + "epoch": 1.350085202670615, + "grad_norm": 13.23792552947998, + "learning_rate": 0.00032642312864186463, + "loss": 0.3989, + "step": 24164 + }, + { + "epoch": 1.3501410732742967, + "grad_norm": 0.4596737027168274, + "learning_rate": 0.0003263951142985209, + "loss": 0.4952, + "step": 24165 + }, + { + "epoch": 1.3501969438779786, + "grad_norm": 0.34271273016929626, + "learning_rate": 0.00032636709995517704, + "loss": 0.4505, + "step": 24166 + }, + { + "epoch": 1.3502528144816606, + "grad_norm": 0.671598494052887, + "learning_rate": 0.0003263390856118333, + "loss": 0.3656, + "step": 24167 + }, + { + "epoch": 1.3503086850853423, + "grad_norm": 0.3608863055706024, + "learning_rate": 0.00032631107126848946, + "loss": 0.4713, + "step": 24168 + }, + { + "epoch": 1.3503645556890242, + "grad_norm": 0.4235893785953522, + "learning_rate": 0.0003262830569251457, + "loss": 0.4524, + "step": 24169 + }, + { + "epoch": 1.3504204262927062, + "grad_norm": 0.5433653593063354, + "learning_rate": 0.00032625504258180187, + "loss": 0.4395, + "step": 24170 + }, + { + "epoch": 1.350476296896388, + "grad_norm": 0.36958810687065125, + "learning_rate": 0.0003262270282384581, + "loss": 0.4677, + "step": 24171 + }, + { + "epoch": 1.3505321675000699, + "grad_norm": 0.41136595606803894, + "learning_rate": 0.0003261990138951143, + "loss": 0.4141, + "step": 24172 + }, + { + "epoch": 1.3505880381037518, + "grad_norm": 0.28011181950569153, + "learning_rate": 0.0003261709995517705, + "loss": 0.3625, + "step": 24173 + }, + { + "epoch": 1.3506439087074336, + "grad_norm": 0.31735214591026306, + "learning_rate": 0.00032614298520842674, + "loss": 0.4238, + "step": 24174 + }, + { + "epoch": 1.3506997793111155, + "grad_norm": 0.8162559270858765, + "learning_rate": 0.0003261149708650829, + "loss": 0.45, + "step": 24175 + }, + { + "epoch": 1.3507556499147975, + "grad_norm": 0.38951683044433594, + "learning_rate": 0.00032608695652173916, + "loss": 0.4099, + "step": 24176 + }, + { + "epoch": 1.3508115205184792, + "grad_norm": 0.42742088437080383, + "learning_rate": 0.0003260589421783953, + "loss": 0.3752, + "step": 24177 + }, + { + "epoch": 1.3508673911221611, + "grad_norm": 0.3354864716529846, + "learning_rate": 0.00032603092783505157, + "loss": 0.388, + "step": 24178 + }, + { + "epoch": 1.3509232617258429, + "grad_norm": 0.4117792546749115, + "learning_rate": 0.0003260029134917077, + "loss": 0.453, + "step": 24179 + }, + { + "epoch": 1.3509791323295248, + "grad_norm": 0.6764058470726013, + "learning_rate": 0.000325974899148364, + "loss": 0.4825, + "step": 24180 + }, + { + "epoch": 1.3510350029332068, + "grad_norm": 0.36990076303482056, + "learning_rate": 0.0003259468848050202, + "loss": 0.4536, + "step": 24181 + }, + { + "epoch": 1.3510908735368885, + "grad_norm": 0.4504822790622711, + "learning_rate": 0.0003259188704616764, + "loss": 0.4906, + "step": 24182 + }, + { + "epoch": 1.3511467441405705, + "grad_norm": 1.050680160522461, + "learning_rate": 0.0003258908561183326, + "loss": 0.3221, + "step": 24183 + }, + { + "epoch": 1.3512026147442522, + "grad_norm": 0.36283165216445923, + "learning_rate": 0.0003258628417749888, + "loss": 0.4327, + "step": 24184 + }, + { + "epoch": 1.3512584853479341, + "grad_norm": 0.6062344312667847, + "learning_rate": 0.000325834827431645, + "loss": 0.4995, + "step": 24185 + }, + { + "epoch": 1.351314355951616, + "grad_norm": 0.5685015916824341, + "learning_rate": 0.0003258068130883012, + "loss": 0.4236, + "step": 24186 + }, + { + "epoch": 1.3513702265552978, + "grad_norm": 0.369754433631897, + "learning_rate": 0.0003257787987449574, + "loss": 0.448, + "step": 24187 + }, + { + "epoch": 1.3514260971589798, + "grad_norm": 1.1082576513290405, + "learning_rate": 0.0003257507844016137, + "loss": 0.5491, + "step": 24188 + }, + { + "epoch": 1.3514819677626617, + "grad_norm": 1.7026315927505493, + "learning_rate": 0.00032572277005826983, + "loss": 0.3898, + "step": 24189 + }, + { + "epoch": 1.3515378383663434, + "grad_norm": 0.3260831832885742, + "learning_rate": 0.0003256947557149261, + "loss": 0.3284, + "step": 24190 + }, + { + "epoch": 1.3515937089700254, + "grad_norm": 1.1437277793884277, + "learning_rate": 0.00032566674137158224, + "loss": 0.3714, + "step": 24191 + }, + { + "epoch": 1.3516495795737073, + "grad_norm": 0.3942250907421112, + "learning_rate": 0.0003256387270282385, + "loss": 0.4128, + "step": 24192 + }, + { + "epoch": 1.351705450177389, + "grad_norm": 0.6129845380783081, + "learning_rate": 0.00032561071268489465, + "loss": 0.4247, + "step": 24193 + }, + { + "epoch": 1.351761320781071, + "grad_norm": 0.7314615249633789, + "learning_rate": 0.00032558269834155086, + "loss": 0.3414, + "step": 24194 + }, + { + "epoch": 1.351817191384753, + "grad_norm": 0.3553287088871002, + "learning_rate": 0.00032555468399820707, + "loss": 0.359, + "step": 24195 + }, + { + "epoch": 1.3518730619884347, + "grad_norm": 0.35312575101852417, + "learning_rate": 0.00032552666965486327, + "loss": 0.3663, + "step": 24196 + }, + { + "epoch": 1.3519289325921167, + "grad_norm": 1.2901912927627563, + "learning_rate": 0.00032549865531151953, + "loss": 0.4707, + "step": 24197 + }, + { + "epoch": 1.3519848031957986, + "grad_norm": 0.3400590717792511, + "learning_rate": 0.0003254706409681757, + "loss": 0.3601, + "step": 24198 + }, + { + "epoch": 1.3520406737994803, + "grad_norm": 2.5228192806243896, + "learning_rate": 0.00032544262662483194, + "loss": 0.4354, + "step": 24199 + }, + { + "epoch": 1.3520965444031623, + "grad_norm": 0.7816308736801147, + "learning_rate": 0.0003254146122814881, + "loss": 0.3674, + "step": 24200 + }, + { + "epoch": 1.3521524150068442, + "grad_norm": 0.4558485448360443, + "learning_rate": 0.00032538659793814436, + "loss": 0.4262, + "step": 24201 + }, + { + "epoch": 1.352208285610526, + "grad_norm": 0.48882460594177246, + "learning_rate": 0.0003253585835948005, + "loss": 0.4766, + "step": 24202 + }, + { + "epoch": 1.352264156214208, + "grad_norm": 0.8024447560310364, + "learning_rate": 0.00032533056925145677, + "loss": 0.5197, + "step": 24203 + }, + { + "epoch": 1.3523200268178899, + "grad_norm": 0.5205121040344238, + "learning_rate": 0.000325302554908113, + "loss": 0.4685, + "step": 24204 + }, + { + "epoch": 1.3523758974215716, + "grad_norm": 0.3009337782859802, + "learning_rate": 0.0003252745405647692, + "loss": 0.3749, + "step": 24205 + }, + { + "epoch": 1.3524317680252536, + "grad_norm": 0.4163579046726227, + "learning_rate": 0.0003252465262214254, + "loss": 0.4809, + "step": 24206 + }, + { + "epoch": 1.3524876386289355, + "grad_norm": 0.3608185052871704, + "learning_rate": 0.0003252185118780816, + "loss": 0.339, + "step": 24207 + }, + { + "epoch": 1.3525435092326172, + "grad_norm": 0.4698604643344879, + "learning_rate": 0.0003251904975347378, + "loss": 0.4653, + "step": 24208 + }, + { + "epoch": 1.3525993798362992, + "grad_norm": 17.252145767211914, + "learning_rate": 0.000325162483191394, + "loss": 0.4481, + "step": 24209 + }, + { + "epoch": 1.3526552504399811, + "grad_norm": 0.604186475276947, + "learning_rate": 0.0003251344688480502, + "loss": 0.4787, + "step": 24210 + }, + { + "epoch": 1.3527111210436629, + "grad_norm": 0.548440158367157, + "learning_rate": 0.0003251064545047064, + "loss": 0.3335, + "step": 24211 + }, + { + "epoch": 1.3527669916473448, + "grad_norm": 0.6337429285049438, + "learning_rate": 0.0003250784401613626, + "loss": 0.4554, + "step": 24212 + }, + { + "epoch": 1.3528228622510265, + "grad_norm": 0.43641430139541626, + "learning_rate": 0.0003250504258180189, + "loss": 0.4122, + "step": 24213 + }, + { + "epoch": 1.3528787328547085, + "grad_norm": 0.7177774906158447, + "learning_rate": 0.00032502241147467503, + "loss": 0.4101, + "step": 24214 + }, + { + "epoch": 1.3529346034583905, + "grad_norm": 0.535201370716095, + "learning_rate": 0.00032499439713133124, + "loss": 0.3703, + "step": 24215 + }, + { + "epoch": 1.3529904740620722, + "grad_norm": 0.5379929542541504, + "learning_rate": 0.00032496638278798744, + "loss": 0.4378, + "step": 24216 + }, + { + "epoch": 1.3530463446657541, + "grad_norm": 0.4119912385940552, + "learning_rate": 0.00032493836844464365, + "loss": 0.3881, + "step": 24217 + }, + { + "epoch": 1.3531022152694359, + "grad_norm": 0.5589600801467896, + "learning_rate": 0.00032491035410129985, + "loss": 0.4237, + "step": 24218 + }, + { + "epoch": 1.3531580858731178, + "grad_norm": 0.49631887674331665, + "learning_rate": 0.00032488233975795606, + "loss": 0.4165, + "step": 24219 + }, + { + "epoch": 1.3532139564767998, + "grad_norm": 0.7281591892242432, + "learning_rate": 0.0003248543254146123, + "loss": 0.5741, + "step": 24220 + }, + { + "epoch": 1.3532698270804815, + "grad_norm": 0.6635600924491882, + "learning_rate": 0.00032482631107126847, + "loss": 0.3516, + "step": 24221 + }, + { + "epoch": 1.3533256976841634, + "grad_norm": 0.8216168284416199, + "learning_rate": 0.00032479829672792473, + "loss": 0.4605, + "step": 24222 + }, + { + "epoch": 1.3533815682878454, + "grad_norm": 0.42729663848876953, + "learning_rate": 0.0003247702823845809, + "loss": 0.3954, + "step": 24223 + }, + { + "epoch": 1.3534374388915271, + "grad_norm": 0.5140583515167236, + "learning_rate": 0.00032474226804123714, + "loss": 0.5363, + "step": 24224 + }, + { + "epoch": 1.353493309495209, + "grad_norm": 0.36371397972106934, + "learning_rate": 0.0003247142536978933, + "loss": 0.3824, + "step": 24225 + }, + { + "epoch": 1.353549180098891, + "grad_norm": 0.423991858959198, + "learning_rate": 0.00032468623935454956, + "loss": 0.3446, + "step": 24226 + }, + { + "epoch": 1.3536050507025728, + "grad_norm": 0.40827539563179016, + "learning_rate": 0.0003246582250112057, + "loss": 0.4881, + "step": 24227 + }, + { + "epoch": 1.3536609213062547, + "grad_norm": 0.5632182955741882, + "learning_rate": 0.00032463021066786197, + "loss": 0.4, + "step": 24228 + }, + { + "epoch": 1.3537167919099367, + "grad_norm": 0.9441452622413635, + "learning_rate": 0.00032460219632451817, + "loss": 0.3975, + "step": 24229 + }, + { + "epoch": 1.3537726625136184, + "grad_norm": 0.45878762006759644, + "learning_rate": 0.0003245741819811744, + "loss": 0.5425, + "step": 24230 + }, + { + "epoch": 1.3538285331173003, + "grad_norm": 0.40865814685821533, + "learning_rate": 0.0003245461676378306, + "loss": 0.4761, + "step": 24231 + }, + { + "epoch": 1.3538844037209823, + "grad_norm": 0.38108330965042114, + "learning_rate": 0.0003245181532944868, + "loss": 0.3807, + "step": 24232 + }, + { + "epoch": 1.353940274324664, + "grad_norm": 0.3842355012893677, + "learning_rate": 0.000324490138951143, + "loss": 0.4167, + "step": 24233 + }, + { + "epoch": 1.353996144928346, + "grad_norm": 1.023078203201294, + "learning_rate": 0.0003244621246077992, + "loss": 0.3928, + "step": 24234 + }, + { + "epoch": 1.354052015532028, + "grad_norm": 0.4083908796310425, + "learning_rate": 0.0003244341102644554, + "loss": 0.4219, + "step": 24235 + }, + { + "epoch": 1.3541078861357096, + "grad_norm": 0.41740983724594116, + "learning_rate": 0.00032440609592111167, + "loss": 0.4183, + "step": 24236 + }, + { + "epoch": 1.3541637567393916, + "grad_norm": 0.3497413396835327, + "learning_rate": 0.0003243780815777678, + "loss": 0.3396, + "step": 24237 + }, + { + "epoch": 1.3542196273430736, + "grad_norm": 2.673888921737671, + "learning_rate": 0.000324350067234424, + "loss": 0.4436, + "step": 24238 + }, + { + "epoch": 1.3542754979467553, + "grad_norm": 0.9925609827041626, + "learning_rate": 0.00032432205289108023, + "loss": 0.453, + "step": 24239 + }, + { + "epoch": 1.3543313685504372, + "grad_norm": 0.47515881061553955, + "learning_rate": 0.00032429403854773644, + "loss": 0.5236, + "step": 24240 + }, + { + "epoch": 1.3543872391541192, + "grad_norm": 0.4732493460178375, + "learning_rate": 0.00032426602420439264, + "loss": 0.4715, + "step": 24241 + }, + { + "epoch": 1.354443109757801, + "grad_norm": 0.4686375856399536, + "learning_rate": 0.00032423800986104885, + "loss": 0.4775, + "step": 24242 + }, + { + "epoch": 1.3544989803614829, + "grad_norm": 0.8262261152267456, + "learning_rate": 0.00032420999551770505, + "loss": 0.5256, + "step": 24243 + }, + { + "epoch": 1.3545548509651648, + "grad_norm": 0.31089723110198975, + "learning_rate": 0.00032418198117436126, + "loss": 0.395, + "step": 24244 + }, + { + "epoch": 1.3546107215688465, + "grad_norm": 0.38349649310112, + "learning_rate": 0.0003241539668310175, + "loss": 0.4039, + "step": 24245 + }, + { + "epoch": 1.3546665921725285, + "grad_norm": 0.37388095259666443, + "learning_rate": 0.00032412595248767367, + "loss": 0.3817, + "step": 24246 + }, + { + "epoch": 1.3547224627762102, + "grad_norm": 0.35042187571525574, + "learning_rate": 0.00032409793814432993, + "loss": 0.3614, + "step": 24247 + }, + { + "epoch": 1.3547783333798922, + "grad_norm": 0.4847610294818878, + "learning_rate": 0.0003240699238009861, + "loss": 0.4579, + "step": 24248 + }, + { + "epoch": 1.3548342039835741, + "grad_norm": 0.5264451503753662, + "learning_rate": 0.00032404190945764234, + "loss": 0.4836, + "step": 24249 + }, + { + "epoch": 1.3548900745872559, + "grad_norm": 0.6056839227676392, + "learning_rate": 0.0003240138951142985, + "loss": 0.5074, + "step": 24250 + }, + { + "epoch": 1.3549459451909378, + "grad_norm": 2.446103572845459, + "learning_rate": 0.00032398588077095475, + "loss": 0.4029, + "step": 24251 + }, + { + "epoch": 1.3550018157946195, + "grad_norm": 0.40597087144851685, + "learning_rate": 0.00032395786642761096, + "loss": 0.3975, + "step": 24252 + }, + { + "epoch": 1.3550576863983015, + "grad_norm": 0.44031190872192383, + "learning_rate": 0.00032392985208426717, + "loss": 0.4954, + "step": 24253 + }, + { + "epoch": 1.3551135570019834, + "grad_norm": 0.5738152861595154, + "learning_rate": 0.00032390183774092337, + "loss": 0.3544, + "step": 24254 + }, + { + "epoch": 1.3551694276056652, + "grad_norm": 0.4629400074481964, + "learning_rate": 0.0003238738233975796, + "loss": 0.5305, + "step": 24255 + }, + { + "epoch": 1.3552252982093471, + "grad_norm": 0.3483300805091858, + "learning_rate": 0.0003238458090542358, + "loss": 0.3497, + "step": 24256 + }, + { + "epoch": 1.355281168813029, + "grad_norm": 0.6422309875488281, + "learning_rate": 0.000323817794710892, + "loss": 0.4174, + "step": 24257 + }, + { + "epoch": 1.3553370394167108, + "grad_norm": 0.43150079250335693, + "learning_rate": 0.0003237897803675482, + "loss": 0.453, + "step": 24258 + }, + { + "epoch": 1.3553929100203927, + "grad_norm": 0.7247979640960693, + "learning_rate": 0.00032376176602420435, + "loss": 0.4106, + "step": 24259 + }, + { + "epoch": 1.3554487806240747, + "grad_norm": 0.31776610016822815, + "learning_rate": 0.0003237337516808606, + "loss": 0.2944, + "step": 24260 + }, + { + "epoch": 1.3555046512277564, + "grad_norm": 3.854092836380005, + "learning_rate": 0.0003237057373375168, + "loss": 0.405, + "step": 24261 + }, + { + "epoch": 1.3555605218314384, + "grad_norm": 0.5932706594467163, + "learning_rate": 0.000323677722994173, + "loss": 0.4699, + "step": 24262 + }, + { + "epoch": 1.3556163924351203, + "grad_norm": 0.519911527633667, + "learning_rate": 0.0003236497086508292, + "loss": 0.3998, + "step": 24263 + }, + { + "epoch": 1.355672263038802, + "grad_norm": 1.1077357530593872, + "learning_rate": 0.00032362169430748543, + "loss": 0.4604, + "step": 24264 + }, + { + "epoch": 1.355728133642484, + "grad_norm": 0.4856237769126892, + "learning_rate": 0.00032359367996414164, + "loss": 0.4101, + "step": 24265 + }, + { + "epoch": 1.355784004246166, + "grad_norm": 0.3805868625640869, + "learning_rate": 0.00032356566562079784, + "loss": 0.3833, + "step": 24266 + }, + { + "epoch": 1.3558398748498477, + "grad_norm": 0.46297067403793335, + "learning_rate": 0.00032353765127745405, + "loss": 0.4336, + "step": 24267 + }, + { + "epoch": 1.3558957454535296, + "grad_norm": 11.88827896118164, + "learning_rate": 0.0003235096369341103, + "loss": 0.4905, + "step": 24268 + }, + { + "epoch": 1.3559516160572116, + "grad_norm": 0.43945106863975525, + "learning_rate": 0.00032348162259076646, + "loss": 0.3642, + "step": 24269 + }, + { + "epoch": 1.3560074866608933, + "grad_norm": 0.4332883059978485, + "learning_rate": 0.0003234536082474227, + "loss": 0.4044, + "step": 24270 + }, + { + "epoch": 1.3560633572645753, + "grad_norm": 0.4519623816013336, + "learning_rate": 0.00032342559390407887, + "loss": 0.4269, + "step": 24271 + }, + { + "epoch": 1.3561192278682572, + "grad_norm": 0.5308061242103577, + "learning_rate": 0.00032339757956073513, + "loss": 0.4366, + "step": 24272 + }, + { + "epoch": 1.356175098471939, + "grad_norm": 0.5764955878257751, + "learning_rate": 0.0003233695652173913, + "loss": 0.4022, + "step": 24273 + }, + { + "epoch": 1.356230969075621, + "grad_norm": 0.5484359264373779, + "learning_rate": 0.00032334155087404754, + "loss": 0.3558, + "step": 24274 + }, + { + "epoch": 1.3562868396793029, + "grad_norm": 0.4250751733779907, + "learning_rate": 0.00032331353653070375, + "loss": 0.4606, + "step": 24275 + }, + { + "epoch": 1.3563427102829846, + "grad_norm": 0.4829278290271759, + "learning_rate": 0.00032328552218735995, + "loss": 0.4758, + "step": 24276 + }, + { + "epoch": 1.3563985808866665, + "grad_norm": 0.39336085319519043, + "learning_rate": 0.00032325750784401616, + "loss": 0.3604, + "step": 24277 + }, + { + "epoch": 1.3564544514903485, + "grad_norm": 0.418163925409317, + "learning_rate": 0.00032322949350067237, + "loss": 0.4046, + "step": 24278 + }, + { + "epoch": 1.3565103220940302, + "grad_norm": 0.5150154232978821, + "learning_rate": 0.00032320147915732857, + "loss": 0.4101, + "step": 24279 + }, + { + "epoch": 1.3565661926977122, + "grad_norm": 0.8980719447135925, + "learning_rate": 0.0003231734648139847, + "loss": 0.4462, + "step": 24280 + }, + { + "epoch": 1.356622063301394, + "grad_norm": 0.38153916597366333, + "learning_rate": 0.000323145450470641, + "loss": 0.3702, + "step": 24281 + }, + { + "epoch": 1.3566779339050759, + "grad_norm": 0.42829594016075134, + "learning_rate": 0.00032311743612729714, + "loss": 0.401, + "step": 24282 + }, + { + "epoch": 1.3567338045087576, + "grad_norm": 0.5368427634239197, + "learning_rate": 0.0003230894217839534, + "loss": 0.3498, + "step": 24283 + }, + { + "epoch": 1.3567896751124395, + "grad_norm": 1.821367621421814, + "learning_rate": 0.0003230614074406096, + "loss": 0.4164, + "step": 24284 + }, + { + "epoch": 1.3568455457161215, + "grad_norm": 0.42636632919311523, + "learning_rate": 0.0003230333930972658, + "loss": 0.3779, + "step": 24285 + }, + { + "epoch": 1.3569014163198032, + "grad_norm": 0.7462283968925476, + "learning_rate": 0.000323005378753922, + "loss": 0.3702, + "step": 24286 + }, + { + "epoch": 1.3569572869234852, + "grad_norm": 0.4659004509449005, + "learning_rate": 0.0003229773644105782, + "loss": 0.4104, + "step": 24287 + }, + { + "epoch": 1.3570131575271671, + "grad_norm": 0.37262365221977234, + "learning_rate": 0.0003229493500672344, + "loss": 0.4173, + "step": 24288 + }, + { + "epoch": 1.3570690281308488, + "grad_norm": 1.4753694534301758, + "learning_rate": 0.00032292133572389063, + "loss": 0.5777, + "step": 24289 + }, + { + "epoch": 1.3571248987345308, + "grad_norm": 0.6836430430412292, + "learning_rate": 0.00032289332138054684, + "loss": 0.4448, + "step": 24290 + }, + { + "epoch": 1.3571807693382127, + "grad_norm": 0.358907014131546, + "learning_rate": 0.0003228653070372031, + "loss": 0.3536, + "step": 24291 + }, + { + "epoch": 1.3572366399418945, + "grad_norm": 1.4453364610671997, + "learning_rate": 0.00032283729269385925, + "loss": 0.5098, + "step": 24292 + }, + { + "epoch": 1.3572925105455764, + "grad_norm": 0.9450571537017822, + "learning_rate": 0.0003228092783505155, + "loss": 0.4665, + "step": 24293 + }, + { + "epoch": 1.3573483811492584, + "grad_norm": 0.4638526737689972, + "learning_rate": 0.00032278126400717166, + "loss": 0.4624, + "step": 24294 + }, + { + "epoch": 1.35740425175294, + "grad_norm": 0.48939061164855957, + "learning_rate": 0.0003227532496638279, + "loss": 0.4554, + "step": 24295 + }, + { + "epoch": 1.357460122356622, + "grad_norm": 0.378181517124176, + "learning_rate": 0.00032272523532048407, + "loss": 0.3947, + "step": 24296 + }, + { + "epoch": 1.357515992960304, + "grad_norm": 0.6646251082420349, + "learning_rate": 0.00032269722097714033, + "loss": 0.4059, + "step": 24297 + }, + { + "epoch": 1.3575718635639857, + "grad_norm": 0.5390288233757019, + "learning_rate": 0.0003226692066337965, + "loss": 0.3757, + "step": 24298 + }, + { + "epoch": 1.3576277341676677, + "grad_norm": 0.7418566942214966, + "learning_rate": 0.00032264119229045274, + "loss": 0.3968, + "step": 24299 + }, + { + "epoch": 1.3576836047713496, + "grad_norm": 0.3556757867336273, + "learning_rate": 0.00032261317794710895, + "loss": 0.3609, + "step": 24300 + }, + { + "epoch": 1.3577394753750314, + "grad_norm": 0.3980995714664459, + "learning_rate": 0.0003225851636037651, + "loss": 0.4922, + "step": 24301 + }, + { + "epoch": 1.3577953459787133, + "grad_norm": 0.4160306751728058, + "learning_rate": 0.00032255714926042136, + "loss": 0.3734, + "step": 24302 + }, + { + "epoch": 1.3578512165823953, + "grad_norm": 0.5129803419113159, + "learning_rate": 0.0003225291349170775, + "loss": 0.4322, + "step": 24303 + }, + { + "epoch": 1.357907087186077, + "grad_norm": 0.5537702441215515, + "learning_rate": 0.00032250112057373377, + "loss": 0.4062, + "step": 24304 + }, + { + "epoch": 1.357962957789759, + "grad_norm": 1.2811039686203003, + "learning_rate": 0.0003224731062303899, + "loss": 0.4495, + "step": 24305 + }, + { + "epoch": 1.358018828393441, + "grad_norm": 0.44477352499961853, + "learning_rate": 0.0003224450918870462, + "loss": 0.4097, + "step": 24306 + }, + { + "epoch": 1.3580746989971226, + "grad_norm": 1.7901041507720947, + "learning_rate": 0.0003224170775437024, + "loss": 0.5222, + "step": 24307 + }, + { + "epoch": 1.3581305696008046, + "grad_norm": 0.3975430428981781, + "learning_rate": 0.0003223890632003586, + "loss": 0.4189, + "step": 24308 + }, + { + "epoch": 1.3581864402044865, + "grad_norm": 0.419068843126297, + "learning_rate": 0.0003223610488570148, + "loss": 0.4452, + "step": 24309 + }, + { + "epoch": 1.3582423108081683, + "grad_norm": 0.574730634689331, + "learning_rate": 0.000322333034513671, + "loss": 0.6384, + "step": 24310 + }, + { + "epoch": 1.3582981814118502, + "grad_norm": 0.6659075021743774, + "learning_rate": 0.0003223050201703272, + "loss": 0.4877, + "step": 24311 + }, + { + "epoch": 1.3583540520155322, + "grad_norm": 0.6732210516929626, + "learning_rate": 0.0003222770058269834, + "loss": 0.4058, + "step": 24312 + }, + { + "epoch": 1.358409922619214, + "grad_norm": 0.42502883076667786, + "learning_rate": 0.0003222489914836396, + "loss": 0.4936, + "step": 24313 + }, + { + "epoch": 1.3584657932228958, + "grad_norm": 0.49028855562210083, + "learning_rate": 0.00032222097714029583, + "loss": 0.4025, + "step": 24314 + }, + { + "epoch": 1.3585216638265776, + "grad_norm": 0.6335682272911072, + "learning_rate": 0.00032219296279695204, + "loss": 0.4908, + "step": 24315 + }, + { + "epoch": 1.3585775344302595, + "grad_norm": 0.4184604287147522, + "learning_rate": 0.0003221649484536083, + "loss": 0.4306, + "step": 24316 + }, + { + "epoch": 1.3586334050339413, + "grad_norm": 0.34322139620780945, + "learning_rate": 0.00032213693411026445, + "loss": 0.3496, + "step": 24317 + }, + { + "epoch": 1.3586892756376232, + "grad_norm": 0.7651280760765076, + "learning_rate": 0.0003221089197669207, + "loss": 0.6447, + "step": 24318 + }, + { + "epoch": 1.3587451462413052, + "grad_norm": 0.35073962807655334, + "learning_rate": 0.00032208090542357686, + "loss": 0.4413, + "step": 24319 + }, + { + "epoch": 1.3588010168449869, + "grad_norm": 0.377806693315506, + "learning_rate": 0.0003220528910802331, + "loss": 0.408, + "step": 24320 + }, + { + "epoch": 1.3588568874486688, + "grad_norm": 0.4348628520965576, + "learning_rate": 0.00032202487673688927, + "loss": 0.4512, + "step": 24321 + }, + { + "epoch": 1.3589127580523508, + "grad_norm": 0.9253932237625122, + "learning_rate": 0.0003219968623935455, + "loss": 0.4369, + "step": 24322 + }, + { + "epoch": 1.3589686286560325, + "grad_norm": 0.3740878403186798, + "learning_rate": 0.00032196884805020174, + "loss": 0.3963, + "step": 24323 + }, + { + "epoch": 1.3590244992597145, + "grad_norm": 0.4741731286048889, + "learning_rate": 0.0003219408337068579, + "loss": 0.4889, + "step": 24324 + }, + { + "epoch": 1.3590803698633964, + "grad_norm": 0.5855237245559692, + "learning_rate": 0.00032191281936351415, + "loss": 0.5421, + "step": 24325 + }, + { + "epoch": 1.3591362404670781, + "grad_norm": 0.343892902135849, + "learning_rate": 0.0003218848050201703, + "loss": 0.4066, + "step": 24326 + }, + { + "epoch": 1.35919211107076, + "grad_norm": 0.54657381772995, + "learning_rate": 0.00032185679067682656, + "loss": 0.3816, + "step": 24327 + }, + { + "epoch": 1.359247981674442, + "grad_norm": 0.3060499429702759, + "learning_rate": 0.0003218287763334827, + "loss": 0.3408, + "step": 24328 + }, + { + "epoch": 1.3593038522781238, + "grad_norm": 0.6541352868080139, + "learning_rate": 0.00032180076199013897, + "loss": 0.3938, + "step": 24329 + }, + { + "epoch": 1.3593597228818057, + "grad_norm": 0.5887869000434875, + "learning_rate": 0.0003217727476467951, + "loss": 0.5504, + "step": 24330 + }, + { + "epoch": 1.3594155934854877, + "grad_norm": 0.3581368327140808, + "learning_rate": 0.0003217447333034514, + "loss": 0.4106, + "step": 24331 + }, + { + "epoch": 1.3594714640891694, + "grad_norm": 0.3977319002151489, + "learning_rate": 0.0003217167189601076, + "loss": 0.4373, + "step": 24332 + }, + { + "epoch": 1.3595273346928514, + "grad_norm": 0.38842007517814636, + "learning_rate": 0.0003216887046167638, + "loss": 0.4057, + "step": 24333 + }, + { + "epoch": 1.3595832052965333, + "grad_norm": 0.3422843813896179, + "learning_rate": 0.00032166069027342, + "loss": 0.5017, + "step": 24334 + }, + { + "epoch": 1.359639075900215, + "grad_norm": 0.45384833216667175, + "learning_rate": 0.0003216326759300762, + "loss": 0.476, + "step": 24335 + }, + { + "epoch": 1.359694946503897, + "grad_norm": 0.8501861095428467, + "learning_rate": 0.0003216046615867324, + "loss": 0.3874, + "step": 24336 + }, + { + "epoch": 1.359750817107579, + "grad_norm": 0.428409606218338, + "learning_rate": 0.0003215766472433886, + "loss": 0.3111, + "step": 24337 + }, + { + "epoch": 1.3598066877112607, + "grad_norm": 1.147321105003357, + "learning_rate": 0.0003215486329000448, + "loss": 0.4013, + "step": 24338 + }, + { + "epoch": 1.3598625583149426, + "grad_norm": 0.4875951111316681, + "learning_rate": 0.0003215206185567011, + "loss": 0.5141, + "step": 24339 + }, + { + "epoch": 1.3599184289186246, + "grad_norm": 0.317350298166275, + "learning_rate": 0.00032149260421335724, + "loss": 0.4251, + "step": 24340 + }, + { + "epoch": 1.3599742995223063, + "grad_norm": 1.1040380001068115, + "learning_rate": 0.0003214645898700135, + "loss": 0.4588, + "step": 24341 + }, + { + "epoch": 1.3600301701259883, + "grad_norm": 0.3913990557193756, + "learning_rate": 0.00032143657552666965, + "loss": 0.4111, + "step": 24342 + }, + { + "epoch": 1.3600860407296702, + "grad_norm": 0.5505000352859497, + "learning_rate": 0.00032140856118332585, + "loss": 0.6686, + "step": 24343 + }, + { + "epoch": 1.360141911333352, + "grad_norm": 0.6947262287139893, + "learning_rate": 0.00032138054683998206, + "loss": 0.4573, + "step": 24344 + }, + { + "epoch": 1.360197781937034, + "grad_norm": 1.2232115268707275, + "learning_rate": 0.00032135253249663826, + "loss": 0.3976, + "step": 24345 + }, + { + "epoch": 1.3602536525407158, + "grad_norm": 0.4365038573741913, + "learning_rate": 0.0003213245181532945, + "loss": 0.5117, + "step": 24346 + }, + { + "epoch": 1.3603095231443976, + "grad_norm": 0.37816643714904785, + "learning_rate": 0.0003212965038099507, + "loss": 0.4411, + "step": 24347 + }, + { + "epoch": 1.3603653937480795, + "grad_norm": 0.5340506434440613, + "learning_rate": 0.00032126848946660694, + "loss": 0.4654, + "step": 24348 + }, + { + "epoch": 1.3604212643517613, + "grad_norm": 0.421539306640625, + "learning_rate": 0.0003212404751232631, + "loss": 0.416, + "step": 24349 + }, + { + "epoch": 1.3604771349554432, + "grad_norm": 0.9257697463035583, + "learning_rate": 0.00032121246077991935, + "loss": 0.3652, + "step": 24350 + }, + { + "epoch": 1.360533005559125, + "grad_norm": 0.540880024433136, + "learning_rate": 0.0003211844464365755, + "loss": 0.3728, + "step": 24351 + }, + { + "epoch": 1.3605888761628069, + "grad_norm": 1.9397530555725098, + "learning_rate": 0.00032115643209323176, + "loss": 0.5034, + "step": 24352 + }, + { + "epoch": 1.3606447467664888, + "grad_norm": 0.25615158677101135, + "learning_rate": 0.0003211284177498879, + "loss": 0.3802, + "step": 24353 + }, + { + "epoch": 1.3607006173701706, + "grad_norm": 0.6919043660163879, + "learning_rate": 0.00032110040340654417, + "loss": 0.4931, + "step": 24354 + }, + { + "epoch": 1.3607564879738525, + "grad_norm": 1.4106855392456055, + "learning_rate": 0.0003210723890632004, + "loss": 0.3911, + "step": 24355 + }, + { + "epoch": 1.3608123585775345, + "grad_norm": 0.4484522044658661, + "learning_rate": 0.0003210443747198566, + "loss": 0.4861, + "step": 24356 + }, + { + "epoch": 1.3608682291812162, + "grad_norm": 0.38122060894966125, + "learning_rate": 0.0003210163603765128, + "loss": 0.4268, + "step": 24357 + }, + { + "epoch": 1.3609240997848981, + "grad_norm": 0.4225742518901825, + "learning_rate": 0.000320988346033169, + "loss": 0.4616, + "step": 24358 + }, + { + "epoch": 1.36097997038858, + "grad_norm": 0.5134007334709167, + "learning_rate": 0.0003209603316898252, + "loss": 0.5141, + "step": 24359 + }, + { + "epoch": 1.3610358409922618, + "grad_norm": 0.5997005701065063, + "learning_rate": 0.0003209323173464814, + "loss": 0.4167, + "step": 24360 + }, + { + "epoch": 1.3610917115959438, + "grad_norm": 0.7501387596130371, + "learning_rate": 0.0003209043030031376, + "loss": 0.4818, + "step": 24361 + }, + { + "epoch": 1.3611475821996257, + "grad_norm": 0.35736963152885437, + "learning_rate": 0.00032087628865979387, + "loss": 0.3724, + "step": 24362 + }, + { + "epoch": 1.3612034528033075, + "grad_norm": 2.7723469734191895, + "learning_rate": 0.00032084827431645, + "loss": 0.4993, + "step": 24363 + }, + { + "epoch": 1.3612593234069894, + "grad_norm": 0.3649037480354309, + "learning_rate": 0.00032082025997310623, + "loss": 0.3982, + "step": 24364 + }, + { + "epoch": 1.3613151940106714, + "grad_norm": 0.4194002151489258, + "learning_rate": 0.00032079224562976243, + "loss": 0.4058, + "step": 24365 + }, + { + "epoch": 1.361371064614353, + "grad_norm": 0.38212907314300537, + "learning_rate": 0.00032076423128641864, + "loss": 0.3584, + "step": 24366 + }, + { + "epoch": 1.361426935218035, + "grad_norm": 0.4749254584312439, + "learning_rate": 0.00032073621694307485, + "loss": 0.3175, + "step": 24367 + }, + { + "epoch": 1.361482805821717, + "grad_norm": 0.36644601821899414, + "learning_rate": 0.00032070820259973105, + "loss": 0.3139, + "step": 24368 + }, + { + "epoch": 1.3615386764253987, + "grad_norm": 0.5241307616233826, + "learning_rate": 0.00032068018825638726, + "loss": 0.4571, + "step": 24369 + }, + { + "epoch": 1.3615945470290807, + "grad_norm": 2.402127742767334, + "learning_rate": 0.00032065217391304346, + "loss": 0.4164, + "step": 24370 + }, + { + "epoch": 1.3616504176327626, + "grad_norm": 0.5321205854415894, + "learning_rate": 0.0003206241595696997, + "loss": 0.5185, + "step": 24371 + }, + { + "epoch": 1.3617062882364444, + "grad_norm": 0.33397766947746277, + "learning_rate": 0.0003205961452263559, + "loss": 0.4048, + "step": 24372 + }, + { + "epoch": 1.3617621588401263, + "grad_norm": 0.7502289414405823, + "learning_rate": 0.00032056813088301214, + "loss": 0.3694, + "step": 24373 + }, + { + "epoch": 1.3618180294438083, + "grad_norm": 0.3493303656578064, + "learning_rate": 0.0003205401165396683, + "loss": 0.3823, + "step": 24374 + }, + { + "epoch": 1.36187390004749, + "grad_norm": 0.35359013080596924, + "learning_rate": 0.00032051210219632455, + "loss": 0.3761, + "step": 24375 + }, + { + "epoch": 1.361929770651172, + "grad_norm": 0.548119843006134, + "learning_rate": 0.0003204840878529807, + "loss": 0.3764, + "step": 24376 + }, + { + "epoch": 1.3619856412548539, + "grad_norm": 0.4145055115222931, + "learning_rate": 0.00032045607350963696, + "loss": 0.4025, + "step": 24377 + }, + { + "epoch": 1.3620415118585356, + "grad_norm": 0.3656717538833618, + "learning_rate": 0.00032042805916629316, + "loss": 0.4046, + "step": 24378 + }, + { + "epoch": 1.3620973824622176, + "grad_norm": 0.4060579836368561, + "learning_rate": 0.00032040004482294937, + "loss": 0.4132, + "step": 24379 + }, + { + "epoch": 1.3621532530658993, + "grad_norm": 1.566423773765564, + "learning_rate": 0.0003203720304796056, + "loss": 0.3618, + "step": 24380 + }, + { + "epoch": 1.3622091236695812, + "grad_norm": 0.45055902004241943, + "learning_rate": 0.0003203440161362618, + "loss": 0.4558, + "step": 24381 + }, + { + "epoch": 1.3622649942732632, + "grad_norm": 0.33534014225006104, + "learning_rate": 0.000320316001792918, + "loss": 0.3408, + "step": 24382 + }, + { + "epoch": 1.362320864876945, + "grad_norm": 0.5152701139450073, + "learning_rate": 0.0003202879874495742, + "loss": 0.4945, + "step": 24383 + }, + { + "epoch": 1.3623767354806269, + "grad_norm": 0.5772975087165833, + "learning_rate": 0.0003202599731062304, + "loss": 0.4663, + "step": 24384 + }, + { + "epoch": 1.3624326060843086, + "grad_norm": 1.2032032012939453, + "learning_rate": 0.00032023195876288655, + "loss": 0.3101, + "step": 24385 + }, + { + "epoch": 1.3624884766879906, + "grad_norm": 0.3928215205669403, + "learning_rate": 0.0003202039444195428, + "loss": 0.4459, + "step": 24386 + }, + { + "epoch": 1.3625443472916725, + "grad_norm": 0.6431586146354675, + "learning_rate": 0.000320175930076199, + "loss": 0.4649, + "step": 24387 + }, + { + "epoch": 1.3626002178953542, + "grad_norm": 0.6574409008026123, + "learning_rate": 0.0003201479157328552, + "loss": 0.4673, + "step": 24388 + }, + { + "epoch": 1.3626560884990362, + "grad_norm": 0.550195574760437, + "learning_rate": 0.00032011990138951143, + "loss": 0.3614, + "step": 24389 + }, + { + "epoch": 1.3627119591027181, + "grad_norm": 0.6298977732658386, + "learning_rate": 0.00032009188704616763, + "loss": 0.4183, + "step": 24390 + }, + { + "epoch": 1.3627678297063999, + "grad_norm": 0.4160238206386566, + "learning_rate": 0.00032006387270282384, + "loss": 0.4784, + "step": 24391 + }, + { + "epoch": 1.3628237003100818, + "grad_norm": 0.8591924905776978, + "learning_rate": 0.00032003585835948005, + "loss": 0.4955, + "step": 24392 + }, + { + "epoch": 1.3628795709137638, + "grad_norm": 0.3940196931362152, + "learning_rate": 0.00032000784401613625, + "loss": 0.3387, + "step": 24393 + }, + { + "epoch": 1.3629354415174455, + "grad_norm": 0.41842904686927795, + "learning_rate": 0.0003199798296727925, + "loss": 0.3928, + "step": 24394 + }, + { + "epoch": 1.3629913121211275, + "grad_norm": 1.0191794633865356, + "learning_rate": 0.00031995181532944866, + "loss": 0.36, + "step": 24395 + }, + { + "epoch": 1.3630471827248094, + "grad_norm": 0.4890260100364685, + "learning_rate": 0.0003199238009861049, + "loss": 0.4439, + "step": 24396 + }, + { + "epoch": 1.3631030533284911, + "grad_norm": 0.2746260166168213, + "learning_rate": 0.0003198957866427611, + "loss": 0.3043, + "step": 24397 + }, + { + "epoch": 1.363158923932173, + "grad_norm": 0.45601698756217957, + "learning_rate": 0.00031986777229941733, + "loss": 0.4562, + "step": 24398 + }, + { + "epoch": 1.363214794535855, + "grad_norm": 0.643912136554718, + "learning_rate": 0.0003198397579560735, + "loss": 0.451, + "step": 24399 + }, + { + "epoch": 1.3632706651395368, + "grad_norm": 0.3844054043292999, + "learning_rate": 0.00031981174361272975, + "loss": 0.3448, + "step": 24400 + }, + { + "epoch": 1.3633265357432187, + "grad_norm": 0.473387748003006, + "learning_rate": 0.00031978372926938595, + "loss": 0.3517, + "step": 24401 + }, + { + "epoch": 1.3633824063469007, + "grad_norm": 1.8933035135269165, + "learning_rate": 0.00031975571492604216, + "loss": 0.5027, + "step": 24402 + }, + { + "epoch": 1.3634382769505824, + "grad_norm": 0.44724252820014954, + "learning_rate": 0.00031972770058269836, + "loss": 0.3692, + "step": 24403 + }, + { + "epoch": 1.3634941475542643, + "grad_norm": 0.40734753012657166, + "learning_rate": 0.00031969968623935457, + "loss": 0.3936, + "step": 24404 + }, + { + "epoch": 1.3635500181579463, + "grad_norm": 0.5492499470710754, + "learning_rate": 0.0003196716718960108, + "loss": 0.3827, + "step": 24405 + }, + { + "epoch": 1.363605888761628, + "grad_norm": 0.5489571690559387, + "learning_rate": 0.000319643657552667, + "loss": 0.3705, + "step": 24406 + }, + { + "epoch": 1.36366175936531, + "grad_norm": 0.41693687438964844, + "learning_rate": 0.0003196156432093232, + "loss": 0.3053, + "step": 24407 + }, + { + "epoch": 1.363717629968992, + "grad_norm": 0.3955787420272827, + "learning_rate": 0.00031958762886597934, + "loss": 0.4665, + "step": 24408 + }, + { + "epoch": 1.3637735005726737, + "grad_norm": 0.4854544401168823, + "learning_rate": 0.0003195596145226356, + "loss": 0.4104, + "step": 24409 + }, + { + "epoch": 1.3638293711763556, + "grad_norm": 0.5305918455123901, + "learning_rate": 0.0003195316001792918, + "loss": 0.5696, + "step": 24410 + }, + { + "epoch": 1.3638852417800376, + "grad_norm": 0.6169124841690063, + "learning_rate": 0.000319503585835948, + "loss": 0.4752, + "step": 24411 + }, + { + "epoch": 1.3639411123837193, + "grad_norm": 0.49418267607688904, + "learning_rate": 0.0003194755714926042, + "loss": 0.4316, + "step": 24412 + }, + { + "epoch": 1.3639969829874012, + "grad_norm": 1.796499252319336, + "learning_rate": 0.0003194475571492604, + "loss": 0.3919, + "step": 24413 + }, + { + "epoch": 1.364052853591083, + "grad_norm": 0.35766321420669556, + "learning_rate": 0.00031941954280591663, + "loss": 0.4031, + "step": 24414 + }, + { + "epoch": 1.364108724194765, + "grad_norm": 5.178258419036865, + "learning_rate": 0.00031939152846257283, + "loss": 0.5225, + "step": 24415 + }, + { + "epoch": 1.3641645947984469, + "grad_norm": 0.5656289458274841, + "learning_rate": 0.00031936351411922904, + "loss": 0.3944, + "step": 24416 + }, + { + "epoch": 1.3642204654021286, + "grad_norm": 0.5849435925483704, + "learning_rate": 0.0003193354997758853, + "loss": 0.4746, + "step": 24417 + }, + { + "epoch": 1.3642763360058106, + "grad_norm": 0.5909810066223145, + "learning_rate": 0.00031930748543254145, + "loss": 0.3716, + "step": 24418 + }, + { + "epoch": 1.3643322066094923, + "grad_norm": 0.4448094069957733, + "learning_rate": 0.0003192794710891977, + "loss": 0.4955, + "step": 24419 + }, + { + "epoch": 1.3643880772131742, + "grad_norm": 0.44039857387542725, + "learning_rate": 0.00031925145674585386, + "loss": 0.4314, + "step": 24420 + }, + { + "epoch": 1.3644439478168562, + "grad_norm": 0.43824002146720886, + "learning_rate": 0.0003192234424025101, + "loss": 0.4628, + "step": 24421 + }, + { + "epoch": 1.364499818420538, + "grad_norm": 0.42289721965789795, + "learning_rate": 0.0003191954280591663, + "loss": 0.4845, + "step": 24422 + }, + { + "epoch": 1.3645556890242199, + "grad_norm": 0.46715983748435974, + "learning_rate": 0.00031916741371582253, + "loss": 0.3353, + "step": 24423 + }, + { + "epoch": 1.3646115596279018, + "grad_norm": 0.45203661918640137, + "learning_rate": 0.0003191393993724787, + "loss": 0.3876, + "step": 24424 + }, + { + "epoch": 1.3646674302315835, + "grad_norm": 0.5540289878845215, + "learning_rate": 0.00031911138502913495, + "loss": 0.4231, + "step": 24425 + }, + { + "epoch": 1.3647233008352655, + "grad_norm": 0.43327796459198, + "learning_rate": 0.00031908337068579115, + "loss": 0.4726, + "step": 24426 + }, + { + "epoch": 1.3647791714389474, + "grad_norm": 0.5657753944396973, + "learning_rate": 0.00031905535634244736, + "loss": 0.4055, + "step": 24427 + }, + { + "epoch": 1.3648350420426292, + "grad_norm": 0.5257159471511841, + "learning_rate": 0.00031902734199910356, + "loss": 0.5747, + "step": 24428 + }, + { + "epoch": 1.3648909126463111, + "grad_norm": 0.5571518540382385, + "learning_rate": 0.0003189993276557597, + "loss": 0.4734, + "step": 24429 + }, + { + "epoch": 1.364946783249993, + "grad_norm": 0.3877604603767395, + "learning_rate": 0.000318971313312416, + "loss": 0.4658, + "step": 24430 + }, + { + "epoch": 1.3650026538536748, + "grad_norm": 0.3992144465446472, + "learning_rate": 0.0003189432989690721, + "loss": 0.4045, + "step": 24431 + }, + { + "epoch": 1.3650585244573568, + "grad_norm": 0.6580184102058411, + "learning_rate": 0.0003189152846257284, + "loss": 0.4721, + "step": 24432 + }, + { + "epoch": 1.3651143950610387, + "grad_norm": 0.7114299535751343, + "learning_rate": 0.0003188872702823846, + "loss": 0.4645, + "step": 24433 + }, + { + "epoch": 1.3651702656647204, + "grad_norm": 0.5155527591705322, + "learning_rate": 0.0003188592559390408, + "loss": 0.397, + "step": 24434 + }, + { + "epoch": 1.3652261362684024, + "grad_norm": 0.40645018219947815, + "learning_rate": 0.000318831241595697, + "loss": 0.3553, + "step": 24435 + }, + { + "epoch": 1.3652820068720843, + "grad_norm": 0.3446824848651886, + "learning_rate": 0.0003188032272523532, + "loss": 0.415, + "step": 24436 + }, + { + "epoch": 1.365337877475766, + "grad_norm": 0.6060294508934021, + "learning_rate": 0.0003187752129090094, + "loss": 0.5251, + "step": 24437 + }, + { + "epoch": 1.365393748079448, + "grad_norm": 1.208666443824768, + "learning_rate": 0.0003187471985656656, + "loss": 0.4394, + "step": 24438 + }, + { + "epoch": 1.36544961868313, + "grad_norm": 0.3877502381801605, + "learning_rate": 0.00031871918422232183, + "loss": 0.3675, + "step": 24439 + }, + { + "epoch": 1.3655054892868117, + "grad_norm": 0.5541069507598877, + "learning_rate": 0.00031869116987897803, + "loss": 0.4208, + "step": 24440 + }, + { + "epoch": 1.3655613598904937, + "grad_norm": 0.7591167688369751, + "learning_rate": 0.00031866315553563424, + "loss": 0.6063, + "step": 24441 + }, + { + "epoch": 1.3656172304941756, + "grad_norm": 0.48690125346183777, + "learning_rate": 0.0003186351411922905, + "loss": 0.4215, + "step": 24442 + }, + { + "epoch": 1.3656731010978573, + "grad_norm": 0.6250282526016235, + "learning_rate": 0.00031860712684894665, + "loss": 0.3417, + "step": 24443 + }, + { + "epoch": 1.3657289717015393, + "grad_norm": 0.4475819766521454, + "learning_rate": 0.0003185791125056029, + "loss": 0.3786, + "step": 24444 + }, + { + "epoch": 1.3657848423052212, + "grad_norm": 0.4864611029624939, + "learning_rate": 0.00031855109816225906, + "loss": 0.3839, + "step": 24445 + }, + { + "epoch": 1.365840712908903, + "grad_norm": 0.4781450033187866, + "learning_rate": 0.0003185230838189153, + "loss": 0.3143, + "step": 24446 + }, + { + "epoch": 1.365896583512585, + "grad_norm": 0.4752298891544342, + "learning_rate": 0.0003184950694755715, + "loss": 0.4541, + "step": 24447 + }, + { + "epoch": 1.3659524541162666, + "grad_norm": 0.41494855284690857, + "learning_rate": 0.00031846705513222773, + "loss": 0.4107, + "step": 24448 + }, + { + "epoch": 1.3660083247199486, + "grad_norm": 0.9757683277130127, + "learning_rate": 0.00031843904078888394, + "loss": 0.63, + "step": 24449 + }, + { + "epoch": 1.3660641953236305, + "grad_norm": 0.29227393865585327, + "learning_rate": 0.0003184110264455401, + "loss": 0.3338, + "step": 24450 + }, + { + "epoch": 1.3661200659273123, + "grad_norm": 0.9545615315437317, + "learning_rate": 0.00031838301210219635, + "loss": 0.4037, + "step": 24451 + }, + { + "epoch": 1.3661759365309942, + "grad_norm": 0.5531363487243652, + "learning_rate": 0.0003183549977588525, + "loss": 0.3876, + "step": 24452 + }, + { + "epoch": 1.366231807134676, + "grad_norm": 0.6908907890319824, + "learning_rate": 0.00031832698341550876, + "loss": 0.3989, + "step": 24453 + }, + { + "epoch": 1.366287677738358, + "grad_norm": 0.6899473667144775, + "learning_rate": 0.0003182989690721649, + "loss": 0.4504, + "step": 24454 + }, + { + "epoch": 1.3663435483420399, + "grad_norm": 0.3862413763999939, + "learning_rate": 0.0003182709547288212, + "loss": 0.5491, + "step": 24455 + }, + { + "epoch": 1.3663994189457216, + "grad_norm": 0.6828871369361877, + "learning_rate": 0.0003182429403854773, + "loss": 0.4465, + "step": 24456 + }, + { + "epoch": 1.3664552895494035, + "grad_norm": 2.7458178997039795, + "learning_rate": 0.0003182149260421336, + "loss": 0.3331, + "step": 24457 + }, + { + "epoch": 1.3665111601530855, + "grad_norm": 0.4984043836593628, + "learning_rate": 0.0003181869116987898, + "loss": 0.3523, + "step": 24458 + }, + { + "epoch": 1.3665670307567672, + "grad_norm": 0.4493887424468994, + "learning_rate": 0.000318158897355446, + "loss": 0.5368, + "step": 24459 + }, + { + "epoch": 1.3666229013604492, + "grad_norm": 0.35766100883483887, + "learning_rate": 0.0003181308830121022, + "loss": 0.3709, + "step": 24460 + }, + { + "epoch": 1.3666787719641311, + "grad_norm": 0.36706987023353577, + "learning_rate": 0.0003181028686687584, + "loss": 0.3365, + "step": 24461 + }, + { + "epoch": 1.3667346425678129, + "grad_norm": 0.3625475764274597, + "learning_rate": 0.0003180748543254146, + "loss": 0.4211, + "step": 24462 + }, + { + "epoch": 1.3667905131714948, + "grad_norm": 0.8112778067588806, + "learning_rate": 0.0003180468399820708, + "loss": 0.5284, + "step": 24463 + }, + { + "epoch": 1.3668463837751768, + "grad_norm": 1.9017622470855713, + "learning_rate": 0.00031801882563872703, + "loss": 0.4106, + "step": 24464 + }, + { + "epoch": 1.3669022543788585, + "grad_norm": 0.4935758709907532, + "learning_rate": 0.0003179908112953833, + "loss": 0.4765, + "step": 24465 + }, + { + "epoch": 1.3669581249825404, + "grad_norm": 0.35609591007232666, + "learning_rate": 0.00031796279695203944, + "loss": 0.4028, + "step": 24466 + }, + { + "epoch": 1.3670139955862224, + "grad_norm": 0.7822611331939697, + "learning_rate": 0.0003179347826086957, + "loss": 0.3196, + "step": 24467 + }, + { + "epoch": 1.3670698661899041, + "grad_norm": 0.5921357274055481, + "learning_rate": 0.00031790676826535185, + "loss": 0.4772, + "step": 24468 + }, + { + "epoch": 1.367125736793586, + "grad_norm": 0.3597450256347656, + "learning_rate": 0.0003178787539220081, + "loss": 0.3248, + "step": 24469 + }, + { + "epoch": 1.367181607397268, + "grad_norm": 0.63299560546875, + "learning_rate": 0.00031785073957866426, + "loss": 0.4704, + "step": 24470 + }, + { + "epoch": 1.3672374780009497, + "grad_norm": 0.5246701836585999, + "learning_rate": 0.00031782272523532047, + "loss": 0.5471, + "step": 24471 + }, + { + "epoch": 1.3672933486046317, + "grad_norm": 0.4346126616001129, + "learning_rate": 0.00031779471089197673, + "loss": 0.3369, + "step": 24472 + }, + { + "epoch": 1.3673492192083136, + "grad_norm": 0.38988882303237915, + "learning_rate": 0.0003177666965486329, + "loss": 0.4367, + "step": 24473 + }, + { + "epoch": 1.3674050898119954, + "grad_norm": 0.36178451776504517, + "learning_rate": 0.00031773868220528914, + "loss": 0.3669, + "step": 24474 + }, + { + "epoch": 1.3674609604156773, + "grad_norm": 2.0618908405303955, + "learning_rate": 0.0003177106678619453, + "loss": 0.4257, + "step": 24475 + }, + { + "epoch": 1.3675168310193593, + "grad_norm": 0.32940733432769775, + "learning_rate": 0.00031768265351860155, + "loss": 0.3122, + "step": 24476 + }, + { + "epoch": 1.367572701623041, + "grad_norm": 0.5555112957954407, + "learning_rate": 0.0003176546391752577, + "loss": 0.3059, + "step": 24477 + }, + { + "epoch": 1.367628572226723, + "grad_norm": 0.4774741232395172, + "learning_rate": 0.00031762662483191396, + "loss": 0.4553, + "step": 24478 + }, + { + "epoch": 1.367684442830405, + "grad_norm": 0.7613052129745483, + "learning_rate": 0.0003175986104885701, + "loss": 0.4089, + "step": 24479 + }, + { + "epoch": 1.3677403134340866, + "grad_norm": 1.7943933010101318, + "learning_rate": 0.0003175705961452264, + "loss": 0.4241, + "step": 24480 + }, + { + "epoch": 1.3677961840377686, + "grad_norm": 3.3346071243286133, + "learning_rate": 0.0003175425818018826, + "loss": 0.5636, + "step": 24481 + }, + { + "epoch": 1.3678520546414503, + "grad_norm": 0.47424086928367615, + "learning_rate": 0.0003175145674585388, + "loss": 0.3975, + "step": 24482 + }, + { + "epoch": 1.3679079252451323, + "grad_norm": 0.903059720993042, + "learning_rate": 0.000317486553115195, + "loss": 0.3841, + "step": 24483 + }, + { + "epoch": 1.3679637958488142, + "grad_norm": 2.0153183937072754, + "learning_rate": 0.0003174585387718512, + "loss": 0.4174, + "step": 24484 + }, + { + "epoch": 1.368019666452496, + "grad_norm": 0.30785036087036133, + "learning_rate": 0.0003174305244285074, + "loss": 0.3455, + "step": 24485 + }, + { + "epoch": 1.368075537056178, + "grad_norm": 0.34346944093704224, + "learning_rate": 0.0003174025100851636, + "loss": 0.3883, + "step": 24486 + }, + { + "epoch": 1.3681314076598596, + "grad_norm": 3.7171478271484375, + "learning_rate": 0.0003173744957418198, + "loss": 0.4244, + "step": 24487 + }, + { + "epoch": 1.3681872782635416, + "grad_norm": 0.3544069230556488, + "learning_rate": 0.0003173464813984761, + "loss": 0.3553, + "step": 24488 + }, + { + "epoch": 1.3682431488672235, + "grad_norm": 0.6728845834732056, + "learning_rate": 0.0003173184670551322, + "loss": 0.4799, + "step": 24489 + }, + { + "epoch": 1.3682990194709053, + "grad_norm": 0.34190934896469116, + "learning_rate": 0.0003172904527117885, + "loss": 0.397, + "step": 24490 + }, + { + "epoch": 1.3683548900745872, + "grad_norm": 2.084429979324341, + "learning_rate": 0.00031726243836844464, + "loss": 0.4437, + "step": 24491 + }, + { + "epoch": 1.3684107606782692, + "grad_norm": 3.2836685180664062, + "learning_rate": 0.00031723442402510084, + "loss": 0.3814, + "step": 24492 + }, + { + "epoch": 1.368466631281951, + "grad_norm": 4.27080774307251, + "learning_rate": 0.00031720640968175705, + "loss": 0.4902, + "step": 24493 + }, + { + "epoch": 1.3685225018856328, + "grad_norm": 0.6161115765571594, + "learning_rate": 0.00031717839533841326, + "loss": 0.4431, + "step": 24494 + }, + { + "epoch": 1.3685783724893148, + "grad_norm": 0.32110321521759033, + "learning_rate": 0.00031715038099506946, + "loss": 0.3564, + "step": 24495 + }, + { + "epoch": 1.3686342430929965, + "grad_norm": 0.5173138976097107, + "learning_rate": 0.00031712236665172567, + "loss": 0.5283, + "step": 24496 + }, + { + "epoch": 1.3686901136966785, + "grad_norm": 0.5653108358383179, + "learning_rate": 0.00031709435230838193, + "loss": 0.4433, + "step": 24497 + }, + { + "epoch": 1.3687459843003604, + "grad_norm": 0.5507844090461731, + "learning_rate": 0.0003170663379650381, + "loss": 0.4625, + "step": 24498 + }, + { + "epoch": 1.3688018549040422, + "grad_norm": 0.5151591300964355, + "learning_rate": 0.00031703832362169434, + "loss": 0.4501, + "step": 24499 + }, + { + "epoch": 1.368857725507724, + "grad_norm": 2.6444830894470215, + "learning_rate": 0.0003170103092783505, + "loss": 0.4479, + "step": 24500 + }, + { + "epoch": 1.368857725507724, + "eval_cer": 0.08642924945173644, + "eval_loss": 0.32431721687316895, + "eval_runtime": 56.0308, + "eval_samples_per_second": 80.991, + "eval_steps_per_second": 5.069, + "eval_wer": 0.34454344850702495, + "step": 24500 + }, + { + "epoch": 1.368913596111406, + "grad_norm": 0.42986148595809937, + "learning_rate": 0.00031698229493500675, + "loss": 0.4887, + "step": 24501 + }, + { + "epoch": 1.3689694667150878, + "grad_norm": 0.4669835865497589, + "learning_rate": 0.0003169542805916629, + "loss": 0.4086, + "step": 24502 + }, + { + "epoch": 1.3690253373187697, + "grad_norm": 0.38927164673805237, + "learning_rate": 0.00031692626624831916, + "loss": 0.4898, + "step": 24503 + }, + { + "epoch": 1.3690812079224517, + "grad_norm": 0.5460061430931091, + "learning_rate": 0.00031689825190497537, + "loss": 0.43, + "step": 24504 + }, + { + "epoch": 1.3691370785261334, + "grad_norm": 0.778160810470581, + "learning_rate": 0.0003168702375616316, + "loss": 0.6369, + "step": 24505 + }, + { + "epoch": 1.3691929491298154, + "grad_norm": 0.4796617925167084, + "learning_rate": 0.0003168422232182878, + "loss": 0.4112, + "step": 24506 + }, + { + "epoch": 1.3692488197334973, + "grad_norm": 0.47398877143859863, + "learning_rate": 0.000316814208874944, + "loss": 0.4974, + "step": 24507 + }, + { + "epoch": 1.369304690337179, + "grad_norm": 0.42609700560569763, + "learning_rate": 0.0003167861945316002, + "loss": 0.3084, + "step": 24508 + }, + { + "epoch": 1.369360560940861, + "grad_norm": 0.37535524368286133, + "learning_rate": 0.0003167581801882564, + "loss": 0.4164, + "step": 24509 + }, + { + "epoch": 1.369416431544543, + "grad_norm": 0.6311372518539429, + "learning_rate": 0.0003167301658449126, + "loss": 0.4327, + "step": 24510 + }, + { + "epoch": 1.3694723021482247, + "grad_norm": 0.4855610728263855, + "learning_rate": 0.0003167021515015688, + "loss": 0.5025, + "step": 24511 + }, + { + "epoch": 1.3695281727519066, + "grad_norm": 0.3720603883266449, + "learning_rate": 0.000316674137158225, + "loss": 0.382, + "step": 24512 + }, + { + "epoch": 1.3695840433555886, + "grad_norm": 0.8351364731788635, + "learning_rate": 0.0003166461228148812, + "loss": 0.4105, + "step": 24513 + }, + { + "epoch": 1.3696399139592703, + "grad_norm": 0.49884194135665894, + "learning_rate": 0.0003166181084715374, + "loss": 0.4532, + "step": 24514 + }, + { + "epoch": 1.3696957845629523, + "grad_norm": 0.27206429839134216, + "learning_rate": 0.00031659009412819363, + "loss": 0.3569, + "step": 24515 + }, + { + "epoch": 1.369751655166634, + "grad_norm": 0.7734537124633789, + "learning_rate": 0.00031656207978484984, + "loss": 0.3471, + "step": 24516 + }, + { + "epoch": 1.369807525770316, + "grad_norm": 1.8109050989151, + "learning_rate": 0.00031653406544150604, + "loss": 0.3834, + "step": 24517 + }, + { + "epoch": 1.369863396373998, + "grad_norm": 0.8112854957580566, + "learning_rate": 0.00031650605109816225, + "loss": 0.4292, + "step": 24518 + }, + { + "epoch": 1.3699192669776796, + "grad_norm": 0.5000078678131104, + "learning_rate": 0.00031647803675481846, + "loss": 0.4374, + "step": 24519 + }, + { + "epoch": 1.3699751375813616, + "grad_norm": 0.45532724261283875, + "learning_rate": 0.0003164500224114747, + "loss": 0.4092, + "step": 24520 + }, + { + "epoch": 1.3700310081850433, + "grad_norm": 0.4365138113498688, + "learning_rate": 0.00031642200806813087, + "loss": 0.3933, + "step": 24521 + }, + { + "epoch": 1.3700868787887253, + "grad_norm": 0.583670973777771, + "learning_rate": 0.00031639399372478713, + "loss": 0.5493, + "step": 24522 + }, + { + "epoch": 1.3701427493924072, + "grad_norm": 0.6384440660476685, + "learning_rate": 0.0003163659793814433, + "loss": 0.4263, + "step": 24523 + }, + { + "epoch": 1.370198619996089, + "grad_norm": 0.3799341022968292, + "learning_rate": 0.00031633796503809954, + "loss": 0.3826, + "step": 24524 + }, + { + "epoch": 1.370254490599771, + "grad_norm": 0.4388483464717865, + "learning_rate": 0.0003163099506947557, + "loss": 0.4866, + "step": 24525 + }, + { + "epoch": 1.3703103612034528, + "grad_norm": 0.47827619314193726, + "learning_rate": 0.00031628193635141195, + "loss": 0.502, + "step": 24526 + }, + { + "epoch": 1.3703662318071346, + "grad_norm": 0.43231451511383057, + "learning_rate": 0.0003162539220080681, + "loss": 0.4382, + "step": 24527 + }, + { + "epoch": 1.3704221024108165, + "grad_norm": 0.40887650847435, + "learning_rate": 0.00031622590766472436, + "loss": 0.4548, + "step": 24528 + }, + { + "epoch": 1.3704779730144985, + "grad_norm": 0.5036583542823792, + "learning_rate": 0.00031619789332138057, + "loss": 0.5724, + "step": 24529 + }, + { + "epoch": 1.3705338436181802, + "grad_norm": 1.9565993547439575, + "learning_rate": 0.0003161698789780368, + "loss": 0.4378, + "step": 24530 + }, + { + "epoch": 1.3705897142218622, + "grad_norm": 0.33863580226898193, + "learning_rate": 0.000316141864634693, + "loss": 0.409, + "step": 24531 + }, + { + "epoch": 1.370645584825544, + "grad_norm": 0.42710286378860474, + "learning_rate": 0.0003161138502913492, + "loss": 0.3959, + "step": 24532 + }, + { + "epoch": 1.3707014554292258, + "grad_norm": 0.5485436320304871, + "learning_rate": 0.0003160858359480054, + "loss": 0.4551, + "step": 24533 + }, + { + "epoch": 1.3707573260329078, + "grad_norm": 0.5192214846611023, + "learning_rate": 0.00031605782160466154, + "loss": 0.3351, + "step": 24534 + }, + { + "epoch": 1.3708131966365897, + "grad_norm": 0.5476623773574829, + "learning_rate": 0.0003160298072613178, + "loss": 0.5557, + "step": 24535 + }, + { + "epoch": 1.3708690672402715, + "grad_norm": 1.3046131134033203, + "learning_rate": 0.000316001792917974, + "loss": 0.5649, + "step": 24536 + }, + { + "epoch": 1.3709249378439534, + "grad_norm": 0.5167523622512817, + "learning_rate": 0.0003159737785746302, + "loss": 0.3481, + "step": 24537 + }, + { + "epoch": 1.3709808084476354, + "grad_norm": 0.5409326553344727, + "learning_rate": 0.0003159457642312864, + "loss": 0.3993, + "step": 24538 + }, + { + "epoch": 1.371036679051317, + "grad_norm": 0.33384689688682556, + "learning_rate": 0.0003159177498879426, + "loss": 0.345, + "step": 24539 + }, + { + "epoch": 1.371092549654999, + "grad_norm": 3.7947487831115723, + "learning_rate": 0.00031588973554459883, + "loss": 0.5126, + "step": 24540 + }, + { + "epoch": 1.371148420258681, + "grad_norm": 0.5459240674972534, + "learning_rate": 0.00031586172120125504, + "loss": 0.5296, + "step": 24541 + }, + { + "epoch": 1.3712042908623627, + "grad_norm": 0.5933355689048767, + "learning_rate": 0.00031583370685791124, + "loss": 0.437, + "step": 24542 + }, + { + "epoch": 1.3712601614660447, + "grad_norm": 0.37017127871513367, + "learning_rate": 0.0003158056925145675, + "loss": 0.4016, + "step": 24543 + }, + { + "epoch": 1.3713160320697266, + "grad_norm": 1.823818325996399, + "learning_rate": 0.00031577767817122366, + "loss": 0.454, + "step": 24544 + }, + { + "epoch": 1.3713719026734084, + "grad_norm": 3.8623602390289307, + "learning_rate": 0.0003157496638278799, + "loss": 0.3546, + "step": 24545 + }, + { + "epoch": 1.3714277732770903, + "grad_norm": 0.32659998536109924, + "learning_rate": 0.00031572164948453607, + "loss": 0.3907, + "step": 24546 + }, + { + "epoch": 1.3714836438807723, + "grad_norm": 0.5224636793136597, + "learning_rate": 0.0003156936351411923, + "loss": 0.409, + "step": 24547 + }, + { + "epoch": 1.371539514484454, + "grad_norm": 0.39653918147087097, + "learning_rate": 0.0003156656207978485, + "loss": 0.3887, + "step": 24548 + }, + { + "epoch": 1.371595385088136, + "grad_norm": 1.1151378154754639, + "learning_rate": 0.00031563760645450474, + "loss": 0.4794, + "step": 24549 + }, + { + "epoch": 1.3716512556918177, + "grad_norm": 0.41958460211753845, + "learning_rate": 0.0003156095921111609, + "loss": 0.5167, + "step": 24550 + }, + { + "epoch": 1.3717071262954996, + "grad_norm": 2.6937193870544434, + "learning_rate": 0.00031558157776781715, + "loss": 0.5326, + "step": 24551 + }, + { + "epoch": 1.3717629968991816, + "grad_norm": 3.233395576477051, + "learning_rate": 0.00031555356342447336, + "loss": 0.3466, + "step": 24552 + }, + { + "epoch": 1.3718188675028633, + "grad_norm": 0.3562087118625641, + "learning_rate": 0.00031552554908112956, + "loss": 0.3859, + "step": 24553 + }, + { + "epoch": 1.3718747381065453, + "grad_norm": 0.39315155148506165, + "learning_rate": 0.00031549753473778577, + "loss": 0.3449, + "step": 24554 + }, + { + "epoch": 1.371930608710227, + "grad_norm": 0.430144339799881, + "learning_rate": 0.0003154695203944419, + "loss": 0.355, + "step": 24555 + }, + { + "epoch": 1.371986479313909, + "grad_norm": 0.7686746716499329, + "learning_rate": 0.0003154415060510982, + "loss": 0.4503, + "step": 24556 + }, + { + "epoch": 1.3720423499175909, + "grad_norm": 0.5789282917976379, + "learning_rate": 0.00031541349170775433, + "loss": 0.3817, + "step": 24557 + }, + { + "epoch": 1.3720982205212726, + "grad_norm": 0.347003698348999, + "learning_rate": 0.0003153854773644106, + "loss": 0.3472, + "step": 24558 + }, + { + "epoch": 1.3721540911249546, + "grad_norm": 0.3490416407585144, + "learning_rate": 0.0003153574630210668, + "loss": 0.4203, + "step": 24559 + }, + { + "epoch": 1.3722099617286365, + "grad_norm": 0.3226853013038635, + "learning_rate": 0.000315329448677723, + "loss": 0.2817, + "step": 24560 + }, + { + "epoch": 1.3722658323323182, + "grad_norm": 0.4693419337272644, + "learning_rate": 0.0003153014343343792, + "loss": 0.4125, + "step": 24561 + }, + { + "epoch": 1.3723217029360002, + "grad_norm": 0.668691873550415, + "learning_rate": 0.0003152734199910354, + "loss": 0.3567, + "step": 24562 + }, + { + "epoch": 1.3723775735396821, + "grad_norm": 1.6839121580123901, + "learning_rate": 0.0003152454056476916, + "loss": 0.442, + "step": 24563 + }, + { + "epoch": 1.3724334441433639, + "grad_norm": 0.3701396584510803, + "learning_rate": 0.0003152173913043478, + "loss": 0.3329, + "step": 24564 + }, + { + "epoch": 1.3724893147470458, + "grad_norm": 0.37789681553840637, + "learning_rate": 0.00031518937696100403, + "loss": 0.3491, + "step": 24565 + }, + { + "epoch": 1.3725451853507278, + "grad_norm": 0.3406659960746765, + "learning_rate": 0.00031516136261766024, + "loss": 0.3867, + "step": 24566 + }, + { + "epoch": 1.3726010559544095, + "grad_norm": 0.39512377977371216, + "learning_rate": 0.00031513334827431644, + "loss": 0.3053, + "step": 24567 + }, + { + "epoch": 1.3726569265580915, + "grad_norm": 0.4900803864002228, + "learning_rate": 0.0003151053339309727, + "loss": 0.5108, + "step": 24568 + }, + { + "epoch": 1.3727127971617734, + "grad_norm": 0.5600206851959229, + "learning_rate": 0.00031507731958762885, + "loss": 0.4388, + "step": 24569 + }, + { + "epoch": 1.3727686677654551, + "grad_norm": 0.4562597870826721, + "learning_rate": 0.0003150493052442851, + "loss": 0.3146, + "step": 24570 + }, + { + "epoch": 1.372824538369137, + "grad_norm": 0.5856838226318359, + "learning_rate": 0.00031502129090094127, + "loss": 0.312, + "step": 24571 + }, + { + "epoch": 1.372880408972819, + "grad_norm": 0.7920502424240112, + "learning_rate": 0.0003149932765575975, + "loss": 0.383, + "step": 24572 + }, + { + "epoch": 1.3729362795765008, + "grad_norm": 0.5450758337974548, + "learning_rate": 0.0003149652622142537, + "loss": 0.3715, + "step": 24573 + }, + { + "epoch": 1.3729921501801827, + "grad_norm": 0.3974796533584595, + "learning_rate": 0.00031493724787090994, + "loss": 0.4859, + "step": 24574 + }, + { + "epoch": 1.3730480207838647, + "grad_norm": 0.4175736606121063, + "learning_rate": 0.00031490923352756614, + "loss": 0.4298, + "step": 24575 + }, + { + "epoch": 1.3731038913875464, + "grad_norm": 0.4208797216415405, + "learning_rate": 0.00031488121918422235, + "loss": 0.4206, + "step": 24576 + }, + { + "epoch": 1.3731597619912284, + "grad_norm": 0.35483700037002563, + "learning_rate": 0.00031485320484087856, + "loss": 0.3654, + "step": 24577 + }, + { + "epoch": 1.3732156325949103, + "grad_norm": 1.0821683406829834, + "learning_rate": 0.0003148251904975347, + "loss": 0.4296, + "step": 24578 + }, + { + "epoch": 1.373271503198592, + "grad_norm": 0.48379945755004883, + "learning_rate": 0.00031479717615419097, + "loss": 0.2852, + "step": 24579 + }, + { + "epoch": 1.373327373802274, + "grad_norm": 0.4581889808177948, + "learning_rate": 0.0003147691618108471, + "loss": 0.5113, + "step": 24580 + }, + { + "epoch": 1.373383244405956, + "grad_norm": 0.44941025972366333, + "learning_rate": 0.0003147411474675034, + "loss": 0.5156, + "step": 24581 + }, + { + "epoch": 1.3734391150096377, + "grad_norm": 0.9124999046325684, + "learning_rate": 0.00031471313312415953, + "loss": 0.5363, + "step": 24582 + }, + { + "epoch": 1.3734949856133196, + "grad_norm": 0.4239540696144104, + "learning_rate": 0.0003146851187808158, + "loss": 0.4535, + "step": 24583 + }, + { + "epoch": 1.3735508562170013, + "grad_norm": 0.37863144278526306, + "learning_rate": 0.000314657104437472, + "loss": 0.408, + "step": 24584 + }, + { + "epoch": 1.3736067268206833, + "grad_norm": 0.40794637799263, + "learning_rate": 0.0003146290900941282, + "loss": 0.4929, + "step": 24585 + }, + { + "epoch": 1.3736625974243653, + "grad_norm": 0.4017622172832489, + "learning_rate": 0.0003146010757507844, + "loss": 0.3836, + "step": 24586 + }, + { + "epoch": 1.373718468028047, + "grad_norm": 0.4469187259674072, + "learning_rate": 0.0003145730614074406, + "loss": 0.455, + "step": 24587 + }, + { + "epoch": 1.373774338631729, + "grad_norm": 1.2451673746109009, + "learning_rate": 0.0003145450470640968, + "loss": 0.3929, + "step": 24588 + }, + { + "epoch": 1.3738302092354107, + "grad_norm": 0.4340880513191223, + "learning_rate": 0.000314517032720753, + "loss": 0.5305, + "step": 24589 + }, + { + "epoch": 1.3738860798390926, + "grad_norm": 1.0066745281219482, + "learning_rate": 0.00031448901837740923, + "loss": 0.3907, + "step": 24590 + }, + { + "epoch": 1.3739419504427746, + "grad_norm": 0.7755605578422546, + "learning_rate": 0.0003144610040340655, + "loss": 0.3674, + "step": 24591 + }, + { + "epoch": 1.3739978210464563, + "grad_norm": 0.44511958956718445, + "learning_rate": 0.00031443298969072164, + "loss": 0.3876, + "step": 24592 + }, + { + "epoch": 1.3740536916501382, + "grad_norm": 0.45490917563438416, + "learning_rate": 0.0003144049753473779, + "loss": 0.4009, + "step": 24593 + }, + { + "epoch": 1.3741095622538202, + "grad_norm": 0.45113837718963623, + "learning_rate": 0.00031437696100403405, + "loss": 0.3643, + "step": 24594 + }, + { + "epoch": 1.374165432857502, + "grad_norm": 0.35423731803894043, + "learning_rate": 0.0003143489466606903, + "loss": 0.5073, + "step": 24595 + }, + { + "epoch": 1.3742213034611839, + "grad_norm": 0.4773174822330475, + "learning_rate": 0.00031432093231734647, + "loss": 0.4409, + "step": 24596 + }, + { + "epoch": 1.3742771740648658, + "grad_norm": 0.42499926686286926, + "learning_rate": 0.0003142929179740027, + "loss": 0.4749, + "step": 24597 + }, + { + "epoch": 1.3743330446685476, + "grad_norm": 0.6116361021995544, + "learning_rate": 0.00031426490363065893, + "loss": 0.4615, + "step": 24598 + }, + { + "epoch": 1.3743889152722295, + "grad_norm": 0.9569025635719299, + "learning_rate": 0.0003142368892873151, + "loss": 0.312, + "step": 24599 + }, + { + "epoch": 1.3744447858759115, + "grad_norm": 1.9866101741790771, + "learning_rate": 0.00031420887494397134, + "loss": 0.4365, + "step": 24600 + }, + { + "epoch": 1.3745006564795932, + "grad_norm": 0.8180204629898071, + "learning_rate": 0.0003141808606006275, + "loss": 0.4339, + "step": 24601 + }, + { + "epoch": 1.3745565270832751, + "grad_norm": 1.3644014596939087, + "learning_rate": 0.00031415284625728375, + "loss": 0.5744, + "step": 24602 + }, + { + "epoch": 1.374612397686957, + "grad_norm": 0.4826418161392212, + "learning_rate": 0.0003141248319139399, + "loss": 0.4019, + "step": 24603 + }, + { + "epoch": 1.3746682682906388, + "grad_norm": 0.5031048059463501, + "learning_rate": 0.00031409681757059617, + "loss": 0.4729, + "step": 24604 + }, + { + "epoch": 1.3747241388943208, + "grad_norm": 0.3608101010322571, + "learning_rate": 0.0003140688032272523, + "loss": 0.4944, + "step": 24605 + }, + { + "epoch": 1.3747800094980027, + "grad_norm": 0.420693963766098, + "learning_rate": 0.0003140407888839086, + "loss": 0.4025, + "step": 24606 + }, + { + "epoch": 1.3748358801016844, + "grad_norm": 0.29743456840515137, + "learning_rate": 0.0003140127745405648, + "loss": 0.373, + "step": 24607 + }, + { + "epoch": 1.3748917507053664, + "grad_norm": 0.5577669739723206, + "learning_rate": 0.000313984760197221, + "loss": 0.4153, + "step": 24608 + }, + { + "epoch": 1.3749476213090484, + "grad_norm": 0.5026376843452454, + "learning_rate": 0.0003139567458538772, + "loss": 0.4006, + "step": 24609 + }, + { + "epoch": 1.37500349191273, + "grad_norm": 0.4107271134853363, + "learning_rate": 0.0003139287315105334, + "loss": 0.3895, + "step": 24610 + }, + { + "epoch": 1.375059362516412, + "grad_norm": 0.38516178727149963, + "learning_rate": 0.0003139007171671896, + "loss": 0.4617, + "step": 24611 + }, + { + "epoch": 1.375115233120094, + "grad_norm": 0.4029679298400879, + "learning_rate": 0.0003138727028238458, + "loss": 0.44, + "step": 24612 + }, + { + "epoch": 1.3751711037237757, + "grad_norm": 0.40646740794181824, + "learning_rate": 0.000313844688480502, + "loss": 0.484, + "step": 24613 + }, + { + "epoch": 1.3752269743274577, + "grad_norm": 0.45417603850364685, + "learning_rate": 0.0003138166741371583, + "loss": 0.516, + "step": 24614 + }, + { + "epoch": 1.3752828449311396, + "grad_norm": 1.8941469192504883, + "learning_rate": 0.00031378865979381443, + "loss": 0.5028, + "step": 24615 + }, + { + "epoch": 1.3753387155348213, + "grad_norm": 0.4492465853691101, + "learning_rate": 0.0003137606454504707, + "loss": 0.4045, + "step": 24616 + }, + { + "epoch": 1.3753945861385033, + "grad_norm": 0.5950848460197449, + "learning_rate": 0.00031373263110712684, + "loss": 0.7257, + "step": 24617 + }, + { + "epoch": 1.375450456742185, + "grad_norm": 0.8576170802116394, + "learning_rate": 0.0003137046167637831, + "loss": 0.5458, + "step": 24618 + }, + { + "epoch": 1.375506327345867, + "grad_norm": 0.9594679474830627, + "learning_rate": 0.00031367660242043925, + "loss": 0.4207, + "step": 24619 + }, + { + "epoch": 1.375562197949549, + "grad_norm": 0.4792385995388031, + "learning_rate": 0.00031364858807709546, + "loss": 0.4573, + "step": 24620 + }, + { + "epoch": 1.3756180685532307, + "grad_norm": 0.5310460329055786, + "learning_rate": 0.00031362057373375167, + "loss": 0.4339, + "step": 24621 + }, + { + "epoch": 1.3756739391569126, + "grad_norm": 0.6589347124099731, + "learning_rate": 0.00031359255939040787, + "loss": 0.4137, + "step": 24622 + }, + { + "epoch": 1.3757298097605943, + "grad_norm": 0.5711861252784729, + "learning_rate": 0.00031356454504706413, + "loss": 0.3879, + "step": 24623 + }, + { + "epoch": 1.3757856803642763, + "grad_norm": 0.6465849280357361, + "learning_rate": 0.0003135365307037203, + "loss": 0.5222, + "step": 24624 + }, + { + "epoch": 1.3758415509679582, + "grad_norm": 0.6434741020202637, + "learning_rate": 0.00031350851636037654, + "loss": 0.386, + "step": 24625 + }, + { + "epoch": 1.37589742157164, + "grad_norm": 0.4712800085544586, + "learning_rate": 0.0003134805020170327, + "loss": 0.403, + "step": 24626 + }, + { + "epoch": 1.375953292175322, + "grad_norm": 0.5076837539672852, + "learning_rate": 0.00031345248767368895, + "loss": 0.4788, + "step": 24627 + }, + { + "epoch": 1.3760091627790039, + "grad_norm": 0.6015979647636414, + "learning_rate": 0.0003134244733303451, + "loss": 0.3854, + "step": 24628 + }, + { + "epoch": 1.3760650333826856, + "grad_norm": 0.5838484764099121, + "learning_rate": 0.00031339645898700137, + "loss": 0.4476, + "step": 24629 + }, + { + "epoch": 1.3761209039863675, + "grad_norm": 0.4089365601539612, + "learning_rate": 0.00031336844464365757, + "loss": 0.4143, + "step": 24630 + }, + { + "epoch": 1.3761767745900495, + "grad_norm": 0.5153072476387024, + "learning_rate": 0.0003133404303003138, + "loss": 0.4045, + "step": 24631 + }, + { + "epoch": 1.3762326451937312, + "grad_norm": 0.42766812443733215, + "learning_rate": 0.00031331241595697, + "loss": 0.3433, + "step": 24632 + }, + { + "epoch": 1.3762885157974132, + "grad_norm": 0.4806351661682129, + "learning_rate": 0.0003132844016136262, + "loss": 0.4923, + "step": 24633 + }, + { + "epoch": 1.3763443864010951, + "grad_norm": 0.4829728305339813, + "learning_rate": 0.0003132563872702824, + "loss": 0.4626, + "step": 24634 + }, + { + "epoch": 1.3764002570047769, + "grad_norm": 0.48325103521347046, + "learning_rate": 0.0003132283729269386, + "loss": 0.4864, + "step": 24635 + }, + { + "epoch": 1.3764561276084588, + "grad_norm": 0.7419774532318115, + "learning_rate": 0.0003132003585835948, + "loss": 0.3311, + "step": 24636 + }, + { + "epoch": 1.3765119982121408, + "grad_norm": 0.4290074110031128, + "learning_rate": 0.000313172344240251, + "loss": 0.42, + "step": 24637 + }, + { + "epoch": 1.3765678688158225, + "grad_norm": 1.4378046989440918, + "learning_rate": 0.0003131443298969072, + "loss": 0.6377, + "step": 24638 + }, + { + "epoch": 1.3766237394195044, + "grad_norm": 0.4767231047153473, + "learning_rate": 0.0003131163155535635, + "loss": 0.42, + "step": 24639 + }, + { + "epoch": 1.3766796100231864, + "grad_norm": 0.5699975490570068, + "learning_rate": 0.00031308830121021963, + "loss": 0.5685, + "step": 24640 + }, + { + "epoch": 1.3767354806268681, + "grad_norm": 0.391075074672699, + "learning_rate": 0.00031306028686687584, + "loss": 0.5145, + "step": 24641 + }, + { + "epoch": 1.37679135123055, + "grad_norm": 0.5889468789100647, + "learning_rate": 0.00031303227252353204, + "loss": 0.6113, + "step": 24642 + }, + { + "epoch": 1.376847221834232, + "grad_norm": 0.38755473494529724, + "learning_rate": 0.00031300425818018825, + "loss": 0.4083, + "step": 24643 + }, + { + "epoch": 1.3769030924379138, + "grad_norm": 2.820420742034912, + "learning_rate": 0.00031297624383684445, + "loss": 0.3764, + "step": 24644 + }, + { + "epoch": 1.3769589630415957, + "grad_norm": 0.8960794806480408, + "learning_rate": 0.00031294822949350066, + "loss": 0.4018, + "step": 24645 + }, + { + "epoch": 1.3770148336452777, + "grad_norm": 0.49177366495132446, + "learning_rate": 0.0003129202151501569, + "loss": 0.5385, + "step": 24646 + }, + { + "epoch": 1.3770707042489594, + "grad_norm": 0.3271487057209015, + "learning_rate": 0.00031289220080681307, + "loss": 0.4843, + "step": 24647 + }, + { + "epoch": 1.3771265748526413, + "grad_norm": 0.34570586681365967, + "learning_rate": 0.00031286418646346933, + "loss": 0.3632, + "step": 24648 + }, + { + "epoch": 1.3771824454563233, + "grad_norm": 0.379562109708786, + "learning_rate": 0.0003128361721201255, + "loss": 0.3958, + "step": 24649 + }, + { + "epoch": 1.377238316060005, + "grad_norm": 3.870682954788208, + "learning_rate": 0.00031280815777678174, + "loss": 0.3791, + "step": 24650 + }, + { + "epoch": 1.377294186663687, + "grad_norm": 0.5895723104476929, + "learning_rate": 0.0003127801434334379, + "loss": 0.4786, + "step": 24651 + }, + { + "epoch": 1.3773500572673687, + "grad_norm": 0.3996473252773285, + "learning_rate": 0.00031275212909009415, + "loss": 0.437, + "step": 24652 + }, + { + "epoch": 1.3774059278710507, + "grad_norm": 0.48510080575942993, + "learning_rate": 0.0003127241147467503, + "loss": 0.2969, + "step": 24653 + }, + { + "epoch": 1.3774617984747326, + "grad_norm": 0.4460342526435852, + "learning_rate": 0.00031269610040340657, + "loss": 0.5024, + "step": 24654 + }, + { + "epoch": 1.3775176690784143, + "grad_norm": 0.5402969717979431, + "learning_rate": 0.00031266808606006277, + "loss": 0.5264, + "step": 24655 + }, + { + "epoch": 1.3775735396820963, + "grad_norm": 0.34899044036865234, + "learning_rate": 0.000312640071716719, + "loss": 0.4226, + "step": 24656 + }, + { + "epoch": 1.377629410285778, + "grad_norm": 2.161604166030884, + "learning_rate": 0.0003126120573733752, + "loss": 0.4102, + "step": 24657 + }, + { + "epoch": 1.37768528088946, + "grad_norm": 0.3708206117153168, + "learning_rate": 0.0003125840430300314, + "loss": 0.4469, + "step": 24658 + }, + { + "epoch": 1.377741151493142, + "grad_norm": 0.38672158122062683, + "learning_rate": 0.0003125560286866876, + "loss": 0.4617, + "step": 24659 + }, + { + "epoch": 1.3777970220968236, + "grad_norm": 0.3253410756587982, + "learning_rate": 0.0003125280143433438, + "loss": 0.3554, + "step": 24660 + }, + { + "epoch": 1.3778528927005056, + "grad_norm": 1.2906049489974976, + "learning_rate": 0.0003125, + "loss": 0.5833, + "step": 24661 + }, + { + "epoch": 1.3779087633041875, + "grad_norm": 0.6199899315834045, + "learning_rate": 0.0003124719856566562, + "loss": 0.3262, + "step": 24662 + }, + { + "epoch": 1.3779646339078693, + "grad_norm": 0.3582860827445984, + "learning_rate": 0.0003124439713133124, + "loss": 0.4458, + "step": 24663 + }, + { + "epoch": 1.3780205045115512, + "grad_norm": 0.3954160809516907, + "learning_rate": 0.0003124159569699686, + "loss": 0.4354, + "step": 24664 + }, + { + "epoch": 1.3780763751152332, + "grad_norm": 0.37739986181259155, + "learning_rate": 0.00031238794262662483, + "loss": 0.5387, + "step": 24665 + }, + { + "epoch": 1.378132245718915, + "grad_norm": 1.7281557321548462, + "learning_rate": 0.00031235992828328104, + "loss": 0.5093, + "step": 24666 + }, + { + "epoch": 1.3781881163225969, + "grad_norm": 0.6539730429649353, + "learning_rate": 0.00031233191393993724, + "loss": 0.3553, + "step": 24667 + }, + { + "epoch": 1.3782439869262788, + "grad_norm": 0.4740387201309204, + "learning_rate": 0.00031230389959659345, + "loss": 0.5195, + "step": 24668 + }, + { + "epoch": 1.3782998575299605, + "grad_norm": 0.7500782012939453, + "learning_rate": 0.0003122758852532497, + "loss": 0.3523, + "step": 24669 + }, + { + "epoch": 1.3783557281336425, + "grad_norm": 0.4838676154613495, + "learning_rate": 0.00031224787090990586, + "loss": 0.4755, + "step": 24670 + }, + { + "epoch": 1.3784115987373244, + "grad_norm": 0.6429794430732727, + "learning_rate": 0.0003122198565665621, + "loss": 0.4742, + "step": 24671 + }, + { + "epoch": 1.3784674693410062, + "grad_norm": 0.5406340956687927, + "learning_rate": 0.00031219184222321827, + "loss": 0.5239, + "step": 24672 + }, + { + "epoch": 1.3785233399446881, + "grad_norm": 0.8493536710739136, + "learning_rate": 0.00031216382787987453, + "loss": 0.6006, + "step": 24673 + }, + { + "epoch": 1.37857921054837, + "grad_norm": 0.43293526768684387, + "learning_rate": 0.0003121358135365307, + "loss": 0.4008, + "step": 24674 + }, + { + "epoch": 1.3786350811520518, + "grad_norm": 0.3905524015426636, + "learning_rate": 0.00031210779919318694, + "loss": 0.3318, + "step": 24675 + }, + { + "epoch": 1.3786909517557338, + "grad_norm": 0.4020029902458191, + "learning_rate": 0.0003120797848498431, + "loss": 0.4561, + "step": 24676 + }, + { + "epoch": 1.3787468223594157, + "grad_norm": 0.6172112226486206, + "learning_rate": 0.00031205177050649935, + "loss": 0.4808, + "step": 24677 + }, + { + "epoch": 1.3788026929630974, + "grad_norm": 0.5582640767097473, + "learning_rate": 0.00031202375616315556, + "loss": 0.3164, + "step": 24678 + }, + { + "epoch": 1.3788585635667794, + "grad_norm": 0.958625316619873, + "learning_rate": 0.00031199574181981177, + "loss": 0.3444, + "step": 24679 + }, + { + "epoch": 1.3789144341704613, + "grad_norm": 2.396491050720215, + "learning_rate": 0.00031196772747646797, + "loss": 0.3737, + "step": 24680 + }, + { + "epoch": 1.378970304774143, + "grad_norm": 0.7368852496147156, + "learning_rate": 0.0003119397131331242, + "loss": 0.4594, + "step": 24681 + }, + { + "epoch": 1.379026175377825, + "grad_norm": 0.37509724497795105, + "learning_rate": 0.0003119116987897804, + "loss": 0.4488, + "step": 24682 + }, + { + "epoch": 1.379082045981507, + "grad_norm": 0.8142114877700806, + "learning_rate": 0.00031188368444643653, + "loss": 0.3473, + "step": 24683 + }, + { + "epoch": 1.3791379165851887, + "grad_norm": 0.37686648964881897, + "learning_rate": 0.0003118556701030928, + "loss": 0.3752, + "step": 24684 + }, + { + "epoch": 1.3791937871888706, + "grad_norm": 0.36844223737716675, + "learning_rate": 0.000311827655759749, + "loss": 0.357, + "step": 24685 + }, + { + "epoch": 1.3792496577925524, + "grad_norm": 0.3968408405780792, + "learning_rate": 0.0003117996414164052, + "loss": 0.469, + "step": 24686 + }, + { + "epoch": 1.3793055283962343, + "grad_norm": 0.5391830801963806, + "learning_rate": 0.0003117716270730614, + "loss": 0.5051, + "step": 24687 + }, + { + "epoch": 1.3793613989999163, + "grad_norm": 0.450968474149704, + "learning_rate": 0.0003117436127297176, + "loss": 0.3463, + "step": 24688 + }, + { + "epoch": 1.379417269603598, + "grad_norm": 0.29674798250198364, + "learning_rate": 0.0003117155983863738, + "loss": 0.3004, + "step": 24689 + }, + { + "epoch": 1.37947314020728, + "grad_norm": 0.35822823643684387, + "learning_rate": 0.00031168758404303003, + "loss": 0.4571, + "step": 24690 + }, + { + "epoch": 1.3795290108109617, + "grad_norm": 0.8103290796279907, + "learning_rate": 0.00031165956969968624, + "loss": 0.6192, + "step": 24691 + }, + { + "epoch": 1.3795848814146436, + "grad_norm": 0.41023385524749756, + "learning_rate": 0.00031163155535634244, + "loss": 0.4147, + "step": 24692 + }, + { + "epoch": 1.3796407520183256, + "grad_norm": 0.3444606065750122, + "learning_rate": 0.00031160354101299865, + "loss": 0.318, + "step": 24693 + }, + { + "epoch": 1.3796966226220073, + "grad_norm": 0.35311198234558105, + "learning_rate": 0.0003115755266696549, + "loss": 0.4077, + "step": 24694 + }, + { + "epoch": 1.3797524932256893, + "grad_norm": 0.47818419337272644, + "learning_rate": 0.00031154751232631106, + "loss": 0.4349, + "step": 24695 + }, + { + "epoch": 1.3798083638293712, + "grad_norm": 1.8174772262573242, + "learning_rate": 0.0003115194979829673, + "loss": 0.4692, + "step": 24696 + }, + { + "epoch": 1.379864234433053, + "grad_norm": 0.4573594629764557, + "learning_rate": 0.00031149148363962347, + "loss": 0.4728, + "step": 24697 + }, + { + "epoch": 1.379920105036735, + "grad_norm": 0.3478325307369232, + "learning_rate": 0.00031146346929627973, + "loss": 0.398, + "step": 24698 + }, + { + "epoch": 1.3799759756404169, + "grad_norm": 0.326210081577301, + "learning_rate": 0.0003114354549529359, + "loss": 0.3113, + "step": 24699 + }, + { + "epoch": 1.3800318462440986, + "grad_norm": 0.3905622959136963, + "learning_rate": 0.00031140744060959214, + "loss": 0.4909, + "step": 24700 + }, + { + "epoch": 1.3800877168477805, + "grad_norm": 0.5066124200820923, + "learning_rate": 0.00031137942626624835, + "loss": 0.3575, + "step": 24701 + }, + { + "epoch": 1.3801435874514625, + "grad_norm": 0.39573854207992554, + "learning_rate": 0.00031135141192290455, + "loss": 0.542, + "step": 24702 + }, + { + "epoch": 1.3801994580551442, + "grad_norm": 0.4236353635787964, + "learning_rate": 0.00031132339757956076, + "loss": 0.3888, + "step": 24703 + }, + { + "epoch": 1.3802553286588262, + "grad_norm": 0.4596301019191742, + "learning_rate": 0.0003112953832362169, + "loss": 0.5343, + "step": 24704 + }, + { + "epoch": 1.3803111992625081, + "grad_norm": 0.5406448245048523, + "learning_rate": 0.00031126736889287317, + "loss": 0.4421, + "step": 24705 + }, + { + "epoch": 1.3803670698661898, + "grad_norm": 0.3544085919857025, + "learning_rate": 0.0003112393545495293, + "loss": 0.3127, + "step": 24706 + }, + { + "epoch": 1.3804229404698718, + "grad_norm": 0.42086732387542725, + "learning_rate": 0.0003112113402061856, + "loss": 0.4737, + "step": 24707 + }, + { + "epoch": 1.3804788110735537, + "grad_norm": 0.3865312337875366, + "learning_rate": 0.00031118332586284173, + "loss": 0.423, + "step": 24708 + }, + { + "epoch": 1.3805346816772355, + "grad_norm": 0.5474509596824646, + "learning_rate": 0.000311155311519498, + "loss": 0.4781, + "step": 24709 + }, + { + "epoch": 1.3805905522809174, + "grad_norm": 1.2002677917480469, + "learning_rate": 0.0003111272971761542, + "loss": 0.3348, + "step": 24710 + }, + { + "epoch": 1.3806464228845994, + "grad_norm": 0.42364153265953064, + "learning_rate": 0.0003110992828328104, + "loss": 0.4284, + "step": 24711 + }, + { + "epoch": 1.380702293488281, + "grad_norm": 0.5934035181999207, + "learning_rate": 0.0003110712684894666, + "loss": 0.4324, + "step": 24712 + }, + { + "epoch": 1.380758164091963, + "grad_norm": 0.34996533393859863, + "learning_rate": 0.0003110432541461228, + "loss": 0.4123, + "step": 24713 + }, + { + "epoch": 1.380814034695645, + "grad_norm": 0.45796820521354675, + "learning_rate": 0.000311015239802779, + "loss": 0.3877, + "step": 24714 + }, + { + "epoch": 1.3808699052993267, + "grad_norm": 0.4490642249584198, + "learning_rate": 0.00031098722545943523, + "loss": 0.4936, + "step": 24715 + }, + { + "epoch": 1.3809257759030087, + "grad_norm": 0.8233450651168823, + "learning_rate": 0.00031095921111609143, + "loss": 0.2964, + "step": 24716 + }, + { + "epoch": 1.3809816465066906, + "grad_norm": 0.47555533051490784, + "learning_rate": 0.0003109311967727477, + "loss": 0.3933, + "step": 24717 + }, + { + "epoch": 1.3810375171103724, + "grad_norm": 0.3766520917415619, + "learning_rate": 0.00031090318242940385, + "loss": 0.4402, + "step": 24718 + }, + { + "epoch": 1.3810933877140543, + "grad_norm": 0.3788241744041443, + "learning_rate": 0.0003108751680860601, + "loss": 0.4958, + "step": 24719 + }, + { + "epoch": 1.381149258317736, + "grad_norm": 0.40071237087249756, + "learning_rate": 0.00031084715374271626, + "loss": 0.3824, + "step": 24720 + }, + { + "epoch": 1.381205128921418, + "grad_norm": 0.8713767528533936, + "learning_rate": 0.0003108191393993725, + "loss": 0.3414, + "step": 24721 + }, + { + "epoch": 1.3812609995250997, + "grad_norm": 0.5823108553886414, + "learning_rate": 0.00031079112505602867, + "loss": 0.4055, + "step": 24722 + }, + { + "epoch": 1.3813168701287817, + "grad_norm": 0.5367872714996338, + "learning_rate": 0.00031076311071268493, + "loss": 0.4779, + "step": 24723 + }, + { + "epoch": 1.3813727407324636, + "grad_norm": 0.6538180112838745, + "learning_rate": 0.0003107350963693411, + "loss": 0.4066, + "step": 24724 + }, + { + "epoch": 1.3814286113361454, + "grad_norm": 0.40228790044784546, + "learning_rate": 0.0003107070820259973, + "loss": 0.4119, + "step": 24725 + }, + { + "epoch": 1.3814844819398273, + "grad_norm": 0.42171916365623474, + "learning_rate": 0.00031067906768265355, + "loss": 0.35, + "step": 24726 + }, + { + "epoch": 1.3815403525435093, + "grad_norm": 0.7368721961975098, + "learning_rate": 0.0003106510533393097, + "loss": 0.3969, + "step": 24727 + }, + { + "epoch": 1.381596223147191, + "grad_norm": 0.6513999700546265, + "learning_rate": 0.00031062303899596596, + "loss": 0.3871, + "step": 24728 + }, + { + "epoch": 1.381652093750873, + "grad_norm": 0.40515831112861633, + "learning_rate": 0.0003105950246526221, + "loss": 0.3317, + "step": 24729 + }, + { + "epoch": 1.381707964354555, + "grad_norm": 6.409379959106445, + "learning_rate": 0.00031056701030927837, + "loss": 0.4049, + "step": 24730 + }, + { + "epoch": 1.3817638349582366, + "grad_norm": 0.5314245223999023, + "learning_rate": 0.0003105389959659345, + "loss": 0.4916, + "step": 24731 + }, + { + "epoch": 1.3818197055619186, + "grad_norm": 0.423173189163208, + "learning_rate": 0.0003105109816225908, + "loss": 0.4031, + "step": 24732 + }, + { + "epoch": 1.3818755761656005, + "grad_norm": 0.36974483728408813, + "learning_rate": 0.000310482967279247, + "loss": 0.5292, + "step": 24733 + }, + { + "epoch": 1.3819314467692823, + "grad_norm": 0.42580220103263855, + "learning_rate": 0.0003104549529359032, + "loss": 0.4566, + "step": 24734 + }, + { + "epoch": 1.3819873173729642, + "grad_norm": 0.43156343698501587, + "learning_rate": 0.0003104269385925594, + "loss": 0.4504, + "step": 24735 + }, + { + "epoch": 1.3820431879766462, + "grad_norm": 0.5517470836639404, + "learning_rate": 0.0003103989242492156, + "loss": 0.4335, + "step": 24736 + }, + { + "epoch": 1.3820990585803279, + "grad_norm": 0.600007951259613, + "learning_rate": 0.0003103709099058718, + "loss": 0.4964, + "step": 24737 + }, + { + "epoch": 1.3821549291840098, + "grad_norm": 0.36744916439056396, + "learning_rate": 0.000310342895562528, + "loss": 0.4288, + "step": 24738 + }, + { + "epoch": 1.3822107997876918, + "grad_norm": 0.3450670540332794, + "learning_rate": 0.0003103148812191842, + "loss": 0.4577, + "step": 24739 + }, + { + "epoch": 1.3822666703913735, + "grad_norm": 0.40114590525627136, + "learning_rate": 0.0003102868668758405, + "loss": 0.4302, + "step": 24740 + }, + { + "epoch": 1.3823225409950555, + "grad_norm": 0.44508710503578186, + "learning_rate": 0.00031025885253249663, + "loss": 0.3937, + "step": 24741 + }, + { + "epoch": 1.3823784115987374, + "grad_norm": 0.46199920773506165, + "learning_rate": 0.0003102308381891529, + "loss": 0.4765, + "step": 24742 + }, + { + "epoch": 1.3824342822024192, + "grad_norm": 0.32329657673835754, + "learning_rate": 0.00031020282384580905, + "loss": 0.3348, + "step": 24743 + }, + { + "epoch": 1.382490152806101, + "grad_norm": 0.5287577509880066, + "learning_rate": 0.0003101748095024653, + "loss": 0.5929, + "step": 24744 + }, + { + "epoch": 1.382546023409783, + "grad_norm": 0.602811336517334, + "learning_rate": 0.00031014679515912146, + "loss": 0.4477, + "step": 24745 + }, + { + "epoch": 1.3826018940134648, + "grad_norm": 0.784876823425293, + "learning_rate": 0.00031011878081577766, + "loss": 0.3657, + "step": 24746 + }, + { + "epoch": 1.3826577646171467, + "grad_norm": 1.1349800825119019, + "learning_rate": 0.00031009076647243387, + "loss": 0.4439, + "step": 24747 + }, + { + "epoch": 1.3827136352208287, + "grad_norm": 0.699588418006897, + "learning_rate": 0.0003100627521290901, + "loss": 0.4067, + "step": 24748 + }, + { + "epoch": 1.3827695058245104, + "grad_norm": 0.4175185561180115, + "learning_rate": 0.00031003473778574634, + "loss": 0.4771, + "step": 24749 + }, + { + "epoch": 1.3828253764281924, + "grad_norm": 1.023662805557251, + "learning_rate": 0.0003100067234424025, + "loss": 0.4743, + "step": 24750 + }, + { + "epoch": 1.3828812470318743, + "grad_norm": 0.6269209384918213, + "learning_rate": 0.00030997870909905875, + "loss": 0.5278, + "step": 24751 + }, + { + "epoch": 1.382937117635556, + "grad_norm": 0.5654877424240112, + "learning_rate": 0.0003099506947557149, + "loss": 0.3452, + "step": 24752 + }, + { + "epoch": 1.382992988239238, + "grad_norm": 0.5950965285301208, + "learning_rate": 0.00030992268041237116, + "loss": 0.5105, + "step": 24753 + }, + { + "epoch": 1.3830488588429197, + "grad_norm": 0.5276561379432678, + "learning_rate": 0.0003098946660690273, + "loss": 0.6772, + "step": 24754 + }, + { + "epoch": 1.3831047294466017, + "grad_norm": 0.43337738513946533, + "learning_rate": 0.00030986665172568357, + "loss": 0.4871, + "step": 24755 + }, + { + "epoch": 1.3831606000502834, + "grad_norm": 0.43242165446281433, + "learning_rate": 0.0003098386373823398, + "loss": 0.4221, + "step": 24756 + }, + { + "epoch": 1.3832164706539654, + "grad_norm": 0.45020347833633423, + "learning_rate": 0.000309810623038996, + "loss": 0.436, + "step": 24757 + }, + { + "epoch": 1.3832723412576473, + "grad_norm": 0.37466108798980713, + "learning_rate": 0.0003097826086956522, + "loss": 0.3717, + "step": 24758 + }, + { + "epoch": 1.383328211861329, + "grad_norm": 0.4293549656867981, + "learning_rate": 0.0003097545943523084, + "loss": 0.4178, + "step": 24759 + }, + { + "epoch": 1.383384082465011, + "grad_norm": 0.7105755805969238, + "learning_rate": 0.0003097265800089646, + "loss": 0.4511, + "step": 24760 + }, + { + "epoch": 1.383439953068693, + "grad_norm": 0.8122818470001221, + "learning_rate": 0.0003096985656656208, + "loss": 0.4657, + "step": 24761 + }, + { + "epoch": 1.3834958236723747, + "grad_norm": 0.6549862623214722, + "learning_rate": 0.000309670551322277, + "loss": 0.4592, + "step": 24762 + }, + { + "epoch": 1.3835516942760566, + "grad_norm": 0.5110135674476624, + "learning_rate": 0.0003096425369789332, + "loss": 0.4966, + "step": 24763 + }, + { + "epoch": 1.3836075648797386, + "grad_norm": 0.626767635345459, + "learning_rate": 0.0003096145226355894, + "loss": 0.4526, + "step": 24764 + }, + { + "epoch": 1.3836634354834203, + "grad_norm": 0.5122875571250916, + "learning_rate": 0.0003095865082922457, + "loss": 0.3598, + "step": 24765 + }, + { + "epoch": 1.3837193060871023, + "grad_norm": 0.4600706100463867, + "learning_rate": 0.00030955849394890183, + "loss": 0.3109, + "step": 24766 + }, + { + "epoch": 1.3837751766907842, + "grad_norm": 0.6822768449783325, + "learning_rate": 0.0003095304796055581, + "loss": 0.496, + "step": 24767 + }, + { + "epoch": 1.383831047294466, + "grad_norm": 0.5417356491088867, + "learning_rate": 0.00030950246526221425, + "loss": 0.4365, + "step": 24768 + }, + { + "epoch": 1.3838869178981479, + "grad_norm": 0.40276139974594116, + "learning_rate": 0.00030947445091887045, + "loss": 0.4846, + "step": 24769 + }, + { + "epoch": 1.3839427885018298, + "grad_norm": 0.4234556555747986, + "learning_rate": 0.00030944643657552666, + "loss": 0.4421, + "step": 24770 + }, + { + "epoch": 1.3839986591055116, + "grad_norm": 0.6059746146202087, + "learning_rate": 0.00030941842223218286, + "loss": 0.3873, + "step": 24771 + }, + { + "epoch": 1.3840545297091935, + "grad_norm": 0.4114874303340912, + "learning_rate": 0.0003093904078888391, + "loss": 0.4347, + "step": 24772 + }, + { + "epoch": 1.3841104003128755, + "grad_norm": 0.41488566994667053, + "learning_rate": 0.0003093623935454953, + "loss": 0.3648, + "step": 24773 + }, + { + "epoch": 1.3841662709165572, + "grad_norm": 0.5201414227485657, + "learning_rate": 0.00030933437920215153, + "loss": 0.4288, + "step": 24774 + }, + { + "epoch": 1.3842221415202391, + "grad_norm": 0.3737536668777466, + "learning_rate": 0.0003093063648588077, + "loss": 0.451, + "step": 24775 + }, + { + "epoch": 1.384278012123921, + "grad_norm": 0.9990293383598328, + "learning_rate": 0.00030927835051546395, + "loss": 0.3886, + "step": 24776 + }, + { + "epoch": 1.3843338827276028, + "grad_norm": 1.174865484237671, + "learning_rate": 0.0003092503361721201, + "loss": 0.5512, + "step": 24777 + }, + { + "epoch": 1.3843897533312848, + "grad_norm": 0.5158865451812744, + "learning_rate": 0.00030922232182877636, + "loss": 0.479, + "step": 24778 + }, + { + "epoch": 1.3844456239349667, + "grad_norm": 0.4083426892757416, + "learning_rate": 0.0003091943074854325, + "loss": 0.4058, + "step": 24779 + }, + { + "epoch": 1.3845014945386485, + "grad_norm": 0.478687584400177, + "learning_rate": 0.00030916629314208877, + "loss": 0.4112, + "step": 24780 + }, + { + "epoch": 1.3845573651423304, + "grad_norm": 0.42723751068115234, + "learning_rate": 0.000309138278798745, + "loss": 0.3935, + "step": 24781 + }, + { + "epoch": 1.3846132357460124, + "grad_norm": 0.61372971534729, + "learning_rate": 0.0003091102644554012, + "loss": 0.4111, + "step": 24782 + }, + { + "epoch": 1.384669106349694, + "grad_norm": 1.7607953548431396, + "learning_rate": 0.0003090822501120574, + "loss": 0.3812, + "step": 24783 + }, + { + "epoch": 1.384724976953376, + "grad_norm": 0.5804696679115295, + "learning_rate": 0.0003090542357687136, + "loss": 0.4334, + "step": 24784 + }, + { + "epoch": 1.384780847557058, + "grad_norm": 0.43137261271476746, + "learning_rate": 0.0003090262214253698, + "loss": 0.551, + "step": 24785 + }, + { + "epoch": 1.3848367181607397, + "grad_norm": 0.44802939891815186, + "learning_rate": 0.000308998207082026, + "loss": 0.3852, + "step": 24786 + }, + { + "epoch": 1.3848925887644217, + "grad_norm": 0.5001193284988403, + "learning_rate": 0.0003089701927386822, + "loss": 0.4088, + "step": 24787 + }, + { + "epoch": 1.3849484593681034, + "grad_norm": 0.3534652590751648, + "learning_rate": 0.00030894217839533847, + "loss": 0.353, + "step": 24788 + }, + { + "epoch": 1.3850043299717854, + "grad_norm": 0.7307066321372986, + "learning_rate": 0.0003089141640519946, + "loss": 0.3535, + "step": 24789 + }, + { + "epoch": 1.385060200575467, + "grad_norm": 0.6070053577423096, + "learning_rate": 0.00030888614970865083, + "loss": 0.4688, + "step": 24790 + }, + { + "epoch": 1.385116071179149, + "grad_norm": 0.4625588655471802, + "learning_rate": 0.00030885813536530703, + "loss": 0.4827, + "step": 24791 + }, + { + "epoch": 1.385171941782831, + "grad_norm": 0.6489719152450562, + "learning_rate": 0.00030883012102196324, + "loss": 0.4883, + "step": 24792 + }, + { + "epoch": 1.3852278123865127, + "grad_norm": 0.47420138120651245, + "learning_rate": 0.00030880210667861945, + "loss": 0.515, + "step": 24793 + }, + { + "epoch": 1.3852836829901947, + "grad_norm": 0.43356555700302124, + "learning_rate": 0.00030877409233527565, + "loss": 0.4478, + "step": 24794 + }, + { + "epoch": 1.3853395535938766, + "grad_norm": 0.5024399161338806, + "learning_rate": 0.0003087460779919319, + "loss": 0.4499, + "step": 24795 + }, + { + "epoch": 1.3853954241975583, + "grad_norm": 0.42101335525512695, + "learning_rate": 0.00030871806364858806, + "loss": 0.4351, + "step": 24796 + }, + { + "epoch": 1.3854512948012403, + "grad_norm": 0.39215901494026184, + "learning_rate": 0.0003086900493052443, + "loss": 0.4814, + "step": 24797 + }, + { + "epoch": 1.3855071654049222, + "grad_norm": 0.47577595710754395, + "learning_rate": 0.0003086620349619005, + "loss": 0.3966, + "step": 24798 + }, + { + "epoch": 1.385563036008604, + "grad_norm": 0.6449511051177979, + "learning_rate": 0.00030863402061855673, + "loss": 0.3873, + "step": 24799 + }, + { + "epoch": 1.385618906612286, + "grad_norm": 1.9734902381896973, + "learning_rate": 0.0003086060062752129, + "loss": 0.489, + "step": 24800 + }, + { + "epoch": 1.3856747772159679, + "grad_norm": 0.36655133962631226, + "learning_rate": 0.00030857799193186915, + "loss": 0.3381, + "step": 24801 + }, + { + "epoch": 1.3857306478196496, + "grad_norm": 0.5470194816589355, + "learning_rate": 0.0003085499775885253, + "loss": 0.4687, + "step": 24802 + }, + { + "epoch": 1.3857865184233316, + "grad_norm": 0.6692386269569397, + "learning_rate": 0.00030852196324518156, + "loss": 0.4396, + "step": 24803 + }, + { + "epoch": 1.3858423890270135, + "grad_norm": 0.4561745226383209, + "learning_rate": 0.00030849394890183776, + "loss": 0.4626, + "step": 24804 + }, + { + "epoch": 1.3858982596306952, + "grad_norm": 0.7501745820045471, + "learning_rate": 0.00030846593455849397, + "loss": 0.4455, + "step": 24805 + }, + { + "epoch": 1.3859541302343772, + "grad_norm": 0.9584851264953613, + "learning_rate": 0.0003084379202151502, + "loss": 0.33, + "step": 24806 + }, + { + "epoch": 1.3860100008380591, + "grad_norm": 0.3281038701534271, + "learning_rate": 0.0003084099058718064, + "loss": 0.4021, + "step": 24807 + }, + { + "epoch": 1.3860658714417409, + "grad_norm": 0.44985297322273254, + "learning_rate": 0.0003083818915284626, + "loss": 0.4114, + "step": 24808 + }, + { + "epoch": 1.3861217420454228, + "grad_norm": 0.4140405058860779, + "learning_rate": 0.0003083538771851188, + "loss": 0.415, + "step": 24809 + }, + { + "epoch": 1.3861776126491048, + "grad_norm": 0.40750837326049805, + "learning_rate": 0.000308325862841775, + "loss": 0.3517, + "step": 24810 + }, + { + "epoch": 1.3862334832527865, + "grad_norm": 0.4187694787979126, + "learning_rate": 0.0003082978484984312, + "loss": 0.3343, + "step": 24811 + }, + { + "epoch": 1.3862893538564685, + "grad_norm": 0.4538898169994354, + "learning_rate": 0.0003082698341550874, + "loss": 0.447, + "step": 24812 + }, + { + "epoch": 1.3863452244601504, + "grad_norm": 0.613899290561676, + "learning_rate": 0.0003082418198117436, + "loss": 0.397, + "step": 24813 + }, + { + "epoch": 1.3864010950638321, + "grad_norm": 0.5018624067306519, + "learning_rate": 0.0003082138054683998, + "loss": 0.4426, + "step": 24814 + }, + { + "epoch": 1.386456965667514, + "grad_norm": 0.3279392719268799, + "learning_rate": 0.00030818579112505603, + "loss": 0.392, + "step": 24815 + }, + { + "epoch": 1.386512836271196, + "grad_norm": 0.3458634316921234, + "learning_rate": 0.00030815777678171223, + "loss": 0.3666, + "step": 24816 + }, + { + "epoch": 1.3865687068748778, + "grad_norm": 0.47198057174682617, + "learning_rate": 0.00030812976243836844, + "loss": 0.4431, + "step": 24817 + }, + { + "epoch": 1.3866245774785597, + "grad_norm": 0.3989321291446686, + "learning_rate": 0.00030810174809502464, + "loss": 0.4796, + "step": 24818 + }, + { + "epoch": 1.3866804480822414, + "grad_norm": 0.4542236626148224, + "learning_rate": 0.00030807373375168085, + "loss": 0.5171, + "step": 24819 + }, + { + "epoch": 1.3867363186859234, + "grad_norm": 0.3885350525379181, + "learning_rate": 0.0003080457194083371, + "loss": 0.4104, + "step": 24820 + }, + { + "epoch": 1.3867921892896053, + "grad_norm": 0.4409312903881073, + "learning_rate": 0.00030801770506499326, + "loss": 0.4473, + "step": 24821 + }, + { + "epoch": 1.386848059893287, + "grad_norm": 6.525620937347412, + "learning_rate": 0.0003079896907216495, + "loss": 0.3248, + "step": 24822 + }, + { + "epoch": 1.386903930496969, + "grad_norm": 0.33308738470077515, + "learning_rate": 0.0003079616763783057, + "loss": 0.3342, + "step": 24823 + }, + { + "epoch": 1.3869598011006508, + "grad_norm": 0.44862914085388184, + "learning_rate": 0.00030793366203496193, + "loss": 0.4845, + "step": 24824 + }, + { + "epoch": 1.3870156717043327, + "grad_norm": 0.9995421171188354, + "learning_rate": 0.0003079056476916181, + "loss": 0.3807, + "step": 24825 + }, + { + "epoch": 1.3870715423080147, + "grad_norm": 0.3043747842311859, + "learning_rate": 0.00030787763334827435, + "loss": 0.3555, + "step": 24826 + }, + { + "epoch": 1.3871274129116964, + "grad_norm": 0.5075585842132568, + "learning_rate": 0.00030784961900493055, + "loss": 0.4806, + "step": 24827 + }, + { + "epoch": 1.3871832835153783, + "grad_norm": 0.9293913841247559, + "learning_rate": 0.00030782160466158676, + "loss": 0.4532, + "step": 24828 + }, + { + "epoch": 1.3872391541190603, + "grad_norm": 0.4325586259365082, + "learning_rate": 0.00030779359031824296, + "loss": 0.5021, + "step": 24829 + }, + { + "epoch": 1.387295024722742, + "grad_norm": 0.37176552414894104, + "learning_rate": 0.00030776557597489917, + "loss": 0.3904, + "step": 24830 + }, + { + "epoch": 1.387350895326424, + "grad_norm": 0.5832306146621704, + "learning_rate": 0.0003077375616315554, + "loss": 0.4577, + "step": 24831 + }, + { + "epoch": 1.387406765930106, + "grad_norm": 0.8751463294029236, + "learning_rate": 0.0003077095472882115, + "loss": 0.3868, + "step": 24832 + }, + { + "epoch": 1.3874626365337877, + "grad_norm": 2.1591989994049072, + "learning_rate": 0.0003076815329448678, + "loss": 0.3794, + "step": 24833 + }, + { + "epoch": 1.3875185071374696, + "grad_norm": 0.4305052161216736, + "learning_rate": 0.00030765351860152394, + "loss": 0.4016, + "step": 24834 + }, + { + "epoch": 1.3875743777411516, + "grad_norm": 0.39178892970085144, + "learning_rate": 0.0003076255042581802, + "loss": 0.4556, + "step": 24835 + }, + { + "epoch": 1.3876302483448333, + "grad_norm": 0.42856159806251526, + "learning_rate": 0.0003075974899148364, + "loss": 0.49, + "step": 24836 + }, + { + "epoch": 1.3876861189485152, + "grad_norm": 0.5551307201385498, + "learning_rate": 0.0003075694755714926, + "loss": 0.6084, + "step": 24837 + }, + { + "epoch": 1.3877419895521972, + "grad_norm": 0.7496421337127686, + "learning_rate": 0.0003075414612281488, + "loss": 0.4322, + "step": 24838 + }, + { + "epoch": 1.387797860155879, + "grad_norm": 0.36586835980415344, + "learning_rate": 0.000307513446884805, + "loss": 0.4133, + "step": 24839 + }, + { + "epoch": 1.3878537307595609, + "grad_norm": 0.4342970550060272, + "learning_rate": 0.0003074854325414612, + "loss": 0.4051, + "step": 24840 + }, + { + "epoch": 1.3879096013632428, + "grad_norm": 0.32720786333084106, + "learning_rate": 0.00030745741819811743, + "loss": 0.4135, + "step": 24841 + }, + { + "epoch": 1.3879654719669245, + "grad_norm": 0.4604443311691284, + "learning_rate": 0.00030742940385477364, + "loss": 0.474, + "step": 24842 + }, + { + "epoch": 1.3880213425706065, + "grad_norm": 7.875121593475342, + "learning_rate": 0.0003074013895114299, + "loss": 0.4717, + "step": 24843 + }, + { + "epoch": 1.3880772131742884, + "grad_norm": 0.7248756885528564, + "learning_rate": 0.00030737337516808605, + "loss": 0.321, + "step": 24844 + }, + { + "epoch": 1.3881330837779702, + "grad_norm": 0.847934365272522, + "learning_rate": 0.0003073453608247423, + "loss": 0.373, + "step": 24845 + }, + { + "epoch": 1.3881889543816521, + "grad_norm": 0.3938896059989929, + "learning_rate": 0.00030731734648139846, + "loss": 0.4649, + "step": 24846 + }, + { + "epoch": 1.388244824985334, + "grad_norm": 1.8317790031433105, + "learning_rate": 0.0003072893321380547, + "loss": 0.3073, + "step": 24847 + }, + { + "epoch": 1.3883006955890158, + "grad_norm": 1.598581075668335, + "learning_rate": 0.0003072613177947109, + "loss": 0.3204, + "step": 24848 + }, + { + "epoch": 1.3883565661926978, + "grad_norm": 0.4276219308376312, + "learning_rate": 0.00030723330345136713, + "loss": 0.3482, + "step": 24849 + }, + { + "epoch": 1.3884124367963797, + "grad_norm": 0.6038588285446167, + "learning_rate": 0.0003072052891080233, + "loss": 0.4048, + "step": 24850 + }, + { + "epoch": 1.3884683074000614, + "grad_norm": 0.5747584104537964, + "learning_rate": 0.00030717727476467955, + "loss": 0.3912, + "step": 24851 + }, + { + "epoch": 1.3885241780037434, + "grad_norm": 6.149479866027832, + "learning_rate": 0.00030714926042133575, + "loss": 0.2539, + "step": 24852 + }, + { + "epoch": 1.3885800486074251, + "grad_norm": 0.43188923597335815, + "learning_rate": 0.0003071212460779919, + "loss": 0.5519, + "step": 24853 + }, + { + "epoch": 1.388635919211107, + "grad_norm": 0.5655919313430786, + "learning_rate": 0.00030709323173464816, + "loss": 0.5038, + "step": 24854 + }, + { + "epoch": 1.388691789814789, + "grad_norm": 0.42350026965141296, + "learning_rate": 0.0003070652173913043, + "loss": 0.4995, + "step": 24855 + }, + { + "epoch": 1.3887476604184708, + "grad_norm": 0.41204598546028137, + "learning_rate": 0.0003070372030479606, + "loss": 0.4701, + "step": 24856 + }, + { + "epoch": 1.3888035310221527, + "grad_norm": 0.40489205718040466, + "learning_rate": 0.0003070091887046167, + "loss": 0.4186, + "step": 24857 + }, + { + "epoch": 1.3888594016258344, + "grad_norm": 0.40741708874702454, + "learning_rate": 0.000306981174361273, + "loss": 0.5181, + "step": 24858 + }, + { + "epoch": 1.3889152722295164, + "grad_norm": 0.38408562541007996, + "learning_rate": 0.0003069531600179292, + "loss": 0.348, + "step": 24859 + }, + { + "epoch": 1.3889711428331983, + "grad_norm": 0.8601895570755005, + "learning_rate": 0.0003069251456745854, + "loss": 0.4431, + "step": 24860 + }, + { + "epoch": 1.38902701343688, + "grad_norm": 1.132491946220398, + "learning_rate": 0.0003068971313312416, + "loss": 0.424, + "step": 24861 + }, + { + "epoch": 1.389082884040562, + "grad_norm": 0.5659751296043396, + "learning_rate": 0.0003068691169878978, + "loss": 0.4036, + "step": 24862 + }, + { + "epoch": 1.389138754644244, + "grad_norm": 0.46242114901542664, + "learning_rate": 0.000306841102644554, + "loss": 0.3477, + "step": 24863 + }, + { + "epoch": 1.3891946252479257, + "grad_norm": 0.5245785117149353, + "learning_rate": 0.0003068130883012102, + "loss": 0.4721, + "step": 24864 + }, + { + "epoch": 1.3892504958516076, + "grad_norm": 0.5338692665100098, + "learning_rate": 0.0003067850739578664, + "loss": 0.4201, + "step": 24865 + }, + { + "epoch": 1.3893063664552896, + "grad_norm": 0.3514336943626404, + "learning_rate": 0.0003067570596145227, + "loss": 0.3109, + "step": 24866 + }, + { + "epoch": 1.3893622370589713, + "grad_norm": 0.7318530678749084, + "learning_rate": 0.00030672904527117884, + "loss": 0.5278, + "step": 24867 + }, + { + "epoch": 1.3894181076626533, + "grad_norm": 0.646131157875061, + "learning_rate": 0.0003067010309278351, + "loss": 0.4471, + "step": 24868 + }, + { + "epoch": 1.3894739782663352, + "grad_norm": 0.556520402431488, + "learning_rate": 0.00030667301658449125, + "loss": 0.4474, + "step": 24869 + }, + { + "epoch": 1.389529848870017, + "grad_norm": 0.5739495754241943, + "learning_rate": 0.0003066450022411475, + "loss": 0.4305, + "step": 24870 + }, + { + "epoch": 1.389585719473699, + "grad_norm": 0.3668670058250427, + "learning_rate": 0.00030661698789780366, + "loss": 0.3241, + "step": 24871 + }, + { + "epoch": 1.3896415900773809, + "grad_norm": 0.42778655886650085, + "learning_rate": 0.0003065889735544599, + "loss": 0.4883, + "step": 24872 + }, + { + "epoch": 1.3896974606810626, + "grad_norm": 0.9569193720817566, + "learning_rate": 0.0003065609592111161, + "loss": 0.4006, + "step": 24873 + }, + { + "epoch": 1.3897533312847445, + "grad_norm": 0.3894733488559723, + "learning_rate": 0.0003065329448677723, + "loss": 0.4197, + "step": 24874 + }, + { + "epoch": 1.3898092018884265, + "grad_norm": 0.4380839467048645, + "learning_rate": 0.00030650493052442854, + "loss": 0.4347, + "step": 24875 + }, + { + "epoch": 1.3898650724921082, + "grad_norm": 0.5179672241210938, + "learning_rate": 0.0003064769161810847, + "loss": 0.553, + "step": 24876 + }, + { + "epoch": 1.3899209430957902, + "grad_norm": 0.4352754056453705, + "learning_rate": 0.00030644890183774095, + "loss": 0.4809, + "step": 24877 + }, + { + "epoch": 1.3899768136994721, + "grad_norm": 0.5594584941864014, + "learning_rate": 0.0003064208874943971, + "loss": 0.4791, + "step": 24878 + }, + { + "epoch": 1.3900326843031539, + "grad_norm": 1.371315598487854, + "learning_rate": 0.00030639287315105336, + "loss": 0.4356, + "step": 24879 + }, + { + "epoch": 1.3900885549068358, + "grad_norm": 0.3722877502441406, + "learning_rate": 0.0003063648588077095, + "loss": 0.4976, + "step": 24880 + }, + { + "epoch": 1.3901444255105178, + "grad_norm": 0.3308373689651489, + "learning_rate": 0.0003063368444643658, + "loss": 0.3537, + "step": 24881 + }, + { + "epoch": 1.3902002961141995, + "grad_norm": 0.4689120054244995, + "learning_rate": 0.000306308830121022, + "loss": 0.494, + "step": 24882 + }, + { + "epoch": 1.3902561667178814, + "grad_norm": 0.4508424997329712, + "learning_rate": 0.0003062808157776782, + "loss": 0.5807, + "step": 24883 + }, + { + "epoch": 1.3903120373215634, + "grad_norm": 0.7504796385765076, + "learning_rate": 0.0003062528014343344, + "loss": 0.4908, + "step": 24884 + }, + { + "epoch": 1.3903679079252451, + "grad_norm": 2.611375331878662, + "learning_rate": 0.0003062247870909906, + "loss": 0.3536, + "step": 24885 + }, + { + "epoch": 1.390423778528927, + "grad_norm": 0.589514970779419, + "learning_rate": 0.0003061967727476468, + "loss": 0.3906, + "step": 24886 + }, + { + "epoch": 1.3904796491326088, + "grad_norm": 0.5351823568344116, + "learning_rate": 0.000306168758404303, + "loss": 0.3878, + "step": 24887 + }, + { + "epoch": 1.3905355197362907, + "grad_norm": 1.2014268636703491, + "learning_rate": 0.0003061407440609592, + "loss": 0.4337, + "step": 24888 + }, + { + "epoch": 1.3905913903399727, + "grad_norm": 0.40491509437561035, + "learning_rate": 0.0003061127297176154, + "loss": 0.3776, + "step": 24889 + }, + { + "epoch": 1.3906472609436544, + "grad_norm": 0.5443685054779053, + "learning_rate": 0.0003060847153742716, + "loss": 0.3944, + "step": 24890 + }, + { + "epoch": 1.3907031315473364, + "grad_norm": 0.6040867567062378, + "learning_rate": 0.0003060567010309279, + "loss": 0.5523, + "step": 24891 + }, + { + "epoch": 1.390759002151018, + "grad_norm": 0.8525095582008362, + "learning_rate": 0.00030602868668758404, + "loss": 0.4063, + "step": 24892 + }, + { + "epoch": 1.3908148727547, + "grad_norm": 0.47105804085731506, + "learning_rate": 0.0003060006723442403, + "loss": 0.5055, + "step": 24893 + }, + { + "epoch": 1.390870743358382, + "grad_norm": 0.43997955322265625, + "learning_rate": 0.00030597265800089645, + "loss": 0.3956, + "step": 24894 + }, + { + "epoch": 1.3909266139620637, + "grad_norm": 0.30303093791007996, + "learning_rate": 0.00030594464365755266, + "loss": 0.3484, + "step": 24895 + }, + { + "epoch": 1.3909824845657457, + "grad_norm": 0.3985835015773773, + "learning_rate": 0.00030591662931420886, + "loss": 0.403, + "step": 24896 + }, + { + "epoch": 1.3910383551694276, + "grad_norm": 0.4107621908187866, + "learning_rate": 0.00030588861497086507, + "loss": 0.4471, + "step": 24897 + }, + { + "epoch": 1.3910942257731094, + "grad_norm": 0.9589910507202148, + "learning_rate": 0.0003058606006275213, + "loss": 0.4202, + "step": 24898 + }, + { + "epoch": 1.3911500963767913, + "grad_norm": 2.7127771377563477, + "learning_rate": 0.0003058325862841775, + "loss": 0.5568, + "step": 24899 + }, + { + "epoch": 1.3912059669804733, + "grad_norm": 1.239359736442566, + "learning_rate": 0.00030580457194083374, + "loss": 0.354, + "step": 24900 + }, + { + "epoch": 1.391261837584155, + "grad_norm": 0.39800119400024414, + "learning_rate": 0.0003057765575974899, + "loss": 0.4355, + "step": 24901 + }, + { + "epoch": 1.391317708187837, + "grad_norm": 1.6763312816619873, + "learning_rate": 0.00030574854325414615, + "loss": 0.4032, + "step": 24902 + }, + { + "epoch": 1.391373578791519, + "grad_norm": 0.8698835372924805, + "learning_rate": 0.0003057205289108023, + "loss": 0.3965, + "step": 24903 + }, + { + "epoch": 1.3914294493952006, + "grad_norm": 0.8748095631599426, + "learning_rate": 0.00030569251456745856, + "loss": 0.4133, + "step": 24904 + }, + { + "epoch": 1.3914853199988826, + "grad_norm": 0.39852118492126465, + "learning_rate": 0.0003056645002241147, + "loss": 0.5164, + "step": 24905 + }, + { + "epoch": 1.3915411906025645, + "grad_norm": 0.4315069615840912, + "learning_rate": 0.000305636485880771, + "loss": 0.3553, + "step": 24906 + }, + { + "epoch": 1.3915970612062463, + "grad_norm": 0.444669246673584, + "learning_rate": 0.0003056084715374272, + "loss": 0.5132, + "step": 24907 + }, + { + "epoch": 1.3916529318099282, + "grad_norm": 0.49098095297813416, + "learning_rate": 0.0003055804571940834, + "loss": 0.4245, + "step": 24908 + }, + { + "epoch": 1.3917088024136102, + "grad_norm": 0.3293941915035248, + "learning_rate": 0.0003055524428507396, + "loss": 0.333, + "step": 24909 + }, + { + "epoch": 1.391764673017292, + "grad_norm": 0.35982751846313477, + "learning_rate": 0.0003055244285073958, + "loss": 0.4224, + "step": 24910 + }, + { + "epoch": 1.3918205436209738, + "grad_norm": 0.7798851132392883, + "learning_rate": 0.000305496414164052, + "loss": 0.4639, + "step": 24911 + }, + { + "epoch": 1.3918764142246558, + "grad_norm": 0.4037436842918396, + "learning_rate": 0.0003054683998207082, + "loss": 0.4228, + "step": 24912 + }, + { + "epoch": 1.3919322848283375, + "grad_norm": 0.9885292053222656, + "learning_rate": 0.0003054403854773644, + "loss": 0.3462, + "step": 24913 + }, + { + "epoch": 1.3919881554320195, + "grad_norm": 0.3871195912361145, + "learning_rate": 0.0003054123711340207, + "loss": 0.4309, + "step": 24914 + }, + { + "epoch": 1.3920440260357014, + "grad_norm": 0.4178009033203125, + "learning_rate": 0.0003053843567906768, + "loss": 0.4467, + "step": 24915 + }, + { + "epoch": 1.3920998966393832, + "grad_norm": 0.4915333092212677, + "learning_rate": 0.00030535634244733303, + "loss": 0.3742, + "step": 24916 + }, + { + "epoch": 1.3921557672430651, + "grad_norm": 0.3863416314125061, + "learning_rate": 0.00030532832810398924, + "loss": 0.3837, + "step": 24917 + }, + { + "epoch": 1.392211637846747, + "grad_norm": 0.3614286184310913, + "learning_rate": 0.00030530031376064544, + "loss": 0.398, + "step": 24918 + }, + { + "epoch": 1.3922675084504288, + "grad_norm": 0.3797650933265686, + "learning_rate": 0.00030527229941730165, + "loss": 0.4028, + "step": 24919 + }, + { + "epoch": 1.3923233790541107, + "grad_norm": 0.9129873514175415, + "learning_rate": 0.00030524428507395785, + "loss": 0.4328, + "step": 24920 + }, + { + "epoch": 1.3923792496577925, + "grad_norm": 0.3196428120136261, + "learning_rate": 0.00030521627073061406, + "loss": 0.2744, + "step": 24921 + }, + { + "epoch": 1.3924351202614744, + "grad_norm": 0.41444841027259827, + "learning_rate": 0.00030518825638727027, + "loss": 0.4228, + "step": 24922 + }, + { + "epoch": 1.3924909908651564, + "grad_norm": 0.5266194939613342, + "learning_rate": 0.0003051602420439265, + "loss": 0.3623, + "step": 24923 + }, + { + "epoch": 1.392546861468838, + "grad_norm": 0.7659940719604492, + "learning_rate": 0.0003051322277005827, + "loss": 0.5844, + "step": 24924 + }, + { + "epoch": 1.39260273207252, + "grad_norm": 0.3607201874256134, + "learning_rate": 0.00030510421335723894, + "loss": 0.4225, + "step": 24925 + }, + { + "epoch": 1.3926586026762018, + "grad_norm": 0.5333848595619202, + "learning_rate": 0.0003050761990138951, + "loss": 0.4319, + "step": 24926 + }, + { + "epoch": 1.3927144732798837, + "grad_norm": 0.7187523245811462, + "learning_rate": 0.00030504818467055135, + "loss": 0.4019, + "step": 24927 + }, + { + "epoch": 1.3927703438835657, + "grad_norm": 0.5226324200630188, + "learning_rate": 0.0003050201703272075, + "loss": 0.4927, + "step": 24928 + }, + { + "epoch": 1.3928262144872474, + "grad_norm": 5.585302829742432, + "learning_rate": 0.00030499215598386376, + "loss": 0.5492, + "step": 24929 + }, + { + "epoch": 1.3928820850909294, + "grad_norm": 0.7850372195243835, + "learning_rate": 0.00030496414164051997, + "loss": 0.3919, + "step": 24930 + }, + { + "epoch": 1.3929379556946113, + "grad_norm": 0.3854898512363434, + "learning_rate": 0.0003049361272971762, + "loss": 0.3278, + "step": 24931 + }, + { + "epoch": 1.392993826298293, + "grad_norm": 0.4110313355922699, + "learning_rate": 0.0003049081129538324, + "loss": 0.5324, + "step": 24932 + }, + { + "epoch": 1.393049696901975, + "grad_norm": 6.397408962249756, + "learning_rate": 0.0003048800986104886, + "loss": 0.4467, + "step": 24933 + }, + { + "epoch": 1.393105567505657, + "grad_norm": 0.363856703042984, + "learning_rate": 0.0003048520842671448, + "loss": 0.4482, + "step": 24934 + }, + { + "epoch": 1.3931614381093387, + "grad_norm": 0.3677813708782196, + "learning_rate": 0.000304824069923801, + "loss": 0.4193, + "step": 24935 + }, + { + "epoch": 1.3932173087130206, + "grad_norm": 0.32812026143074036, + "learning_rate": 0.0003047960555804572, + "loss": 0.3518, + "step": 24936 + }, + { + "epoch": 1.3932731793167026, + "grad_norm": 0.5118632912635803, + "learning_rate": 0.00030476804123711346, + "loss": 0.4202, + "step": 24937 + }, + { + "epoch": 1.3933290499203843, + "grad_norm": 0.49738961458206177, + "learning_rate": 0.0003047400268937696, + "loss": 0.4147, + "step": 24938 + }, + { + "epoch": 1.3933849205240663, + "grad_norm": 0.29511594772338867, + "learning_rate": 0.0003047120125504258, + "loss": 0.3371, + "step": 24939 + }, + { + "epoch": 1.3934407911277482, + "grad_norm": 2.1739253997802734, + "learning_rate": 0.000304683998207082, + "loss": 0.4076, + "step": 24940 + }, + { + "epoch": 1.39349666173143, + "grad_norm": 0.6608916521072388, + "learning_rate": 0.00030465598386373823, + "loss": 0.4129, + "step": 24941 + }, + { + "epoch": 1.393552532335112, + "grad_norm": 0.40961605310440063, + "learning_rate": 0.00030462796952039444, + "loss": 0.372, + "step": 24942 + }, + { + "epoch": 1.3936084029387938, + "grad_norm": 0.41363325715065, + "learning_rate": 0.00030459995517705064, + "loss": 0.3925, + "step": 24943 + }, + { + "epoch": 1.3936642735424756, + "grad_norm": 0.545559287071228, + "learning_rate": 0.00030457194083370685, + "loss": 0.4512, + "step": 24944 + }, + { + "epoch": 1.3937201441461575, + "grad_norm": 0.46408116817474365, + "learning_rate": 0.00030454392649036305, + "loss": 0.4812, + "step": 24945 + }, + { + "epoch": 1.3937760147498395, + "grad_norm": 0.541314959526062, + "learning_rate": 0.0003045159121470193, + "loss": 0.4767, + "step": 24946 + }, + { + "epoch": 1.3938318853535212, + "grad_norm": 0.590434730052948, + "learning_rate": 0.00030448789780367547, + "loss": 0.4085, + "step": 24947 + }, + { + "epoch": 1.3938877559572032, + "grad_norm": 0.5001686215400696, + "learning_rate": 0.0003044598834603317, + "loss": 0.5457, + "step": 24948 + }, + { + "epoch": 1.393943626560885, + "grad_norm": 1.282332181930542, + "learning_rate": 0.0003044318691169879, + "loss": 0.417, + "step": 24949 + }, + { + "epoch": 1.3939994971645668, + "grad_norm": 0.422426700592041, + "learning_rate": 0.00030440385477364414, + "loss": 0.4128, + "step": 24950 + }, + { + "epoch": 1.3940553677682488, + "grad_norm": 0.9699578285217285, + "learning_rate": 0.0003043758404303003, + "loss": 0.517, + "step": 24951 + }, + { + "epoch": 1.3941112383719307, + "grad_norm": 0.43999385833740234, + "learning_rate": 0.00030434782608695655, + "loss": 0.5031, + "step": 24952 + }, + { + "epoch": 1.3941671089756125, + "grad_norm": 0.3361496925354004, + "learning_rate": 0.00030431981174361276, + "loss": 0.4476, + "step": 24953 + }, + { + "epoch": 1.3942229795792944, + "grad_norm": 0.4084065854549408, + "learning_rate": 0.00030429179740026896, + "loss": 0.5072, + "step": 24954 + }, + { + "epoch": 1.3942788501829761, + "grad_norm": 0.4951719343662262, + "learning_rate": 0.00030426378305692517, + "loss": 0.4489, + "step": 24955 + }, + { + "epoch": 1.394334720786658, + "grad_norm": 0.41947662830352783, + "learning_rate": 0.00030423576871358137, + "loss": 0.4135, + "step": 24956 + }, + { + "epoch": 1.39439059139034, + "grad_norm": 0.34856921434402466, + "learning_rate": 0.0003042077543702376, + "loss": 0.3577, + "step": 24957 + }, + { + "epoch": 1.3944464619940218, + "grad_norm": 0.3592633903026581, + "learning_rate": 0.0003041797400268938, + "loss": 0.4511, + "step": 24958 + }, + { + "epoch": 1.3945023325977037, + "grad_norm": 0.3624826967716217, + "learning_rate": 0.00030415172568355, + "loss": 0.3607, + "step": 24959 + }, + { + "epoch": 1.3945582032013855, + "grad_norm": 0.48899465799331665, + "learning_rate": 0.00030412371134020614, + "loss": 0.476, + "step": 24960 + }, + { + "epoch": 1.3946140738050674, + "grad_norm": 0.3461707830429077, + "learning_rate": 0.0003040956969968624, + "loss": 0.3802, + "step": 24961 + }, + { + "epoch": 1.3946699444087494, + "grad_norm": 0.42009231448173523, + "learning_rate": 0.0003040676826535186, + "loss": 0.4518, + "step": 24962 + }, + { + "epoch": 1.394725815012431, + "grad_norm": 0.3520447015762329, + "learning_rate": 0.0003040396683101748, + "loss": 0.3983, + "step": 24963 + }, + { + "epoch": 1.394781685616113, + "grad_norm": 9.095337867736816, + "learning_rate": 0.000304011653966831, + "loss": 0.4055, + "step": 24964 + }, + { + "epoch": 1.394837556219795, + "grad_norm": 0.47877588868141174, + "learning_rate": 0.0003039836396234872, + "loss": 0.6039, + "step": 24965 + }, + { + "epoch": 1.3948934268234767, + "grad_norm": 0.5836473107337952, + "learning_rate": 0.00030395562528014343, + "loss": 0.3419, + "step": 24966 + }, + { + "epoch": 1.3949492974271587, + "grad_norm": 1.3174247741699219, + "learning_rate": 0.00030392761093679964, + "loss": 0.4488, + "step": 24967 + }, + { + "epoch": 1.3950051680308406, + "grad_norm": 0.5349089503288269, + "learning_rate": 0.00030389959659345584, + "loss": 0.3932, + "step": 24968 + }, + { + "epoch": 1.3950610386345224, + "grad_norm": 0.3397807776927948, + "learning_rate": 0.0003038715822501121, + "loss": 0.4245, + "step": 24969 + }, + { + "epoch": 1.3951169092382043, + "grad_norm": 0.5275771021842957, + "learning_rate": 0.00030384356790676825, + "loss": 0.6574, + "step": 24970 + }, + { + "epoch": 1.3951727798418863, + "grad_norm": 0.6051958203315735, + "learning_rate": 0.0003038155535634245, + "loss": 0.5035, + "step": 24971 + }, + { + "epoch": 1.395228650445568, + "grad_norm": 0.39965328574180603, + "learning_rate": 0.00030378753922008067, + "loss": 0.4878, + "step": 24972 + }, + { + "epoch": 1.39528452104925, + "grad_norm": 0.5942613482475281, + "learning_rate": 0.0003037595248767369, + "loss": 0.332, + "step": 24973 + }, + { + "epoch": 1.3953403916529319, + "grad_norm": 0.3920553922653198, + "learning_rate": 0.0003037315105333931, + "loss": 0.35, + "step": 24974 + }, + { + "epoch": 1.3953962622566136, + "grad_norm": 0.5647568106651306, + "learning_rate": 0.00030370349619004934, + "loss": 0.4287, + "step": 24975 + }, + { + "epoch": 1.3954521328602956, + "grad_norm": 0.5253053307533264, + "learning_rate": 0.0003036754818467055, + "loss": 0.4012, + "step": 24976 + }, + { + "epoch": 1.3955080034639775, + "grad_norm": 0.3782784342765808, + "learning_rate": 0.00030364746750336175, + "loss": 0.3623, + "step": 24977 + }, + { + "epoch": 1.3955638740676592, + "grad_norm": 0.9400460124015808, + "learning_rate": 0.00030361945316001795, + "loss": 0.4568, + "step": 24978 + }, + { + "epoch": 1.3956197446713412, + "grad_norm": 0.4861321747303009, + "learning_rate": 0.00030359143881667416, + "loss": 0.4096, + "step": 24979 + }, + { + "epoch": 1.3956756152750232, + "grad_norm": 0.6963964700698853, + "learning_rate": 0.00030356342447333037, + "loss": 0.4536, + "step": 24980 + }, + { + "epoch": 1.3957314858787049, + "grad_norm": 0.4914814829826355, + "learning_rate": 0.0003035354101299865, + "loss": 0.4118, + "step": 24981 + }, + { + "epoch": 1.3957873564823868, + "grad_norm": 0.4753936529159546, + "learning_rate": 0.0003035073957866428, + "loss": 0.3937, + "step": 24982 + }, + { + "epoch": 1.3958432270860688, + "grad_norm": 0.35393211245536804, + "learning_rate": 0.00030347938144329893, + "loss": 0.4156, + "step": 24983 + }, + { + "epoch": 1.3958990976897505, + "grad_norm": 0.3499584496021271, + "learning_rate": 0.0003034513670999552, + "loss": 0.4098, + "step": 24984 + }, + { + "epoch": 1.3959549682934325, + "grad_norm": 0.5299673676490784, + "learning_rate": 0.0003034233527566114, + "loss": 0.5571, + "step": 24985 + }, + { + "epoch": 1.3960108388971144, + "grad_norm": 0.2910718321800232, + "learning_rate": 0.0003033953384132676, + "loss": 0.2762, + "step": 24986 + }, + { + "epoch": 1.3960667095007961, + "grad_norm": 0.3922654688358307, + "learning_rate": 0.0003033673240699238, + "loss": 0.3624, + "step": 24987 + }, + { + "epoch": 1.396122580104478, + "grad_norm": 0.37554869055747986, + "learning_rate": 0.00030333930972658, + "loss": 0.4796, + "step": 24988 + }, + { + "epoch": 1.3961784507081598, + "grad_norm": 0.7936713695526123, + "learning_rate": 0.0003033112953832362, + "loss": 0.3166, + "step": 24989 + }, + { + "epoch": 1.3962343213118418, + "grad_norm": 0.4347393810749054, + "learning_rate": 0.0003032832810398924, + "loss": 0.4116, + "step": 24990 + }, + { + "epoch": 1.3962901919155237, + "grad_norm": 0.3888291120529175, + "learning_rate": 0.00030325526669654863, + "loss": 0.4002, + "step": 24991 + }, + { + "epoch": 1.3963460625192055, + "grad_norm": 1.9732400178909302, + "learning_rate": 0.0003032272523532049, + "loss": 0.4979, + "step": 24992 + }, + { + "epoch": 1.3964019331228874, + "grad_norm": 0.9401376843452454, + "learning_rate": 0.00030319923800986104, + "loss": 0.4123, + "step": 24993 + }, + { + "epoch": 1.3964578037265691, + "grad_norm": 0.8861599564552307, + "learning_rate": 0.0003031712236665173, + "loss": 0.4956, + "step": 24994 + }, + { + "epoch": 1.396513674330251, + "grad_norm": 0.658684492111206, + "learning_rate": 0.00030314320932317345, + "loss": 0.4356, + "step": 24995 + }, + { + "epoch": 1.396569544933933, + "grad_norm": 0.4755181670188904, + "learning_rate": 0.0003031151949798297, + "loss": 0.3774, + "step": 24996 + }, + { + "epoch": 1.3966254155376148, + "grad_norm": 0.5694617629051208, + "learning_rate": 0.00030308718063648587, + "loss": 0.5174, + "step": 24997 + }, + { + "epoch": 1.3966812861412967, + "grad_norm": 0.329653799533844, + "learning_rate": 0.0003030591662931421, + "loss": 0.3545, + "step": 24998 + }, + { + "epoch": 1.3967371567449787, + "grad_norm": 0.45617973804473877, + "learning_rate": 0.0003030311519497983, + "loss": 0.4228, + "step": 24999 + }, + { + "epoch": 1.3967930273486604, + "grad_norm": 1.1318237781524658, + "learning_rate": 0.00030300313760645454, + "loss": 0.4601, + "step": 25000 + }, + { + "epoch": 1.3967930273486604, + "eval_cer": 0.08578551711346055, + "eval_loss": 0.3196848928928375, + "eval_runtime": 56.2477, + "eval_samples_per_second": 80.679, + "eval_steps_per_second": 5.049, + "eval_wer": 0.34046535012988466, + "step": 25000 + }, + { + "epoch": 1.3968488979523423, + "grad_norm": 1.3776681423187256, + "learning_rate": 0.00030297512326311074, + "loss": 0.5128, + "step": 25001 + }, + { + "epoch": 1.3969047685560243, + "grad_norm": 0.3506929576396942, + "learning_rate": 0.0003029471089197669, + "loss": 0.398, + "step": 25002 + }, + { + "epoch": 1.396960639159706, + "grad_norm": 2.8316407203674316, + "learning_rate": 0.00030291909457642315, + "loss": 0.5487, + "step": 25003 + }, + { + "epoch": 1.397016509763388, + "grad_norm": 0.3636258542537689, + "learning_rate": 0.0003028910802330793, + "loss": 0.4148, + "step": 25004 + }, + { + "epoch": 1.39707238036707, + "grad_norm": 0.42588022351264954, + "learning_rate": 0.00030286306588973557, + "loss": 0.4556, + "step": 25005 + }, + { + "epoch": 1.3971282509707517, + "grad_norm": 0.6541616916656494, + "learning_rate": 0.0003028350515463917, + "loss": 0.3587, + "step": 25006 + }, + { + "epoch": 1.3971841215744336, + "grad_norm": 1.1411908864974976, + "learning_rate": 0.000302807037203048, + "loss": 0.5816, + "step": 25007 + }, + { + "epoch": 1.3972399921781156, + "grad_norm": 0.39364391565322876, + "learning_rate": 0.0003027790228597042, + "loss": 0.3778, + "step": 25008 + }, + { + "epoch": 1.3972958627817973, + "grad_norm": 0.40545782446861267, + "learning_rate": 0.0003027510085163604, + "loss": 0.4908, + "step": 25009 + }, + { + "epoch": 1.3973517333854792, + "grad_norm": 0.34452396631240845, + "learning_rate": 0.0003027229941730166, + "loss": 0.3763, + "step": 25010 + }, + { + "epoch": 1.3974076039891612, + "grad_norm": 0.827235221862793, + "learning_rate": 0.0003026949798296728, + "loss": 0.4268, + "step": 25011 + }, + { + "epoch": 1.397463474592843, + "grad_norm": 0.5505421161651611, + "learning_rate": 0.000302666965486329, + "loss": 0.412, + "step": 25012 + }, + { + "epoch": 1.3975193451965249, + "grad_norm": 0.5731505155563354, + "learning_rate": 0.0003026389511429852, + "loss": 0.4873, + "step": 25013 + }, + { + "epoch": 1.3975752158002068, + "grad_norm": 0.4122897982597351, + "learning_rate": 0.0003026109367996414, + "loss": 0.396, + "step": 25014 + }, + { + "epoch": 1.3976310864038886, + "grad_norm": 2.1813368797302246, + "learning_rate": 0.0003025829224562976, + "loss": 0.4467, + "step": 25015 + }, + { + "epoch": 1.3976869570075705, + "grad_norm": 0.5911661386489868, + "learning_rate": 0.00030255490811295383, + "loss": 0.3619, + "step": 25016 + }, + { + "epoch": 1.3977428276112525, + "grad_norm": 0.4471738338470459, + "learning_rate": 0.0003025268937696101, + "loss": 0.4423, + "step": 25017 + }, + { + "epoch": 1.3977986982149342, + "grad_norm": 0.4140910506248474, + "learning_rate": 0.00030249887942626624, + "loss": 0.3818, + "step": 25018 + }, + { + "epoch": 1.3978545688186161, + "grad_norm": 0.7444854378700256, + "learning_rate": 0.0003024708650829225, + "loss": 0.4681, + "step": 25019 + }, + { + "epoch": 1.397910439422298, + "grad_norm": 1.1878191232681274, + "learning_rate": 0.00030244285073957865, + "loss": 0.4699, + "step": 25020 + }, + { + "epoch": 1.3979663100259798, + "grad_norm": 0.6135988831520081, + "learning_rate": 0.0003024148363962349, + "loss": 0.6032, + "step": 25021 + }, + { + "epoch": 1.3980221806296618, + "grad_norm": 0.3481011986732483, + "learning_rate": 0.00030238682205289106, + "loss": 0.4684, + "step": 25022 + }, + { + "epoch": 1.3980780512333435, + "grad_norm": 0.5249390006065369, + "learning_rate": 0.00030235880770954727, + "loss": 0.5604, + "step": 25023 + }, + { + "epoch": 1.3981339218370255, + "grad_norm": 0.550870954990387, + "learning_rate": 0.00030233079336620353, + "loss": 0.3722, + "step": 25024 + }, + { + "epoch": 1.3981897924407074, + "grad_norm": 0.41084885597229004, + "learning_rate": 0.0003023027790228597, + "loss": 0.4082, + "step": 25025 + }, + { + "epoch": 1.3982456630443891, + "grad_norm": 0.44016745686531067, + "learning_rate": 0.00030227476467951594, + "loss": 0.4084, + "step": 25026 + }, + { + "epoch": 1.398301533648071, + "grad_norm": 0.63351970911026, + "learning_rate": 0.0003022467503361721, + "loss": 0.3968, + "step": 25027 + }, + { + "epoch": 1.3983574042517528, + "grad_norm": 0.49211013317108154, + "learning_rate": 0.00030221873599282835, + "loss": 0.3355, + "step": 25028 + }, + { + "epoch": 1.3984132748554348, + "grad_norm": 0.4929969012737274, + "learning_rate": 0.0003021907216494845, + "loss": 0.4793, + "step": 25029 + }, + { + "epoch": 1.3984691454591167, + "grad_norm": 0.41224581003189087, + "learning_rate": 0.00030216270730614077, + "loss": 0.6008, + "step": 25030 + }, + { + "epoch": 1.3985250160627984, + "grad_norm": 1.1294101476669312, + "learning_rate": 0.0003021346929627969, + "loss": 0.3608, + "step": 25031 + }, + { + "epoch": 1.3985808866664804, + "grad_norm": 0.8247612118721008, + "learning_rate": 0.0003021066786194532, + "loss": 0.4556, + "step": 25032 + }, + { + "epoch": 1.3986367572701623, + "grad_norm": 0.454879492521286, + "learning_rate": 0.0003020786642761094, + "loss": 0.5366, + "step": 25033 + }, + { + "epoch": 1.398692627873844, + "grad_norm": 0.8277721405029297, + "learning_rate": 0.0003020506499327656, + "loss": 0.5165, + "step": 25034 + }, + { + "epoch": 1.398748498477526, + "grad_norm": 1.4426103830337524, + "learning_rate": 0.0003020226355894218, + "loss": 0.4189, + "step": 25035 + }, + { + "epoch": 1.398804369081208, + "grad_norm": 0.4736434817314148, + "learning_rate": 0.000301994621246078, + "loss": 0.4622, + "step": 25036 + }, + { + "epoch": 1.3988602396848897, + "grad_norm": 0.6247780919075012, + "learning_rate": 0.0003019666069027342, + "loss": 0.5095, + "step": 25037 + }, + { + "epoch": 1.3989161102885717, + "grad_norm": 1.3226778507232666, + "learning_rate": 0.0003019385925593904, + "loss": 0.4631, + "step": 25038 + }, + { + "epoch": 1.3989719808922536, + "grad_norm": 1.3239349126815796, + "learning_rate": 0.0003019105782160466, + "loss": 0.3978, + "step": 25039 + }, + { + "epoch": 1.3990278514959353, + "grad_norm": 0.9859894514083862, + "learning_rate": 0.0003018825638727029, + "loss": 0.4014, + "step": 25040 + }, + { + "epoch": 1.3990837220996173, + "grad_norm": 1.144464373588562, + "learning_rate": 0.00030185454952935903, + "loss": 0.4391, + "step": 25041 + }, + { + "epoch": 1.3991395927032992, + "grad_norm": 0.5746455192565918, + "learning_rate": 0.0003018265351860153, + "loss": 0.5541, + "step": 25042 + }, + { + "epoch": 1.399195463306981, + "grad_norm": 0.4079228639602661, + "learning_rate": 0.00030179852084267144, + "loss": 0.4467, + "step": 25043 + }, + { + "epoch": 1.399251333910663, + "grad_norm": 0.666991651058197, + "learning_rate": 0.00030177050649932765, + "loss": 0.5247, + "step": 25044 + }, + { + "epoch": 1.3993072045143449, + "grad_norm": 1.6851742267608643, + "learning_rate": 0.00030174249215598385, + "loss": 0.3648, + "step": 25045 + }, + { + "epoch": 1.3993630751180266, + "grad_norm": 0.941083550453186, + "learning_rate": 0.00030171447781264006, + "loss": 0.3645, + "step": 25046 + }, + { + "epoch": 1.3994189457217086, + "grad_norm": 0.8927803039550781, + "learning_rate": 0.00030168646346929626, + "loss": 0.4917, + "step": 25047 + }, + { + "epoch": 1.3994748163253905, + "grad_norm": 3.037759780883789, + "learning_rate": 0.00030165844912595247, + "loss": 0.4368, + "step": 25048 + }, + { + "epoch": 1.3995306869290722, + "grad_norm": 0.44673222303390503, + "learning_rate": 0.00030163043478260873, + "loss": 0.3343, + "step": 25049 + }, + { + "epoch": 1.3995865575327542, + "grad_norm": 0.37730085849761963, + "learning_rate": 0.0003016024204392649, + "loss": 0.331, + "step": 25050 + }, + { + "epoch": 1.3996424281364361, + "grad_norm": 0.6113755106925964, + "learning_rate": 0.00030157440609592114, + "loss": 0.4764, + "step": 25051 + }, + { + "epoch": 1.3996982987401179, + "grad_norm": 0.3797110915184021, + "learning_rate": 0.0003015463917525773, + "loss": 0.3486, + "step": 25052 + }, + { + "epoch": 1.3997541693437998, + "grad_norm": 0.5265967845916748, + "learning_rate": 0.00030151837740923355, + "loss": 0.4062, + "step": 25053 + }, + { + "epoch": 1.3998100399474818, + "grad_norm": 0.39122119545936584, + "learning_rate": 0.0003014903630658897, + "loss": 0.4728, + "step": 25054 + }, + { + "epoch": 1.3998659105511635, + "grad_norm": 0.3981761336326599, + "learning_rate": 0.00030146234872254597, + "loss": 0.4766, + "step": 25055 + }, + { + "epoch": 1.3999217811548454, + "grad_norm": 0.4168090224266052, + "learning_rate": 0.00030143433437920217, + "loss": 0.3995, + "step": 25056 + }, + { + "epoch": 1.3999776517585272, + "grad_norm": 0.6330571174621582, + "learning_rate": 0.0003014063200358584, + "loss": 0.3792, + "step": 25057 + }, + { + "epoch": 1.4000335223622091, + "grad_norm": 1.384414792060852, + "learning_rate": 0.0003013783056925146, + "loss": 0.3726, + "step": 25058 + }, + { + "epoch": 1.400089392965891, + "grad_norm": 0.4148305356502533, + "learning_rate": 0.0003013502913491708, + "loss": 0.4993, + "step": 25059 + }, + { + "epoch": 1.4001452635695728, + "grad_norm": 0.5033206939697266, + "learning_rate": 0.000301322277005827, + "loss": 0.3624, + "step": 25060 + }, + { + "epoch": 1.4002011341732548, + "grad_norm": 0.4744735360145569, + "learning_rate": 0.0003012942626624832, + "loss": 0.3685, + "step": 25061 + }, + { + "epoch": 1.4002570047769365, + "grad_norm": 0.5262047648429871, + "learning_rate": 0.0003012662483191394, + "loss": 0.4967, + "step": 25062 + }, + { + "epoch": 1.4003128753806184, + "grad_norm": 0.72247314453125, + "learning_rate": 0.00030123823397579567, + "loss": 0.3408, + "step": 25063 + }, + { + "epoch": 1.4003687459843004, + "grad_norm": 0.591766893863678, + "learning_rate": 0.0003012102196324518, + "loss": 0.3441, + "step": 25064 + }, + { + "epoch": 1.4004246165879821, + "grad_norm": 0.486727774143219, + "learning_rate": 0.000301182205289108, + "loss": 0.3965, + "step": 25065 + }, + { + "epoch": 1.400480487191664, + "grad_norm": 0.5620769262313843, + "learning_rate": 0.00030115419094576423, + "loss": 0.4614, + "step": 25066 + }, + { + "epoch": 1.400536357795346, + "grad_norm": 1.0921201705932617, + "learning_rate": 0.00030112617660242043, + "loss": 0.5205, + "step": 25067 + }, + { + "epoch": 1.4005922283990277, + "grad_norm": 0.3826963007450104, + "learning_rate": 0.00030109816225907664, + "loss": 0.5334, + "step": 25068 + }, + { + "epoch": 1.4006480990027097, + "grad_norm": 0.7020829319953918, + "learning_rate": 0.00030107014791573285, + "loss": 0.4445, + "step": 25069 + }, + { + "epoch": 1.4007039696063917, + "grad_norm": 0.803946316242218, + "learning_rate": 0.00030104213357238905, + "loss": 0.4336, + "step": 25070 + }, + { + "epoch": 1.4007598402100734, + "grad_norm": 0.4607584476470947, + "learning_rate": 0.00030101411922904526, + "loss": 0.415, + "step": 25071 + }, + { + "epoch": 1.4008157108137553, + "grad_norm": 1.4815397262573242, + "learning_rate": 0.0003009861048857015, + "loss": 0.432, + "step": 25072 + }, + { + "epoch": 1.4008715814174373, + "grad_norm": 0.4135887622833252, + "learning_rate": 0.00030095809054235767, + "loss": 0.4194, + "step": 25073 + }, + { + "epoch": 1.400927452021119, + "grad_norm": 0.6319279074668884, + "learning_rate": 0.00030093007619901393, + "loss": 0.485, + "step": 25074 + }, + { + "epoch": 1.400983322624801, + "grad_norm": 0.535199761390686, + "learning_rate": 0.0003009020618556701, + "loss": 0.3679, + "step": 25075 + }, + { + "epoch": 1.401039193228483, + "grad_norm": 1.0585347414016724, + "learning_rate": 0.00030087404751232634, + "loss": 0.4669, + "step": 25076 + }, + { + "epoch": 1.4010950638321646, + "grad_norm": 1.0955965518951416, + "learning_rate": 0.0003008460331689825, + "loss": 0.4182, + "step": 25077 + }, + { + "epoch": 1.4011509344358466, + "grad_norm": 0.47204506397247314, + "learning_rate": 0.00030081801882563875, + "loss": 0.4524, + "step": 25078 + }, + { + "epoch": 1.4012068050395285, + "grad_norm": 0.44933828711509705, + "learning_rate": 0.00030079000448229496, + "loss": 0.4067, + "step": 25079 + }, + { + "epoch": 1.4012626756432103, + "grad_norm": 0.4127874970436096, + "learning_rate": 0.00030076199013895116, + "loss": 0.551, + "step": 25080 + }, + { + "epoch": 1.4013185462468922, + "grad_norm": 0.5538652539253235, + "learning_rate": 0.00030073397579560737, + "loss": 0.3024, + "step": 25081 + }, + { + "epoch": 1.4013744168505742, + "grad_norm": 0.42446768283843994, + "learning_rate": 0.0003007059614522636, + "loss": 0.4962, + "step": 25082 + }, + { + "epoch": 1.401430287454256, + "grad_norm": 0.6328679323196411, + "learning_rate": 0.0003006779471089198, + "loss": 0.3843, + "step": 25083 + }, + { + "epoch": 1.4014861580579379, + "grad_norm": 0.45018723607063293, + "learning_rate": 0.000300649932765576, + "loss": 0.381, + "step": 25084 + }, + { + "epoch": 1.4015420286616198, + "grad_norm": 0.36486929655075073, + "learning_rate": 0.0003006219184222322, + "loss": 0.3848, + "step": 25085 + }, + { + "epoch": 1.4015978992653015, + "grad_norm": 0.3322540819644928, + "learning_rate": 0.00030059390407888835, + "loss": 0.42, + "step": 25086 + }, + { + "epoch": 1.4016537698689835, + "grad_norm": 0.40913936495780945, + "learning_rate": 0.0003005658897355446, + "loss": 0.4265, + "step": 25087 + }, + { + "epoch": 1.4017096404726654, + "grad_norm": 0.5079473257064819, + "learning_rate": 0.0003005378753922008, + "loss": 0.3819, + "step": 25088 + }, + { + "epoch": 1.4017655110763472, + "grad_norm": 0.7415079474449158, + "learning_rate": 0.000300509861048857, + "loss": 0.4425, + "step": 25089 + }, + { + "epoch": 1.4018213816800291, + "grad_norm": 0.628671407699585, + "learning_rate": 0.0003004818467055132, + "loss": 0.4083, + "step": 25090 + }, + { + "epoch": 1.4018772522837109, + "grad_norm": 1.1421531438827515, + "learning_rate": 0.00030045383236216943, + "loss": 0.4358, + "step": 25091 + }, + { + "epoch": 1.4019331228873928, + "grad_norm": 0.5757814645767212, + "learning_rate": 0.00030042581801882563, + "loss": 0.5036, + "step": 25092 + }, + { + "epoch": 1.4019889934910748, + "grad_norm": 0.4398746192455292, + "learning_rate": 0.00030039780367548184, + "loss": 0.2969, + "step": 25093 + }, + { + "epoch": 1.4020448640947565, + "grad_norm": 0.4445653259754181, + "learning_rate": 0.00030036978933213805, + "loss": 0.3887, + "step": 25094 + }, + { + "epoch": 1.4021007346984384, + "grad_norm": 0.4579963684082031, + "learning_rate": 0.0003003417749887943, + "loss": 0.3837, + "step": 25095 + }, + { + "epoch": 1.4021566053021202, + "grad_norm": 0.3385140001773834, + "learning_rate": 0.00030031376064545046, + "loss": 0.3726, + "step": 25096 + }, + { + "epoch": 1.4022124759058021, + "grad_norm": 0.9967763423919678, + "learning_rate": 0.0003002857463021067, + "loss": 0.4368, + "step": 25097 + }, + { + "epoch": 1.402268346509484, + "grad_norm": 0.7004317045211792, + "learning_rate": 0.00030025773195876287, + "loss": 0.4572, + "step": 25098 + }, + { + "epoch": 1.4023242171131658, + "grad_norm": 0.3549564480781555, + "learning_rate": 0.00030022971761541913, + "loss": 0.3853, + "step": 25099 + }, + { + "epoch": 1.4023800877168477, + "grad_norm": 1.2007534503936768, + "learning_rate": 0.0003002017032720753, + "loss": 0.5011, + "step": 25100 + }, + { + "epoch": 1.4024359583205297, + "grad_norm": 0.6796981692314148, + "learning_rate": 0.00030017368892873154, + "loss": 0.4454, + "step": 25101 + }, + { + "epoch": 1.4024918289242114, + "grad_norm": 0.6401402354240417, + "learning_rate": 0.0003001456745853877, + "loss": 0.4836, + "step": 25102 + }, + { + "epoch": 1.4025476995278934, + "grad_norm": 1.5979936122894287, + "learning_rate": 0.00030011766024204395, + "loss": 0.3379, + "step": 25103 + }, + { + "epoch": 1.4026035701315753, + "grad_norm": 0.33705195784568787, + "learning_rate": 0.00030008964589870016, + "loss": 0.3094, + "step": 25104 + }, + { + "epoch": 1.402659440735257, + "grad_norm": 0.3688066005706787, + "learning_rate": 0.00030006163155535636, + "loss": 0.4805, + "step": 25105 + }, + { + "epoch": 1.402715311338939, + "grad_norm": 0.6044827699661255, + "learning_rate": 0.00030003361721201257, + "loss": 0.5095, + "step": 25106 + }, + { + "epoch": 1.402771181942621, + "grad_norm": 0.41226860880851746, + "learning_rate": 0.0003000056028686688, + "loss": 0.3885, + "step": 25107 + }, + { + "epoch": 1.4028270525463027, + "grad_norm": 0.6578903198242188, + "learning_rate": 0.000299977588525325, + "loss": 0.439, + "step": 25108 + }, + { + "epoch": 1.4028829231499846, + "grad_norm": 0.47292742133140564, + "learning_rate": 0.00029994957418198113, + "loss": 0.5118, + "step": 25109 + }, + { + "epoch": 1.4029387937536666, + "grad_norm": 0.4324246048927307, + "learning_rate": 0.0002999215598386374, + "loss": 0.3469, + "step": 25110 + }, + { + "epoch": 1.4029946643573483, + "grad_norm": 0.5991880297660828, + "learning_rate": 0.0002998935454952936, + "loss": 0.444, + "step": 25111 + }, + { + "epoch": 1.4030505349610303, + "grad_norm": 0.3820249140262604, + "learning_rate": 0.0002998655311519498, + "loss": 0.3758, + "step": 25112 + }, + { + "epoch": 1.4031064055647122, + "grad_norm": 0.5816817879676819, + "learning_rate": 0.000299837516808606, + "loss": 0.3465, + "step": 25113 + }, + { + "epoch": 1.403162276168394, + "grad_norm": 0.4065574109554291, + "learning_rate": 0.0002998095024652622, + "loss": 0.4061, + "step": 25114 + }, + { + "epoch": 1.403218146772076, + "grad_norm": 6.2636919021606445, + "learning_rate": 0.0002997814881219184, + "loss": 0.2839, + "step": 25115 + }, + { + "epoch": 1.4032740173757579, + "grad_norm": 0.9023560881614685, + "learning_rate": 0.00029975347377857463, + "loss": 0.3719, + "step": 25116 + }, + { + "epoch": 1.4033298879794396, + "grad_norm": 0.8885036110877991, + "learning_rate": 0.00029972545943523083, + "loss": 0.3754, + "step": 25117 + }, + { + "epoch": 1.4033857585831215, + "grad_norm": 0.5798556208610535, + "learning_rate": 0.00029969744509188704, + "loss": 0.4152, + "step": 25118 + }, + { + "epoch": 1.4034416291868035, + "grad_norm": 0.5757834315299988, + "learning_rate": 0.00029966943074854325, + "loss": 0.4738, + "step": 25119 + }, + { + "epoch": 1.4034974997904852, + "grad_norm": 0.6275804042816162, + "learning_rate": 0.0002996414164051995, + "loss": 0.372, + "step": 25120 + }, + { + "epoch": 1.4035533703941672, + "grad_norm": 0.3119231164455414, + "learning_rate": 0.00029961340206185566, + "loss": 0.3797, + "step": 25121 + }, + { + "epoch": 1.4036092409978491, + "grad_norm": 0.38687002658843994, + "learning_rate": 0.0002995853877185119, + "loss": 0.5036, + "step": 25122 + }, + { + "epoch": 1.4036651116015308, + "grad_norm": 0.3970722556114197, + "learning_rate": 0.00029955737337516807, + "loss": 0.4172, + "step": 25123 + }, + { + "epoch": 1.4037209822052128, + "grad_norm": 0.7774612903594971, + "learning_rate": 0.00029952935903182433, + "loss": 0.4021, + "step": 25124 + }, + { + "epoch": 1.4037768528088945, + "grad_norm": 0.49840083718299866, + "learning_rate": 0.0002995013446884805, + "loss": 0.4162, + "step": 25125 + }, + { + "epoch": 1.4038327234125765, + "grad_norm": 2.1427741050720215, + "learning_rate": 0.00029947333034513674, + "loss": 0.386, + "step": 25126 + }, + { + "epoch": 1.4038885940162584, + "grad_norm": 0.48059675097465515, + "learning_rate": 0.00029944531600179295, + "loss": 0.3999, + "step": 25127 + }, + { + "epoch": 1.4039444646199402, + "grad_norm": 0.491432249546051, + "learning_rate": 0.00029941730165844915, + "loss": 0.3384, + "step": 25128 + }, + { + "epoch": 1.404000335223622, + "grad_norm": 0.4887419641017914, + "learning_rate": 0.00029938928731510536, + "loss": 0.4428, + "step": 25129 + }, + { + "epoch": 1.4040562058273038, + "grad_norm": 0.683412492275238, + "learning_rate": 0.0002993612729717615, + "loss": 0.495, + "step": 25130 + }, + { + "epoch": 1.4041120764309858, + "grad_norm": 0.4753853678703308, + "learning_rate": 0.00029933325862841777, + "loss": 0.3353, + "step": 25131 + }, + { + "epoch": 1.4041679470346677, + "grad_norm": 0.5847434997558594, + "learning_rate": 0.0002993052442850739, + "loss": 0.4934, + "step": 25132 + }, + { + "epoch": 1.4042238176383495, + "grad_norm": 0.36977583169937134, + "learning_rate": 0.0002992772299417302, + "loss": 0.3594, + "step": 25133 + }, + { + "epoch": 1.4042796882420314, + "grad_norm": 0.4121890068054199, + "learning_rate": 0.00029924921559838633, + "loss": 0.4702, + "step": 25134 + }, + { + "epoch": 1.4043355588457134, + "grad_norm": 0.49399226903915405, + "learning_rate": 0.0002992212012550426, + "loss": 0.5006, + "step": 25135 + }, + { + "epoch": 1.404391429449395, + "grad_norm": 0.5308624505996704, + "learning_rate": 0.0002991931869116988, + "loss": 0.4051, + "step": 25136 + }, + { + "epoch": 1.404447300053077, + "grad_norm": 1.136225700378418, + "learning_rate": 0.000299165172568355, + "loss": 0.4247, + "step": 25137 + }, + { + "epoch": 1.404503170656759, + "grad_norm": 0.8912349939346313, + "learning_rate": 0.0002991371582250112, + "loss": 0.4207, + "step": 25138 + }, + { + "epoch": 1.4045590412604407, + "grad_norm": 0.42600852251052856, + "learning_rate": 0.0002991091438816674, + "loss": 0.3232, + "step": 25139 + }, + { + "epoch": 1.4046149118641227, + "grad_norm": 0.5082436800003052, + "learning_rate": 0.0002990811295383236, + "loss": 0.435, + "step": 25140 + }, + { + "epoch": 1.4046707824678046, + "grad_norm": 0.42656606435775757, + "learning_rate": 0.00029905311519497983, + "loss": 0.4238, + "step": 25141 + }, + { + "epoch": 1.4047266530714864, + "grad_norm": 0.8770961761474609, + "learning_rate": 0.00029902510085163603, + "loss": 0.4306, + "step": 25142 + }, + { + "epoch": 1.4047825236751683, + "grad_norm": 0.4579281508922577, + "learning_rate": 0.0002989970865082923, + "loss": 0.4547, + "step": 25143 + }, + { + "epoch": 1.4048383942788503, + "grad_norm": 0.538448691368103, + "learning_rate": 0.00029896907216494845, + "loss": 0.3754, + "step": 25144 + }, + { + "epoch": 1.404894264882532, + "grad_norm": 1.2509400844573975, + "learning_rate": 0.0002989410578216047, + "loss": 0.4437, + "step": 25145 + }, + { + "epoch": 1.404950135486214, + "grad_norm": 0.47208601236343384, + "learning_rate": 0.00029891304347826086, + "loss": 0.425, + "step": 25146 + }, + { + "epoch": 1.405006006089896, + "grad_norm": 0.42725035548210144, + "learning_rate": 0.0002988850291349171, + "loss": 0.4011, + "step": 25147 + }, + { + "epoch": 1.4050618766935776, + "grad_norm": 0.3421425223350525, + "learning_rate": 0.00029885701479157327, + "loss": 0.3907, + "step": 25148 + }, + { + "epoch": 1.4051177472972596, + "grad_norm": 0.4268074631690979, + "learning_rate": 0.00029882900044822953, + "loss": 0.5222, + "step": 25149 + }, + { + "epoch": 1.4051736179009415, + "grad_norm": 0.42771658301353455, + "learning_rate": 0.00029880098610488573, + "loss": 0.4474, + "step": 25150 + }, + { + "epoch": 1.4052294885046233, + "grad_norm": 0.4449639916419983, + "learning_rate": 0.0002987729717615419, + "loss": 0.4699, + "step": 25151 + }, + { + "epoch": 1.4052853591083052, + "grad_norm": 0.40266236662864685, + "learning_rate": 0.00029874495741819815, + "loss": 0.3341, + "step": 25152 + }, + { + "epoch": 1.4053412297119872, + "grad_norm": 0.43466171622276306, + "learning_rate": 0.0002987169430748543, + "loss": 0.37, + "step": 25153 + }, + { + "epoch": 1.405397100315669, + "grad_norm": 0.6099838614463806, + "learning_rate": 0.00029868892873151056, + "loss": 0.4726, + "step": 25154 + }, + { + "epoch": 1.4054529709193508, + "grad_norm": 0.30465689301490784, + "learning_rate": 0.0002986609143881667, + "loss": 0.4125, + "step": 25155 + }, + { + "epoch": 1.4055088415230328, + "grad_norm": 2.115309238433838, + "learning_rate": 0.00029863290004482297, + "loss": 0.3391, + "step": 25156 + }, + { + "epoch": 1.4055647121267145, + "grad_norm": 0.31397542357444763, + "learning_rate": 0.0002986048857014791, + "loss": 0.383, + "step": 25157 + }, + { + "epoch": 1.4056205827303965, + "grad_norm": 0.38147634267807007, + "learning_rate": 0.0002985768713581354, + "loss": 0.4084, + "step": 25158 + }, + { + "epoch": 1.4056764533340782, + "grad_norm": 0.6167344450950623, + "learning_rate": 0.0002985488570147916, + "loss": 0.4469, + "step": 25159 + }, + { + "epoch": 1.4057323239377602, + "grad_norm": 0.36887264251708984, + "learning_rate": 0.0002985208426714478, + "loss": 0.3881, + "step": 25160 + }, + { + "epoch": 1.4057881945414419, + "grad_norm": 0.3033626675605774, + "learning_rate": 0.000298492828328104, + "loss": 0.4171, + "step": 25161 + }, + { + "epoch": 1.4058440651451238, + "grad_norm": 0.3924283981323242, + "learning_rate": 0.0002984648139847602, + "loss": 0.4495, + "step": 25162 + }, + { + "epoch": 1.4058999357488058, + "grad_norm": 0.30496537685394287, + "learning_rate": 0.0002984367996414164, + "loss": 0.4428, + "step": 25163 + }, + { + "epoch": 1.4059558063524875, + "grad_norm": 0.3981260657310486, + "learning_rate": 0.0002984087852980726, + "loss": 0.4476, + "step": 25164 + }, + { + "epoch": 1.4060116769561695, + "grad_norm": 0.4186633229255676, + "learning_rate": 0.0002983807709547288, + "loss": 0.5545, + "step": 25165 + }, + { + "epoch": 1.4060675475598514, + "grad_norm": 0.4967566728591919, + "learning_rate": 0.0002983527566113851, + "loss": 0.4144, + "step": 25166 + }, + { + "epoch": 1.4061234181635331, + "grad_norm": 3.167351722717285, + "learning_rate": 0.00029832474226804123, + "loss": 0.4041, + "step": 25167 + }, + { + "epoch": 1.406179288767215, + "grad_norm": 0.5566815733909607, + "learning_rate": 0.0002982967279246975, + "loss": 0.4334, + "step": 25168 + }, + { + "epoch": 1.406235159370897, + "grad_norm": 1.6940306425094604, + "learning_rate": 0.00029826871358135364, + "loss": 0.4265, + "step": 25169 + }, + { + "epoch": 1.4062910299745788, + "grad_norm": 0.33249741792678833, + "learning_rate": 0.0002982406992380099, + "loss": 0.4638, + "step": 25170 + }, + { + "epoch": 1.4063469005782607, + "grad_norm": 0.581244945526123, + "learning_rate": 0.00029821268489466606, + "loss": 0.4283, + "step": 25171 + }, + { + "epoch": 1.4064027711819427, + "grad_norm": 0.4349924623966217, + "learning_rate": 0.00029818467055132226, + "loss": 0.4247, + "step": 25172 + }, + { + "epoch": 1.4064586417856244, + "grad_norm": 0.43883100152015686, + "learning_rate": 0.00029815665620797847, + "loss": 0.456, + "step": 25173 + }, + { + "epoch": 1.4065145123893064, + "grad_norm": 1.0891770124435425, + "learning_rate": 0.0002981286418646347, + "loss": 0.4108, + "step": 25174 + }, + { + "epoch": 1.4065703829929883, + "grad_norm": 0.3191690146923065, + "learning_rate": 0.00029810062752129093, + "loss": 0.4118, + "step": 25175 + }, + { + "epoch": 1.40662625359667, + "grad_norm": 0.5398943424224854, + "learning_rate": 0.0002980726131779471, + "loss": 0.3896, + "step": 25176 + }, + { + "epoch": 1.406682124200352, + "grad_norm": 0.3924348056316376, + "learning_rate": 0.00029804459883460335, + "loss": 0.4947, + "step": 25177 + }, + { + "epoch": 1.406737994804034, + "grad_norm": 1.6702051162719727, + "learning_rate": 0.0002980165844912595, + "loss": 0.5856, + "step": 25178 + }, + { + "epoch": 1.4067938654077157, + "grad_norm": 1.1574705839157104, + "learning_rate": 0.00029798857014791576, + "loss": 0.5634, + "step": 25179 + }, + { + "epoch": 1.4068497360113976, + "grad_norm": 0.41054028272628784, + "learning_rate": 0.0002979605558045719, + "loss": 0.4395, + "step": 25180 + }, + { + "epoch": 1.4069056066150796, + "grad_norm": 0.3366006314754486, + "learning_rate": 0.00029793254146122817, + "loss": 0.3315, + "step": 25181 + }, + { + "epoch": 1.4069614772187613, + "grad_norm": 0.4928637444972992, + "learning_rate": 0.0002979045271178844, + "loss": 0.3735, + "step": 25182 + }, + { + "epoch": 1.4070173478224433, + "grad_norm": 0.40383803844451904, + "learning_rate": 0.0002978765127745406, + "loss": 0.3714, + "step": 25183 + }, + { + "epoch": 1.4070732184261252, + "grad_norm": 0.3480403423309326, + "learning_rate": 0.0002978484984311968, + "loss": 0.4115, + "step": 25184 + }, + { + "epoch": 1.407129089029807, + "grad_norm": 0.596836507320404, + "learning_rate": 0.000297820484087853, + "loss": 0.373, + "step": 25185 + }, + { + "epoch": 1.4071849596334889, + "grad_norm": 0.4613744914531708, + "learning_rate": 0.0002977924697445092, + "loss": 0.3561, + "step": 25186 + }, + { + "epoch": 1.4072408302371708, + "grad_norm": 0.735103189945221, + "learning_rate": 0.0002977644554011654, + "loss": 0.4648, + "step": 25187 + }, + { + "epoch": 1.4072967008408526, + "grad_norm": 0.40909186005592346, + "learning_rate": 0.0002977364410578216, + "loss": 0.4336, + "step": 25188 + }, + { + "epoch": 1.4073525714445345, + "grad_norm": 0.32672134041786194, + "learning_rate": 0.0002977084267144778, + "loss": 0.4025, + "step": 25189 + }, + { + "epoch": 1.4074084420482165, + "grad_norm": 3.275927782058716, + "learning_rate": 0.000297680412371134, + "loss": 0.3569, + "step": 25190 + }, + { + "epoch": 1.4074643126518982, + "grad_norm": 0.5139309167861938, + "learning_rate": 0.0002976523980277903, + "loss": 0.5081, + "step": 25191 + }, + { + "epoch": 1.4075201832555801, + "grad_norm": 0.6199195981025696, + "learning_rate": 0.00029762438368444643, + "loss": 0.4084, + "step": 25192 + }, + { + "epoch": 1.4075760538592619, + "grad_norm": 0.601423978805542, + "learning_rate": 0.00029759636934110264, + "loss": 0.3678, + "step": 25193 + }, + { + "epoch": 1.4076319244629438, + "grad_norm": 0.36335325241088867, + "learning_rate": 0.00029756835499775884, + "loss": 0.3692, + "step": 25194 + }, + { + "epoch": 1.4076877950666256, + "grad_norm": 0.7380579710006714, + "learning_rate": 0.00029754034065441505, + "loss": 0.4428, + "step": 25195 + }, + { + "epoch": 1.4077436656703075, + "grad_norm": 0.4872835576534271, + "learning_rate": 0.00029751232631107126, + "loss": 0.3101, + "step": 25196 + }, + { + "epoch": 1.4077995362739895, + "grad_norm": 0.44083118438720703, + "learning_rate": 0.00029748431196772746, + "loss": 0.4295, + "step": 25197 + }, + { + "epoch": 1.4078554068776712, + "grad_norm": 0.46854788064956665, + "learning_rate": 0.0002974562976243837, + "loss": 0.3767, + "step": 25198 + }, + { + "epoch": 1.4079112774813531, + "grad_norm": 1.6823570728302002, + "learning_rate": 0.0002974282832810399, + "loss": 0.4615, + "step": 25199 + }, + { + "epoch": 1.407967148085035, + "grad_norm": 0.4369940459728241, + "learning_rate": 0.00029740026893769613, + "loss": 0.4003, + "step": 25200 + }, + { + "epoch": 1.4080230186887168, + "grad_norm": 0.40352103114128113, + "learning_rate": 0.0002973722545943523, + "loss": 0.392, + "step": 25201 + }, + { + "epoch": 1.4080788892923988, + "grad_norm": 0.6673204302787781, + "learning_rate": 0.00029734424025100855, + "loss": 0.411, + "step": 25202 + }, + { + "epoch": 1.4081347598960807, + "grad_norm": 0.4723840653896332, + "learning_rate": 0.0002973162259076647, + "loss": 0.3695, + "step": 25203 + }, + { + "epoch": 1.4081906304997625, + "grad_norm": 0.3060612380504608, + "learning_rate": 0.00029728821156432096, + "loss": 0.3371, + "step": 25204 + }, + { + "epoch": 1.4082465011034444, + "grad_norm": 1.1576935052871704, + "learning_rate": 0.0002972601972209771, + "loss": 0.396, + "step": 25205 + }, + { + "epoch": 1.4083023717071264, + "grad_norm": 3.4371888637542725, + "learning_rate": 0.00029723218287763337, + "loss": 0.576, + "step": 25206 + }, + { + "epoch": 1.408358242310808, + "grad_norm": 4.898423671722412, + "learning_rate": 0.0002972041685342896, + "loss": 0.4847, + "step": 25207 + }, + { + "epoch": 1.40841411291449, + "grad_norm": 0.3769655227661133, + "learning_rate": 0.0002971761541909458, + "loss": 0.3906, + "step": 25208 + }, + { + "epoch": 1.408469983518172, + "grad_norm": 4.159917831420898, + "learning_rate": 0.000297148139847602, + "loss": 0.4123, + "step": 25209 + }, + { + "epoch": 1.4085258541218537, + "grad_norm": 0.5102370977401733, + "learning_rate": 0.0002971201255042582, + "loss": 0.3402, + "step": 25210 + }, + { + "epoch": 1.4085817247255357, + "grad_norm": 0.8205966949462891, + "learning_rate": 0.0002970921111609144, + "loss": 0.5335, + "step": 25211 + }, + { + "epoch": 1.4086375953292176, + "grad_norm": 2.2895541191101074, + "learning_rate": 0.0002970640968175706, + "loss": 0.3295, + "step": 25212 + }, + { + "epoch": 1.4086934659328993, + "grad_norm": 0.4740789532661438, + "learning_rate": 0.0002970360824742268, + "loss": 0.4009, + "step": 25213 + }, + { + "epoch": 1.4087493365365813, + "grad_norm": 0.3984726071357727, + "learning_rate": 0.000297008068130883, + "loss": 0.3515, + "step": 25214 + }, + { + "epoch": 1.4088052071402632, + "grad_norm": 0.39429280161857605, + "learning_rate": 0.0002969800537875392, + "loss": 0.3777, + "step": 25215 + }, + { + "epoch": 1.408861077743945, + "grad_norm": 0.4409060776233673, + "learning_rate": 0.0002969520394441954, + "loss": 0.3815, + "step": 25216 + }, + { + "epoch": 1.408916948347627, + "grad_norm": 0.40877828001976013, + "learning_rate": 0.00029692402510085163, + "loss": 0.4314, + "step": 25217 + }, + { + "epoch": 1.4089728189513089, + "grad_norm": 1.7531801462173462, + "learning_rate": 0.00029689601075750784, + "loss": 0.3247, + "step": 25218 + }, + { + "epoch": 1.4090286895549906, + "grad_norm": 0.32984256744384766, + "learning_rate": 0.00029686799641416404, + "loss": 0.3799, + "step": 25219 + }, + { + "epoch": 1.4090845601586726, + "grad_norm": 0.3579524755477905, + "learning_rate": 0.00029683998207082025, + "loss": 0.4367, + "step": 25220 + }, + { + "epoch": 1.4091404307623545, + "grad_norm": 0.4236724078655243, + "learning_rate": 0.0002968119677274765, + "loss": 0.4577, + "step": 25221 + }, + { + "epoch": 1.4091963013660362, + "grad_norm": 9.319859504699707, + "learning_rate": 0.00029678395338413266, + "loss": 0.4019, + "step": 25222 + }, + { + "epoch": 1.4092521719697182, + "grad_norm": 0.3541973829269409, + "learning_rate": 0.0002967559390407889, + "loss": 0.3876, + "step": 25223 + }, + { + "epoch": 1.4093080425734001, + "grad_norm": 0.6266674995422363, + "learning_rate": 0.0002967279246974451, + "loss": 0.2599, + "step": 25224 + }, + { + "epoch": 1.4093639131770819, + "grad_norm": 0.4076622426509857, + "learning_rate": 0.00029669991035410133, + "loss": 0.4827, + "step": 25225 + }, + { + "epoch": 1.4094197837807638, + "grad_norm": 0.5092754364013672, + "learning_rate": 0.0002966718960107575, + "loss": 0.3652, + "step": 25226 + }, + { + "epoch": 1.4094756543844456, + "grad_norm": 0.5385376214981079, + "learning_rate": 0.00029664388166741374, + "loss": 0.5766, + "step": 25227 + }, + { + "epoch": 1.4095315249881275, + "grad_norm": 0.46156376600265503, + "learning_rate": 0.0002966158673240699, + "loss": 0.4528, + "step": 25228 + }, + { + "epoch": 1.4095873955918092, + "grad_norm": 0.39187049865722656, + "learning_rate": 0.00029658785298072616, + "loss": 0.3923, + "step": 25229 + }, + { + "epoch": 1.4096432661954912, + "grad_norm": 0.7072697281837463, + "learning_rate": 0.00029655983863738236, + "loss": 0.4883, + "step": 25230 + }, + { + "epoch": 1.4096991367991731, + "grad_norm": 0.41067060828208923, + "learning_rate": 0.00029653182429403857, + "loss": 0.3933, + "step": 25231 + }, + { + "epoch": 1.4097550074028549, + "grad_norm": 0.43857648968696594, + "learning_rate": 0.0002965038099506948, + "loss": 0.3756, + "step": 25232 + }, + { + "epoch": 1.4098108780065368, + "grad_norm": 0.455666184425354, + "learning_rate": 0.000296475795607351, + "loss": 0.3679, + "step": 25233 + }, + { + "epoch": 1.4098667486102188, + "grad_norm": 0.5868924260139465, + "learning_rate": 0.0002964477812640072, + "loss": 0.4603, + "step": 25234 + }, + { + "epoch": 1.4099226192139005, + "grad_norm": 0.6364417672157288, + "learning_rate": 0.00029641976692066334, + "loss": 0.3668, + "step": 25235 + }, + { + "epoch": 1.4099784898175824, + "grad_norm": 0.3583075702190399, + "learning_rate": 0.0002963917525773196, + "loss": 0.3626, + "step": 25236 + }, + { + "epoch": 1.4100343604212644, + "grad_norm": 0.518810510635376, + "learning_rate": 0.0002963637382339758, + "loss": 0.4977, + "step": 25237 + }, + { + "epoch": 1.4100902310249461, + "grad_norm": 0.464847594499588, + "learning_rate": 0.000296335723890632, + "loss": 0.4933, + "step": 25238 + }, + { + "epoch": 1.410146101628628, + "grad_norm": 0.5794022083282471, + "learning_rate": 0.0002963077095472882, + "loss": 0.2786, + "step": 25239 + }, + { + "epoch": 1.41020197223231, + "grad_norm": 0.4520609974861145, + "learning_rate": 0.0002962796952039444, + "loss": 0.5423, + "step": 25240 + }, + { + "epoch": 1.4102578428359918, + "grad_norm": 0.516660749912262, + "learning_rate": 0.0002962516808606006, + "loss": 0.4312, + "step": 25241 + }, + { + "epoch": 1.4103137134396737, + "grad_norm": 0.6438189744949341, + "learning_rate": 0.00029622366651725683, + "loss": 0.6561, + "step": 25242 + }, + { + "epoch": 1.4103695840433557, + "grad_norm": Infinity, + "learning_rate": 0.00029622366651725683, + "loss": 0.4827, + "step": 25243 + }, + { + "epoch": 1.4104254546470374, + "grad_norm": 0.4525085389614105, + "learning_rate": 0.00029619565217391304, + "loss": 0.3932, + "step": 25244 + }, + { + "epoch": 1.4104813252507193, + "grad_norm": 0.3550153374671936, + "learning_rate": 0.00029616763783056924, + "loss": 0.3094, + "step": 25245 + }, + { + "epoch": 1.4105371958544013, + "grad_norm": 0.5975397825241089, + "learning_rate": 0.00029613962348722545, + "loss": 0.4218, + "step": 25246 + }, + { + "epoch": 1.410593066458083, + "grad_norm": 0.44618692994117737, + "learning_rate": 0.0002961116091438817, + "loss": 0.3799, + "step": 25247 + }, + { + "epoch": 1.410648937061765, + "grad_norm": 0.434089332818985, + "learning_rate": 0.00029608359480053786, + "loss": 0.4532, + "step": 25248 + }, + { + "epoch": 1.410704807665447, + "grad_norm": 0.37324076890945435, + "learning_rate": 0.0002960555804571941, + "loss": 0.4923, + "step": 25249 + }, + { + "epoch": 1.4107606782691287, + "grad_norm": 0.43275490403175354, + "learning_rate": 0.00029602756611385027, + "loss": 0.3116, + "step": 25250 + }, + { + "epoch": 1.4108165488728106, + "grad_norm": 0.5941612720489502, + "learning_rate": 0.00029599955177050653, + "loss": 0.486, + "step": 25251 + }, + { + "epoch": 1.4108724194764926, + "grad_norm": 0.40476927161216736, + "learning_rate": 0.0002959715374271627, + "loss": 0.4533, + "step": 25252 + }, + { + "epoch": 1.4109282900801743, + "grad_norm": 0.4684789776802063, + "learning_rate": 0.00029594352308381894, + "loss": 0.4065, + "step": 25253 + }, + { + "epoch": 1.4109841606838562, + "grad_norm": 0.8710064888000488, + "learning_rate": 0.00029591550874047515, + "loss": 0.7943, + "step": 25254 + }, + { + "epoch": 1.4110400312875382, + "grad_norm": 0.48292461037635803, + "learning_rate": 0.00029588749439713136, + "loss": 0.3952, + "step": 25255 + }, + { + "epoch": 1.41109590189122, + "grad_norm": 0.4828588664531708, + "learning_rate": 0.00029585948005378756, + "loss": 0.454, + "step": 25256 + }, + { + "epoch": 1.4111517724949019, + "grad_norm": 0.8257144093513489, + "learning_rate": 0.0002958314657104437, + "loss": 0.4172, + "step": 25257 + }, + { + "epoch": 1.4112076430985836, + "grad_norm": 0.4873869717121124, + "learning_rate": 0.0002958034513671, + "loss": 0.3575, + "step": 25258 + }, + { + "epoch": 1.4112635137022655, + "grad_norm": 0.5339638590812683, + "learning_rate": 0.0002957754370237561, + "loss": 0.4362, + "step": 25259 + }, + { + "epoch": 1.4113193843059475, + "grad_norm": 0.4832000732421875, + "learning_rate": 0.0002957474226804124, + "loss": 0.4637, + "step": 25260 + }, + { + "epoch": 1.4113752549096292, + "grad_norm": 1.832234263420105, + "learning_rate": 0.00029571940833706854, + "loss": 0.509, + "step": 25261 + }, + { + "epoch": 1.4114311255133112, + "grad_norm": 0.4459955394268036, + "learning_rate": 0.0002956913939937248, + "loss": 0.4492, + "step": 25262 + }, + { + "epoch": 1.411486996116993, + "grad_norm": 0.5033721923828125, + "learning_rate": 0.000295663379650381, + "loss": 0.4852, + "step": 25263 + }, + { + "epoch": 1.4115428667206749, + "grad_norm": 0.9396177530288696, + "learning_rate": 0.0002956353653070372, + "loss": 0.4073, + "step": 25264 + }, + { + "epoch": 1.4115987373243568, + "grad_norm": 0.546764075756073, + "learning_rate": 0.0002956073509636934, + "loss": 0.4143, + "step": 25265 + }, + { + "epoch": 1.4116546079280385, + "grad_norm": 0.47128617763519287, + "learning_rate": 0.0002955793366203496, + "loss": 0.4385, + "step": 25266 + }, + { + "epoch": 1.4117104785317205, + "grad_norm": 0.4672614634037018, + "learning_rate": 0.0002955513222770058, + "loss": 0.5842, + "step": 25267 + }, + { + "epoch": 1.4117663491354024, + "grad_norm": 2.1540591716766357, + "learning_rate": 0.00029552330793366203, + "loss": 0.5215, + "step": 25268 + }, + { + "epoch": 1.4118222197390842, + "grad_norm": 2.698937177658081, + "learning_rate": 0.00029549529359031824, + "loss": 0.4846, + "step": 25269 + }, + { + "epoch": 1.4118780903427661, + "grad_norm": 0.5223460793495178, + "learning_rate": 0.0002954672792469745, + "loss": 0.4847, + "step": 25270 + }, + { + "epoch": 1.411933960946448, + "grad_norm": 0.662784993648529, + "learning_rate": 0.00029543926490363065, + "loss": 0.6816, + "step": 25271 + }, + { + "epoch": 1.4119898315501298, + "grad_norm": 0.43286046385765076, + "learning_rate": 0.0002954112505602869, + "loss": 0.423, + "step": 25272 + }, + { + "epoch": 1.4120457021538118, + "grad_norm": 0.4523671567440033, + "learning_rate": 0.00029538323621694306, + "loss": 0.4091, + "step": 25273 + }, + { + "epoch": 1.4121015727574937, + "grad_norm": 0.4348929822444916, + "learning_rate": 0.0002953552218735993, + "loss": 0.418, + "step": 25274 + }, + { + "epoch": 1.4121574433611754, + "grad_norm": 1.0836001634597778, + "learning_rate": 0.00029532720753025547, + "loss": 0.4392, + "step": 25275 + }, + { + "epoch": 1.4122133139648574, + "grad_norm": 0.3758975863456726, + "learning_rate": 0.00029529919318691173, + "loss": 0.417, + "step": 25276 + }, + { + "epoch": 1.4122691845685393, + "grad_norm": 0.5048877596855164, + "learning_rate": 0.00029527117884356794, + "loss": 0.4361, + "step": 25277 + }, + { + "epoch": 1.412325055172221, + "grad_norm": 2.0892817974090576, + "learning_rate": 0.00029524316450022414, + "loss": 0.4566, + "step": 25278 + }, + { + "epoch": 1.412380925775903, + "grad_norm": 0.4856647551059723, + "learning_rate": 0.00029521515015688035, + "loss": 0.4151, + "step": 25279 + }, + { + "epoch": 1.412436796379585, + "grad_norm": 0.69875568151474, + "learning_rate": 0.0002951871358135365, + "loss": 0.3837, + "step": 25280 + }, + { + "epoch": 1.4124926669832667, + "grad_norm": 0.3520853817462921, + "learning_rate": 0.00029515912147019276, + "loss": 0.4022, + "step": 25281 + }, + { + "epoch": 1.4125485375869486, + "grad_norm": 0.3322727084159851, + "learning_rate": 0.0002951311071268489, + "loss": 0.373, + "step": 25282 + }, + { + "epoch": 1.4126044081906306, + "grad_norm": 1.617799162864685, + "learning_rate": 0.0002951030927835052, + "loss": 0.5363, + "step": 25283 + }, + { + "epoch": 1.4126602787943123, + "grad_norm": 0.8305922150611877, + "learning_rate": 0.0002950750784401613, + "loss": 0.4014, + "step": 25284 + }, + { + "epoch": 1.4127161493979943, + "grad_norm": 0.4150320291519165, + "learning_rate": 0.0002950470640968176, + "loss": 0.4711, + "step": 25285 + }, + { + "epoch": 1.4127720200016762, + "grad_norm": 0.4078699052333832, + "learning_rate": 0.0002950190497534738, + "loss": 0.3428, + "step": 25286 + }, + { + "epoch": 1.412827890605358, + "grad_norm": 0.7741267681121826, + "learning_rate": 0.00029499103541013, + "loss": 0.3948, + "step": 25287 + }, + { + "epoch": 1.41288376120904, + "grad_norm": 0.33236879110336304, + "learning_rate": 0.0002949630210667862, + "loss": 0.4083, + "step": 25288 + }, + { + "epoch": 1.4129396318127219, + "grad_norm": 0.6009236574172974, + "learning_rate": 0.0002949350067234424, + "loss": 0.4159, + "step": 25289 + }, + { + "epoch": 1.4129955024164036, + "grad_norm": 0.5308520197868347, + "learning_rate": 0.0002949069923800986, + "loss": 0.398, + "step": 25290 + }, + { + "epoch": 1.4130513730200855, + "grad_norm": 0.7535595893859863, + "learning_rate": 0.0002948789780367548, + "loss": 0.3895, + "step": 25291 + }, + { + "epoch": 1.4131072436237673, + "grad_norm": 0.47969672083854675, + "learning_rate": 0.000294850963693411, + "loss": 0.4723, + "step": 25292 + }, + { + "epoch": 1.4131631142274492, + "grad_norm": 0.38218480348587036, + "learning_rate": 0.0002948229493500673, + "loss": 0.4489, + "step": 25293 + }, + { + "epoch": 1.4132189848311312, + "grad_norm": 0.918692409992218, + "learning_rate": 0.00029479493500672344, + "loss": 0.4621, + "step": 25294 + }, + { + "epoch": 1.413274855434813, + "grad_norm": 1.6545624732971191, + "learning_rate": 0.0002947669206633797, + "loss": 0.3802, + "step": 25295 + }, + { + "epoch": 1.4133307260384949, + "grad_norm": 0.4619746804237366, + "learning_rate": 0.00029473890632003585, + "loss": 0.4241, + "step": 25296 + }, + { + "epoch": 1.4133865966421766, + "grad_norm": 0.8072317242622375, + "learning_rate": 0.0002947108919766921, + "loss": 0.4866, + "step": 25297 + }, + { + "epoch": 1.4134424672458585, + "grad_norm": 0.4704931974411011, + "learning_rate": 0.00029468287763334826, + "loss": 0.4388, + "step": 25298 + }, + { + "epoch": 1.4134983378495405, + "grad_norm": 0.4008505642414093, + "learning_rate": 0.0002946548632900045, + "loss": 0.3844, + "step": 25299 + }, + { + "epoch": 1.4135542084532222, + "grad_norm": 0.45994922518730164, + "learning_rate": 0.00029462684894666067, + "loss": 0.3613, + "step": 25300 + }, + { + "epoch": 1.4136100790569042, + "grad_norm": 0.49017399549484253, + "learning_rate": 0.0002945988346033169, + "loss": 0.4159, + "step": 25301 + }, + { + "epoch": 1.4136659496605861, + "grad_norm": 0.41449323296546936, + "learning_rate": 0.00029457082025997314, + "loss": 0.4384, + "step": 25302 + }, + { + "epoch": 1.4137218202642678, + "grad_norm": 0.35741615295410156, + "learning_rate": 0.0002945428059166293, + "loss": 0.31, + "step": 25303 + }, + { + "epoch": 1.4137776908679498, + "grad_norm": 0.4889402389526367, + "learning_rate": 0.00029451479157328555, + "loss": 0.4603, + "step": 25304 + }, + { + "epoch": 1.4138335614716318, + "grad_norm": 0.4618699550628662, + "learning_rate": 0.0002944867772299417, + "loss": 0.3809, + "step": 25305 + }, + { + "epoch": 1.4138894320753135, + "grad_norm": 0.4635239839553833, + "learning_rate": 0.00029445876288659796, + "loss": 0.4099, + "step": 25306 + }, + { + "epoch": 1.4139453026789954, + "grad_norm": 0.45996442437171936, + "learning_rate": 0.0002944307485432541, + "loss": 0.3729, + "step": 25307 + }, + { + "epoch": 1.4140011732826774, + "grad_norm": 0.3698520064353943, + "learning_rate": 0.00029440273419991037, + "loss": 0.3429, + "step": 25308 + }, + { + "epoch": 1.414057043886359, + "grad_norm": 3.251147747039795, + "learning_rate": 0.0002943747198565666, + "loss": 0.6189, + "step": 25309 + }, + { + "epoch": 1.414112914490041, + "grad_norm": 0.6577363014221191, + "learning_rate": 0.0002943467055132228, + "loss": 0.4897, + "step": 25310 + }, + { + "epoch": 1.414168785093723, + "grad_norm": 0.48485827445983887, + "learning_rate": 0.000294318691169879, + "loss": 0.5134, + "step": 25311 + }, + { + "epoch": 1.4142246556974047, + "grad_norm": 0.5219417810440063, + "learning_rate": 0.0002942906768265352, + "loss": 0.4443, + "step": 25312 + }, + { + "epoch": 1.4142805263010867, + "grad_norm": 0.5322735905647278, + "learning_rate": 0.0002942626624831914, + "loss": 0.3914, + "step": 25313 + }, + { + "epoch": 1.4143363969047686, + "grad_norm": 0.450514554977417, + "learning_rate": 0.0002942346481398476, + "loss": 0.4026, + "step": 25314 + }, + { + "epoch": 1.4143922675084504, + "grad_norm": 0.5256046652793884, + "learning_rate": 0.0002942066337965038, + "loss": 0.4891, + "step": 25315 + }, + { + "epoch": 1.4144481381121323, + "grad_norm": 0.4611338973045349, + "learning_rate": 0.00029417861945316, + "loss": 0.4082, + "step": 25316 + }, + { + "epoch": 1.4145040087158143, + "grad_norm": 0.29622581601142883, + "learning_rate": 0.0002941506051098162, + "loss": 0.3684, + "step": 25317 + }, + { + "epoch": 1.414559879319496, + "grad_norm": 0.6681001782417297, + "learning_rate": 0.0002941225907664725, + "loss": 0.417, + "step": 25318 + }, + { + "epoch": 1.414615749923178, + "grad_norm": 0.4406169056892395, + "learning_rate": 0.00029409457642312864, + "loss": 0.4765, + "step": 25319 + }, + { + "epoch": 1.41467162052686, + "grad_norm": 0.3293468952178955, + "learning_rate": 0.0002940665620797849, + "loss": 0.3628, + "step": 25320 + }, + { + "epoch": 1.4147274911305416, + "grad_norm": 0.6254953145980835, + "learning_rate": 0.00029403854773644105, + "loss": 0.4862, + "step": 25321 + }, + { + "epoch": 1.4147833617342236, + "grad_norm": 0.5870314836502075, + "learning_rate": 0.00029401053339309725, + "loss": 0.423, + "step": 25322 + }, + { + "epoch": 1.4148392323379055, + "grad_norm": 0.44004252552986145, + "learning_rate": 0.00029398251904975346, + "loss": 0.3122, + "step": 25323 + }, + { + "epoch": 1.4148951029415873, + "grad_norm": 0.6655048727989197, + "learning_rate": 0.00029395450470640967, + "loss": 0.4169, + "step": 25324 + }, + { + "epoch": 1.4149509735452692, + "grad_norm": 0.6879778504371643, + "learning_rate": 0.0002939264903630659, + "loss": 0.516, + "step": 25325 + }, + { + "epoch": 1.415006844148951, + "grad_norm": 1.9169622659683228, + "learning_rate": 0.0002938984760197221, + "loss": 0.3664, + "step": 25326 + }, + { + "epoch": 1.415062714752633, + "grad_norm": 0.5461876392364502, + "learning_rate": 0.00029387046167637834, + "loss": 0.5584, + "step": 25327 + }, + { + "epoch": 1.4151185853563149, + "grad_norm": 0.38999226689338684, + "learning_rate": 0.0002938424473330345, + "loss": 0.4626, + "step": 25328 + }, + { + "epoch": 1.4151744559599966, + "grad_norm": 1.393233299255371, + "learning_rate": 0.00029381443298969075, + "loss": 0.4752, + "step": 25329 + }, + { + "epoch": 1.4152303265636785, + "grad_norm": 0.4343869686126709, + "learning_rate": 0.0002937864186463469, + "loss": 0.4499, + "step": 25330 + }, + { + "epoch": 1.4152861971673603, + "grad_norm": 0.31542596220970154, + "learning_rate": 0.00029375840430300316, + "loss": 0.4191, + "step": 25331 + }, + { + "epoch": 1.4153420677710422, + "grad_norm": 1.395193099975586, + "learning_rate": 0.0002937303899596593, + "loss": 0.3694, + "step": 25332 + }, + { + "epoch": 1.4153979383747242, + "grad_norm": 0.7387697100639343, + "learning_rate": 0.00029370237561631557, + "loss": 0.4648, + "step": 25333 + }, + { + "epoch": 1.415453808978406, + "grad_norm": 0.42126572132110596, + "learning_rate": 0.0002936743612729718, + "loss": 0.5057, + "step": 25334 + }, + { + "epoch": 1.4155096795820878, + "grad_norm": 0.4728471338748932, + "learning_rate": 0.000293646346929628, + "loss": 0.4222, + "step": 25335 + }, + { + "epoch": 1.4155655501857698, + "grad_norm": 0.38324445486068726, + "learning_rate": 0.0002936183325862842, + "loss": 0.437, + "step": 25336 + }, + { + "epoch": 1.4156214207894515, + "grad_norm": 0.3893141448497772, + "learning_rate": 0.0002935903182429404, + "loss": 0.3504, + "step": 25337 + }, + { + "epoch": 1.4156772913931335, + "grad_norm": 0.6711946725845337, + "learning_rate": 0.0002935623038995966, + "loss": 0.6158, + "step": 25338 + }, + { + "epoch": 1.4157331619968154, + "grad_norm": 0.5026379823684692, + "learning_rate": 0.0002935342895562528, + "loss": 0.6926, + "step": 25339 + }, + { + "epoch": 1.4157890326004972, + "grad_norm": 1.118318796157837, + "learning_rate": 0.000293506275212909, + "loss": 0.5026, + "step": 25340 + }, + { + "epoch": 1.415844903204179, + "grad_norm": 0.46188780665397644, + "learning_rate": 0.0002934782608695653, + "loss": 0.3751, + "step": 25341 + }, + { + "epoch": 1.415900773807861, + "grad_norm": 0.6214801073074341, + "learning_rate": 0.0002934502465262214, + "loss": 0.4922, + "step": 25342 + }, + { + "epoch": 1.4159566444115428, + "grad_norm": 0.42922037839889526, + "learning_rate": 0.00029342223218287763, + "loss": 0.4115, + "step": 25343 + }, + { + "epoch": 1.4160125150152247, + "grad_norm": 3.019184112548828, + "learning_rate": 0.00029339421783953384, + "loss": 0.5221, + "step": 25344 + }, + { + "epoch": 1.4160683856189067, + "grad_norm": 0.808844804763794, + "learning_rate": 0.00029336620349619004, + "loss": 0.4106, + "step": 25345 + }, + { + "epoch": 1.4161242562225884, + "grad_norm": 0.42245909571647644, + "learning_rate": 0.00029333818915284625, + "loss": 0.4686, + "step": 25346 + }, + { + "epoch": 1.4161801268262704, + "grad_norm": 0.9617237448692322, + "learning_rate": 0.00029331017480950245, + "loss": 0.3877, + "step": 25347 + }, + { + "epoch": 1.4162359974299523, + "grad_norm": 0.37163978815078735, + "learning_rate": 0.0002932821604661587, + "loss": 0.3391, + "step": 25348 + }, + { + "epoch": 1.416291868033634, + "grad_norm": 0.400189608335495, + "learning_rate": 0.00029325414612281487, + "loss": 0.3999, + "step": 25349 + }, + { + "epoch": 1.416347738637316, + "grad_norm": 0.3418673574924469, + "learning_rate": 0.0002932261317794711, + "loss": 0.4344, + "step": 25350 + }, + { + "epoch": 1.416403609240998, + "grad_norm": 0.48953473567962646, + "learning_rate": 0.0002931981174361273, + "loss": 0.4036, + "step": 25351 + }, + { + "epoch": 1.4164594798446797, + "grad_norm": 0.4328266382217407, + "learning_rate": 0.00029317010309278354, + "loss": 0.4156, + "step": 25352 + }, + { + "epoch": 1.4165153504483616, + "grad_norm": 0.5705244541168213, + "learning_rate": 0.0002931420887494397, + "loss": 0.5496, + "step": 25353 + }, + { + "epoch": 1.4165712210520436, + "grad_norm": 0.47917869687080383, + "learning_rate": 0.00029311407440609595, + "loss": 0.5407, + "step": 25354 + }, + { + "epoch": 1.4166270916557253, + "grad_norm": 0.7136543989181519, + "learning_rate": 0.0002930860600627521, + "loss": 0.414, + "step": 25355 + }, + { + "epoch": 1.4166829622594073, + "grad_norm": 0.7759119272232056, + "learning_rate": 0.00029305804571940836, + "loss": 0.395, + "step": 25356 + }, + { + "epoch": 1.4167388328630892, + "grad_norm": 0.33319219946861267, + "learning_rate": 0.00029303003137606457, + "loss": 0.3829, + "step": 25357 + }, + { + "epoch": 1.416794703466771, + "grad_norm": 0.4059865474700928, + "learning_rate": 0.00029300201703272077, + "loss": 0.3987, + "step": 25358 + }, + { + "epoch": 1.416850574070453, + "grad_norm": 0.5560084581375122, + "learning_rate": 0.000292974002689377, + "loss": 0.473, + "step": 25359 + }, + { + "epoch": 1.4169064446741346, + "grad_norm": 0.3597427010536194, + "learning_rate": 0.0002929459883460332, + "loss": 0.3425, + "step": 25360 + }, + { + "epoch": 1.4169623152778166, + "grad_norm": 0.3420138955116272, + "learning_rate": 0.0002929179740026894, + "loss": 0.3628, + "step": 25361 + }, + { + "epoch": 1.4170181858814985, + "grad_norm": 4.377434730529785, + "learning_rate": 0.0002928899596593456, + "loss": 0.3913, + "step": 25362 + }, + { + "epoch": 1.4170740564851803, + "grad_norm": 0.5498926043510437, + "learning_rate": 0.0002928619453160018, + "loss": 0.3948, + "step": 25363 + }, + { + "epoch": 1.4171299270888622, + "grad_norm": 0.42704421281814575, + "learning_rate": 0.000292833930972658, + "loss": 0.451, + "step": 25364 + }, + { + "epoch": 1.417185797692544, + "grad_norm": 0.9248833060264587, + "learning_rate": 0.0002928059166293142, + "loss": 0.3974, + "step": 25365 + }, + { + "epoch": 1.4172416682962259, + "grad_norm": 0.495794415473938, + "learning_rate": 0.0002927779022859704, + "loss": 0.3997, + "step": 25366 + }, + { + "epoch": 1.4172975388999078, + "grad_norm": 0.4068528711795807, + "learning_rate": 0.0002927498879426266, + "loss": 0.4, + "step": 25367 + }, + { + "epoch": 1.4173534095035896, + "grad_norm": 0.37579867243766785, + "learning_rate": 0.00029272187359928283, + "loss": 0.4514, + "step": 25368 + }, + { + "epoch": 1.4174092801072715, + "grad_norm": 2.9611833095550537, + "learning_rate": 0.00029269385925593904, + "loss": 0.4315, + "step": 25369 + }, + { + "epoch": 1.4174651507109535, + "grad_norm": 0.36025163531303406, + "learning_rate": 0.00029266584491259524, + "loss": 0.3776, + "step": 25370 + }, + { + "epoch": 1.4175210213146352, + "grad_norm": 0.5586057901382446, + "learning_rate": 0.00029263783056925145, + "loss": 0.4032, + "step": 25371 + }, + { + "epoch": 1.4175768919183172, + "grad_norm": 0.6224256753921509, + "learning_rate": 0.00029260981622590765, + "loss": 0.3937, + "step": 25372 + }, + { + "epoch": 1.417632762521999, + "grad_norm": 0.3785489797592163, + "learning_rate": 0.0002925818018825639, + "loss": 0.4635, + "step": 25373 + }, + { + "epoch": 1.4176886331256808, + "grad_norm": 0.33229365944862366, + "learning_rate": 0.00029255378753922006, + "loss": 0.4398, + "step": 25374 + }, + { + "epoch": 1.4177445037293628, + "grad_norm": 0.5013099312782288, + "learning_rate": 0.0002925257731958763, + "loss": 0.3594, + "step": 25375 + }, + { + "epoch": 1.4178003743330447, + "grad_norm": 5.57098913192749, + "learning_rate": 0.0002924977588525325, + "loss": 0.3687, + "step": 25376 + }, + { + "epoch": 1.4178562449367265, + "grad_norm": 0.3506752550601959, + "learning_rate": 0.00029246974450918874, + "loss": 0.4279, + "step": 25377 + }, + { + "epoch": 1.4179121155404084, + "grad_norm": 0.4275366961956024, + "learning_rate": 0.0002924417301658449, + "loss": 0.4406, + "step": 25378 + }, + { + "epoch": 1.4179679861440904, + "grad_norm": 0.33816975355148315, + "learning_rate": 0.00029241371582250115, + "loss": 0.365, + "step": 25379 + }, + { + "epoch": 1.418023856747772, + "grad_norm": 0.5344197750091553, + "learning_rate": 0.00029238570147915735, + "loss": 0.3693, + "step": 25380 + }, + { + "epoch": 1.418079727351454, + "grad_norm": 0.9041814208030701, + "learning_rate": 0.00029235768713581356, + "loss": 0.4069, + "step": 25381 + }, + { + "epoch": 1.418135597955136, + "grad_norm": 1.2378853559494019, + "learning_rate": 0.00029232967279246977, + "loss": 0.6403, + "step": 25382 + }, + { + "epoch": 1.4181914685588177, + "grad_norm": 0.38541361689567566, + "learning_rate": 0.00029230165844912597, + "loss": 0.3578, + "step": 25383 + }, + { + "epoch": 1.4182473391624997, + "grad_norm": 0.460164338350296, + "learning_rate": 0.0002922736441057822, + "loss": 0.3932, + "step": 25384 + }, + { + "epoch": 1.4183032097661816, + "grad_norm": 0.4129196107387543, + "learning_rate": 0.00029224562976243833, + "loss": 0.4243, + "step": 25385 + }, + { + "epoch": 1.4183590803698634, + "grad_norm": 0.37825295329093933, + "learning_rate": 0.0002922176154190946, + "loss": 0.4283, + "step": 25386 + }, + { + "epoch": 1.4184149509735453, + "grad_norm": 0.8050940036773682, + "learning_rate": 0.00029218960107575074, + "loss": 0.3565, + "step": 25387 + }, + { + "epoch": 1.4184708215772273, + "grad_norm": 1.3069223165512085, + "learning_rate": 0.000292161586732407, + "loss": 0.5522, + "step": 25388 + }, + { + "epoch": 1.418526692180909, + "grad_norm": 0.7189136147499084, + "learning_rate": 0.0002921335723890632, + "loss": 0.3603, + "step": 25389 + }, + { + "epoch": 1.418582562784591, + "grad_norm": 0.41668710112571716, + "learning_rate": 0.0002921055580457194, + "loss": 0.3984, + "step": 25390 + }, + { + "epoch": 1.418638433388273, + "grad_norm": 0.5063287019729614, + "learning_rate": 0.0002920775437023756, + "loss": 0.4685, + "step": 25391 + }, + { + "epoch": 1.4186943039919546, + "grad_norm": 1.2213770151138306, + "learning_rate": 0.0002920495293590318, + "loss": 0.3827, + "step": 25392 + }, + { + "epoch": 1.4187501745956366, + "grad_norm": 0.5245025753974915, + "learning_rate": 0.00029202151501568803, + "loss": 0.4321, + "step": 25393 + }, + { + "epoch": 1.4188060451993183, + "grad_norm": 0.4517768919467926, + "learning_rate": 0.00029199350067234424, + "loss": 0.3218, + "step": 25394 + }, + { + "epoch": 1.4188619158030003, + "grad_norm": 0.5038626194000244, + "learning_rate": 0.00029196548632900044, + "loss": 0.3573, + "step": 25395 + }, + { + "epoch": 1.4189177864066822, + "grad_norm": 0.5676859021186829, + "learning_rate": 0.0002919374719856567, + "loss": 0.3783, + "step": 25396 + }, + { + "epoch": 1.418973657010364, + "grad_norm": 0.349449098110199, + "learning_rate": 0.00029190945764231285, + "loss": 0.4192, + "step": 25397 + }, + { + "epoch": 1.4190295276140459, + "grad_norm": 1.3807650804519653, + "learning_rate": 0.0002918814432989691, + "loss": 0.6221, + "step": 25398 + }, + { + "epoch": 1.4190853982177276, + "grad_norm": 0.4278123080730438, + "learning_rate": 0.00029185342895562526, + "loss": 0.3257, + "step": 25399 + }, + { + "epoch": 1.4191412688214096, + "grad_norm": 0.3480389416217804, + "learning_rate": 0.0002918254146122815, + "loss": 0.3488, + "step": 25400 + }, + { + "epoch": 1.4191971394250915, + "grad_norm": 0.9813069105148315, + "learning_rate": 0.0002917974002689377, + "loss": 0.3334, + "step": 25401 + }, + { + "epoch": 1.4192530100287732, + "grad_norm": 0.6606032252311707, + "learning_rate": 0.00029176938592559394, + "loss": 0.5574, + "step": 25402 + }, + { + "epoch": 1.4193088806324552, + "grad_norm": 0.27764391899108887, + "learning_rate": 0.0002917413715822501, + "loss": 0.3364, + "step": 25403 + }, + { + "epoch": 1.4193647512361371, + "grad_norm": 5.817492961883545, + "learning_rate": 0.00029171335723890635, + "loss": 0.4343, + "step": 25404 + }, + { + "epoch": 1.4194206218398189, + "grad_norm": 0.5628882050514221, + "learning_rate": 0.00029168534289556255, + "loss": 0.4781, + "step": 25405 + }, + { + "epoch": 1.4194764924435008, + "grad_norm": 0.3939323425292969, + "learning_rate": 0.0002916573285522187, + "loss": 0.4352, + "step": 25406 + }, + { + "epoch": 1.4195323630471828, + "grad_norm": 0.5740794539451599, + "learning_rate": 0.00029162931420887497, + "loss": 0.6142, + "step": 25407 + }, + { + "epoch": 1.4195882336508645, + "grad_norm": 0.4453948438167572, + "learning_rate": 0.0002916012998655311, + "loss": 0.4739, + "step": 25408 + }, + { + "epoch": 1.4196441042545465, + "grad_norm": 0.7252997756004333, + "learning_rate": 0.0002915732855221874, + "loss": 0.4018, + "step": 25409 + }, + { + "epoch": 1.4196999748582284, + "grad_norm": 0.5948705673217773, + "learning_rate": 0.00029154527117884353, + "loss": 0.4166, + "step": 25410 + }, + { + "epoch": 1.4197558454619101, + "grad_norm": 0.5045588612556458, + "learning_rate": 0.0002915172568354998, + "loss": 0.4171, + "step": 25411 + }, + { + "epoch": 1.419811716065592, + "grad_norm": 0.3631840944290161, + "learning_rate": 0.000291489242492156, + "loss": 0.4065, + "step": 25412 + }, + { + "epoch": 1.419867586669274, + "grad_norm": 0.3071323037147522, + "learning_rate": 0.0002914612281488122, + "loss": 0.3845, + "step": 25413 + }, + { + "epoch": 1.4199234572729558, + "grad_norm": 0.43856391310691833, + "learning_rate": 0.0002914332138054684, + "loss": 0.471, + "step": 25414 + }, + { + "epoch": 1.4199793278766377, + "grad_norm": 0.3853822648525238, + "learning_rate": 0.0002914051994621246, + "loss": 0.3787, + "step": 25415 + }, + { + "epoch": 1.4200351984803197, + "grad_norm": 0.3946545422077179, + "learning_rate": 0.0002913771851187808, + "loss": 0.3843, + "step": 25416 + }, + { + "epoch": 1.4200910690840014, + "grad_norm": 4.331936359405518, + "learning_rate": 0.000291349170775437, + "loss": 0.5388, + "step": 25417 + }, + { + "epoch": 1.4201469396876834, + "grad_norm": 2.421652317047119, + "learning_rate": 0.00029132115643209323, + "loss": 0.4059, + "step": 25418 + }, + { + "epoch": 1.4202028102913653, + "grad_norm": 1.345258355140686, + "learning_rate": 0.0002912931420887495, + "loss": 0.3878, + "step": 25419 + }, + { + "epoch": 1.420258680895047, + "grad_norm": 0.40249335765838623, + "learning_rate": 0.00029126512774540564, + "loss": 0.45, + "step": 25420 + }, + { + "epoch": 1.420314551498729, + "grad_norm": 0.4264683127403259, + "learning_rate": 0.0002912371134020619, + "loss": 0.3346, + "step": 25421 + }, + { + "epoch": 1.420370422102411, + "grad_norm": 0.36769384145736694, + "learning_rate": 0.00029120909905871805, + "loss": 0.3175, + "step": 25422 + }, + { + "epoch": 1.4204262927060927, + "grad_norm": 0.4750441908836365, + "learning_rate": 0.0002911810847153743, + "loss": 0.5432, + "step": 25423 + }, + { + "epoch": 1.4204821633097746, + "grad_norm": 0.4189533591270447, + "learning_rate": 0.00029115307037203046, + "loss": 0.4067, + "step": 25424 + }, + { + "epoch": 1.4205380339134566, + "grad_norm": 0.39574477076530457, + "learning_rate": 0.0002911250560286867, + "loss": 0.4989, + "step": 25425 + }, + { + "epoch": 1.4205939045171383, + "grad_norm": 0.7196701765060425, + "learning_rate": 0.0002910970416853429, + "loss": 0.3637, + "step": 25426 + }, + { + "epoch": 1.4206497751208202, + "grad_norm": 0.30444520711898804, + "learning_rate": 0.0002910690273419991, + "loss": 0.3856, + "step": 25427 + }, + { + "epoch": 1.420705645724502, + "grad_norm": 0.4391741156578064, + "learning_rate": 0.00029104101299865534, + "loss": 0.3956, + "step": 25428 + }, + { + "epoch": 1.420761516328184, + "grad_norm": 2.5714917182922363, + "learning_rate": 0.0002910129986553115, + "loss": 0.4581, + "step": 25429 + }, + { + "epoch": 1.4208173869318659, + "grad_norm": 0.5592794418334961, + "learning_rate": 0.00029098498431196775, + "loss": 0.4544, + "step": 25430 + }, + { + "epoch": 1.4208732575355476, + "grad_norm": 0.38417306542396545, + "learning_rate": 0.0002909569699686239, + "loss": 0.4149, + "step": 25431 + }, + { + "epoch": 1.4209291281392296, + "grad_norm": 2.0664000511169434, + "learning_rate": 0.00029092895562528016, + "loss": 0.4281, + "step": 25432 + }, + { + "epoch": 1.4209849987429113, + "grad_norm": 0.5301982164382935, + "learning_rate": 0.0002909009412819363, + "loss": 0.4404, + "step": 25433 + }, + { + "epoch": 1.4210408693465932, + "grad_norm": 0.3858567774295807, + "learning_rate": 0.0002908729269385926, + "loss": 0.3661, + "step": 25434 + }, + { + "epoch": 1.4210967399502752, + "grad_norm": 0.4178643226623535, + "learning_rate": 0.0002908449125952488, + "loss": 0.5224, + "step": 25435 + }, + { + "epoch": 1.421152610553957, + "grad_norm": 0.4791751801967621, + "learning_rate": 0.000290816898251905, + "loss": 0.3913, + "step": 25436 + }, + { + "epoch": 1.4212084811576389, + "grad_norm": 0.40266111493110657, + "learning_rate": 0.0002907888839085612, + "loss": 0.5943, + "step": 25437 + }, + { + "epoch": 1.4212643517613208, + "grad_norm": 0.6722913980484009, + "learning_rate": 0.0002907608695652174, + "loss": 0.3412, + "step": 25438 + }, + { + "epoch": 1.4213202223650026, + "grad_norm": 0.33675968647003174, + "learning_rate": 0.0002907328552218736, + "loss": 0.4261, + "step": 25439 + }, + { + "epoch": 1.4213760929686845, + "grad_norm": 0.476284921169281, + "learning_rate": 0.0002907048408785298, + "loss": 0.3643, + "step": 25440 + }, + { + "epoch": 1.4214319635723665, + "grad_norm": 3.527056932449341, + "learning_rate": 0.000290676826535186, + "loss": 0.4895, + "step": 25441 + }, + { + "epoch": 1.4214878341760482, + "grad_norm": 2.638477087020874, + "learning_rate": 0.0002906488121918422, + "loss": 0.405, + "step": 25442 + }, + { + "epoch": 1.4215437047797301, + "grad_norm": 0.5411128997802734, + "learning_rate": 0.00029062079784849843, + "loss": 0.454, + "step": 25443 + }, + { + "epoch": 1.421599575383412, + "grad_norm": 0.6498481631278992, + "learning_rate": 0.0002905927835051547, + "loss": 0.6743, + "step": 25444 + }, + { + "epoch": 1.4216554459870938, + "grad_norm": 0.4985831379890442, + "learning_rate": 0.00029056476916181084, + "loss": 0.3721, + "step": 25445 + }, + { + "epoch": 1.4217113165907758, + "grad_norm": 0.4296768605709076, + "learning_rate": 0.0002905367548184671, + "loss": 0.3329, + "step": 25446 + }, + { + "epoch": 1.4217671871944577, + "grad_norm": 0.621747612953186, + "learning_rate": 0.00029050874047512325, + "loss": 0.4611, + "step": 25447 + }, + { + "epoch": 1.4218230577981394, + "grad_norm": 1.1796104907989502, + "learning_rate": 0.0002904807261317795, + "loss": 0.4104, + "step": 25448 + }, + { + "epoch": 1.4218789284018214, + "grad_norm": 2.0157530307769775, + "learning_rate": 0.00029045271178843566, + "loss": 0.433, + "step": 25449 + }, + { + "epoch": 1.4219347990055033, + "grad_norm": 0.3481171429157257, + "learning_rate": 0.00029042469744509187, + "loss": 0.399, + "step": 25450 + }, + { + "epoch": 1.421990669609185, + "grad_norm": 1.9303622245788574, + "learning_rate": 0.00029039668310174813, + "loss": 0.3304, + "step": 25451 + }, + { + "epoch": 1.422046540212867, + "grad_norm": 15.021668434143066, + "learning_rate": 0.0002903686687584043, + "loss": 0.3982, + "step": 25452 + }, + { + "epoch": 1.422102410816549, + "grad_norm": 0.582679033279419, + "learning_rate": 0.00029034065441506054, + "loss": 0.5706, + "step": 25453 + }, + { + "epoch": 1.4221582814202307, + "grad_norm": 0.5455203652381897, + "learning_rate": 0.0002903126400717167, + "loss": 0.4962, + "step": 25454 + }, + { + "epoch": 1.4222141520239127, + "grad_norm": 0.5289329290390015, + "learning_rate": 0.00029028462572837295, + "loss": 0.4454, + "step": 25455 + }, + { + "epoch": 1.4222700226275946, + "grad_norm": 0.6536529064178467, + "learning_rate": 0.0002902566113850291, + "loss": 0.4274, + "step": 25456 + }, + { + "epoch": 1.4223258932312763, + "grad_norm": 0.5803133249282837, + "learning_rate": 0.00029022859704168536, + "loss": 0.3847, + "step": 25457 + }, + { + "epoch": 1.4223817638349583, + "grad_norm": 0.5410629510879517, + "learning_rate": 0.0002902005826983415, + "loss": 0.4302, + "step": 25458 + }, + { + "epoch": 1.4224376344386402, + "grad_norm": 0.3772728145122528, + "learning_rate": 0.0002901725683549978, + "loss": 0.3987, + "step": 25459 + }, + { + "epoch": 1.422493505042322, + "grad_norm": 0.40224719047546387, + "learning_rate": 0.000290144554011654, + "loss": 0.44, + "step": 25460 + }, + { + "epoch": 1.422549375646004, + "grad_norm": 0.4489666521549225, + "learning_rate": 0.0002901165396683102, + "loss": 0.4098, + "step": 25461 + }, + { + "epoch": 1.4226052462496857, + "grad_norm": 0.6130057573318481, + "learning_rate": 0.0002900885253249664, + "loss": 0.3376, + "step": 25462 + }, + { + "epoch": 1.4226611168533676, + "grad_norm": 0.311115562915802, + "learning_rate": 0.0002900605109816226, + "loss": 0.3318, + "step": 25463 + }, + { + "epoch": 1.4227169874570496, + "grad_norm": 0.40908172726631165, + "learning_rate": 0.0002900324966382788, + "loss": 0.5304, + "step": 25464 + }, + { + "epoch": 1.4227728580607313, + "grad_norm": 0.4104495644569397, + "learning_rate": 0.000290004482294935, + "loss": 0.4127, + "step": 25465 + }, + { + "epoch": 1.4228287286644132, + "grad_norm": 0.33365944027900696, + "learning_rate": 0.0002899764679515912, + "loss": 0.3832, + "step": 25466 + }, + { + "epoch": 1.422884599268095, + "grad_norm": 0.7053511142730713, + "learning_rate": 0.0002899484536082475, + "loss": 0.4047, + "step": 25467 + }, + { + "epoch": 1.422940469871777, + "grad_norm": 0.5030487775802612, + "learning_rate": 0.00028992043926490363, + "loss": 0.3746, + "step": 25468 + }, + { + "epoch": 1.4229963404754589, + "grad_norm": 0.6047595143318176, + "learning_rate": 0.0002898924249215599, + "loss": 0.3586, + "step": 25469 + }, + { + "epoch": 1.4230522110791406, + "grad_norm": 0.5674085021018982, + "learning_rate": 0.00028986441057821604, + "loss": 0.4135, + "step": 25470 + }, + { + "epoch": 1.4231080816828225, + "grad_norm": 0.28867530822753906, + "learning_rate": 0.00028983639623487225, + "loss": 0.3936, + "step": 25471 + }, + { + "epoch": 1.4231639522865045, + "grad_norm": 0.3970608115196228, + "learning_rate": 0.00028980838189152845, + "loss": 0.3942, + "step": 25472 + }, + { + "epoch": 1.4232198228901862, + "grad_norm": 1.850643277168274, + "learning_rate": 0.00028978036754818466, + "loss": 0.4274, + "step": 25473 + }, + { + "epoch": 1.4232756934938682, + "grad_norm": 0.4165058732032776, + "learning_rate": 0.0002897523532048409, + "loss": 0.5171, + "step": 25474 + }, + { + "epoch": 1.4233315640975501, + "grad_norm": 0.7398850321769714, + "learning_rate": 0.00028972433886149707, + "loss": 0.3089, + "step": 25475 + }, + { + "epoch": 1.4233874347012319, + "grad_norm": 0.48951125144958496, + "learning_rate": 0.00028969632451815333, + "loss": 0.4401, + "step": 25476 + }, + { + "epoch": 1.4234433053049138, + "grad_norm": 0.7286028265953064, + "learning_rate": 0.0002896683101748095, + "loss": 0.4693, + "step": 25477 + }, + { + "epoch": 1.4234991759085958, + "grad_norm": 0.3817266821861267, + "learning_rate": 0.00028964029583146574, + "loss": 0.3398, + "step": 25478 + }, + { + "epoch": 1.4235550465122775, + "grad_norm": 0.6839649677276611, + "learning_rate": 0.0002896122814881219, + "loss": 0.5933, + "step": 25479 + }, + { + "epoch": 1.4236109171159594, + "grad_norm": 0.3611190915107727, + "learning_rate": 0.00028958426714477815, + "loss": 0.4282, + "step": 25480 + }, + { + "epoch": 1.4236667877196414, + "grad_norm": 0.5921741127967834, + "learning_rate": 0.0002895562528014343, + "loss": 0.419, + "step": 25481 + }, + { + "epoch": 1.4237226583233231, + "grad_norm": 2.1053993701934814, + "learning_rate": 0.00028952823845809056, + "loss": 0.4249, + "step": 25482 + }, + { + "epoch": 1.423778528927005, + "grad_norm": 0.4932815730571747, + "learning_rate": 0.00028950022411474677, + "loss": 0.604, + "step": 25483 + }, + { + "epoch": 1.423834399530687, + "grad_norm": 0.3796716332435608, + "learning_rate": 0.000289472209771403, + "loss": 0.369, + "step": 25484 + }, + { + "epoch": 1.4238902701343688, + "grad_norm": 0.4119528830051422, + "learning_rate": 0.0002894441954280592, + "loss": 0.3173, + "step": 25485 + }, + { + "epoch": 1.4239461407380507, + "grad_norm": 0.4525563418865204, + "learning_rate": 0.0002894161810847154, + "loss": 0.4035, + "step": 25486 + }, + { + "epoch": 1.4240020113417327, + "grad_norm": 0.364399790763855, + "learning_rate": 0.0002893881667413716, + "loss": 0.3974, + "step": 25487 + }, + { + "epoch": 1.4240578819454144, + "grad_norm": 0.40776491165161133, + "learning_rate": 0.0002893601523980278, + "loss": 0.3696, + "step": 25488 + }, + { + "epoch": 1.4241137525490963, + "grad_norm": 1.3397902250289917, + "learning_rate": 0.000289332138054684, + "loss": 0.3842, + "step": 25489 + }, + { + "epoch": 1.4241696231527783, + "grad_norm": 0.36536699533462524, + "learning_rate": 0.00028930412371134026, + "loss": 0.4418, + "step": 25490 + }, + { + "epoch": 1.42422549375646, + "grad_norm": 0.4378626346588135, + "learning_rate": 0.0002892761093679964, + "loss": 0.3626, + "step": 25491 + }, + { + "epoch": 1.424281364360142, + "grad_norm": 0.3109069764614105, + "learning_rate": 0.0002892480950246526, + "loss": 0.3978, + "step": 25492 + }, + { + "epoch": 1.424337234963824, + "grad_norm": 0.7592468857765198, + "learning_rate": 0.00028922008068130883, + "loss": 0.437, + "step": 25493 + }, + { + "epoch": 1.4243931055675056, + "grad_norm": 0.3265565037727356, + "learning_rate": 0.00028919206633796503, + "loss": 0.3832, + "step": 25494 + }, + { + "epoch": 1.4244489761711876, + "grad_norm": 0.6603091955184937, + "learning_rate": 0.00028916405199462124, + "loss": 0.4154, + "step": 25495 + }, + { + "epoch": 1.4245048467748693, + "grad_norm": 0.4394338130950928, + "learning_rate": 0.00028913603765127745, + "loss": 0.4349, + "step": 25496 + }, + { + "epoch": 1.4245607173785513, + "grad_norm": 0.49744677543640137, + "learning_rate": 0.00028910802330793365, + "loss": 0.4683, + "step": 25497 + }, + { + "epoch": 1.4246165879822332, + "grad_norm": 0.3251411020755768, + "learning_rate": 0.00028908000896458986, + "loss": 0.2945, + "step": 25498 + }, + { + "epoch": 1.424672458585915, + "grad_norm": 1.0808240175247192, + "learning_rate": 0.0002890519946212461, + "loss": 0.595, + "step": 25499 + }, + { + "epoch": 1.424728329189597, + "grad_norm": 0.48918384313583374, + "learning_rate": 0.00028902398027790227, + "loss": 0.4091, + "step": 25500 + }, + { + "epoch": 1.424728329189597, + "eval_cer": 0.08514724013398361, + "eval_loss": 0.3219001293182373, + "eval_runtime": 56.1623, + "eval_samples_per_second": 80.802, + "eval_steps_per_second": 5.057, + "eval_wer": 0.3371134884500433, + "step": 25500 + }, + { + "epoch": 1.4247841997932786, + "grad_norm": 0.35011470317840576, + "learning_rate": 0.00028899596593455853, + "loss": 0.3218, + "step": 25501 + }, + { + "epoch": 1.4248400703969606, + "grad_norm": 0.4353053569793701, + "learning_rate": 0.0002889679515912147, + "loss": 0.3786, + "step": 25502 + }, + { + "epoch": 1.4248959410006425, + "grad_norm": 1.2122390270233154, + "learning_rate": 0.00028893993724787094, + "loss": 0.5471, + "step": 25503 + }, + { + "epoch": 1.4249518116043243, + "grad_norm": 0.5849142670631409, + "learning_rate": 0.0002889119229045271, + "loss": 0.3558, + "step": 25504 + }, + { + "epoch": 1.4250076822080062, + "grad_norm": 0.6091597080230713, + "learning_rate": 0.00028888390856118335, + "loss": 0.4629, + "step": 25505 + }, + { + "epoch": 1.4250635528116882, + "grad_norm": 2.8950307369232178, + "learning_rate": 0.00028885589421783956, + "loss": 0.4009, + "step": 25506 + }, + { + "epoch": 1.42511942341537, + "grad_norm": 0.8560590147972107, + "learning_rate": 0.00028882787987449576, + "loss": 0.3321, + "step": 25507 + }, + { + "epoch": 1.4251752940190519, + "grad_norm": 1.089881420135498, + "learning_rate": 0.00028879986553115197, + "loss": 0.3824, + "step": 25508 + }, + { + "epoch": 1.4252311646227338, + "grad_norm": 0.3353920876979828, + "learning_rate": 0.0002887718511878082, + "loss": 0.4406, + "step": 25509 + }, + { + "epoch": 1.4252870352264155, + "grad_norm": 0.42105981707572937, + "learning_rate": 0.0002887438368444644, + "loss": 0.401, + "step": 25510 + }, + { + "epoch": 1.4253429058300975, + "grad_norm": 0.43882450461387634, + "learning_rate": 0.0002887158225011206, + "loss": 0.3895, + "step": 25511 + }, + { + "epoch": 1.4253987764337794, + "grad_norm": 0.3930583596229553, + "learning_rate": 0.0002886878081577768, + "loss": 0.4304, + "step": 25512 + }, + { + "epoch": 1.4254546470374612, + "grad_norm": 0.32005947828292847, + "learning_rate": 0.00028865979381443294, + "loss": 0.3207, + "step": 25513 + }, + { + "epoch": 1.4255105176411431, + "grad_norm": 0.7226079106330872, + "learning_rate": 0.0002886317794710892, + "loss": 0.3966, + "step": 25514 + }, + { + "epoch": 1.425566388244825, + "grad_norm": 0.4687442183494568, + "learning_rate": 0.0002886037651277454, + "loss": 0.565, + "step": 25515 + }, + { + "epoch": 1.4256222588485068, + "grad_norm": 0.5709344744682312, + "learning_rate": 0.0002885757507844016, + "loss": 0.4691, + "step": 25516 + }, + { + "epoch": 1.4256781294521887, + "grad_norm": 0.36718645691871643, + "learning_rate": 0.0002885477364410578, + "loss": 0.46, + "step": 25517 + }, + { + "epoch": 1.4257340000558707, + "grad_norm": 0.37390270829200745, + "learning_rate": 0.00028851972209771403, + "loss": 0.5342, + "step": 25518 + }, + { + "epoch": 1.4257898706595524, + "grad_norm": 0.42574772238731384, + "learning_rate": 0.00028849170775437023, + "loss": 0.5103, + "step": 25519 + }, + { + "epoch": 1.4258457412632344, + "grad_norm": 0.37539491057395935, + "learning_rate": 0.00028846369341102644, + "loss": 0.3458, + "step": 25520 + }, + { + "epoch": 1.4259016118669163, + "grad_norm": 0.48678603768348694, + "learning_rate": 0.00028843567906768265, + "loss": 0.3733, + "step": 25521 + }, + { + "epoch": 1.425957482470598, + "grad_norm": 0.47108492255210876, + "learning_rate": 0.0002884076647243389, + "loss": 0.3459, + "step": 25522 + }, + { + "epoch": 1.42601335307428, + "grad_norm": 0.46822819113731384, + "learning_rate": 0.00028837965038099506, + "loss": 0.4179, + "step": 25523 + }, + { + "epoch": 1.426069223677962, + "grad_norm": 4.186408042907715, + "learning_rate": 0.0002883516360376513, + "loss": 0.3148, + "step": 25524 + }, + { + "epoch": 1.4261250942816437, + "grad_norm": 0.38902348279953003, + "learning_rate": 0.00028832362169430747, + "loss": 0.3549, + "step": 25525 + }, + { + "epoch": 1.4261809648853256, + "grad_norm": 0.5360889434814453, + "learning_rate": 0.00028829560735096373, + "loss": 0.5105, + "step": 25526 + }, + { + "epoch": 1.4262368354890076, + "grad_norm": 0.3890235424041748, + "learning_rate": 0.0002882675930076199, + "loss": 0.4552, + "step": 25527 + }, + { + "epoch": 1.4262927060926893, + "grad_norm": 0.7125121355056763, + "learning_rate": 0.00028823957866427614, + "loss": 0.3502, + "step": 25528 + }, + { + "epoch": 1.4263485766963713, + "grad_norm": 0.9954025149345398, + "learning_rate": 0.0002882115643209323, + "loss": 0.4379, + "step": 25529 + }, + { + "epoch": 1.426404447300053, + "grad_norm": 0.5401270985603333, + "learning_rate": 0.00028818354997758855, + "loss": 0.3895, + "step": 25530 + }, + { + "epoch": 1.426460317903735, + "grad_norm": 0.3956260085105896, + "learning_rate": 0.00028815553563424476, + "loss": 0.3699, + "step": 25531 + }, + { + "epoch": 1.426516188507417, + "grad_norm": 0.41201671957969666, + "learning_rate": 0.00028812752129090096, + "loss": 0.4379, + "step": 25532 + }, + { + "epoch": 1.4265720591110986, + "grad_norm": 0.36473554372787476, + "learning_rate": 0.00028809950694755717, + "loss": 0.3816, + "step": 25533 + }, + { + "epoch": 1.4266279297147806, + "grad_norm": 0.33641380071640015, + "learning_rate": 0.0002880714926042133, + "loss": 0.4158, + "step": 25534 + }, + { + "epoch": 1.4266838003184623, + "grad_norm": 1.8433700799942017, + "learning_rate": 0.0002880434782608696, + "loss": 0.3313, + "step": 25535 + }, + { + "epoch": 1.4267396709221443, + "grad_norm": 0.3718568682670593, + "learning_rate": 0.00028801546391752573, + "loss": 0.3636, + "step": 25536 + }, + { + "epoch": 1.4267955415258262, + "grad_norm": 0.3887096345424652, + "learning_rate": 0.000287987449574182, + "loss": 0.3518, + "step": 25537 + }, + { + "epoch": 1.426851412129508, + "grad_norm": 0.6843027472496033, + "learning_rate": 0.0002879594352308382, + "loss": 0.4689, + "step": 25538 + }, + { + "epoch": 1.42690728273319, + "grad_norm": 5.137730121612549, + "learning_rate": 0.0002879314208874944, + "loss": 0.4453, + "step": 25539 + }, + { + "epoch": 1.4269631533368718, + "grad_norm": 0.40621915459632874, + "learning_rate": 0.0002879034065441506, + "loss": 0.3904, + "step": 25540 + }, + { + "epoch": 1.4270190239405536, + "grad_norm": 0.49227938055992126, + "learning_rate": 0.0002878753922008068, + "loss": 0.4676, + "step": 25541 + }, + { + "epoch": 1.4270748945442355, + "grad_norm": 0.5761269330978394, + "learning_rate": 0.000287847377857463, + "loss": 0.4224, + "step": 25542 + }, + { + "epoch": 1.4271307651479175, + "grad_norm": 0.45996078848838806, + "learning_rate": 0.00028781936351411923, + "loss": 0.3555, + "step": 25543 + }, + { + "epoch": 1.4271866357515992, + "grad_norm": 1.1656804084777832, + "learning_rate": 0.00028779134917077543, + "loss": 0.4626, + "step": 25544 + }, + { + "epoch": 1.4272425063552812, + "grad_norm": 0.3911478817462921, + "learning_rate": 0.0002877633348274317, + "loss": 0.3508, + "step": 25545 + }, + { + "epoch": 1.427298376958963, + "grad_norm": 0.8286311030387878, + "learning_rate": 0.00028773532048408784, + "loss": 0.3652, + "step": 25546 + }, + { + "epoch": 1.4273542475626448, + "grad_norm": 0.9688900709152222, + "learning_rate": 0.0002877073061407441, + "loss": 0.3772, + "step": 25547 + }, + { + "epoch": 1.4274101181663268, + "grad_norm": 0.40926507115364075, + "learning_rate": 0.00028767929179740026, + "loss": 0.3734, + "step": 25548 + }, + { + "epoch": 1.4274659887700087, + "grad_norm": 1.8863152265548706, + "learning_rate": 0.0002876512774540565, + "loss": 0.4793, + "step": 25549 + }, + { + "epoch": 1.4275218593736905, + "grad_norm": 0.8279812335968018, + "learning_rate": 0.00028762326311071267, + "loss": 0.3388, + "step": 25550 + }, + { + "epoch": 1.4275777299773724, + "grad_norm": 0.5027724504470825, + "learning_rate": 0.00028759524876736893, + "loss": 0.46, + "step": 25551 + }, + { + "epoch": 1.4276336005810544, + "grad_norm": 0.3062552511692047, + "learning_rate": 0.0002875672344240251, + "loss": 0.404, + "step": 25552 + }, + { + "epoch": 1.427689471184736, + "grad_norm": 0.48964107036590576, + "learning_rate": 0.00028753922008068134, + "loss": 0.3088, + "step": 25553 + }, + { + "epoch": 1.427745341788418, + "grad_norm": 0.399536669254303, + "learning_rate": 0.00028751120573733755, + "loss": 0.3478, + "step": 25554 + }, + { + "epoch": 1.4278012123921, + "grad_norm": 1.094277262687683, + "learning_rate": 0.0002874831913939937, + "loss": 0.4603, + "step": 25555 + }, + { + "epoch": 1.4278570829957817, + "grad_norm": 0.3602153956890106, + "learning_rate": 0.00028745517705064996, + "loss": 0.3704, + "step": 25556 + }, + { + "epoch": 1.4279129535994637, + "grad_norm": 0.42085781693458557, + "learning_rate": 0.0002874271627073061, + "loss": 0.3849, + "step": 25557 + }, + { + "epoch": 1.4279688242031456, + "grad_norm": 0.6733229160308838, + "learning_rate": 0.00028739914836396237, + "loss": 0.4133, + "step": 25558 + }, + { + "epoch": 1.4280246948068274, + "grad_norm": 0.4493616223335266, + "learning_rate": 0.0002873711340206185, + "loss": 0.4599, + "step": 25559 + }, + { + "epoch": 1.4280805654105093, + "grad_norm": 0.38915467262268066, + "learning_rate": 0.0002873431196772748, + "loss": 0.3854, + "step": 25560 + }, + { + "epoch": 1.4281364360141913, + "grad_norm": 0.3758406639099121, + "learning_rate": 0.000287315105333931, + "loss": 0.3892, + "step": 25561 + }, + { + "epoch": 1.428192306617873, + "grad_norm": 0.3665432333946228, + "learning_rate": 0.0002872870909905872, + "loss": 0.3736, + "step": 25562 + }, + { + "epoch": 1.428248177221555, + "grad_norm": 0.5153552889823914, + "learning_rate": 0.0002872590766472434, + "loss": 0.4548, + "step": 25563 + }, + { + "epoch": 1.4283040478252367, + "grad_norm": 0.3836689889431, + "learning_rate": 0.0002872310623038996, + "loss": 0.3443, + "step": 25564 + }, + { + "epoch": 1.4283599184289186, + "grad_norm": 2.007498025894165, + "learning_rate": 0.0002872030479605558, + "loss": 0.4257, + "step": 25565 + }, + { + "epoch": 1.4284157890326006, + "grad_norm": 0.4939742982387543, + "learning_rate": 0.000287175033617212, + "loss": 0.2934, + "step": 25566 + }, + { + "epoch": 1.4284716596362823, + "grad_norm": 0.508540689945221, + "learning_rate": 0.0002871470192738682, + "loss": 0.5121, + "step": 25567 + }, + { + "epoch": 1.4285275302399643, + "grad_norm": 0.388118177652359, + "learning_rate": 0.0002871190049305244, + "loss": 0.4741, + "step": 25568 + }, + { + "epoch": 1.428583400843646, + "grad_norm": 0.5957990884780884, + "learning_rate": 0.00028709099058718063, + "loss": 0.4066, + "step": 25569 + }, + { + "epoch": 1.428639271447328, + "grad_norm": 1.0515859127044678, + "learning_rate": 0.0002870629762438369, + "loss": 0.3371, + "step": 25570 + }, + { + "epoch": 1.42869514205101, + "grad_norm": 0.43791547417640686, + "learning_rate": 0.00028703496190049304, + "loss": 0.4621, + "step": 25571 + }, + { + "epoch": 1.4287510126546916, + "grad_norm": 0.7147131562232971, + "learning_rate": 0.0002870069475571493, + "loss": 0.5034, + "step": 25572 + }, + { + "epoch": 1.4288068832583736, + "grad_norm": 0.7382535934448242, + "learning_rate": 0.00028697893321380546, + "loss": 0.374, + "step": 25573 + }, + { + "epoch": 1.4288627538620555, + "grad_norm": 0.47115442156791687, + "learning_rate": 0.0002869509188704617, + "loss": 0.3594, + "step": 25574 + }, + { + "epoch": 1.4289186244657373, + "grad_norm": 0.6174131035804749, + "learning_rate": 0.00028692290452711787, + "loss": 0.2591, + "step": 25575 + }, + { + "epoch": 1.4289744950694192, + "grad_norm": 0.5376176238059998, + "learning_rate": 0.0002868948901837741, + "loss": 0.3694, + "step": 25576 + }, + { + "epoch": 1.4290303656731012, + "grad_norm": 0.651934027671814, + "learning_rate": 0.00028686687584043033, + "loss": 0.4109, + "step": 25577 + }, + { + "epoch": 1.4290862362767829, + "grad_norm": 0.4709267020225525, + "learning_rate": 0.0002868388614970865, + "loss": 0.6212, + "step": 25578 + }, + { + "epoch": 1.4291421068804648, + "grad_norm": 0.5251539349555969, + "learning_rate": 0.00028681084715374274, + "loss": 0.4649, + "step": 25579 + }, + { + "epoch": 1.4291979774841468, + "grad_norm": 2.8078339099884033, + "learning_rate": 0.0002867828328103989, + "loss": 0.4889, + "step": 25580 + }, + { + "epoch": 1.4292538480878285, + "grad_norm": 0.36791232228279114, + "learning_rate": 0.00028675481846705516, + "loss": 0.3674, + "step": 25581 + }, + { + "epoch": 1.4293097186915105, + "grad_norm": 0.6811491250991821, + "learning_rate": 0.0002867268041237113, + "loss": 0.5704, + "step": 25582 + }, + { + "epoch": 1.4293655892951924, + "grad_norm": 0.8714993000030518, + "learning_rate": 0.00028669878978036757, + "loss": 0.4085, + "step": 25583 + }, + { + "epoch": 1.4294214598988741, + "grad_norm": 0.4673130512237549, + "learning_rate": 0.0002866707754370237, + "loss": 0.4217, + "step": 25584 + }, + { + "epoch": 1.429477330502556, + "grad_norm": 0.49375179409980774, + "learning_rate": 0.00028664276109368, + "loss": 0.3641, + "step": 25585 + }, + { + "epoch": 1.429533201106238, + "grad_norm": 1.0057095289230347, + "learning_rate": 0.0002866147467503362, + "loss": 0.3997, + "step": 25586 + }, + { + "epoch": 1.4295890717099198, + "grad_norm": 0.36103275418281555, + "learning_rate": 0.0002865867324069924, + "loss": 0.3615, + "step": 25587 + }, + { + "epoch": 1.4296449423136017, + "grad_norm": 0.5148767232894897, + "learning_rate": 0.0002865587180636486, + "loss": 0.4475, + "step": 25588 + }, + { + "epoch": 1.4297008129172837, + "grad_norm": 0.6091818809509277, + "learning_rate": 0.0002865307037203048, + "loss": 0.5171, + "step": 25589 + }, + { + "epoch": 1.4297566835209654, + "grad_norm": 0.5523170828819275, + "learning_rate": 0.000286502689376961, + "loss": 0.432, + "step": 25590 + }, + { + "epoch": 1.4298125541246474, + "grad_norm": 0.4402397572994232, + "learning_rate": 0.0002864746750336172, + "loss": 0.412, + "step": 25591 + }, + { + "epoch": 1.4298684247283293, + "grad_norm": 0.3701607584953308, + "learning_rate": 0.0002864466606902734, + "loss": 0.3635, + "step": 25592 + }, + { + "epoch": 1.429924295332011, + "grad_norm": 0.41311657428741455, + "learning_rate": 0.0002864186463469297, + "loss": 0.4296, + "step": 25593 + }, + { + "epoch": 1.429980165935693, + "grad_norm": 0.5336693525314331, + "learning_rate": 0.00028639063200358583, + "loss": 0.3925, + "step": 25594 + }, + { + "epoch": 1.430036036539375, + "grad_norm": 0.4230882227420807, + "learning_rate": 0.0002863626176602421, + "loss": 0.4281, + "step": 25595 + }, + { + "epoch": 1.4300919071430567, + "grad_norm": 0.49133968353271484, + "learning_rate": 0.00028633460331689824, + "loss": 0.4971, + "step": 25596 + }, + { + "epoch": 1.4301477777467386, + "grad_norm": 0.38664862513542175, + "learning_rate": 0.00028630658897355445, + "loss": 0.5, + "step": 25597 + }, + { + "epoch": 1.4302036483504204, + "grad_norm": 0.650442361831665, + "learning_rate": 0.00028627857463021066, + "loss": 0.56, + "step": 25598 + }, + { + "epoch": 1.4302595189541023, + "grad_norm": 0.4343176782131195, + "learning_rate": 0.00028625056028686686, + "loss": 0.477, + "step": 25599 + }, + { + "epoch": 1.430315389557784, + "grad_norm": 0.42309051752090454, + "learning_rate": 0.00028622254594352307, + "loss": 0.4344, + "step": 25600 + }, + { + "epoch": 1.430371260161466, + "grad_norm": 0.3386980891227722, + "learning_rate": 0.00028619453160017927, + "loss": 0.3951, + "step": 25601 + }, + { + "epoch": 1.430427130765148, + "grad_norm": 0.7808191180229187, + "learning_rate": 0.00028616651725683553, + "loss": 0.5029, + "step": 25602 + }, + { + "epoch": 1.4304830013688297, + "grad_norm": 0.6249203681945801, + "learning_rate": 0.0002861385029134917, + "loss": 0.5398, + "step": 25603 + }, + { + "epoch": 1.4305388719725116, + "grad_norm": 0.4913393259048462, + "learning_rate": 0.00028611048857014794, + "loss": 0.3685, + "step": 25604 + }, + { + "epoch": 1.4305947425761936, + "grad_norm": 0.4134288728237152, + "learning_rate": 0.0002860824742268041, + "loss": 0.3586, + "step": 25605 + }, + { + "epoch": 1.4306506131798753, + "grad_norm": 0.9465571045875549, + "learning_rate": 0.00028605445988346036, + "loss": 0.5535, + "step": 25606 + }, + { + "epoch": 1.4307064837835572, + "grad_norm": 2.4837636947631836, + "learning_rate": 0.0002860264455401165, + "loss": 0.4068, + "step": 25607 + }, + { + "epoch": 1.4307623543872392, + "grad_norm": 0.4947718679904938, + "learning_rate": 0.00028599843119677277, + "loss": 0.455, + "step": 25608 + }, + { + "epoch": 1.430818224990921, + "grad_norm": 0.46444302797317505, + "learning_rate": 0.000285970416853429, + "loss": 0.5564, + "step": 25609 + }, + { + "epoch": 1.4308740955946029, + "grad_norm": 0.9095009565353394, + "learning_rate": 0.0002859424025100852, + "loss": 0.4198, + "step": 25610 + }, + { + "epoch": 1.4309299661982848, + "grad_norm": 3.4160749912261963, + "learning_rate": 0.0002859143881667414, + "loss": 0.4017, + "step": 25611 + }, + { + "epoch": 1.4309858368019666, + "grad_norm": 0.38843661546707153, + "learning_rate": 0.0002858863738233976, + "loss": 0.3595, + "step": 25612 + }, + { + "epoch": 1.4310417074056485, + "grad_norm": 0.30632835626602173, + "learning_rate": 0.0002858583594800538, + "loss": 0.3766, + "step": 25613 + }, + { + "epoch": 1.4310975780093305, + "grad_norm": 0.7384442090988159, + "learning_rate": 0.00028583034513671, + "loss": 0.4942, + "step": 25614 + }, + { + "epoch": 1.4311534486130122, + "grad_norm": 1.4157463312149048, + "learning_rate": 0.0002858023307933662, + "loss": 0.5301, + "step": 25615 + }, + { + "epoch": 1.4312093192166941, + "grad_norm": 1.4259231090545654, + "learning_rate": 0.00028577431645002247, + "loss": 0.4313, + "step": 25616 + }, + { + "epoch": 1.431265189820376, + "grad_norm": 4.532521724700928, + "learning_rate": 0.0002857463021066786, + "loss": 0.4639, + "step": 25617 + }, + { + "epoch": 1.4313210604240578, + "grad_norm": 0.3932497203350067, + "learning_rate": 0.0002857182877633349, + "loss": 0.4023, + "step": 25618 + }, + { + "epoch": 1.4313769310277398, + "grad_norm": 0.396651953458786, + "learning_rate": 0.00028569027341999103, + "loss": 0.4453, + "step": 25619 + }, + { + "epoch": 1.4314328016314217, + "grad_norm": 0.4059649109840393, + "learning_rate": 0.00028566225907664724, + "loss": 0.4454, + "step": 25620 + }, + { + "epoch": 1.4314886722351035, + "grad_norm": 0.378387987613678, + "learning_rate": 0.00028563424473330344, + "loss": 0.4702, + "step": 25621 + }, + { + "epoch": 1.4315445428387854, + "grad_norm": 3.2000327110290527, + "learning_rate": 0.00028560623038995965, + "loss": 0.4978, + "step": 25622 + }, + { + "epoch": 1.4316004134424674, + "grad_norm": 0.4810851216316223, + "learning_rate": 0.00028557821604661586, + "loss": 0.3795, + "step": 25623 + }, + { + "epoch": 1.431656284046149, + "grad_norm": 0.36724910140037537, + "learning_rate": 0.00028555020170327206, + "loss": 0.4056, + "step": 25624 + }, + { + "epoch": 1.431712154649831, + "grad_norm": 0.3824397027492523, + "learning_rate": 0.0002855221873599283, + "loss": 0.4618, + "step": 25625 + }, + { + "epoch": 1.431768025253513, + "grad_norm": 0.6851532459259033, + "learning_rate": 0.00028549417301658447, + "loss": 0.5065, + "step": 25626 + }, + { + "epoch": 1.4318238958571947, + "grad_norm": 0.3900564908981323, + "learning_rate": 0.00028546615867324073, + "loss": 0.4607, + "step": 25627 + }, + { + "epoch": 1.4318797664608767, + "grad_norm": 0.37553369998931885, + "learning_rate": 0.0002854381443298969, + "loss": 0.3591, + "step": 25628 + }, + { + "epoch": 1.4319356370645586, + "grad_norm": 0.5182043313980103, + "learning_rate": 0.00028541012998655314, + "loss": 0.4555, + "step": 25629 + }, + { + "epoch": 1.4319915076682403, + "grad_norm": 0.513724684715271, + "learning_rate": 0.0002853821156432093, + "loss": 0.4404, + "step": 25630 + }, + { + "epoch": 1.4320473782719223, + "grad_norm": 0.5755179524421692, + "learning_rate": 0.00028535410129986556, + "loss": 0.3356, + "step": 25631 + }, + { + "epoch": 1.432103248875604, + "grad_norm": 0.5537492632865906, + "learning_rate": 0.00028532608695652176, + "loss": 0.4356, + "step": 25632 + }, + { + "epoch": 1.432159119479286, + "grad_norm": 1.6373518705368042, + "learning_rate": 0.00028529807261317797, + "loss": 0.4019, + "step": 25633 + }, + { + "epoch": 1.4322149900829677, + "grad_norm": 0.4362389147281647, + "learning_rate": 0.0002852700582698342, + "loss": 0.4037, + "step": 25634 + }, + { + "epoch": 1.4322708606866497, + "grad_norm": 0.36114662885665894, + "learning_rate": 0.0002852420439264904, + "loss": 0.2995, + "step": 25635 + }, + { + "epoch": 1.4323267312903316, + "grad_norm": 0.6472115516662598, + "learning_rate": 0.0002852140295831466, + "loss": 0.4952, + "step": 25636 + }, + { + "epoch": 1.4323826018940133, + "grad_norm": 0.686759352684021, + "learning_rate": 0.0002851860152398028, + "loss": 0.4496, + "step": 25637 + }, + { + "epoch": 1.4324384724976953, + "grad_norm": 1.1258457899093628, + "learning_rate": 0.000285158000896459, + "loss": 0.6347, + "step": 25638 + }, + { + "epoch": 1.4324943431013772, + "grad_norm": 0.7735172510147095, + "learning_rate": 0.0002851299865531152, + "loss": 0.5243, + "step": 25639 + }, + { + "epoch": 1.432550213705059, + "grad_norm": 0.7587034702301025, + "learning_rate": 0.0002851019722097714, + "loss": 0.3507, + "step": 25640 + }, + { + "epoch": 1.432606084308741, + "grad_norm": 0.5768136382102966, + "learning_rate": 0.0002850739578664276, + "loss": 0.4795, + "step": 25641 + }, + { + "epoch": 1.4326619549124229, + "grad_norm": 0.7939239740371704, + "learning_rate": 0.0002850459435230838, + "loss": 0.3835, + "step": 25642 + }, + { + "epoch": 1.4327178255161046, + "grad_norm": 0.4696371853351593, + "learning_rate": 0.00028501792917974, + "loss": 0.386, + "step": 25643 + }, + { + "epoch": 1.4327736961197866, + "grad_norm": 0.53228759765625, + "learning_rate": 0.00028498991483639623, + "loss": 0.4137, + "step": 25644 + }, + { + "epoch": 1.4328295667234685, + "grad_norm": 0.376912385225296, + "learning_rate": 0.00028496190049305244, + "loss": 0.4353, + "step": 25645 + }, + { + "epoch": 1.4328854373271502, + "grad_norm": 0.3298324942588806, + "learning_rate": 0.00028493388614970864, + "loss": 0.3522, + "step": 25646 + }, + { + "epoch": 1.4329413079308322, + "grad_norm": 0.4238959848880768, + "learning_rate": 0.00028490587180636485, + "loss": 0.4371, + "step": 25647 + }, + { + "epoch": 1.4329971785345141, + "grad_norm": 0.42588573694229126, + "learning_rate": 0.0002848778574630211, + "loss": 0.4674, + "step": 25648 + }, + { + "epoch": 1.4330530491381959, + "grad_norm": 0.42496001720428467, + "learning_rate": 0.00028484984311967726, + "loss": 0.4375, + "step": 25649 + }, + { + "epoch": 1.4331089197418778, + "grad_norm": 0.5343071222305298, + "learning_rate": 0.0002848218287763335, + "loss": 0.4824, + "step": 25650 + }, + { + "epoch": 1.4331647903455598, + "grad_norm": 0.32160282135009766, + "learning_rate": 0.00028479381443298967, + "loss": 0.5047, + "step": 25651 + }, + { + "epoch": 1.4332206609492415, + "grad_norm": 0.41109660267829895, + "learning_rate": 0.00028476580008964593, + "loss": 0.2991, + "step": 25652 + }, + { + "epoch": 1.4332765315529234, + "grad_norm": 0.8022966980934143, + "learning_rate": 0.0002847377857463021, + "loss": 0.3719, + "step": 25653 + }, + { + "epoch": 1.4333324021566054, + "grad_norm": 1.3224780559539795, + "learning_rate": 0.00028470977140295834, + "loss": 0.4212, + "step": 25654 + }, + { + "epoch": 1.4333882727602871, + "grad_norm": 18.47213363647461, + "learning_rate": 0.0002846817570596145, + "loss": 0.4854, + "step": 25655 + }, + { + "epoch": 1.433444143363969, + "grad_norm": 0.4164747893810272, + "learning_rate": 0.00028465374271627076, + "loss": 0.4179, + "step": 25656 + }, + { + "epoch": 1.433500013967651, + "grad_norm": 0.43065935373306274, + "learning_rate": 0.00028462572837292696, + "loss": 0.3656, + "step": 25657 + }, + { + "epoch": 1.4335558845713328, + "grad_norm": 0.6105392575263977, + "learning_rate": 0.00028459771402958317, + "loss": 0.4629, + "step": 25658 + }, + { + "epoch": 1.4336117551750147, + "grad_norm": 1.3146700859069824, + "learning_rate": 0.00028456969968623937, + "loss": 0.4481, + "step": 25659 + }, + { + "epoch": 1.4336676257786967, + "grad_norm": 0.3521309494972229, + "learning_rate": 0.0002845416853428956, + "loss": 0.3635, + "step": 25660 + }, + { + "epoch": 1.4337234963823784, + "grad_norm": 0.40936678647994995, + "learning_rate": 0.0002845136709995518, + "loss": 0.4611, + "step": 25661 + }, + { + "epoch": 1.4337793669860603, + "grad_norm": 0.4082617461681366, + "learning_rate": 0.00028448565665620794, + "loss": 0.3421, + "step": 25662 + }, + { + "epoch": 1.4338352375897423, + "grad_norm": 0.5691753029823303, + "learning_rate": 0.0002844576423128642, + "loss": 0.4964, + "step": 25663 + }, + { + "epoch": 1.433891108193424, + "grad_norm": 0.3810172975063324, + "learning_rate": 0.0002844296279695204, + "loss": 0.4027, + "step": 25664 + }, + { + "epoch": 1.433946978797106, + "grad_norm": 0.3449004888534546, + "learning_rate": 0.0002844016136261766, + "loss": 0.3577, + "step": 25665 + }, + { + "epoch": 1.4340028494007877, + "grad_norm": 0.4047762155532837, + "learning_rate": 0.0002843735992828328, + "loss": 0.3618, + "step": 25666 + }, + { + "epoch": 1.4340587200044697, + "grad_norm": 0.4244455099105835, + "learning_rate": 0.000284345584939489, + "loss": 0.3893, + "step": 25667 + }, + { + "epoch": 1.4341145906081514, + "grad_norm": 0.35627514123916626, + "learning_rate": 0.0002843175705961452, + "loss": 0.5045, + "step": 25668 + }, + { + "epoch": 1.4341704612118333, + "grad_norm": 0.41866984963417053, + "learning_rate": 0.00028428955625280143, + "loss": 0.4358, + "step": 25669 + }, + { + "epoch": 1.4342263318155153, + "grad_norm": 0.9309684634208679, + "learning_rate": 0.00028426154190945764, + "loss": 0.4364, + "step": 25670 + }, + { + "epoch": 1.434282202419197, + "grad_norm": 0.4762499928474426, + "learning_rate": 0.0002842335275661139, + "loss": 0.5662, + "step": 25671 + }, + { + "epoch": 1.434338073022879, + "grad_norm": 0.4706338346004486, + "learning_rate": 0.00028420551322277005, + "loss": 0.337, + "step": 25672 + }, + { + "epoch": 1.434393943626561, + "grad_norm": 0.4233732223510742, + "learning_rate": 0.0002841774988794263, + "loss": 0.3396, + "step": 25673 + }, + { + "epoch": 1.4344498142302426, + "grad_norm": 0.6048129796981812, + "learning_rate": 0.00028414948453608246, + "loss": 0.4036, + "step": 25674 + }, + { + "epoch": 1.4345056848339246, + "grad_norm": 0.7692992687225342, + "learning_rate": 0.0002841214701927387, + "loss": 0.4707, + "step": 25675 + }, + { + "epoch": 1.4345615554376066, + "grad_norm": 0.7843669056892395, + "learning_rate": 0.00028409345584939487, + "loss": 0.457, + "step": 25676 + }, + { + "epoch": 1.4346174260412883, + "grad_norm": 0.5191477537155151, + "learning_rate": 0.00028406544150605113, + "loss": 0.5791, + "step": 25677 + }, + { + "epoch": 1.4346732966449702, + "grad_norm": 0.4497368037700653, + "learning_rate": 0.0002840374271627073, + "loss": 0.4693, + "step": 25678 + }, + { + "epoch": 1.4347291672486522, + "grad_norm": 0.4479810893535614, + "learning_rate": 0.00028400941281936354, + "loss": 0.3945, + "step": 25679 + }, + { + "epoch": 1.434785037852334, + "grad_norm": 0.34298765659332275, + "learning_rate": 0.00028398139847601975, + "loss": 0.4217, + "step": 25680 + }, + { + "epoch": 1.4348409084560159, + "grad_norm": 0.32204097509384155, + "learning_rate": 0.00028395338413267595, + "loss": 0.4179, + "step": 25681 + }, + { + "epoch": 1.4348967790596978, + "grad_norm": 0.5449739098548889, + "learning_rate": 0.00028392536978933216, + "loss": 0.3481, + "step": 25682 + }, + { + "epoch": 1.4349526496633795, + "grad_norm": 0.5109368562698364, + "learning_rate": 0.0002838973554459883, + "loss": 0.3864, + "step": 25683 + }, + { + "epoch": 1.4350085202670615, + "grad_norm": 0.958243727684021, + "learning_rate": 0.00028386934110264457, + "loss": 0.3263, + "step": 25684 + }, + { + "epoch": 1.4350643908707434, + "grad_norm": 0.42391237616539, + "learning_rate": 0.0002838413267593007, + "loss": 0.3785, + "step": 25685 + }, + { + "epoch": 1.4351202614744252, + "grad_norm": 0.6057214736938477, + "learning_rate": 0.000283813312415957, + "loss": 0.3801, + "step": 25686 + }, + { + "epoch": 1.4351761320781071, + "grad_norm": 0.47762197256088257, + "learning_rate": 0.0002837852980726132, + "loss": 0.3457, + "step": 25687 + }, + { + "epoch": 1.435232002681789, + "grad_norm": 0.39959031343460083, + "learning_rate": 0.0002837572837292694, + "loss": 0.3405, + "step": 25688 + }, + { + "epoch": 1.4352878732854708, + "grad_norm": 0.34265220165252686, + "learning_rate": 0.0002837292693859256, + "loss": 0.4169, + "step": 25689 + }, + { + "epoch": 1.4353437438891528, + "grad_norm": 0.49791789054870605, + "learning_rate": 0.0002837012550425818, + "loss": 0.413, + "step": 25690 + }, + { + "epoch": 1.4353996144928347, + "grad_norm": 0.31605565547943115, + "learning_rate": 0.000283673240699238, + "loss": 0.289, + "step": 25691 + }, + { + "epoch": 1.4354554850965164, + "grad_norm": 7.118356704711914, + "learning_rate": 0.0002836452263558942, + "loss": 0.6054, + "step": 25692 + }, + { + "epoch": 1.4355113557001984, + "grad_norm": 3.8606464862823486, + "learning_rate": 0.0002836172120125504, + "loss": 0.3897, + "step": 25693 + }, + { + "epoch": 1.4355672263038803, + "grad_norm": 0.5286182165145874, + "learning_rate": 0.00028358919766920663, + "loss": 0.4415, + "step": 25694 + }, + { + "epoch": 1.435623096907562, + "grad_norm": 2.915724515914917, + "learning_rate": 0.00028356118332586284, + "loss": 0.4108, + "step": 25695 + }, + { + "epoch": 1.435678967511244, + "grad_norm": 0.35722190141677856, + "learning_rate": 0.0002835331689825191, + "loss": 0.3508, + "step": 25696 + }, + { + "epoch": 1.4357348381149257, + "grad_norm": 0.47840359807014465, + "learning_rate": 0.00028350515463917525, + "loss": 0.4664, + "step": 25697 + }, + { + "epoch": 1.4357907087186077, + "grad_norm": 0.4052661657333374, + "learning_rate": 0.0002834771402958315, + "loss": 0.3266, + "step": 25698 + }, + { + "epoch": 1.4358465793222897, + "grad_norm": 2.717552423477173, + "learning_rate": 0.00028344912595248766, + "loss": 0.4496, + "step": 25699 + }, + { + "epoch": 1.4359024499259714, + "grad_norm": 2.1743524074554443, + "learning_rate": 0.0002834211116091439, + "loss": 0.437, + "step": 25700 + }, + { + "epoch": 1.4359583205296533, + "grad_norm": 0.7136190533638, + "learning_rate": 0.00028339309726580007, + "loss": 0.4452, + "step": 25701 + }, + { + "epoch": 1.436014191133335, + "grad_norm": 0.6987035870552063, + "learning_rate": 0.00028336508292245633, + "loss": 0.3904, + "step": 25702 + }, + { + "epoch": 1.436070061737017, + "grad_norm": 0.46045607328414917, + "learning_rate": 0.00028333706857911254, + "loss": 0.3766, + "step": 25703 + }, + { + "epoch": 1.436125932340699, + "grad_norm": 0.41911664605140686, + "learning_rate": 0.0002833090542357687, + "loss": 0.4501, + "step": 25704 + }, + { + "epoch": 1.4361818029443807, + "grad_norm": 0.6499769687652588, + "learning_rate": 0.00028328103989242495, + "loss": 0.3501, + "step": 25705 + }, + { + "epoch": 1.4362376735480626, + "grad_norm": 0.5377281904220581, + "learning_rate": 0.0002832530255490811, + "loss": 0.3721, + "step": 25706 + }, + { + "epoch": 1.4362935441517446, + "grad_norm": 0.40210863947868347, + "learning_rate": 0.00028322501120573736, + "loss": 0.3862, + "step": 25707 + }, + { + "epoch": 1.4363494147554263, + "grad_norm": 0.7827403545379639, + "learning_rate": 0.0002831969968623935, + "loss": 0.5397, + "step": 25708 + }, + { + "epoch": 1.4364052853591083, + "grad_norm": 0.3338254392147064, + "learning_rate": 0.00028316898251904977, + "loss": 0.4261, + "step": 25709 + }, + { + "epoch": 1.4364611559627902, + "grad_norm": 0.5155331492424011, + "learning_rate": 0.0002831409681757059, + "loss": 0.4, + "step": 25710 + }, + { + "epoch": 1.436517026566472, + "grad_norm": 0.4933198392391205, + "learning_rate": 0.0002831129538323622, + "loss": 0.4243, + "step": 25711 + }, + { + "epoch": 1.436572897170154, + "grad_norm": 0.45094597339630127, + "learning_rate": 0.0002830849394890184, + "loss": 0.3446, + "step": 25712 + }, + { + "epoch": 1.4366287677738359, + "grad_norm": 0.6447983384132385, + "learning_rate": 0.0002830569251456746, + "loss": 0.4107, + "step": 25713 + }, + { + "epoch": 1.4366846383775176, + "grad_norm": 0.37861984968185425, + "learning_rate": 0.0002830289108023308, + "loss": 0.4883, + "step": 25714 + }, + { + "epoch": 1.4367405089811995, + "grad_norm": 1.2462234497070312, + "learning_rate": 0.000283000896458987, + "loss": 0.426, + "step": 25715 + }, + { + "epoch": 1.4367963795848815, + "grad_norm": 0.5921631455421448, + "learning_rate": 0.0002829728821156432, + "loss": 0.3698, + "step": 25716 + }, + { + "epoch": 1.4368522501885632, + "grad_norm": 0.4235615134239197, + "learning_rate": 0.0002829448677722994, + "loss": 0.4013, + "step": 25717 + }, + { + "epoch": 1.4369081207922452, + "grad_norm": 5.579555988311768, + "learning_rate": 0.0002829168534289556, + "loss": 0.4661, + "step": 25718 + }, + { + "epoch": 1.4369639913959271, + "grad_norm": 0.5799793004989624, + "learning_rate": 0.0002828888390856119, + "loss": 0.402, + "step": 25719 + }, + { + "epoch": 1.4370198619996088, + "grad_norm": 0.3431016504764557, + "learning_rate": 0.00028286082474226804, + "loss": 0.4793, + "step": 25720 + }, + { + "epoch": 1.4370757326032908, + "grad_norm": 0.5501025319099426, + "learning_rate": 0.0002828328103989243, + "loss": 0.4031, + "step": 25721 + }, + { + "epoch": 1.4371316032069728, + "grad_norm": 0.47957977652549744, + "learning_rate": 0.00028280479605558045, + "loss": 0.3895, + "step": 25722 + }, + { + "epoch": 1.4371874738106545, + "grad_norm": 0.669540286064148, + "learning_rate": 0.0002827767817122367, + "loss": 0.448, + "step": 25723 + }, + { + "epoch": 1.4372433444143364, + "grad_norm": 0.37787190079689026, + "learning_rate": 0.00028274876736889286, + "loss": 0.4436, + "step": 25724 + }, + { + "epoch": 1.4372992150180184, + "grad_norm": 0.4939820170402527, + "learning_rate": 0.00028272075302554907, + "loss": 0.5, + "step": 25725 + }, + { + "epoch": 1.4373550856217001, + "grad_norm": 0.4952257573604584, + "learning_rate": 0.00028269273868220527, + "loss": 0.4593, + "step": 25726 + }, + { + "epoch": 1.437410956225382, + "grad_norm": 0.4698506295681, + "learning_rate": 0.0002826647243388615, + "loss": 0.3813, + "step": 25727 + }, + { + "epoch": 1.437466826829064, + "grad_norm": 0.45937949419021606, + "learning_rate": 0.00028263670999551774, + "loss": 0.4588, + "step": 25728 + }, + { + "epoch": 1.4375226974327457, + "grad_norm": 0.404796838760376, + "learning_rate": 0.0002826086956521739, + "loss": 0.4389, + "step": 25729 + }, + { + "epoch": 1.4375785680364277, + "grad_norm": 0.5282415151596069, + "learning_rate": 0.00028258068130883015, + "loss": 0.6055, + "step": 25730 + }, + { + "epoch": 1.4376344386401094, + "grad_norm": 3.8311920166015625, + "learning_rate": 0.0002825526669654863, + "loss": 0.4873, + "step": 25731 + }, + { + "epoch": 1.4376903092437914, + "grad_norm": 0.4963023066520691, + "learning_rate": 0.00028252465262214256, + "loss": 0.5218, + "step": 25732 + }, + { + "epoch": 1.4377461798474733, + "grad_norm": 0.5332184433937073, + "learning_rate": 0.0002824966382787987, + "loss": 0.4764, + "step": 25733 + }, + { + "epoch": 1.437802050451155, + "grad_norm": 0.47405049204826355, + "learning_rate": 0.00028246862393545497, + "loss": 0.4134, + "step": 25734 + }, + { + "epoch": 1.437857921054837, + "grad_norm": 0.412855327129364, + "learning_rate": 0.0002824406095921112, + "loss": 0.5106, + "step": 25735 + }, + { + "epoch": 1.4379137916585187, + "grad_norm": 0.48093298077583313, + "learning_rate": 0.0002824125952487674, + "loss": 0.4016, + "step": 25736 + }, + { + "epoch": 1.4379696622622007, + "grad_norm": 0.6421946883201599, + "learning_rate": 0.0002823845809054236, + "loss": 0.5148, + "step": 25737 + }, + { + "epoch": 1.4380255328658826, + "grad_norm": 0.7398240566253662, + "learning_rate": 0.0002823565665620798, + "loss": 0.3103, + "step": 25738 + }, + { + "epoch": 1.4380814034695644, + "grad_norm": 0.40673282742500305, + "learning_rate": 0.000282328552218736, + "loss": 0.3717, + "step": 25739 + }, + { + "epoch": 1.4381372740732463, + "grad_norm": 0.6428828239440918, + "learning_rate": 0.0002823005378753922, + "loss": 0.4354, + "step": 25740 + }, + { + "epoch": 1.4381931446769283, + "grad_norm": 4.97114896774292, + "learning_rate": 0.0002822725235320484, + "loss": 0.4738, + "step": 25741 + }, + { + "epoch": 1.43824901528061, + "grad_norm": 0.5536156892776489, + "learning_rate": 0.00028224450918870467, + "loss": 0.4937, + "step": 25742 + }, + { + "epoch": 1.438304885884292, + "grad_norm": 0.3495059013366699, + "learning_rate": 0.0002822164948453608, + "loss": 0.4552, + "step": 25743 + }, + { + "epoch": 1.438360756487974, + "grad_norm": 0.42387720942497253, + "learning_rate": 0.0002821884805020171, + "loss": 0.3967, + "step": 25744 + }, + { + "epoch": 1.4384166270916556, + "grad_norm": 0.3929635286331177, + "learning_rate": 0.00028216046615867324, + "loss": 0.3642, + "step": 25745 + }, + { + "epoch": 1.4384724976953376, + "grad_norm": 5.30753231048584, + "learning_rate": 0.00028213245181532944, + "loss": 0.3664, + "step": 25746 + }, + { + "epoch": 1.4385283682990195, + "grad_norm": 0.5126060247421265, + "learning_rate": 0.00028210443747198565, + "loss": 0.4421, + "step": 25747 + }, + { + "epoch": 1.4385842389027013, + "grad_norm": 0.5700439214706421, + "learning_rate": 0.00028207642312864185, + "loss": 0.4872, + "step": 25748 + }, + { + "epoch": 1.4386401095063832, + "grad_norm": 0.3755549490451813, + "learning_rate": 0.00028204840878529806, + "loss": 0.4786, + "step": 25749 + }, + { + "epoch": 1.4386959801100652, + "grad_norm": 0.3953537344932556, + "learning_rate": 0.00028202039444195426, + "loss": 0.4714, + "step": 25750 + }, + { + "epoch": 1.438751850713747, + "grad_norm": 0.6060662865638733, + "learning_rate": 0.0002819923800986105, + "loss": 0.3291, + "step": 25751 + }, + { + "epoch": 1.4388077213174288, + "grad_norm": 0.41647040843963623, + "learning_rate": 0.0002819643657552667, + "loss": 0.4362, + "step": 25752 + }, + { + "epoch": 1.4388635919211108, + "grad_norm": 0.3750888407230377, + "learning_rate": 0.00028193635141192294, + "loss": 0.4424, + "step": 25753 + }, + { + "epoch": 1.4389194625247925, + "grad_norm": 0.4093138575553894, + "learning_rate": 0.0002819083370685791, + "loss": 0.5137, + "step": 25754 + }, + { + "epoch": 1.4389753331284745, + "grad_norm": 0.6382489204406738, + "learning_rate": 0.00028188032272523535, + "loss": 0.3441, + "step": 25755 + }, + { + "epoch": 1.4390312037321564, + "grad_norm": 3.389427661895752, + "learning_rate": 0.0002818523083818915, + "loss": 0.4706, + "step": 25756 + }, + { + "epoch": 1.4390870743358382, + "grad_norm": 0.8003926873207092, + "learning_rate": 0.00028182429403854776, + "loss": 0.507, + "step": 25757 + }, + { + "epoch": 1.43914294493952, + "grad_norm": 4.2473907470703125, + "learning_rate": 0.00028179627969520397, + "loss": 0.5458, + "step": 25758 + }, + { + "epoch": 1.439198815543202, + "grad_norm": 0.5231404304504395, + "learning_rate": 0.00028176826535186017, + "loss": 0.59, + "step": 25759 + }, + { + "epoch": 1.4392546861468838, + "grad_norm": 0.5320442914962769, + "learning_rate": 0.0002817402510085164, + "loss": 0.4025, + "step": 25760 + }, + { + "epoch": 1.4393105567505657, + "grad_norm": 0.4367045760154724, + "learning_rate": 0.0002817122366651726, + "loss": 0.3893, + "step": 25761 + }, + { + "epoch": 1.4393664273542477, + "grad_norm": 0.46448367834091187, + "learning_rate": 0.0002816842223218288, + "loss": 0.4333, + "step": 25762 + }, + { + "epoch": 1.4394222979579294, + "grad_norm": 0.42958909273147583, + "learning_rate": 0.000281656207978485, + "loss": 0.3717, + "step": 25763 + }, + { + "epoch": 1.4394781685616114, + "grad_norm": 0.49900203943252563, + "learning_rate": 0.0002816281936351412, + "loss": 0.3032, + "step": 25764 + }, + { + "epoch": 1.439534039165293, + "grad_norm": 0.27316176891326904, + "learning_rate": 0.0002816001792917974, + "loss": 0.3189, + "step": 25765 + }, + { + "epoch": 1.439589909768975, + "grad_norm": 0.543822705745697, + "learning_rate": 0.0002815721649484536, + "loss": 0.3752, + "step": 25766 + }, + { + "epoch": 1.439645780372657, + "grad_norm": 0.47034740447998047, + "learning_rate": 0.0002815441506051098, + "loss": 0.4187, + "step": 25767 + }, + { + "epoch": 1.4397016509763387, + "grad_norm": 0.3178834021091461, + "learning_rate": 0.000281516136261766, + "loss": 0.4019, + "step": 25768 + }, + { + "epoch": 1.4397575215800207, + "grad_norm": 0.3402908742427826, + "learning_rate": 0.00028148812191842223, + "loss": 0.4278, + "step": 25769 + }, + { + "epoch": 1.4398133921837024, + "grad_norm": 0.4646540582180023, + "learning_rate": 0.00028146010757507844, + "loss": 0.4702, + "step": 25770 + }, + { + "epoch": 1.4398692627873844, + "grad_norm": 0.41966912150382996, + "learning_rate": 0.00028143209323173464, + "loss": 0.4134, + "step": 25771 + }, + { + "epoch": 1.4399251333910663, + "grad_norm": 0.7642876505851746, + "learning_rate": 0.00028140407888839085, + "loss": 0.3637, + "step": 25772 + }, + { + "epoch": 1.439981003994748, + "grad_norm": 0.42575204372406006, + "learning_rate": 0.00028137606454504705, + "loss": 0.4237, + "step": 25773 + }, + { + "epoch": 1.44003687459843, + "grad_norm": 0.5192739963531494, + "learning_rate": 0.0002813480502017033, + "loss": 0.4013, + "step": 25774 + }, + { + "epoch": 1.440092745202112, + "grad_norm": 0.783706784248352, + "learning_rate": 0.00028132003585835946, + "loss": 0.3889, + "step": 25775 + }, + { + "epoch": 1.4401486158057937, + "grad_norm": 0.4173177480697632, + "learning_rate": 0.0002812920215150157, + "loss": 0.4044, + "step": 25776 + }, + { + "epoch": 1.4402044864094756, + "grad_norm": 0.5970402956008911, + "learning_rate": 0.0002812640071716719, + "loss": 0.4586, + "step": 25777 + }, + { + "epoch": 1.4402603570131576, + "grad_norm": 0.3617291748523712, + "learning_rate": 0.00028123599282832814, + "loss": 0.46, + "step": 25778 + }, + { + "epoch": 1.4403162276168393, + "grad_norm": 0.9968316555023193, + "learning_rate": 0.0002812079784849843, + "loss": 0.3466, + "step": 25779 + }, + { + "epoch": 1.4403720982205213, + "grad_norm": 3.154629945755005, + "learning_rate": 0.00028117996414164055, + "loss": 0.3668, + "step": 25780 + }, + { + "epoch": 1.4404279688242032, + "grad_norm": 2.996654987335205, + "learning_rate": 0.0002811519497982967, + "loss": 0.3801, + "step": 25781 + }, + { + "epoch": 1.440483839427885, + "grad_norm": 0.8468272089958191, + "learning_rate": 0.00028112393545495296, + "loss": 0.4514, + "step": 25782 + }, + { + "epoch": 1.440539710031567, + "grad_norm": 0.6900931596755981, + "learning_rate": 0.00028109592111160916, + "loss": 0.4599, + "step": 25783 + }, + { + "epoch": 1.4405955806352488, + "grad_norm": 0.5172414779663086, + "learning_rate": 0.00028106790676826537, + "loss": 0.3996, + "step": 25784 + }, + { + "epoch": 1.4406514512389306, + "grad_norm": 0.44374504685401917, + "learning_rate": 0.0002810398924249216, + "loss": 0.4459, + "step": 25785 + }, + { + "epoch": 1.4407073218426125, + "grad_norm": 0.6314961910247803, + "learning_rate": 0.0002810118780815778, + "loss": 0.5287, + "step": 25786 + }, + { + "epoch": 1.4407631924462945, + "grad_norm": 0.4883423447608948, + "learning_rate": 0.000280983863738234, + "loss": 0.5633, + "step": 25787 + }, + { + "epoch": 1.4408190630499762, + "grad_norm": 0.8612371683120728, + "learning_rate": 0.0002809558493948902, + "loss": 0.3371, + "step": 25788 + }, + { + "epoch": 1.4408749336536582, + "grad_norm": 3.639069080352783, + "learning_rate": 0.0002809278350515464, + "loss": 0.4384, + "step": 25789 + }, + { + "epoch": 1.44093080425734, + "grad_norm": 0.9355509877204895, + "learning_rate": 0.0002808998207082026, + "loss": 0.3646, + "step": 25790 + }, + { + "epoch": 1.4409866748610218, + "grad_norm": 0.46318912506103516, + "learning_rate": 0.0002808718063648588, + "loss": 0.3653, + "step": 25791 + }, + { + "epoch": 1.4410425454647038, + "grad_norm": 0.33467572927474976, + "learning_rate": 0.000280843792021515, + "loss": 0.4163, + "step": 25792 + }, + { + "epoch": 1.4410984160683857, + "grad_norm": 0.3771304190158844, + "learning_rate": 0.0002808157776781712, + "loss": 0.5369, + "step": 25793 + }, + { + "epoch": 1.4411542866720675, + "grad_norm": 0.4003654420375824, + "learning_rate": 0.00028078776333482743, + "loss": 0.3844, + "step": 25794 + }, + { + "epoch": 1.4412101572757494, + "grad_norm": 0.4106546938419342, + "learning_rate": 0.00028075974899148363, + "loss": 0.3443, + "step": 25795 + }, + { + "epoch": 1.4412660278794314, + "grad_norm": 0.3353961110115051, + "learning_rate": 0.00028073173464813984, + "loss": 0.4518, + "step": 25796 + }, + { + "epoch": 1.441321898483113, + "grad_norm": 0.839235246181488, + "learning_rate": 0.00028070372030479605, + "loss": 0.512, + "step": 25797 + }, + { + "epoch": 1.441377769086795, + "grad_norm": 1.758514404296875, + "learning_rate": 0.00028067570596145225, + "loss": 0.3244, + "step": 25798 + }, + { + "epoch": 1.4414336396904768, + "grad_norm": 5.077045917510986, + "learning_rate": 0.0002806476916181085, + "loss": 0.3972, + "step": 25799 + }, + { + "epoch": 1.4414895102941587, + "grad_norm": 0.4326013922691345, + "learning_rate": 0.00028061967727476466, + "loss": 0.446, + "step": 25800 + }, + { + "epoch": 1.4415453808978407, + "grad_norm": 0.47398751974105835, + "learning_rate": 0.0002805916629314209, + "loss": 0.3613, + "step": 25801 + }, + { + "epoch": 1.4416012515015224, + "grad_norm": 0.33436810970306396, + "learning_rate": 0.0002805636485880771, + "loss": 0.3353, + "step": 25802 + }, + { + "epoch": 1.4416571221052044, + "grad_norm": 0.4418369233608246, + "learning_rate": 0.00028053563424473334, + "loss": 0.4005, + "step": 25803 + }, + { + "epoch": 1.441712992708886, + "grad_norm": 0.48347947001457214, + "learning_rate": 0.0002805076199013895, + "loss": 0.3846, + "step": 25804 + }, + { + "epoch": 1.441768863312568, + "grad_norm": 0.27763688564300537, + "learning_rate": 0.00028047960555804575, + "loss": 0.3159, + "step": 25805 + }, + { + "epoch": 1.44182473391625, + "grad_norm": 0.5690096020698547, + "learning_rate": 0.00028045159121470195, + "loss": 0.3591, + "step": 25806 + }, + { + "epoch": 1.4418806045199317, + "grad_norm": 1.0925372838974, + "learning_rate": 0.00028042357687135816, + "loss": 0.3966, + "step": 25807 + }, + { + "epoch": 1.4419364751236137, + "grad_norm": 0.7413789629936218, + "learning_rate": 0.00028039556252801436, + "loss": 0.4386, + "step": 25808 + }, + { + "epoch": 1.4419923457272956, + "grad_norm": 0.7861416339874268, + "learning_rate": 0.00028036754818467057, + "loss": 0.3825, + "step": 25809 + }, + { + "epoch": 1.4420482163309774, + "grad_norm": 0.5086455345153809, + "learning_rate": 0.0002803395338413268, + "loss": 0.437, + "step": 25810 + }, + { + "epoch": 1.4421040869346593, + "grad_norm": 2.2084736824035645, + "learning_rate": 0.00028031151949798293, + "loss": 0.4322, + "step": 25811 + }, + { + "epoch": 1.4421599575383413, + "grad_norm": 0.5100639462471008, + "learning_rate": 0.0002802835051546392, + "loss": 0.5214, + "step": 25812 + }, + { + "epoch": 1.442215828142023, + "grad_norm": 0.4597476124763489, + "learning_rate": 0.00028025549081129534, + "loss": 0.4411, + "step": 25813 + }, + { + "epoch": 1.442271698745705, + "grad_norm": 0.3360564410686493, + "learning_rate": 0.0002802274764679516, + "loss": 0.3551, + "step": 25814 + }, + { + "epoch": 1.4423275693493869, + "grad_norm": 0.8032421469688416, + "learning_rate": 0.0002801994621246078, + "loss": 0.5624, + "step": 25815 + }, + { + "epoch": 1.4423834399530686, + "grad_norm": 0.8069425821304321, + "learning_rate": 0.000280171447781264, + "loss": 0.4815, + "step": 25816 + }, + { + "epoch": 1.4424393105567506, + "grad_norm": 0.5405459403991699, + "learning_rate": 0.0002801434334379202, + "loss": 0.4111, + "step": 25817 + }, + { + "epoch": 1.4424951811604325, + "grad_norm": 0.45779159665107727, + "learning_rate": 0.0002801154190945764, + "loss": 0.3959, + "step": 25818 + }, + { + "epoch": 1.4425510517641142, + "grad_norm": 0.5164960026741028, + "learning_rate": 0.00028008740475123263, + "loss": 0.5724, + "step": 25819 + }, + { + "epoch": 1.4426069223677962, + "grad_norm": 0.4927295744419098, + "learning_rate": 0.00028005939040788883, + "loss": 0.5555, + "step": 25820 + }, + { + "epoch": 1.4426627929714781, + "grad_norm": 0.47097262740135193, + "learning_rate": 0.00028003137606454504, + "loss": 0.4465, + "step": 25821 + }, + { + "epoch": 1.4427186635751599, + "grad_norm": 0.4650045335292816, + "learning_rate": 0.0002800033617212013, + "loss": 0.4472, + "step": 25822 + }, + { + "epoch": 1.4427745341788418, + "grad_norm": 0.3566451668739319, + "learning_rate": 0.00027997534737785745, + "loss": 0.4346, + "step": 25823 + }, + { + "epoch": 1.4428304047825238, + "grad_norm": 0.3844127357006073, + "learning_rate": 0.0002799473330345137, + "loss": 0.3833, + "step": 25824 + }, + { + "epoch": 1.4428862753862055, + "grad_norm": 0.49906766414642334, + "learning_rate": 0.00027991931869116986, + "loss": 0.4442, + "step": 25825 + }, + { + "epoch": 1.4429421459898875, + "grad_norm": 0.6098069548606873, + "learning_rate": 0.0002798913043478261, + "loss": 0.4043, + "step": 25826 + }, + { + "epoch": 1.4429980165935694, + "grad_norm": 0.4441903233528137, + "learning_rate": 0.0002798632900044823, + "loss": 0.422, + "step": 25827 + }, + { + "epoch": 1.4430538871972511, + "grad_norm": 0.3718082010746002, + "learning_rate": 0.00027983527566113854, + "loss": 0.3599, + "step": 25828 + }, + { + "epoch": 1.443109757800933, + "grad_norm": 9.138996124267578, + "learning_rate": 0.00027980726131779474, + "loss": 0.3958, + "step": 25829 + }, + { + "epoch": 1.443165628404615, + "grad_norm": 0.44802001118659973, + "learning_rate": 0.00027977924697445095, + "loss": 0.3717, + "step": 25830 + }, + { + "epoch": 1.4432214990082968, + "grad_norm": 0.4017542898654938, + "learning_rate": 0.00027975123263110715, + "loss": 0.4035, + "step": 25831 + }, + { + "epoch": 1.4432773696119787, + "grad_norm": 0.39042386412620544, + "learning_rate": 0.0002797232182877633, + "loss": 0.4457, + "step": 25832 + }, + { + "epoch": 1.4433332402156605, + "grad_norm": 1.1134326457977295, + "learning_rate": 0.00027969520394441956, + "loss": 0.4525, + "step": 25833 + }, + { + "epoch": 1.4433891108193424, + "grad_norm": 1.0425100326538086, + "learning_rate": 0.0002796671896010757, + "loss": 0.5501, + "step": 25834 + }, + { + "epoch": 1.4434449814230244, + "grad_norm": 1.7370672225952148, + "learning_rate": 0.000279639175257732, + "loss": 0.3489, + "step": 25835 + }, + { + "epoch": 1.443500852026706, + "grad_norm": 0.3866599500179291, + "learning_rate": 0.00027961116091438813, + "loss": 0.3613, + "step": 25836 + }, + { + "epoch": 1.443556722630388, + "grad_norm": 0.5685406923294067, + "learning_rate": 0.0002795831465710444, + "loss": 0.4039, + "step": 25837 + }, + { + "epoch": 1.4436125932340698, + "grad_norm": 0.33820387721061707, + "learning_rate": 0.0002795551322277006, + "loss": 0.4625, + "step": 25838 + }, + { + "epoch": 1.4436684638377517, + "grad_norm": 0.7385107278823853, + "learning_rate": 0.0002795271178843568, + "loss": 0.4541, + "step": 25839 + }, + { + "epoch": 1.4437243344414337, + "grad_norm": 0.3130318522453308, + "learning_rate": 0.000279499103541013, + "loss": 0.3363, + "step": 25840 + }, + { + "epoch": 1.4437802050451154, + "grad_norm": 1.5798331499099731, + "learning_rate": 0.0002794710891976692, + "loss": 0.3338, + "step": 25841 + }, + { + "epoch": 1.4438360756487973, + "grad_norm": 0.772828221321106, + "learning_rate": 0.0002794430748543254, + "loss": 0.5598, + "step": 25842 + }, + { + "epoch": 1.4438919462524793, + "grad_norm": 0.5065878033638, + "learning_rate": 0.0002794150605109816, + "loss": 0.3583, + "step": 25843 + }, + { + "epoch": 1.443947816856161, + "grad_norm": 0.3707405924797058, + "learning_rate": 0.00027938704616763783, + "loss": 0.3032, + "step": 25844 + }, + { + "epoch": 1.444003687459843, + "grad_norm": 0.37432861328125, + "learning_rate": 0.0002793590318242941, + "loss": 0.4059, + "step": 25845 + }, + { + "epoch": 1.444059558063525, + "grad_norm": 0.9331777095794678, + "learning_rate": 0.00027933101748095024, + "loss": 0.4725, + "step": 25846 + }, + { + "epoch": 1.4441154286672067, + "grad_norm": 0.6486799716949463, + "learning_rate": 0.0002793030031376065, + "loss": 0.4048, + "step": 25847 + }, + { + "epoch": 1.4441712992708886, + "grad_norm": 0.6591295003890991, + "learning_rate": 0.00027927498879426265, + "loss": 0.4642, + "step": 25848 + }, + { + "epoch": 1.4442271698745706, + "grad_norm": 0.43216508626937866, + "learning_rate": 0.0002792469744509189, + "loss": 0.3908, + "step": 25849 + }, + { + "epoch": 1.4442830404782523, + "grad_norm": 0.4329882562160492, + "learning_rate": 0.00027921896010757506, + "loss": 0.427, + "step": 25850 + }, + { + "epoch": 1.4443389110819342, + "grad_norm": 0.6598688960075378, + "learning_rate": 0.0002791909457642313, + "loss": 0.6538, + "step": 25851 + }, + { + "epoch": 1.4443947816856162, + "grad_norm": 0.36005645990371704, + "learning_rate": 0.0002791629314208875, + "loss": 0.442, + "step": 25852 + }, + { + "epoch": 1.444450652289298, + "grad_norm": 0.5382002592086792, + "learning_rate": 0.0002791349170775437, + "loss": 0.4251, + "step": 25853 + }, + { + "epoch": 1.4445065228929799, + "grad_norm": 0.49164873361587524, + "learning_rate": 0.00027910690273419994, + "loss": 0.555, + "step": 25854 + }, + { + "epoch": 1.4445623934966618, + "grad_norm": 0.3399268090724945, + "learning_rate": 0.0002790788883908561, + "loss": 0.3832, + "step": 25855 + }, + { + "epoch": 1.4446182641003436, + "grad_norm": 0.3754825294017792, + "learning_rate": 0.00027905087404751235, + "loss": 0.3274, + "step": 25856 + }, + { + "epoch": 1.4446741347040255, + "grad_norm": 0.49084705114364624, + "learning_rate": 0.0002790228597041685, + "loss": 0.4017, + "step": 25857 + }, + { + "epoch": 1.4447300053077075, + "grad_norm": 3.7324798107147217, + "learning_rate": 0.00027899484536082476, + "loss": 0.3167, + "step": 25858 + }, + { + "epoch": 1.4447858759113892, + "grad_norm": 2.1274807453155518, + "learning_rate": 0.0002789668310174809, + "loss": 0.4384, + "step": 25859 + }, + { + "epoch": 1.4448417465150711, + "grad_norm": 1.9736326932907104, + "learning_rate": 0.0002789388166741372, + "loss": 0.7037, + "step": 25860 + }, + { + "epoch": 1.444897617118753, + "grad_norm": 1.0530766248703003, + "learning_rate": 0.0002789108023307934, + "loss": 0.438, + "step": 25861 + }, + { + "epoch": 1.4449534877224348, + "grad_norm": 0.508939802646637, + "learning_rate": 0.0002788827879874496, + "loss": 0.3784, + "step": 25862 + }, + { + "epoch": 1.4450093583261168, + "grad_norm": 0.32777395844459534, + "learning_rate": 0.0002788547736441058, + "loss": 0.325, + "step": 25863 + }, + { + "epoch": 1.4450652289297987, + "grad_norm": 1.037164330482483, + "learning_rate": 0.000278826759300762, + "loss": 0.5592, + "step": 25864 + }, + { + "epoch": 1.4451210995334804, + "grad_norm": 0.6736197471618652, + "learning_rate": 0.0002787987449574182, + "loss": 0.4901, + "step": 25865 + }, + { + "epoch": 1.4451769701371624, + "grad_norm": 0.7057804465293884, + "learning_rate": 0.0002787707306140744, + "loss": 0.5113, + "step": 25866 + }, + { + "epoch": 1.4452328407408441, + "grad_norm": 0.42618313431739807, + "learning_rate": 0.0002787427162707306, + "loss": 0.3126, + "step": 25867 + }, + { + "epoch": 1.445288711344526, + "grad_norm": 0.5441707968711853, + "learning_rate": 0.0002787147019273868, + "loss": 0.4017, + "step": 25868 + }, + { + "epoch": 1.445344581948208, + "grad_norm": 0.8736864328384399, + "learning_rate": 0.00027868668758404303, + "loss": 0.5468, + "step": 25869 + }, + { + "epoch": 1.4454004525518898, + "grad_norm": 0.5235812067985535, + "learning_rate": 0.0002786586732406993, + "loss": 0.4912, + "step": 25870 + }, + { + "epoch": 1.4454563231555717, + "grad_norm": 0.43292486667633057, + "learning_rate": 0.00027863065889735544, + "loss": 0.3484, + "step": 25871 + }, + { + "epoch": 1.4455121937592534, + "grad_norm": 1.3673516511917114, + "learning_rate": 0.0002786026445540117, + "loss": 0.4448, + "step": 25872 + }, + { + "epoch": 1.4455680643629354, + "grad_norm": 0.4635019302368164, + "learning_rate": 0.00027857463021066785, + "loss": 0.4753, + "step": 25873 + }, + { + "epoch": 1.4456239349666173, + "grad_norm": 0.3860451281070709, + "learning_rate": 0.00027854661586732406, + "loss": 0.4041, + "step": 25874 + }, + { + "epoch": 1.445679805570299, + "grad_norm": 0.4396926760673523, + "learning_rate": 0.00027851860152398026, + "loss": 0.5607, + "step": 25875 + }, + { + "epoch": 1.445735676173981, + "grad_norm": 0.6736778616905212, + "learning_rate": 0.00027849058718063647, + "loss": 0.4121, + "step": 25876 + }, + { + "epoch": 1.445791546777663, + "grad_norm": 0.443109929561615, + "learning_rate": 0.00027846257283729273, + "loss": 0.4054, + "step": 25877 + }, + { + "epoch": 1.4458474173813447, + "grad_norm": 0.3522387146949768, + "learning_rate": 0.0002784345584939489, + "loss": 0.5022, + "step": 25878 + }, + { + "epoch": 1.4459032879850267, + "grad_norm": 0.5051218867301941, + "learning_rate": 0.00027840654415060514, + "loss": 0.3904, + "step": 25879 + }, + { + "epoch": 1.4459591585887086, + "grad_norm": 0.5098929405212402, + "learning_rate": 0.0002783785298072613, + "loss": 0.4346, + "step": 25880 + }, + { + "epoch": 1.4460150291923903, + "grad_norm": 0.3898889422416687, + "learning_rate": 0.00027835051546391755, + "loss": 0.4917, + "step": 25881 + }, + { + "epoch": 1.4460708997960723, + "grad_norm": 1.2048770189285278, + "learning_rate": 0.0002783225011205737, + "loss": 0.4165, + "step": 25882 + }, + { + "epoch": 1.4461267703997542, + "grad_norm": 4.578648090362549, + "learning_rate": 0.00027829448677722996, + "loss": 0.3543, + "step": 25883 + }, + { + "epoch": 1.446182641003436, + "grad_norm": 0.490121990442276, + "learning_rate": 0.0002782664724338861, + "loss": 0.3784, + "step": 25884 + }, + { + "epoch": 1.446238511607118, + "grad_norm": 0.3235437273979187, + "learning_rate": 0.0002782384580905424, + "loss": 0.3745, + "step": 25885 + }, + { + "epoch": 1.4462943822107999, + "grad_norm": 4.760403633117676, + "learning_rate": 0.0002782104437471986, + "loss": 0.4349, + "step": 25886 + }, + { + "epoch": 1.4463502528144816, + "grad_norm": 0.4807650148868561, + "learning_rate": 0.0002781824294038548, + "loss": 0.3139, + "step": 25887 + }, + { + "epoch": 1.4464061234181635, + "grad_norm": 0.4217633903026581, + "learning_rate": 0.000278154415060511, + "loss": 0.4187, + "step": 25888 + }, + { + "epoch": 1.4464619940218455, + "grad_norm": 0.3995607793331146, + "learning_rate": 0.0002781264007171672, + "loss": 0.4421, + "step": 25889 + }, + { + "epoch": 1.4465178646255272, + "grad_norm": 2.61922287940979, + "learning_rate": 0.0002780983863738234, + "loss": 0.4578, + "step": 25890 + }, + { + "epoch": 1.4465737352292092, + "grad_norm": 0.47092410922050476, + "learning_rate": 0.0002780703720304796, + "loss": 0.4539, + "step": 25891 + }, + { + "epoch": 1.4466296058328911, + "grad_norm": 0.41976815462112427, + "learning_rate": 0.0002780423576871358, + "loss": 0.5587, + "step": 25892 + }, + { + "epoch": 1.4466854764365729, + "grad_norm": 0.5950008630752563, + "learning_rate": 0.0002780143433437921, + "loss": 0.4441, + "step": 25893 + }, + { + "epoch": 1.4467413470402548, + "grad_norm": 0.7094340920448303, + "learning_rate": 0.00027798632900044823, + "loss": 0.4374, + "step": 25894 + }, + { + "epoch": 1.4467972176439368, + "grad_norm": 0.3659481108188629, + "learning_rate": 0.00027795831465710443, + "loss": 0.4959, + "step": 25895 + }, + { + "epoch": 1.4468530882476185, + "grad_norm": 0.41396671533584595, + "learning_rate": 0.00027793030031376064, + "loss": 0.4631, + "step": 25896 + }, + { + "epoch": 1.4469089588513004, + "grad_norm": 0.5116416811943054, + "learning_rate": 0.00027790228597041684, + "loss": 0.3512, + "step": 25897 + }, + { + "epoch": 1.4469648294549824, + "grad_norm": 0.35110318660736084, + "learning_rate": 0.00027787427162707305, + "loss": 0.4168, + "step": 25898 + }, + { + "epoch": 1.4470207000586641, + "grad_norm": 0.4483712315559387, + "learning_rate": 0.00027784625728372926, + "loss": 0.3728, + "step": 25899 + }, + { + "epoch": 1.447076570662346, + "grad_norm": 0.6483471393585205, + "learning_rate": 0.0002778182429403855, + "loss": 0.551, + "step": 25900 + }, + { + "epoch": 1.4471324412660278, + "grad_norm": 0.44197601079940796, + "learning_rate": 0.00027779022859704167, + "loss": 0.4288, + "step": 25901 + }, + { + "epoch": 1.4471883118697098, + "grad_norm": 0.6049637198448181, + "learning_rate": 0.00027776221425369793, + "loss": 0.6515, + "step": 25902 + }, + { + "epoch": 1.4472441824733917, + "grad_norm": 0.33587414026260376, + "learning_rate": 0.0002777341999103541, + "loss": 0.3405, + "step": 25903 + }, + { + "epoch": 1.4473000530770734, + "grad_norm": 0.4622876048088074, + "learning_rate": 0.00027770618556701034, + "loss": 0.3819, + "step": 25904 + }, + { + "epoch": 1.4473559236807554, + "grad_norm": 0.43857622146606445, + "learning_rate": 0.0002776781712236665, + "loss": 0.437, + "step": 25905 + }, + { + "epoch": 1.4474117942844371, + "grad_norm": 0.48071736097335815, + "learning_rate": 0.00027765015688032275, + "loss": 0.5615, + "step": 25906 + }, + { + "epoch": 1.447467664888119, + "grad_norm": 0.43923094868659973, + "learning_rate": 0.0002776221425369789, + "loss": 0.3637, + "step": 25907 + }, + { + "epoch": 1.447523535491801, + "grad_norm": 1.0419713258743286, + "learning_rate": 0.00027759412819363516, + "loss": 0.4525, + "step": 25908 + }, + { + "epoch": 1.4475794060954827, + "grad_norm": 0.40748468041419983, + "learning_rate": 0.00027756611385029137, + "loss": 0.5189, + "step": 25909 + }, + { + "epoch": 1.4476352766991647, + "grad_norm": 0.6802159547805786, + "learning_rate": 0.0002775380995069476, + "loss": 0.3985, + "step": 25910 + }, + { + "epoch": 1.4476911473028466, + "grad_norm": 1.103584885597229, + "learning_rate": 0.0002775100851636038, + "loss": 0.4462, + "step": 25911 + }, + { + "epoch": 1.4477470179065284, + "grad_norm": 0.7355383038520813, + "learning_rate": 0.00027748207082026, + "loss": 0.3719, + "step": 25912 + }, + { + "epoch": 1.4478028885102103, + "grad_norm": 0.4663538634777069, + "learning_rate": 0.0002774540564769162, + "loss": 0.4366, + "step": 25913 + }, + { + "epoch": 1.4478587591138923, + "grad_norm": 0.4733201861381531, + "learning_rate": 0.0002774260421335724, + "loss": 0.5056, + "step": 25914 + }, + { + "epoch": 1.447914629717574, + "grad_norm": 0.8254550695419312, + "learning_rate": 0.0002773980277902286, + "loss": 0.4309, + "step": 25915 + }, + { + "epoch": 1.447970500321256, + "grad_norm": 0.37905392050743103, + "learning_rate": 0.0002773700134468848, + "loss": 0.3466, + "step": 25916 + }, + { + "epoch": 1.448026370924938, + "grad_norm": 0.5183680653572083, + "learning_rate": 0.000277341999103541, + "loss": 0.5104, + "step": 25917 + }, + { + "epoch": 1.4480822415286196, + "grad_norm": 0.3524991571903229, + "learning_rate": 0.0002773139847601972, + "loss": 0.4066, + "step": 25918 + }, + { + "epoch": 1.4481381121323016, + "grad_norm": 6.587028503417969, + "learning_rate": 0.0002772859704168534, + "loss": 0.4106, + "step": 25919 + }, + { + "epoch": 1.4481939827359835, + "grad_norm": 0.7168499231338501, + "learning_rate": 0.00027725795607350963, + "loss": 0.4259, + "step": 25920 + }, + { + "epoch": 1.4482498533396653, + "grad_norm": 0.7212789058685303, + "learning_rate": 0.00027722994173016584, + "loss": 0.455, + "step": 25921 + }, + { + "epoch": 1.4483057239433472, + "grad_norm": 0.4659196436405182, + "learning_rate": 0.00027720192738682204, + "loss": 0.356, + "step": 25922 + }, + { + "epoch": 1.4483615945470292, + "grad_norm": 0.43143609166145325, + "learning_rate": 0.00027717391304347825, + "loss": 0.5291, + "step": 25923 + }, + { + "epoch": 1.448417465150711, + "grad_norm": 1.0044957399368286, + "learning_rate": 0.00027714589870013446, + "loss": 0.4744, + "step": 25924 + }, + { + "epoch": 1.4484733357543929, + "grad_norm": 2.0654313564300537, + "learning_rate": 0.0002771178843567907, + "loss": 0.3026, + "step": 25925 + }, + { + "epoch": 1.4485292063580748, + "grad_norm": 0.6870397329330444, + "learning_rate": 0.00027708987001344687, + "loss": 0.4077, + "step": 25926 + }, + { + "epoch": 1.4485850769617565, + "grad_norm": 0.4260483682155609, + "learning_rate": 0.00027706185567010313, + "loss": 0.4262, + "step": 25927 + }, + { + "epoch": 1.4486409475654385, + "grad_norm": 0.3788306713104248, + "learning_rate": 0.0002770338413267593, + "loss": 0.3944, + "step": 25928 + }, + { + "epoch": 1.4486968181691204, + "grad_norm": 0.4825703501701355, + "learning_rate": 0.00027700582698341554, + "loss": 0.4357, + "step": 25929 + }, + { + "epoch": 1.4487526887728022, + "grad_norm": 0.6372263431549072, + "learning_rate": 0.0002769778126400717, + "loss": 0.5372, + "step": 25930 + }, + { + "epoch": 1.4488085593764841, + "grad_norm": 0.381690114736557, + "learning_rate": 0.00027694979829672795, + "loss": 0.4368, + "step": 25931 + }, + { + "epoch": 1.448864429980166, + "grad_norm": 0.4775625467300415, + "learning_rate": 0.00027692178395338416, + "loss": 0.3715, + "step": 25932 + }, + { + "epoch": 1.4489203005838478, + "grad_norm": 0.5191200375556946, + "learning_rate": 0.00027689376961004036, + "loss": 0.5044, + "step": 25933 + }, + { + "epoch": 1.4489761711875297, + "grad_norm": 0.47004619240760803, + "learning_rate": 0.00027686575526669657, + "loss": 0.4426, + "step": 25934 + }, + { + "epoch": 1.4490320417912115, + "grad_norm": 0.48735690116882324, + "learning_rate": 0.0002768377409233528, + "loss": 0.4185, + "step": 25935 + }, + { + "epoch": 1.4490879123948934, + "grad_norm": 0.4121435284614563, + "learning_rate": 0.000276809726580009, + "loss": 0.3938, + "step": 25936 + }, + { + "epoch": 1.4491437829985754, + "grad_norm": 0.4103309214115143, + "learning_rate": 0.00027678171223666513, + "loss": 0.3712, + "step": 25937 + }, + { + "epoch": 1.449199653602257, + "grad_norm": 0.8193264603614807, + "learning_rate": 0.0002767536978933214, + "loss": 0.3983, + "step": 25938 + }, + { + "epoch": 1.449255524205939, + "grad_norm": 0.40949127078056335, + "learning_rate": 0.00027672568354997754, + "loss": 0.436, + "step": 25939 + }, + { + "epoch": 1.4493113948096208, + "grad_norm": 0.4749462306499481, + "learning_rate": 0.0002766976692066338, + "loss": 0.4031, + "step": 25940 + }, + { + "epoch": 1.4493672654133027, + "grad_norm": 1.3078604936599731, + "learning_rate": 0.00027666965486329, + "loss": 0.5059, + "step": 25941 + }, + { + "epoch": 1.4494231360169847, + "grad_norm": 4.293729305267334, + "learning_rate": 0.0002766416405199462, + "loss": 0.5056, + "step": 25942 + }, + { + "epoch": 1.4494790066206664, + "grad_norm": 1.2929327487945557, + "learning_rate": 0.0002766136261766024, + "loss": 0.5224, + "step": 25943 + }, + { + "epoch": 1.4495348772243484, + "grad_norm": 0.5666354894638062, + "learning_rate": 0.0002765856118332586, + "loss": 0.434, + "step": 25944 + }, + { + "epoch": 1.4495907478280303, + "grad_norm": 1.9142258167266846, + "learning_rate": 0.00027655759748991483, + "loss": 0.4662, + "step": 25945 + }, + { + "epoch": 1.449646618431712, + "grad_norm": 0.46869006752967834, + "learning_rate": 0.00027652958314657104, + "loss": 0.3661, + "step": 25946 + }, + { + "epoch": 1.449702489035394, + "grad_norm": 0.3499148488044739, + "learning_rate": 0.00027650156880322724, + "loss": 0.3712, + "step": 25947 + }, + { + "epoch": 1.449758359639076, + "grad_norm": 0.5397231578826904, + "learning_rate": 0.0002764735544598835, + "loss": 0.443, + "step": 25948 + }, + { + "epoch": 1.4498142302427577, + "grad_norm": 0.6407825946807861, + "learning_rate": 0.00027644554011653966, + "loss": 0.5393, + "step": 25949 + }, + { + "epoch": 1.4498701008464396, + "grad_norm": 0.4587763845920563, + "learning_rate": 0.0002764175257731959, + "loss": 0.415, + "step": 25950 + }, + { + "epoch": 1.4499259714501216, + "grad_norm": 0.402183473110199, + "learning_rate": 0.00027638951142985207, + "loss": 0.3925, + "step": 25951 + }, + { + "epoch": 1.4499818420538033, + "grad_norm": 0.4355466961860657, + "learning_rate": 0.00027636149708650833, + "loss": 0.432, + "step": 25952 + }, + { + "epoch": 1.4500377126574853, + "grad_norm": 0.36189329624176025, + "learning_rate": 0.0002763334827431645, + "loss": 0.4702, + "step": 25953 + }, + { + "epoch": 1.4500935832611672, + "grad_norm": 0.5015860795974731, + "learning_rate": 0.00027630546839982074, + "loss": 0.3229, + "step": 25954 + }, + { + "epoch": 1.450149453864849, + "grad_norm": 0.8244081139564514, + "learning_rate": 0.00027627745405647694, + "loss": 0.3732, + "step": 25955 + }, + { + "epoch": 1.450205324468531, + "grad_norm": 0.38075315952301025, + "learning_rate": 0.00027624943971313315, + "loss": 0.3729, + "step": 25956 + }, + { + "epoch": 1.4502611950722128, + "grad_norm": 3.840789556503296, + "learning_rate": 0.00027622142536978936, + "loss": 0.35, + "step": 25957 + }, + { + "epoch": 1.4503170656758946, + "grad_norm": 3.448505401611328, + "learning_rate": 0.00027619341102644556, + "loss": 0.413, + "step": 25958 + }, + { + "epoch": 1.4503729362795765, + "grad_norm": 0.3538975715637207, + "learning_rate": 0.00027616539668310177, + "loss": 0.4237, + "step": 25959 + }, + { + "epoch": 1.4504288068832585, + "grad_norm": 0.54448002576828, + "learning_rate": 0.0002761373823397579, + "loss": 0.5209, + "step": 25960 + }, + { + "epoch": 1.4504846774869402, + "grad_norm": 0.3890562355518341, + "learning_rate": 0.0002761093679964142, + "loss": 0.4689, + "step": 25961 + }, + { + "epoch": 1.4505405480906222, + "grad_norm": 0.6504620909690857, + "learning_rate": 0.00027608135365307033, + "loss": 0.5034, + "step": 25962 + }, + { + "epoch": 1.4505964186943041, + "grad_norm": 0.8403568267822266, + "learning_rate": 0.0002760533393097266, + "loss": 0.4216, + "step": 25963 + }, + { + "epoch": 1.4506522892979858, + "grad_norm": 0.43074315786361694, + "learning_rate": 0.0002760253249663828, + "loss": 0.4651, + "step": 25964 + }, + { + "epoch": 1.4507081599016678, + "grad_norm": 0.45079055428504944, + "learning_rate": 0.000275997310623039, + "loss": 0.5029, + "step": 25965 + }, + { + "epoch": 1.4507640305053497, + "grad_norm": 0.35876035690307617, + "learning_rate": 0.0002759692962796952, + "loss": 0.4133, + "step": 25966 + }, + { + "epoch": 1.4508199011090315, + "grad_norm": 0.6193006038665771, + "learning_rate": 0.0002759412819363514, + "loss": 0.4018, + "step": 25967 + }, + { + "epoch": 1.4508757717127134, + "grad_norm": 0.30247780680656433, + "learning_rate": 0.0002759132675930076, + "loss": 0.3746, + "step": 25968 + }, + { + "epoch": 1.4509316423163952, + "grad_norm": 0.4808914363384247, + "learning_rate": 0.0002758852532496638, + "loss": 0.375, + "step": 25969 + }, + { + "epoch": 1.450987512920077, + "grad_norm": 0.3433404564857483, + "learning_rate": 0.00027585723890632003, + "loss": 0.3546, + "step": 25970 + }, + { + "epoch": 1.451043383523759, + "grad_norm": 0.5104790329933167, + "learning_rate": 0.0002758292245629763, + "loss": 0.3678, + "step": 25971 + }, + { + "epoch": 1.4510992541274408, + "grad_norm": 0.5136494040489197, + "learning_rate": 0.00027580121021963244, + "loss": 0.5045, + "step": 25972 + }, + { + "epoch": 1.4511551247311227, + "grad_norm": 0.45848348736763, + "learning_rate": 0.0002757731958762887, + "loss": 0.3212, + "step": 25973 + }, + { + "epoch": 1.4512109953348045, + "grad_norm": 0.395846426486969, + "learning_rate": 0.00027574518153294486, + "loss": 0.3995, + "step": 25974 + }, + { + "epoch": 1.4512668659384864, + "grad_norm": 0.43474680185317993, + "learning_rate": 0.0002757171671896011, + "loss": 0.3817, + "step": 25975 + }, + { + "epoch": 1.4513227365421684, + "grad_norm": 0.4928224980831146, + "learning_rate": 0.00027568915284625727, + "loss": 0.3619, + "step": 25976 + }, + { + "epoch": 1.45137860714585, + "grad_norm": 1.0808464288711548, + "learning_rate": 0.0002756611385029135, + "loss": 0.5027, + "step": 25977 + }, + { + "epoch": 1.451434477749532, + "grad_norm": 0.6784989237785339, + "learning_rate": 0.0002756331241595697, + "loss": 0.3157, + "step": 25978 + }, + { + "epoch": 1.451490348353214, + "grad_norm": 0.842096209526062, + "learning_rate": 0.00027560510981622594, + "loss": 0.4635, + "step": 25979 + }, + { + "epoch": 1.4515462189568957, + "grad_norm": 0.48060551285743713, + "learning_rate": 0.00027557709547288214, + "loss": 0.3999, + "step": 25980 + }, + { + "epoch": 1.4516020895605777, + "grad_norm": 0.4227439761161804, + "learning_rate": 0.0002755490811295383, + "loss": 0.5501, + "step": 25981 + }, + { + "epoch": 1.4516579601642596, + "grad_norm": 2.616748809814453, + "learning_rate": 0.00027552106678619456, + "loss": 0.3881, + "step": 25982 + }, + { + "epoch": 1.4517138307679414, + "grad_norm": 0.36667802929878235, + "learning_rate": 0.0002754930524428507, + "loss": 0.3218, + "step": 25983 + }, + { + "epoch": 1.4517697013716233, + "grad_norm": 0.46715661883354187, + "learning_rate": 0.00027546503809950697, + "loss": 0.3352, + "step": 25984 + }, + { + "epoch": 1.4518255719753053, + "grad_norm": 0.8051064014434814, + "learning_rate": 0.0002754370237561631, + "loss": 0.4785, + "step": 25985 + }, + { + "epoch": 1.451881442578987, + "grad_norm": 0.7173269391059875, + "learning_rate": 0.0002754090094128194, + "loss": 0.3551, + "step": 25986 + }, + { + "epoch": 1.451937313182669, + "grad_norm": 0.4591195285320282, + "learning_rate": 0.0002753809950694756, + "loss": 0.5564, + "step": 25987 + }, + { + "epoch": 1.451993183786351, + "grad_norm": 4.114699363708496, + "learning_rate": 0.0002753529807261318, + "loss": 0.354, + "step": 25988 + }, + { + "epoch": 1.4520490543900326, + "grad_norm": 1.0218424797058105, + "learning_rate": 0.000275324966382788, + "loss": 0.4955, + "step": 25989 + }, + { + "epoch": 1.4521049249937146, + "grad_norm": 0.33964163064956665, + "learning_rate": 0.0002752969520394442, + "loss": 0.3517, + "step": 25990 + }, + { + "epoch": 1.4521607955973965, + "grad_norm": 0.4378121793270111, + "learning_rate": 0.0002752689376961004, + "loss": 0.3945, + "step": 25991 + }, + { + "epoch": 1.4522166662010783, + "grad_norm": 0.4164026379585266, + "learning_rate": 0.0002752409233527566, + "loss": 0.52, + "step": 25992 + }, + { + "epoch": 1.4522725368047602, + "grad_norm": 0.3483008146286011, + "learning_rate": 0.0002752129090094128, + "loss": 0.3303, + "step": 25993 + }, + { + "epoch": 1.4523284074084422, + "grad_norm": 0.832281231880188, + "learning_rate": 0.000275184894666069, + "loss": 0.3029, + "step": 25994 + }, + { + "epoch": 1.4523842780121239, + "grad_norm": 0.3139343857765198, + "learning_rate": 0.00027515688032272523, + "loss": 0.3892, + "step": 25995 + }, + { + "epoch": 1.4524401486158058, + "grad_norm": 0.566857099533081, + "learning_rate": 0.0002751288659793815, + "loss": 0.4814, + "step": 25996 + }, + { + "epoch": 1.4524960192194878, + "grad_norm": 0.3745422065258026, + "learning_rate": 0.00027510085163603764, + "loss": 0.3223, + "step": 25997 + }, + { + "epoch": 1.4525518898231695, + "grad_norm": 3.5283329486846924, + "learning_rate": 0.0002750728372926939, + "loss": 0.458, + "step": 25998 + }, + { + "epoch": 1.4526077604268515, + "grad_norm": 0.6429724097251892, + "learning_rate": 0.00027504482294935005, + "loss": 0.5042, + "step": 25999 + }, + { + "epoch": 1.4526636310305334, + "grad_norm": 0.5833585262298584, + "learning_rate": 0.0002750168086060063, + "loss": 0.3548, + "step": 26000 + }, + { + "epoch": 1.4526636310305334, + "eval_cer": 0.08557275812030157, + "eval_loss": 0.32069674134254456, + "eval_runtime": 56.224, + "eval_samples_per_second": 80.713, + "eval_steps_per_second": 5.051, + "eval_wer": 0.34166643389849444, + "step": 26000 + }, + { + "epoch": 1.4527195016342151, + "grad_norm": 0.5964205861091614, + "learning_rate": 0.00027498879426266247, + "loss": 0.3318, + "step": 26001 + }, + { + "epoch": 1.452775372237897, + "grad_norm": 0.7173854112625122, + "learning_rate": 0.00027496077991931867, + "loss": 0.4777, + "step": 26002 + }, + { + "epoch": 1.4528312428415788, + "grad_norm": 0.5307384729385376, + "learning_rate": 0.00027493276557597493, + "loss": 0.3695, + "step": 26003 + }, + { + "epoch": 1.4528871134452608, + "grad_norm": 0.5460808277130127, + "learning_rate": 0.0002749047512326311, + "loss": 0.364, + "step": 26004 + }, + { + "epoch": 1.4529429840489427, + "grad_norm": 0.42633017897605896, + "learning_rate": 0.00027487673688928734, + "loss": 0.3516, + "step": 26005 + }, + { + "epoch": 1.4529988546526245, + "grad_norm": 1.3672116994857788, + "learning_rate": 0.0002748487225459435, + "loss": 0.4639, + "step": 26006 + }, + { + "epoch": 1.4530547252563064, + "grad_norm": 0.4235997498035431, + "learning_rate": 0.00027482070820259976, + "loss": 0.3972, + "step": 26007 + }, + { + "epoch": 1.4531105958599881, + "grad_norm": 0.39259904623031616, + "learning_rate": 0.0002747926938592559, + "loss": 0.3897, + "step": 26008 + }, + { + "epoch": 1.45316646646367, + "grad_norm": 0.2905145585536957, + "learning_rate": 0.00027476467951591217, + "loss": 0.376, + "step": 26009 + }, + { + "epoch": 1.453222337067352, + "grad_norm": 0.8316758871078491, + "learning_rate": 0.0002747366651725683, + "loss": 0.3238, + "step": 26010 + }, + { + "epoch": 1.4532782076710338, + "grad_norm": 0.7312795519828796, + "learning_rate": 0.0002747086508292246, + "loss": 0.3557, + "step": 26011 + }, + { + "epoch": 1.4533340782747157, + "grad_norm": 0.4684583842754364, + "learning_rate": 0.0002746806364858808, + "loss": 0.4553, + "step": 26012 + }, + { + "epoch": 1.4533899488783977, + "grad_norm": 1.1078592538833618, + "learning_rate": 0.000274652622142537, + "loss": 0.4046, + "step": 26013 + }, + { + "epoch": 1.4534458194820794, + "grad_norm": 0.479493111371994, + "learning_rate": 0.0002746246077991932, + "loss": 0.4702, + "step": 26014 + }, + { + "epoch": 1.4535016900857614, + "grad_norm": 0.6654062271118164, + "learning_rate": 0.0002745965934558494, + "loss": 0.4215, + "step": 26015 + }, + { + "epoch": 1.4535575606894433, + "grad_norm": 0.40118199586868286, + "learning_rate": 0.0002745685791125056, + "loss": 0.431, + "step": 26016 + }, + { + "epoch": 1.453613431293125, + "grad_norm": 0.815926194190979, + "learning_rate": 0.0002745405647691618, + "loss": 0.4579, + "step": 26017 + }, + { + "epoch": 1.453669301896807, + "grad_norm": 0.3398641347885132, + "learning_rate": 0.000274512550425818, + "loss": 0.3312, + "step": 26018 + }, + { + "epoch": 1.453725172500489, + "grad_norm": 0.7360523343086243, + "learning_rate": 0.0002744845360824743, + "loss": 0.3759, + "step": 26019 + }, + { + "epoch": 1.4537810431041707, + "grad_norm": 2.5592591762542725, + "learning_rate": 0.00027445652173913043, + "loss": 0.3933, + "step": 26020 + }, + { + "epoch": 1.4538369137078526, + "grad_norm": 0.36946940422058105, + "learning_rate": 0.0002744285073957867, + "loss": 0.4025, + "step": 26021 + }, + { + "epoch": 1.4538927843115346, + "grad_norm": 0.3854958713054657, + "learning_rate": 0.00027440049305244284, + "loss": 0.3796, + "step": 26022 + }, + { + "epoch": 1.4539486549152163, + "grad_norm": 0.3383450210094452, + "learning_rate": 0.00027437247870909905, + "loss": 0.3334, + "step": 26023 + }, + { + "epoch": 1.4540045255188982, + "grad_norm": 0.6823750734329224, + "learning_rate": 0.00027434446436575525, + "loss": 0.3886, + "step": 26024 + }, + { + "epoch": 1.4540603961225802, + "grad_norm": 0.357427179813385, + "learning_rate": 0.00027431645002241146, + "loss": 0.4045, + "step": 26025 + }, + { + "epoch": 1.454116266726262, + "grad_norm": 0.33861806988716125, + "learning_rate": 0.0002742884356790677, + "loss": 0.3337, + "step": 26026 + }, + { + "epoch": 1.4541721373299439, + "grad_norm": 0.3521299362182617, + "learning_rate": 0.00027426042133572387, + "loss": 0.4152, + "step": 26027 + }, + { + "epoch": 1.4542280079336258, + "grad_norm": 0.4976790249347687, + "learning_rate": 0.00027423240699238013, + "loss": 0.402, + "step": 26028 + }, + { + "epoch": 1.4542838785373076, + "grad_norm": 0.36532384157180786, + "learning_rate": 0.0002742043926490363, + "loss": 0.3572, + "step": 26029 + }, + { + "epoch": 1.4543397491409895, + "grad_norm": 0.4457390606403351, + "learning_rate": 0.00027417637830569254, + "loss": 0.4582, + "step": 26030 + }, + { + "epoch": 1.4543956197446715, + "grad_norm": 0.30314984917640686, + "learning_rate": 0.0002741483639623487, + "loss": 0.3158, + "step": 26031 + }, + { + "epoch": 1.4544514903483532, + "grad_norm": 0.34685835242271423, + "learning_rate": 0.00027412034961900496, + "loss": 0.3725, + "step": 26032 + }, + { + "epoch": 1.4545073609520351, + "grad_norm": 0.4230630695819855, + "learning_rate": 0.0002740923352756611, + "loss": 0.3458, + "step": 26033 + }, + { + "epoch": 1.454563231555717, + "grad_norm": 0.36657071113586426, + "learning_rate": 0.00027406432093231737, + "loss": 0.37, + "step": 26034 + }, + { + "epoch": 1.4546191021593988, + "grad_norm": 0.3911707103252411, + "learning_rate": 0.00027403630658897357, + "loss": 0.4939, + "step": 26035 + }, + { + "epoch": 1.4546749727630808, + "grad_norm": 0.4189978837966919, + "learning_rate": 0.0002740082922456298, + "loss": 0.5132, + "step": 26036 + }, + { + "epoch": 1.4547308433667625, + "grad_norm": 0.41575050354003906, + "learning_rate": 0.000273980277902286, + "loss": 0.3277, + "step": 26037 + }, + { + "epoch": 1.4547867139704445, + "grad_norm": 0.3735716938972473, + "learning_rate": 0.0002739522635589422, + "loss": 0.3587, + "step": 26038 + }, + { + "epoch": 1.4548425845741262, + "grad_norm": 0.3520765006542206, + "learning_rate": 0.0002739242492155984, + "loss": 0.4394, + "step": 26039 + }, + { + "epoch": 1.4548984551778081, + "grad_norm": 1.9142006635665894, + "learning_rate": 0.0002738962348722546, + "loss": 0.4208, + "step": 26040 + }, + { + "epoch": 1.45495432578149, + "grad_norm": 4.655470848083496, + "learning_rate": 0.0002738682205289108, + "loss": 0.377, + "step": 26041 + }, + { + "epoch": 1.4550101963851718, + "grad_norm": 0.43441078066825867, + "learning_rate": 0.00027384020618556707, + "loss": 0.3767, + "step": 26042 + }, + { + "epoch": 1.4550660669888538, + "grad_norm": 0.3187900185585022, + "learning_rate": 0.0002738121918422232, + "loss": 0.3813, + "step": 26043 + }, + { + "epoch": 1.4551219375925357, + "grad_norm": 0.45603224635124207, + "learning_rate": 0.0002737841774988794, + "loss": 0.4935, + "step": 26044 + }, + { + "epoch": 1.4551778081962174, + "grad_norm": 0.3255827724933624, + "learning_rate": 0.00027375616315553563, + "loss": 0.3562, + "step": 26045 + }, + { + "epoch": 1.4552336787998994, + "grad_norm": 0.8669347167015076, + "learning_rate": 0.00027372814881219184, + "loss": 0.4843, + "step": 26046 + }, + { + "epoch": 1.4552895494035814, + "grad_norm": 0.5832328200340271, + "learning_rate": 0.00027370013446884804, + "loss": 0.3911, + "step": 26047 + }, + { + "epoch": 1.455345420007263, + "grad_norm": 0.5411698222160339, + "learning_rate": 0.00027367212012550425, + "loss": 0.3422, + "step": 26048 + }, + { + "epoch": 1.455401290610945, + "grad_norm": 0.43537142872810364, + "learning_rate": 0.00027364410578216045, + "loss": 0.4943, + "step": 26049 + }, + { + "epoch": 1.455457161214627, + "grad_norm": 0.4833305776119232, + "learning_rate": 0.00027361609143881666, + "loss": 0.5257, + "step": 26050 + }, + { + "epoch": 1.4555130318183087, + "grad_norm": 0.720405101776123, + "learning_rate": 0.0002735880770954729, + "loss": 0.4782, + "step": 26051 + }, + { + "epoch": 1.4555689024219907, + "grad_norm": 0.4192240834236145, + "learning_rate": 0.00027356006275212907, + "loss": 0.4327, + "step": 26052 + }, + { + "epoch": 1.4556247730256726, + "grad_norm": 0.5121405124664307, + "learning_rate": 0.00027353204840878533, + "loss": 0.5858, + "step": 26053 + }, + { + "epoch": 1.4556806436293543, + "grad_norm": 0.5927819609642029, + "learning_rate": 0.0002735040340654415, + "loss": 0.4325, + "step": 26054 + }, + { + "epoch": 1.4557365142330363, + "grad_norm": 0.8511234521865845, + "learning_rate": 0.00027347601972209774, + "loss": 0.55, + "step": 26055 + }, + { + "epoch": 1.4557923848367182, + "grad_norm": 1.0126527547836304, + "learning_rate": 0.0002734480053787539, + "loss": 0.4429, + "step": 26056 + }, + { + "epoch": 1.4558482554404, + "grad_norm": 0.5598774552345276, + "learning_rate": 0.00027341999103541015, + "loss": 0.5428, + "step": 26057 + }, + { + "epoch": 1.455904126044082, + "grad_norm": 1.1602352857589722, + "learning_rate": 0.00027339197669206636, + "loss": 0.4313, + "step": 26058 + }, + { + "epoch": 1.4559599966477639, + "grad_norm": 0.30792680382728577, + "learning_rate": 0.00027336396234872257, + "loss": 0.3432, + "step": 26059 + }, + { + "epoch": 1.4560158672514456, + "grad_norm": 0.48012828826904297, + "learning_rate": 0.00027333594800537877, + "loss": 0.4216, + "step": 26060 + }, + { + "epoch": 1.4560717378551276, + "grad_norm": 0.3667505383491516, + "learning_rate": 0.000273307933662035, + "loss": 0.4324, + "step": 26061 + }, + { + "epoch": 1.4561276084588095, + "grad_norm": 0.7316762804985046, + "learning_rate": 0.0002732799193186912, + "loss": 0.4718, + "step": 26062 + }, + { + "epoch": 1.4561834790624912, + "grad_norm": 0.4104806184768677, + "learning_rate": 0.0002732519049753474, + "loss": 0.3892, + "step": 26063 + }, + { + "epoch": 1.4562393496661732, + "grad_norm": 0.5451052188873291, + "learning_rate": 0.0002732238906320036, + "loss": 0.5467, + "step": 26064 + }, + { + "epoch": 1.4562952202698551, + "grad_norm": 0.35609036684036255, + "learning_rate": 0.00027319587628865975, + "loss": 0.4276, + "step": 26065 + }, + { + "epoch": 1.4563510908735369, + "grad_norm": 1.1090257167816162, + "learning_rate": 0.000273167861945316, + "loss": 0.4597, + "step": 26066 + }, + { + "epoch": 1.4564069614772188, + "grad_norm": 0.31876876950263977, + "learning_rate": 0.0002731398476019722, + "loss": 0.392, + "step": 26067 + }, + { + "epoch": 1.4564628320809008, + "grad_norm": 0.43427568674087524, + "learning_rate": 0.0002731118332586284, + "loss": 0.4929, + "step": 26068 + }, + { + "epoch": 1.4565187026845825, + "grad_norm": 5.206859588623047, + "learning_rate": 0.0002730838189152846, + "loss": 0.4119, + "step": 26069 + }, + { + "epoch": 1.4565745732882645, + "grad_norm": 0.3151867985725403, + "learning_rate": 0.00027305580457194083, + "loss": 0.3161, + "step": 26070 + }, + { + "epoch": 1.4566304438919462, + "grad_norm": 0.4428328573703766, + "learning_rate": 0.00027302779022859704, + "loss": 0.4065, + "step": 26071 + }, + { + "epoch": 1.4566863144956281, + "grad_norm": 0.6960854530334473, + "learning_rate": 0.00027299977588525324, + "loss": 0.4009, + "step": 26072 + }, + { + "epoch": 1.4567421850993099, + "grad_norm": 0.8380762934684753, + "learning_rate": 0.00027297176154190945, + "loss": 0.2911, + "step": 26073 + }, + { + "epoch": 1.4567980557029918, + "grad_norm": 0.44007742404937744, + "learning_rate": 0.0002729437471985657, + "loss": 0.3589, + "step": 26074 + }, + { + "epoch": 1.4568539263066738, + "grad_norm": 0.4428408443927765, + "learning_rate": 0.00027291573285522186, + "loss": 0.4929, + "step": 26075 + }, + { + "epoch": 1.4569097969103555, + "grad_norm": 0.5709370374679565, + "learning_rate": 0.0002728877185118781, + "loss": 0.4581, + "step": 26076 + }, + { + "epoch": 1.4569656675140374, + "grad_norm": 0.44359681010246277, + "learning_rate": 0.00027285970416853427, + "loss": 0.5578, + "step": 26077 + }, + { + "epoch": 1.4570215381177194, + "grad_norm": 0.4141961634159088, + "learning_rate": 0.00027283168982519053, + "loss": 0.4666, + "step": 26078 + }, + { + "epoch": 1.4570774087214011, + "grad_norm": 0.4027453064918518, + "learning_rate": 0.0002728036754818467, + "loss": 0.4366, + "step": 26079 + }, + { + "epoch": 1.457133279325083, + "grad_norm": 0.40738916397094727, + "learning_rate": 0.00027277566113850294, + "loss": 0.3856, + "step": 26080 + }, + { + "epoch": 1.457189149928765, + "grad_norm": 0.48103925585746765, + "learning_rate": 0.0002727476467951591, + "loss": 0.4172, + "step": 26081 + }, + { + "epoch": 1.4572450205324468, + "grad_norm": 0.47552913427352905, + "learning_rate": 0.00027271963245181535, + "loss": 0.5218, + "step": 26082 + }, + { + "epoch": 1.4573008911361287, + "grad_norm": 0.5418100953102112, + "learning_rate": 0.00027269161810847156, + "loss": 0.3295, + "step": 26083 + }, + { + "epoch": 1.4573567617398107, + "grad_norm": 1.1390886306762695, + "learning_rate": 0.00027266360376512777, + "loss": 0.3931, + "step": 26084 + }, + { + "epoch": 1.4574126323434924, + "grad_norm": 0.43464040756225586, + "learning_rate": 0.00027263558942178397, + "loss": 0.4021, + "step": 26085 + }, + { + "epoch": 1.4574685029471743, + "grad_norm": 4.94527530670166, + "learning_rate": 0.0002726075750784401, + "loss": 0.482, + "step": 26086 + }, + { + "epoch": 1.4575243735508563, + "grad_norm": 0.29462525248527527, + "learning_rate": 0.0002725795607350964, + "loss": 0.28, + "step": 26087 + }, + { + "epoch": 1.457580244154538, + "grad_norm": 0.4744241535663605, + "learning_rate": 0.00027255154639175253, + "loss": 0.466, + "step": 26088 + }, + { + "epoch": 1.45763611475822, + "grad_norm": 0.4073913097381592, + "learning_rate": 0.0002725235320484088, + "loss": 0.2759, + "step": 26089 + }, + { + "epoch": 1.457691985361902, + "grad_norm": 0.49415886402130127, + "learning_rate": 0.000272495517705065, + "loss": 0.4335, + "step": 26090 + }, + { + "epoch": 1.4577478559655836, + "grad_norm": 0.520814061164856, + "learning_rate": 0.0002724675033617212, + "loss": 0.4514, + "step": 26091 + }, + { + "epoch": 1.4578037265692656, + "grad_norm": 0.6070795655250549, + "learning_rate": 0.0002724394890183774, + "loss": 0.481, + "step": 26092 + }, + { + "epoch": 1.4578595971729476, + "grad_norm": 0.860490083694458, + "learning_rate": 0.0002724114746750336, + "loss": 0.4068, + "step": 26093 + }, + { + "epoch": 1.4579154677766293, + "grad_norm": 0.46506431698799133, + "learning_rate": 0.0002723834603316898, + "loss": 0.506, + "step": 26094 + }, + { + "epoch": 1.4579713383803112, + "grad_norm": 0.372999370098114, + "learning_rate": 0.00027235544598834603, + "loss": 0.3543, + "step": 26095 + }, + { + "epoch": 1.4580272089839932, + "grad_norm": 0.6070287227630615, + "learning_rate": 0.00027232743164500224, + "loss": 0.3719, + "step": 26096 + }, + { + "epoch": 1.458083079587675, + "grad_norm": 0.3662227690219879, + "learning_rate": 0.0002722994173016585, + "loss": 0.4143, + "step": 26097 + }, + { + "epoch": 1.4581389501913569, + "grad_norm": 0.4598543643951416, + "learning_rate": 0.00027227140295831465, + "loss": 0.3858, + "step": 26098 + }, + { + "epoch": 1.4581948207950388, + "grad_norm": 0.37283143401145935, + "learning_rate": 0.0002722433886149709, + "loss": 0.416, + "step": 26099 + }, + { + "epoch": 1.4582506913987205, + "grad_norm": 0.48974183201789856, + "learning_rate": 0.00027221537427162706, + "loss": 0.4775, + "step": 26100 + }, + { + "epoch": 1.4583065620024025, + "grad_norm": 0.35626134276390076, + "learning_rate": 0.0002721873599282833, + "loss": 0.5164, + "step": 26101 + }, + { + "epoch": 1.4583624326060844, + "grad_norm": 0.4046962857246399, + "learning_rate": 0.00027215934558493947, + "loss": 0.4901, + "step": 26102 + }, + { + "epoch": 1.4584183032097662, + "grad_norm": 0.4010716378688812, + "learning_rate": 0.00027213133124159573, + "loss": 0.4433, + "step": 26103 + }, + { + "epoch": 1.4584741738134481, + "grad_norm": 0.37631386518478394, + "learning_rate": 0.0002721033168982519, + "loss": 0.4212, + "step": 26104 + }, + { + "epoch": 1.4585300444171299, + "grad_norm": 0.39065420627593994, + "learning_rate": 0.00027207530255490814, + "loss": 0.3771, + "step": 26105 + }, + { + "epoch": 1.4585859150208118, + "grad_norm": 0.3869036138057709, + "learning_rate": 0.00027204728821156435, + "loss": 0.5105, + "step": 26106 + }, + { + "epoch": 1.4586417856244935, + "grad_norm": 0.3347383737564087, + "learning_rate": 0.0002720192738682205, + "loss": 0.4484, + "step": 26107 + }, + { + "epoch": 1.4586976562281755, + "grad_norm": 0.42592155933380127, + "learning_rate": 0.00027199125952487676, + "loss": 0.4192, + "step": 26108 + }, + { + "epoch": 1.4587535268318574, + "grad_norm": 0.3131970763206482, + "learning_rate": 0.0002719632451815329, + "loss": 0.3452, + "step": 26109 + }, + { + "epoch": 1.4588093974355392, + "grad_norm": 0.5231389403343201, + "learning_rate": 0.00027193523083818917, + "loss": 0.3784, + "step": 26110 + }, + { + "epoch": 1.4588652680392211, + "grad_norm": 0.5454223155975342, + "learning_rate": 0.0002719072164948453, + "loss": 0.5029, + "step": 26111 + }, + { + "epoch": 1.458921138642903, + "grad_norm": 0.6813825964927673, + "learning_rate": 0.0002718792021515016, + "loss": 0.4269, + "step": 26112 + }, + { + "epoch": 1.4589770092465848, + "grad_norm": 0.46261337399482727, + "learning_rate": 0.0002718511878081578, + "loss": 0.3845, + "step": 26113 + }, + { + "epoch": 1.4590328798502668, + "grad_norm": 0.41386327147483826, + "learning_rate": 0.000271823173464814, + "loss": 0.3955, + "step": 26114 + }, + { + "epoch": 1.4590887504539487, + "grad_norm": 3.624962091445923, + "learning_rate": 0.0002717951591214702, + "loss": 0.4666, + "step": 26115 + }, + { + "epoch": 1.4591446210576304, + "grad_norm": 0.6284210681915283, + "learning_rate": 0.0002717671447781264, + "loss": 0.4857, + "step": 26116 + }, + { + "epoch": 1.4592004916613124, + "grad_norm": 0.44108763337135315, + "learning_rate": 0.0002717391304347826, + "loss": 0.3462, + "step": 26117 + }, + { + "epoch": 1.4592563622649943, + "grad_norm": 0.5056160092353821, + "learning_rate": 0.0002717111160914388, + "loss": 0.4952, + "step": 26118 + }, + { + "epoch": 1.459312232868676, + "grad_norm": 0.629004716873169, + "learning_rate": 0.000271683101748095, + "loss": 0.397, + "step": 26119 + }, + { + "epoch": 1.459368103472358, + "grad_norm": 0.39985111355781555, + "learning_rate": 0.00027165508740475123, + "loss": 0.3888, + "step": 26120 + }, + { + "epoch": 1.45942397407604, + "grad_norm": 0.5125189423561096, + "learning_rate": 0.00027162707306140744, + "loss": 0.449, + "step": 26121 + }, + { + "epoch": 1.4594798446797217, + "grad_norm": 0.343862920999527, + "learning_rate": 0.0002715990587180637, + "loss": 0.4, + "step": 26122 + }, + { + "epoch": 1.4595357152834036, + "grad_norm": 0.34052592515945435, + "learning_rate": 0.00027157104437471985, + "loss": 0.4335, + "step": 26123 + }, + { + "epoch": 1.4595915858870856, + "grad_norm": 1.4199320077896118, + "learning_rate": 0.0002715430300313761, + "loss": 0.3792, + "step": 26124 + }, + { + "epoch": 1.4596474564907673, + "grad_norm": 0.647232711315155, + "learning_rate": 0.00027151501568803226, + "loss": 0.3454, + "step": 26125 + }, + { + "epoch": 1.4597033270944493, + "grad_norm": 0.37799206376075745, + "learning_rate": 0.0002714870013446885, + "loss": 0.5199, + "step": 26126 + }, + { + "epoch": 1.4597591976981312, + "grad_norm": 0.7069239616394043, + "learning_rate": 0.00027145898700134467, + "loss": 0.5947, + "step": 26127 + }, + { + "epoch": 1.459815068301813, + "grad_norm": 1.0344213247299194, + "learning_rate": 0.00027143097265800093, + "loss": 0.7018, + "step": 26128 + }, + { + "epoch": 1.459870938905495, + "grad_norm": 0.7129703164100647, + "learning_rate": 0.00027140295831465714, + "loss": 0.4906, + "step": 26129 + }, + { + "epoch": 1.4599268095091769, + "grad_norm": 1.7607262134552002, + "learning_rate": 0.0002713749439713133, + "loss": 0.7075, + "step": 26130 + }, + { + "epoch": 1.4599826801128586, + "grad_norm": 0.5819432735443115, + "learning_rate": 0.00027134692962796955, + "loss": 0.4008, + "step": 26131 + }, + { + "epoch": 1.4600385507165405, + "grad_norm": 0.8393914103507996, + "learning_rate": 0.0002713189152846257, + "loss": 0.6285, + "step": 26132 + }, + { + "epoch": 1.4600944213202225, + "grad_norm": 3.8774666786193848, + "learning_rate": 0.00027129090094128196, + "loss": 0.4434, + "step": 26133 + }, + { + "epoch": 1.4601502919239042, + "grad_norm": 0.5302733778953552, + "learning_rate": 0.0002712628865979381, + "loss": 0.3903, + "step": 26134 + }, + { + "epoch": 1.4602061625275862, + "grad_norm": 1.703210711479187, + "learning_rate": 0.00027123487225459437, + "loss": 0.3873, + "step": 26135 + }, + { + "epoch": 1.460262033131268, + "grad_norm": 0.5338876247406006, + "learning_rate": 0.0002712068579112505, + "loss": 0.4214, + "step": 26136 + }, + { + "epoch": 1.4603179037349499, + "grad_norm": 0.5218629837036133, + "learning_rate": 0.0002711788435679068, + "loss": 0.5126, + "step": 26137 + }, + { + "epoch": 1.4603737743386318, + "grad_norm": 0.3802683651447296, + "learning_rate": 0.000271150829224563, + "loss": 0.4492, + "step": 26138 + }, + { + "epoch": 1.4604296449423135, + "grad_norm": 0.3950273096561432, + "learning_rate": 0.0002711228148812192, + "loss": 0.3986, + "step": 26139 + }, + { + "epoch": 1.4604855155459955, + "grad_norm": 0.44753026962280273, + "learning_rate": 0.0002710948005378754, + "loss": 0.3637, + "step": 26140 + }, + { + "epoch": 1.4605413861496772, + "grad_norm": 0.4119139015674591, + "learning_rate": 0.0002710667861945316, + "loss": 0.3902, + "step": 26141 + }, + { + "epoch": 1.4605972567533592, + "grad_norm": 0.34608691930770874, + "learning_rate": 0.0002710387718511878, + "loss": 0.4224, + "step": 26142 + }, + { + "epoch": 1.4606531273570411, + "grad_norm": 0.6005557179450989, + "learning_rate": 0.000271010757507844, + "loss": 0.371, + "step": 26143 + }, + { + "epoch": 1.4607089979607228, + "grad_norm": 0.7789295315742493, + "learning_rate": 0.0002709827431645002, + "loss": 0.4688, + "step": 26144 + }, + { + "epoch": 1.4607648685644048, + "grad_norm": 0.3623631000518799, + "learning_rate": 0.0002709547288211565, + "loss": 0.3894, + "step": 26145 + }, + { + "epoch": 1.4608207391680867, + "grad_norm": 9.482156753540039, + "learning_rate": 0.00027092671447781263, + "loss": 0.4649, + "step": 26146 + }, + { + "epoch": 1.4608766097717685, + "grad_norm": 1.9276000261306763, + "learning_rate": 0.0002708987001344689, + "loss": 0.3659, + "step": 26147 + }, + { + "epoch": 1.4609324803754504, + "grad_norm": 0.4596257507801056, + "learning_rate": 0.00027087068579112505, + "loss": 0.3984, + "step": 26148 + }, + { + "epoch": 1.4609883509791324, + "grad_norm": 0.33717119693756104, + "learning_rate": 0.0002708426714477813, + "loss": 0.4205, + "step": 26149 + }, + { + "epoch": 1.461044221582814, + "grad_norm": 0.632051408290863, + "learning_rate": 0.00027081465710443746, + "loss": 0.426, + "step": 26150 + }, + { + "epoch": 1.461100092186496, + "grad_norm": 1.4473837614059448, + "learning_rate": 0.00027078664276109366, + "loss": 0.397, + "step": 26151 + }, + { + "epoch": 1.461155962790178, + "grad_norm": 0.3308489918708801, + "learning_rate": 0.0002707586284177499, + "loss": 0.3074, + "step": 26152 + }, + { + "epoch": 1.4612118333938597, + "grad_norm": 0.4332271218299866, + "learning_rate": 0.0002707306140744061, + "loss": 0.3866, + "step": 26153 + }, + { + "epoch": 1.4612677039975417, + "grad_norm": 0.8309876322746277, + "learning_rate": 0.00027070259973106234, + "loss": 0.4128, + "step": 26154 + }, + { + "epoch": 1.4613235746012236, + "grad_norm": 0.5213522911071777, + "learning_rate": 0.0002706745853877185, + "loss": 0.3783, + "step": 26155 + }, + { + "epoch": 1.4613794452049054, + "grad_norm": 0.47853967547416687, + "learning_rate": 0.00027064657104437475, + "loss": 0.4292, + "step": 26156 + }, + { + "epoch": 1.4614353158085873, + "grad_norm": 0.43343299627304077, + "learning_rate": 0.0002706185567010309, + "loss": 0.4602, + "step": 26157 + }, + { + "epoch": 1.4614911864122693, + "grad_norm": 0.5082798004150391, + "learning_rate": 0.00027059054235768716, + "loss": 0.6337, + "step": 26158 + }, + { + "epoch": 1.461547057015951, + "grad_norm": 0.5692682266235352, + "learning_rate": 0.0002705625280143433, + "loss": 0.4173, + "step": 26159 + }, + { + "epoch": 1.461602927619633, + "grad_norm": 0.6406525373458862, + "learning_rate": 0.00027053451367099957, + "loss": 0.4515, + "step": 26160 + }, + { + "epoch": 1.461658798223315, + "grad_norm": 0.4172689616680145, + "learning_rate": 0.0002705064993276558, + "loss": 0.4927, + "step": 26161 + }, + { + "epoch": 1.4617146688269966, + "grad_norm": 0.6994619369506836, + "learning_rate": 0.000270478484984312, + "loss": 0.4027, + "step": 26162 + }, + { + "epoch": 1.4617705394306786, + "grad_norm": 0.7628791928291321, + "learning_rate": 0.0002704504706409682, + "loss": 0.3967, + "step": 26163 + }, + { + "epoch": 1.4618264100343605, + "grad_norm": 0.49111366271972656, + "learning_rate": 0.0002704224562976244, + "loss": 0.4894, + "step": 26164 + }, + { + "epoch": 1.4618822806380423, + "grad_norm": 0.45332029461860657, + "learning_rate": 0.0002703944419542806, + "loss": 0.4028, + "step": 26165 + }, + { + "epoch": 1.4619381512417242, + "grad_norm": 0.3481979966163635, + "learning_rate": 0.0002703664276109368, + "loss": 0.3241, + "step": 26166 + }, + { + "epoch": 1.4619940218454062, + "grad_norm": 0.36659321188926697, + "learning_rate": 0.000270338413267593, + "loss": 0.4813, + "step": 26167 + }, + { + "epoch": 1.462049892449088, + "grad_norm": 0.43677830696105957, + "learning_rate": 0.00027031039892424927, + "loss": 0.4853, + "step": 26168 + }, + { + "epoch": 1.4621057630527698, + "grad_norm": 0.5276451110839844, + "learning_rate": 0.0002702823845809054, + "loss": 0.5669, + "step": 26169 + }, + { + "epoch": 1.4621616336564516, + "grad_norm": 0.39547306299209595, + "learning_rate": 0.0002702543702375617, + "loss": 0.4882, + "step": 26170 + }, + { + "epoch": 1.4622175042601335, + "grad_norm": 0.7133668065071106, + "learning_rate": 0.00027022635589421783, + "loss": 0.4285, + "step": 26171 + }, + { + "epoch": 1.4622733748638155, + "grad_norm": 0.40569061040878296, + "learning_rate": 0.00027019834155087404, + "loss": 0.4246, + "step": 26172 + }, + { + "epoch": 1.4623292454674972, + "grad_norm": 0.627311110496521, + "learning_rate": 0.00027017032720753025, + "loss": 0.4839, + "step": 26173 + }, + { + "epoch": 1.4623851160711792, + "grad_norm": 0.5133434534072876, + "learning_rate": 0.00027014231286418645, + "loss": 0.392, + "step": 26174 + }, + { + "epoch": 1.4624409866748609, + "grad_norm": 0.6777279376983643, + "learning_rate": 0.00027011429852084266, + "loss": 0.5144, + "step": 26175 + }, + { + "epoch": 1.4624968572785428, + "grad_norm": 0.3508586585521698, + "learning_rate": 0.00027008628417749886, + "loss": 0.32, + "step": 26176 + }, + { + "epoch": 1.4625527278822248, + "grad_norm": 0.3341846764087677, + "learning_rate": 0.0002700582698341551, + "loss": 0.4027, + "step": 26177 + }, + { + "epoch": 1.4626085984859065, + "grad_norm": 0.3462032079696655, + "learning_rate": 0.0002700302554908113, + "loss": 0.2915, + "step": 26178 + }, + { + "epoch": 1.4626644690895885, + "grad_norm": 0.3724822402000427, + "learning_rate": 0.00027000224114746754, + "loss": 0.4825, + "step": 26179 + }, + { + "epoch": 1.4627203396932704, + "grad_norm": 0.39909127354621887, + "learning_rate": 0.0002699742268041237, + "loss": 0.4172, + "step": 26180 + }, + { + "epoch": 1.4627762102969522, + "grad_norm": 0.34065189957618713, + "learning_rate": 0.00026994621246077995, + "loss": 0.4187, + "step": 26181 + }, + { + "epoch": 1.462832080900634, + "grad_norm": 1.1036605834960938, + "learning_rate": 0.0002699181981174361, + "loss": 0.479, + "step": 26182 + }, + { + "epoch": 1.462887951504316, + "grad_norm": 0.6284537315368652, + "learning_rate": 0.00026989018377409236, + "loss": 0.4321, + "step": 26183 + }, + { + "epoch": 1.4629438221079978, + "grad_norm": 0.4042433500289917, + "learning_rate": 0.00026986216943074856, + "loss": 0.4567, + "step": 26184 + }, + { + "epoch": 1.4629996927116797, + "grad_norm": 0.39882951974868774, + "learning_rate": 0.00026983415508740477, + "loss": 0.3886, + "step": 26185 + }, + { + "epoch": 1.4630555633153617, + "grad_norm": 0.3500574827194214, + "learning_rate": 0.000269806140744061, + "loss": 0.2687, + "step": 26186 + }, + { + "epoch": 1.4631114339190434, + "grad_norm": 1.948663353919983, + "learning_rate": 0.0002697781264007172, + "loss": 0.454, + "step": 26187 + }, + { + "epoch": 1.4631673045227254, + "grad_norm": 0.4872284531593323, + "learning_rate": 0.0002697501120573734, + "loss": 0.4826, + "step": 26188 + }, + { + "epoch": 1.4632231751264073, + "grad_norm": 0.3434900641441345, + "learning_rate": 0.0002697220977140296, + "loss": 0.3283, + "step": 26189 + }, + { + "epoch": 1.463279045730089, + "grad_norm": 0.4575725793838501, + "learning_rate": 0.0002696940833706858, + "loss": 0.4167, + "step": 26190 + }, + { + "epoch": 1.463334916333771, + "grad_norm": 2.029345750808716, + "learning_rate": 0.000269666069027342, + "loss": 0.4605, + "step": 26191 + }, + { + "epoch": 1.463390786937453, + "grad_norm": 0.3914542496204376, + "learning_rate": 0.0002696380546839982, + "loss": 0.455, + "step": 26192 + }, + { + "epoch": 1.4634466575411347, + "grad_norm": 1.995374321937561, + "learning_rate": 0.0002696100403406544, + "loss": 0.414, + "step": 26193 + }, + { + "epoch": 1.4635025281448166, + "grad_norm": 0.5663999915122986, + "learning_rate": 0.0002695820259973106, + "loss": 0.5413, + "step": 26194 + }, + { + "epoch": 1.4635583987484986, + "grad_norm": 0.4158914089202881, + "learning_rate": 0.00026955401165396683, + "loss": 0.3942, + "step": 26195 + }, + { + "epoch": 1.4636142693521803, + "grad_norm": 0.5109186768531799, + "learning_rate": 0.00026952599731062303, + "loss": 0.3542, + "step": 26196 + }, + { + "epoch": 1.4636701399558623, + "grad_norm": 1.1970486640930176, + "learning_rate": 0.00026949798296727924, + "loss": 0.4434, + "step": 26197 + }, + { + "epoch": 1.4637260105595442, + "grad_norm": 0.45222747325897217, + "learning_rate": 0.00026946996862393545, + "loss": 0.6525, + "step": 26198 + }, + { + "epoch": 1.463781881163226, + "grad_norm": 0.40911802649497986, + "learning_rate": 0.00026944195428059165, + "loss": 0.4128, + "step": 26199 + }, + { + "epoch": 1.463837751766908, + "grad_norm": 0.4332362711429596, + "learning_rate": 0.0002694139399372479, + "loss": 0.596, + "step": 26200 + }, + { + "epoch": 1.4638936223705898, + "grad_norm": 0.3107750415802002, + "learning_rate": 0.00026938592559390406, + "loss": 0.3498, + "step": 26201 + }, + { + "epoch": 1.4639494929742716, + "grad_norm": 0.6140823364257812, + "learning_rate": 0.0002693579112505603, + "loss": 0.4475, + "step": 26202 + }, + { + "epoch": 1.4640053635779535, + "grad_norm": 0.813179075717926, + "learning_rate": 0.0002693298969072165, + "loss": 0.4135, + "step": 26203 + }, + { + "epoch": 1.4640612341816353, + "grad_norm": 0.49855318665504456, + "learning_rate": 0.00026930188256387273, + "loss": 0.3955, + "step": 26204 + }, + { + "epoch": 1.4641171047853172, + "grad_norm": 1.6067999601364136, + "learning_rate": 0.0002692738682205289, + "loss": 0.4711, + "step": 26205 + }, + { + "epoch": 1.4641729753889992, + "grad_norm": 0.6728524565696716, + "learning_rate": 0.00026924585387718515, + "loss": 0.3761, + "step": 26206 + }, + { + "epoch": 1.4642288459926809, + "grad_norm": 0.38056209683418274, + "learning_rate": 0.0002692178395338413, + "loss": 0.4184, + "step": 26207 + }, + { + "epoch": 1.4642847165963628, + "grad_norm": 0.38635480403900146, + "learning_rate": 0.00026918982519049756, + "loss": 0.5458, + "step": 26208 + }, + { + "epoch": 1.4643405872000446, + "grad_norm": 1.2499618530273438, + "learning_rate": 0.00026916181084715376, + "loss": 0.5261, + "step": 26209 + }, + { + "epoch": 1.4643964578037265, + "grad_norm": 0.40952280163764954, + "learning_rate": 0.00026913379650380997, + "loss": 0.4399, + "step": 26210 + }, + { + "epoch": 1.4644523284074085, + "grad_norm": 0.37047022581100464, + "learning_rate": 0.0002691057821604662, + "loss": 0.4337, + "step": 26211 + }, + { + "epoch": 1.4645081990110902, + "grad_norm": 0.37660661339759827, + "learning_rate": 0.0002690777678171224, + "loss": 0.3686, + "step": 26212 + }, + { + "epoch": 1.4645640696147721, + "grad_norm": 0.822748064994812, + "learning_rate": 0.0002690497534737786, + "loss": 0.3794, + "step": 26213 + }, + { + "epoch": 1.464619940218454, + "grad_norm": 0.39422863721847534, + "learning_rate": 0.00026902173913043474, + "loss": 0.3819, + "step": 26214 + }, + { + "epoch": 1.4646758108221358, + "grad_norm": 0.6443850994110107, + "learning_rate": 0.000268993724787091, + "loss": 0.4957, + "step": 26215 + }, + { + "epoch": 1.4647316814258178, + "grad_norm": 0.390278160572052, + "learning_rate": 0.0002689657104437472, + "loss": 0.3504, + "step": 26216 + }, + { + "epoch": 1.4647875520294997, + "grad_norm": 0.34775280952453613, + "learning_rate": 0.0002689376961004034, + "loss": 0.3506, + "step": 26217 + }, + { + "epoch": 1.4648434226331815, + "grad_norm": 0.4257044494152069, + "learning_rate": 0.0002689096817570596, + "loss": 0.452, + "step": 26218 + }, + { + "epoch": 1.4648992932368634, + "grad_norm": 0.35025542974472046, + "learning_rate": 0.0002688816674137158, + "loss": 0.3845, + "step": 26219 + }, + { + "epoch": 1.4649551638405454, + "grad_norm": 0.3435302972793579, + "learning_rate": 0.00026885365307037203, + "loss": 0.3504, + "step": 26220 + }, + { + "epoch": 1.465011034444227, + "grad_norm": 0.5028291344642639, + "learning_rate": 0.00026882563872702823, + "loss": 0.4595, + "step": 26221 + }, + { + "epoch": 1.465066905047909, + "grad_norm": 0.5022467374801636, + "learning_rate": 0.00026879762438368444, + "loss": 0.4146, + "step": 26222 + }, + { + "epoch": 1.465122775651591, + "grad_norm": 0.4304402470588684, + "learning_rate": 0.0002687696100403407, + "loss": 0.4167, + "step": 26223 + }, + { + "epoch": 1.4651786462552727, + "grad_norm": 0.4991121292114258, + "learning_rate": 0.00026874159569699685, + "loss": 0.5295, + "step": 26224 + }, + { + "epoch": 1.4652345168589547, + "grad_norm": 0.9513658881187439, + "learning_rate": 0.0002687135813536531, + "loss": 0.3842, + "step": 26225 + }, + { + "epoch": 1.4652903874626366, + "grad_norm": 3.159173011779785, + "learning_rate": 0.00026868556701030926, + "loss": 0.3976, + "step": 26226 + }, + { + "epoch": 1.4653462580663184, + "grad_norm": 0.89540034532547, + "learning_rate": 0.0002686575526669655, + "loss": 0.3922, + "step": 26227 + }, + { + "epoch": 1.4654021286700003, + "grad_norm": 0.593191921710968, + "learning_rate": 0.0002686295383236217, + "loss": 0.3773, + "step": 26228 + }, + { + "epoch": 1.4654579992736823, + "grad_norm": 6.629914283752441, + "learning_rate": 0.00026860152398027793, + "loss": 0.4034, + "step": 26229 + }, + { + "epoch": 1.465513869877364, + "grad_norm": 0.3611897826194763, + "learning_rate": 0.0002685735096369341, + "loss": 0.4025, + "step": 26230 + }, + { + "epoch": 1.465569740481046, + "grad_norm": 0.42098695039749146, + "learning_rate": 0.00026854549529359035, + "loss": 0.4442, + "step": 26231 + }, + { + "epoch": 1.4656256110847279, + "grad_norm": 1.0501673221588135, + "learning_rate": 0.00026851748095024655, + "loss": 0.5337, + "step": 26232 + }, + { + "epoch": 1.4656814816884096, + "grad_norm": 0.40565305948257446, + "learning_rate": 0.00026848946660690276, + "loss": 0.3755, + "step": 26233 + }, + { + "epoch": 1.4657373522920916, + "grad_norm": 0.33408212661743164, + "learning_rate": 0.00026846145226355896, + "loss": 0.3902, + "step": 26234 + }, + { + "epoch": 1.4657932228957735, + "grad_norm": 0.6624436974525452, + "learning_rate": 0.0002684334379202151, + "loss": 0.355, + "step": 26235 + }, + { + "epoch": 1.4658490934994552, + "grad_norm": 0.3754214942455292, + "learning_rate": 0.0002684054235768714, + "loss": 0.3639, + "step": 26236 + }, + { + "epoch": 1.4659049641031372, + "grad_norm": 0.47834596037864685, + "learning_rate": 0.0002683774092335275, + "loss": 0.4036, + "step": 26237 + }, + { + "epoch": 1.465960834706819, + "grad_norm": 0.6750750541687012, + "learning_rate": 0.0002683493948901838, + "loss": 0.3993, + "step": 26238 + }, + { + "epoch": 1.4660167053105009, + "grad_norm": 0.5637394189834595, + "learning_rate": 0.00026832138054684, + "loss": 0.4521, + "step": 26239 + }, + { + "epoch": 1.4660725759141828, + "grad_norm": 0.27914562821388245, + "learning_rate": 0.0002682933662034962, + "loss": 0.3147, + "step": 26240 + }, + { + "epoch": 1.4661284465178646, + "grad_norm": 0.3772856891155243, + "learning_rate": 0.0002682653518601524, + "loss": 0.4642, + "step": 26241 + }, + { + "epoch": 1.4661843171215465, + "grad_norm": 2.2359044551849365, + "learning_rate": 0.0002682373375168086, + "loss": 0.391, + "step": 26242 + }, + { + "epoch": 1.4662401877252282, + "grad_norm": 0.4663480818271637, + "learning_rate": 0.0002682093231734648, + "loss": 0.418, + "step": 26243 + }, + { + "epoch": 1.4662960583289102, + "grad_norm": 0.41921266913414, + "learning_rate": 0.000268181308830121, + "loss": 0.3204, + "step": 26244 + }, + { + "epoch": 1.4663519289325921, + "grad_norm": 0.43896231055259705, + "learning_rate": 0.00026815329448677723, + "loss": 0.3747, + "step": 26245 + }, + { + "epoch": 1.4664077995362739, + "grad_norm": 0.39348313212394714, + "learning_rate": 0.00026812528014343343, + "loss": 0.4807, + "step": 26246 + }, + { + "epoch": 1.4664636701399558, + "grad_norm": 0.8048584461212158, + "learning_rate": 0.00026809726580008964, + "loss": 0.4941, + "step": 26247 + }, + { + "epoch": 1.4665195407436378, + "grad_norm": 0.7695173621177673, + "learning_rate": 0.0002680692514567459, + "loss": 0.5419, + "step": 26248 + }, + { + "epoch": 1.4665754113473195, + "grad_norm": 0.3883902430534363, + "learning_rate": 0.00026804123711340205, + "loss": 0.4183, + "step": 26249 + }, + { + "epoch": 1.4666312819510015, + "grad_norm": 4.409517765045166, + "learning_rate": 0.0002680132227700583, + "loss": 0.4931, + "step": 26250 + }, + { + "epoch": 1.4666871525546834, + "grad_norm": 0.8479477763175964, + "learning_rate": 0.00026798520842671446, + "loss": 0.4459, + "step": 26251 + }, + { + "epoch": 1.4667430231583651, + "grad_norm": 0.8310754299163818, + "learning_rate": 0.0002679571940833707, + "loss": 0.4144, + "step": 26252 + }, + { + "epoch": 1.466798893762047, + "grad_norm": 0.4339667856693268, + "learning_rate": 0.0002679291797400269, + "loss": 0.4576, + "step": 26253 + }, + { + "epoch": 1.466854764365729, + "grad_norm": 0.3329693078994751, + "learning_rate": 0.00026790116539668313, + "loss": 0.4869, + "step": 26254 + }, + { + "epoch": 1.4669106349694108, + "grad_norm": 0.44866228103637695, + "learning_rate": 0.00026787315105333934, + "loss": 0.4087, + "step": 26255 + }, + { + "epoch": 1.4669665055730927, + "grad_norm": 0.7456190586090088, + "learning_rate": 0.0002678451367099955, + "loss": 0.4398, + "step": 26256 + }, + { + "epoch": 1.4670223761767747, + "grad_norm": 1.6116846799850464, + "learning_rate": 0.00026781712236665175, + "loss": 0.4128, + "step": 26257 + }, + { + "epoch": 1.4670782467804564, + "grad_norm": 0.4456803798675537, + "learning_rate": 0.0002677891080233079, + "loss": 0.3568, + "step": 26258 + }, + { + "epoch": 1.4671341173841383, + "grad_norm": 0.5235687494277954, + "learning_rate": 0.00026776109367996416, + "loss": 0.4189, + "step": 26259 + }, + { + "epoch": 1.4671899879878203, + "grad_norm": 0.40907323360443115, + "learning_rate": 0.0002677330793366203, + "loss": 0.3115, + "step": 26260 + }, + { + "epoch": 1.467245858591502, + "grad_norm": 0.7234904170036316, + "learning_rate": 0.0002677050649932766, + "loss": 0.4128, + "step": 26261 + }, + { + "epoch": 1.467301729195184, + "grad_norm": 0.5074568390846252, + "learning_rate": 0.0002676770506499327, + "loss": 0.3856, + "step": 26262 + }, + { + "epoch": 1.467357599798866, + "grad_norm": 0.4060390293598175, + "learning_rate": 0.000267649036306589, + "loss": 0.4582, + "step": 26263 + }, + { + "epoch": 1.4674134704025477, + "grad_norm": 0.416525274515152, + "learning_rate": 0.0002676210219632452, + "loss": 0.4048, + "step": 26264 + }, + { + "epoch": 1.4674693410062296, + "grad_norm": 0.41986435651779175, + "learning_rate": 0.0002675930076199014, + "loss": 0.4109, + "step": 26265 + }, + { + "epoch": 1.4675252116099116, + "grad_norm": 0.6886189579963684, + "learning_rate": 0.0002675649932765576, + "loss": 0.3989, + "step": 26266 + }, + { + "epoch": 1.4675810822135933, + "grad_norm": 0.514624834060669, + "learning_rate": 0.0002675369789332138, + "loss": 0.3737, + "step": 26267 + }, + { + "epoch": 1.4676369528172752, + "grad_norm": 0.3409751355648041, + "learning_rate": 0.00026750896458987, + "loss": 0.3645, + "step": 26268 + }, + { + "epoch": 1.4676928234209572, + "grad_norm": 0.3791182339191437, + "learning_rate": 0.0002674809502465262, + "loss": 0.353, + "step": 26269 + }, + { + "epoch": 1.467748694024639, + "grad_norm": 0.34900152683258057, + "learning_rate": 0.0002674529359031824, + "loss": 0.3774, + "step": 26270 + }, + { + "epoch": 1.4678045646283209, + "grad_norm": 1.1005713939666748, + "learning_rate": 0.0002674249215598387, + "loss": 0.4084, + "step": 26271 + }, + { + "epoch": 1.4678604352320026, + "grad_norm": 9.38562297821045, + "learning_rate": 0.00026739690721649484, + "loss": 0.433, + "step": 26272 + }, + { + "epoch": 1.4679163058356846, + "grad_norm": 0.3635374903678894, + "learning_rate": 0.0002673688928731511, + "loss": 0.4288, + "step": 26273 + }, + { + "epoch": 1.4679721764393665, + "grad_norm": 0.33690980076789856, + "learning_rate": 0.00026734087852980725, + "loss": 0.3894, + "step": 26274 + }, + { + "epoch": 1.4680280470430482, + "grad_norm": 0.3227465748786926, + "learning_rate": 0.0002673128641864635, + "loss": 0.3692, + "step": 26275 + }, + { + "epoch": 1.4680839176467302, + "grad_norm": 0.904056966304779, + "learning_rate": 0.00026728484984311966, + "loss": 0.3475, + "step": 26276 + }, + { + "epoch": 1.468139788250412, + "grad_norm": 0.6361174583435059, + "learning_rate": 0.00026725683549977587, + "loss": 0.5279, + "step": 26277 + }, + { + "epoch": 1.4681956588540939, + "grad_norm": 0.36645564436912537, + "learning_rate": 0.0002672288211564321, + "loss": 0.3795, + "step": 26278 + }, + { + "epoch": 1.4682515294577758, + "grad_norm": 0.5163374543190002, + "learning_rate": 0.0002672008068130883, + "loss": 0.5082, + "step": 26279 + }, + { + "epoch": 1.4683074000614575, + "grad_norm": 0.44423195719718933, + "learning_rate": 0.00026717279246974454, + "loss": 0.4222, + "step": 26280 + }, + { + "epoch": 1.4683632706651395, + "grad_norm": 0.46150606870651245, + "learning_rate": 0.0002671447781264007, + "loss": 0.429, + "step": 26281 + }, + { + "epoch": 1.4684191412688214, + "grad_norm": 0.353159099817276, + "learning_rate": 0.00026711676378305695, + "loss": 0.3636, + "step": 26282 + }, + { + "epoch": 1.4684750118725032, + "grad_norm": 0.36518755555152893, + "learning_rate": 0.0002670887494397131, + "loss": 0.4162, + "step": 26283 + }, + { + "epoch": 1.4685308824761851, + "grad_norm": 0.9647290706634521, + "learning_rate": 0.00026706073509636936, + "loss": 0.4124, + "step": 26284 + }, + { + "epoch": 1.468586753079867, + "grad_norm": 0.38658425211906433, + "learning_rate": 0.0002670327207530255, + "loss": 0.4027, + "step": 26285 + }, + { + "epoch": 1.4686426236835488, + "grad_norm": 0.49353620409965515, + "learning_rate": 0.0002670047064096818, + "loss": 0.4166, + "step": 26286 + }, + { + "epoch": 1.4686984942872308, + "grad_norm": 1.527569055557251, + "learning_rate": 0.000266976692066338, + "loss": 0.407, + "step": 26287 + }, + { + "epoch": 1.4687543648909127, + "grad_norm": 0.37786296010017395, + "learning_rate": 0.0002669486777229942, + "loss": 0.3641, + "step": 26288 + }, + { + "epoch": 1.4688102354945944, + "grad_norm": 0.999712347984314, + "learning_rate": 0.0002669206633796504, + "loss": 0.3503, + "step": 26289 + }, + { + "epoch": 1.4688661060982764, + "grad_norm": 0.4827325642108917, + "learning_rate": 0.0002668926490363066, + "loss": 0.3421, + "step": 26290 + }, + { + "epoch": 1.4689219767019583, + "grad_norm": 0.7512935400009155, + "learning_rate": 0.0002668646346929628, + "loss": 0.4535, + "step": 26291 + }, + { + "epoch": 1.46897784730564, + "grad_norm": 0.33881962299346924, + "learning_rate": 0.000266836620349619, + "loss": 0.4133, + "step": 26292 + }, + { + "epoch": 1.469033717909322, + "grad_norm": 0.4055050015449524, + "learning_rate": 0.0002668086060062752, + "loss": 0.4451, + "step": 26293 + }, + { + "epoch": 1.469089588513004, + "grad_norm": 0.8526102304458618, + "learning_rate": 0.0002667805916629315, + "loss": 0.4546, + "step": 26294 + }, + { + "epoch": 1.4691454591166857, + "grad_norm": 10.561302185058594, + "learning_rate": 0.0002667525773195876, + "loss": 0.6302, + "step": 26295 + }, + { + "epoch": 1.4692013297203677, + "grad_norm": 0.34296858310699463, + "learning_rate": 0.0002667245629762439, + "loss": 0.3093, + "step": 26296 + }, + { + "epoch": 1.4692572003240496, + "grad_norm": 0.47810137271881104, + "learning_rate": 0.00026669654863290004, + "loss": 0.3806, + "step": 26297 + }, + { + "epoch": 1.4693130709277313, + "grad_norm": 0.45630761981010437, + "learning_rate": 0.0002666685342895563, + "loss": 0.4185, + "step": 26298 + }, + { + "epoch": 1.4693689415314133, + "grad_norm": 4.990628242492676, + "learning_rate": 0.00026664051994621245, + "loss": 0.461, + "step": 26299 + }, + { + "epoch": 1.4694248121350952, + "grad_norm": 0.34585973620414734, + "learning_rate": 0.00026661250560286866, + "loss": 0.3948, + "step": 26300 + }, + { + "epoch": 1.469480682738777, + "grad_norm": 0.7228346467018127, + "learning_rate": 0.00026658449125952486, + "loss": 0.376, + "step": 26301 + }, + { + "epoch": 1.469536553342459, + "grad_norm": 0.47204089164733887, + "learning_rate": 0.00026655647691618107, + "loss": 0.4165, + "step": 26302 + }, + { + "epoch": 1.4695924239461409, + "grad_norm": 2.0584778785705566, + "learning_rate": 0.00026652846257283733, + "loss": 0.3634, + "step": 26303 + }, + { + "epoch": 1.4696482945498226, + "grad_norm": 3.083925247192383, + "learning_rate": 0.0002665004482294935, + "loss": 0.3719, + "step": 26304 + }, + { + "epoch": 1.4697041651535045, + "grad_norm": 0.5280430912971497, + "learning_rate": 0.00026647243388614974, + "loss": 0.4092, + "step": 26305 + }, + { + "epoch": 1.4697600357571863, + "grad_norm": 0.3925227224826813, + "learning_rate": 0.0002664444195428059, + "loss": 0.44, + "step": 26306 + }, + { + "epoch": 1.4698159063608682, + "grad_norm": 1.409358263015747, + "learning_rate": 0.00026641640519946215, + "loss": 0.4368, + "step": 26307 + }, + { + "epoch": 1.4698717769645502, + "grad_norm": 0.5309264659881592, + "learning_rate": 0.0002663883908561183, + "loss": 0.5376, + "step": 26308 + }, + { + "epoch": 1.469927647568232, + "grad_norm": 0.40777257084846497, + "learning_rate": 0.00026636037651277456, + "loss": 0.4521, + "step": 26309 + }, + { + "epoch": 1.4699835181719139, + "grad_norm": 0.4915851652622223, + "learning_rate": 0.00026633236216943077, + "loss": 0.4005, + "step": 26310 + }, + { + "epoch": 1.4700393887755956, + "grad_norm": 0.5612964034080505, + "learning_rate": 0.000266304347826087, + "loss": 0.5235, + "step": 26311 + }, + { + "epoch": 1.4700952593792775, + "grad_norm": 0.4264354705810547, + "learning_rate": 0.0002662763334827432, + "loss": 0.5984, + "step": 26312 + }, + { + "epoch": 1.4701511299829595, + "grad_norm": 0.38931334018707275, + "learning_rate": 0.0002662483191393994, + "loss": 0.4725, + "step": 26313 + }, + { + "epoch": 1.4702070005866412, + "grad_norm": 0.45987898111343384, + "learning_rate": 0.0002662203047960556, + "loss": 0.3653, + "step": 26314 + }, + { + "epoch": 1.4702628711903232, + "grad_norm": 2.822554111480713, + "learning_rate": 0.0002661922904527118, + "loss": 0.3601, + "step": 26315 + }, + { + "epoch": 1.4703187417940051, + "grad_norm": 0.438111811876297, + "learning_rate": 0.000266164276109368, + "loss": 0.4194, + "step": 26316 + }, + { + "epoch": 1.4703746123976869, + "grad_norm": 0.44649970531463623, + "learning_rate": 0.0002661362617660242, + "loss": 0.4994, + "step": 26317 + }, + { + "epoch": 1.4704304830013688, + "grad_norm": 0.40809929370880127, + "learning_rate": 0.0002661082474226804, + "loss": 0.5393, + "step": 26318 + }, + { + "epoch": 1.4704863536050508, + "grad_norm": 0.96542888879776, + "learning_rate": 0.0002660802330793367, + "loss": 0.7012, + "step": 26319 + }, + { + "epoch": 1.4705422242087325, + "grad_norm": 0.6647967100143433, + "learning_rate": 0.0002660522187359928, + "loss": 0.4478, + "step": 26320 + }, + { + "epoch": 1.4705980948124144, + "grad_norm": 0.3826344609260559, + "learning_rate": 0.00026602420439264903, + "loss": 0.4022, + "step": 26321 + }, + { + "epoch": 1.4706539654160964, + "grad_norm": 0.35198962688446045, + "learning_rate": 0.00026599619004930524, + "loss": 0.3464, + "step": 26322 + }, + { + "epoch": 1.4707098360197781, + "grad_norm": 0.4285905063152313, + "learning_rate": 0.00026596817570596144, + "loss": 0.4687, + "step": 26323 + }, + { + "epoch": 1.47076570662346, + "grad_norm": 0.8588554263114929, + "learning_rate": 0.00026594016136261765, + "loss": 0.4631, + "step": 26324 + }, + { + "epoch": 1.470821577227142, + "grad_norm": 0.5923973321914673, + "learning_rate": 0.00026591214701927386, + "loss": 0.3773, + "step": 26325 + }, + { + "epoch": 1.4708774478308237, + "grad_norm": 0.6686844229698181, + "learning_rate": 0.0002658841326759301, + "loss": 0.4978, + "step": 26326 + }, + { + "epoch": 1.4709333184345057, + "grad_norm": 0.35557103157043457, + "learning_rate": 0.00026585611833258627, + "loss": 0.441, + "step": 26327 + }, + { + "epoch": 1.4709891890381877, + "grad_norm": 0.41091451048851013, + "learning_rate": 0.0002658281039892425, + "loss": 0.4195, + "step": 26328 + }, + { + "epoch": 1.4710450596418694, + "grad_norm": 0.39132770895957947, + "learning_rate": 0.0002658000896458987, + "loss": 0.3937, + "step": 26329 + }, + { + "epoch": 1.4711009302455513, + "grad_norm": 0.43539923429489136, + "learning_rate": 0.00026577207530255494, + "loss": 0.4276, + "step": 26330 + }, + { + "epoch": 1.4711568008492333, + "grad_norm": 0.686616837978363, + "learning_rate": 0.0002657440609592111, + "loss": 0.4051, + "step": 26331 + }, + { + "epoch": 1.471212671452915, + "grad_norm": 0.6483572721481323, + "learning_rate": 0.00026571604661586735, + "loss": 0.3373, + "step": 26332 + }, + { + "epoch": 1.471268542056597, + "grad_norm": 0.4443967938423157, + "learning_rate": 0.0002656880322725235, + "loss": 0.4334, + "step": 26333 + }, + { + "epoch": 1.471324412660279, + "grad_norm": 0.3869313895702362, + "learning_rate": 0.00026566001792917976, + "loss": 0.3899, + "step": 26334 + }, + { + "epoch": 1.4713802832639606, + "grad_norm": 0.3495972752571106, + "learning_rate": 0.00026563200358583597, + "loss": 0.3691, + "step": 26335 + }, + { + "epoch": 1.4714361538676426, + "grad_norm": 0.4062095880508423, + "learning_rate": 0.0002656039892424922, + "loss": 0.3882, + "step": 26336 + }, + { + "epoch": 1.4714920244713245, + "grad_norm": 23.19703483581543, + "learning_rate": 0.0002655759748991484, + "loss": 0.317, + "step": 26337 + }, + { + "epoch": 1.4715478950750063, + "grad_norm": 1.6983460187911987, + "learning_rate": 0.0002655479605558046, + "loss": 0.4898, + "step": 26338 + }, + { + "epoch": 1.4716037656786882, + "grad_norm": 0.8121536374092102, + "learning_rate": 0.0002655199462124608, + "loss": 0.4894, + "step": 26339 + }, + { + "epoch": 1.47165963628237, + "grad_norm": 0.45068851113319397, + "learning_rate": 0.000265491931869117, + "loss": 0.4216, + "step": 26340 + }, + { + "epoch": 1.471715506886052, + "grad_norm": 0.4742927849292755, + "learning_rate": 0.0002654639175257732, + "loss": 0.3687, + "step": 26341 + }, + { + "epoch": 1.4717713774897339, + "grad_norm": 0.5178678035736084, + "learning_rate": 0.0002654359031824294, + "loss": 0.3855, + "step": 26342 + }, + { + "epoch": 1.4718272480934156, + "grad_norm": 0.48079004883766174, + "learning_rate": 0.0002654078888390856, + "loss": 0.3968, + "step": 26343 + }, + { + "epoch": 1.4718831186970975, + "grad_norm": 0.3496006727218628, + "learning_rate": 0.0002653798744957418, + "loss": 0.4018, + "step": 26344 + }, + { + "epoch": 1.4719389893007793, + "grad_norm": 0.7446387410163879, + "learning_rate": 0.000265351860152398, + "loss": 0.4673, + "step": 26345 + }, + { + "epoch": 1.4719948599044612, + "grad_norm": 0.5086300373077393, + "learning_rate": 0.00026532384580905423, + "loss": 0.4881, + "step": 26346 + }, + { + "epoch": 1.4720507305081432, + "grad_norm": 0.5237719416618347, + "learning_rate": 0.00026529583146571044, + "loss": 0.3603, + "step": 26347 + }, + { + "epoch": 1.472106601111825, + "grad_norm": 2.3500566482543945, + "learning_rate": 0.00026526781712236664, + "loss": 0.5356, + "step": 26348 + }, + { + "epoch": 1.4721624717155068, + "grad_norm": 0.4004875123500824, + "learning_rate": 0.0002652398027790229, + "loss": 0.4143, + "step": 26349 + }, + { + "epoch": 1.4722183423191888, + "grad_norm": 0.38075748085975647, + "learning_rate": 0.00026521178843567905, + "loss": 0.3343, + "step": 26350 + }, + { + "epoch": 1.4722742129228705, + "grad_norm": 0.3712274432182312, + "learning_rate": 0.0002651837740923353, + "loss": 0.3488, + "step": 26351 + }, + { + "epoch": 1.4723300835265525, + "grad_norm": 0.5136805176734924, + "learning_rate": 0.00026515575974899147, + "loss": 0.4884, + "step": 26352 + }, + { + "epoch": 1.4723859541302344, + "grad_norm": 0.640019416809082, + "learning_rate": 0.0002651277454056477, + "loss": 0.4039, + "step": 26353 + }, + { + "epoch": 1.4724418247339162, + "grad_norm": 0.3487688899040222, + "learning_rate": 0.0002650997310623039, + "loss": 0.4191, + "step": 26354 + }, + { + "epoch": 1.4724976953375981, + "grad_norm": 3.1720077991485596, + "learning_rate": 0.00026507171671896014, + "loss": 0.4774, + "step": 26355 + }, + { + "epoch": 1.47255356594128, + "grad_norm": 0.5712063312530518, + "learning_rate": 0.0002650437023756163, + "loss": 0.3637, + "step": 26356 + }, + { + "epoch": 1.4726094365449618, + "grad_norm": 0.4158768653869629, + "learning_rate": 0.00026501568803227255, + "loss": 0.3418, + "step": 26357 + }, + { + "epoch": 1.4726653071486437, + "grad_norm": 0.44613656401634216, + "learning_rate": 0.00026498767368892876, + "loss": 0.34, + "step": 26358 + }, + { + "epoch": 1.4727211777523257, + "grad_norm": 8.611661911010742, + "learning_rate": 0.00026495965934558496, + "loss": 0.4454, + "step": 26359 + }, + { + "epoch": 1.4727770483560074, + "grad_norm": 0.4627591073513031, + "learning_rate": 0.00026493164500224117, + "loss": 0.4329, + "step": 26360 + }, + { + "epoch": 1.4728329189596894, + "grad_norm": 0.45280832052230835, + "learning_rate": 0.0002649036306588974, + "loss": 0.3808, + "step": 26361 + }, + { + "epoch": 1.4728887895633713, + "grad_norm": 0.46500346064567566, + "learning_rate": 0.0002648756163155536, + "loss": 0.4424, + "step": 26362 + }, + { + "epoch": 1.472944660167053, + "grad_norm": 0.8442691564559937, + "learning_rate": 0.00026484760197220973, + "loss": 0.3875, + "step": 26363 + }, + { + "epoch": 1.473000530770735, + "grad_norm": 0.2932359278202057, + "learning_rate": 0.000264819587628866, + "loss": 0.3293, + "step": 26364 + }, + { + "epoch": 1.473056401374417, + "grad_norm": 0.4744786322116852, + "learning_rate": 0.0002647915732855222, + "loss": 0.4534, + "step": 26365 + }, + { + "epoch": 1.4731122719780987, + "grad_norm": 0.42231225967407227, + "learning_rate": 0.0002647635589421784, + "loss": 0.432, + "step": 26366 + }, + { + "epoch": 1.4731681425817806, + "grad_norm": 0.44417327642440796, + "learning_rate": 0.0002647355445988346, + "loss": 0.4193, + "step": 26367 + }, + { + "epoch": 1.4732240131854626, + "grad_norm": 0.5431002974510193, + "learning_rate": 0.0002647075302554908, + "loss": 0.4165, + "step": 26368 + }, + { + "epoch": 1.4732798837891443, + "grad_norm": 0.3917212188243866, + "learning_rate": 0.000264679515912147, + "loss": 0.3546, + "step": 26369 + }, + { + "epoch": 1.4733357543928263, + "grad_norm": 0.4318731427192688, + "learning_rate": 0.0002646515015688032, + "loss": 0.4566, + "step": 26370 + }, + { + "epoch": 1.4733916249965082, + "grad_norm": 0.3251017928123474, + "learning_rate": 0.00026462348722545943, + "loss": 0.3122, + "step": 26371 + }, + { + "epoch": 1.47344749560019, + "grad_norm": 2.7688639163970947, + "learning_rate": 0.00026459547288211564, + "loss": 0.4211, + "step": 26372 + }, + { + "epoch": 1.473503366203872, + "grad_norm": 0.48193782567977905, + "learning_rate": 0.00026456745853877184, + "loss": 0.3822, + "step": 26373 + }, + { + "epoch": 1.4735592368075536, + "grad_norm": 0.5072848200798035, + "learning_rate": 0.0002645394441954281, + "loss": 0.2934, + "step": 26374 + }, + { + "epoch": 1.4736151074112356, + "grad_norm": 0.5467243790626526, + "learning_rate": 0.00026451142985208425, + "loss": 0.3969, + "step": 26375 + }, + { + "epoch": 1.4736709780149175, + "grad_norm": 0.46888023614883423, + "learning_rate": 0.0002644834155087405, + "loss": 0.4043, + "step": 26376 + }, + { + "epoch": 1.4737268486185993, + "grad_norm": 0.37985095381736755, + "learning_rate": 0.00026445540116539667, + "loss": 0.4168, + "step": 26377 + }, + { + "epoch": 1.4737827192222812, + "grad_norm": 0.570374608039856, + "learning_rate": 0.0002644273868220529, + "loss": 0.5959, + "step": 26378 + }, + { + "epoch": 1.473838589825963, + "grad_norm": 0.6611078381538391, + "learning_rate": 0.0002643993724787091, + "loss": 0.4249, + "step": 26379 + }, + { + "epoch": 1.473894460429645, + "grad_norm": 0.3269822895526886, + "learning_rate": 0.00026437135813536534, + "loss": 0.3024, + "step": 26380 + }, + { + "epoch": 1.4739503310333268, + "grad_norm": 0.3457227945327759, + "learning_rate": 0.00026434334379202154, + "loss": 0.3378, + "step": 26381 + }, + { + "epoch": 1.4740062016370086, + "grad_norm": 0.6063132286071777, + "learning_rate": 0.00026431532944867775, + "loss": 0.4138, + "step": 26382 + }, + { + "epoch": 1.4740620722406905, + "grad_norm": 0.4628012478351593, + "learning_rate": 0.00026428731510533396, + "loss": 0.3946, + "step": 26383 + }, + { + "epoch": 1.4741179428443725, + "grad_norm": 0.5043282508850098, + "learning_rate": 0.0002642593007619901, + "loss": 0.4754, + "step": 26384 + }, + { + "epoch": 1.4741738134480542, + "grad_norm": 0.3548203408718109, + "learning_rate": 0.00026423128641864637, + "loss": 0.4413, + "step": 26385 + }, + { + "epoch": 1.4742296840517362, + "grad_norm": 1.768875002861023, + "learning_rate": 0.0002642032720753025, + "loss": 0.4149, + "step": 26386 + }, + { + "epoch": 1.474285554655418, + "grad_norm": 0.4505743384361267, + "learning_rate": 0.0002641752577319588, + "loss": 0.485, + "step": 26387 + }, + { + "epoch": 1.4743414252590998, + "grad_norm": 0.5132591128349304, + "learning_rate": 0.00026414724338861493, + "loss": 0.4602, + "step": 26388 + }, + { + "epoch": 1.4743972958627818, + "grad_norm": 0.48554468154907227, + "learning_rate": 0.0002641192290452712, + "loss": 0.355, + "step": 26389 + }, + { + "epoch": 1.4744531664664637, + "grad_norm": 0.42286747694015503, + "learning_rate": 0.0002640912147019274, + "loss": 0.4012, + "step": 26390 + }, + { + "epoch": 1.4745090370701455, + "grad_norm": 0.33950597047805786, + "learning_rate": 0.0002640632003585836, + "loss": 0.4386, + "step": 26391 + }, + { + "epoch": 1.4745649076738274, + "grad_norm": 0.6717396378517151, + "learning_rate": 0.0002640351860152398, + "loss": 0.4629, + "step": 26392 + }, + { + "epoch": 1.4746207782775094, + "grad_norm": 0.421053022146225, + "learning_rate": 0.000264007171671896, + "loss": 0.5066, + "step": 26393 + }, + { + "epoch": 1.474676648881191, + "grad_norm": 0.5041390061378479, + "learning_rate": 0.0002639791573285522, + "loss": 0.3305, + "step": 26394 + }, + { + "epoch": 1.474732519484873, + "grad_norm": 0.411080539226532, + "learning_rate": 0.0002639511429852084, + "loss": 0.4579, + "step": 26395 + }, + { + "epoch": 1.474788390088555, + "grad_norm": 0.37828531861305237, + "learning_rate": 0.00026392312864186463, + "loss": 0.3905, + "step": 26396 + }, + { + "epoch": 1.4748442606922367, + "grad_norm": 0.3905521333217621, + "learning_rate": 0.0002638951142985209, + "loss": 0.4606, + "step": 26397 + }, + { + "epoch": 1.4749001312959187, + "grad_norm": 0.6436390280723572, + "learning_rate": 0.00026386709995517704, + "loss": 0.541, + "step": 26398 + }, + { + "epoch": 1.4749560018996006, + "grad_norm": 0.6765650510787964, + "learning_rate": 0.0002638390856118333, + "loss": 0.5429, + "step": 26399 + }, + { + "epoch": 1.4750118725032824, + "grad_norm": 0.5531436800956726, + "learning_rate": 0.00026381107126848945, + "loss": 0.5375, + "step": 26400 + }, + { + "epoch": 1.4750677431069643, + "grad_norm": 0.5072423815727234, + "learning_rate": 0.0002637830569251457, + "loss": 0.4914, + "step": 26401 + }, + { + "epoch": 1.4751236137106463, + "grad_norm": 0.44933074712753296, + "learning_rate": 0.00026375504258180187, + "loss": 0.4568, + "step": 26402 + }, + { + "epoch": 1.475179484314328, + "grad_norm": 0.6067904829978943, + "learning_rate": 0.0002637270282384581, + "loss": 0.4045, + "step": 26403 + }, + { + "epoch": 1.47523535491801, + "grad_norm": 0.3569679260253906, + "learning_rate": 0.0002636990138951143, + "loss": 0.4374, + "step": 26404 + }, + { + "epoch": 1.475291225521692, + "grad_norm": 0.9895254373550415, + "learning_rate": 0.0002636709995517705, + "loss": 0.4072, + "step": 26405 + }, + { + "epoch": 1.4753470961253736, + "grad_norm": 0.44951459765434265, + "learning_rate": 0.00026364298520842674, + "loss": 0.3353, + "step": 26406 + }, + { + "epoch": 1.4754029667290556, + "grad_norm": 0.5404253602027893, + "learning_rate": 0.0002636149708650829, + "loss": 0.4275, + "step": 26407 + }, + { + "epoch": 1.4754588373327373, + "grad_norm": 0.4961984157562256, + "learning_rate": 0.00026358695652173915, + "loss": 0.497, + "step": 26408 + }, + { + "epoch": 1.4755147079364193, + "grad_norm": 0.8926873207092285, + "learning_rate": 0.0002635589421783953, + "loss": 0.3523, + "step": 26409 + }, + { + "epoch": 1.4755705785401012, + "grad_norm": 0.638465940952301, + "learning_rate": 0.00026353092783505157, + "loss": 0.3613, + "step": 26410 + }, + { + "epoch": 1.475626449143783, + "grad_norm": 0.7444284558296204, + "learning_rate": 0.0002635029134917077, + "loss": 0.472, + "step": 26411 + }, + { + "epoch": 1.4756823197474649, + "grad_norm": 0.40081003308296204, + "learning_rate": 0.000263474899148364, + "loss": 0.3846, + "step": 26412 + }, + { + "epoch": 1.4757381903511466, + "grad_norm": 1.0543814897537231, + "learning_rate": 0.0002634468848050202, + "loss": 0.4015, + "step": 26413 + }, + { + "epoch": 1.4757940609548286, + "grad_norm": 0.8591623306274414, + "learning_rate": 0.0002634188704616764, + "loss": 0.394, + "step": 26414 + }, + { + "epoch": 1.4758499315585105, + "grad_norm": 0.5256332159042358, + "learning_rate": 0.0002633908561183326, + "loss": 0.4562, + "step": 26415 + }, + { + "epoch": 1.4759058021621922, + "grad_norm": 0.364409476518631, + "learning_rate": 0.0002633628417749888, + "loss": 0.5164, + "step": 26416 + }, + { + "epoch": 1.4759616727658742, + "grad_norm": 0.4266495108604431, + "learning_rate": 0.000263334827431645, + "loss": 0.4421, + "step": 26417 + }, + { + "epoch": 1.4760175433695562, + "grad_norm": 0.9864848256111145, + "learning_rate": 0.0002633068130883012, + "loss": 0.3549, + "step": 26418 + }, + { + "epoch": 1.4760734139732379, + "grad_norm": 0.5484050512313843, + "learning_rate": 0.0002632787987449574, + "loss": 0.5267, + "step": 26419 + }, + { + "epoch": 1.4761292845769198, + "grad_norm": 0.4118669033050537, + "learning_rate": 0.0002632507844016137, + "loss": 0.382, + "step": 26420 + }, + { + "epoch": 1.4761851551806018, + "grad_norm": 0.7808704972267151, + "learning_rate": 0.00026322277005826983, + "loss": 0.4528, + "step": 26421 + }, + { + "epoch": 1.4762410257842835, + "grad_norm": 0.500434935092926, + "learning_rate": 0.0002631947557149261, + "loss": 0.5366, + "step": 26422 + }, + { + "epoch": 1.4762968963879655, + "grad_norm": 0.4678138494491577, + "learning_rate": 0.00026316674137158224, + "loss": 0.4418, + "step": 26423 + }, + { + "epoch": 1.4763527669916474, + "grad_norm": 0.8523640632629395, + "learning_rate": 0.0002631387270282385, + "loss": 0.4948, + "step": 26424 + }, + { + "epoch": 1.4764086375953291, + "grad_norm": 0.4543602764606476, + "learning_rate": 0.00026311071268489465, + "loss": 0.3773, + "step": 26425 + }, + { + "epoch": 1.476464508199011, + "grad_norm": 0.4541991651058197, + "learning_rate": 0.00026308269834155086, + "loss": 0.5124, + "step": 26426 + }, + { + "epoch": 1.476520378802693, + "grad_norm": 0.32079100608825684, + "learning_rate": 0.00026305468399820707, + "loss": 0.3925, + "step": 26427 + }, + { + "epoch": 1.4765762494063748, + "grad_norm": 0.31349262595176697, + "learning_rate": 0.00026302666965486327, + "loss": 0.3807, + "step": 26428 + }, + { + "epoch": 1.4766321200100567, + "grad_norm": 0.4943876266479492, + "learning_rate": 0.00026299865531151953, + "loss": 0.5065, + "step": 26429 + }, + { + "epoch": 1.4766879906137387, + "grad_norm": 2.1346256732940674, + "learning_rate": 0.0002629706409681757, + "loss": 0.408, + "step": 26430 + }, + { + "epoch": 1.4767438612174204, + "grad_norm": 0.4542417526245117, + "learning_rate": 0.00026294262662483194, + "loss": 0.3844, + "step": 26431 + }, + { + "epoch": 1.4767997318211024, + "grad_norm": 0.5103306174278259, + "learning_rate": 0.0002629146122814881, + "loss": 0.3689, + "step": 26432 + }, + { + "epoch": 1.4768556024247843, + "grad_norm": 0.37451428174972534, + "learning_rate": 0.00026288659793814435, + "loss": 0.3786, + "step": 26433 + }, + { + "epoch": 1.476911473028466, + "grad_norm": 0.34077632427215576, + "learning_rate": 0.0002628585835948005, + "loss": 0.4153, + "step": 26434 + }, + { + "epoch": 1.476967343632148, + "grad_norm": 0.4727122485637665, + "learning_rate": 0.00026283056925145677, + "loss": 0.3832, + "step": 26435 + }, + { + "epoch": 1.47702321423583, + "grad_norm": 0.38306179642677307, + "learning_rate": 0.00026280255490811297, + "loss": 0.442, + "step": 26436 + }, + { + "epoch": 1.4770790848395117, + "grad_norm": 0.5622496008872986, + "learning_rate": 0.0002627745405647692, + "loss": 0.4479, + "step": 26437 + }, + { + "epoch": 1.4771349554431936, + "grad_norm": 0.37294790148735046, + "learning_rate": 0.0002627465262214254, + "loss": 0.4865, + "step": 26438 + }, + { + "epoch": 1.4771908260468756, + "grad_norm": 0.5226739645004272, + "learning_rate": 0.0002627185118780816, + "loss": 0.3744, + "step": 26439 + }, + { + "epoch": 1.4772466966505573, + "grad_norm": 0.6166883111000061, + "learning_rate": 0.0002626904975347378, + "loss": 0.3967, + "step": 26440 + }, + { + "epoch": 1.4773025672542393, + "grad_norm": 0.5062513947486877, + "learning_rate": 0.000262662483191394, + "loss": 0.4746, + "step": 26441 + }, + { + "epoch": 1.477358437857921, + "grad_norm": 0.36058658361434937, + "learning_rate": 0.0002626344688480502, + "loss": 0.4338, + "step": 26442 + }, + { + "epoch": 1.477414308461603, + "grad_norm": 0.3264825642108917, + "learning_rate": 0.0002626064545047064, + "loss": 0.3492, + "step": 26443 + }, + { + "epoch": 1.4774701790652849, + "grad_norm": 0.4864605963230133, + "learning_rate": 0.0002625784401613626, + "loss": 0.5211, + "step": 26444 + }, + { + "epoch": 1.4775260496689666, + "grad_norm": 0.4813495874404907, + "learning_rate": 0.0002625504258180189, + "loss": 0.4736, + "step": 26445 + }, + { + "epoch": 1.4775819202726486, + "grad_norm": 0.6361761093139648, + "learning_rate": 0.00026252241147467503, + "loss": 0.4175, + "step": 26446 + }, + { + "epoch": 1.4776377908763303, + "grad_norm": 0.6053617596626282, + "learning_rate": 0.00026249439713133124, + "loss": 0.3715, + "step": 26447 + }, + { + "epoch": 1.4776936614800122, + "grad_norm": 0.4077340066432953, + "learning_rate": 0.00026246638278798744, + "loss": 0.4059, + "step": 26448 + }, + { + "epoch": 1.4777495320836942, + "grad_norm": 0.364370733499527, + "learning_rate": 0.00026243836844464365, + "loss": 0.4124, + "step": 26449 + }, + { + "epoch": 1.477805402687376, + "grad_norm": 0.6130596399307251, + "learning_rate": 0.00026241035410129985, + "loss": 0.5471, + "step": 26450 + }, + { + "epoch": 1.4778612732910579, + "grad_norm": 0.4201354384422302, + "learning_rate": 0.00026238233975795606, + "loss": 0.3732, + "step": 26451 + }, + { + "epoch": 1.4779171438947398, + "grad_norm": 0.49926289916038513, + "learning_rate": 0.0002623543254146123, + "loss": 0.5004, + "step": 26452 + }, + { + "epoch": 1.4779730144984216, + "grad_norm": 0.3611316680908203, + "learning_rate": 0.00026232631107126847, + "loss": 0.4939, + "step": 26453 + }, + { + "epoch": 1.4780288851021035, + "grad_norm": 0.7750571370124817, + "learning_rate": 0.00026229829672792473, + "loss": 0.4814, + "step": 26454 + }, + { + "epoch": 1.4780847557057855, + "grad_norm": 0.4070029556751251, + "learning_rate": 0.0002622702823845809, + "loss": 0.4476, + "step": 26455 + }, + { + "epoch": 1.4781406263094672, + "grad_norm": 0.5537377595901489, + "learning_rate": 0.00026224226804123714, + "loss": 0.4659, + "step": 26456 + }, + { + "epoch": 1.4781964969131491, + "grad_norm": 0.4101667106151581, + "learning_rate": 0.0002622142536978933, + "loss": 0.3549, + "step": 26457 + }, + { + "epoch": 1.478252367516831, + "grad_norm": 0.7471737861633301, + "learning_rate": 0.00026218623935454955, + "loss": 0.4594, + "step": 26458 + }, + { + "epoch": 1.4783082381205128, + "grad_norm": 0.6697216629981995, + "learning_rate": 0.0002621582250112057, + "loss": 0.4012, + "step": 26459 + }, + { + "epoch": 1.4783641087241948, + "grad_norm": 0.38304826617240906, + "learning_rate": 0.00026213021066786197, + "loss": 0.3843, + "step": 26460 + }, + { + "epoch": 1.4784199793278767, + "grad_norm": 0.928489089012146, + "learning_rate": 0.00026210219632451817, + "loss": 0.5456, + "step": 26461 + }, + { + "epoch": 1.4784758499315585, + "grad_norm": 2.630584478378296, + "learning_rate": 0.0002620741819811744, + "loss": 0.4634, + "step": 26462 + }, + { + "epoch": 1.4785317205352404, + "grad_norm": 0.5347984433174133, + "learning_rate": 0.0002620461676378306, + "loss": 0.3974, + "step": 26463 + }, + { + "epoch": 1.4785875911389224, + "grad_norm": 0.4883876144886017, + "learning_rate": 0.0002620181532944868, + "loss": 0.4261, + "step": 26464 + }, + { + "epoch": 1.478643461742604, + "grad_norm": 0.670250415802002, + "learning_rate": 0.000261990138951143, + "loss": 0.4131, + "step": 26465 + }, + { + "epoch": 1.478699332346286, + "grad_norm": 0.4481145739555359, + "learning_rate": 0.0002619621246077992, + "loss": 0.3723, + "step": 26466 + }, + { + "epoch": 1.478755202949968, + "grad_norm": 0.8334583640098572, + "learning_rate": 0.0002619341102644554, + "loss": 0.3948, + "step": 26467 + }, + { + "epoch": 1.4788110735536497, + "grad_norm": 0.4461888074874878, + "learning_rate": 0.00026190609592111167, + "loss": 0.4669, + "step": 26468 + }, + { + "epoch": 1.4788669441573317, + "grad_norm": 0.5045279860496521, + "learning_rate": 0.0002618780815777678, + "loss": 0.3996, + "step": 26469 + }, + { + "epoch": 1.4789228147610136, + "grad_norm": 0.4536230266094208, + "learning_rate": 0.000261850067234424, + "loss": 0.3586, + "step": 26470 + }, + { + "epoch": 1.4789786853646953, + "grad_norm": 0.72591233253479, + "learning_rate": 0.00026182205289108023, + "loss": 0.4482, + "step": 26471 + }, + { + "epoch": 1.4790345559683773, + "grad_norm": 0.6419498920440674, + "learning_rate": 0.00026179403854773644, + "loss": 0.4868, + "step": 26472 + }, + { + "epoch": 1.4790904265720592, + "grad_norm": 0.519203782081604, + "learning_rate": 0.00026176602420439264, + "loss": 0.4382, + "step": 26473 + }, + { + "epoch": 1.479146297175741, + "grad_norm": 0.5573003888130188, + "learning_rate": 0.00026173800986104885, + "loss": 0.4389, + "step": 26474 + }, + { + "epoch": 1.479202167779423, + "grad_norm": 1.4441336393356323, + "learning_rate": 0.00026170999551770505, + "loss": 0.3946, + "step": 26475 + }, + { + "epoch": 1.4792580383831047, + "grad_norm": 0.4013555347919464, + "learning_rate": 0.00026168198117436126, + "loss": 0.38, + "step": 26476 + }, + { + "epoch": 1.4793139089867866, + "grad_norm": 3.6703908443450928, + "learning_rate": 0.0002616539668310175, + "loss": 0.3964, + "step": 26477 + }, + { + "epoch": 1.4793697795904683, + "grad_norm": 0.6233845353126526, + "learning_rate": 0.00026162595248767367, + "loss": 0.4621, + "step": 26478 + }, + { + "epoch": 1.4794256501941503, + "grad_norm": 0.3398776352405548, + "learning_rate": 0.00026159793814432993, + "loss": 0.3627, + "step": 26479 + }, + { + "epoch": 1.4794815207978322, + "grad_norm": 0.4846530556678772, + "learning_rate": 0.0002615699238009861, + "loss": 0.5286, + "step": 26480 + }, + { + "epoch": 1.479537391401514, + "grad_norm": 0.7470941543579102, + "learning_rate": 0.00026154190945764234, + "loss": 0.5045, + "step": 26481 + }, + { + "epoch": 1.479593262005196, + "grad_norm": 0.9913563132286072, + "learning_rate": 0.0002615138951142985, + "loss": 0.4588, + "step": 26482 + }, + { + "epoch": 1.4796491326088779, + "grad_norm": 0.9132216572761536, + "learning_rate": 0.00026148588077095475, + "loss": 0.3871, + "step": 26483 + }, + { + "epoch": 1.4797050032125596, + "grad_norm": 0.3353070020675659, + "learning_rate": 0.00026145786642761096, + "loss": 0.3937, + "step": 26484 + }, + { + "epoch": 1.4797608738162416, + "grad_norm": 0.4605007469654083, + "learning_rate": 0.00026142985208426717, + "loss": 0.5017, + "step": 26485 + }, + { + "epoch": 1.4798167444199235, + "grad_norm": 0.6902381181716919, + "learning_rate": 0.00026140183774092337, + "loss": 0.4318, + "step": 26486 + }, + { + "epoch": 1.4798726150236052, + "grad_norm": 0.7749120593070984, + "learning_rate": 0.0002613738233975796, + "loss": 0.6142, + "step": 26487 + }, + { + "epoch": 1.4799284856272872, + "grad_norm": 0.45147737860679626, + "learning_rate": 0.0002613458090542358, + "loss": 0.4267, + "step": 26488 + }, + { + "epoch": 1.4799843562309691, + "grad_norm": 0.3949648439884186, + "learning_rate": 0.000261317794710892, + "loss": 0.3643, + "step": 26489 + }, + { + "epoch": 1.4800402268346509, + "grad_norm": 0.39968228340148926, + "learning_rate": 0.0002612897803675482, + "loss": 0.3851, + "step": 26490 + }, + { + "epoch": 1.4800960974383328, + "grad_norm": 0.4786425828933716, + "learning_rate": 0.00026126176602420435, + "loss": 0.4204, + "step": 26491 + }, + { + "epoch": 1.4801519680420148, + "grad_norm": 0.6673839092254639, + "learning_rate": 0.0002612337516808606, + "loss": 0.4547, + "step": 26492 + }, + { + "epoch": 1.4802078386456965, + "grad_norm": 0.6267326474189758, + "learning_rate": 0.0002612057373375168, + "loss": 0.4415, + "step": 26493 + }, + { + "epoch": 1.4802637092493784, + "grad_norm": 0.37584325671195984, + "learning_rate": 0.000261177722994173, + "loss": 0.4374, + "step": 26494 + }, + { + "epoch": 1.4803195798530604, + "grad_norm": 0.4062919020652771, + "learning_rate": 0.0002611497086508292, + "loss": 0.2902, + "step": 26495 + }, + { + "epoch": 1.4803754504567421, + "grad_norm": 0.42754366993904114, + "learning_rate": 0.00026112169430748543, + "loss": 0.3343, + "step": 26496 + }, + { + "epoch": 1.480431321060424, + "grad_norm": 0.7397001385688782, + "learning_rate": 0.00026109367996414163, + "loss": 0.5219, + "step": 26497 + }, + { + "epoch": 1.480487191664106, + "grad_norm": 1.1893954277038574, + "learning_rate": 0.00026106566562079784, + "loss": 0.3498, + "step": 26498 + }, + { + "epoch": 1.4805430622677878, + "grad_norm": 0.7353065013885498, + "learning_rate": 0.00026103765127745405, + "loss": 0.4705, + "step": 26499 + }, + { + "epoch": 1.4805989328714697, + "grad_norm": 0.6637658476829529, + "learning_rate": 0.0002610096369341103, + "loss": 0.4587, + "step": 26500 + }, + { + "epoch": 1.4805989328714697, + "eval_cer": 0.08511450798118993, + "eval_loss": 0.31833064556121826, + "eval_runtime": 56.324, + "eval_samples_per_second": 80.57, + "eval_steps_per_second": 5.042, + "eval_wer": 0.33596826904276417, + "step": 26500 + }, + { + "epoch": 1.4806548034751517, + "grad_norm": 0.4005553126335144, + "learning_rate": 0.00026098162259076646, + "loss": 0.4402, + "step": 26501 + }, + { + "epoch": 1.4807106740788334, + "grad_norm": 0.6042585968971252, + "learning_rate": 0.0002609536082474227, + "loss": 0.513, + "step": 26502 + }, + { + "epoch": 1.4807665446825153, + "grad_norm": 0.7191950082778931, + "learning_rate": 0.00026092559390407887, + "loss": 0.5313, + "step": 26503 + }, + { + "epoch": 1.4808224152861973, + "grad_norm": 0.35561397671699524, + "learning_rate": 0.00026089757956073513, + "loss": 0.38, + "step": 26504 + }, + { + "epoch": 1.480878285889879, + "grad_norm": 0.3701673150062561, + "learning_rate": 0.0002608695652173913, + "loss": 0.3746, + "step": 26505 + }, + { + "epoch": 1.480934156493561, + "grad_norm": 0.3924461901187897, + "learning_rate": 0.00026084155087404754, + "loss": 0.3686, + "step": 26506 + }, + { + "epoch": 1.480990027097243, + "grad_norm": 0.4536309838294983, + "learning_rate": 0.00026081353653070375, + "loss": 0.5123, + "step": 26507 + }, + { + "epoch": 1.4810458977009247, + "grad_norm": 0.37347865104675293, + "learning_rate": 0.00026078552218735995, + "loss": 0.4999, + "step": 26508 + }, + { + "epoch": 1.4811017683046066, + "grad_norm": 0.3604888319969177, + "learning_rate": 0.00026075750784401616, + "loss": 0.4235, + "step": 26509 + }, + { + "epoch": 1.4811576389082883, + "grad_norm": 0.8306170701980591, + "learning_rate": 0.00026072949350067236, + "loss": 0.3886, + "step": 26510 + }, + { + "epoch": 1.4812135095119703, + "grad_norm": 0.644685685634613, + "learning_rate": 0.00026070147915732857, + "loss": 0.5324, + "step": 26511 + }, + { + "epoch": 1.481269380115652, + "grad_norm": 0.9753699898719788, + "learning_rate": 0.0002606734648139847, + "loss": 0.46, + "step": 26512 + }, + { + "epoch": 1.481325250719334, + "grad_norm": 0.312921941280365, + "learning_rate": 0.000260645450470641, + "loss": 0.3773, + "step": 26513 + }, + { + "epoch": 1.481381121323016, + "grad_norm": 2.4449613094329834, + "learning_rate": 0.00026061743612729713, + "loss": 0.325, + "step": 26514 + }, + { + "epoch": 1.4814369919266976, + "grad_norm": 0.3593847155570984, + "learning_rate": 0.0002605894217839534, + "loss": 0.378, + "step": 26515 + }, + { + "epoch": 1.4814928625303796, + "grad_norm": 1.120644211769104, + "learning_rate": 0.0002605614074406096, + "loss": 0.3666, + "step": 26516 + }, + { + "epoch": 1.4815487331340615, + "grad_norm": 0.4982132017612457, + "learning_rate": 0.0002605333930972658, + "loss": 0.3821, + "step": 26517 + }, + { + "epoch": 1.4816046037377433, + "grad_norm": 0.6062407493591309, + "learning_rate": 0.000260505378753922, + "loss": 0.4767, + "step": 26518 + }, + { + "epoch": 1.4816604743414252, + "grad_norm": 0.4250244200229645, + "learning_rate": 0.0002604773644105782, + "loss": 0.4815, + "step": 26519 + }, + { + "epoch": 1.4817163449451072, + "grad_norm": 0.9288024306297302, + "learning_rate": 0.0002604493500672344, + "loss": 0.392, + "step": 26520 + }, + { + "epoch": 1.481772215548789, + "grad_norm": 0.4592486321926117, + "learning_rate": 0.00026042133572389063, + "loss": 0.4708, + "step": 26521 + }, + { + "epoch": 1.4818280861524709, + "grad_norm": 0.44552525877952576, + "learning_rate": 0.00026039332138054683, + "loss": 0.426, + "step": 26522 + }, + { + "epoch": 1.4818839567561528, + "grad_norm": 0.47622808814048767, + "learning_rate": 0.0002603653070372031, + "loss": 0.4652, + "step": 26523 + }, + { + "epoch": 1.4819398273598345, + "grad_norm": 0.3879536986351013, + "learning_rate": 0.00026033729269385925, + "loss": 0.433, + "step": 26524 + }, + { + "epoch": 1.4819956979635165, + "grad_norm": 0.35368138551712036, + "learning_rate": 0.0002603092783505155, + "loss": 0.3449, + "step": 26525 + }, + { + "epoch": 1.4820515685671984, + "grad_norm": 0.44238173961639404, + "learning_rate": 0.00026028126400717166, + "loss": 0.5114, + "step": 26526 + }, + { + "epoch": 1.4821074391708802, + "grad_norm": 2.8682918548583984, + "learning_rate": 0.0002602532496638279, + "loss": 0.4869, + "step": 26527 + }, + { + "epoch": 1.4821633097745621, + "grad_norm": 0.3939684331417084, + "learning_rate": 0.00026022523532048407, + "loss": 0.3854, + "step": 26528 + }, + { + "epoch": 1.482219180378244, + "grad_norm": 0.39718419313430786, + "learning_rate": 0.00026019722097714033, + "loss": 0.4389, + "step": 26529 + }, + { + "epoch": 1.4822750509819258, + "grad_norm": 0.4972341060638428, + "learning_rate": 0.0002601692066337965, + "loss": 0.4761, + "step": 26530 + }, + { + "epoch": 1.4823309215856078, + "grad_norm": 0.3806841969490051, + "learning_rate": 0.00026014119229045274, + "loss": 0.437, + "step": 26531 + }, + { + "epoch": 1.4823867921892897, + "grad_norm": 0.6194173097610474, + "learning_rate": 0.00026011317794710895, + "loss": 0.4093, + "step": 26532 + }, + { + "epoch": 1.4824426627929714, + "grad_norm": 0.31615522503852844, + "learning_rate": 0.0002600851636037651, + "loss": 0.425, + "step": 26533 + }, + { + "epoch": 1.4824985333966534, + "grad_norm": 0.4520016014575958, + "learning_rate": 0.00026005714926042136, + "loss": 0.4422, + "step": 26534 + }, + { + "epoch": 1.4825544040003353, + "grad_norm": 3.0606892108917236, + "learning_rate": 0.0002600291349170775, + "loss": 0.4781, + "step": 26535 + }, + { + "epoch": 1.482610274604017, + "grad_norm": 0.6503085494041443, + "learning_rate": 0.00026000112057373377, + "loss": 0.3959, + "step": 26536 + }, + { + "epoch": 1.482666145207699, + "grad_norm": 0.5710634589195251, + "learning_rate": 0.0002599731062303899, + "loss": 0.3458, + "step": 26537 + }, + { + "epoch": 1.482722015811381, + "grad_norm": 0.4185311794281006, + "learning_rate": 0.0002599450918870462, + "loss": 0.3828, + "step": 26538 + }, + { + "epoch": 1.4827778864150627, + "grad_norm": 0.36645209789276123, + "learning_rate": 0.0002599170775437024, + "loss": 0.3641, + "step": 26539 + }, + { + "epoch": 1.4828337570187446, + "grad_norm": 0.36189502477645874, + "learning_rate": 0.0002598890632003586, + "loss": 0.378, + "step": 26540 + }, + { + "epoch": 1.4828896276224266, + "grad_norm": 0.37118709087371826, + "learning_rate": 0.0002598610488570148, + "loss": 0.4096, + "step": 26541 + }, + { + "epoch": 1.4829454982261083, + "grad_norm": 0.2967815399169922, + "learning_rate": 0.000259833034513671, + "loss": 0.4211, + "step": 26542 + }, + { + "epoch": 1.4830013688297903, + "grad_norm": 2.716198205947876, + "learning_rate": 0.0002598050201703272, + "loss": 0.5353, + "step": 26543 + }, + { + "epoch": 1.483057239433472, + "grad_norm": 0.4377076029777527, + "learning_rate": 0.0002597770058269834, + "loss": 0.3564, + "step": 26544 + }, + { + "epoch": 1.483113110037154, + "grad_norm": 0.3956722319126129, + "learning_rate": 0.0002597489914836396, + "loss": 0.5297, + "step": 26545 + }, + { + "epoch": 1.4831689806408357, + "grad_norm": 0.4477604627609253, + "learning_rate": 0.00025972097714029583, + "loss": 0.383, + "step": 26546 + }, + { + "epoch": 1.4832248512445176, + "grad_norm": 0.42635729908943176, + "learning_rate": 0.00025969296279695203, + "loss": 0.4357, + "step": 26547 + }, + { + "epoch": 1.4832807218481996, + "grad_norm": 0.3337779939174652, + "learning_rate": 0.0002596649484536083, + "loss": 0.4524, + "step": 26548 + }, + { + "epoch": 1.4833365924518813, + "grad_norm": 1.2523871660232544, + "learning_rate": 0.00025963693411026445, + "loss": 0.4474, + "step": 26549 + }, + { + "epoch": 1.4833924630555633, + "grad_norm": 0.37108203768730164, + "learning_rate": 0.0002596089197669207, + "loss": 0.3444, + "step": 26550 + }, + { + "epoch": 1.4834483336592452, + "grad_norm": 0.6855967044830322, + "learning_rate": 0.00025958090542357686, + "loss": 0.3677, + "step": 26551 + }, + { + "epoch": 1.483504204262927, + "grad_norm": 0.525210440158844, + "learning_rate": 0.0002595528910802331, + "loss": 0.3242, + "step": 26552 + }, + { + "epoch": 1.483560074866609, + "grad_norm": 1.2016615867614746, + "learning_rate": 0.00025952487673688927, + "loss": 0.4296, + "step": 26553 + }, + { + "epoch": 1.4836159454702909, + "grad_norm": 0.39813247323036194, + "learning_rate": 0.0002594968623935455, + "loss": 0.4149, + "step": 26554 + }, + { + "epoch": 1.4836718160739726, + "grad_norm": 0.6326019763946533, + "learning_rate": 0.00025946884805020173, + "loss": 0.4231, + "step": 26555 + }, + { + "epoch": 1.4837276866776545, + "grad_norm": 0.6367979049682617, + "learning_rate": 0.0002594408337068579, + "loss": 0.3561, + "step": 26556 + }, + { + "epoch": 1.4837835572813365, + "grad_norm": 0.3780694305896759, + "learning_rate": 0.00025941281936351415, + "loss": 0.4271, + "step": 26557 + }, + { + "epoch": 1.4838394278850182, + "grad_norm": 0.3898679316043854, + "learning_rate": 0.0002593848050201703, + "loss": 0.3578, + "step": 26558 + }, + { + "epoch": 1.4838952984887002, + "grad_norm": 2.282057762145996, + "learning_rate": 0.00025935679067682656, + "loss": 0.3617, + "step": 26559 + }, + { + "epoch": 1.4839511690923821, + "grad_norm": 1.0444315671920776, + "learning_rate": 0.0002593287763334827, + "loss": 0.3951, + "step": 26560 + }, + { + "epoch": 1.4840070396960638, + "grad_norm": 0.36778539419174194, + "learning_rate": 0.00025930076199013897, + "loss": 0.4083, + "step": 26561 + }, + { + "epoch": 1.4840629102997458, + "grad_norm": 0.44153866171836853, + "learning_rate": 0.0002592727476467951, + "loss": 0.4075, + "step": 26562 + }, + { + "epoch": 1.4841187809034277, + "grad_norm": 0.48620250821113586, + "learning_rate": 0.0002592447333034514, + "loss": 0.4456, + "step": 26563 + }, + { + "epoch": 1.4841746515071095, + "grad_norm": 0.39481061697006226, + "learning_rate": 0.0002592167189601076, + "loss": 0.4601, + "step": 26564 + }, + { + "epoch": 1.4842305221107914, + "grad_norm": 0.429703027009964, + "learning_rate": 0.0002591887046167638, + "loss": 0.4168, + "step": 26565 + }, + { + "epoch": 1.4842863927144734, + "grad_norm": 0.40719929337501526, + "learning_rate": 0.00025916069027342, + "loss": 0.483, + "step": 26566 + }, + { + "epoch": 1.484342263318155, + "grad_norm": 0.6063939332962036, + "learning_rate": 0.0002591326759300762, + "loss": 0.4448, + "step": 26567 + }, + { + "epoch": 1.484398133921837, + "grad_norm": 0.4440094232559204, + "learning_rate": 0.0002591046615867324, + "loss": 0.4344, + "step": 26568 + }, + { + "epoch": 1.484454004525519, + "grad_norm": 0.5654959082603455, + "learning_rate": 0.0002590766472433886, + "loss": 0.3401, + "step": 26569 + }, + { + "epoch": 1.4845098751292007, + "grad_norm": 0.46724334359169006, + "learning_rate": 0.0002590486329000448, + "loss": 0.4202, + "step": 26570 + }, + { + "epoch": 1.4845657457328827, + "grad_norm": 0.3739475905895233, + "learning_rate": 0.0002590206185567011, + "loss": 0.4481, + "step": 26571 + }, + { + "epoch": 1.4846216163365646, + "grad_norm": 1.2769032716751099, + "learning_rate": 0.00025899260421335723, + "loss": 0.5619, + "step": 26572 + }, + { + "epoch": 1.4846774869402464, + "grad_norm": 0.3603380024433136, + "learning_rate": 0.0002589645898700135, + "loss": 0.3647, + "step": 26573 + }, + { + "epoch": 1.4847333575439283, + "grad_norm": 0.6922237277030945, + "learning_rate": 0.00025893657552666965, + "loss": 0.5738, + "step": 26574 + }, + { + "epoch": 1.48478922814761, + "grad_norm": 0.6826455593109131, + "learning_rate": 0.00025890856118332585, + "loss": 0.3995, + "step": 26575 + }, + { + "epoch": 1.484845098751292, + "grad_norm": 0.4184112846851349, + "learning_rate": 0.00025888054683998206, + "loss": 0.4555, + "step": 26576 + }, + { + "epoch": 1.484900969354974, + "grad_norm": 0.452779620885849, + "learning_rate": 0.00025885253249663826, + "loss": 0.3041, + "step": 26577 + }, + { + "epoch": 1.4849568399586557, + "grad_norm": 0.4487784802913666, + "learning_rate": 0.0002588245181532945, + "loss": 0.4139, + "step": 26578 + }, + { + "epoch": 1.4850127105623376, + "grad_norm": 0.36686792969703674, + "learning_rate": 0.0002587965038099507, + "loss": 0.4064, + "step": 26579 + }, + { + "epoch": 1.4850685811660194, + "grad_norm": 0.6255607604980469, + "learning_rate": 0.00025876848946660693, + "loss": 0.4907, + "step": 26580 + }, + { + "epoch": 1.4851244517697013, + "grad_norm": 0.9148398637771606, + "learning_rate": 0.0002587404751232631, + "loss": 0.4416, + "step": 26581 + }, + { + "epoch": 1.4851803223733833, + "grad_norm": 0.4949028491973877, + "learning_rate": 0.00025871246077991935, + "loss": 0.3628, + "step": 26582 + }, + { + "epoch": 1.485236192977065, + "grad_norm": 0.43781930208206177, + "learning_rate": 0.0002586844464365755, + "loss": 0.3495, + "step": 26583 + }, + { + "epoch": 1.485292063580747, + "grad_norm": 0.384892076253891, + "learning_rate": 0.00025865643209323176, + "loss": 0.4038, + "step": 26584 + }, + { + "epoch": 1.485347934184429, + "grad_norm": 0.4262393116950989, + "learning_rate": 0.0002586284177498879, + "loss": 0.4806, + "step": 26585 + }, + { + "epoch": 1.4854038047881106, + "grad_norm": 0.8553087711334229, + "learning_rate": 0.00025860040340654417, + "loss": 0.4663, + "step": 26586 + }, + { + "epoch": 1.4854596753917926, + "grad_norm": 0.3357478678226471, + "learning_rate": 0.0002585723890632004, + "loss": 0.3471, + "step": 26587 + }, + { + "epoch": 1.4855155459954745, + "grad_norm": 4.692857265472412, + "learning_rate": 0.0002585443747198566, + "loss": 0.4393, + "step": 26588 + }, + { + "epoch": 1.4855714165991563, + "grad_norm": 0.3443882465362549, + "learning_rate": 0.0002585163603765128, + "loss": 0.4541, + "step": 26589 + }, + { + "epoch": 1.4856272872028382, + "grad_norm": 0.4472997188568115, + "learning_rate": 0.000258488346033169, + "loss": 0.4416, + "step": 26590 + }, + { + "epoch": 1.4856831578065202, + "grad_norm": 6.6955485343933105, + "learning_rate": 0.0002584603316898252, + "loss": 0.3753, + "step": 26591 + }, + { + "epoch": 1.485739028410202, + "grad_norm": 0.5781491994857788, + "learning_rate": 0.0002584323173464814, + "loss": 0.5838, + "step": 26592 + }, + { + "epoch": 1.4857948990138838, + "grad_norm": 0.4999300241470337, + "learning_rate": 0.0002584043030031376, + "loss": 0.5723, + "step": 26593 + }, + { + "epoch": 1.4858507696175658, + "grad_norm": 0.4302813708782196, + "learning_rate": 0.00025837628865979387, + "loss": 0.3608, + "step": 26594 + }, + { + "epoch": 1.4859066402212475, + "grad_norm": 0.4544837474822998, + "learning_rate": 0.00025834827431645, + "loss": 0.4062, + "step": 26595 + }, + { + "epoch": 1.4859625108249295, + "grad_norm": 1.2106529474258423, + "learning_rate": 0.00025832025997310623, + "loss": 0.4095, + "step": 26596 + }, + { + "epoch": 1.4860183814286114, + "grad_norm": 0.6010449528694153, + "learning_rate": 0.00025829224562976243, + "loss": 0.5016, + "step": 26597 + }, + { + "epoch": 1.4860742520322932, + "grad_norm": 0.36018848419189453, + "learning_rate": 0.00025826423128641864, + "loss": 0.3669, + "step": 26598 + }, + { + "epoch": 1.486130122635975, + "grad_norm": 0.4510480761528015, + "learning_rate": 0.00025823621694307484, + "loss": 0.4192, + "step": 26599 + }, + { + "epoch": 1.486185993239657, + "grad_norm": 0.5642402768135071, + "learning_rate": 0.00025820820259973105, + "loss": 0.3891, + "step": 26600 + }, + { + "epoch": 1.4862418638433388, + "grad_norm": 0.5728122591972351, + "learning_rate": 0.00025818018825638726, + "loss": 0.4621, + "step": 26601 + }, + { + "epoch": 1.4862977344470207, + "grad_norm": 0.4300774931907654, + "learning_rate": 0.00025815217391304346, + "loss": 0.4546, + "step": 26602 + }, + { + "epoch": 1.4863536050507027, + "grad_norm": 0.8043560981750488, + "learning_rate": 0.0002581241595696997, + "loss": 0.3501, + "step": 26603 + }, + { + "epoch": 1.4864094756543844, + "grad_norm": 0.3926759958267212, + "learning_rate": 0.0002580961452263559, + "loss": 0.4356, + "step": 26604 + }, + { + "epoch": 1.4864653462580664, + "grad_norm": 0.36940962076187134, + "learning_rate": 0.00025806813088301213, + "loss": 0.4949, + "step": 26605 + }, + { + "epoch": 1.4865212168617483, + "grad_norm": 0.5866392850875854, + "learning_rate": 0.0002580401165396683, + "loss": 0.3354, + "step": 26606 + }, + { + "epoch": 1.48657708746543, + "grad_norm": 0.35162657499313354, + "learning_rate": 0.00025801210219632455, + "loss": 0.389, + "step": 26607 + }, + { + "epoch": 1.486632958069112, + "grad_norm": 0.4365301728248596, + "learning_rate": 0.0002579840878529807, + "loss": 0.4182, + "step": 26608 + }, + { + "epoch": 1.4866888286727937, + "grad_norm": 0.4496482312679291, + "learning_rate": 0.00025795607350963696, + "loss": 0.5229, + "step": 26609 + }, + { + "epoch": 1.4867446992764757, + "grad_norm": 0.5931729674339294, + "learning_rate": 0.00025792805916629316, + "loss": 0.372, + "step": 26610 + }, + { + "epoch": 1.4868005698801576, + "grad_norm": 0.4675448536872864, + "learning_rate": 0.00025790004482294937, + "loss": 0.4977, + "step": 26611 + }, + { + "epoch": 1.4868564404838394, + "grad_norm": 0.34026384353637695, + "learning_rate": 0.0002578720304796056, + "loss": 0.3696, + "step": 26612 + }, + { + "epoch": 1.4869123110875213, + "grad_norm": 0.3526687026023865, + "learning_rate": 0.0002578440161362618, + "loss": 0.3509, + "step": 26613 + }, + { + "epoch": 1.486968181691203, + "grad_norm": 0.49977144598960876, + "learning_rate": 0.000257816001792918, + "loss": 0.3396, + "step": 26614 + }, + { + "epoch": 1.487024052294885, + "grad_norm": 0.35519540309906006, + "learning_rate": 0.0002577879874495742, + "loss": 0.4218, + "step": 26615 + }, + { + "epoch": 1.487079922898567, + "grad_norm": 0.5122429728507996, + "learning_rate": 0.0002577599731062304, + "loss": 0.5152, + "step": 26616 + }, + { + "epoch": 1.4871357935022487, + "grad_norm": 0.46152424812316895, + "learning_rate": 0.00025773195876288655, + "loss": 0.4484, + "step": 26617 + }, + { + "epoch": 1.4871916641059306, + "grad_norm": 6.79889440536499, + "learning_rate": 0.0002577039444195428, + "loss": 0.3534, + "step": 26618 + }, + { + "epoch": 1.4872475347096126, + "grad_norm": 0.43492263555526733, + "learning_rate": 0.000257675930076199, + "loss": 0.4301, + "step": 26619 + }, + { + "epoch": 1.4873034053132943, + "grad_norm": 0.4016093909740448, + "learning_rate": 0.0002576479157328552, + "loss": 0.4314, + "step": 26620 + }, + { + "epoch": 1.4873592759169763, + "grad_norm": 0.346035361289978, + "learning_rate": 0.00025761990138951143, + "loss": 0.3912, + "step": 26621 + }, + { + "epoch": 1.4874151465206582, + "grad_norm": 0.4892879128456116, + "learning_rate": 0.00025759188704616763, + "loss": 0.3532, + "step": 26622 + }, + { + "epoch": 1.48747101712434, + "grad_norm": 0.537389874458313, + "learning_rate": 0.00025756387270282384, + "loss": 0.2851, + "step": 26623 + }, + { + "epoch": 1.4875268877280219, + "grad_norm": 0.42179128527641296, + "learning_rate": 0.00025753585835948004, + "loss": 0.4901, + "step": 26624 + }, + { + "epoch": 1.4875827583317038, + "grad_norm": 0.3558533191680908, + "learning_rate": 0.00025750784401613625, + "loss": 0.4826, + "step": 26625 + }, + { + "epoch": 1.4876386289353856, + "grad_norm": 0.4095960259437561, + "learning_rate": 0.0002574798296727925, + "loss": 0.4247, + "step": 26626 + }, + { + "epoch": 1.4876944995390675, + "grad_norm": 0.47004368901252747, + "learning_rate": 0.00025745181532944866, + "loss": 0.3568, + "step": 26627 + }, + { + "epoch": 1.4877503701427495, + "grad_norm": 0.4006637632846832, + "learning_rate": 0.0002574238009861049, + "loss": 0.3814, + "step": 26628 + }, + { + "epoch": 1.4878062407464312, + "grad_norm": 0.35570746660232544, + "learning_rate": 0.0002573957866427611, + "loss": 0.4588, + "step": 26629 + }, + { + "epoch": 1.4878621113501131, + "grad_norm": 0.5322726964950562, + "learning_rate": 0.00025736777229941733, + "loss": 0.4675, + "step": 26630 + }, + { + "epoch": 1.487917981953795, + "grad_norm": 0.39304983615875244, + "learning_rate": 0.0002573397579560735, + "loss": 0.4591, + "step": 26631 + }, + { + "epoch": 1.4879738525574768, + "grad_norm": 0.9072357416152954, + "learning_rate": 0.00025731174361272975, + "loss": 0.363, + "step": 26632 + }, + { + "epoch": 1.4880297231611588, + "grad_norm": 0.6091480255126953, + "learning_rate": 0.00025728372926938595, + "loss": 0.4506, + "step": 26633 + }, + { + "epoch": 1.4880855937648407, + "grad_norm": 0.43041351437568665, + "learning_rate": 0.00025725571492604216, + "loss": 0.425, + "step": 26634 + }, + { + "epoch": 1.4881414643685225, + "grad_norm": 0.6625931859016418, + "learning_rate": 0.00025722770058269836, + "loss": 0.5634, + "step": 26635 + }, + { + "epoch": 1.4881973349722044, + "grad_norm": 0.38587820529937744, + "learning_rate": 0.00025719968623935457, + "loss": 0.4192, + "step": 26636 + }, + { + "epoch": 1.4882532055758864, + "grad_norm": 0.543972909450531, + "learning_rate": 0.0002571716718960108, + "loss": 0.4597, + "step": 26637 + }, + { + "epoch": 1.488309076179568, + "grad_norm": 0.4898339509963989, + "learning_rate": 0.000257143657552667, + "loss": 0.3808, + "step": 26638 + }, + { + "epoch": 1.48836494678325, + "grad_norm": 0.5838319659233093, + "learning_rate": 0.0002571156432093232, + "loss": 0.4014, + "step": 26639 + }, + { + "epoch": 1.488420817386932, + "grad_norm": 0.38977858424186707, + "learning_rate": 0.00025708762886597934, + "loss": 0.3937, + "step": 26640 + }, + { + "epoch": 1.4884766879906137, + "grad_norm": 0.5624439120292664, + "learning_rate": 0.0002570596145226356, + "loss": 0.4025, + "step": 26641 + }, + { + "epoch": 1.4885325585942957, + "grad_norm": 2.484994649887085, + "learning_rate": 0.0002570316001792918, + "loss": 0.3723, + "step": 26642 + }, + { + "epoch": 1.4885884291979774, + "grad_norm": 0.5403811931610107, + "learning_rate": 0.000257003585835948, + "loss": 0.3633, + "step": 26643 + }, + { + "epoch": 1.4886442998016594, + "grad_norm": 0.42583614587783813, + "learning_rate": 0.0002569755714926042, + "loss": 0.3623, + "step": 26644 + }, + { + "epoch": 1.4887001704053413, + "grad_norm": 0.5823318958282471, + "learning_rate": 0.0002569475571492604, + "loss": 0.4373, + "step": 26645 + }, + { + "epoch": 1.488756041009023, + "grad_norm": 0.4718870520591736, + "learning_rate": 0.0002569195428059166, + "loss": 0.3485, + "step": 26646 + }, + { + "epoch": 1.488811911612705, + "grad_norm": 1.7120457887649536, + "learning_rate": 0.00025689152846257283, + "loss": 0.6182, + "step": 26647 + }, + { + "epoch": 1.4888677822163867, + "grad_norm": 1.2921549081802368, + "learning_rate": 0.00025686351411922904, + "loss": 0.5285, + "step": 26648 + }, + { + "epoch": 1.4889236528200687, + "grad_norm": 0.4813506603240967, + "learning_rate": 0.0002568354997758853, + "loss": 0.423, + "step": 26649 + }, + { + "epoch": 1.4889795234237506, + "grad_norm": 0.4937189817428589, + "learning_rate": 0.00025680748543254145, + "loss": 0.4371, + "step": 26650 + }, + { + "epoch": 1.4890353940274323, + "grad_norm": 0.40537914633750916, + "learning_rate": 0.0002567794710891977, + "loss": 0.3735, + "step": 26651 + }, + { + "epoch": 1.4890912646311143, + "grad_norm": 0.3756500780582428, + "learning_rate": 0.00025675145674585386, + "loss": 0.4318, + "step": 26652 + }, + { + "epoch": 1.4891471352347962, + "grad_norm": 0.35722973942756653, + "learning_rate": 0.0002567234424025101, + "loss": 0.3738, + "step": 26653 + }, + { + "epoch": 1.489203005838478, + "grad_norm": 0.5713325142860413, + "learning_rate": 0.0002566954280591663, + "loss": 0.3961, + "step": 26654 + }, + { + "epoch": 1.48925887644216, + "grad_norm": 1.1735258102416992, + "learning_rate": 0.00025666741371582253, + "loss": 0.4373, + "step": 26655 + }, + { + "epoch": 1.4893147470458419, + "grad_norm": 0.7385091185569763, + "learning_rate": 0.0002566393993724787, + "loss": 0.3825, + "step": 26656 + }, + { + "epoch": 1.4893706176495236, + "grad_norm": 0.4710679054260254, + "learning_rate": 0.00025661138502913494, + "loss": 0.2717, + "step": 26657 + }, + { + "epoch": 1.4894264882532056, + "grad_norm": 0.7978706359863281, + "learning_rate": 0.00025658337068579115, + "loss": 0.437, + "step": 26658 + }, + { + "epoch": 1.4894823588568875, + "grad_norm": 0.9042708277702332, + "learning_rate": 0.00025655535634244736, + "loss": 0.3002, + "step": 26659 + }, + { + "epoch": 1.4895382294605692, + "grad_norm": 0.3737684190273285, + "learning_rate": 0.00025652734199910356, + "loss": 0.4182, + "step": 26660 + }, + { + "epoch": 1.4895941000642512, + "grad_norm": 0.8557690382003784, + "learning_rate": 0.0002564993276557597, + "loss": 0.4047, + "step": 26661 + }, + { + "epoch": 1.4896499706679331, + "grad_norm": 8.143767356872559, + "learning_rate": 0.000256471313312416, + "loss": 0.5156, + "step": 26662 + }, + { + "epoch": 1.4897058412716149, + "grad_norm": 0.5170104503631592, + "learning_rate": 0.0002564432989690721, + "loss": 0.4218, + "step": 26663 + }, + { + "epoch": 1.4897617118752968, + "grad_norm": 0.6713389158248901, + "learning_rate": 0.0002564152846257284, + "loss": 0.5607, + "step": 26664 + }, + { + "epoch": 1.4898175824789788, + "grad_norm": 0.5852422714233398, + "learning_rate": 0.0002563872702823846, + "loss": 0.5275, + "step": 26665 + }, + { + "epoch": 1.4898734530826605, + "grad_norm": 0.3137337267398834, + "learning_rate": 0.0002563592559390408, + "loss": 0.3583, + "step": 26666 + }, + { + "epoch": 1.4899293236863425, + "grad_norm": 0.4676213264465332, + "learning_rate": 0.000256331241595697, + "loss": 0.5068, + "step": 26667 + }, + { + "epoch": 1.4899851942900244, + "grad_norm": 0.6073673367500305, + "learning_rate": 0.0002563032272523532, + "loss": 0.411, + "step": 26668 + }, + { + "epoch": 1.4900410648937061, + "grad_norm": 1.0331838130950928, + "learning_rate": 0.0002562752129090094, + "loss": 0.3852, + "step": 26669 + }, + { + "epoch": 1.490096935497388, + "grad_norm": 0.4333665072917938, + "learning_rate": 0.0002562471985656656, + "loss": 0.3903, + "step": 26670 + }, + { + "epoch": 1.49015280610107, + "grad_norm": 2.6315090656280518, + "learning_rate": 0.0002562191842223218, + "loss": 0.5624, + "step": 26671 + }, + { + "epoch": 1.4902086767047518, + "grad_norm": 0.4048861265182495, + "learning_rate": 0.00025619116987897803, + "loss": 0.4053, + "step": 26672 + }, + { + "epoch": 1.4902645473084337, + "grad_norm": 0.45314767956733704, + "learning_rate": 0.00025616315553563424, + "loss": 0.4179, + "step": 26673 + }, + { + "epoch": 1.4903204179121157, + "grad_norm": 0.8396989107131958, + "learning_rate": 0.0002561351411922905, + "loss": 0.3629, + "step": 26674 + }, + { + "epoch": 1.4903762885157974, + "grad_norm": 0.4561412036418915, + "learning_rate": 0.00025610712684894665, + "loss": 0.424, + "step": 26675 + }, + { + "epoch": 1.4904321591194793, + "grad_norm": 0.5468929409980774, + "learning_rate": 0.0002560791125056029, + "loss": 0.3886, + "step": 26676 + }, + { + "epoch": 1.490488029723161, + "grad_norm": 0.5606470108032227, + "learning_rate": 0.00025605109816225906, + "loss": 0.4566, + "step": 26677 + }, + { + "epoch": 1.490543900326843, + "grad_norm": 0.828107476234436, + "learning_rate": 0.0002560230838189153, + "loss": 0.3968, + "step": 26678 + }, + { + "epoch": 1.490599770930525, + "grad_norm": 0.6531714797019958, + "learning_rate": 0.00025599506947557147, + "loss": 0.4841, + "step": 26679 + }, + { + "epoch": 1.4906556415342067, + "grad_norm": 0.528183102607727, + "learning_rate": 0.00025596705513222773, + "loss": 0.3273, + "step": 26680 + }, + { + "epoch": 1.4907115121378887, + "grad_norm": 0.546356201171875, + "learning_rate": 0.00025593904078888394, + "loss": 0.4263, + "step": 26681 + }, + { + "epoch": 1.4907673827415704, + "grad_norm": 0.33705422282218933, + "learning_rate": 0.0002559110264455401, + "loss": 0.2822, + "step": 26682 + }, + { + "epoch": 1.4908232533452523, + "grad_norm": 1.060929536819458, + "learning_rate": 0.00025588301210219635, + "loss": 0.4233, + "step": 26683 + }, + { + "epoch": 1.4908791239489343, + "grad_norm": 0.38586780428886414, + "learning_rate": 0.0002558549977588525, + "loss": 0.3762, + "step": 26684 + }, + { + "epoch": 1.490934994552616, + "grad_norm": 0.6183493733406067, + "learning_rate": 0.00025582698341550876, + "loss": 0.4608, + "step": 26685 + }, + { + "epoch": 1.490990865156298, + "grad_norm": 2.2866508960723877, + "learning_rate": 0.0002557989690721649, + "loss": 0.4533, + "step": 26686 + }, + { + "epoch": 1.49104673575998, + "grad_norm": 0.35418975353240967, + "learning_rate": 0.0002557709547288212, + "loss": 0.4031, + "step": 26687 + }, + { + "epoch": 1.4911026063636617, + "grad_norm": 0.3565312325954437, + "learning_rate": 0.0002557429403854773, + "loss": 0.4029, + "step": 26688 + }, + { + "epoch": 1.4911584769673436, + "grad_norm": 0.8709076642990112, + "learning_rate": 0.0002557149260421336, + "loss": 0.5098, + "step": 26689 + }, + { + "epoch": 1.4912143475710256, + "grad_norm": 0.48974838852882385, + "learning_rate": 0.0002556869116987898, + "loss": 0.448, + "step": 26690 + }, + { + "epoch": 1.4912702181747073, + "grad_norm": 0.7439846992492676, + "learning_rate": 0.000255658897355446, + "loss": 0.4395, + "step": 26691 + }, + { + "epoch": 1.4913260887783892, + "grad_norm": 0.36167165637016296, + "learning_rate": 0.0002556308830121022, + "loss": 0.4238, + "step": 26692 + }, + { + "epoch": 1.4913819593820712, + "grad_norm": 0.358607679605484, + "learning_rate": 0.0002556028686687584, + "loss": 0.4585, + "step": 26693 + }, + { + "epoch": 1.491437829985753, + "grad_norm": 1.8632802963256836, + "learning_rate": 0.0002555748543254146, + "loss": 0.3888, + "step": 26694 + }, + { + "epoch": 1.4914937005894349, + "grad_norm": 0.419323205947876, + "learning_rate": 0.0002555468399820708, + "loss": 0.2912, + "step": 26695 + }, + { + "epoch": 1.4915495711931168, + "grad_norm": 0.47683385014533997, + "learning_rate": 0.000255518825638727, + "loss": 0.3837, + "step": 26696 + }, + { + "epoch": 1.4916054417967985, + "grad_norm": 0.45728346705436707, + "learning_rate": 0.0002554908112953833, + "loss": 0.4633, + "step": 26697 + }, + { + "epoch": 1.4916613124004805, + "grad_norm": 0.5028613805770874, + "learning_rate": 0.00025546279695203944, + "loss": 0.3945, + "step": 26698 + }, + { + "epoch": 1.4917171830041625, + "grad_norm": 0.7220109701156616, + "learning_rate": 0.0002554347826086957, + "loss": 0.4015, + "step": 26699 + }, + { + "epoch": 1.4917730536078442, + "grad_norm": 0.422605037689209, + "learning_rate": 0.00025540676826535185, + "loss": 0.4297, + "step": 26700 + }, + { + "epoch": 1.4918289242115261, + "grad_norm": 0.47918936610221863, + "learning_rate": 0.0002553787539220081, + "loss": 0.5682, + "step": 26701 + }, + { + "epoch": 1.491884794815208, + "grad_norm": 0.34402933716773987, + "learning_rate": 0.00025535073957866426, + "loss": 0.3268, + "step": 26702 + }, + { + "epoch": 1.4919406654188898, + "grad_norm": 0.4105924069881439, + "learning_rate": 0.00025532272523532047, + "loss": 0.414, + "step": 26703 + }, + { + "epoch": 1.4919965360225718, + "grad_norm": 0.4572731554508209, + "learning_rate": 0.0002552947108919767, + "loss": 0.4168, + "step": 26704 + }, + { + "epoch": 1.4920524066262537, + "grad_norm": 0.40404167771339417, + "learning_rate": 0.0002552666965486329, + "loss": 0.479, + "step": 26705 + }, + { + "epoch": 1.4921082772299354, + "grad_norm": 1.7251135110855103, + "learning_rate": 0.00025523868220528914, + "loss": 0.3426, + "step": 26706 + }, + { + "epoch": 1.4921641478336174, + "grad_norm": 0.6116678714752197, + "learning_rate": 0.0002552106678619453, + "loss": 0.4334, + "step": 26707 + }, + { + "epoch": 1.4922200184372993, + "grad_norm": 5.577293872833252, + "learning_rate": 0.00025518265351860155, + "loss": 0.46, + "step": 26708 + }, + { + "epoch": 1.492275889040981, + "grad_norm": 0.39714348316192627, + "learning_rate": 0.0002551546391752577, + "loss": 0.4116, + "step": 26709 + }, + { + "epoch": 1.492331759644663, + "grad_norm": 0.3580404818058014, + "learning_rate": 0.00025512662483191396, + "loss": 0.4338, + "step": 26710 + }, + { + "epoch": 1.4923876302483448, + "grad_norm": 0.43159741163253784, + "learning_rate": 0.0002550986104885701, + "loss": 0.4557, + "step": 26711 + }, + { + "epoch": 1.4924435008520267, + "grad_norm": 0.5062522292137146, + "learning_rate": 0.0002550705961452264, + "loss": 0.5298, + "step": 26712 + }, + { + "epoch": 1.4924993714557087, + "grad_norm": 0.35188910365104675, + "learning_rate": 0.0002550425818018826, + "loss": 0.3069, + "step": 26713 + }, + { + "epoch": 1.4925552420593904, + "grad_norm": 1.3954689502716064, + "learning_rate": 0.0002550145674585388, + "loss": 0.3163, + "step": 26714 + }, + { + "epoch": 1.4926111126630723, + "grad_norm": 0.6140140295028687, + "learning_rate": 0.000254986553115195, + "loss": 0.3425, + "step": 26715 + }, + { + "epoch": 1.492666983266754, + "grad_norm": 0.3714824914932251, + "learning_rate": 0.0002549585387718512, + "loss": 0.4429, + "step": 26716 + }, + { + "epoch": 1.492722853870436, + "grad_norm": 0.34966781735420227, + "learning_rate": 0.0002549305244285074, + "loss": 0.3719, + "step": 26717 + }, + { + "epoch": 1.492778724474118, + "grad_norm": 1.0264346599578857, + "learning_rate": 0.0002549025100851636, + "loss": 0.3771, + "step": 26718 + }, + { + "epoch": 1.4928345950777997, + "grad_norm": 0.34758853912353516, + "learning_rate": 0.0002548744957418198, + "loss": 0.4023, + "step": 26719 + }, + { + "epoch": 1.4928904656814816, + "grad_norm": 0.5211666226387024, + "learning_rate": 0.0002548464813984761, + "loss": 0.4097, + "step": 26720 + }, + { + "epoch": 1.4929463362851636, + "grad_norm": 0.4293036460876465, + "learning_rate": 0.0002548184670551322, + "loss": 0.4073, + "step": 26721 + }, + { + "epoch": 1.4930022068888453, + "grad_norm": 0.4667908847332001, + "learning_rate": 0.0002547904527117885, + "loss": 0.5026, + "step": 26722 + }, + { + "epoch": 1.4930580774925273, + "grad_norm": 0.78385990858078, + "learning_rate": 0.00025476243836844464, + "loss": 0.3786, + "step": 26723 + }, + { + "epoch": 1.4931139480962092, + "grad_norm": 0.31666406989097595, + "learning_rate": 0.00025473442402510084, + "loss": 0.37, + "step": 26724 + }, + { + "epoch": 1.493169818699891, + "grad_norm": 0.43022093176841736, + "learning_rate": 0.00025470640968175705, + "loss": 0.3892, + "step": 26725 + }, + { + "epoch": 1.493225689303573, + "grad_norm": 0.5415105819702148, + "learning_rate": 0.00025467839533841325, + "loss": 0.4722, + "step": 26726 + }, + { + "epoch": 1.4932815599072549, + "grad_norm": 0.7569164037704468, + "learning_rate": 0.00025465038099506946, + "loss": 0.4229, + "step": 26727 + }, + { + "epoch": 1.4933374305109366, + "grad_norm": 0.3055840730667114, + "learning_rate": 0.00025462236665172567, + "loss": 0.3587, + "step": 26728 + }, + { + "epoch": 1.4933933011146185, + "grad_norm": 0.5724349021911621, + "learning_rate": 0.0002545943523083819, + "loss": 0.4385, + "step": 26729 + }, + { + "epoch": 1.4934491717183005, + "grad_norm": 1.492093563079834, + "learning_rate": 0.0002545663379650381, + "loss": 0.4554, + "step": 26730 + }, + { + "epoch": 1.4935050423219822, + "grad_norm": 0.42512914538383484, + "learning_rate": 0.00025453832362169434, + "loss": 0.4301, + "step": 26731 + }, + { + "epoch": 1.4935609129256642, + "grad_norm": 0.39698469638824463, + "learning_rate": 0.0002545103092783505, + "loss": 0.3163, + "step": 26732 + }, + { + "epoch": 1.4936167835293461, + "grad_norm": 0.5998199582099915, + "learning_rate": 0.00025448229493500675, + "loss": 0.4763, + "step": 26733 + }, + { + "epoch": 1.4936726541330279, + "grad_norm": 0.39718499779701233, + "learning_rate": 0.0002544542805916629, + "loss": 0.3809, + "step": 26734 + }, + { + "epoch": 1.4937285247367098, + "grad_norm": 0.5684711337089539, + "learning_rate": 0.00025442626624831916, + "loss": 0.4844, + "step": 26735 + }, + { + "epoch": 1.4937843953403918, + "grad_norm": 0.39866381883621216, + "learning_rate": 0.00025439825190497537, + "loss": 0.3469, + "step": 26736 + }, + { + "epoch": 1.4938402659440735, + "grad_norm": 0.3576541841030121, + "learning_rate": 0.00025437023756163157, + "loss": 0.4023, + "step": 26737 + }, + { + "epoch": 1.4938961365477554, + "grad_norm": 0.5050910711288452, + "learning_rate": 0.0002543422232182878, + "loss": 0.4346, + "step": 26738 + }, + { + "epoch": 1.4939520071514374, + "grad_norm": 0.7432433366775513, + "learning_rate": 0.000254314208874944, + "loss": 0.4393, + "step": 26739 + }, + { + "epoch": 1.4940078777551191, + "grad_norm": 0.6504354476928711, + "learning_rate": 0.0002542861945316002, + "loss": 0.4927, + "step": 26740 + }, + { + "epoch": 1.494063748358801, + "grad_norm": 0.5390623807907104, + "learning_rate": 0.0002542581801882564, + "loss": 0.3502, + "step": 26741 + }, + { + "epoch": 1.494119618962483, + "grad_norm": 0.3769565224647522, + "learning_rate": 0.0002542301658449126, + "loss": 0.4007, + "step": 26742 + }, + { + "epoch": 1.4941754895661647, + "grad_norm": 0.6334998607635498, + "learning_rate": 0.0002542021515015688, + "loss": 0.3504, + "step": 26743 + }, + { + "epoch": 1.4942313601698467, + "grad_norm": 0.4568713903427124, + "learning_rate": 0.000254174137158225, + "loss": 0.4547, + "step": 26744 + }, + { + "epoch": 1.4942872307735284, + "grad_norm": 0.5503087639808655, + "learning_rate": 0.0002541461228148812, + "loss": 0.466, + "step": 26745 + }, + { + "epoch": 1.4943431013772104, + "grad_norm": 0.4282681941986084, + "learning_rate": 0.0002541181084715374, + "loss": 0.4646, + "step": 26746 + }, + { + "epoch": 1.4943989719808923, + "grad_norm": 0.4194110631942749, + "learning_rate": 0.00025409009412819363, + "loss": 0.4551, + "step": 26747 + }, + { + "epoch": 1.494454842584574, + "grad_norm": 0.44805434346199036, + "learning_rate": 0.00025406207978484984, + "loss": 0.4482, + "step": 26748 + }, + { + "epoch": 1.494510713188256, + "grad_norm": 0.36196109652519226, + "learning_rate": 0.00025403406544150604, + "loss": 0.3249, + "step": 26749 + }, + { + "epoch": 1.4945665837919377, + "grad_norm": 0.9410011768341064, + "learning_rate": 0.00025400605109816225, + "loss": 0.3742, + "step": 26750 + }, + { + "epoch": 1.4946224543956197, + "grad_norm": 0.3696025609970093, + "learning_rate": 0.00025397803675481845, + "loss": 0.3721, + "step": 26751 + }, + { + "epoch": 1.4946783249993016, + "grad_norm": 0.3963746428489685, + "learning_rate": 0.0002539500224114747, + "loss": 0.4799, + "step": 26752 + }, + { + "epoch": 1.4947341956029834, + "grad_norm": 0.41466936469078064, + "learning_rate": 0.00025392200806813087, + "loss": 0.4641, + "step": 26753 + }, + { + "epoch": 1.4947900662066653, + "grad_norm": 0.4015507102012634, + "learning_rate": 0.0002538939937247871, + "loss": 0.4495, + "step": 26754 + }, + { + "epoch": 1.4948459368103473, + "grad_norm": 0.4969639182090759, + "learning_rate": 0.0002538659793814433, + "loss": 0.4899, + "step": 26755 + }, + { + "epoch": 1.494901807414029, + "grad_norm": 0.4021916687488556, + "learning_rate": 0.00025383796503809954, + "loss": 0.3805, + "step": 26756 + }, + { + "epoch": 1.494957678017711, + "grad_norm": 0.5541976690292358, + "learning_rate": 0.0002538099506947557, + "loss": 0.5128, + "step": 26757 + }, + { + "epoch": 1.495013548621393, + "grad_norm": 0.49642840027809143, + "learning_rate": 0.00025378193635141195, + "loss": 0.415, + "step": 26758 + }, + { + "epoch": 1.4950694192250746, + "grad_norm": 0.3424910008907318, + "learning_rate": 0.0002537539220080681, + "loss": 0.3106, + "step": 26759 + }, + { + "epoch": 1.4951252898287566, + "grad_norm": 1.2445027828216553, + "learning_rate": 0.00025372590766472436, + "loss": 0.3765, + "step": 26760 + }, + { + "epoch": 1.4951811604324385, + "grad_norm": 0.4387858510017395, + "learning_rate": 0.00025369789332138057, + "loss": 0.3725, + "step": 26761 + }, + { + "epoch": 1.4952370310361203, + "grad_norm": 0.3227299451828003, + "learning_rate": 0.00025366987897803677, + "loss": 0.3782, + "step": 26762 + }, + { + "epoch": 1.4952929016398022, + "grad_norm": 0.5885306596755981, + "learning_rate": 0.000253641864634693, + "loss": 0.5173, + "step": 26763 + }, + { + "epoch": 1.4953487722434842, + "grad_norm": 1.0874147415161133, + "learning_rate": 0.0002536138502913492, + "loss": 0.4519, + "step": 26764 + }, + { + "epoch": 1.495404642847166, + "grad_norm": 2.81321382522583, + "learning_rate": 0.0002535858359480054, + "loss": 0.3513, + "step": 26765 + }, + { + "epoch": 1.4954605134508479, + "grad_norm": 0.40452951192855835, + "learning_rate": 0.00025355782160466154, + "loss": 0.3436, + "step": 26766 + }, + { + "epoch": 1.4955163840545298, + "grad_norm": 0.4962192475795746, + "learning_rate": 0.0002535298072613178, + "loss": 0.3886, + "step": 26767 + }, + { + "epoch": 1.4955722546582115, + "grad_norm": 0.5819253325462341, + "learning_rate": 0.000253501792917974, + "loss": 0.4709, + "step": 26768 + }, + { + "epoch": 1.4956281252618935, + "grad_norm": 0.39016321301460266, + "learning_rate": 0.0002534737785746302, + "loss": 0.3103, + "step": 26769 + }, + { + "epoch": 1.4956839958655754, + "grad_norm": 0.44245100021362305, + "learning_rate": 0.0002534457642312864, + "loss": 0.399, + "step": 26770 + }, + { + "epoch": 1.4957398664692572, + "grad_norm": 0.46596240997314453, + "learning_rate": 0.0002534177498879426, + "loss": 0.5126, + "step": 26771 + }, + { + "epoch": 1.4957957370729391, + "grad_norm": 4.668961048126221, + "learning_rate": 0.00025338973554459883, + "loss": 0.3808, + "step": 26772 + }, + { + "epoch": 1.495851607676621, + "grad_norm": 0.64085453748703, + "learning_rate": 0.00025336172120125504, + "loss": 0.4706, + "step": 26773 + }, + { + "epoch": 1.4959074782803028, + "grad_norm": 0.403486043214798, + "learning_rate": 0.00025333370685791124, + "loss": 0.3412, + "step": 26774 + }, + { + "epoch": 1.4959633488839847, + "grad_norm": 0.8471749424934387, + "learning_rate": 0.0002533056925145675, + "loss": 0.3934, + "step": 26775 + }, + { + "epoch": 1.4960192194876667, + "grad_norm": 1.6908457279205322, + "learning_rate": 0.00025327767817122365, + "loss": 0.4437, + "step": 26776 + }, + { + "epoch": 1.4960750900913484, + "grad_norm": 0.41323354840278625, + "learning_rate": 0.0002532496638278799, + "loss": 0.3235, + "step": 26777 + }, + { + "epoch": 1.4961309606950304, + "grad_norm": 2.6207761764526367, + "learning_rate": 0.00025322164948453607, + "loss": 0.3371, + "step": 26778 + }, + { + "epoch": 1.496186831298712, + "grad_norm": 0.35324352979660034, + "learning_rate": 0.0002531936351411923, + "loss": 0.4586, + "step": 26779 + }, + { + "epoch": 1.496242701902394, + "grad_norm": 0.401948481798172, + "learning_rate": 0.0002531656207978485, + "loss": 0.3783, + "step": 26780 + }, + { + "epoch": 1.496298572506076, + "grad_norm": 0.6726385354995728, + "learning_rate": 0.00025313760645450474, + "loss": 0.42, + "step": 26781 + }, + { + "epoch": 1.4963544431097577, + "grad_norm": 0.3814324736595154, + "learning_rate": 0.0002531095921111609, + "loss": 0.3728, + "step": 26782 + }, + { + "epoch": 1.4964103137134397, + "grad_norm": 0.4794893264770508, + "learning_rate": 0.00025308157776781715, + "loss": 0.5035, + "step": 26783 + }, + { + "epoch": 1.4964661843171214, + "grad_norm": 8.668697357177734, + "learning_rate": 0.00025305356342447335, + "loss": 0.4336, + "step": 26784 + }, + { + "epoch": 1.4965220549208034, + "grad_norm": 0.7615892887115479, + "learning_rate": 0.00025302554908112956, + "loss": 0.386, + "step": 26785 + }, + { + "epoch": 1.4965779255244853, + "grad_norm": 0.8289125561714172, + "learning_rate": 0.00025299753473778577, + "loss": 0.3676, + "step": 26786 + }, + { + "epoch": 1.496633796128167, + "grad_norm": 0.35531213879585266, + "learning_rate": 0.0002529695203944419, + "loss": 0.4346, + "step": 26787 + }, + { + "epoch": 1.496689666731849, + "grad_norm": 0.6611014604568481, + "learning_rate": 0.0002529415060510982, + "loss": 0.361, + "step": 26788 + }, + { + "epoch": 1.496745537335531, + "grad_norm": 0.4272010028362274, + "learning_rate": 0.00025291349170775433, + "loss": 0.4411, + "step": 26789 + }, + { + "epoch": 1.4968014079392127, + "grad_norm": 0.42303383350372314, + "learning_rate": 0.0002528854773644106, + "loss": 0.3441, + "step": 26790 + }, + { + "epoch": 1.4968572785428946, + "grad_norm": 0.49185511469841003, + "learning_rate": 0.0002528574630210668, + "loss": 0.4526, + "step": 26791 + }, + { + "epoch": 1.4969131491465766, + "grad_norm": 0.4611146152019501, + "learning_rate": 0.000252829448677723, + "loss": 0.4626, + "step": 26792 + }, + { + "epoch": 1.4969690197502583, + "grad_norm": 1.0374372005462646, + "learning_rate": 0.0002528014343343792, + "loss": 0.8466, + "step": 26793 + }, + { + "epoch": 1.4970248903539403, + "grad_norm": 0.5631198287010193, + "learning_rate": 0.0002527734199910354, + "loss": 0.3772, + "step": 26794 + }, + { + "epoch": 1.4970807609576222, + "grad_norm": 0.5647053122520447, + "learning_rate": 0.0002527454056476916, + "loss": 0.4667, + "step": 26795 + }, + { + "epoch": 1.497136631561304, + "grad_norm": 0.4094533622264862, + "learning_rate": 0.0002527173913043478, + "loss": 0.4329, + "step": 26796 + }, + { + "epoch": 1.497192502164986, + "grad_norm": 0.4864872097969055, + "learning_rate": 0.00025268937696100403, + "loss": 0.3373, + "step": 26797 + }, + { + "epoch": 1.4972483727686678, + "grad_norm": 0.41223862767219543, + "learning_rate": 0.00025266136261766024, + "loss": 0.3646, + "step": 26798 + }, + { + "epoch": 1.4973042433723496, + "grad_norm": 0.9023059606552124, + "learning_rate": 0.00025263334827431644, + "loss": 0.3421, + "step": 26799 + }, + { + "epoch": 1.4973601139760315, + "grad_norm": 0.7244061827659607, + "learning_rate": 0.0002526053339309727, + "loss": 0.3669, + "step": 26800 + }, + { + "epoch": 1.4974159845797135, + "grad_norm": 0.9449537396430969, + "learning_rate": 0.00025257731958762885, + "loss": 0.3363, + "step": 26801 + }, + { + "epoch": 1.4974718551833952, + "grad_norm": 0.3927737772464752, + "learning_rate": 0.0002525493052442851, + "loss": 0.3327, + "step": 26802 + }, + { + "epoch": 1.4975277257870772, + "grad_norm": 0.49535098671913147, + "learning_rate": 0.00025252129090094126, + "loss": 0.3729, + "step": 26803 + }, + { + "epoch": 1.497583596390759, + "grad_norm": 2.454453468322754, + "learning_rate": 0.0002524932765575975, + "loss": 0.3695, + "step": 26804 + }, + { + "epoch": 1.4976394669944408, + "grad_norm": 0.5012763142585754, + "learning_rate": 0.0002524652622142537, + "loss": 0.4649, + "step": 26805 + }, + { + "epoch": 1.4976953375981228, + "grad_norm": 0.3799877166748047, + "learning_rate": 0.00025243724787090994, + "loss": 0.3685, + "step": 26806 + }, + { + "epoch": 1.4977512082018047, + "grad_norm": 0.4021853804588318, + "learning_rate": 0.00025240923352756614, + "loss": 0.3351, + "step": 26807 + }, + { + "epoch": 1.4978070788054865, + "grad_norm": 0.7069501280784607, + "learning_rate": 0.00025238121918422235, + "loss": 0.4327, + "step": 26808 + }, + { + "epoch": 1.4978629494091684, + "grad_norm": 0.43786948919296265, + "learning_rate": 0.00025235320484087855, + "loss": 0.3459, + "step": 26809 + }, + { + "epoch": 1.4979188200128504, + "grad_norm": 0.5495749711990356, + "learning_rate": 0.0002523251904975347, + "loss": 0.462, + "step": 26810 + }, + { + "epoch": 1.497974690616532, + "grad_norm": 1.305078148841858, + "learning_rate": 0.00025229717615419097, + "loss": 0.4393, + "step": 26811 + }, + { + "epoch": 1.498030561220214, + "grad_norm": 0.38759201765060425, + "learning_rate": 0.0002522691618108471, + "loss": 0.4909, + "step": 26812 + }, + { + "epoch": 1.4980864318238958, + "grad_norm": 0.48106691241264343, + "learning_rate": 0.0002522411474675034, + "loss": 0.3818, + "step": 26813 + }, + { + "epoch": 1.4981423024275777, + "grad_norm": 0.3979330360889435, + "learning_rate": 0.00025221313312415953, + "loss": 0.3688, + "step": 26814 + }, + { + "epoch": 1.4981981730312597, + "grad_norm": 0.6733072996139526, + "learning_rate": 0.0002521851187808158, + "loss": 0.4352, + "step": 26815 + }, + { + "epoch": 1.4982540436349414, + "grad_norm": 0.35582906007766724, + "learning_rate": 0.000252157104437472, + "loss": 0.3862, + "step": 26816 + }, + { + "epoch": 1.4983099142386234, + "grad_norm": 0.6817499995231628, + "learning_rate": 0.0002521290900941282, + "loss": 0.4514, + "step": 26817 + }, + { + "epoch": 1.498365784842305, + "grad_norm": 0.5793691873550415, + "learning_rate": 0.0002521010757507844, + "loss": 0.6408, + "step": 26818 + }, + { + "epoch": 1.498421655445987, + "grad_norm": 0.34764865040779114, + "learning_rate": 0.0002520730614074406, + "loss": 0.3863, + "step": 26819 + }, + { + "epoch": 1.498477526049669, + "grad_norm": 0.8499041795730591, + "learning_rate": 0.0002520450470640968, + "loss": 0.3578, + "step": 26820 + }, + { + "epoch": 1.4985333966533507, + "grad_norm": 0.6021517515182495, + "learning_rate": 0.000252017032720753, + "loss": 0.3947, + "step": 26821 + }, + { + "epoch": 1.4985892672570327, + "grad_norm": 0.36220452189445496, + "learning_rate": 0.00025198901837740923, + "loss": 0.3897, + "step": 26822 + }, + { + "epoch": 1.4986451378607146, + "grad_norm": 0.5120118856430054, + "learning_rate": 0.0002519610040340655, + "loss": 0.4381, + "step": 26823 + }, + { + "epoch": 1.4987010084643964, + "grad_norm": 0.45829781889915466, + "learning_rate": 0.00025193298969072164, + "loss": 0.4654, + "step": 26824 + }, + { + "epoch": 1.4987568790680783, + "grad_norm": 0.5812274813652039, + "learning_rate": 0.0002519049753473779, + "loss": 0.3835, + "step": 26825 + }, + { + "epoch": 1.4988127496717603, + "grad_norm": 0.7362874150276184, + "learning_rate": 0.00025187696100403405, + "loss": 0.4899, + "step": 26826 + }, + { + "epoch": 1.498868620275442, + "grad_norm": 0.5076290965080261, + "learning_rate": 0.0002518489466606903, + "loss": 0.3271, + "step": 26827 + }, + { + "epoch": 1.498924490879124, + "grad_norm": 0.5328342318534851, + "learning_rate": 0.00025182093231734646, + "loss": 0.3426, + "step": 26828 + }, + { + "epoch": 1.498980361482806, + "grad_norm": 0.6255783438682556, + "learning_rate": 0.0002517929179740027, + "loss": 0.5323, + "step": 26829 + }, + { + "epoch": 1.4990362320864876, + "grad_norm": 0.8801692128181458, + "learning_rate": 0.00025176490363065893, + "loss": 0.4405, + "step": 26830 + }, + { + "epoch": 1.4990921026901696, + "grad_norm": 1.0084162950515747, + "learning_rate": 0.0002517368892873151, + "loss": 0.3877, + "step": 26831 + }, + { + "epoch": 1.4991479732938515, + "grad_norm": 0.36970028281211853, + "learning_rate": 0.00025170887494397134, + "loss": 0.4445, + "step": 26832 + }, + { + "epoch": 1.4992038438975333, + "grad_norm": 0.4852658212184906, + "learning_rate": 0.0002516808606006275, + "loss": 0.4181, + "step": 26833 + }, + { + "epoch": 1.4992597145012152, + "grad_norm": 0.9572538137435913, + "learning_rate": 0.00025165284625728375, + "loss": 0.4245, + "step": 26834 + }, + { + "epoch": 1.4993155851048972, + "grad_norm": 0.5234407186508179, + "learning_rate": 0.0002516248319139399, + "loss": 0.4038, + "step": 26835 + }, + { + "epoch": 1.4993714557085789, + "grad_norm": 0.3817599415779114, + "learning_rate": 0.00025159681757059617, + "loss": 0.4439, + "step": 26836 + }, + { + "epoch": 1.4994273263122608, + "grad_norm": 0.3587530553340912, + "learning_rate": 0.0002515688032272523, + "loss": 0.3496, + "step": 26837 + }, + { + "epoch": 1.4994831969159428, + "grad_norm": 0.5755777955055237, + "learning_rate": 0.0002515407888839086, + "loss": 0.3892, + "step": 26838 + }, + { + "epoch": 1.4995390675196245, + "grad_norm": 0.9101813435554504, + "learning_rate": 0.0002515127745405648, + "loss": 0.5849, + "step": 26839 + }, + { + "epoch": 1.4995949381233065, + "grad_norm": 0.3551722466945648, + "learning_rate": 0.000251484760197221, + "loss": 0.432, + "step": 26840 + }, + { + "epoch": 1.4996508087269884, + "grad_norm": 0.9124963283538818, + "learning_rate": 0.0002514567458538772, + "loss": 0.2891, + "step": 26841 + }, + { + "epoch": 1.4997066793306701, + "grad_norm": 0.4420229494571686, + "learning_rate": 0.0002514287315105334, + "loss": 0.3695, + "step": 26842 + }, + { + "epoch": 1.499762549934352, + "grad_norm": 0.3830171227455139, + "learning_rate": 0.0002514007171671896, + "loss": 0.2587, + "step": 26843 + }, + { + "epoch": 1.499818420538034, + "grad_norm": 0.4684363305568695, + "learning_rate": 0.0002513727028238458, + "loss": 0.4485, + "step": 26844 + }, + { + "epoch": 1.4998742911417158, + "grad_norm": 0.8428595662117004, + "learning_rate": 0.000251344688480502, + "loss": 0.4122, + "step": 26845 + }, + { + "epoch": 1.4999301617453977, + "grad_norm": 0.7388940453529358, + "learning_rate": 0.0002513166741371583, + "loss": 0.3281, + "step": 26846 + }, + { + "epoch": 1.4999860323490795, + "grad_norm": 0.8338161110877991, + "learning_rate": 0.00025128865979381443, + "loss": 0.4723, + "step": 26847 + }, + { + "epoch": 1.5000419029527614, + "grad_norm": 1.7829195261001587, + "learning_rate": 0.0002512606454504707, + "loss": 0.4925, + "step": 26848 + }, + { + "epoch": 1.5000977735564431, + "grad_norm": 0.65754634141922, + "learning_rate": 0.00025123263110712684, + "loss": 0.5004, + "step": 26849 + }, + { + "epoch": 1.5001536441601253, + "grad_norm": 0.440282940864563, + "learning_rate": 0.0002512046167637831, + "loss": 0.4088, + "step": 26850 + }, + { + "epoch": 1.500209514763807, + "grad_norm": 0.4693745970726013, + "learning_rate": 0.00025117660242043925, + "loss": 0.3866, + "step": 26851 + }, + { + "epoch": 1.5002653853674888, + "grad_norm": 0.48971274495124817, + "learning_rate": 0.00025114858807709546, + "loss": 0.4773, + "step": 26852 + }, + { + "epoch": 1.5003212559711707, + "grad_norm": 0.32900750637054443, + "learning_rate": 0.00025112057373375166, + "loss": 0.3591, + "step": 26853 + }, + { + "epoch": 1.5003771265748527, + "grad_norm": 0.36227089166641235, + "learning_rate": 0.00025109255939040787, + "loss": 0.4491, + "step": 26854 + }, + { + "epoch": 1.5004329971785344, + "grad_norm": 0.828117847442627, + "learning_rate": 0.00025106454504706413, + "loss": 0.4352, + "step": 26855 + }, + { + "epoch": 1.5004888677822164, + "grad_norm": 3.953965425491333, + "learning_rate": 0.0002510365307037203, + "loss": 0.5904, + "step": 26856 + }, + { + "epoch": 1.5005447383858983, + "grad_norm": 1.2978026866912842, + "learning_rate": 0.00025100851636037654, + "loss": 0.3899, + "step": 26857 + }, + { + "epoch": 1.50060060898958, + "grad_norm": 0.39078405499458313, + "learning_rate": 0.0002509805020170327, + "loss": 0.3471, + "step": 26858 + }, + { + "epoch": 1.500656479593262, + "grad_norm": 0.9083840847015381, + "learning_rate": 0.00025095248767368895, + "loss": 0.4525, + "step": 26859 + }, + { + "epoch": 1.500712350196944, + "grad_norm": 0.6555735468864441, + "learning_rate": 0.0002509244733303451, + "loss": 0.4172, + "step": 26860 + }, + { + "epoch": 1.5007682208006257, + "grad_norm": 1.8371074199676514, + "learning_rate": 0.00025089645898700136, + "loss": 0.439, + "step": 26861 + }, + { + "epoch": 1.5008240914043076, + "grad_norm": 0.33928659558296204, + "learning_rate": 0.00025086844464365757, + "loss": 0.3526, + "step": 26862 + }, + { + "epoch": 1.5008799620079896, + "grad_norm": 0.3924770653247833, + "learning_rate": 0.0002508404303003138, + "loss": 0.3842, + "step": 26863 + }, + { + "epoch": 1.5009358326116713, + "grad_norm": 0.399757444858551, + "learning_rate": 0.00025081241595697, + "loss": 0.3906, + "step": 26864 + }, + { + "epoch": 1.5009917032153532, + "grad_norm": 0.39820748567581177, + "learning_rate": 0.0002507844016136262, + "loss": 0.4104, + "step": 26865 + }, + { + "epoch": 1.5010475738190352, + "grad_norm": 0.4327450692653656, + "learning_rate": 0.0002507563872702824, + "loss": 0.5127, + "step": 26866 + }, + { + "epoch": 1.501103444422717, + "grad_norm": 0.5312962532043457, + "learning_rate": 0.0002507283729269386, + "loss": 0.3507, + "step": 26867 + }, + { + "epoch": 1.5011593150263989, + "grad_norm": 0.3515913486480713, + "learning_rate": 0.0002507003585835948, + "loss": 0.4339, + "step": 26868 + }, + { + "epoch": 1.5012151856300808, + "grad_norm": 1.0790802240371704, + "learning_rate": 0.000250672344240251, + "loss": 0.3606, + "step": 26869 + }, + { + "epoch": 1.5012710562337626, + "grad_norm": 0.3548915982246399, + "learning_rate": 0.0002506443298969072, + "loss": 0.4885, + "step": 26870 + }, + { + "epoch": 1.5013269268374445, + "grad_norm": 0.3532642722129822, + "learning_rate": 0.0002506163155535635, + "loss": 0.3208, + "step": 26871 + }, + { + "epoch": 1.5013827974411265, + "grad_norm": 0.8566687703132629, + "learning_rate": 0.00025058830121021963, + "loss": 0.5829, + "step": 26872 + }, + { + "epoch": 1.5014386680448082, + "grad_norm": 1.3550080060958862, + "learning_rate": 0.00025056028686687583, + "loss": 0.4443, + "step": 26873 + }, + { + "epoch": 1.5014945386484901, + "grad_norm": 0.4764050841331482, + "learning_rate": 0.00025053227252353204, + "loss": 0.3453, + "step": 26874 + }, + { + "epoch": 1.501550409252172, + "grad_norm": 0.6770615577697754, + "learning_rate": 0.00025050425818018825, + "loss": 0.5042, + "step": 26875 + }, + { + "epoch": 1.5016062798558538, + "grad_norm": 0.5381422638893127, + "learning_rate": 0.00025047624383684445, + "loss": 0.6067, + "step": 26876 + }, + { + "epoch": 1.5016621504595355, + "grad_norm": 0.5162345767021179, + "learning_rate": 0.00025044822949350066, + "loss": 0.3932, + "step": 26877 + }, + { + "epoch": 1.5017180210632177, + "grad_norm": 0.6494522094726562, + "learning_rate": 0.0002504202151501569, + "loss": 0.4519, + "step": 26878 + }, + { + "epoch": 1.5017738916668995, + "grad_norm": 0.4049936830997467, + "learning_rate": 0.00025039220080681307, + "loss": 0.3784, + "step": 26879 + }, + { + "epoch": 1.5018297622705812, + "grad_norm": 0.5431889891624451, + "learning_rate": 0.00025036418646346933, + "loss": 0.4168, + "step": 26880 + }, + { + "epoch": 1.5018856328742634, + "grad_norm": 0.6850423216819763, + "learning_rate": 0.0002503361721201255, + "loss": 0.5469, + "step": 26881 + }, + { + "epoch": 1.501941503477945, + "grad_norm": 0.3912980258464813, + "learning_rate": 0.00025030815777678174, + "loss": 0.378, + "step": 26882 + }, + { + "epoch": 1.5019973740816268, + "grad_norm": 0.42680010199546814, + "learning_rate": 0.0002502801434334379, + "loss": 0.4623, + "step": 26883 + }, + { + "epoch": 1.502053244685309, + "grad_norm": 0.5227128863334656, + "learning_rate": 0.00025025212909009415, + "loss": 0.3218, + "step": 26884 + }, + { + "epoch": 1.5021091152889907, + "grad_norm": 0.4229763150215149, + "learning_rate": 0.0002502241147467503, + "loss": 0.3861, + "step": 26885 + }, + { + "epoch": 1.5021649858926724, + "grad_norm": 0.5803403258323669, + "learning_rate": 0.00025019610040340656, + "loss": 0.4646, + "step": 26886 + }, + { + "epoch": 1.5022208564963544, + "grad_norm": 0.5116835832595825, + "learning_rate": 0.00025016808606006277, + "loss": 0.5995, + "step": 26887 + }, + { + "epoch": 1.5022767271000363, + "grad_norm": 0.5051725506782532, + "learning_rate": 0.000250140071716719, + "loss": 0.4468, + "step": 26888 + }, + { + "epoch": 1.502332597703718, + "grad_norm": 0.4681498110294342, + "learning_rate": 0.0002501120573733752, + "loss": 0.4495, + "step": 26889 + }, + { + "epoch": 1.5023884683074, + "grad_norm": 0.39097756147384644, + "learning_rate": 0.0002500840430300314, + "loss": 0.4352, + "step": 26890 + }, + { + "epoch": 1.502444338911082, + "grad_norm": 0.3767452538013458, + "learning_rate": 0.0002500560286866876, + "loss": 0.4815, + "step": 26891 + }, + { + "epoch": 1.5025002095147637, + "grad_norm": 0.4612888991832733, + "learning_rate": 0.0002500280143433438, + "loss": 0.3887, + "step": 26892 + }, + { + "epoch": 1.5025560801184457, + "grad_norm": 0.31755709648132324, + "learning_rate": 0.00025, + "loss": 0.3199, + "step": 26893 + }, + { + "epoch": 1.5026119507221276, + "grad_norm": 0.6343958973884583, + "learning_rate": 0.0002499719856566562, + "loss": 0.6962, + "step": 26894 + }, + { + "epoch": 1.5026678213258093, + "grad_norm": 0.3070787191390991, + "learning_rate": 0.0002499439713133124, + "loss": 0.3675, + "step": 26895 + }, + { + "epoch": 1.5027236919294913, + "grad_norm": 0.5094442367553711, + "learning_rate": 0.0002499159569699686, + "loss": 0.4746, + "step": 26896 + }, + { + "epoch": 1.5027795625331732, + "grad_norm": 0.5122941732406616, + "learning_rate": 0.00024988794262662483, + "loss": 0.4509, + "step": 26897 + }, + { + "epoch": 1.502835433136855, + "grad_norm": 0.5493853688240051, + "learning_rate": 0.00024985992828328103, + "loss": 0.3631, + "step": 26898 + }, + { + "epoch": 1.502891303740537, + "grad_norm": 0.39324238896369934, + "learning_rate": 0.00024983191393993724, + "loss": 0.3644, + "step": 26899 + }, + { + "epoch": 1.5029471743442189, + "grad_norm": 0.474071204662323, + "learning_rate": 0.00024980389959659345, + "loss": 0.4807, + "step": 26900 + }, + { + "epoch": 1.5030030449479006, + "grad_norm": 0.39632388949394226, + "learning_rate": 0.0002497758852532497, + "loss": 0.4163, + "step": 26901 + }, + { + "epoch": 1.5030589155515826, + "grad_norm": 3.1251676082611084, + "learning_rate": 0.0002497478709099059, + "loss": 0.5174, + "step": 26902 + }, + { + "epoch": 1.5031147861552645, + "grad_norm": 0.6689712405204773, + "learning_rate": 0.0002497198565665621, + "loss": 0.3444, + "step": 26903 + }, + { + "epoch": 1.5031706567589462, + "grad_norm": 0.430950790643692, + "learning_rate": 0.00024969184222321827, + "loss": 0.4505, + "step": 26904 + }, + { + "epoch": 1.5032265273626282, + "grad_norm": 7.625525951385498, + "learning_rate": 0.0002496638278798745, + "loss": 0.4471, + "step": 26905 + }, + { + "epoch": 1.5032823979663101, + "grad_norm": 0.30869778990745544, + "learning_rate": 0.0002496358135365307, + "loss": 0.5584, + "step": 26906 + }, + { + "epoch": 1.5033382685699919, + "grad_norm": 0.4613724648952484, + "learning_rate": 0.0002496077991931869, + "loss": 0.4157, + "step": 26907 + }, + { + "epoch": 1.5033941391736738, + "grad_norm": 0.4255533814430237, + "learning_rate": 0.0002495797848498431, + "loss": 0.3703, + "step": 26908 + }, + { + "epoch": 1.5034500097773558, + "grad_norm": 0.9760921597480774, + "learning_rate": 0.00024955177050649935, + "loss": 0.3071, + "step": 26909 + }, + { + "epoch": 1.5035058803810375, + "grad_norm": 0.3604421615600586, + "learning_rate": 0.00024952375616315556, + "loss": 0.4758, + "step": 26910 + }, + { + "epoch": 1.5035617509847192, + "grad_norm": 1.3267323970794678, + "learning_rate": 0.00024949574181981176, + "loss": 0.5525, + "step": 26911 + }, + { + "epoch": 1.5036176215884014, + "grad_norm": 0.38306695222854614, + "learning_rate": 0.00024946772747646797, + "loss": 0.404, + "step": 26912 + }, + { + "epoch": 1.5036734921920831, + "grad_norm": 0.40343818068504333, + "learning_rate": 0.0002494397131331242, + "loss": 0.4328, + "step": 26913 + }, + { + "epoch": 1.5037293627957649, + "grad_norm": 0.3909597098827362, + "learning_rate": 0.0002494116987897804, + "loss": 0.3293, + "step": 26914 + }, + { + "epoch": 1.503785233399447, + "grad_norm": 1.4953429698944092, + "learning_rate": 0.0002493836844464366, + "loss": 0.3966, + "step": 26915 + }, + { + "epoch": 1.5038411040031288, + "grad_norm": 0.44817689061164856, + "learning_rate": 0.0002493556701030928, + "loss": 0.4685, + "step": 26916 + }, + { + "epoch": 1.5038969746068105, + "grad_norm": 0.3953607678413391, + "learning_rate": 0.000249327655759749, + "loss": 0.3893, + "step": 26917 + }, + { + "epoch": 1.5039528452104927, + "grad_norm": 0.3193623721599579, + "learning_rate": 0.0002492996414164052, + "loss": 0.3711, + "step": 26918 + }, + { + "epoch": 1.5040087158141744, + "grad_norm": 0.2646198570728302, + "learning_rate": 0.0002492716270730614, + "loss": 0.381, + "step": 26919 + }, + { + "epoch": 1.5040645864178561, + "grad_norm": 0.3984130620956421, + "learning_rate": 0.0002492436127297176, + "loss": 0.4821, + "step": 26920 + }, + { + "epoch": 1.504120457021538, + "grad_norm": 0.37887921929359436, + "learning_rate": 0.0002492155983863738, + "loss": 0.4007, + "step": 26921 + }, + { + "epoch": 1.50417632762522, + "grad_norm": 0.5033379793167114, + "learning_rate": 0.00024918758404303003, + "loss": 0.6242, + "step": 26922 + }, + { + "epoch": 1.5042321982289018, + "grad_norm": 0.38671520352363586, + "learning_rate": 0.00024915956969968623, + "loss": 0.4636, + "step": 26923 + }, + { + "epoch": 1.5042880688325837, + "grad_norm": 0.3848573863506317, + "learning_rate": 0.00024913155535634244, + "loss": 0.4916, + "step": 26924 + }, + { + "epoch": 1.5043439394362657, + "grad_norm": 0.7977062463760376, + "learning_rate": 0.00024910354101299865, + "loss": 0.5593, + "step": 26925 + }, + { + "epoch": 1.5043998100399474, + "grad_norm": 0.41417357325553894, + "learning_rate": 0.00024907552666965485, + "loss": 0.3403, + "step": 26926 + }, + { + "epoch": 1.5044556806436293, + "grad_norm": 0.3602253496646881, + "learning_rate": 0.00024904751232631106, + "loss": 0.436, + "step": 26927 + }, + { + "epoch": 1.5045115512473113, + "grad_norm": 0.3580031991004944, + "learning_rate": 0.00024901949798296726, + "loss": 0.3891, + "step": 26928 + }, + { + "epoch": 1.504567421850993, + "grad_norm": 0.3379653990268707, + "learning_rate": 0.00024899148363962347, + "loss": 0.4007, + "step": 26929 + }, + { + "epoch": 1.504623292454675, + "grad_norm": 2.0517921447753906, + "learning_rate": 0.0002489634692962797, + "loss": 0.4217, + "step": 26930 + }, + { + "epoch": 1.504679163058357, + "grad_norm": 0.39091917872428894, + "learning_rate": 0.0002489354549529359, + "loss": 0.4703, + "step": 26931 + }, + { + "epoch": 1.5047350336620386, + "grad_norm": 0.4265705943107605, + "learning_rate": 0.0002489074406095921, + "loss": 0.4426, + "step": 26932 + }, + { + "epoch": 1.5047909042657206, + "grad_norm": 0.38513970375061035, + "learning_rate": 0.00024887942626624835, + "loss": 0.4087, + "step": 26933 + }, + { + "epoch": 1.5048467748694025, + "grad_norm": 1.4046039581298828, + "learning_rate": 0.00024885141192290455, + "loss": 0.5384, + "step": 26934 + }, + { + "epoch": 1.5049026454730843, + "grad_norm": 0.32906007766723633, + "learning_rate": 0.00024882339757956076, + "loss": 0.3146, + "step": 26935 + }, + { + "epoch": 1.5049585160767662, + "grad_norm": 0.5282392501831055, + "learning_rate": 0.00024879538323621696, + "loss": 0.4855, + "step": 26936 + }, + { + "epoch": 1.5050143866804482, + "grad_norm": 0.5737109184265137, + "learning_rate": 0.00024876736889287317, + "loss": 0.4169, + "step": 26937 + }, + { + "epoch": 1.50507025728413, + "grad_norm": 0.32580500841140747, + "learning_rate": 0.0002487393545495294, + "loss": 0.4426, + "step": 26938 + }, + { + "epoch": 1.5051261278878119, + "grad_norm": 9.303754806518555, + "learning_rate": 0.0002487113402061856, + "loss": 0.4022, + "step": 26939 + }, + { + "epoch": 1.5051819984914938, + "grad_norm": 0.5449299216270447, + "learning_rate": 0.0002486833258628418, + "loss": 0.4825, + "step": 26940 + }, + { + "epoch": 1.5052378690951755, + "grad_norm": 0.40863800048828125, + "learning_rate": 0.000248655311519498, + "loss": 0.4095, + "step": 26941 + }, + { + "epoch": 1.5052937396988575, + "grad_norm": 0.39164960384368896, + "learning_rate": 0.0002486272971761542, + "loss": 0.3985, + "step": 26942 + }, + { + "epoch": 1.5053496103025394, + "grad_norm": 9.233859062194824, + "learning_rate": 0.0002485992828328104, + "loss": 0.3981, + "step": 26943 + }, + { + "epoch": 1.5054054809062212, + "grad_norm": 0.5761954188346863, + "learning_rate": 0.0002485712684894666, + "loss": 0.3183, + "step": 26944 + }, + { + "epoch": 1.505461351509903, + "grad_norm": 0.28006941080093384, + "learning_rate": 0.0002485432541461228, + "loss": 0.3038, + "step": 26945 + }, + { + "epoch": 1.505517222113585, + "grad_norm": 0.4060983657836914, + "learning_rate": 0.000248515239802779, + "loss": 0.3852, + "step": 26946 + }, + { + "epoch": 1.5055730927172668, + "grad_norm": 1.0632413625717163, + "learning_rate": 0.00024848722545943523, + "loss": 0.3354, + "step": 26947 + }, + { + "epoch": 1.5056289633209485, + "grad_norm": 0.42573145031929016, + "learning_rate": 0.00024845921111609143, + "loss": 0.3652, + "step": 26948 + }, + { + "epoch": 1.5056848339246307, + "grad_norm": 0.45780324935913086, + "learning_rate": 0.00024843119677274764, + "loss": 0.4778, + "step": 26949 + }, + { + "epoch": 1.5057407045283124, + "grad_norm": 0.8628089427947998, + "learning_rate": 0.00024840318242940385, + "loss": 0.5412, + "step": 26950 + }, + { + "epoch": 1.5057965751319942, + "grad_norm": 0.39254772663116455, + "learning_rate": 0.00024837516808606005, + "loss": 0.3675, + "step": 26951 + }, + { + "epoch": 1.5058524457356763, + "grad_norm": 0.37553566694259644, + "learning_rate": 0.00024834715374271626, + "loss": 0.3435, + "step": 26952 + }, + { + "epoch": 1.505908316339358, + "grad_norm": 0.828495979309082, + "learning_rate": 0.00024831913939937246, + "loss": 0.3752, + "step": 26953 + }, + { + "epoch": 1.5059641869430398, + "grad_norm": 0.38802415132522583, + "learning_rate": 0.00024829112505602867, + "loss": 0.4153, + "step": 26954 + }, + { + "epoch": 1.5060200575467217, + "grad_norm": 0.6248673796653748, + "learning_rate": 0.0002482631107126849, + "loss": 0.3764, + "step": 26955 + }, + { + "epoch": 1.5060759281504037, + "grad_norm": 0.3504869043827057, + "learning_rate": 0.0002482350963693411, + "loss": 0.3955, + "step": 26956 + }, + { + "epoch": 1.5061317987540854, + "grad_norm": 0.3283778429031372, + "learning_rate": 0.00024820708202599734, + "loss": 0.374, + "step": 26957 + }, + { + "epoch": 1.5061876693577674, + "grad_norm": 0.41627317667007446, + "learning_rate": 0.00024817906768265355, + "loss": 0.4033, + "step": 26958 + }, + { + "epoch": 1.5062435399614493, + "grad_norm": 0.4031420052051544, + "learning_rate": 0.00024815105333930975, + "loss": 0.4152, + "step": 26959 + }, + { + "epoch": 1.506299410565131, + "grad_norm": 0.3391091525554657, + "learning_rate": 0.00024812303899596596, + "loss": 0.319, + "step": 26960 + }, + { + "epoch": 1.506355281168813, + "grad_norm": 0.8430882096290588, + "learning_rate": 0.00024809502465262216, + "loss": 0.5805, + "step": 26961 + }, + { + "epoch": 1.506411151772495, + "grad_norm": 1.7542567253112793, + "learning_rate": 0.00024806701030927837, + "loss": 0.4738, + "step": 26962 + }, + { + "epoch": 1.5064670223761767, + "grad_norm": 0.42080286145210266, + "learning_rate": 0.0002480389959659346, + "loss": 0.4069, + "step": 26963 + }, + { + "epoch": 1.5065228929798586, + "grad_norm": 0.3703868091106415, + "learning_rate": 0.0002480109816225908, + "loss": 0.3831, + "step": 26964 + }, + { + "epoch": 1.5065787635835406, + "grad_norm": 0.3433639407157898, + "learning_rate": 0.000247982967279247, + "loss": 0.4035, + "step": 26965 + }, + { + "epoch": 1.5066346341872223, + "grad_norm": 0.4285292327404022, + "learning_rate": 0.0002479549529359032, + "loss": 0.367, + "step": 26966 + }, + { + "epoch": 1.5066905047909043, + "grad_norm": 0.42199525237083435, + "learning_rate": 0.0002479269385925594, + "loss": 0.3483, + "step": 26967 + }, + { + "epoch": 1.5067463753945862, + "grad_norm": 0.5355505347251892, + "learning_rate": 0.0002478989242492156, + "loss": 0.407, + "step": 26968 + }, + { + "epoch": 1.506802245998268, + "grad_norm": 0.6926929950714111, + "learning_rate": 0.0002478709099058718, + "loss": 0.3507, + "step": 26969 + }, + { + "epoch": 1.50685811660195, + "grad_norm": 0.44767507910728455, + "learning_rate": 0.000247842895562528, + "loss": 0.494, + "step": 26970 + }, + { + "epoch": 1.5069139872056319, + "grad_norm": 0.4231330454349518, + "learning_rate": 0.0002478148812191842, + "loss": 0.3197, + "step": 26971 + }, + { + "epoch": 1.5069698578093136, + "grad_norm": 0.4554014503955841, + "learning_rate": 0.00024778686687584043, + "loss": 0.4123, + "step": 26972 + }, + { + "epoch": 1.5070257284129955, + "grad_norm": 0.38721901178359985, + "learning_rate": 0.00024775885253249663, + "loss": 0.391, + "step": 26973 + }, + { + "epoch": 1.5070815990166775, + "grad_norm": 0.41614896059036255, + "learning_rate": 0.00024773083818915284, + "loss": 0.3161, + "step": 26974 + }, + { + "epoch": 1.5071374696203592, + "grad_norm": 0.44681647419929504, + "learning_rate": 0.00024770282384580904, + "loss": 0.4185, + "step": 26975 + }, + { + "epoch": 1.5071933402240412, + "grad_norm": 0.36234691739082336, + "learning_rate": 0.00024767480950246525, + "loss": 0.4547, + "step": 26976 + }, + { + "epoch": 1.5072492108277231, + "grad_norm": 0.5695242285728455, + "learning_rate": 0.00024764679515912146, + "loss": 0.3728, + "step": 26977 + }, + { + "epoch": 1.5073050814314048, + "grad_norm": 0.3392972946166992, + "learning_rate": 0.00024761878081577766, + "loss": 0.4141, + "step": 26978 + }, + { + "epoch": 1.5073609520350866, + "grad_norm": 0.45705336332321167, + "learning_rate": 0.00024759076647243387, + "loss": 0.3686, + "step": 26979 + }, + { + "epoch": 1.5074168226387687, + "grad_norm": 0.3834809362888336, + "learning_rate": 0.00024756275212909013, + "loss": 0.5109, + "step": 26980 + }, + { + "epoch": 1.5074726932424505, + "grad_norm": 1.1632710695266724, + "learning_rate": 0.00024753473778574633, + "loss": 0.3162, + "step": 26981 + }, + { + "epoch": 1.5075285638461322, + "grad_norm": 0.6275104880332947, + "learning_rate": 0.00024750672344240254, + "loss": 0.3423, + "step": 26982 + }, + { + "epoch": 1.5075844344498144, + "grad_norm": 10.467479705810547, + "learning_rate": 0.00024747870909905875, + "loss": 0.2626, + "step": 26983 + }, + { + "epoch": 1.507640305053496, + "grad_norm": 0.4909029006958008, + "learning_rate": 0.00024745069475571495, + "loss": 0.4173, + "step": 26984 + }, + { + "epoch": 1.5076961756571778, + "grad_norm": 0.2817886471748352, + "learning_rate": 0.00024742268041237116, + "loss": 0.3217, + "step": 26985 + }, + { + "epoch": 1.50775204626086, + "grad_norm": 0.8009860515594482, + "learning_rate": 0.00024739466606902736, + "loss": 0.4839, + "step": 26986 + }, + { + "epoch": 1.5078079168645417, + "grad_norm": 0.3429698646068573, + "learning_rate": 0.00024736665172568357, + "loss": 0.4227, + "step": 26987 + }, + { + "epoch": 1.5078637874682235, + "grad_norm": 0.3478972911834717, + "learning_rate": 0.0002473386373823398, + "loss": 0.3308, + "step": 26988 + }, + { + "epoch": 1.5079196580719054, + "grad_norm": 0.6188536286354065, + "learning_rate": 0.000247310623038996, + "loss": 0.4783, + "step": 26989 + }, + { + "epoch": 1.5079755286755874, + "grad_norm": 0.49401384592056274, + "learning_rate": 0.0002472826086956522, + "loss": 0.4708, + "step": 26990 + }, + { + "epoch": 1.508031399279269, + "grad_norm": 1.733994722366333, + "learning_rate": 0.0002472545943523084, + "loss": 0.4298, + "step": 26991 + }, + { + "epoch": 1.508087269882951, + "grad_norm": 0.38596290349960327, + "learning_rate": 0.0002472265800089646, + "loss": 0.3382, + "step": 26992 + }, + { + "epoch": 1.508143140486633, + "grad_norm": 0.41127198934555054, + "learning_rate": 0.0002471985656656208, + "loss": 0.5046, + "step": 26993 + }, + { + "epoch": 1.5081990110903147, + "grad_norm": 0.3501875698566437, + "learning_rate": 0.000247170551322277, + "loss": 0.34, + "step": 26994 + }, + { + "epoch": 1.5082548816939967, + "grad_norm": 0.41806313395500183, + "learning_rate": 0.0002471425369789332, + "loss": 0.3109, + "step": 26995 + }, + { + "epoch": 1.5083107522976786, + "grad_norm": 0.46726787090301514, + "learning_rate": 0.0002471145226355894, + "loss": 0.455, + "step": 26996 + }, + { + "epoch": 1.5083666229013604, + "grad_norm": 1.07365083694458, + "learning_rate": 0.0002470865082922456, + "loss": 0.3959, + "step": 26997 + }, + { + "epoch": 1.5084224935050423, + "grad_norm": 0.4257749617099762, + "learning_rate": 0.00024705849394890183, + "loss": 0.4144, + "step": 26998 + }, + { + "epoch": 1.5084783641087243, + "grad_norm": 6.374155521392822, + "learning_rate": 0.00024703047960555804, + "loss": 0.4216, + "step": 26999 + }, + { + "epoch": 1.508534234712406, + "grad_norm": 0.5878410339355469, + "learning_rate": 0.00024700246526221424, + "loss": 0.51, + "step": 27000 + }, + { + "epoch": 1.508534234712406, + "eval_cer": 0.0853054455391531, + "eval_loss": 0.3195560574531555, + "eval_runtime": 56.4682, + "eval_samples_per_second": 80.364, + "eval_steps_per_second": 5.029, + "eval_wer": 0.33929219854194015, + "step": 27000 + }, + { + "epoch": 1.508590105316088, + "grad_norm": 0.4482194185256958, + "learning_rate": 0.00024697445091887045, + "loss": 0.4449, + "step": 27001 + }, + { + "epoch": 1.50864597591977, + "grad_norm": 0.3808304965496063, + "learning_rate": 0.00024694643657552666, + "loss": 0.3732, + "step": 27002 + }, + { + "epoch": 1.5087018465234516, + "grad_norm": 0.5555459856987, + "learning_rate": 0.00024691842223218286, + "loss": 0.4642, + "step": 27003 + }, + { + "epoch": 1.5087577171271336, + "grad_norm": 0.5625905990600586, + "learning_rate": 0.0002468904078888391, + "loss": 0.4796, + "step": 27004 + }, + { + "epoch": 1.5088135877308155, + "grad_norm": 0.4406568706035614, + "learning_rate": 0.00024686239354549533, + "loss": 0.3675, + "step": 27005 + }, + { + "epoch": 1.5088694583344973, + "grad_norm": 0.44478192925453186, + "learning_rate": 0.00024683437920215153, + "loss": 0.35, + "step": 27006 + }, + { + "epoch": 1.5089253289381792, + "grad_norm": 1.5182297229766846, + "learning_rate": 0.00024680636485880774, + "loss": 0.6306, + "step": 27007 + }, + { + "epoch": 1.5089811995418612, + "grad_norm": 0.5168110132217407, + "learning_rate": 0.00024677835051546395, + "loss": 0.4105, + "step": 27008 + }, + { + "epoch": 1.509037070145543, + "grad_norm": 0.4453426003456116, + "learning_rate": 0.00024675033617212015, + "loss": 0.2764, + "step": 27009 + }, + { + "epoch": 1.5090929407492248, + "grad_norm": 0.6154159903526306, + "learning_rate": 0.0002467223218287763, + "loss": 0.3607, + "step": 27010 + }, + { + "epoch": 1.5091488113529068, + "grad_norm": 0.6085834503173828, + "learning_rate": 0.0002466943074854325, + "loss": 0.3422, + "step": 27011 + }, + { + "epoch": 1.5092046819565885, + "grad_norm": 0.48112109303474426, + "learning_rate": 0.00024666629314208877, + "loss": 0.4371, + "step": 27012 + }, + { + "epoch": 1.5092605525602703, + "grad_norm": 0.32010573148727417, + "learning_rate": 0.000246638278798745, + "loss": 0.3359, + "step": 27013 + }, + { + "epoch": 1.5093164231639524, + "grad_norm": 0.6963441967964172, + "learning_rate": 0.0002466102644554012, + "loss": 0.4529, + "step": 27014 + }, + { + "epoch": 1.5093722937676342, + "grad_norm": 0.3968566358089447, + "learning_rate": 0.0002465822501120574, + "loss": 0.4598, + "step": 27015 + }, + { + "epoch": 1.5094281643713159, + "grad_norm": 0.6322019100189209, + "learning_rate": 0.0002465542357687136, + "loss": 0.5339, + "step": 27016 + }, + { + "epoch": 1.509484034974998, + "grad_norm": 0.5965427160263062, + "learning_rate": 0.0002465262214253698, + "loss": 0.4562, + "step": 27017 + }, + { + "epoch": 1.5095399055786798, + "grad_norm": 0.9315035939216614, + "learning_rate": 0.000246498207082026, + "loss": 0.3508, + "step": 27018 + }, + { + "epoch": 1.5095957761823615, + "grad_norm": 0.4965234100818634, + "learning_rate": 0.0002464701927386822, + "loss": 0.3678, + "step": 27019 + }, + { + "epoch": 1.5096516467860437, + "grad_norm": 0.43672096729278564, + "learning_rate": 0.0002464421783953384, + "loss": 0.3753, + "step": 27020 + }, + { + "epoch": 1.5097075173897254, + "grad_norm": 0.3645946681499481, + "learning_rate": 0.0002464141640519946, + "loss": 0.3911, + "step": 27021 + }, + { + "epoch": 1.5097633879934071, + "grad_norm": 0.38443121314048767, + "learning_rate": 0.0002463861497086508, + "loss": 0.4734, + "step": 27022 + }, + { + "epoch": 1.509819258597089, + "grad_norm": 0.7099629044532776, + "learning_rate": 0.00024635813536530703, + "loss": 0.4231, + "step": 27023 + }, + { + "epoch": 1.509875129200771, + "grad_norm": 1.5020612478256226, + "learning_rate": 0.00024633012102196324, + "loss": 0.3919, + "step": 27024 + }, + { + "epoch": 1.5099309998044528, + "grad_norm": 0.7240928411483765, + "learning_rate": 0.00024630210667861944, + "loss": 0.5072, + "step": 27025 + }, + { + "epoch": 1.5099868704081347, + "grad_norm": 0.41580912470817566, + "learning_rate": 0.00024627409233527565, + "loss": 0.3844, + "step": 27026 + }, + { + "epoch": 1.5100427410118167, + "grad_norm": 0.34367209672927856, + "learning_rate": 0.0002462460779919319, + "loss": 0.4039, + "step": 27027 + }, + { + "epoch": 1.5100986116154984, + "grad_norm": 0.524825394153595, + "learning_rate": 0.0002462180636485881, + "loss": 0.4139, + "step": 27028 + }, + { + "epoch": 1.5101544822191804, + "grad_norm": 0.5175501704216003, + "learning_rate": 0.0002461900493052443, + "loss": 0.338, + "step": 27029 + }, + { + "epoch": 1.5102103528228623, + "grad_norm": 0.4177793860435486, + "learning_rate": 0.00024616203496190053, + "loss": 0.3938, + "step": 27030 + }, + { + "epoch": 1.510266223426544, + "grad_norm": 0.37870556116104126, + "learning_rate": 0.00024613402061855673, + "loss": 0.4746, + "step": 27031 + }, + { + "epoch": 1.510322094030226, + "grad_norm": 1.5935587882995605, + "learning_rate": 0.0002461060062752129, + "loss": 0.4321, + "step": 27032 + }, + { + "epoch": 1.510377964633908, + "grad_norm": 0.36850541830062866, + "learning_rate": 0.0002460779919318691, + "loss": 0.543, + "step": 27033 + }, + { + "epoch": 1.5104338352375897, + "grad_norm": 1.2693642377853394, + "learning_rate": 0.0002460499775885253, + "loss": 0.4053, + "step": 27034 + }, + { + "epoch": 1.5104897058412716, + "grad_norm": 0.29438140988349915, + "learning_rate": 0.00024602196324518156, + "loss": 0.4025, + "step": 27035 + }, + { + "epoch": 1.5105455764449536, + "grad_norm": 0.7878247499465942, + "learning_rate": 0.00024599394890183776, + "loss": 0.4068, + "step": 27036 + }, + { + "epoch": 1.5106014470486353, + "grad_norm": 0.3499389886856079, + "learning_rate": 0.00024596593455849397, + "loss": 0.4896, + "step": 27037 + }, + { + "epoch": 1.5106573176523173, + "grad_norm": 0.43918174505233765, + "learning_rate": 0.0002459379202151502, + "loss": 0.5224, + "step": 27038 + }, + { + "epoch": 1.5107131882559992, + "grad_norm": 0.39809513092041016, + "learning_rate": 0.0002459099058718064, + "loss": 0.3748, + "step": 27039 + }, + { + "epoch": 1.510769058859681, + "grad_norm": 0.8200459480285645, + "learning_rate": 0.0002458818915284626, + "loss": 0.4451, + "step": 27040 + }, + { + "epoch": 1.5108249294633629, + "grad_norm": 0.44147002696990967, + "learning_rate": 0.0002458538771851188, + "loss": 0.4088, + "step": 27041 + }, + { + "epoch": 1.5108808000670448, + "grad_norm": 0.6864660978317261, + "learning_rate": 0.000245825862841775, + "loss": 0.4742, + "step": 27042 + }, + { + "epoch": 1.5109366706707266, + "grad_norm": 0.36334890127182007, + "learning_rate": 0.0002457978484984312, + "loss": 0.3569, + "step": 27043 + }, + { + "epoch": 1.5109925412744085, + "grad_norm": 0.42825350165367126, + "learning_rate": 0.0002457698341550874, + "loss": 0.3916, + "step": 27044 + }, + { + "epoch": 1.5110484118780905, + "grad_norm": 0.3704744577407837, + "learning_rate": 0.0002457418198117436, + "loss": 0.3437, + "step": 27045 + }, + { + "epoch": 1.5111042824817722, + "grad_norm": 0.48977985978126526, + "learning_rate": 0.0002457138054683998, + "loss": 0.5458, + "step": 27046 + }, + { + "epoch": 1.511160153085454, + "grad_norm": 0.45982348918914795, + "learning_rate": 0.000245685791125056, + "loss": 0.4293, + "step": 27047 + }, + { + "epoch": 1.511216023689136, + "grad_norm": 0.4038778841495514, + "learning_rate": 0.00024565777678171223, + "loss": 0.3325, + "step": 27048 + }, + { + "epoch": 1.5112718942928178, + "grad_norm": 0.6409962773323059, + "learning_rate": 0.00024562976243836844, + "loss": 0.4, + "step": 27049 + }, + { + "epoch": 1.5113277648964996, + "grad_norm": 0.411016583442688, + "learning_rate": 0.00024560174809502464, + "loss": 0.3769, + "step": 27050 + }, + { + "epoch": 1.5113836355001817, + "grad_norm": 0.44771191477775574, + "learning_rate": 0.0002455737337516809, + "loss": 0.4724, + "step": 27051 + }, + { + "epoch": 1.5114395061038635, + "grad_norm": 0.5697089433670044, + "learning_rate": 0.0002455457194083371, + "loss": 0.377, + "step": 27052 + }, + { + "epoch": 1.5114953767075452, + "grad_norm": 0.36244869232177734, + "learning_rate": 0.00024551770506499326, + "loss": 0.3827, + "step": 27053 + }, + { + "epoch": 1.5115512473112274, + "grad_norm": 0.4392862915992737, + "learning_rate": 0.00024548969072164947, + "loss": 0.5573, + "step": 27054 + }, + { + "epoch": 1.511607117914909, + "grad_norm": 0.4702557325363159, + "learning_rate": 0.00024546167637830567, + "loss": 0.3945, + "step": 27055 + }, + { + "epoch": 1.5116629885185908, + "grad_norm": 0.43900948762893677, + "learning_rate": 0.0002454336620349619, + "loss": 0.5131, + "step": 27056 + }, + { + "epoch": 1.5117188591222728, + "grad_norm": 3.5984761714935303, + "learning_rate": 0.0002454056476916181, + "loss": 0.422, + "step": 27057 + }, + { + "epoch": 1.5117747297259547, + "grad_norm": 0.44077104330062866, + "learning_rate": 0.0002453776333482743, + "loss": 0.3602, + "step": 27058 + }, + { + "epoch": 1.5118306003296365, + "grad_norm": 1.110576868057251, + "learning_rate": 0.00024534961900493055, + "loss": 0.3781, + "step": 27059 + }, + { + "epoch": 1.5118864709333184, + "grad_norm": 0.5029183030128479, + "learning_rate": 0.00024532160466158676, + "loss": 0.4983, + "step": 27060 + }, + { + "epoch": 1.5119423415370004, + "grad_norm": 0.490611732006073, + "learning_rate": 0.00024529359031824296, + "loss": 0.3825, + "step": 27061 + }, + { + "epoch": 1.511998212140682, + "grad_norm": 0.6204043030738831, + "learning_rate": 0.00024526557597489917, + "loss": 0.4053, + "step": 27062 + }, + { + "epoch": 1.512054082744364, + "grad_norm": 0.4024357795715332, + "learning_rate": 0.0002452375616315554, + "loss": 0.3665, + "step": 27063 + }, + { + "epoch": 1.512109953348046, + "grad_norm": 0.39138492941856384, + "learning_rate": 0.0002452095472882116, + "loss": 0.4625, + "step": 27064 + }, + { + "epoch": 1.5121658239517277, + "grad_norm": 0.3724555969238281, + "learning_rate": 0.0002451815329448678, + "loss": 0.4212, + "step": 27065 + }, + { + "epoch": 1.5122216945554097, + "grad_norm": 0.41143038868904114, + "learning_rate": 0.000245153518601524, + "loss": 0.4032, + "step": 27066 + }, + { + "epoch": 1.5122775651590916, + "grad_norm": 0.379128098487854, + "learning_rate": 0.0002451255042581802, + "loss": 0.3706, + "step": 27067 + }, + { + "epoch": 1.5123334357627733, + "grad_norm": 0.9741653800010681, + "learning_rate": 0.0002450974899148364, + "loss": 0.5, + "step": 27068 + }, + { + "epoch": 1.5123893063664553, + "grad_norm": 0.39965659379959106, + "learning_rate": 0.0002450694755714926, + "loss": 0.3965, + "step": 27069 + }, + { + "epoch": 1.5124451769701373, + "grad_norm": 0.3079405426979065, + "learning_rate": 0.0002450414612281488, + "loss": 0.3871, + "step": 27070 + }, + { + "epoch": 1.512501047573819, + "grad_norm": 0.45544326305389404, + "learning_rate": 0.000245013446884805, + "loss": 0.404, + "step": 27071 + }, + { + "epoch": 1.512556918177501, + "grad_norm": 0.626377284526825, + "learning_rate": 0.0002449854325414612, + "loss": 0.4852, + "step": 27072 + }, + { + "epoch": 1.5126127887811829, + "grad_norm": 0.9889181852340698, + "learning_rate": 0.00024495741819811743, + "loss": 0.4824, + "step": 27073 + }, + { + "epoch": 1.5126686593848646, + "grad_norm": 0.4551949203014374, + "learning_rate": 0.00024492940385477364, + "loss": 0.4315, + "step": 27074 + }, + { + "epoch": 1.5127245299885466, + "grad_norm": 0.4214625954627991, + "learning_rate": 0.00024490138951142984, + "loss": 0.4046, + "step": 27075 + }, + { + "epoch": 1.5127804005922285, + "grad_norm": 1.3767069578170776, + "learning_rate": 0.00024487337516808605, + "loss": 0.3225, + "step": 27076 + }, + { + "epoch": 1.5128362711959102, + "grad_norm": 0.42790183424949646, + "learning_rate": 0.00024484536082474225, + "loss": 0.3493, + "step": 27077 + }, + { + "epoch": 1.5128921417995922, + "grad_norm": 0.40144529938697815, + "learning_rate": 0.00024481734648139846, + "loss": 0.4784, + "step": 27078 + }, + { + "epoch": 1.5129480124032741, + "grad_norm": 0.37630516290664673, + "learning_rate": 0.00024478933213805467, + "loss": 0.373, + "step": 27079 + }, + { + "epoch": 1.5130038830069559, + "grad_norm": 0.3876764476299286, + "learning_rate": 0.00024476131779471087, + "loss": 0.5324, + "step": 27080 + }, + { + "epoch": 1.5130597536106376, + "grad_norm": 0.8416647911071777, + "learning_rate": 0.0002447333034513671, + "loss": 0.6173, + "step": 27081 + }, + { + "epoch": 1.5131156242143198, + "grad_norm": 0.5880663394927979, + "learning_rate": 0.0002447052891080233, + "loss": 0.3724, + "step": 27082 + }, + { + "epoch": 1.5131714948180015, + "grad_norm": 0.3469715714454651, + "learning_rate": 0.00024467727476467954, + "loss": 0.4311, + "step": 27083 + }, + { + "epoch": 1.5132273654216832, + "grad_norm": 1.3049145936965942, + "learning_rate": 0.00024464926042133575, + "loss": 0.48, + "step": 27084 + }, + { + "epoch": 1.5132832360253654, + "grad_norm": 0.4694821238517761, + "learning_rate": 0.00024462124607799196, + "loss": 0.4386, + "step": 27085 + }, + { + "epoch": 1.5133391066290471, + "grad_norm": 0.8144823908805847, + "learning_rate": 0.00024459323173464816, + "loss": 0.3954, + "step": 27086 + }, + { + "epoch": 1.5133949772327289, + "grad_norm": 0.9790603518486023, + "learning_rate": 0.00024456521739130437, + "loss": 0.4534, + "step": 27087 + }, + { + "epoch": 1.5134508478364108, + "grad_norm": 0.30708813667297363, + "learning_rate": 0.00024453720304796057, + "loss": 0.3248, + "step": 27088 + }, + { + "epoch": 1.5135067184400928, + "grad_norm": 0.38596439361572266, + "learning_rate": 0.0002445091887046168, + "loss": 0.4345, + "step": 27089 + }, + { + "epoch": 1.5135625890437745, + "grad_norm": 0.8033664226531982, + "learning_rate": 0.000244481174361273, + "loss": 0.3711, + "step": 27090 + }, + { + "epoch": 1.5136184596474564, + "grad_norm": 0.4427851438522339, + "learning_rate": 0.0002444531600179292, + "loss": 0.3631, + "step": 27091 + }, + { + "epoch": 1.5136743302511384, + "grad_norm": 1.8250809907913208, + "learning_rate": 0.0002444251456745854, + "loss": 0.3753, + "step": 27092 + }, + { + "epoch": 1.5137302008548201, + "grad_norm": 0.5121492147445679, + "learning_rate": 0.0002443971313312416, + "loss": 0.4181, + "step": 27093 + }, + { + "epoch": 1.513786071458502, + "grad_norm": 0.3618110120296478, + "learning_rate": 0.0002443691169878978, + "loss": 0.4483, + "step": 27094 + }, + { + "epoch": 1.513841942062184, + "grad_norm": 0.43435707688331604, + "learning_rate": 0.000244341102644554, + "loss": 0.3525, + "step": 27095 + }, + { + "epoch": 1.5138978126658658, + "grad_norm": 0.4009908139705658, + "learning_rate": 0.0002443130883012102, + "loss": 0.4568, + "step": 27096 + }, + { + "epoch": 1.5139536832695477, + "grad_norm": 0.32286298274993896, + "learning_rate": 0.0002442850739578664, + "loss": 0.4208, + "step": 27097 + }, + { + "epoch": 1.5140095538732297, + "grad_norm": 0.5144447088241577, + "learning_rate": 0.00024425705961452263, + "loss": 0.4316, + "step": 27098 + }, + { + "epoch": 1.5140654244769114, + "grad_norm": 0.4843514561653137, + "learning_rate": 0.00024422904527117884, + "loss": 0.467, + "step": 27099 + }, + { + "epoch": 1.5141212950805933, + "grad_norm": 0.3887917399406433, + "learning_rate": 0.00024420103092783504, + "loss": 0.4865, + "step": 27100 + }, + { + "epoch": 1.5141771656842753, + "grad_norm": 0.6219039559364319, + "learning_rate": 0.00024417301658449125, + "loss": 0.4692, + "step": 27101 + }, + { + "epoch": 1.514233036287957, + "grad_norm": 0.6141810417175293, + "learning_rate": 0.00024414500224114745, + "loss": 0.4716, + "step": 27102 + }, + { + "epoch": 1.514288906891639, + "grad_norm": 0.7266367077827454, + "learning_rate": 0.0002441169878978037, + "loss": 0.5011, + "step": 27103 + }, + { + "epoch": 1.514344777495321, + "grad_norm": 0.5091022849082947, + "learning_rate": 0.0002440889735544599, + "loss": 0.4868, + "step": 27104 + }, + { + "epoch": 1.5144006480990027, + "grad_norm": 0.3314410150051117, + "learning_rate": 0.0002440609592111161, + "loss": 0.3903, + "step": 27105 + }, + { + "epoch": 1.5144565187026846, + "grad_norm": 0.3650238811969757, + "learning_rate": 0.0002440329448677723, + "loss": 0.389, + "step": 27106 + }, + { + "epoch": 1.5145123893063666, + "grad_norm": 0.46140021085739136, + "learning_rate": 0.0002440049305244285, + "loss": 0.5131, + "step": 27107 + }, + { + "epoch": 1.5145682599100483, + "grad_norm": 0.5174919962882996, + "learning_rate": 0.00024397691618108472, + "loss": 0.4107, + "step": 27108 + }, + { + "epoch": 1.5146241305137302, + "grad_norm": 0.38509660959243774, + "learning_rate": 0.00024394890183774092, + "loss": 0.4258, + "step": 27109 + }, + { + "epoch": 1.5146800011174122, + "grad_norm": 0.3678419589996338, + "learning_rate": 0.00024392088749439716, + "loss": 0.3161, + "step": 27110 + }, + { + "epoch": 1.514735871721094, + "grad_norm": 0.3924303352832794, + "learning_rate": 0.00024389287315105336, + "loss": 0.4461, + "step": 27111 + }, + { + "epoch": 1.5147917423247759, + "grad_norm": 0.8905881643295288, + "learning_rate": 0.00024386485880770957, + "loss": 0.5185, + "step": 27112 + }, + { + "epoch": 1.5148476129284578, + "grad_norm": 0.5715407133102417, + "learning_rate": 0.00024383684446436577, + "loss": 0.4378, + "step": 27113 + }, + { + "epoch": 1.5149034835321395, + "grad_norm": 0.2933935523033142, + "learning_rate": 0.00024380883012102198, + "loss": 0.3104, + "step": 27114 + }, + { + "epoch": 1.5149593541358213, + "grad_norm": 0.6731224060058594, + "learning_rate": 0.00024378081577767818, + "loss": 0.4341, + "step": 27115 + }, + { + "epoch": 1.5150152247395035, + "grad_norm": 2.7699618339538574, + "learning_rate": 0.0002437528014343344, + "loss": 0.45, + "step": 27116 + }, + { + "epoch": 1.5150710953431852, + "grad_norm": 0.2843782901763916, + "learning_rate": 0.00024372478709099057, + "loss": 0.3123, + "step": 27117 + }, + { + "epoch": 1.515126965946867, + "grad_norm": 0.6976133584976196, + "learning_rate": 0.0002436967727476468, + "loss": 0.4398, + "step": 27118 + }, + { + "epoch": 1.515182836550549, + "grad_norm": 0.49650177359580994, + "learning_rate": 0.000243668758404303, + "loss": 0.4256, + "step": 27119 + }, + { + "epoch": 1.5152387071542308, + "grad_norm": 0.3423368036746979, + "learning_rate": 0.0002436407440609592, + "loss": 0.4155, + "step": 27120 + }, + { + "epoch": 1.5152945777579125, + "grad_norm": 0.39330625534057617, + "learning_rate": 0.00024361272971761542, + "loss": 0.4154, + "step": 27121 + }, + { + "epoch": 1.5153504483615945, + "grad_norm": 0.30349960923194885, + "learning_rate": 0.00024358471537427162, + "loss": 0.3372, + "step": 27122 + }, + { + "epoch": 1.5154063189652764, + "grad_norm": 0.3262613117694855, + "learning_rate": 0.00024355670103092783, + "loss": 0.3178, + "step": 27123 + }, + { + "epoch": 1.5154621895689582, + "grad_norm": 1.339996099472046, + "learning_rate": 0.00024352868668758404, + "loss": 0.3571, + "step": 27124 + }, + { + "epoch": 1.5155180601726401, + "grad_norm": 0.37302109599113464, + "learning_rate": 0.00024350067234424024, + "loss": 0.417, + "step": 27125 + }, + { + "epoch": 1.515573930776322, + "grad_norm": 0.6653668880462646, + "learning_rate": 0.00024347265800089648, + "loss": 0.5733, + "step": 27126 + }, + { + "epoch": 1.5156298013800038, + "grad_norm": 0.324188768863678, + "learning_rate": 0.00024344464365755268, + "loss": 0.4779, + "step": 27127 + }, + { + "epoch": 1.5156856719836858, + "grad_norm": 0.3209812343120575, + "learning_rate": 0.0002434166293142089, + "loss": 0.2875, + "step": 27128 + }, + { + "epoch": 1.5157415425873677, + "grad_norm": 1.203560471534729, + "learning_rate": 0.0002433886149708651, + "loss": 0.5733, + "step": 27129 + }, + { + "epoch": 1.5157974131910494, + "grad_norm": 2.35331392288208, + "learning_rate": 0.0002433606006275213, + "loss": 0.2954, + "step": 27130 + }, + { + "epoch": 1.5158532837947314, + "grad_norm": 0.8161532282829285, + "learning_rate": 0.0002433325862841775, + "loss": 0.4424, + "step": 27131 + }, + { + "epoch": 1.5159091543984133, + "grad_norm": 0.7651375532150269, + "learning_rate": 0.0002433045719408337, + "loss": 0.4919, + "step": 27132 + }, + { + "epoch": 1.515965025002095, + "grad_norm": 0.4796311855316162, + "learning_rate": 0.00024327655759748992, + "loss": 0.4906, + "step": 27133 + }, + { + "epoch": 1.516020895605777, + "grad_norm": 1.5950208902359009, + "learning_rate": 0.00024324854325414615, + "loss": 0.4939, + "step": 27134 + }, + { + "epoch": 1.516076766209459, + "grad_norm": 0.40782442688941956, + "learning_rate": 0.00024322052891080235, + "loss": 0.4517, + "step": 27135 + }, + { + "epoch": 1.5161326368131407, + "grad_norm": 0.4317890703678131, + "learning_rate": 0.00024319251456745856, + "loss": 0.3484, + "step": 27136 + }, + { + "epoch": 1.5161885074168227, + "grad_norm": 0.4910604655742645, + "learning_rate": 0.00024316450022411477, + "loss": 0.4111, + "step": 27137 + }, + { + "epoch": 1.5162443780205046, + "grad_norm": 0.5279322266578674, + "learning_rate": 0.00024313648588077094, + "loss": 0.4792, + "step": 27138 + }, + { + "epoch": 1.5163002486241863, + "grad_norm": 0.41844961047172546, + "learning_rate": 0.00024310847153742715, + "loss": 0.4058, + "step": 27139 + }, + { + "epoch": 1.5163561192278683, + "grad_norm": 0.42013028264045715, + "learning_rate": 0.00024308045719408336, + "loss": 0.3323, + "step": 27140 + }, + { + "epoch": 1.5164119898315502, + "grad_norm": 0.8312857747077942, + "learning_rate": 0.00024305244285073956, + "loss": 0.4284, + "step": 27141 + }, + { + "epoch": 1.516467860435232, + "grad_norm": 0.3590991795063019, + "learning_rate": 0.0002430244285073958, + "loss": 0.3955, + "step": 27142 + }, + { + "epoch": 1.516523731038914, + "grad_norm": 0.41079193353652954, + "learning_rate": 0.000242996414164052, + "loss": 0.3956, + "step": 27143 + }, + { + "epoch": 1.5165796016425959, + "grad_norm": 0.4293532073497772, + "learning_rate": 0.0002429683998207082, + "loss": 0.4843, + "step": 27144 + }, + { + "epoch": 1.5166354722462776, + "grad_norm": 0.41068822145462036, + "learning_rate": 0.0002429403854773644, + "loss": 0.4805, + "step": 27145 + }, + { + "epoch": 1.5166913428499595, + "grad_norm": 0.3926202058792114, + "learning_rate": 0.00024291237113402062, + "loss": 0.3674, + "step": 27146 + }, + { + "epoch": 1.5167472134536415, + "grad_norm": 0.6428609490394592, + "learning_rate": 0.00024288435679067682, + "loss": 0.3417, + "step": 27147 + }, + { + "epoch": 1.5168030840573232, + "grad_norm": 0.7835080027580261, + "learning_rate": 0.00024285634244733303, + "loss": 0.4161, + "step": 27148 + }, + { + "epoch": 1.516858954661005, + "grad_norm": 4.73732852935791, + "learning_rate": 0.00024282832810398924, + "loss": 0.4018, + "step": 27149 + }, + { + "epoch": 1.5169148252646871, + "grad_norm": 1.0399593114852905, + "learning_rate": 0.00024280031376064547, + "loss": 0.4277, + "step": 27150 + }, + { + "epoch": 1.5169706958683689, + "grad_norm": 0.44896602630615234, + "learning_rate": 0.00024277229941730167, + "loss": 0.3561, + "step": 27151 + }, + { + "epoch": 1.5170265664720506, + "grad_norm": 0.9666321277618408, + "learning_rate": 0.00024274428507395788, + "loss": 0.5465, + "step": 27152 + }, + { + "epoch": 1.5170824370757328, + "grad_norm": 0.37628069519996643, + "learning_rate": 0.00024271627073061409, + "loss": 0.4221, + "step": 27153 + }, + { + "epoch": 1.5171383076794145, + "grad_norm": 0.43613582849502563, + "learning_rate": 0.0002426882563872703, + "loss": 0.5211, + "step": 27154 + }, + { + "epoch": 1.5171941782830962, + "grad_norm": 1.3975069522857666, + "learning_rate": 0.0002426602420439265, + "loss": 0.3612, + "step": 27155 + }, + { + "epoch": 1.5172500488867782, + "grad_norm": 0.3440871238708496, + "learning_rate": 0.0002426322277005827, + "loss": 0.3652, + "step": 27156 + }, + { + "epoch": 1.5173059194904601, + "grad_norm": 0.5806730389595032, + "learning_rate": 0.0002426042133572389, + "loss": 0.5057, + "step": 27157 + }, + { + "epoch": 1.5173617900941418, + "grad_norm": 0.7122828960418701, + "learning_rate": 0.00024257619901389514, + "loss": 0.4605, + "step": 27158 + }, + { + "epoch": 1.5174176606978238, + "grad_norm": 0.3906717598438263, + "learning_rate": 0.00024254818467055132, + "loss": 0.381, + "step": 27159 + }, + { + "epoch": 1.5174735313015058, + "grad_norm": 0.37187737226486206, + "learning_rate": 0.00024252017032720753, + "loss": 0.3719, + "step": 27160 + }, + { + "epoch": 1.5175294019051875, + "grad_norm": 0.37941598892211914, + "learning_rate": 0.00024249215598386373, + "loss": 0.4661, + "step": 27161 + }, + { + "epoch": 1.5175852725088694, + "grad_norm": 0.3282466232776642, + "learning_rate": 0.00024246414164051994, + "loss": 0.3371, + "step": 27162 + }, + { + "epoch": 1.5176411431125514, + "grad_norm": 3.65228533744812, + "learning_rate": 0.00024243612729717614, + "loss": 0.3718, + "step": 27163 + }, + { + "epoch": 1.5176970137162331, + "grad_norm": 1.411009430885315, + "learning_rate": 0.00024240811295383235, + "loss": 0.3755, + "step": 27164 + }, + { + "epoch": 1.517752884319915, + "grad_norm": 1.1351687908172607, + "learning_rate": 0.00024238009861048856, + "loss": 0.4486, + "step": 27165 + }, + { + "epoch": 1.517808754923597, + "grad_norm": 0.3753211200237274, + "learning_rate": 0.0002423520842671448, + "loss": 0.3641, + "step": 27166 + }, + { + "epoch": 1.5178646255272787, + "grad_norm": 0.5559199452400208, + "learning_rate": 0.000242324069923801, + "loss": 0.3824, + "step": 27167 + }, + { + "epoch": 1.5179204961309607, + "grad_norm": 2.889558792114258, + "learning_rate": 0.0002422960555804572, + "loss": 0.4731, + "step": 27168 + }, + { + "epoch": 1.5179763667346426, + "grad_norm": 0.4988475739955902, + "learning_rate": 0.0002422680412371134, + "loss": 0.4655, + "step": 27169 + }, + { + "epoch": 1.5180322373383244, + "grad_norm": 0.35066306591033936, + "learning_rate": 0.0002422400268937696, + "loss": 0.3153, + "step": 27170 + }, + { + "epoch": 1.5180881079420063, + "grad_norm": 0.40172672271728516, + "learning_rate": 0.00024221201255042582, + "loss": 0.4435, + "step": 27171 + }, + { + "epoch": 1.5181439785456883, + "grad_norm": 0.46777743101119995, + "learning_rate": 0.00024218399820708202, + "loss": 0.4554, + "step": 27172 + }, + { + "epoch": 1.51819984914937, + "grad_norm": 1.952703833580017, + "learning_rate": 0.00024215598386373826, + "loss": 0.3712, + "step": 27173 + }, + { + "epoch": 1.518255719753052, + "grad_norm": 0.8110021352767944, + "learning_rate": 0.00024212796952039446, + "loss": 0.3866, + "step": 27174 + }, + { + "epoch": 1.518311590356734, + "grad_norm": 0.46172335743904114, + "learning_rate": 0.00024209995517705067, + "loss": 0.3903, + "step": 27175 + }, + { + "epoch": 1.5183674609604156, + "grad_norm": 0.4133976995944977, + "learning_rate": 0.00024207194083370687, + "loss": 0.4154, + "step": 27176 + }, + { + "epoch": 1.5184233315640976, + "grad_norm": 0.6370513439178467, + "learning_rate": 0.00024204392649036308, + "loss": 0.4399, + "step": 27177 + }, + { + "epoch": 1.5184792021677795, + "grad_norm": 0.892169177532196, + "learning_rate": 0.00024201591214701929, + "loss": 0.5222, + "step": 27178 + }, + { + "epoch": 1.5185350727714613, + "grad_norm": 0.8814449310302734, + "learning_rate": 0.0002419878978036755, + "loss": 0.434, + "step": 27179 + }, + { + "epoch": 1.5185909433751432, + "grad_norm": 0.5433261394500732, + "learning_rate": 0.00024195988346033167, + "loss": 0.4279, + "step": 27180 + }, + { + "epoch": 1.5186468139788252, + "grad_norm": 0.44577229022979736, + "learning_rate": 0.0002419318691169879, + "loss": 0.4748, + "step": 27181 + }, + { + "epoch": 1.518702684582507, + "grad_norm": 0.5548958778381348, + "learning_rate": 0.0002419038547736441, + "loss": 0.2608, + "step": 27182 + }, + { + "epoch": 1.5187585551861886, + "grad_norm": 0.5621219277381897, + "learning_rate": 0.00024187584043030032, + "loss": 0.4155, + "step": 27183 + }, + { + "epoch": 1.5188144257898708, + "grad_norm": 0.306785523891449, + "learning_rate": 0.00024184782608695652, + "loss": 0.3557, + "step": 27184 + }, + { + "epoch": 1.5188702963935525, + "grad_norm": 0.3933010697364807, + "learning_rate": 0.00024181981174361273, + "loss": 0.3331, + "step": 27185 + }, + { + "epoch": 1.5189261669972343, + "grad_norm": 17.214885711669922, + "learning_rate": 0.00024179179740026893, + "loss": 0.5183, + "step": 27186 + }, + { + "epoch": 1.5189820376009164, + "grad_norm": 1.356013536453247, + "learning_rate": 0.00024176378305692514, + "loss": 0.3752, + "step": 27187 + }, + { + "epoch": 1.5190379082045982, + "grad_norm": 0.4321078658103943, + "learning_rate": 0.00024173576871358134, + "loss": 0.4827, + "step": 27188 + }, + { + "epoch": 1.51909377880828, + "grad_norm": 0.41295820474624634, + "learning_rate": 0.00024170775437023758, + "loss": 0.3504, + "step": 27189 + }, + { + "epoch": 1.5191496494119618, + "grad_norm": 2.2101762294769287, + "learning_rate": 0.00024167974002689378, + "loss": 0.3761, + "step": 27190 + }, + { + "epoch": 1.5192055200156438, + "grad_norm": 0.5014694333076477, + "learning_rate": 0.00024165172568355, + "loss": 0.4846, + "step": 27191 + }, + { + "epoch": 1.5192613906193255, + "grad_norm": 0.8893139958381653, + "learning_rate": 0.0002416237113402062, + "loss": 0.461, + "step": 27192 + }, + { + "epoch": 1.5193172612230075, + "grad_norm": 0.4117484390735626, + "learning_rate": 0.0002415956969968624, + "loss": 0.4039, + "step": 27193 + }, + { + "epoch": 1.5193731318266894, + "grad_norm": 0.8676986694335938, + "learning_rate": 0.0002415676826535186, + "loss": 0.4756, + "step": 27194 + }, + { + "epoch": 1.5194290024303712, + "grad_norm": 0.3161623477935791, + "learning_rate": 0.0002415396683101748, + "loss": 0.4283, + "step": 27195 + }, + { + "epoch": 1.519484873034053, + "grad_norm": 0.8111993074417114, + "learning_rate": 0.00024151165396683102, + "loss": 0.4306, + "step": 27196 + }, + { + "epoch": 1.519540743637735, + "grad_norm": 0.5442301034927368, + "learning_rate": 0.00024148363962348725, + "loss": 0.3389, + "step": 27197 + }, + { + "epoch": 1.5195966142414168, + "grad_norm": 0.5490463376045227, + "learning_rate": 0.00024145562528014346, + "loss": 0.5693, + "step": 27198 + }, + { + "epoch": 1.5196524848450987, + "grad_norm": 3.3546581268310547, + "learning_rate": 0.00024142761093679966, + "loss": 0.4498, + "step": 27199 + }, + { + "epoch": 1.5197083554487807, + "grad_norm": 0.6803393959999084, + "learning_rate": 0.00024139959659345587, + "loss": 0.4636, + "step": 27200 + }, + { + "epoch": 1.5197642260524624, + "grad_norm": 0.4123590886592865, + "learning_rate": 0.00024137158225011207, + "loss": 0.377, + "step": 27201 + }, + { + "epoch": 1.5198200966561444, + "grad_norm": 3.5688085556030273, + "learning_rate": 0.00024134356790676825, + "loss": 0.4055, + "step": 27202 + }, + { + "epoch": 1.5198759672598263, + "grad_norm": 0.4313047230243683, + "learning_rate": 0.00024131555356342446, + "loss": 0.5136, + "step": 27203 + }, + { + "epoch": 1.519931837863508, + "grad_norm": 0.3454136550426483, + "learning_rate": 0.00024128753922008066, + "loss": 0.345, + "step": 27204 + }, + { + "epoch": 1.51998770846719, + "grad_norm": 1.1606401205062866, + "learning_rate": 0.0002412595248767369, + "loss": 0.3591, + "step": 27205 + }, + { + "epoch": 1.520043579070872, + "grad_norm": 4.206499099731445, + "learning_rate": 0.0002412315105333931, + "loss": 0.3727, + "step": 27206 + }, + { + "epoch": 1.5200994496745537, + "grad_norm": 0.5074648857116699, + "learning_rate": 0.0002412034961900493, + "loss": 0.4157, + "step": 27207 + }, + { + "epoch": 1.5201553202782356, + "grad_norm": 2.423895835876465, + "learning_rate": 0.00024117548184670551, + "loss": 0.4028, + "step": 27208 + }, + { + "epoch": 1.5202111908819176, + "grad_norm": 0.4261736273765564, + "learning_rate": 0.00024114746750336172, + "loss": 0.3944, + "step": 27209 + }, + { + "epoch": 1.5202670614855993, + "grad_norm": 1.2130545377731323, + "learning_rate": 0.00024111945316001793, + "loss": 0.5239, + "step": 27210 + }, + { + "epoch": 1.5203229320892813, + "grad_norm": 1.020094394683838, + "learning_rate": 0.00024109143881667413, + "loss": 0.3774, + "step": 27211 + }, + { + "epoch": 1.5203788026929632, + "grad_norm": 1.8251856565475464, + "learning_rate": 0.00024106342447333034, + "loss": 0.5076, + "step": 27212 + }, + { + "epoch": 1.520434673296645, + "grad_norm": 0.6420210599899292, + "learning_rate": 0.00024103541012998657, + "loss": 0.5172, + "step": 27213 + }, + { + "epoch": 1.520490543900327, + "grad_norm": 0.30073729157447815, + "learning_rate": 0.00024100739578664278, + "loss": 0.4141, + "step": 27214 + }, + { + "epoch": 1.5205464145040088, + "grad_norm": 0.3832685649394989, + "learning_rate": 0.00024097938144329898, + "loss": 0.3279, + "step": 27215 + }, + { + "epoch": 1.5206022851076906, + "grad_norm": 0.6287106275558472, + "learning_rate": 0.0002409513670999552, + "loss": 0.4354, + "step": 27216 + }, + { + "epoch": 1.5206581557113723, + "grad_norm": 0.4485498368740082, + "learning_rate": 0.0002409233527566114, + "loss": 0.3952, + "step": 27217 + }, + { + "epoch": 1.5207140263150545, + "grad_norm": 0.4482787847518921, + "learning_rate": 0.0002408953384132676, + "loss": 0.3674, + "step": 27218 + }, + { + "epoch": 1.5207698969187362, + "grad_norm": 0.8484033346176147, + "learning_rate": 0.0002408673240699238, + "loss": 0.4685, + "step": 27219 + }, + { + "epoch": 1.520825767522418, + "grad_norm": 0.8580420613288879, + "learning_rate": 0.00024083930972658, + "loss": 0.3828, + "step": 27220 + }, + { + "epoch": 1.5208816381261, + "grad_norm": 0.881024956703186, + "learning_rate": 0.00024081129538323624, + "loss": 0.4071, + "step": 27221 + }, + { + "epoch": 1.5209375087297818, + "grad_norm": 0.5705154538154602, + "learning_rate": 0.00024078328103989245, + "loss": 0.5158, + "step": 27222 + }, + { + "epoch": 1.5209933793334636, + "grad_norm": 0.28629469871520996, + "learning_rate": 0.00024075526669654863, + "loss": 0.3458, + "step": 27223 + }, + { + "epoch": 1.5210492499371455, + "grad_norm": 0.48127472400665283, + "learning_rate": 0.00024072725235320483, + "loss": 0.4754, + "step": 27224 + }, + { + "epoch": 1.5211051205408275, + "grad_norm": 0.3959304094314575, + "learning_rate": 0.00024069923800986104, + "loss": 0.3333, + "step": 27225 + }, + { + "epoch": 1.5211609911445092, + "grad_norm": 0.4963681995868683, + "learning_rate": 0.00024067122366651725, + "loss": 0.4642, + "step": 27226 + }, + { + "epoch": 1.5212168617481912, + "grad_norm": 1.6926106214523315, + "learning_rate": 0.00024064320932317345, + "loss": 0.4643, + "step": 27227 + }, + { + "epoch": 1.521272732351873, + "grad_norm": 0.6406708359718323, + "learning_rate": 0.00024061519497982966, + "loss": 0.425, + "step": 27228 + }, + { + "epoch": 1.5213286029555548, + "grad_norm": 0.6254609227180481, + "learning_rate": 0.0002405871806364859, + "loss": 0.2557, + "step": 27229 + }, + { + "epoch": 1.5213844735592368, + "grad_norm": 0.5214444398880005, + "learning_rate": 0.0002405591662931421, + "loss": 0.4129, + "step": 27230 + }, + { + "epoch": 1.5214403441629187, + "grad_norm": 0.3805098533630371, + "learning_rate": 0.0002405311519497983, + "loss": 0.4325, + "step": 27231 + }, + { + "epoch": 1.5214962147666005, + "grad_norm": 0.74355149269104, + "learning_rate": 0.0002405031376064545, + "loss": 0.4557, + "step": 27232 + }, + { + "epoch": 1.5215520853702824, + "grad_norm": 0.5801365971565247, + "learning_rate": 0.00024047512326311071, + "loss": 0.3843, + "step": 27233 + }, + { + "epoch": 1.5216079559739644, + "grad_norm": 0.31522423028945923, + "learning_rate": 0.00024044710891976692, + "loss": 0.3226, + "step": 27234 + }, + { + "epoch": 1.521663826577646, + "grad_norm": 0.4987167716026306, + "learning_rate": 0.00024041909457642313, + "loss": 0.4829, + "step": 27235 + }, + { + "epoch": 1.521719697181328, + "grad_norm": 0.5285966992378235, + "learning_rate": 0.00024039108023307936, + "loss": 0.4991, + "step": 27236 + }, + { + "epoch": 1.52177556778501, + "grad_norm": 0.46623602509498596, + "learning_rate": 0.00024036306588973556, + "loss": 0.4636, + "step": 27237 + }, + { + "epoch": 1.5218314383886917, + "grad_norm": 0.3739365041255951, + "learning_rate": 0.00024033505154639177, + "loss": 0.479, + "step": 27238 + }, + { + "epoch": 1.5218873089923737, + "grad_norm": 0.45569539070129395, + "learning_rate": 0.00024030703720304798, + "loss": 0.3303, + "step": 27239 + }, + { + "epoch": 1.5219431795960556, + "grad_norm": 0.3092219829559326, + "learning_rate": 0.00024027902285970418, + "loss": 0.285, + "step": 27240 + }, + { + "epoch": 1.5219990501997374, + "grad_norm": 0.4516395330429077, + "learning_rate": 0.0002402510085163604, + "loss": 0.37, + "step": 27241 + }, + { + "epoch": 1.5220549208034193, + "grad_norm": 0.3658222258090973, + "learning_rate": 0.0002402229941730166, + "loss": 0.5068, + "step": 27242 + }, + { + "epoch": 1.5221107914071013, + "grad_norm": 0.5049994587898254, + "learning_rate": 0.0002401949798296728, + "loss": 0.472, + "step": 27243 + }, + { + "epoch": 1.522166662010783, + "grad_norm": 0.382595032453537, + "learning_rate": 0.000240166965486329, + "loss": 0.312, + "step": 27244 + }, + { + "epoch": 1.522222532614465, + "grad_norm": 1.3169244527816772, + "learning_rate": 0.0002401389511429852, + "loss": 0.4908, + "step": 27245 + }, + { + "epoch": 1.522278403218147, + "grad_norm": 0.6699212193489075, + "learning_rate": 0.00024011093679964142, + "loss": 0.3783, + "step": 27246 + }, + { + "epoch": 1.5223342738218286, + "grad_norm": 0.35343071818351746, + "learning_rate": 0.00024008292245629762, + "loss": 0.4121, + "step": 27247 + }, + { + "epoch": 1.5223901444255106, + "grad_norm": 0.4251079857349396, + "learning_rate": 0.00024005490811295383, + "loss": 0.6218, + "step": 27248 + }, + { + "epoch": 1.5224460150291925, + "grad_norm": 0.3154706656932831, + "learning_rate": 0.00024002689376961003, + "loss": 0.3979, + "step": 27249 + }, + { + "epoch": 1.5225018856328743, + "grad_norm": 0.5241888761520386, + "learning_rate": 0.00023999887942626624, + "loss": 0.2873, + "step": 27250 + }, + { + "epoch": 1.522557756236556, + "grad_norm": 6.531786918640137, + "learning_rate": 0.00023997086508292245, + "loss": 0.3557, + "step": 27251 + }, + { + "epoch": 1.5226136268402382, + "grad_norm": 0.7775462865829468, + "learning_rate": 0.00023994285073957868, + "loss": 0.3995, + "step": 27252 + }, + { + "epoch": 1.5226694974439199, + "grad_norm": 0.39683786034584045, + "learning_rate": 0.00023991483639623488, + "loss": 0.4937, + "step": 27253 + }, + { + "epoch": 1.5227253680476016, + "grad_norm": 1.8581801652908325, + "learning_rate": 0.0002398868220528911, + "loss": 0.4951, + "step": 27254 + }, + { + "epoch": 1.5227812386512838, + "grad_norm": 0.5609967708587646, + "learning_rate": 0.0002398588077095473, + "loss": 0.4132, + "step": 27255 + }, + { + "epoch": 1.5228371092549655, + "grad_norm": 0.5056455135345459, + "learning_rate": 0.0002398307933662035, + "loss": 0.3755, + "step": 27256 + }, + { + "epoch": 1.5228929798586472, + "grad_norm": 0.5220693945884705, + "learning_rate": 0.0002398027790228597, + "loss": 0.4355, + "step": 27257 + }, + { + "epoch": 1.5229488504623292, + "grad_norm": 0.7116255760192871, + "learning_rate": 0.00023977476467951591, + "loss": 0.4629, + "step": 27258 + }, + { + "epoch": 1.5230047210660111, + "grad_norm": 0.36034637689590454, + "learning_rate": 0.00023974675033617212, + "loss": 0.4113, + "step": 27259 + }, + { + "epoch": 1.5230605916696929, + "grad_norm": 0.4844285249710083, + "learning_rate": 0.00023971873599282835, + "loss": 0.3788, + "step": 27260 + }, + { + "epoch": 1.5231164622733748, + "grad_norm": 0.4711947441101074, + "learning_rate": 0.00023969072164948456, + "loss": 0.387, + "step": 27261 + }, + { + "epoch": 1.5231723328770568, + "grad_norm": 0.46270930767059326, + "learning_rate": 0.00023966270730614076, + "loss": 0.5029, + "step": 27262 + }, + { + "epoch": 1.5232282034807385, + "grad_norm": 2.663978338241577, + "learning_rate": 0.00023963469296279697, + "loss": 0.29, + "step": 27263 + }, + { + "epoch": 1.5232840740844205, + "grad_norm": 0.35066667199134827, + "learning_rate": 0.00023960667861945318, + "loss": 0.3573, + "step": 27264 + }, + { + "epoch": 1.5233399446881024, + "grad_norm": 0.5547683835029602, + "learning_rate": 0.00023957866427610935, + "loss": 0.356, + "step": 27265 + }, + { + "epoch": 1.5233958152917841, + "grad_norm": 0.38472259044647217, + "learning_rate": 0.00023955064993276556, + "loss": 0.4606, + "step": 27266 + }, + { + "epoch": 1.523451685895466, + "grad_norm": 0.38955891132354736, + "learning_rate": 0.00023952263558942177, + "loss": 0.3161, + "step": 27267 + }, + { + "epoch": 1.523507556499148, + "grad_norm": 0.8533410429954529, + "learning_rate": 0.000239494621246078, + "loss": 0.3931, + "step": 27268 + }, + { + "epoch": 1.5235634271028298, + "grad_norm": 0.4077841341495514, + "learning_rate": 0.0002394666069027342, + "loss": 0.4223, + "step": 27269 + }, + { + "epoch": 1.5236192977065117, + "grad_norm": 0.4235222637653351, + "learning_rate": 0.0002394385925593904, + "loss": 0.3495, + "step": 27270 + }, + { + "epoch": 1.5236751683101937, + "grad_norm": 0.4391299784183502, + "learning_rate": 0.00023941057821604662, + "loss": 0.3839, + "step": 27271 + }, + { + "epoch": 1.5237310389138754, + "grad_norm": 1.2513521909713745, + "learning_rate": 0.00023938256387270282, + "loss": 0.5157, + "step": 27272 + }, + { + "epoch": 1.5237869095175574, + "grad_norm": 0.5578777194023132, + "learning_rate": 0.00023935454952935903, + "loss": 0.3968, + "step": 27273 + }, + { + "epoch": 1.5238427801212393, + "grad_norm": 0.6113776564598083, + "learning_rate": 0.00023932653518601523, + "loss": 0.4159, + "step": 27274 + }, + { + "epoch": 1.523898650724921, + "grad_norm": 1.2151367664337158, + "learning_rate": 0.00023929852084267144, + "loss": 0.5477, + "step": 27275 + }, + { + "epoch": 1.523954521328603, + "grad_norm": 0.6326130628585815, + "learning_rate": 0.00023927050649932767, + "loss": 0.4869, + "step": 27276 + }, + { + "epoch": 1.524010391932285, + "grad_norm": 1.2779505252838135, + "learning_rate": 0.00023924249215598388, + "loss": 0.4378, + "step": 27277 + }, + { + "epoch": 1.5240662625359667, + "grad_norm": 0.4508815109729767, + "learning_rate": 0.00023921447781264008, + "loss": 0.4521, + "step": 27278 + }, + { + "epoch": 1.5241221331396486, + "grad_norm": 0.6910430192947388, + "learning_rate": 0.0002391864634692963, + "loss": 0.4205, + "step": 27279 + }, + { + "epoch": 1.5241780037433306, + "grad_norm": 0.4239903390407562, + "learning_rate": 0.0002391584491259525, + "loss": 0.4388, + "step": 27280 + }, + { + "epoch": 1.5242338743470123, + "grad_norm": 0.7480160593986511, + "learning_rate": 0.0002391304347826087, + "loss": 0.3943, + "step": 27281 + }, + { + "epoch": 1.5242897449506942, + "grad_norm": 0.6098573207855225, + "learning_rate": 0.0002391024204392649, + "loss": 0.4623, + "step": 27282 + }, + { + "epoch": 1.5243456155543762, + "grad_norm": 0.3301045894622803, + "learning_rate": 0.0002390744060959211, + "loss": 0.3374, + "step": 27283 + }, + { + "epoch": 1.524401486158058, + "grad_norm": 0.3464714586734772, + "learning_rate": 0.00023904639175257735, + "loss": 0.4421, + "step": 27284 + }, + { + "epoch": 1.5244573567617397, + "grad_norm": 0.7185529470443726, + "learning_rate": 0.00023901837740923355, + "loss": 0.4491, + "step": 27285 + }, + { + "epoch": 1.5245132273654218, + "grad_norm": 0.4127735197544098, + "learning_rate": 0.00023899036306588976, + "loss": 0.4248, + "step": 27286 + }, + { + "epoch": 1.5245690979691036, + "grad_norm": 0.5250301957130432, + "learning_rate": 0.00023896234872254594, + "loss": 0.4353, + "step": 27287 + }, + { + "epoch": 1.5246249685727853, + "grad_norm": 0.7550703287124634, + "learning_rate": 0.00023893433437920214, + "loss": 0.6152, + "step": 27288 + }, + { + "epoch": 1.5246808391764675, + "grad_norm": 0.5265851616859436, + "learning_rate": 0.00023890632003585835, + "loss": 0.5194, + "step": 27289 + }, + { + "epoch": 1.5247367097801492, + "grad_norm": 0.3241077959537506, + "learning_rate": 0.00023887830569251455, + "loss": 0.4554, + "step": 27290 + }, + { + "epoch": 1.524792580383831, + "grad_norm": 6.772311687469482, + "learning_rate": 0.00023885029134917076, + "loss": 0.3664, + "step": 27291 + }, + { + "epoch": 1.5248484509875129, + "grad_norm": 0.6479354500770569, + "learning_rate": 0.000238822277005827, + "loss": 0.415, + "step": 27292 + }, + { + "epoch": 1.5249043215911948, + "grad_norm": 0.407571405172348, + "learning_rate": 0.0002387942626624832, + "loss": 0.3858, + "step": 27293 + }, + { + "epoch": 1.5249601921948766, + "grad_norm": 0.935457706451416, + "learning_rate": 0.0002387662483191394, + "loss": 0.3616, + "step": 27294 + }, + { + "epoch": 1.5250160627985585, + "grad_norm": 0.449314147233963, + "learning_rate": 0.0002387382339757956, + "loss": 0.3904, + "step": 27295 + }, + { + "epoch": 1.5250719334022405, + "grad_norm": 0.7116770148277283, + "learning_rate": 0.00023871021963245182, + "loss": 0.5155, + "step": 27296 + }, + { + "epoch": 1.5251278040059222, + "grad_norm": 0.817043662071228, + "learning_rate": 0.00023868220528910802, + "loss": 0.4328, + "step": 27297 + }, + { + "epoch": 1.5251836746096041, + "grad_norm": 0.54438316822052, + "learning_rate": 0.00023865419094576423, + "loss": 0.4364, + "step": 27298 + }, + { + "epoch": 1.525239545213286, + "grad_norm": 0.4430907368659973, + "learning_rate": 0.00023862617660242046, + "loss": 0.3644, + "step": 27299 + }, + { + "epoch": 1.5252954158169678, + "grad_norm": 0.33324187994003296, + "learning_rate": 0.00023859816225907667, + "loss": 0.3279, + "step": 27300 + }, + { + "epoch": 1.5253512864206498, + "grad_norm": 0.5131190419197083, + "learning_rate": 0.00023857014791573287, + "loss": 0.3388, + "step": 27301 + }, + { + "epoch": 1.5254071570243317, + "grad_norm": 0.6104937791824341, + "learning_rate": 0.00023854213357238908, + "loss": 0.5806, + "step": 27302 + }, + { + "epoch": 1.5254630276280134, + "grad_norm": 1.2106472253799438, + "learning_rate": 0.00023851411922904528, + "loss": 0.4833, + "step": 27303 + }, + { + "epoch": 1.5255188982316954, + "grad_norm": 0.4978826344013214, + "learning_rate": 0.0002384861048857015, + "loss": 0.5191, + "step": 27304 + }, + { + "epoch": 1.5255747688353773, + "grad_norm": 0.4879085421562195, + "learning_rate": 0.0002384580905423577, + "loss": 0.4656, + "step": 27305 + }, + { + "epoch": 1.525630639439059, + "grad_norm": 0.7316271662712097, + "learning_rate": 0.0002384300761990139, + "loss": 0.3883, + "step": 27306 + }, + { + "epoch": 1.525686510042741, + "grad_norm": 0.5241630673408508, + "learning_rate": 0.00023840206185567013, + "loss": 0.3922, + "step": 27307 + }, + { + "epoch": 1.525742380646423, + "grad_norm": 0.5267336964607239, + "learning_rate": 0.0002383740475123263, + "loss": 0.5005, + "step": 27308 + }, + { + "epoch": 1.5257982512501047, + "grad_norm": 0.34401991963386536, + "learning_rate": 0.00023834603316898252, + "loss": 0.3822, + "step": 27309 + }, + { + "epoch": 1.5258541218537867, + "grad_norm": 0.44951409101486206, + "learning_rate": 0.00023831801882563872, + "loss": 0.4271, + "step": 27310 + }, + { + "epoch": 1.5259099924574686, + "grad_norm": 0.6743469834327698, + "learning_rate": 0.00023829000448229493, + "loss": 0.5238, + "step": 27311 + }, + { + "epoch": 1.5259658630611503, + "grad_norm": 0.4588659107685089, + "learning_rate": 0.00023826199013895114, + "loss": 0.3834, + "step": 27312 + }, + { + "epoch": 1.5260217336648323, + "grad_norm": 0.35519474744796753, + "learning_rate": 0.00023823397579560734, + "loss": 0.3592, + "step": 27313 + }, + { + "epoch": 1.5260776042685142, + "grad_norm": 2.773869276046753, + "learning_rate": 0.00023820596145226355, + "loss": 0.4444, + "step": 27314 + }, + { + "epoch": 1.526133474872196, + "grad_norm": 1.9681897163391113, + "learning_rate": 0.00023817794710891978, + "loss": 0.4331, + "step": 27315 + }, + { + "epoch": 1.5261893454758777, + "grad_norm": 0.49987781047821045, + "learning_rate": 0.000238149932765576, + "loss": 0.4098, + "step": 27316 + }, + { + "epoch": 1.5262452160795599, + "grad_norm": 0.5058838725090027, + "learning_rate": 0.0002381219184222322, + "loss": 0.3749, + "step": 27317 + }, + { + "epoch": 1.5263010866832416, + "grad_norm": 1.3968210220336914, + "learning_rate": 0.0002380939040788884, + "loss": 0.5373, + "step": 27318 + }, + { + "epoch": 1.5263569572869233, + "grad_norm": 3.7920291423797607, + "learning_rate": 0.0002380658897355446, + "loss": 0.4, + "step": 27319 + }, + { + "epoch": 1.5264128278906055, + "grad_norm": 0.3518010377883911, + "learning_rate": 0.0002380378753922008, + "loss": 0.4078, + "step": 27320 + }, + { + "epoch": 1.5264686984942872, + "grad_norm": 0.3984849154949188, + "learning_rate": 0.00023800986104885702, + "loss": 0.379, + "step": 27321 + }, + { + "epoch": 1.526524569097969, + "grad_norm": 0.6074727177619934, + "learning_rate": 0.00023798184670551322, + "loss": 0.4756, + "step": 27322 + }, + { + "epoch": 1.5265804397016511, + "grad_norm": 2.2903807163238525, + "learning_rate": 0.00023795383236216945, + "loss": 0.4846, + "step": 27323 + }, + { + "epoch": 1.5266363103053329, + "grad_norm": 0.6180073022842407, + "learning_rate": 0.00023792581801882566, + "loss": 0.3153, + "step": 27324 + }, + { + "epoch": 1.5266921809090146, + "grad_norm": 0.5676213502883911, + "learning_rate": 0.00023789780367548187, + "loss": 0.4451, + "step": 27325 + }, + { + "epoch": 1.5267480515126965, + "grad_norm": 0.48676666617393494, + "learning_rate": 0.00023786978933213807, + "loss": 0.3782, + "step": 27326 + }, + { + "epoch": 1.5268039221163785, + "grad_norm": 0.5178574919700623, + "learning_rate": 0.00023784177498879428, + "loss": 0.3251, + "step": 27327 + }, + { + "epoch": 1.5268597927200602, + "grad_norm": 0.42892494797706604, + "learning_rate": 0.00023781376064545048, + "loss": 0.4358, + "step": 27328 + }, + { + "epoch": 1.5269156633237422, + "grad_norm": 0.32155096530914307, + "learning_rate": 0.00023778574630210666, + "loss": 0.3629, + "step": 27329 + }, + { + "epoch": 1.5269715339274241, + "grad_norm": 0.39960145950317383, + "learning_rate": 0.00023775773195876287, + "loss": 0.421, + "step": 27330 + }, + { + "epoch": 1.5270274045311059, + "grad_norm": 0.3750848174095154, + "learning_rate": 0.0002377297176154191, + "loss": 0.4591, + "step": 27331 + }, + { + "epoch": 1.5270832751347878, + "grad_norm": 0.46726882457733154, + "learning_rate": 0.0002377017032720753, + "loss": 0.5071, + "step": 27332 + }, + { + "epoch": 1.5271391457384698, + "grad_norm": 0.482266366481781, + "learning_rate": 0.0002376736889287315, + "loss": 0.4512, + "step": 27333 + }, + { + "epoch": 1.5271950163421515, + "grad_norm": 0.3758215606212616, + "learning_rate": 0.00023764567458538772, + "loss": 0.3538, + "step": 27334 + }, + { + "epoch": 1.5272508869458334, + "grad_norm": 0.4092075824737549, + "learning_rate": 0.00023761766024204392, + "loss": 0.3558, + "step": 27335 + }, + { + "epoch": 1.5273067575495154, + "grad_norm": 0.4452613294124603, + "learning_rate": 0.00023758964589870013, + "loss": 0.4405, + "step": 27336 + }, + { + "epoch": 1.5273626281531971, + "grad_norm": 0.41175830364227295, + "learning_rate": 0.00023756163155535634, + "loss": 0.4131, + "step": 27337 + }, + { + "epoch": 1.527418498756879, + "grad_norm": 1.7354158163070679, + "learning_rate": 0.00023753361721201254, + "loss": 0.4067, + "step": 27338 + }, + { + "epoch": 1.527474369360561, + "grad_norm": 1.5373127460479736, + "learning_rate": 0.00023750560286866877, + "loss": 0.3617, + "step": 27339 + }, + { + "epoch": 1.5275302399642428, + "grad_norm": 0.5049653053283691, + "learning_rate": 0.00023747758852532498, + "loss": 0.4695, + "step": 27340 + }, + { + "epoch": 1.5275861105679247, + "grad_norm": 0.7412675023078918, + "learning_rate": 0.00023744957418198119, + "loss": 0.4577, + "step": 27341 + }, + { + "epoch": 1.5276419811716067, + "grad_norm": 0.3931983411312103, + "learning_rate": 0.0002374215598386374, + "loss": 0.3911, + "step": 27342 + }, + { + "epoch": 1.5276978517752884, + "grad_norm": 0.323533833026886, + "learning_rate": 0.0002373935454952936, + "loss": 0.383, + "step": 27343 + }, + { + "epoch": 1.5277537223789703, + "grad_norm": 0.971325695514679, + "learning_rate": 0.0002373655311519498, + "loss": 0.4722, + "step": 27344 + }, + { + "epoch": 1.5278095929826523, + "grad_norm": 1.5397576093673706, + "learning_rate": 0.000237337516808606, + "loss": 0.3808, + "step": 27345 + }, + { + "epoch": 1.527865463586334, + "grad_norm": 0.35811102390289307, + "learning_rate": 0.00023730950246526222, + "loss": 0.3621, + "step": 27346 + }, + { + "epoch": 1.527921334190016, + "grad_norm": 0.7605679035186768, + "learning_rate": 0.00023728148812191845, + "loss": 0.3335, + "step": 27347 + }, + { + "epoch": 1.527977204793698, + "grad_norm": 0.32589849829673767, + "learning_rate": 0.00023725347377857465, + "loss": 0.3135, + "step": 27348 + }, + { + "epoch": 1.5280330753973796, + "grad_norm": 0.36841490864753723, + "learning_rate": 0.00023722545943523086, + "loss": 0.3998, + "step": 27349 + }, + { + "epoch": 1.5280889460010614, + "grad_norm": 0.4044669568538666, + "learning_rate": 0.00023719744509188704, + "loss": 0.4084, + "step": 27350 + }, + { + "epoch": 1.5281448166047435, + "grad_norm": 0.37612974643707275, + "learning_rate": 0.00023716943074854324, + "loss": 0.5037, + "step": 27351 + }, + { + "epoch": 1.5282006872084253, + "grad_norm": 0.5608786344528198, + "learning_rate": 0.00023714141640519945, + "loss": 0.3865, + "step": 27352 + }, + { + "epoch": 1.528256557812107, + "grad_norm": 0.46288275718688965, + "learning_rate": 0.00023711340206185566, + "loss": 0.3629, + "step": 27353 + }, + { + "epoch": 1.5283124284157892, + "grad_norm": 0.5720546245574951, + "learning_rate": 0.00023708538771851186, + "loss": 0.5229, + "step": 27354 + }, + { + "epoch": 1.528368299019471, + "grad_norm": 0.33582043647766113, + "learning_rate": 0.0002370573733751681, + "loss": 0.3471, + "step": 27355 + }, + { + "epoch": 1.5284241696231526, + "grad_norm": 0.3635227680206299, + "learning_rate": 0.0002370293590318243, + "loss": 0.4391, + "step": 27356 + }, + { + "epoch": 1.5284800402268348, + "grad_norm": 0.5867956280708313, + "learning_rate": 0.0002370013446884805, + "loss": 0.3139, + "step": 27357 + }, + { + "epoch": 1.5285359108305165, + "grad_norm": 2.9725468158721924, + "learning_rate": 0.0002369733303451367, + "loss": 0.3492, + "step": 27358 + }, + { + "epoch": 1.5285917814341983, + "grad_norm": 0.5167023539543152, + "learning_rate": 0.00023694531600179292, + "loss": 0.4745, + "step": 27359 + }, + { + "epoch": 1.5286476520378802, + "grad_norm": 0.4001023769378662, + "learning_rate": 0.00023691730165844912, + "loss": 0.4118, + "step": 27360 + }, + { + "epoch": 1.5287035226415622, + "grad_norm": 0.48002195358276367, + "learning_rate": 0.00023688928731510533, + "loss": 0.3178, + "step": 27361 + }, + { + "epoch": 1.528759393245244, + "grad_norm": 1.1133397817611694, + "learning_rate": 0.00023686127297176154, + "loss": 0.4236, + "step": 27362 + }, + { + "epoch": 1.5288152638489259, + "grad_norm": 0.3960559070110321, + "learning_rate": 0.00023683325862841777, + "loss": 0.396, + "step": 27363 + }, + { + "epoch": 1.5288711344526078, + "grad_norm": 0.3602199852466583, + "learning_rate": 0.00023680524428507397, + "loss": 0.3648, + "step": 27364 + }, + { + "epoch": 1.5289270050562895, + "grad_norm": 0.5824289321899414, + "learning_rate": 0.00023677722994173018, + "loss": 0.493, + "step": 27365 + }, + { + "epoch": 1.5289828756599715, + "grad_norm": 0.402782142162323, + "learning_rate": 0.00023674921559838639, + "loss": 0.3548, + "step": 27366 + }, + { + "epoch": 1.5290387462636534, + "grad_norm": 0.3671169579029083, + "learning_rate": 0.0002367212012550426, + "loss": 0.4472, + "step": 27367 + }, + { + "epoch": 1.5290946168673352, + "grad_norm": 0.687039315700531, + "learning_rate": 0.0002366931869116988, + "loss": 0.4338, + "step": 27368 + }, + { + "epoch": 1.5291504874710171, + "grad_norm": 0.7018346786499023, + "learning_rate": 0.000236665172568355, + "loss": 0.3757, + "step": 27369 + }, + { + "epoch": 1.529206358074699, + "grad_norm": 0.4044056534767151, + "learning_rate": 0.00023663715822501124, + "loss": 0.479, + "step": 27370 + }, + { + "epoch": 1.5292622286783808, + "grad_norm": 3.311774253845215, + "learning_rate": 0.00023660914388166744, + "loss": 0.4719, + "step": 27371 + }, + { + "epoch": 1.5293180992820627, + "grad_norm": 0.4448700547218323, + "learning_rate": 0.00023658112953832362, + "loss": 0.5002, + "step": 27372 + }, + { + "epoch": 1.5293739698857447, + "grad_norm": 0.4703613519668579, + "learning_rate": 0.00023655311519497983, + "loss": 0.4738, + "step": 27373 + }, + { + "epoch": 1.5294298404894264, + "grad_norm": 0.31173786520957947, + "learning_rate": 0.00023652510085163603, + "loss": 0.3567, + "step": 27374 + }, + { + "epoch": 1.5294857110931084, + "grad_norm": 0.5787282586097717, + "learning_rate": 0.00023649708650829224, + "loss": 0.4674, + "step": 27375 + }, + { + "epoch": 1.5295415816967903, + "grad_norm": 0.3723186254501343, + "learning_rate": 0.00023646907216494844, + "loss": 0.3711, + "step": 27376 + }, + { + "epoch": 1.529597452300472, + "grad_norm": 0.735572874546051, + "learning_rate": 0.00023644105782160465, + "loss": 0.5299, + "step": 27377 + }, + { + "epoch": 1.529653322904154, + "grad_norm": 0.36910638213157654, + "learning_rate": 0.00023641304347826088, + "loss": 0.4711, + "step": 27378 + }, + { + "epoch": 1.529709193507836, + "grad_norm": 0.5050048232078552, + "learning_rate": 0.0002363850291349171, + "loss": 0.4727, + "step": 27379 + }, + { + "epoch": 1.5297650641115177, + "grad_norm": 0.6158594489097595, + "learning_rate": 0.0002363570147915733, + "loss": 0.4326, + "step": 27380 + }, + { + "epoch": 1.5298209347151996, + "grad_norm": 0.37799951434135437, + "learning_rate": 0.0002363290004482295, + "loss": 0.3876, + "step": 27381 + }, + { + "epoch": 1.5298768053188816, + "grad_norm": 0.45996320247650146, + "learning_rate": 0.0002363009861048857, + "loss": 0.4096, + "step": 27382 + }, + { + "epoch": 1.5299326759225633, + "grad_norm": 0.39318376779556274, + "learning_rate": 0.0002362729717615419, + "loss": 0.4225, + "step": 27383 + }, + { + "epoch": 1.529988546526245, + "grad_norm": 3.304091453552246, + "learning_rate": 0.00023624495741819812, + "loss": 0.3455, + "step": 27384 + }, + { + "epoch": 1.5300444171299272, + "grad_norm": 0.5451069474220276, + "learning_rate": 0.00023621694307485432, + "loss": 0.3438, + "step": 27385 + }, + { + "epoch": 1.530100287733609, + "grad_norm": 0.7334466576576233, + "learning_rate": 0.00023618892873151056, + "loss": 0.3755, + "step": 27386 + }, + { + "epoch": 1.5301561583372907, + "grad_norm": 0.3504617512226105, + "learning_rate": 0.00023616091438816676, + "loss": 0.4517, + "step": 27387 + }, + { + "epoch": 1.5302120289409729, + "grad_norm": 0.44291990995407104, + "learning_rate": 0.00023613290004482297, + "loss": 0.421, + "step": 27388 + }, + { + "epoch": 1.5302678995446546, + "grad_norm": 0.47006988525390625, + "learning_rate": 0.00023610488570147917, + "loss": 0.549, + "step": 27389 + }, + { + "epoch": 1.5303237701483363, + "grad_norm": 0.3876210153102875, + "learning_rate": 0.00023607687135813538, + "loss": 0.3268, + "step": 27390 + }, + { + "epoch": 1.5303796407520185, + "grad_norm": 0.48824620246887207, + "learning_rate": 0.00023604885701479159, + "loss": 0.4173, + "step": 27391 + }, + { + "epoch": 1.5304355113557002, + "grad_norm": 0.3244960308074951, + "learning_rate": 0.0002360208426714478, + "loss": 0.3804, + "step": 27392 + }, + { + "epoch": 1.530491381959382, + "grad_norm": 0.3911602199077606, + "learning_rate": 0.00023599282832810397, + "loss": 0.3918, + "step": 27393 + }, + { + "epoch": 1.530547252563064, + "grad_norm": 10.666038513183594, + "learning_rate": 0.0002359648139847602, + "loss": 0.387, + "step": 27394 + }, + { + "epoch": 1.5306031231667458, + "grad_norm": 0.4663597047328949, + "learning_rate": 0.0002359367996414164, + "loss": 0.354, + "step": 27395 + }, + { + "epoch": 1.5306589937704276, + "grad_norm": 0.39560529589653015, + "learning_rate": 0.00023590878529807261, + "loss": 0.4699, + "step": 27396 + }, + { + "epoch": 1.5307148643741095, + "grad_norm": 0.44028550386428833, + "learning_rate": 0.00023588077095472882, + "loss": 0.4371, + "step": 27397 + }, + { + "epoch": 1.5307707349777915, + "grad_norm": 0.4705107808113098, + "learning_rate": 0.00023585275661138503, + "loss": 0.4011, + "step": 27398 + }, + { + "epoch": 1.5308266055814732, + "grad_norm": 0.37401288747787476, + "learning_rate": 0.00023582474226804123, + "loss": 0.3962, + "step": 27399 + }, + { + "epoch": 1.5308824761851552, + "grad_norm": 0.7094566226005554, + "learning_rate": 0.00023579672792469744, + "loss": 0.4385, + "step": 27400 + }, + { + "epoch": 1.5309383467888371, + "grad_norm": 1.0348548889160156, + "learning_rate": 0.00023576871358135364, + "loss": 0.4736, + "step": 27401 + }, + { + "epoch": 1.5309942173925188, + "grad_norm": 0.39111101627349854, + "learning_rate": 0.00023574069923800988, + "loss": 0.4088, + "step": 27402 + }, + { + "epoch": 1.5310500879962008, + "grad_norm": 0.5588521957397461, + "learning_rate": 0.00023571268489466608, + "loss": 0.524, + "step": 27403 + }, + { + "epoch": 1.5311059585998827, + "grad_norm": 0.36690229177474976, + "learning_rate": 0.0002356846705513223, + "loss": 0.3898, + "step": 27404 + }, + { + "epoch": 1.5311618292035645, + "grad_norm": 0.3586207926273346, + "learning_rate": 0.0002356566562079785, + "loss": 0.4286, + "step": 27405 + }, + { + "epoch": 1.5312176998072464, + "grad_norm": 0.45563575625419617, + "learning_rate": 0.0002356286418646347, + "loss": 0.5369, + "step": 27406 + }, + { + "epoch": 1.5312735704109284, + "grad_norm": 0.4706072509288788, + "learning_rate": 0.0002356006275212909, + "loss": 0.3579, + "step": 27407 + }, + { + "epoch": 1.53132944101461, + "grad_norm": 0.45935189723968506, + "learning_rate": 0.0002355726131779471, + "loss": 0.3315, + "step": 27408 + }, + { + "epoch": 1.531385311618292, + "grad_norm": 0.32796910405158997, + "learning_rate": 0.00023554459883460332, + "loss": 0.4214, + "step": 27409 + }, + { + "epoch": 1.531441182221974, + "grad_norm": 0.4664836823940277, + "learning_rate": 0.00023551658449125955, + "loss": 0.351, + "step": 27410 + }, + { + "epoch": 1.5314970528256557, + "grad_norm": 0.5364111661911011, + "learning_rate": 0.00023548857014791576, + "loss": 0.352, + "step": 27411 + }, + { + "epoch": 1.5315529234293377, + "grad_norm": 3.1562259197235107, + "learning_rate": 0.00023546055580457196, + "loss": 0.4077, + "step": 27412 + }, + { + "epoch": 1.5316087940330196, + "grad_norm": 0.33343079686164856, + "learning_rate": 0.00023543254146122817, + "loss": 0.4817, + "step": 27413 + }, + { + "epoch": 1.5316646646367014, + "grad_norm": 0.3561510145664215, + "learning_rate": 0.00023540452711788435, + "loss": 0.4451, + "step": 27414 + }, + { + "epoch": 1.5317205352403833, + "grad_norm": 0.4865967035293579, + "learning_rate": 0.00023537651277454055, + "loss": 0.3804, + "step": 27415 + }, + { + "epoch": 1.5317764058440653, + "grad_norm": 0.4242435097694397, + "learning_rate": 0.00023534849843119676, + "loss": 0.521, + "step": 27416 + }, + { + "epoch": 1.531832276447747, + "grad_norm": 0.41643038392066956, + "learning_rate": 0.00023532048408785296, + "loss": 0.451, + "step": 27417 + }, + { + "epoch": 1.5318881470514287, + "grad_norm": 0.4274826645851135, + "learning_rate": 0.0002352924697445092, + "loss": 0.4455, + "step": 27418 + }, + { + "epoch": 1.531944017655111, + "grad_norm": 0.36480361223220825, + "learning_rate": 0.0002352644554011654, + "loss": 0.3341, + "step": 27419 + }, + { + "epoch": 1.5319998882587926, + "grad_norm": 0.5687849521636963, + "learning_rate": 0.0002352364410578216, + "loss": 0.4437, + "step": 27420 + }, + { + "epoch": 1.5320557588624744, + "grad_norm": 0.3752278685569763, + "learning_rate": 0.00023520842671447781, + "loss": 0.3792, + "step": 27421 + }, + { + "epoch": 1.5321116294661565, + "grad_norm": 2.0016541481018066, + "learning_rate": 0.00023518041237113402, + "loss": 0.3689, + "step": 27422 + }, + { + "epoch": 1.5321675000698383, + "grad_norm": 0.43280458450317383, + "learning_rate": 0.00023515239802779023, + "loss": 0.4559, + "step": 27423 + }, + { + "epoch": 1.53222337067352, + "grad_norm": 2.5214788913726807, + "learning_rate": 0.00023512438368444643, + "loss": 0.5372, + "step": 27424 + }, + { + "epoch": 1.5322792412772022, + "grad_norm": 0.5882480144500732, + "learning_rate": 0.00023509636934110264, + "loss": 0.4011, + "step": 27425 + }, + { + "epoch": 1.532335111880884, + "grad_norm": 0.4555791914463043, + "learning_rate": 0.00023506835499775887, + "loss": 0.3741, + "step": 27426 + }, + { + "epoch": 1.5323909824845656, + "grad_norm": 3.234168529510498, + "learning_rate": 0.00023504034065441508, + "loss": 0.4314, + "step": 27427 + }, + { + "epoch": 1.5324468530882476, + "grad_norm": 0.3413326144218445, + "learning_rate": 0.00023501232631107128, + "loss": 0.3899, + "step": 27428 + }, + { + "epoch": 1.5325027236919295, + "grad_norm": 0.527764618396759, + "learning_rate": 0.0002349843119677275, + "loss": 0.4045, + "step": 27429 + }, + { + "epoch": 1.5325585942956113, + "grad_norm": 0.4015065133571625, + "learning_rate": 0.0002349562976243837, + "loss": 0.4071, + "step": 27430 + }, + { + "epoch": 1.5326144648992932, + "grad_norm": 0.3646553158760071, + "learning_rate": 0.0002349282832810399, + "loss": 0.4413, + "step": 27431 + }, + { + "epoch": 1.5326703355029752, + "grad_norm": 0.3845968246459961, + "learning_rate": 0.0002349002689376961, + "loss": 0.508, + "step": 27432 + }, + { + "epoch": 1.5327262061066569, + "grad_norm": 0.39861032366752625, + "learning_rate": 0.00023487225459435234, + "loss": 0.4588, + "step": 27433 + }, + { + "epoch": 1.5327820767103388, + "grad_norm": 2.75730037689209, + "learning_rate": 0.00023484424025100854, + "loss": 0.5303, + "step": 27434 + }, + { + "epoch": 1.5328379473140208, + "grad_norm": 2.6257734298706055, + "learning_rate": 0.00023481622590766472, + "loss": 0.4654, + "step": 27435 + }, + { + "epoch": 1.5328938179177025, + "grad_norm": 0.5221206545829773, + "learning_rate": 0.00023478821156432093, + "loss": 0.6659, + "step": 27436 + }, + { + "epoch": 1.5329496885213845, + "grad_norm": 0.6166841387748718, + "learning_rate": 0.00023476019722097713, + "loss": 0.4191, + "step": 27437 + }, + { + "epoch": 1.5330055591250664, + "grad_norm": 0.39675429463386536, + "learning_rate": 0.00023473218287763334, + "loss": 0.3873, + "step": 27438 + }, + { + "epoch": 1.5330614297287481, + "grad_norm": 0.37266433238983154, + "learning_rate": 0.00023470416853428955, + "loss": 0.4376, + "step": 27439 + }, + { + "epoch": 1.53311730033243, + "grad_norm": 0.3813243508338928, + "learning_rate": 0.00023467615419094575, + "loss": 0.4332, + "step": 27440 + }, + { + "epoch": 1.533173170936112, + "grad_norm": 0.3930411636829376, + "learning_rate": 0.00023464813984760198, + "loss": 0.5279, + "step": 27441 + }, + { + "epoch": 1.5332290415397938, + "grad_norm": 0.43105819821357727, + "learning_rate": 0.0002346201255042582, + "loss": 0.4131, + "step": 27442 + }, + { + "epoch": 1.5332849121434757, + "grad_norm": 0.4237760603427887, + "learning_rate": 0.0002345921111609144, + "loss": 0.4237, + "step": 27443 + }, + { + "epoch": 1.5333407827471577, + "grad_norm": 0.3138270974159241, + "learning_rate": 0.0002345640968175706, + "loss": 0.3726, + "step": 27444 + }, + { + "epoch": 1.5333966533508394, + "grad_norm": 1.620635986328125, + "learning_rate": 0.0002345360824742268, + "loss": 0.5398, + "step": 27445 + }, + { + "epoch": 1.5334525239545214, + "grad_norm": 0.4389100670814514, + "learning_rate": 0.00023450806813088301, + "loss": 0.3738, + "step": 27446 + }, + { + "epoch": 1.5335083945582033, + "grad_norm": 0.5327972769737244, + "learning_rate": 0.00023448005378753922, + "loss": 0.3472, + "step": 27447 + }, + { + "epoch": 1.533564265161885, + "grad_norm": 0.7254643440246582, + "learning_rate": 0.00023445203944419543, + "loss": 0.4292, + "step": 27448 + }, + { + "epoch": 1.533620135765567, + "grad_norm": 0.36029964685440063, + "learning_rate": 0.00023442402510085166, + "loss": 0.3558, + "step": 27449 + }, + { + "epoch": 1.533676006369249, + "grad_norm": 0.5378834009170532, + "learning_rate": 0.00023439601075750786, + "loss": 0.2783, + "step": 27450 + }, + { + "epoch": 1.5337318769729307, + "grad_norm": 0.4236155152320862, + "learning_rate": 0.00023436799641416407, + "loss": 0.4444, + "step": 27451 + }, + { + "epoch": 1.5337877475766124, + "grad_norm": 0.4755271375179291, + "learning_rate": 0.00023433998207082028, + "loss": 0.3945, + "step": 27452 + }, + { + "epoch": 1.5338436181802946, + "grad_norm": 0.321257621049881, + "learning_rate": 0.00023431196772747648, + "loss": 0.4042, + "step": 27453 + }, + { + "epoch": 1.5338994887839763, + "grad_norm": 0.5416561365127563, + "learning_rate": 0.0002342839533841327, + "loss": 0.4707, + "step": 27454 + }, + { + "epoch": 1.533955359387658, + "grad_norm": 0.5831932425498962, + "learning_rate": 0.0002342559390407889, + "loss": 0.4222, + "step": 27455 + }, + { + "epoch": 1.5340112299913402, + "grad_norm": 0.5861297845840454, + "learning_rate": 0.0002342279246974451, + "loss": 0.4615, + "step": 27456 + }, + { + "epoch": 1.534067100595022, + "grad_norm": 0.6165740489959717, + "learning_rate": 0.0002341999103541013, + "loss": 0.5604, + "step": 27457 + }, + { + "epoch": 1.5341229711987037, + "grad_norm": 0.5912654399871826, + "learning_rate": 0.0002341718960107575, + "loss": 0.6785, + "step": 27458 + }, + { + "epoch": 1.5341788418023858, + "grad_norm": 0.42080438137054443, + "learning_rate": 0.00023414388166741372, + "loss": 0.442, + "step": 27459 + }, + { + "epoch": 1.5342347124060676, + "grad_norm": 0.4513460397720337, + "learning_rate": 0.00023411586732406992, + "loss": 0.3088, + "step": 27460 + }, + { + "epoch": 1.5342905830097493, + "grad_norm": 1.7957381010055542, + "learning_rate": 0.00023408785298072613, + "loss": 0.3531, + "step": 27461 + }, + { + "epoch": 1.5343464536134312, + "grad_norm": 0.36487171053886414, + "learning_rate": 0.00023405983863738233, + "loss": 0.4525, + "step": 27462 + }, + { + "epoch": 1.5344023242171132, + "grad_norm": 0.9620201587677002, + "learning_rate": 0.00023403182429403854, + "loss": 0.3502, + "step": 27463 + }, + { + "epoch": 1.534458194820795, + "grad_norm": 0.4686712920665741, + "learning_rate": 0.00023400380995069475, + "loss": 0.6006, + "step": 27464 + }, + { + "epoch": 1.5345140654244769, + "grad_norm": 0.5589626431465149, + "learning_rate": 0.00023397579560735098, + "loss": 0.4329, + "step": 27465 + }, + { + "epoch": 1.5345699360281588, + "grad_norm": 0.985014021396637, + "learning_rate": 0.00023394778126400718, + "loss": 0.3801, + "step": 27466 + }, + { + "epoch": 1.5346258066318406, + "grad_norm": 0.4419470429420471, + "learning_rate": 0.0002339197669206634, + "loss": 0.5707, + "step": 27467 + }, + { + "epoch": 1.5346816772355225, + "grad_norm": 0.5171631574630737, + "learning_rate": 0.0002338917525773196, + "loss": 0.491, + "step": 27468 + }, + { + "epoch": 1.5347375478392045, + "grad_norm": 0.4629291296005249, + "learning_rate": 0.0002338637382339758, + "loss": 0.4065, + "step": 27469 + }, + { + "epoch": 1.5347934184428862, + "grad_norm": 0.39730697870254517, + "learning_rate": 0.000233835723890632, + "loss": 0.3685, + "step": 27470 + }, + { + "epoch": 1.5348492890465681, + "grad_norm": 0.4713270962238312, + "learning_rate": 0.0002338077095472882, + "loss": 0.3381, + "step": 27471 + }, + { + "epoch": 1.53490515965025, + "grad_norm": 0.5966078639030457, + "learning_rate": 0.00023377969520394442, + "loss": 0.3905, + "step": 27472 + }, + { + "epoch": 1.5349610302539318, + "grad_norm": 0.691716194152832, + "learning_rate": 0.00023375168086060065, + "loss": 0.4021, + "step": 27473 + }, + { + "epoch": 1.5350169008576138, + "grad_norm": 4.528242111206055, + "learning_rate": 0.00023372366651725686, + "loss": 0.5202, + "step": 27474 + }, + { + "epoch": 1.5350727714612957, + "grad_norm": 0.6732977032661438, + "learning_rate": 0.00023369565217391306, + "loss": 0.3311, + "step": 27475 + }, + { + "epoch": 1.5351286420649775, + "grad_norm": 0.5775501132011414, + "learning_rate": 0.00023366763783056927, + "loss": 0.3319, + "step": 27476 + }, + { + "epoch": 1.5351845126686594, + "grad_norm": 0.7006282806396484, + "learning_rate": 0.00023363962348722548, + "loss": 0.5661, + "step": 27477 + }, + { + "epoch": 1.5352403832723414, + "grad_norm": 0.3624022305011749, + "learning_rate": 0.00023361160914388165, + "loss": 0.4606, + "step": 27478 + }, + { + "epoch": 1.535296253876023, + "grad_norm": 0.580739438533783, + "learning_rate": 0.00023358359480053786, + "loss": 0.5301, + "step": 27479 + }, + { + "epoch": 1.535352124479705, + "grad_norm": 0.5690389275550842, + "learning_rate": 0.00023355558045719407, + "loss": 0.4312, + "step": 27480 + }, + { + "epoch": 1.535407995083387, + "grad_norm": 0.49294936656951904, + "learning_rate": 0.0002335275661138503, + "loss": 0.3262, + "step": 27481 + }, + { + "epoch": 1.5354638656870687, + "grad_norm": 0.4345843493938446, + "learning_rate": 0.0002334995517705065, + "loss": 0.3529, + "step": 27482 + }, + { + "epoch": 1.5355197362907507, + "grad_norm": 1.6802916526794434, + "learning_rate": 0.0002334715374271627, + "loss": 0.3439, + "step": 27483 + }, + { + "epoch": 1.5355756068944326, + "grad_norm": 1.2338780164718628, + "learning_rate": 0.00023344352308381892, + "loss": 0.4125, + "step": 27484 + }, + { + "epoch": 1.5356314774981144, + "grad_norm": 0.44626331329345703, + "learning_rate": 0.00023341550874047512, + "loss": 0.2875, + "step": 27485 + }, + { + "epoch": 1.535687348101796, + "grad_norm": 0.8931185007095337, + "learning_rate": 0.00023338749439713133, + "loss": 0.5286, + "step": 27486 + }, + { + "epoch": 1.5357432187054783, + "grad_norm": 0.2938445806503296, + "learning_rate": 0.00023335948005378753, + "loss": 0.4004, + "step": 27487 + }, + { + "epoch": 1.53579908930916, + "grad_norm": 0.8167943954467773, + "learning_rate": 0.00023333146571044374, + "loss": 0.4812, + "step": 27488 + }, + { + "epoch": 1.5358549599128417, + "grad_norm": 0.3010959029197693, + "learning_rate": 0.00023330345136709997, + "loss": 0.3194, + "step": 27489 + }, + { + "epoch": 1.5359108305165239, + "grad_norm": 0.3520642817020416, + "learning_rate": 0.00023327543702375618, + "loss": 0.3737, + "step": 27490 + }, + { + "epoch": 1.5359667011202056, + "grad_norm": 0.6298880577087402, + "learning_rate": 0.00023324742268041238, + "loss": 0.6498, + "step": 27491 + }, + { + "epoch": 1.5360225717238873, + "grad_norm": 0.4807088375091553, + "learning_rate": 0.0002332194083370686, + "loss": 0.4017, + "step": 27492 + }, + { + "epoch": 1.5360784423275695, + "grad_norm": 0.30987268686294556, + "learning_rate": 0.0002331913939937248, + "loss": 0.3779, + "step": 27493 + }, + { + "epoch": 1.5361343129312512, + "grad_norm": 0.8740503191947937, + "learning_rate": 0.000233163379650381, + "loss": 0.4175, + "step": 27494 + }, + { + "epoch": 1.536190183534933, + "grad_norm": 1.7666741609573364, + "learning_rate": 0.0002331353653070372, + "loss": 0.4497, + "step": 27495 + }, + { + "epoch": 1.536246054138615, + "grad_norm": 0.6066797971725464, + "learning_rate": 0.0002331073509636934, + "loss": 0.3332, + "step": 27496 + }, + { + "epoch": 1.5363019247422969, + "grad_norm": 0.42059358954429626, + "learning_rate": 0.00023307933662034965, + "loss": 0.3883, + "step": 27497 + }, + { + "epoch": 1.5363577953459786, + "grad_norm": 0.39016667008399963, + "learning_rate": 0.00023305132227700585, + "loss": 0.534, + "step": 27498 + }, + { + "epoch": 1.5364136659496606, + "grad_norm": 0.803303599357605, + "learning_rate": 0.00023302330793366203, + "loss": 0.4219, + "step": 27499 + }, + { + "epoch": 1.5364695365533425, + "grad_norm": 2.3216474056243896, + "learning_rate": 0.00023299529359031824, + "loss": 0.4705, + "step": 27500 + }, + { + "epoch": 1.5364695365533425, + "eval_cer": 0.08529453482155522, + "eval_loss": 0.3186691105365753, + "eval_runtime": 56.2951, + "eval_samples_per_second": 80.611, + "eval_steps_per_second": 5.045, + "eval_wer": 0.3382307756766571, + "step": 27500 + }, + { + "epoch": 1.5365254071570242, + "grad_norm": 0.494113028049469, + "learning_rate": 0.00023296727924697444, + "loss": 0.45, + "step": 27501 + }, + { + "epoch": 1.5365812777607062, + "grad_norm": 0.6037323474884033, + "learning_rate": 0.00023293926490363065, + "loss": 0.4606, + "step": 27502 + }, + { + "epoch": 1.5366371483643881, + "grad_norm": 0.4332275092601776, + "learning_rate": 0.00023291125056028685, + "loss": 0.4276, + "step": 27503 + }, + { + "epoch": 1.5366930189680699, + "grad_norm": 0.4101077616214752, + "learning_rate": 0.00023288323621694306, + "loss": 0.4414, + "step": 27504 + }, + { + "epoch": 1.5367488895717518, + "grad_norm": 0.44350945949554443, + "learning_rate": 0.0002328552218735993, + "loss": 0.4454, + "step": 27505 + }, + { + "epoch": 1.5368047601754338, + "grad_norm": 0.4070098102092743, + "learning_rate": 0.0002328272075302555, + "loss": 0.2896, + "step": 27506 + }, + { + "epoch": 1.5368606307791155, + "grad_norm": 0.38871029019355774, + "learning_rate": 0.0002327991931869117, + "loss": 0.3377, + "step": 27507 + }, + { + "epoch": 1.5369165013827975, + "grad_norm": 0.43424955010414124, + "learning_rate": 0.0002327711788435679, + "loss": 0.4379, + "step": 27508 + }, + { + "epoch": 1.5369723719864794, + "grad_norm": 0.4650273621082306, + "learning_rate": 0.00023274316450022412, + "loss": 0.4208, + "step": 27509 + }, + { + "epoch": 1.5370282425901611, + "grad_norm": 0.5060387849807739, + "learning_rate": 0.00023271515015688032, + "loss": 0.4661, + "step": 27510 + }, + { + "epoch": 1.537084113193843, + "grad_norm": 0.5224813222885132, + "learning_rate": 0.00023268713581353653, + "loss": 0.5605, + "step": 27511 + }, + { + "epoch": 1.537139983797525, + "grad_norm": 0.3879586160182953, + "learning_rate": 0.00023265912147019276, + "loss": 0.4414, + "step": 27512 + }, + { + "epoch": 1.5371958544012068, + "grad_norm": 0.6288626790046692, + "learning_rate": 0.00023263110712684897, + "loss": 0.6417, + "step": 27513 + }, + { + "epoch": 1.5372517250048887, + "grad_norm": 0.5006847381591797, + "learning_rate": 0.00023260309278350517, + "loss": 0.394, + "step": 27514 + }, + { + "epoch": 1.5373075956085707, + "grad_norm": 0.5906814932823181, + "learning_rate": 0.00023257507844016138, + "loss": 0.3928, + "step": 27515 + }, + { + "epoch": 1.5373634662122524, + "grad_norm": 0.2950955927371979, + "learning_rate": 0.00023254706409681758, + "loss": 0.3837, + "step": 27516 + }, + { + "epoch": 1.5374193368159343, + "grad_norm": 0.3769006133079529, + "learning_rate": 0.0002325190497534738, + "loss": 0.4137, + "step": 27517 + }, + { + "epoch": 1.5374752074196163, + "grad_norm": 0.5452934503555298, + "learning_rate": 0.00023249103541013, + "loss": 0.4628, + "step": 27518 + }, + { + "epoch": 1.537531078023298, + "grad_norm": 1.450360655784607, + "learning_rate": 0.0002324630210667862, + "loss": 0.4457, + "step": 27519 + }, + { + "epoch": 1.5375869486269798, + "grad_norm": 0.745002031326294, + "learning_rate": 0.0002324350067234424, + "loss": 0.735, + "step": 27520 + }, + { + "epoch": 1.537642819230662, + "grad_norm": 0.3759136199951172, + "learning_rate": 0.0002324069923800986, + "loss": 0.3622, + "step": 27521 + }, + { + "epoch": 1.5376986898343437, + "grad_norm": 0.30311641097068787, + "learning_rate": 0.00023237897803675482, + "loss": 0.3062, + "step": 27522 + }, + { + "epoch": 1.5377545604380254, + "grad_norm": 0.39120161533355713, + "learning_rate": 0.00023235096369341102, + "loss": 0.3279, + "step": 27523 + }, + { + "epoch": 1.5378104310417076, + "grad_norm": 0.4499354958534241, + "learning_rate": 0.00023232294935006723, + "loss": 0.3538, + "step": 27524 + }, + { + "epoch": 1.5378663016453893, + "grad_norm": 0.7148920893669128, + "learning_rate": 0.00023229493500672344, + "loss": 0.4703, + "step": 27525 + }, + { + "epoch": 1.537922172249071, + "grad_norm": 0.36438384652137756, + "learning_rate": 0.00023226692066337964, + "loss": 0.3443, + "step": 27526 + }, + { + "epoch": 1.537978042852753, + "grad_norm": 0.5129734873771667, + "learning_rate": 0.00023223890632003585, + "loss": 0.4511, + "step": 27527 + }, + { + "epoch": 1.538033913456435, + "grad_norm": 0.46238964796066284, + "learning_rate": 0.00023221089197669208, + "loss": 0.3983, + "step": 27528 + }, + { + "epoch": 1.5380897840601166, + "grad_norm": 0.40962615609169006, + "learning_rate": 0.00023218287763334829, + "loss": 0.6018, + "step": 27529 + }, + { + "epoch": 1.5381456546637986, + "grad_norm": 0.9188861846923828, + "learning_rate": 0.0002321548632900045, + "loss": 0.4701, + "step": 27530 + }, + { + "epoch": 1.5382015252674806, + "grad_norm": 1.358036994934082, + "learning_rate": 0.0002321268489466607, + "loss": 0.8115, + "step": 27531 + }, + { + "epoch": 1.5382573958711623, + "grad_norm": 0.4823645055294037, + "learning_rate": 0.0002320988346033169, + "loss": 0.4023, + "step": 27532 + }, + { + "epoch": 1.5383132664748442, + "grad_norm": 1.1705881357192993, + "learning_rate": 0.0002320708202599731, + "loss": 0.3216, + "step": 27533 + }, + { + "epoch": 1.5383691370785262, + "grad_norm": 0.5875759124755859, + "learning_rate": 0.00023204280591662932, + "loss": 0.476, + "step": 27534 + }, + { + "epoch": 1.538425007682208, + "grad_norm": 1.1825973987579346, + "learning_rate": 0.00023201479157328552, + "loss": 0.3518, + "step": 27535 + }, + { + "epoch": 1.5384808782858899, + "grad_norm": 0.5413728952407837, + "learning_rate": 0.00023198677722994175, + "loss": 0.4609, + "step": 27536 + }, + { + "epoch": 1.5385367488895718, + "grad_norm": 0.4377173185348511, + "learning_rate": 0.00023195876288659796, + "loss": 0.4251, + "step": 27537 + }, + { + "epoch": 1.5385926194932535, + "grad_norm": 0.5596523284912109, + "learning_rate": 0.00023193074854325417, + "loss": 0.4282, + "step": 27538 + }, + { + "epoch": 1.5386484900969355, + "grad_norm": 0.6493672728538513, + "learning_rate": 0.00023190273419991037, + "loss": 0.4273, + "step": 27539 + }, + { + "epoch": 1.5387043607006174, + "grad_norm": 0.7271520495414734, + "learning_rate": 0.00023187471985656658, + "loss": 0.4745, + "step": 27540 + }, + { + "epoch": 1.5387602313042992, + "grad_norm": 0.4018402397632599, + "learning_rate": 0.00023184670551322278, + "loss": 0.3271, + "step": 27541 + }, + { + "epoch": 1.5388161019079811, + "grad_norm": 0.4253070652484894, + "learning_rate": 0.00023181869116987896, + "loss": 0.5063, + "step": 27542 + }, + { + "epoch": 1.538871972511663, + "grad_norm": 0.36615094542503357, + "learning_rate": 0.00023179067682653517, + "loss": 0.4349, + "step": 27543 + }, + { + "epoch": 1.5389278431153448, + "grad_norm": 0.3087890148162842, + "learning_rate": 0.0002317626624831914, + "loss": 0.324, + "step": 27544 + }, + { + "epoch": 1.5389837137190268, + "grad_norm": 0.602391242980957, + "learning_rate": 0.0002317346481398476, + "loss": 0.4292, + "step": 27545 + }, + { + "epoch": 1.5390395843227087, + "grad_norm": 0.3364918828010559, + "learning_rate": 0.0002317066337965038, + "loss": 0.4219, + "step": 27546 + }, + { + "epoch": 1.5390954549263904, + "grad_norm": 0.5546260476112366, + "learning_rate": 0.00023167861945316002, + "loss": 0.4189, + "step": 27547 + }, + { + "epoch": 1.5391513255300724, + "grad_norm": 0.49278536438941956, + "learning_rate": 0.00023165060510981622, + "loss": 0.4604, + "step": 27548 + }, + { + "epoch": 1.5392071961337543, + "grad_norm": 0.5288643836975098, + "learning_rate": 0.00023162259076647243, + "loss": 0.4571, + "step": 27549 + }, + { + "epoch": 1.539263066737436, + "grad_norm": 0.4069465100765228, + "learning_rate": 0.00023159457642312864, + "loss": 0.4033, + "step": 27550 + }, + { + "epoch": 1.539318937341118, + "grad_norm": 0.33532825112342834, + "learning_rate": 0.00023156656207978484, + "loss": 0.293, + "step": 27551 + }, + { + "epoch": 1.5393748079448, + "grad_norm": 1.5887110233306885, + "learning_rate": 0.00023153854773644107, + "loss": 0.528, + "step": 27552 + }, + { + "epoch": 1.5394306785484817, + "grad_norm": 0.30992117524147034, + "learning_rate": 0.00023151053339309728, + "loss": 0.3385, + "step": 27553 + }, + { + "epoch": 1.5394865491521634, + "grad_norm": 0.3817209005355835, + "learning_rate": 0.00023148251904975349, + "loss": 0.3601, + "step": 27554 + }, + { + "epoch": 1.5395424197558456, + "grad_norm": 0.5966953039169312, + "learning_rate": 0.0002314545047064097, + "loss": 0.3668, + "step": 27555 + }, + { + "epoch": 1.5395982903595273, + "grad_norm": 0.4220139682292938, + "learning_rate": 0.0002314264903630659, + "loss": 0.412, + "step": 27556 + }, + { + "epoch": 1.539654160963209, + "grad_norm": 0.3914194107055664, + "learning_rate": 0.0002313984760197221, + "loss": 0.3784, + "step": 27557 + }, + { + "epoch": 1.5397100315668912, + "grad_norm": 0.5292800664901733, + "learning_rate": 0.0002313704616763783, + "loss": 0.4774, + "step": 27558 + }, + { + "epoch": 1.539765902170573, + "grad_norm": 2.194343090057373, + "learning_rate": 0.00023134244733303451, + "loss": 0.8257, + "step": 27559 + }, + { + "epoch": 1.5398217727742547, + "grad_norm": 1.1186237335205078, + "learning_rate": 0.00023131443298969075, + "loss": 0.4424, + "step": 27560 + }, + { + "epoch": 1.5398776433779366, + "grad_norm": 0.5960226058959961, + "learning_rate": 0.00023128641864634695, + "loss": 0.5502, + "step": 27561 + }, + { + "epoch": 1.5399335139816186, + "grad_norm": 0.6120216846466064, + "learning_rate": 0.00023125840430300316, + "loss": 0.4049, + "step": 27562 + }, + { + "epoch": 1.5399893845853003, + "grad_norm": 0.4798792600631714, + "learning_rate": 0.00023123038995965934, + "loss": 0.443, + "step": 27563 + }, + { + "epoch": 1.5400452551889823, + "grad_norm": 0.574873685836792, + "learning_rate": 0.00023120237561631554, + "loss": 0.4274, + "step": 27564 + }, + { + "epoch": 1.5401011257926642, + "grad_norm": 0.3942379057407379, + "learning_rate": 0.00023117436127297175, + "loss": 0.4628, + "step": 27565 + }, + { + "epoch": 1.540156996396346, + "grad_norm": 1.1070128679275513, + "learning_rate": 0.00023114634692962796, + "loss": 0.3515, + "step": 27566 + }, + { + "epoch": 1.540212867000028, + "grad_norm": 1.5602879524230957, + "learning_rate": 0.00023111833258628416, + "loss": 0.3859, + "step": 27567 + }, + { + "epoch": 1.5402687376037099, + "grad_norm": 0.5834882855415344, + "learning_rate": 0.0002310903182429404, + "loss": 0.5184, + "step": 27568 + }, + { + "epoch": 1.5403246082073916, + "grad_norm": 0.7957922220230103, + "learning_rate": 0.0002310623038995966, + "loss": 0.6139, + "step": 27569 + }, + { + "epoch": 1.5403804788110735, + "grad_norm": 0.4530293941497803, + "learning_rate": 0.0002310342895562528, + "loss": 0.5869, + "step": 27570 + }, + { + "epoch": 1.5404363494147555, + "grad_norm": 0.40378686785697937, + "learning_rate": 0.000231006275212909, + "loss": 0.3247, + "step": 27571 + }, + { + "epoch": 1.5404922200184372, + "grad_norm": 0.43189945816993713, + "learning_rate": 0.00023097826086956522, + "loss": 0.448, + "step": 27572 + }, + { + "epoch": 1.5405480906221192, + "grad_norm": 0.47544360160827637, + "learning_rate": 0.00023095024652622142, + "loss": 0.4523, + "step": 27573 + }, + { + "epoch": 1.5406039612258011, + "grad_norm": 0.37102705240249634, + "learning_rate": 0.00023092223218287763, + "loss": 0.4633, + "step": 27574 + }, + { + "epoch": 1.5406598318294829, + "grad_norm": 0.42863553762435913, + "learning_rate": 0.00023089421783953386, + "loss": 0.3893, + "step": 27575 + }, + { + "epoch": 1.5407157024331648, + "grad_norm": 1.0332551002502441, + "learning_rate": 0.00023086620349619007, + "loss": 0.3717, + "step": 27576 + }, + { + "epoch": 1.5407715730368468, + "grad_norm": 1.6335337162017822, + "learning_rate": 0.00023083818915284627, + "loss": 0.5059, + "step": 27577 + }, + { + "epoch": 1.5408274436405285, + "grad_norm": 0.4141237139701843, + "learning_rate": 0.00023081017480950248, + "loss": 0.4117, + "step": 27578 + }, + { + "epoch": 1.5408833142442104, + "grad_norm": 0.5088288187980652, + "learning_rate": 0.00023078216046615869, + "loss": 0.3895, + "step": 27579 + }, + { + "epoch": 1.5409391848478924, + "grad_norm": 0.5090013146400452, + "learning_rate": 0.0002307541461228149, + "loss": 0.5296, + "step": 27580 + }, + { + "epoch": 1.5409950554515741, + "grad_norm": 0.3680362105369568, + "learning_rate": 0.0002307261317794711, + "loss": 0.3428, + "step": 27581 + }, + { + "epoch": 1.541050926055256, + "grad_norm": 0.4031364321708679, + "learning_rate": 0.0002306981174361273, + "loss": 0.3477, + "step": 27582 + }, + { + "epoch": 1.541106796658938, + "grad_norm": 0.36048513650894165, + "learning_rate": 0.00023067010309278354, + "loss": 0.3391, + "step": 27583 + }, + { + "epoch": 1.5411626672626197, + "grad_norm": 0.30398234724998474, + "learning_rate": 0.00023064208874943971, + "loss": 0.3426, + "step": 27584 + }, + { + "epoch": 1.5412185378663017, + "grad_norm": 0.5259988903999329, + "learning_rate": 0.00023061407440609592, + "loss": 0.4002, + "step": 27585 + }, + { + "epoch": 1.5412744084699836, + "grad_norm": 12.595959663391113, + "learning_rate": 0.00023058606006275213, + "loss": 0.4719, + "step": 27586 + }, + { + "epoch": 1.5413302790736654, + "grad_norm": 1.9347001314163208, + "learning_rate": 0.00023055804571940833, + "loss": 0.4052, + "step": 27587 + }, + { + "epoch": 1.541386149677347, + "grad_norm": 0.7816470265388489, + "learning_rate": 0.00023053003137606454, + "loss": 0.3659, + "step": 27588 + }, + { + "epoch": 1.5414420202810293, + "grad_norm": 0.35352519154548645, + "learning_rate": 0.00023050201703272074, + "loss": 0.3901, + "step": 27589 + }, + { + "epoch": 1.541497890884711, + "grad_norm": 0.493817001581192, + "learning_rate": 0.00023047400268937695, + "loss": 0.4298, + "step": 27590 + }, + { + "epoch": 1.5415537614883927, + "grad_norm": 0.7851975560188293, + "learning_rate": 0.00023044598834603318, + "loss": 0.4057, + "step": 27591 + }, + { + "epoch": 1.541609632092075, + "grad_norm": 0.5500963926315308, + "learning_rate": 0.0002304179740026894, + "loss": 0.4374, + "step": 27592 + }, + { + "epoch": 1.5416655026957566, + "grad_norm": 0.3936786651611328, + "learning_rate": 0.0002303899596593456, + "loss": 0.473, + "step": 27593 + }, + { + "epoch": 1.5417213732994384, + "grad_norm": 0.5077191591262817, + "learning_rate": 0.0002303619453160018, + "loss": 0.4529, + "step": 27594 + }, + { + "epoch": 1.5417772439031203, + "grad_norm": 0.4299165904521942, + "learning_rate": 0.000230333930972658, + "loss": 0.2952, + "step": 27595 + }, + { + "epoch": 1.5418331145068023, + "grad_norm": 0.3848229944705963, + "learning_rate": 0.0002303059166293142, + "loss": 0.4143, + "step": 27596 + }, + { + "epoch": 1.541888985110484, + "grad_norm": 0.733787477016449, + "learning_rate": 0.00023027790228597042, + "loss": 0.4074, + "step": 27597 + }, + { + "epoch": 1.541944855714166, + "grad_norm": 0.7014439105987549, + "learning_rate": 0.00023024988794262662, + "loss": 0.5762, + "step": 27598 + }, + { + "epoch": 1.542000726317848, + "grad_norm": 0.9976742267608643, + "learning_rate": 0.00023022187359928286, + "loss": 0.3973, + "step": 27599 + }, + { + "epoch": 1.5420565969215296, + "grad_norm": 0.46945247054100037, + "learning_rate": 0.00023019385925593906, + "loss": 0.4933, + "step": 27600 + }, + { + "epoch": 1.5421124675252116, + "grad_norm": 0.5064682364463806, + "learning_rate": 0.00023016584491259527, + "loss": 0.3958, + "step": 27601 + }, + { + "epoch": 1.5421683381288935, + "grad_norm": 0.33980628848075867, + "learning_rate": 0.00023013783056925147, + "loss": 0.3833, + "step": 27602 + }, + { + "epoch": 1.5422242087325753, + "grad_norm": 0.6344878077507019, + "learning_rate": 0.00023010981622590768, + "loss": 0.3749, + "step": 27603 + }, + { + "epoch": 1.5422800793362572, + "grad_norm": 0.5433211326599121, + "learning_rate": 0.00023008180188256388, + "loss": 0.5286, + "step": 27604 + }, + { + "epoch": 1.5423359499399392, + "grad_norm": 0.47659093141555786, + "learning_rate": 0.00023005378753922006, + "loss": 0.4234, + "step": 27605 + }, + { + "epoch": 1.542391820543621, + "grad_norm": 0.5995973944664001, + "learning_rate": 0.00023002577319587627, + "loss": 0.408, + "step": 27606 + }, + { + "epoch": 1.5424476911473028, + "grad_norm": 0.5446817874908447, + "learning_rate": 0.0002299977588525325, + "loss": 0.4656, + "step": 27607 + }, + { + "epoch": 1.5425035617509848, + "grad_norm": 0.4610634446144104, + "learning_rate": 0.0002299697445091887, + "loss": 0.4334, + "step": 27608 + }, + { + "epoch": 1.5425594323546665, + "grad_norm": 0.4370999336242676, + "learning_rate": 0.00022994173016584491, + "loss": 0.4005, + "step": 27609 + }, + { + "epoch": 1.5426153029583485, + "grad_norm": 0.4844972491264343, + "learning_rate": 0.00022991371582250112, + "loss": 0.397, + "step": 27610 + }, + { + "epoch": 1.5426711735620304, + "grad_norm": 0.510716438293457, + "learning_rate": 0.00022988570147915733, + "loss": 0.3759, + "step": 27611 + }, + { + "epoch": 1.5427270441657122, + "grad_norm": 0.45736414194107056, + "learning_rate": 0.00022985768713581353, + "loss": 0.4851, + "step": 27612 + }, + { + "epoch": 1.542782914769394, + "grad_norm": 0.42386987805366516, + "learning_rate": 0.00022982967279246974, + "loss": 0.3695, + "step": 27613 + }, + { + "epoch": 1.542838785373076, + "grad_norm": 0.46335387229919434, + "learning_rate": 0.00022980165844912594, + "loss": 0.3744, + "step": 27614 + }, + { + "epoch": 1.5428946559767578, + "grad_norm": 0.7516588568687439, + "learning_rate": 0.00022977364410578218, + "loss": 0.357, + "step": 27615 + }, + { + "epoch": 1.5429505265804397, + "grad_norm": 0.4699917137622833, + "learning_rate": 0.00022974562976243838, + "loss": 0.3979, + "step": 27616 + }, + { + "epoch": 1.5430063971841217, + "grad_norm": 0.41228529810905457, + "learning_rate": 0.0002297176154190946, + "loss": 0.4426, + "step": 27617 + }, + { + "epoch": 1.5430622677878034, + "grad_norm": 0.4842509627342224, + "learning_rate": 0.0002296896010757508, + "loss": 0.4068, + "step": 27618 + }, + { + "epoch": 1.5431181383914854, + "grad_norm": 1.5057295560836792, + "learning_rate": 0.000229661586732407, + "loss": 0.4158, + "step": 27619 + }, + { + "epoch": 1.5431740089951673, + "grad_norm": 0.5585851669311523, + "learning_rate": 0.0002296335723890632, + "loss": 0.4594, + "step": 27620 + }, + { + "epoch": 1.543229879598849, + "grad_norm": 2.392158269882202, + "learning_rate": 0.0002296055580457194, + "loss": 0.408, + "step": 27621 + }, + { + "epoch": 1.5432857502025308, + "grad_norm": 0.6300001740455627, + "learning_rate": 0.00022957754370237562, + "loss": 0.3222, + "step": 27622 + }, + { + "epoch": 1.543341620806213, + "grad_norm": 0.4549967348575592, + "learning_rate": 0.00022954952935903185, + "loss": 0.3762, + "step": 27623 + }, + { + "epoch": 1.5433974914098947, + "grad_norm": 0.4562463164329529, + "learning_rate": 0.00022952151501568806, + "loss": 0.4095, + "step": 27624 + }, + { + "epoch": 1.5434533620135764, + "grad_norm": 0.408584862947464, + "learning_rate": 0.00022949350067234426, + "loss": 0.3424, + "step": 27625 + }, + { + "epoch": 1.5435092326172586, + "grad_norm": 0.5869095921516418, + "learning_rate": 0.00022946548632900047, + "loss": 0.4352, + "step": 27626 + }, + { + "epoch": 1.5435651032209403, + "grad_norm": 0.5660213232040405, + "learning_rate": 0.00022943747198565665, + "loss": 0.5764, + "step": 27627 + }, + { + "epoch": 1.543620973824622, + "grad_norm": 0.42193305492401123, + "learning_rate": 0.00022940945764231285, + "loss": 0.4136, + "step": 27628 + }, + { + "epoch": 1.543676844428304, + "grad_norm": 0.4202708601951599, + "learning_rate": 0.00022938144329896906, + "loss": 0.3867, + "step": 27629 + }, + { + "epoch": 1.543732715031986, + "grad_norm": 0.637697160243988, + "learning_rate": 0.00022935342895562526, + "loss": 0.5013, + "step": 27630 + }, + { + "epoch": 1.5437885856356677, + "grad_norm": 0.703191876411438, + "learning_rate": 0.0002293254146122815, + "loss": 0.3827, + "step": 27631 + }, + { + "epoch": 1.5438444562393496, + "grad_norm": 2.306234359741211, + "learning_rate": 0.0002292974002689377, + "loss": 0.3819, + "step": 27632 + }, + { + "epoch": 1.5439003268430316, + "grad_norm": 0.36323845386505127, + "learning_rate": 0.0002292693859255939, + "loss": 0.3622, + "step": 27633 + }, + { + "epoch": 1.5439561974467133, + "grad_norm": 0.4840041697025299, + "learning_rate": 0.00022924137158225011, + "loss": 0.5368, + "step": 27634 + }, + { + "epoch": 1.5440120680503953, + "grad_norm": 0.6467951536178589, + "learning_rate": 0.00022921335723890632, + "loss": 0.411, + "step": 27635 + }, + { + "epoch": 1.5440679386540772, + "grad_norm": 0.29317033290863037, + "learning_rate": 0.00022918534289556253, + "loss": 0.3746, + "step": 27636 + }, + { + "epoch": 1.544123809257759, + "grad_norm": 0.3975266218185425, + "learning_rate": 0.00022915732855221873, + "loss": 0.404, + "step": 27637 + }, + { + "epoch": 1.544179679861441, + "grad_norm": 0.4187052249908447, + "learning_rate": 0.00022912931420887496, + "loss": 0.4098, + "step": 27638 + }, + { + "epoch": 1.5442355504651228, + "grad_norm": 0.42089951038360596, + "learning_rate": 0.00022910129986553117, + "loss": 0.3745, + "step": 27639 + }, + { + "epoch": 1.5442914210688046, + "grad_norm": 0.7868187427520752, + "learning_rate": 0.00022907328552218738, + "loss": 0.4206, + "step": 27640 + }, + { + "epoch": 1.5443472916724865, + "grad_norm": 0.36734479665756226, + "learning_rate": 0.00022904527117884358, + "loss": 0.3634, + "step": 27641 + }, + { + "epoch": 1.5444031622761685, + "grad_norm": 0.333950936794281, + "learning_rate": 0.0002290172568354998, + "loss": 0.3941, + "step": 27642 + }, + { + "epoch": 1.5444590328798502, + "grad_norm": 0.3796703517436981, + "learning_rate": 0.000228989242492156, + "loss": 0.4969, + "step": 27643 + }, + { + "epoch": 1.5445149034835322, + "grad_norm": 0.3857804238796234, + "learning_rate": 0.0002289612281488122, + "loss": 0.5309, + "step": 27644 + }, + { + "epoch": 1.544570774087214, + "grad_norm": 0.5112529397010803, + "learning_rate": 0.0002289332138054684, + "loss": 0.444, + "step": 27645 + }, + { + "epoch": 1.5446266446908958, + "grad_norm": 0.31150326132774353, + "learning_rate": 0.00022890519946212464, + "loss": 0.3767, + "step": 27646 + }, + { + "epoch": 1.5446825152945778, + "grad_norm": 0.43337300419807434, + "learning_rate": 0.00022887718511878084, + "loss": 0.5163, + "step": 27647 + }, + { + "epoch": 1.5447383858982597, + "grad_norm": 0.5081244707107544, + "learning_rate": 0.00022884917077543702, + "loss": 0.4513, + "step": 27648 + }, + { + "epoch": 1.5447942565019415, + "grad_norm": 0.8009651899337769, + "learning_rate": 0.00022882115643209323, + "loss": 0.482, + "step": 27649 + }, + { + "epoch": 1.5448501271056234, + "grad_norm": 0.38049161434173584, + "learning_rate": 0.00022879314208874943, + "loss": 0.3926, + "step": 27650 + }, + { + "epoch": 1.5449059977093054, + "grad_norm": 0.3877321779727936, + "learning_rate": 0.00022876512774540564, + "loss": 0.404, + "step": 27651 + }, + { + "epoch": 1.544961868312987, + "grad_norm": 0.4820086658000946, + "learning_rate": 0.00022873711340206185, + "loss": 0.418, + "step": 27652 + }, + { + "epoch": 1.545017738916669, + "grad_norm": 0.41081708669662476, + "learning_rate": 0.00022870909905871805, + "loss": 0.3226, + "step": 27653 + }, + { + "epoch": 1.545073609520351, + "grad_norm": 0.4198032319545746, + "learning_rate": 0.00022868108471537428, + "loss": 0.4859, + "step": 27654 + }, + { + "epoch": 1.5451294801240327, + "grad_norm": 0.5313719511032104, + "learning_rate": 0.0002286530703720305, + "loss": 0.4422, + "step": 27655 + }, + { + "epoch": 1.5451853507277145, + "grad_norm": 0.47633010149002075, + "learning_rate": 0.0002286250560286867, + "loss": 0.4401, + "step": 27656 + }, + { + "epoch": 1.5452412213313966, + "grad_norm": 2.8420212268829346, + "learning_rate": 0.0002285970416853429, + "loss": 0.3786, + "step": 27657 + }, + { + "epoch": 1.5452970919350784, + "grad_norm": 0.3094244599342346, + "learning_rate": 0.0002285690273419991, + "loss": 0.3578, + "step": 27658 + }, + { + "epoch": 1.54535296253876, + "grad_norm": 0.32181575894355774, + "learning_rate": 0.0002285410129986553, + "loss": 0.3256, + "step": 27659 + }, + { + "epoch": 1.5454088331424423, + "grad_norm": 0.8276990056037903, + "learning_rate": 0.00022851299865531152, + "loss": 0.4382, + "step": 27660 + }, + { + "epoch": 1.545464703746124, + "grad_norm": 0.4770498275756836, + "learning_rate": 0.00022848498431196772, + "loss": 0.3783, + "step": 27661 + }, + { + "epoch": 1.5455205743498057, + "grad_norm": 0.46266502141952515, + "learning_rate": 0.00022845696996862396, + "loss": 0.4159, + "step": 27662 + }, + { + "epoch": 1.5455764449534877, + "grad_norm": 0.4825572371482849, + "learning_rate": 0.00022842895562528016, + "loss": 0.4075, + "step": 27663 + }, + { + "epoch": 1.5456323155571696, + "grad_norm": 0.5578285455703735, + "learning_rate": 0.00022840094128193637, + "loss": 0.4296, + "step": 27664 + }, + { + "epoch": 1.5456881861608514, + "grad_norm": 0.32182633876800537, + "learning_rate": 0.00022837292693859258, + "loss": 0.3158, + "step": 27665 + }, + { + "epoch": 1.5457440567645333, + "grad_norm": 0.32534730434417725, + "learning_rate": 0.00022834491259524878, + "loss": 0.4015, + "step": 27666 + }, + { + "epoch": 1.5457999273682153, + "grad_norm": 0.33604684472084045, + "learning_rate": 0.000228316898251905, + "loss": 0.38, + "step": 27667 + }, + { + "epoch": 1.545855797971897, + "grad_norm": 0.4227394759654999, + "learning_rate": 0.0002282888839085612, + "loss": 0.3945, + "step": 27668 + }, + { + "epoch": 1.545911668575579, + "grad_norm": 0.39972755312919617, + "learning_rate": 0.00022826086956521737, + "loss": 0.4285, + "step": 27669 + }, + { + "epoch": 1.5459675391792609, + "grad_norm": 0.7499232292175293, + "learning_rate": 0.0002282328552218736, + "loss": 0.3222, + "step": 27670 + }, + { + "epoch": 1.5460234097829426, + "grad_norm": 0.4011319875717163, + "learning_rate": 0.0002282048408785298, + "loss": 0.3618, + "step": 27671 + }, + { + "epoch": 1.5460792803866246, + "grad_norm": 0.581159770488739, + "learning_rate": 0.00022817682653518602, + "loss": 0.462, + "step": 27672 + }, + { + "epoch": 1.5461351509903065, + "grad_norm": 0.519847571849823, + "learning_rate": 0.00022814881219184222, + "loss": 0.3695, + "step": 27673 + }, + { + "epoch": 1.5461910215939882, + "grad_norm": 0.6838935613632202, + "learning_rate": 0.00022812079784849843, + "loss": 0.3954, + "step": 27674 + }, + { + "epoch": 1.5462468921976702, + "grad_norm": 0.43074238300323486, + "learning_rate": 0.00022809278350515463, + "loss": 0.4059, + "step": 27675 + }, + { + "epoch": 1.5463027628013521, + "grad_norm": 7.66457462310791, + "learning_rate": 0.00022806476916181084, + "loss": 0.3276, + "step": 27676 + }, + { + "epoch": 1.5463586334050339, + "grad_norm": 0.5757789611816406, + "learning_rate": 0.00022803675481846704, + "loss": 0.5581, + "step": 27677 + }, + { + "epoch": 1.5464145040087158, + "grad_norm": 0.37280550599098206, + "learning_rate": 0.00022800874047512328, + "loss": 0.4521, + "step": 27678 + }, + { + "epoch": 1.5464703746123978, + "grad_norm": 0.35531356930732727, + "learning_rate": 0.00022798072613177948, + "loss": 0.3232, + "step": 27679 + }, + { + "epoch": 1.5465262452160795, + "grad_norm": 3.5133488178253174, + "learning_rate": 0.0002279527117884357, + "loss": 0.3731, + "step": 27680 + }, + { + "epoch": 1.5465821158197615, + "grad_norm": 0.3921346366405487, + "learning_rate": 0.0002279246974450919, + "loss": 0.3214, + "step": 27681 + }, + { + "epoch": 1.5466379864234434, + "grad_norm": 0.3559415340423584, + "learning_rate": 0.0002278966831017481, + "loss": 0.3576, + "step": 27682 + }, + { + "epoch": 1.5466938570271251, + "grad_norm": 2.2182154655456543, + "learning_rate": 0.0002278686687584043, + "loss": 0.444, + "step": 27683 + }, + { + "epoch": 1.546749727630807, + "grad_norm": 0.5462942719459534, + "learning_rate": 0.0002278406544150605, + "loss": 0.3877, + "step": 27684 + }, + { + "epoch": 1.546805598234489, + "grad_norm": 0.39241528511047363, + "learning_rate": 0.00022781264007171672, + "loss": 0.4171, + "step": 27685 + }, + { + "epoch": 1.5468614688381708, + "grad_norm": 0.6368483901023865, + "learning_rate": 0.00022778462572837295, + "loss": 0.4438, + "step": 27686 + }, + { + "epoch": 1.5469173394418527, + "grad_norm": 0.376973420381546, + "learning_rate": 0.00022775661138502916, + "loss": 0.4217, + "step": 27687 + }, + { + "epoch": 1.5469732100455347, + "grad_norm": 0.41598036885261536, + "learning_rate": 0.00022772859704168536, + "loss": 0.527, + "step": 27688 + }, + { + "epoch": 1.5470290806492164, + "grad_norm": 0.9523904919624329, + "learning_rate": 0.00022770058269834157, + "loss": 0.4572, + "step": 27689 + }, + { + "epoch": 1.5470849512528981, + "grad_norm": 0.5732561349868774, + "learning_rate": 0.00022767256835499775, + "loss": 0.4674, + "step": 27690 + }, + { + "epoch": 1.5471408218565803, + "grad_norm": 0.959956705570221, + "learning_rate": 0.00022764455401165395, + "loss": 0.4186, + "step": 27691 + }, + { + "epoch": 1.547196692460262, + "grad_norm": 0.4005146622657776, + "learning_rate": 0.00022761653966831016, + "loss": 0.3507, + "step": 27692 + }, + { + "epoch": 1.5472525630639438, + "grad_norm": 0.35791510343551636, + "learning_rate": 0.00022758852532496637, + "loss": 0.3722, + "step": 27693 + }, + { + "epoch": 1.547308433667626, + "grad_norm": 0.3763650953769684, + "learning_rate": 0.0002275605109816226, + "loss": 0.427, + "step": 27694 + }, + { + "epoch": 1.5473643042713077, + "grad_norm": 0.3474894165992737, + "learning_rate": 0.0002275324966382788, + "loss": 0.4943, + "step": 27695 + }, + { + "epoch": 1.5474201748749894, + "grad_norm": 0.36806520819664, + "learning_rate": 0.000227504482294935, + "loss": 0.4071, + "step": 27696 + }, + { + "epoch": 1.5474760454786713, + "grad_norm": 0.5212587118148804, + "learning_rate": 0.00022747646795159122, + "loss": 0.3272, + "step": 27697 + }, + { + "epoch": 1.5475319160823533, + "grad_norm": 0.35121914744377136, + "learning_rate": 0.00022744845360824742, + "loss": 0.3781, + "step": 27698 + }, + { + "epoch": 1.547587786686035, + "grad_norm": 0.463652104139328, + "learning_rate": 0.00022742043926490363, + "loss": 0.3998, + "step": 27699 + }, + { + "epoch": 1.547643657289717, + "grad_norm": 0.392190545797348, + "learning_rate": 0.00022739242492155983, + "loss": 0.4285, + "step": 27700 + }, + { + "epoch": 1.547699527893399, + "grad_norm": 0.8651539087295532, + "learning_rate": 0.00022736441057821604, + "loss": 0.4428, + "step": 27701 + }, + { + "epoch": 1.5477553984970807, + "grad_norm": 16.56949234008789, + "learning_rate": 0.00022733639623487227, + "loss": 0.4866, + "step": 27702 + }, + { + "epoch": 1.5478112691007626, + "grad_norm": 0.3296395540237427, + "learning_rate": 0.00022730838189152848, + "loss": 0.3574, + "step": 27703 + }, + { + "epoch": 1.5478671397044446, + "grad_norm": 0.41527628898620605, + "learning_rate": 0.00022728036754818468, + "loss": 0.2959, + "step": 27704 + }, + { + "epoch": 1.5479230103081263, + "grad_norm": 0.37894049286842346, + "learning_rate": 0.0002272523532048409, + "loss": 0.4082, + "step": 27705 + }, + { + "epoch": 1.5479788809118082, + "grad_norm": 0.4484429955482483, + "learning_rate": 0.0002272243388614971, + "loss": 0.3911, + "step": 27706 + }, + { + "epoch": 1.5480347515154902, + "grad_norm": 0.3509270250797272, + "learning_rate": 0.0002271963245181533, + "loss": 0.3828, + "step": 27707 + }, + { + "epoch": 1.548090622119172, + "grad_norm": 0.7783543467521667, + "learning_rate": 0.0002271683101748095, + "loss": 0.3911, + "step": 27708 + }, + { + "epoch": 1.5481464927228539, + "grad_norm": 0.38224613666534424, + "learning_rate": 0.00022714029583146574, + "loss": 0.3586, + "step": 27709 + }, + { + "epoch": 1.5482023633265358, + "grad_norm": 0.5029722452163696, + "learning_rate": 0.00022711228148812195, + "loss": 0.5038, + "step": 27710 + }, + { + "epoch": 1.5482582339302176, + "grad_norm": 0.8338936567306519, + "learning_rate": 0.00022708426714477815, + "loss": 0.4618, + "step": 27711 + }, + { + "epoch": 1.5483141045338995, + "grad_norm": 1.0079330205917358, + "learning_rate": 0.00022705625280143433, + "loss": 0.343, + "step": 27712 + }, + { + "epoch": 1.5483699751375815, + "grad_norm": 0.4418821930885315, + "learning_rate": 0.00022702823845809054, + "loss": 0.4692, + "step": 27713 + }, + { + "epoch": 1.5484258457412632, + "grad_norm": 0.38807713985443115, + "learning_rate": 0.00022700022411474674, + "loss": 0.4505, + "step": 27714 + }, + { + "epoch": 1.5484817163449451, + "grad_norm": 0.6811273694038391, + "learning_rate": 0.00022697220977140295, + "loss": 0.5198, + "step": 27715 + }, + { + "epoch": 1.548537586948627, + "grad_norm": 0.4975146949291229, + "learning_rate": 0.00022694419542805915, + "loss": 0.4518, + "step": 27716 + }, + { + "epoch": 1.5485934575523088, + "grad_norm": 0.4368152618408203, + "learning_rate": 0.00022691618108471539, + "loss": 0.4284, + "step": 27717 + }, + { + "epoch": 1.5486493281559908, + "grad_norm": 0.48681578040122986, + "learning_rate": 0.0002268881667413716, + "loss": 0.4248, + "step": 27718 + }, + { + "epoch": 1.5487051987596727, + "grad_norm": 0.8182348012924194, + "learning_rate": 0.0002268601523980278, + "loss": 0.3539, + "step": 27719 + }, + { + "epoch": 1.5487610693633544, + "grad_norm": 0.402319997549057, + "learning_rate": 0.000226832138054684, + "loss": 0.49, + "step": 27720 + }, + { + "epoch": 1.5488169399670364, + "grad_norm": 0.5039536952972412, + "learning_rate": 0.0002268041237113402, + "loss": 0.3741, + "step": 27721 + }, + { + "epoch": 1.5488728105707184, + "grad_norm": 0.41232094168663025, + "learning_rate": 0.00022677610936799642, + "loss": 0.4657, + "step": 27722 + }, + { + "epoch": 1.5489286811744, + "grad_norm": 0.7597317099571228, + "learning_rate": 0.00022674809502465262, + "loss": 0.3582, + "step": 27723 + }, + { + "epoch": 1.5489845517780818, + "grad_norm": 0.5055459141731262, + "learning_rate": 0.00022672008068130883, + "loss": 0.4781, + "step": 27724 + }, + { + "epoch": 1.549040422381764, + "grad_norm": 0.37984704971313477, + "learning_rate": 0.00022669206633796506, + "loss": 0.3961, + "step": 27725 + }, + { + "epoch": 1.5490962929854457, + "grad_norm": 4.585656642913818, + "learning_rate": 0.00022666405199462127, + "loss": 0.4294, + "step": 27726 + }, + { + "epoch": 1.5491521635891274, + "grad_norm": 3.1337192058563232, + "learning_rate": 0.00022663603765127747, + "loss": 0.4092, + "step": 27727 + }, + { + "epoch": 1.5492080341928096, + "grad_norm": 0.6271639466285706, + "learning_rate": 0.00022660802330793368, + "loss": 0.4417, + "step": 27728 + }, + { + "epoch": 1.5492639047964913, + "grad_norm": 0.43780839443206787, + "learning_rate": 0.00022658000896458988, + "loss": 0.4942, + "step": 27729 + }, + { + "epoch": 1.549319775400173, + "grad_norm": 0.8571069240570068, + "learning_rate": 0.0002265519946212461, + "loss": 0.5771, + "step": 27730 + }, + { + "epoch": 1.549375646003855, + "grad_norm": 0.7636734843254089, + "learning_rate": 0.0002265239802779023, + "loss": 0.6685, + "step": 27731 + }, + { + "epoch": 1.549431516607537, + "grad_norm": 0.5937906503677368, + "learning_rate": 0.0002264959659345585, + "loss": 0.3538, + "step": 27732 + }, + { + "epoch": 1.5494873872112187, + "grad_norm": 0.3964087963104248, + "learning_rate": 0.0002264679515912147, + "loss": 0.3275, + "step": 27733 + }, + { + "epoch": 1.5495432578149007, + "grad_norm": 1.0223655700683594, + "learning_rate": 0.0002264399372478709, + "loss": 0.4464, + "step": 27734 + }, + { + "epoch": 1.5495991284185826, + "grad_norm": 0.3642975389957428, + "learning_rate": 0.00022641192290452712, + "loss": 0.3359, + "step": 27735 + }, + { + "epoch": 1.5496549990222643, + "grad_norm": 0.31322726607322693, + "learning_rate": 0.00022638390856118332, + "loss": 0.3599, + "step": 27736 + }, + { + "epoch": 1.5497108696259463, + "grad_norm": 0.554109513759613, + "learning_rate": 0.00022635589421783953, + "loss": 0.5348, + "step": 27737 + }, + { + "epoch": 1.5497667402296282, + "grad_norm": 0.5273289084434509, + "learning_rate": 0.00022632787987449574, + "loss": 0.4445, + "step": 27738 + }, + { + "epoch": 1.54982261083331, + "grad_norm": 0.49803876876831055, + "learning_rate": 0.00022629986553115194, + "loss": 0.4201, + "step": 27739 + }, + { + "epoch": 1.549878481436992, + "grad_norm": 0.4096674621105194, + "learning_rate": 0.00022627185118780815, + "loss": 0.3909, + "step": 27740 + }, + { + "epoch": 1.5499343520406739, + "grad_norm": 0.5639722943305969, + "learning_rate": 0.00022624383684446438, + "loss": 0.3657, + "step": 27741 + }, + { + "epoch": 1.5499902226443556, + "grad_norm": 0.591784656047821, + "learning_rate": 0.00022621582250112059, + "loss": 0.475, + "step": 27742 + }, + { + "epoch": 1.5500460932480375, + "grad_norm": 0.7738181352615356, + "learning_rate": 0.0002261878081577768, + "loss": 0.581, + "step": 27743 + }, + { + "epoch": 1.5501019638517195, + "grad_norm": 0.9245391488075256, + "learning_rate": 0.000226159793814433, + "loss": 0.7106, + "step": 27744 + }, + { + "epoch": 1.5501578344554012, + "grad_norm": 0.4792414903640747, + "learning_rate": 0.0002261317794710892, + "loss": 0.3696, + "step": 27745 + }, + { + "epoch": 1.5502137050590832, + "grad_norm": 0.4074365198612213, + "learning_rate": 0.0002261037651277454, + "loss": 0.3933, + "step": 27746 + }, + { + "epoch": 1.5502695756627651, + "grad_norm": 0.3219234347343445, + "learning_rate": 0.00022607575078440161, + "loss": 0.3828, + "step": 27747 + }, + { + "epoch": 1.5503254462664469, + "grad_norm": 0.31490156054496765, + "learning_rate": 0.00022604773644105782, + "loss": 0.3954, + "step": 27748 + }, + { + "epoch": 1.5503813168701288, + "grad_norm": 0.39040109515190125, + "learning_rate": 0.00022601972209771405, + "loss": 0.4516, + "step": 27749 + }, + { + "epoch": 1.5504371874738108, + "grad_norm": 0.35525479912757874, + "learning_rate": 0.00022599170775437026, + "loss": 0.3876, + "step": 27750 + }, + { + "epoch": 1.5504930580774925, + "grad_norm": 0.3690975606441498, + "learning_rate": 0.00022596369341102647, + "loss": 0.4649, + "step": 27751 + }, + { + "epoch": 1.5505489286811744, + "grad_norm": 0.3278123438358307, + "learning_rate": 0.00022593567906768267, + "loss": 0.3227, + "step": 27752 + }, + { + "epoch": 1.5506047992848564, + "grad_norm": 0.6176265478134155, + "learning_rate": 0.00022590766472433888, + "loss": 0.4683, + "step": 27753 + }, + { + "epoch": 1.5506606698885381, + "grad_norm": 0.636761486530304, + "learning_rate": 0.00022587965038099506, + "loss": 0.3476, + "step": 27754 + }, + { + "epoch": 1.5507165404922199, + "grad_norm": 0.4006495177745819, + "learning_rate": 0.00022585163603765126, + "loss": 0.3709, + "step": 27755 + }, + { + "epoch": 1.550772411095902, + "grad_norm": 0.43662455677986145, + "learning_rate": 0.00022582362169430747, + "loss": 0.6644, + "step": 27756 + }, + { + "epoch": 1.5508282816995838, + "grad_norm": 0.5051088333129883, + "learning_rate": 0.0002257956073509637, + "loss": 0.3795, + "step": 27757 + }, + { + "epoch": 1.5508841523032655, + "grad_norm": 1.2502954006195068, + "learning_rate": 0.0002257675930076199, + "loss": 0.5654, + "step": 27758 + }, + { + "epoch": 1.5509400229069477, + "grad_norm": 0.3720149099826813, + "learning_rate": 0.0002257395786642761, + "loss": 0.3954, + "step": 27759 + }, + { + "epoch": 1.5509958935106294, + "grad_norm": 0.41748929023742676, + "learning_rate": 0.00022571156432093232, + "loss": 0.4666, + "step": 27760 + }, + { + "epoch": 1.5510517641143111, + "grad_norm": 1.7419819831848145, + "learning_rate": 0.00022568354997758852, + "loss": 0.3727, + "step": 27761 + }, + { + "epoch": 1.5511076347179933, + "grad_norm": 0.7746831178665161, + "learning_rate": 0.00022565553563424473, + "loss": 0.2924, + "step": 27762 + }, + { + "epoch": 1.551163505321675, + "grad_norm": 0.34369394183158875, + "learning_rate": 0.00022562752129090093, + "loss": 0.3423, + "step": 27763 + }, + { + "epoch": 1.5512193759253567, + "grad_norm": 0.5290240049362183, + "learning_rate": 0.00022559950694755714, + "loss": 0.4258, + "step": 27764 + }, + { + "epoch": 1.5512752465290387, + "grad_norm": 0.46754541993141174, + "learning_rate": 0.00022557149260421337, + "loss": 0.4517, + "step": 27765 + }, + { + "epoch": 1.5513311171327206, + "grad_norm": 4.10866117477417, + "learning_rate": 0.00022554347826086958, + "loss": 0.3894, + "step": 27766 + }, + { + "epoch": 1.5513869877364024, + "grad_norm": 0.5014000535011292, + "learning_rate": 0.00022551546391752579, + "loss": 0.4205, + "step": 27767 + }, + { + "epoch": 1.5514428583400843, + "grad_norm": 0.45143991708755493, + "learning_rate": 0.000225487449574182, + "loss": 0.3801, + "step": 27768 + }, + { + "epoch": 1.5514987289437663, + "grad_norm": 0.4778406023979187, + "learning_rate": 0.0002254594352308382, + "loss": 0.4215, + "step": 27769 + }, + { + "epoch": 1.551554599547448, + "grad_norm": 0.4967866539955139, + "learning_rate": 0.0002254314208874944, + "loss": 0.398, + "step": 27770 + }, + { + "epoch": 1.55161047015113, + "grad_norm": 0.7771312594413757, + "learning_rate": 0.0002254034065441506, + "loss": 0.4647, + "step": 27771 + }, + { + "epoch": 1.551666340754812, + "grad_norm": 0.4630691707134247, + "learning_rate": 0.00022537539220080684, + "loss": 0.3401, + "step": 27772 + }, + { + "epoch": 1.5517222113584936, + "grad_norm": 0.3230164051055908, + "learning_rate": 0.00022534737785746305, + "loss": 0.3884, + "step": 27773 + }, + { + "epoch": 1.5517780819621756, + "grad_norm": 0.41936689615249634, + "learning_rate": 0.00022531936351411925, + "loss": 0.382, + "step": 27774 + }, + { + "epoch": 1.5518339525658575, + "grad_norm": 0.5210415720939636, + "learning_rate": 0.00022529134917077543, + "loss": 0.4595, + "step": 27775 + }, + { + "epoch": 1.5518898231695393, + "grad_norm": 0.3723212480545044, + "learning_rate": 0.00022526333482743164, + "loss": 0.3197, + "step": 27776 + }, + { + "epoch": 1.5519456937732212, + "grad_norm": 1.7563360929489136, + "learning_rate": 0.00022523532048408784, + "loss": 0.3441, + "step": 27777 + }, + { + "epoch": 1.5520015643769032, + "grad_norm": 0.6746132373809814, + "learning_rate": 0.00022520730614074405, + "loss": 0.629, + "step": 27778 + }, + { + "epoch": 1.552057434980585, + "grad_norm": 1.3947373628616333, + "learning_rate": 0.00022517929179740025, + "loss": 0.4773, + "step": 27779 + }, + { + "epoch": 1.5521133055842669, + "grad_norm": 2.2690300941467285, + "learning_rate": 0.0002251512774540565, + "loss": 0.424, + "step": 27780 + }, + { + "epoch": 1.5521691761879488, + "grad_norm": 1.2841054201126099, + "learning_rate": 0.0002251232631107127, + "loss": 0.5265, + "step": 27781 + }, + { + "epoch": 1.5522250467916305, + "grad_norm": 0.5672566294670105, + "learning_rate": 0.0002250952487673689, + "loss": 0.4445, + "step": 27782 + }, + { + "epoch": 1.5522809173953125, + "grad_norm": 1.3538535833358765, + "learning_rate": 0.0002250672344240251, + "loss": 0.4753, + "step": 27783 + }, + { + "epoch": 1.5523367879989944, + "grad_norm": 0.5844594240188599, + "learning_rate": 0.0002250392200806813, + "loss": 0.4175, + "step": 27784 + }, + { + "epoch": 1.5523926586026762, + "grad_norm": 0.38843849301338196, + "learning_rate": 0.00022501120573733752, + "loss": 0.4482, + "step": 27785 + }, + { + "epoch": 1.5524485292063581, + "grad_norm": 1.5684584379196167, + "learning_rate": 0.00022498319139399372, + "loss": 0.4213, + "step": 27786 + }, + { + "epoch": 1.55250439981004, + "grad_norm": 0.8363824486732483, + "learning_rate": 0.00022495517705064993, + "loss": 0.3792, + "step": 27787 + }, + { + "epoch": 1.5525602704137218, + "grad_norm": 0.6568323373794556, + "learning_rate": 0.00022492716270730616, + "loss": 0.4233, + "step": 27788 + }, + { + "epoch": 1.5526161410174035, + "grad_norm": 0.5353951454162598, + "learning_rate": 0.00022489914836396237, + "loss": 0.4392, + "step": 27789 + }, + { + "epoch": 1.5526720116210857, + "grad_norm": 0.3730304539203644, + "learning_rate": 0.00022487113402061857, + "loss": 0.483, + "step": 27790 + }, + { + "epoch": 1.5527278822247674, + "grad_norm": 0.522258996963501, + "learning_rate": 0.00022484311967727478, + "loss": 0.3907, + "step": 27791 + }, + { + "epoch": 1.5527837528284492, + "grad_norm": 0.28715744614601135, + "learning_rate": 0.00022481510533393098, + "loss": 0.3259, + "step": 27792 + }, + { + "epoch": 1.5528396234321313, + "grad_norm": 0.3882676064968109, + "learning_rate": 0.0002247870909905872, + "loss": 0.3947, + "step": 27793 + }, + { + "epoch": 1.552895494035813, + "grad_norm": 0.4883400797843933, + "learning_rate": 0.0002247590766472434, + "loss": 0.4205, + "step": 27794 + }, + { + "epoch": 1.5529513646394948, + "grad_norm": 0.6912471055984497, + "learning_rate": 0.0002247310623038996, + "loss": 0.3816, + "step": 27795 + }, + { + "epoch": 1.553007235243177, + "grad_norm": 0.6976539492607117, + "learning_rate": 0.00022470304796055584, + "loss": 0.3729, + "step": 27796 + }, + { + "epoch": 1.5530631058468587, + "grad_norm": 1.3650357723236084, + "learning_rate": 0.00022467503361721201, + "loss": 0.5016, + "step": 27797 + }, + { + "epoch": 1.5531189764505404, + "grad_norm": 0.339224636554718, + "learning_rate": 0.00022464701927386822, + "loss": 0.3764, + "step": 27798 + }, + { + "epoch": 1.5531748470542224, + "grad_norm": 0.374708890914917, + "learning_rate": 0.00022461900493052443, + "loss": 0.3541, + "step": 27799 + }, + { + "epoch": 1.5532307176579043, + "grad_norm": 0.3480576276779175, + "learning_rate": 0.00022459099058718063, + "loss": 0.4288, + "step": 27800 + }, + { + "epoch": 1.553286588261586, + "grad_norm": 0.31329721212387085, + "learning_rate": 0.00022456297624383684, + "loss": 0.3703, + "step": 27801 + }, + { + "epoch": 1.553342458865268, + "grad_norm": 0.4585150182247162, + "learning_rate": 0.00022453496190049304, + "loss": 0.3024, + "step": 27802 + }, + { + "epoch": 1.55339832946895, + "grad_norm": 0.6808285117149353, + "learning_rate": 0.00022450694755714925, + "loss": 0.4105, + "step": 27803 + }, + { + "epoch": 1.5534542000726317, + "grad_norm": 0.49719539284706116, + "learning_rate": 0.00022447893321380548, + "loss": 0.3742, + "step": 27804 + }, + { + "epoch": 1.5535100706763136, + "grad_norm": 0.5015256404876709, + "learning_rate": 0.0002244509188704617, + "loss": 0.466, + "step": 27805 + }, + { + "epoch": 1.5535659412799956, + "grad_norm": 1.5547596216201782, + "learning_rate": 0.0002244229045271179, + "loss": 0.4896, + "step": 27806 + }, + { + "epoch": 1.5536218118836773, + "grad_norm": 0.3561072051525116, + "learning_rate": 0.0002243948901837741, + "loss": 0.3937, + "step": 27807 + }, + { + "epoch": 1.5536776824873593, + "grad_norm": 0.47490665316581726, + "learning_rate": 0.0002243668758404303, + "loss": 0.4376, + "step": 27808 + }, + { + "epoch": 1.5537335530910412, + "grad_norm": 0.636273980140686, + "learning_rate": 0.0002243388614970865, + "loss": 0.5605, + "step": 27809 + }, + { + "epoch": 1.553789423694723, + "grad_norm": 0.3298848867416382, + "learning_rate": 0.00022431084715374272, + "loss": 0.3906, + "step": 27810 + }, + { + "epoch": 1.553845294298405, + "grad_norm": 0.6185240745544434, + "learning_rate": 0.00022428283281039892, + "loss": 0.3962, + "step": 27811 + }, + { + "epoch": 1.5539011649020869, + "grad_norm": 0.6284756064414978, + "learning_rate": 0.00022425481846705516, + "loss": 0.5845, + "step": 27812 + }, + { + "epoch": 1.5539570355057686, + "grad_norm": 0.5087429285049438, + "learning_rate": 0.00022422680412371136, + "loss": 0.4907, + "step": 27813 + }, + { + "epoch": 1.5540129061094505, + "grad_norm": 0.48090463876724243, + "learning_rate": 0.00022419878978036757, + "loss": 0.3607, + "step": 27814 + }, + { + "epoch": 1.5540687767131325, + "grad_norm": 0.36729034781455994, + "learning_rate": 0.00022417077543702377, + "loss": 0.4305, + "step": 27815 + }, + { + "epoch": 1.5541246473168142, + "grad_norm": 3.060041666030884, + "learning_rate": 0.00022414276109367998, + "loss": 0.4347, + "step": 27816 + }, + { + "epoch": 1.5541805179204962, + "grad_norm": 0.4871310293674469, + "learning_rate": 0.00022411474675033618, + "loss": 0.3242, + "step": 27817 + }, + { + "epoch": 1.5542363885241781, + "grad_norm": 0.3779184818267822, + "learning_rate": 0.00022408673240699236, + "loss": 0.4165, + "step": 27818 + }, + { + "epoch": 1.5542922591278598, + "grad_norm": 0.42644569277763367, + "learning_rate": 0.00022405871806364857, + "loss": 0.4416, + "step": 27819 + }, + { + "epoch": 1.5543481297315418, + "grad_norm": 0.5947522521018982, + "learning_rate": 0.0002240307037203048, + "loss": 0.4943, + "step": 27820 + }, + { + "epoch": 1.5544040003352237, + "grad_norm": 0.5552598834037781, + "learning_rate": 0.000224002689376961, + "loss": 0.4556, + "step": 27821 + }, + { + "epoch": 1.5544598709389055, + "grad_norm": 0.3822673559188843, + "learning_rate": 0.0002239746750336172, + "loss": 0.4205, + "step": 27822 + }, + { + "epoch": 1.5545157415425872, + "grad_norm": 0.35517802834510803, + "learning_rate": 0.00022394666069027342, + "loss": 0.3564, + "step": 27823 + }, + { + "epoch": 1.5545716121462694, + "grad_norm": 0.49859362840652466, + "learning_rate": 0.00022391864634692963, + "loss": 0.4022, + "step": 27824 + }, + { + "epoch": 1.554627482749951, + "grad_norm": 0.39390161633491516, + "learning_rate": 0.00022389063200358583, + "loss": 0.4761, + "step": 27825 + }, + { + "epoch": 1.5546833533536328, + "grad_norm": 0.4513009488582611, + "learning_rate": 0.00022386261766024204, + "loss": 0.4751, + "step": 27826 + }, + { + "epoch": 1.554739223957315, + "grad_norm": 0.4418204426765442, + "learning_rate": 0.00022383460331689824, + "loss": 0.3647, + "step": 27827 + }, + { + "epoch": 1.5547950945609967, + "grad_norm": 0.49769890308380127, + "learning_rate": 0.00022380658897355448, + "loss": 0.3348, + "step": 27828 + }, + { + "epoch": 1.5548509651646785, + "grad_norm": 0.43676936626434326, + "learning_rate": 0.00022377857463021068, + "loss": 0.3845, + "step": 27829 + }, + { + "epoch": 1.5549068357683606, + "grad_norm": 0.4104653000831604, + "learning_rate": 0.0002237505602868669, + "loss": 0.4227, + "step": 27830 + }, + { + "epoch": 1.5549627063720424, + "grad_norm": 4.7332634925842285, + "learning_rate": 0.0002237225459435231, + "loss": 0.4563, + "step": 27831 + }, + { + "epoch": 1.555018576975724, + "grad_norm": 0.39292722940444946, + "learning_rate": 0.0002236945316001793, + "loss": 0.3689, + "step": 27832 + }, + { + "epoch": 1.555074447579406, + "grad_norm": 0.4170571267604828, + "learning_rate": 0.0002236665172568355, + "loss": 0.4786, + "step": 27833 + }, + { + "epoch": 1.555130318183088, + "grad_norm": 0.43309634923934937, + "learning_rate": 0.0002236385029134917, + "loss": 0.4563, + "step": 27834 + }, + { + "epoch": 1.5551861887867697, + "grad_norm": 0.31393367052078247, + "learning_rate": 0.00022361048857014792, + "loss": 0.3506, + "step": 27835 + }, + { + "epoch": 1.5552420593904517, + "grad_norm": 0.39643946290016174, + "learning_rate": 0.00022358247422680415, + "loss": 0.4837, + "step": 27836 + }, + { + "epoch": 1.5552979299941336, + "grad_norm": 2.569502830505371, + "learning_rate": 0.00022355445988346035, + "loss": 0.4176, + "step": 27837 + }, + { + "epoch": 1.5553538005978154, + "grad_norm": 0.4869304299354553, + "learning_rate": 0.00022352644554011656, + "loss": 0.5018, + "step": 27838 + }, + { + "epoch": 1.5554096712014973, + "grad_norm": 5.366434097290039, + "learning_rate": 0.00022349843119677274, + "loss": 0.4611, + "step": 27839 + }, + { + "epoch": 1.5554655418051793, + "grad_norm": 0.4666939079761505, + "learning_rate": 0.00022347041685342895, + "loss": 0.4325, + "step": 27840 + }, + { + "epoch": 1.555521412408861, + "grad_norm": 0.3744693696498871, + "learning_rate": 0.00022344240251008515, + "loss": 0.4477, + "step": 27841 + }, + { + "epoch": 1.555577283012543, + "grad_norm": 0.41699907183647156, + "learning_rate": 0.00022341438816674136, + "loss": 0.4242, + "step": 27842 + }, + { + "epoch": 1.555633153616225, + "grad_norm": 0.3345234990119934, + "learning_rate": 0.00022338637382339756, + "loss": 0.3685, + "step": 27843 + }, + { + "epoch": 1.5556890242199066, + "grad_norm": 0.5590550899505615, + "learning_rate": 0.0002233583594800538, + "loss": 0.409, + "step": 27844 + }, + { + "epoch": 1.5557448948235886, + "grad_norm": 1.908947467803955, + "learning_rate": 0.00022333034513671, + "loss": 0.2994, + "step": 27845 + }, + { + "epoch": 1.5558007654272705, + "grad_norm": 0.5028336644172668, + "learning_rate": 0.0002233023307933662, + "loss": 0.4097, + "step": 27846 + }, + { + "epoch": 1.5558566360309523, + "grad_norm": 0.676910400390625, + "learning_rate": 0.0002232743164500224, + "loss": 0.7551, + "step": 27847 + }, + { + "epoch": 1.5559125066346342, + "grad_norm": 0.41420456767082214, + "learning_rate": 0.00022324630210667862, + "loss": 0.3678, + "step": 27848 + }, + { + "epoch": 1.5559683772383162, + "grad_norm": 0.4720274806022644, + "learning_rate": 0.00022321828776333482, + "loss": 0.4985, + "step": 27849 + }, + { + "epoch": 1.5560242478419979, + "grad_norm": 0.45960769057273865, + "learning_rate": 0.00022319027341999103, + "loss": 0.4655, + "step": 27850 + }, + { + "epoch": 1.5560801184456798, + "grad_norm": 0.4269457757472992, + "learning_rate": 0.00022316225907664726, + "loss": 0.3846, + "step": 27851 + }, + { + "epoch": 1.5561359890493618, + "grad_norm": 0.43795156478881836, + "learning_rate": 0.00022313424473330347, + "loss": 0.4644, + "step": 27852 + }, + { + "epoch": 1.5561918596530435, + "grad_norm": 0.3819417953491211, + "learning_rate": 0.00022310623038995968, + "loss": 0.4011, + "step": 27853 + }, + { + "epoch": 1.5562477302567255, + "grad_norm": 0.3569195568561554, + "learning_rate": 0.00022307821604661588, + "loss": 0.3755, + "step": 27854 + }, + { + "epoch": 1.5563036008604074, + "grad_norm": 0.42759114503860474, + "learning_rate": 0.0002230502017032721, + "loss": 0.4518, + "step": 27855 + }, + { + "epoch": 1.5563594714640892, + "grad_norm": 0.45298150181770325, + "learning_rate": 0.0002230221873599283, + "loss": 0.4845, + "step": 27856 + }, + { + "epoch": 1.5564153420677709, + "grad_norm": 0.3739667236804962, + "learning_rate": 0.0002229941730165845, + "loss": 0.4893, + "step": 27857 + }, + { + "epoch": 1.556471212671453, + "grad_norm": 0.9898538589477539, + "learning_rate": 0.0002229661586732407, + "loss": 0.4266, + "step": 27858 + }, + { + "epoch": 1.5565270832751348, + "grad_norm": 2.084681272506714, + "learning_rate": 0.00022293814432989694, + "loss": 0.4044, + "step": 27859 + }, + { + "epoch": 1.5565829538788165, + "grad_norm": 0.3847338855266571, + "learning_rate": 0.00022291012998655312, + "loss": 0.3142, + "step": 27860 + }, + { + "epoch": 1.5566388244824987, + "grad_norm": 1.9584487676620483, + "learning_rate": 0.00022288211564320932, + "loss": 0.4524, + "step": 27861 + }, + { + "epoch": 1.5566946950861804, + "grad_norm": 0.6915044188499451, + "learning_rate": 0.00022285410129986553, + "loss": 0.379, + "step": 27862 + }, + { + "epoch": 1.5567505656898621, + "grad_norm": 0.48093539476394653, + "learning_rate": 0.00022282608695652173, + "loss": 0.4254, + "step": 27863 + }, + { + "epoch": 1.5568064362935443, + "grad_norm": 1.2253575325012207, + "learning_rate": 0.00022279807261317794, + "loss": 0.3245, + "step": 27864 + }, + { + "epoch": 1.556862306897226, + "grad_norm": 2.099454402923584, + "learning_rate": 0.00022277005826983414, + "loss": 0.4663, + "step": 27865 + }, + { + "epoch": 1.5569181775009078, + "grad_norm": 0.4891197085380554, + "learning_rate": 0.00022274204392649035, + "loss": 0.3896, + "step": 27866 + }, + { + "epoch": 1.5569740481045897, + "grad_norm": 0.520697295665741, + "learning_rate": 0.00022271402958314658, + "loss": 0.4551, + "step": 27867 + }, + { + "epoch": 1.5570299187082717, + "grad_norm": 0.5367410778999329, + "learning_rate": 0.0002226860152398028, + "loss": 0.378, + "step": 27868 + }, + { + "epoch": 1.5570857893119534, + "grad_norm": 0.43118903040885925, + "learning_rate": 0.000222658000896459, + "loss": 0.426, + "step": 27869 + }, + { + "epoch": 1.5571416599156354, + "grad_norm": 0.518278956413269, + "learning_rate": 0.0002226299865531152, + "loss": 0.4689, + "step": 27870 + }, + { + "epoch": 1.5571975305193173, + "grad_norm": 0.6233494877815247, + "learning_rate": 0.0002226019722097714, + "loss": 0.6558, + "step": 27871 + }, + { + "epoch": 1.557253401122999, + "grad_norm": 0.47422629594802856, + "learning_rate": 0.0002225739578664276, + "loss": 0.4555, + "step": 27872 + }, + { + "epoch": 1.557309271726681, + "grad_norm": 0.44967809319496155, + "learning_rate": 0.00022254594352308382, + "loss": 0.3184, + "step": 27873 + }, + { + "epoch": 1.557365142330363, + "grad_norm": 0.40457063913345337, + "learning_rate": 0.00022251792917974002, + "loss": 0.4659, + "step": 27874 + }, + { + "epoch": 1.5574210129340447, + "grad_norm": 1.9319628477096558, + "learning_rate": 0.00022248991483639626, + "loss": 0.541, + "step": 27875 + }, + { + "epoch": 1.5574768835377266, + "grad_norm": 0.40028631687164307, + "learning_rate": 0.00022246190049305246, + "loss": 0.4167, + "step": 27876 + }, + { + "epoch": 1.5575327541414086, + "grad_norm": 0.5583202838897705, + "learning_rate": 0.00022243388614970867, + "loss": 0.3465, + "step": 27877 + }, + { + "epoch": 1.5575886247450903, + "grad_norm": 0.5293304324150085, + "learning_rate": 0.00022240587180636487, + "loss": 0.4149, + "step": 27878 + }, + { + "epoch": 1.5576444953487723, + "grad_norm": 0.414290189743042, + "learning_rate": 0.00022237785746302108, + "loss": 0.4277, + "step": 27879 + }, + { + "epoch": 1.5577003659524542, + "grad_norm": 0.43186402320861816, + "learning_rate": 0.00022234984311967729, + "loss": 0.4782, + "step": 27880 + }, + { + "epoch": 1.557756236556136, + "grad_norm": 0.5376913547515869, + "learning_rate": 0.0002223218287763335, + "loss": 0.3447, + "step": 27881 + }, + { + "epoch": 1.5578121071598179, + "grad_norm": 0.4073699712753296, + "learning_rate": 0.00022229381443298967, + "loss": 0.3342, + "step": 27882 + }, + { + "epoch": 1.5578679777634998, + "grad_norm": 0.9715790748596191, + "learning_rate": 0.0002222658000896459, + "loss": 0.4123, + "step": 27883 + }, + { + "epoch": 1.5579238483671816, + "grad_norm": 2.371649980545044, + "learning_rate": 0.0002222377857463021, + "loss": 0.4933, + "step": 27884 + }, + { + "epoch": 1.5579797189708635, + "grad_norm": 0.4362573027610779, + "learning_rate": 0.00022220977140295832, + "loss": 0.3861, + "step": 27885 + }, + { + "epoch": 1.5580355895745455, + "grad_norm": 1.2144173383712769, + "learning_rate": 0.00022218175705961452, + "loss": 0.4155, + "step": 27886 + }, + { + "epoch": 1.5580914601782272, + "grad_norm": 0.7401503324508667, + "learning_rate": 0.00022215374271627073, + "loss": 0.3743, + "step": 27887 + }, + { + "epoch": 1.5581473307819091, + "grad_norm": 0.34899818897247314, + "learning_rate": 0.00022212572837292693, + "loss": 0.4247, + "step": 27888 + }, + { + "epoch": 1.558203201385591, + "grad_norm": 0.4065112769603729, + "learning_rate": 0.00022209771402958314, + "loss": 0.3847, + "step": 27889 + }, + { + "epoch": 1.5582590719892728, + "grad_norm": 0.3681662380695343, + "learning_rate": 0.00022206969968623934, + "loss": 0.2861, + "step": 27890 + }, + { + "epoch": 1.5583149425929546, + "grad_norm": 2.5317037105560303, + "learning_rate": 0.00022204168534289558, + "loss": 0.3997, + "step": 27891 + }, + { + "epoch": 1.5583708131966367, + "grad_norm": 0.40800854563713074, + "learning_rate": 0.00022201367099955178, + "loss": 0.4576, + "step": 27892 + }, + { + "epoch": 1.5584266838003185, + "grad_norm": 0.4038293659687042, + "learning_rate": 0.000221985656656208, + "loss": 0.3508, + "step": 27893 + }, + { + "epoch": 1.5584825544040002, + "grad_norm": 0.6172147989273071, + "learning_rate": 0.0002219576423128642, + "loss": 0.5852, + "step": 27894 + }, + { + "epoch": 1.5585384250076824, + "grad_norm": 0.3672626316547394, + "learning_rate": 0.0002219296279695204, + "loss": 0.2969, + "step": 27895 + }, + { + "epoch": 1.558594295611364, + "grad_norm": 0.4717801511287689, + "learning_rate": 0.0002219016136261766, + "loss": 0.4578, + "step": 27896 + }, + { + "epoch": 1.5586501662150458, + "grad_norm": 0.6867666840553284, + "learning_rate": 0.0002218735992828328, + "loss": 0.3682, + "step": 27897 + }, + { + "epoch": 1.558706036818728, + "grad_norm": 0.4880625307559967, + "learning_rate": 0.00022184558493948902, + "loss": 0.396, + "step": 27898 + }, + { + "epoch": 1.5587619074224097, + "grad_norm": 0.37047410011291504, + "learning_rate": 0.00022181757059614525, + "loss": 0.4338, + "step": 27899 + }, + { + "epoch": 1.5588177780260914, + "grad_norm": 0.39366984367370605, + "learning_rate": 0.00022178955625280146, + "loss": 0.44, + "step": 27900 + }, + { + "epoch": 1.5588736486297734, + "grad_norm": 0.5216104388237, + "learning_rate": 0.00022176154190945766, + "loss": 0.48, + "step": 27901 + }, + { + "epoch": 1.5589295192334554, + "grad_norm": 0.4577905237674713, + "learning_rate": 0.00022173352756611387, + "loss": 0.4446, + "step": 27902 + }, + { + "epoch": 1.558985389837137, + "grad_norm": 0.5937201380729675, + "learning_rate": 0.00022170551322277005, + "loss": 0.4112, + "step": 27903 + }, + { + "epoch": 1.559041260440819, + "grad_norm": 0.47465428709983826, + "learning_rate": 0.00022167749887942625, + "loss": 0.5438, + "step": 27904 + }, + { + "epoch": 1.559097131044501, + "grad_norm": 0.808074951171875, + "learning_rate": 0.00022164948453608246, + "loss": 0.3605, + "step": 27905 + }, + { + "epoch": 1.5591530016481827, + "grad_norm": 0.44995176792144775, + "learning_rate": 0.00022162147019273866, + "loss": 0.461, + "step": 27906 + }, + { + "epoch": 1.5592088722518647, + "grad_norm": 0.38658061623573303, + "learning_rate": 0.0002215934558493949, + "loss": 0.368, + "step": 27907 + }, + { + "epoch": 1.5592647428555466, + "grad_norm": 0.34679463505744934, + "learning_rate": 0.0002215654415060511, + "loss": 0.3462, + "step": 27908 + }, + { + "epoch": 1.5593206134592283, + "grad_norm": 0.48853740096092224, + "learning_rate": 0.0002215374271627073, + "loss": 0.3493, + "step": 27909 + }, + { + "epoch": 1.5593764840629103, + "grad_norm": 0.34291452169418335, + "learning_rate": 0.00022150941281936351, + "loss": 0.3522, + "step": 27910 + }, + { + "epoch": 1.5594323546665922, + "grad_norm": 0.5008469223976135, + "learning_rate": 0.00022148139847601972, + "loss": 0.4065, + "step": 27911 + }, + { + "epoch": 1.559488225270274, + "grad_norm": 1.485838770866394, + "learning_rate": 0.00022145338413267593, + "loss": 0.3617, + "step": 27912 + }, + { + "epoch": 1.559544095873956, + "grad_norm": 0.3673437535762787, + "learning_rate": 0.00022142536978933213, + "loss": 0.3415, + "step": 27913 + }, + { + "epoch": 1.5595999664776379, + "grad_norm": 0.930901288986206, + "learning_rate": 0.00022139735544598837, + "loss": 0.347, + "step": 27914 + }, + { + "epoch": 1.5596558370813196, + "grad_norm": 0.46726223826408386, + "learning_rate": 0.00022136934110264457, + "loss": 0.4956, + "step": 27915 + }, + { + "epoch": 1.5597117076850016, + "grad_norm": 0.3618989586830139, + "learning_rate": 0.00022134132675930078, + "loss": 0.3089, + "step": 27916 + }, + { + "epoch": 1.5597675782886835, + "grad_norm": 0.9292025566101074, + "learning_rate": 0.00022131331241595698, + "loss": 0.3834, + "step": 27917 + }, + { + "epoch": 1.5598234488923652, + "grad_norm": 0.37328287959098816, + "learning_rate": 0.0002212852980726132, + "loss": 0.3388, + "step": 27918 + }, + { + "epoch": 1.5598793194960472, + "grad_norm": 0.9778773188591003, + "learning_rate": 0.0002212572837292694, + "loss": 0.3967, + "step": 27919 + }, + { + "epoch": 1.5599351900997291, + "grad_norm": 0.387905091047287, + "learning_rate": 0.0002212292693859256, + "loss": 0.3547, + "step": 27920 + }, + { + "epoch": 1.5599910607034109, + "grad_norm": 0.5396341681480408, + "learning_rate": 0.0002212012550425818, + "loss": 0.5364, + "step": 27921 + }, + { + "epoch": 1.5600469313070928, + "grad_norm": 0.3466677665710449, + "learning_rate": 0.00022117324069923804, + "loss": 0.4126, + "step": 27922 + }, + { + "epoch": 1.5601028019107748, + "grad_norm": 0.44071587920188904, + "learning_rate": 0.00022114522635589424, + "loss": 0.3954, + "step": 27923 + }, + { + "epoch": 1.5601586725144565, + "grad_norm": 0.4400257468223572, + "learning_rate": 0.00022111721201255042, + "loss": 0.5692, + "step": 27924 + }, + { + "epoch": 1.5602145431181382, + "grad_norm": 0.37065741419792175, + "learning_rate": 0.00022108919766920663, + "loss": 0.3589, + "step": 27925 + }, + { + "epoch": 1.5602704137218204, + "grad_norm": 0.661099374294281, + "learning_rate": 0.00022106118332586284, + "loss": 0.4914, + "step": 27926 + }, + { + "epoch": 1.5603262843255021, + "grad_norm": 0.3368808925151825, + "learning_rate": 0.00022103316898251904, + "loss": 0.4339, + "step": 27927 + }, + { + "epoch": 1.5603821549291839, + "grad_norm": 0.6323128342628479, + "learning_rate": 0.00022100515463917525, + "loss": 0.3811, + "step": 27928 + }, + { + "epoch": 1.560438025532866, + "grad_norm": 0.3779548704624176, + "learning_rate": 0.00022097714029583145, + "loss": 0.4546, + "step": 27929 + }, + { + "epoch": 1.5604938961365478, + "grad_norm": 0.8508569598197937, + "learning_rate": 0.00022094912595248769, + "loss": 0.4515, + "step": 27930 + }, + { + "epoch": 1.5605497667402295, + "grad_norm": 0.44117456674575806, + "learning_rate": 0.0002209211116091439, + "loss": 0.3735, + "step": 27931 + }, + { + "epoch": 1.5606056373439117, + "grad_norm": 0.4118061065673828, + "learning_rate": 0.0002208930972658001, + "loss": 0.4288, + "step": 27932 + }, + { + "epoch": 1.5606615079475934, + "grad_norm": 1.8194656372070312, + "learning_rate": 0.0002208650829224563, + "loss": 0.4333, + "step": 27933 + }, + { + "epoch": 1.5607173785512751, + "grad_norm": 0.45951327681541443, + "learning_rate": 0.0002208370685791125, + "loss": 0.4413, + "step": 27934 + }, + { + "epoch": 1.560773249154957, + "grad_norm": 1.0747466087341309, + "learning_rate": 0.00022080905423576871, + "loss": 0.4127, + "step": 27935 + }, + { + "epoch": 1.560829119758639, + "grad_norm": 0.5585138201713562, + "learning_rate": 0.00022078103989242492, + "loss": 0.3419, + "step": 27936 + }, + { + "epoch": 1.5608849903623208, + "grad_norm": 1.0594756603240967, + "learning_rate": 0.00022075302554908113, + "loss": 0.5028, + "step": 27937 + }, + { + "epoch": 1.5609408609660027, + "grad_norm": 0.3870279788970947, + "learning_rate": 0.00022072501120573736, + "loss": 0.3752, + "step": 27938 + }, + { + "epoch": 1.5609967315696847, + "grad_norm": 1.1915411949157715, + "learning_rate": 0.00022069699686239356, + "loss": 0.3995, + "step": 27939 + }, + { + "epoch": 1.5610526021733664, + "grad_norm": 0.29949671030044556, + "learning_rate": 0.00022066898251904977, + "loss": 0.3921, + "step": 27940 + }, + { + "epoch": 1.5611084727770483, + "grad_norm": 0.6453526616096497, + "learning_rate": 0.00022064096817570598, + "loss": 0.6151, + "step": 27941 + }, + { + "epoch": 1.5611643433807303, + "grad_norm": 1.825904369354248, + "learning_rate": 0.00022061295383236218, + "loss": 0.3656, + "step": 27942 + }, + { + "epoch": 1.561220213984412, + "grad_norm": 1.7123562097549438, + "learning_rate": 0.0002205849394890184, + "loss": 0.3447, + "step": 27943 + }, + { + "epoch": 1.561276084588094, + "grad_norm": 0.41849836707115173, + "learning_rate": 0.0002205569251456746, + "loss": 0.3776, + "step": 27944 + }, + { + "epoch": 1.561331955191776, + "grad_norm": 0.5982191562652588, + "learning_rate": 0.00022052891080233077, + "loss": 0.5491, + "step": 27945 + }, + { + "epoch": 1.5613878257954577, + "grad_norm": 8.694527626037598, + "learning_rate": 0.000220500896458987, + "loss": 0.424, + "step": 27946 + }, + { + "epoch": 1.5614436963991396, + "grad_norm": 0.5989289879798889, + "learning_rate": 0.0002204728821156432, + "loss": 0.4156, + "step": 27947 + }, + { + "epoch": 1.5614995670028216, + "grad_norm": 0.4880196452140808, + "learning_rate": 0.00022044486777229942, + "loss": 0.3671, + "step": 27948 + }, + { + "epoch": 1.5615554376065033, + "grad_norm": 0.3703460693359375, + "learning_rate": 0.00022041685342895562, + "loss": 0.443, + "step": 27949 + }, + { + "epoch": 1.5616113082101852, + "grad_norm": 0.6340950131416321, + "learning_rate": 0.00022038883908561183, + "loss": 0.4155, + "step": 27950 + }, + { + "epoch": 1.5616671788138672, + "grad_norm": 0.4834566116333008, + "learning_rate": 0.00022036082474226803, + "loss": 0.4353, + "step": 27951 + }, + { + "epoch": 1.561723049417549, + "grad_norm": 0.6279585957527161, + "learning_rate": 0.00022033281039892424, + "loss": 0.4408, + "step": 27952 + }, + { + "epoch": 1.5617789200212309, + "grad_norm": 0.43205368518829346, + "learning_rate": 0.00022030479605558045, + "loss": 0.3891, + "step": 27953 + }, + { + "epoch": 1.5618347906249128, + "grad_norm": 0.3627742528915405, + "learning_rate": 0.00022027678171223668, + "loss": 0.476, + "step": 27954 + }, + { + "epoch": 1.5618906612285945, + "grad_norm": 1.9554296731948853, + "learning_rate": 0.00022024876736889289, + "loss": 0.745, + "step": 27955 + }, + { + "epoch": 1.5619465318322765, + "grad_norm": 0.3054269254207611, + "learning_rate": 0.0002202207530255491, + "loss": 0.3131, + "step": 27956 + }, + { + "epoch": 1.5620024024359584, + "grad_norm": 0.49643149971961975, + "learning_rate": 0.0002201927386822053, + "loss": 0.3755, + "step": 27957 + }, + { + "epoch": 1.5620582730396402, + "grad_norm": 0.40398985147476196, + "learning_rate": 0.0002201647243388615, + "loss": 0.3948, + "step": 27958 + }, + { + "epoch": 1.562114143643322, + "grad_norm": 0.5674588084220886, + "learning_rate": 0.0002201367099955177, + "loss": 0.3453, + "step": 27959 + }, + { + "epoch": 1.562170014247004, + "grad_norm": 0.5199033617973328, + "learning_rate": 0.00022010869565217391, + "loss": 0.401, + "step": 27960 + }, + { + "epoch": 1.5622258848506858, + "grad_norm": 0.43146491050720215, + "learning_rate": 0.00022008068130883012, + "loss": 0.4121, + "step": 27961 + }, + { + "epoch": 1.5622817554543675, + "grad_norm": 0.37658241391181946, + "learning_rate": 0.00022005266696548635, + "loss": 0.3815, + "step": 27962 + }, + { + "epoch": 1.5623376260580497, + "grad_norm": 0.6446071267127991, + "learning_rate": 0.00022002465262214256, + "loss": 0.4091, + "step": 27963 + }, + { + "epoch": 1.5623934966617314, + "grad_norm": 0.4006809890270233, + "learning_rate": 0.00021999663827879876, + "loss": 0.5539, + "step": 27964 + }, + { + "epoch": 1.5624493672654132, + "grad_norm": 0.43603983521461487, + "learning_rate": 0.00021996862393545497, + "loss": 0.3059, + "step": 27965 + }, + { + "epoch": 1.5625052378690951, + "grad_norm": 0.38807040452957153, + "learning_rate": 0.00021994060959211118, + "loss": 0.429, + "step": 27966 + }, + { + "epoch": 1.562561108472777, + "grad_norm": 0.9343730807304382, + "learning_rate": 0.00021991259524876735, + "loss": 0.4446, + "step": 27967 + }, + { + "epoch": 1.5626169790764588, + "grad_norm": 2.5859668254852295, + "learning_rate": 0.00021988458090542356, + "loss": 0.4425, + "step": 27968 + }, + { + "epoch": 1.5626728496801408, + "grad_norm": 0.5909152030944824, + "learning_rate": 0.00021985656656207977, + "loss": 0.4821, + "step": 27969 + }, + { + "epoch": 1.5627287202838227, + "grad_norm": 0.7047179937362671, + "learning_rate": 0.000219828552218736, + "loss": 0.5238, + "step": 27970 + }, + { + "epoch": 1.5627845908875044, + "grad_norm": 0.5274574160575867, + "learning_rate": 0.0002198005378753922, + "loss": 0.5286, + "step": 27971 + }, + { + "epoch": 1.5628404614911864, + "grad_norm": 0.38091281056404114, + "learning_rate": 0.0002197725235320484, + "loss": 0.333, + "step": 27972 + }, + { + "epoch": 1.5628963320948683, + "grad_norm": 1.4722204208374023, + "learning_rate": 0.00021974450918870462, + "loss": 0.4923, + "step": 27973 + }, + { + "epoch": 1.56295220269855, + "grad_norm": 0.5483881831169128, + "learning_rate": 0.00021971649484536082, + "loss": 0.3544, + "step": 27974 + }, + { + "epoch": 1.563008073302232, + "grad_norm": 0.5198408961296082, + "learning_rate": 0.00021968848050201703, + "loss": 0.361, + "step": 27975 + }, + { + "epoch": 1.563063943905914, + "grad_norm": 0.6344133019447327, + "learning_rate": 0.00021966046615867323, + "loss": 0.4467, + "step": 27976 + }, + { + "epoch": 1.5631198145095957, + "grad_norm": 0.7599327564239502, + "learning_rate": 0.00021963245181532947, + "loss": 0.3908, + "step": 27977 + }, + { + "epoch": 1.5631756851132776, + "grad_norm": 0.4582282602787018, + "learning_rate": 0.00021960443747198567, + "loss": 0.4712, + "step": 27978 + }, + { + "epoch": 1.5632315557169596, + "grad_norm": 1.7420086860656738, + "learning_rate": 0.00021957642312864188, + "loss": 0.425, + "step": 27979 + }, + { + "epoch": 1.5632874263206413, + "grad_norm": 1.2159398794174194, + "learning_rate": 0.00021954840878529808, + "loss": 0.4041, + "step": 27980 + }, + { + "epoch": 1.5633432969243233, + "grad_norm": 0.44137492775917053, + "learning_rate": 0.0002195203944419543, + "loss": 0.4336, + "step": 27981 + }, + { + "epoch": 1.5633991675280052, + "grad_norm": 0.6439188122749329, + "learning_rate": 0.0002194923800986105, + "loss": 0.3832, + "step": 27982 + }, + { + "epoch": 1.563455038131687, + "grad_norm": 0.5098703503608704, + "learning_rate": 0.0002194643657552667, + "loss": 0.3877, + "step": 27983 + }, + { + "epoch": 1.563510908735369, + "grad_norm": 0.5937493443489075, + "learning_rate": 0.0002194363514119229, + "loss": 0.4212, + "step": 27984 + }, + { + "epoch": 1.5635667793390509, + "grad_norm": 0.5363062024116516, + "learning_rate": 0.00021940833706857914, + "loss": 0.3864, + "step": 27985 + }, + { + "epoch": 1.5636226499427326, + "grad_norm": 0.3386871814727783, + "learning_rate": 0.00021938032272523535, + "loss": 0.3509, + "step": 27986 + }, + { + "epoch": 1.5636785205464145, + "grad_norm": 0.7709883451461792, + "learning_rate": 0.00021935230838189155, + "loss": 0.3865, + "step": 27987 + }, + { + "epoch": 1.5637343911500965, + "grad_norm": 0.716331958770752, + "learning_rate": 0.00021932429403854773, + "loss": 0.4125, + "step": 27988 + }, + { + "epoch": 1.5637902617537782, + "grad_norm": 0.5194579362869263, + "learning_rate": 0.00021929627969520394, + "loss": 0.4318, + "step": 27989 + }, + { + "epoch": 1.5638461323574602, + "grad_norm": 0.6823799014091492, + "learning_rate": 0.00021926826535186014, + "loss": 0.4389, + "step": 27990 + }, + { + "epoch": 1.5639020029611421, + "grad_norm": 1.9007734060287476, + "learning_rate": 0.00021924025100851635, + "loss": 0.3006, + "step": 27991 + }, + { + "epoch": 1.5639578735648239, + "grad_norm": 0.6996992826461792, + "learning_rate": 0.00021921223666517255, + "loss": 0.2983, + "step": 27992 + }, + { + "epoch": 1.5640137441685056, + "grad_norm": 0.7907154560089111, + "learning_rate": 0.0002191842223218288, + "loss": 0.5955, + "step": 27993 + }, + { + "epoch": 1.5640696147721878, + "grad_norm": 0.33944177627563477, + "learning_rate": 0.000219156207978485, + "loss": 0.3986, + "step": 27994 + }, + { + "epoch": 1.5641254853758695, + "grad_norm": 0.36721542477607727, + "learning_rate": 0.0002191281936351412, + "loss": 0.3727, + "step": 27995 + }, + { + "epoch": 1.5641813559795512, + "grad_norm": 1.1125446557998657, + "learning_rate": 0.0002191001792917974, + "loss": 0.5507, + "step": 27996 + }, + { + "epoch": 1.5642372265832334, + "grad_norm": 0.38297519087791443, + "learning_rate": 0.0002190721649484536, + "loss": 0.3567, + "step": 27997 + }, + { + "epoch": 1.5642930971869151, + "grad_norm": 0.45195338129997253, + "learning_rate": 0.00021904415060510982, + "loss": 0.3659, + "step": 27998 + }, + { + "epoch": 1.5643489677905968, + "grad_norm": 0.49000421166419983, + "learning_rate": 0.00021901613626176602, + "loss": 0.3826, + "step": 27999 + }, + { + "epoch": 1.5644048383942788, + "grad_norm": 0.3864416480064392, + "learning_rate": 0.00021898812191842223, + "loss": 0.5046, + "step": 28000 + }, + { + "epoch": 1.5644048383942788, + "eval_cer": 0.08432348095534244, + "eval_loss": 0.3193748891353607, + "eval_runtime": 56.3445, + "eval_samples_per_second": 80.54, + "eval_steps_per_second": 5.04, + "eval_wer": 0.3337057624088713, + "step": 28000 + }, + { + "epoch": 1.5644607089979607, + "grad_norm": 0.3856048583984375, + "learning_rate": 0.00021896010757507846, + "loss": 0.448, + "step": 28001 + }, + { + "epoch": 1.5645165796016425, + "grad_norm": 0.7290375232696533, + "learning_rate": 0.00021893209323173467, + "loss": 0.38, + "step": 28002 + }, + { + "epoch": 1.5645724502053244, + "grad_norm": 0.47865837812423706, + "learning_rate": 0.00021890407888839087, + "loss": 0.5588, + "step": 28003 + }, + { + "epoch": 1.5646283208090064, + "grad_norm": 0.28958427906036377, + "learning_rate": 0.00021887606454504708, + "loss": 0.3307, + "step": 28004 + }, + { + "epoch": 1.564684191412688, + "grad_norm": 0.44012248516082764, + "learning_rate": 0.00021884805020170328, + "loss": 0.4121, + "step": 28005 + }, + { + "epoch": 1.56474006201637, + "grad_norm": 0.781109094619751, + "learning_rate": 0.0002188200358583595, + "loss": 0.3944, + "step": 28006 + }, + { + "epoch": 1.564795932620052, + "grad_norm": 0.537790060043335, + "learning_rate": 0.0002187920215150157, + "loss": 0.4588, + "step": 28007 + }, + { + "epoch": 1.5648518032237337, + "grad_norm": 0.33561182022094727, + "learning_rate": 0.0002187640071716719, + "loss": 0.3853, + "step": 28008 + }, + { + "epoch": 1.5649076738274157, + "grad_norm": 0.4028438925743103, + "learning_rate": 0.0002187359928283281, + "loss": 0.4396, + "step": 28009 + }, + { + "epoch": 1.5649635444310976, + "grad_norm": 0.37504348158836365, + "learning_rate": 0.0002187079784849843, + "loss": 0.3613, + "step": 28010 + }, + { + "epoch": 1.5650194150347794, + "grad_norm": 0.3992330729961395, + "learning_rate": 0.00021867996414164052, + "loss": 0.3541, + "step": 28011 + }, + { + "epoch": 1.5650752856384613, + "grad_norm": 0.3834443986415863, + "learning_rate": 0.00021865194979829672, + "loss": 0.4092, + "step": 28012 + }, + { + "epoch": 1.5651311562421433, + "grad_norm": 0.6637932062149048, + "learning_rate": 0.00021862393545495293, + "loss": 0.3949, + "step": 28013 + }, + { + "epoch": 1.565187026845825, + "grad_norm": 0.42909717559814453, + "learning_rate": 0.00021859592111160914, + "loss": 0.4607, + "step": 28014 + }, + { + "epoch": 1.565242897449507, + "grad_norm": 0.4572175145149231, + "learning_rate": 0.00021856790676826534, + "loss": 0.4024, + "step": 28015 + }, + { + "epoch": 1.565298768053189, + "grad_norm": 0.35807931423187256, + "learning_rate": 0.00021853989242492155, + "loss": 0.3184, + "step": 28016 + }, + { + "epoch": 1.5653546386568706, + "grad_norm": 1.223483681678772, + "learning_rate": 0.00021851187808157778, + "loss": 0.5194, + "step": 28017 + }, + { + "epoch": 1.5654105092605526, + "grad_norm": 0.37996891140937805, + "learning_rate": 0.000218483863738234, + "loss": 0.2505, + "step": 28018 + }, + { + "epoch": 1.5654663798642345, + "grad_norm": 0.854547381401062, + "learning_rate": 0.0002184558493948902, + "loss": 0.3915, + "step": 28019 + }, + { + "epoch": 1.5655222504679163, + "grad_norm": 0.5434274077415466, + "learning_rate": 0.0002184278350515464, + "loss": 0.3458, + "step": 28020 + }, + { + "epoch": 1.5655781210715982, + "grad_norm": 0.5103692412376404, + "learning_rate": 0.0002183998207082026, + "loss": 0.4129, + "step": 28021 + }, + { + "epoch": 1.5656339916752802, + "grad_norm": 0.30027779936790466, + "learning_rate": 0.0002183718063648588, + "loss": 0.3865, + "step": 28022 + }, + { + "epoch": 1.565689862278962, + "grad_norm": 0.36522310972213745, + "learning_rate": 0.00021834379202151502, + "loss": 0.5184, + "step": 28023 + }, + { + "epoch": 1.5657457328826438, + "grad_norm": 0.9711365699768066, + "learning_rate": 0.00021831577767817122, + "loss": 0.3854, + "step": 28024 + }, + { + "epoch": 1.5658016034863258, + "grad_norm": 1.0923025608062744, + "learning_rate": 0.00021828776333482745, + "loss": 0.4354, + "step": 28025 + }, + { + "epoch": 1.5658574740900075, + "grad_norm": 0.354219526052475, + "learning_rate": 0.00021825974899148366, + "loss": 0.3417, + "step": 28026 + }, + { + "epoch": 1.5659133446936893, + "grad_norm": 0.6516990661621094, + "learning_rate": 0.00021823173464813987, + "loss": 0.3619, + "step": 28027 + }, + { + "epoch": 1.5659692152973714, + "grad_norm": 0.43487656116485596, + "learning_rate": 0.00021820372030479607, + "loss": 0.4966, + "step": 28028 + }, + { + "epoch": 1.5660250859010532, + "grad_norm": 1.3294583559036255, + "learning_rate": 0.00021817570596145228, + "loss": 0.5348, + "step": 28029 + }, + { + "epoch": 1.566080956504735, + "grad_norm": 0.7751559019088745, + "learning_rate": 0.00021814769161810846, + "loss": 0.4123, + "step": 28030 + }, + { + "epoch": 1.566136827108417, + "grad_norm": 0.3905678391456604, + "learning_rate": 0.00021811967727476466, + "loss": 0.4455, + "step": 28031 + }, + { + "epoch": 1.5661926977120988, + "grad_norm": 0.40456050634384155, + "learning_rate": 0.00021809166293142087, + "loss": 0.4731, + "step": 28032 + }, + { + "epoch": 1.5662485683157805, + "grad_norm": 0.4156266152858734, + "learning_rate": 0.0002180636485880771, + "loss": 0.479, + "step": 28033 + }, + { + "epoch": 1.5663044389194625, + "grad_norm": 11.762178421020508, + "learning_rate": 0.0002180356342447333, + "loss": 0.6118, + "step": 28034 + }, + { + "epoch": 1.5663603095231444, + "grad_norm": 0.8897059559822083, + "learning_rate": 0.0002180076199013895, + "loss": 0.3943, + "step": 28035 + }, + { + "epoch": 1.5664161801268262, + "grad_norm": 0.44851070642471313, + "learning_rate": 0.00021797960555804572, + "loss": 0.3827, + "step": 28036 + }, + { + "epoch": 1.566472050730508, + "grad_norm": 1.3507859706878662, + "learning_rate": 0.00021795159121470192, + "loss": 0.3871, + "step": 28037 + }, + { + "epoch": 1.56652792133419, + "grad_norm": 0.4262169301509857, + "learning_rate": 0.00021792357687135813, + "loss": 0.4048, + "step": 28038 + }, + { + "epoch": 1.5665837919378718, + "grad_norm": 0.6300052404403687, + "learning_rate": 0.00021789556252801434, + "loss": 0.3897, + "step": 28039 + }, + { + "epoch": 1.5666396625415537, + "grad_norm": 0.7541008591651917, + "learning_rate": 0.00021786754818467054, + "loss": 0.4283, + "step": 28040 + }, + { + "epoch": 1.5666955331452357, + "grad_norm": 0.44931650161743164, + "learning_rate": 0.00021783953384132677, + "loss": 0.3363, + "step": 28041 + }, + { + "epoch": 1.5667514037489174, + "grad_norm": 0.9536195993423462, + "learning_rate": 0.00021781151949798298, + "loss": 0.4321, + "step": 28042 + }, + { + "epoch": 1.5668072743525994, + "grad_norm": 0.5392744541168213, + "learning_rate": 0.0002177835051546392, + "loss": 0.4648, + "step": 28043 + }, + { + "epoch": 1.5668631449562813, + "grad_norm": 1.28699791431427, + "learning_rate": 0.0002177554908112954, + "loss": 0.3132, + "step": 28044 + }, + { + "epoch": 1.566919015559963, + "grad_norm": 3.3238134384155273, + "learning_rate": 0.0002177274764679516, + "loss": 0.385, + "step": 28045 + }, + { + "epoch": 1.566974886163645, + "grad_norm": 0.30162113904953003, + "learning_rate": 0.0002176994621246078, + "loss": 0.4489, + "step": 28046 + }, + { + "epoch": 1.567030756767327, + "grad_norm": 0.35757848620414734, + "learning_rate": 0.000217671447781264, + "loss": 0.4497, + "step": 28047 + }, + { + "epoch": 1.5670866273710087, + "grad_norm": 0.46153542399406433, + "learning_rate": 0.00021764343343792024, + "loss": 0.6103, + "step": 28048 + }, + { + "epoch": 1.5671424979746906, + "grad_norm": 0.3661847412586212, + "learning_rate": 0.00021761541909457645, + "loss": 0.3712, + "step": 28049 + }, + { + "epoch": 1.5671983685783726, + "grad_norm": 1.7007319927215576, + "learning_rate": 0.00021758740475123265, + "loss": 0.4199, + "step": 28050 + }, + { + "epoch": 1.5672542391820543, + "grad_norm": 0.9392789006233215, + "learning_rate": 0.00021755939040788883, + "loss": 0.4083, + "step": 28051 + }, + { + "epoch": 1.5673101097857363, + "grad_norm": 0.5345627069473267, + "learning_rate": 0.00021753137606454504, + "loss": 0.4054, + "step": 28052 + }, + { + "epoch": 1.5673659803894182, + "grad_norm": 0.3528691828250885, + "learning_rate": 0.00021750336172120124, + "loss": 0.3334, + "step": 28053 + }, + { + "epoch": 1.5674218509931, + "grad_norm": 0.4060620367527008, + "learning_rate": 0.00021747534737785745, + "loss": 0.4513, + "step": 28054 + }, + { + "epoch": 1.567477721596782, + "grad_norm": 0.9715126156806946, + "learning_rate": 0.00021744733303451366, + "loss": 0.3908, + "step": 28055 + }, + { + "epoch": 1.5675335922004638, + "grad_norm": 0.5250000357627869, + "learning_rate": 0.0002174193186911699, + "loss": 0.3937, + "step": 28056 + }, + { + "epoch": 1.5675894628041456, + "grad_norm": 0.46161365509033203, + "learning_rate": 0.0002173913043478261, + "loss": 0.3397, + "step": 28057 + }, + { + "epoch": 1.5676453334078275, + "grad_norm": 0.41274040937423706, + "learning_rate": 0.0002173632900044823, + "loss": 0.4057, + "step": 28058 + }, + { + "epoch": 1.5677012040115095, + "grad_norm": 0.34507516026496887, + "learning_rate": 0.0002173352756611385, + "loss": 0.4366, + "step": 28059 + }, + { + "epoch": 1.5677570746151912, + "grad_norm": 0.3797546923160553, + "learning_rate": 0.0002173072613177947, + "loss": 0.4779, + "step": 28060 + }, + { + "epoch": 1.567812945218873, + "grad_norm": 0.3603557348251343, + "learning_rate": 0.00021727924697445092, + "loss": 0.4167, + "step": 28061 + }, + { + "epoch": 1.567868815822555, + "grad_norm": 0.41335564851760864, + "learning_rate": 0.00021725123263110712, + "loss": 0.4085, + "step": 28062 + }, + { + "epoch": 1.5679246864262368, + "grad_norm": 0.35623911023139954, + "learning_rate": 0.00021722321828776333, + "loss": 0.4649, + "step": 28063 + }, + { + "epoch": 1.5679805570299186, + "grad_norm": 0.3632233440876007, + "learning_rate": 0.00021719520394441956, + "loss": 0.3783, + "step": 28064 + }, + { + "epoch": 1.5680364276336007, + "grad_norm": 3.6254470348358154, + "learning_rate": 0.00021716718960107577, + "loss": 0.3511, + "step": 28065 + }, + { + "epoch": 1.5680922982372825, + "grad_norm": 0.269849568605423, + "learning_rate": 0.00021713917525773197, + "loss": 0.3153, + "step": 28066 + }, + { + "epoch": 1.5681481688409642, + "grad_norm": 1.0021249055862427, + "learning_rate": 0.00021711116091438818, + "loss": 0.4058, + "step": 28067 + }, + { + "epoch": 1.5682040394446461, + "grad_norm": 1.3669520616531372, + "learning_rate": 0.00021708314657104439, + "loss": 0.4139, + "step": 28068 + }, + { + "epoch": 1.568259910048328, + "grad_norm": 0.34770381450653076, + "learning_rate": 0.0002170551322277006, + "loss": 0.2867, + "step": 28069 + }, + { + "epoch": 1.5683157806520098, + "grad_norm": 0.8175486922264099, + "learning_rate": 0.0002170271178843568, + "loss": 0.5005, + "step": 28070 + }, + { + "epoch": 1.5683716512556918, + "grad_norm": 0.47642871737480164, + "learning_rate": 0.000216999103541013, + "loss": 0.4685, + "step": 28071 + }, + { + "epoch": 1.5684275218593737, + "grad_norm": 0.4907744228839874, + "learning_rate": 0.00021697108919766924, + "loss": 0.4029, + "step": 28072 + }, + { + "epoch": 1.5684833924630555, + "grad_norm": 0.412636399269104, + "learning_rate": 0.00021694307485432542, + "loss": 0.5054, + "step": 28073 + }, + { + "epoch": 1.5685392630667374, + "grad_norm": 0.4425790011882782, + "learning_rate": 0.00021691506051098162, + "loss": 0.4297, + "step": 28074 + }, + { + "epoch": 1.5685951336704194, + "grad_norm": 1.0023887157440186, + "learning_rate": 0.00021688704616763783, + "loss": 0.4579, + "step": 28075 + }, + { + "epoch": 1.568651004274101, + "grad_norm": 0.5070473551750183, + "learning_rate": 0.00021685903182429403, + "loss": 0.5241, + "step": 28076 + }, + { + "epoch": 1.568706874877783, + "grad_norm": 0.3793185353279114, + "learning_rate": 0.00021683101748095024, + "loss": 0.3484, + "step": 28077 + }, + { + "epoch": 1.568762745481465, + "grad_norm": 1.3627146482467651, + "learning_rate": 0.00021680300313760644, + "loss": 0.3626, + "step": 28078 + }, + { + "epoch": 1.5688186160851467, + "grad_norm": 1.3278921842575073, + "learning_rate": 0.00021677498879426265, + "loss": 0.4271, + "step": 28079 + }, + { + "epoch": 1.5688744866888287, + "grad_norm": 0.5253027677536011, + "learning_rate": 0.00021674697445091888, + "loss": 0.472, + "step": 28080 + }, + { + "epoch": 1.5689303572925106, + "grad_norm": 1.4357789754867554, + "learning_rate": 0.0002167189601075751, + "loss": 0.5141, + "step": 28081 + }, + { + "epoch": 1.5689862278961924, + "grad_norm": 0.4282044470310211, + "learning_rate": 0.0002166909457642313, + "loss": 0.4383, + "step": 28082 + }, + { + "epoch": 1.5690420984998743, + "grad_norm": 0.49623173475265503, + "learning_rate": 0.0002166629314208875, + "loss": 0.3823, + "step": 28083 + }, + { + "epoch": 1.5690979691035563, + "grad_norm": 0.438210129737854, + "learning_rate": 0.0002166349170775437, + "loss": 0.3481, + "step": 28084 + }, + { + "epoch": 1.569153839707238, + "grad_norm": 0.6386687755584717, + "learning_rate": 0.0002166069027341999, + "loss": 0.4991, + "step": 28085 + }, + { + "epoch": 1.56920971031092, + "grad_norm": 2.115889072418213, + "learning_rate": 0.00021657888839085612, + "loss": 0.5128, + "step": 28086 + }, + { + "epoch": 1.5692655809146019, + "grad_norm": 0.44735920429229736, + "learning_rate": 0.00021655087404751232, + "loss": 0.4148, + "step": 28087 + }, + { + "epoch": 1.5693214515182836, + "grad_norm": 0.4042760729789734, + "learning_rate": 0.00021652285970416856, + "loss": 0.4467, + "step": 28088 + }, + { + "epoch": 1.5693773221219656, + "grad_norm": 0.6852260231971741, + "learning_rate": 0.00021649484536082476, + "loss": 0.4541, + "step": 28089 + }, + { + "epoch": 1.5694331927256475, + "grad_norm": 0.4396273195743561, + "learning_rate": 0.00021646683101748097, + "loss": 0.3938, + "step": 28090 + }, + { + "epoch": 1.5694890633293292, + "grad_norm": 1.4993504285812378, + "learning_rate": 0.00021643881667413717, + "loss": 0.3387, + "step": 28091 + }, + { + "epoch": 1.5695449339330112, + "grad_norm": 1.4420117139816284, + "learning_rate": 0.00021641080233079338, + "loss": 0.6576, + "step": 28092 + }, + { + "epoch": 1.5696008045366932, + "grad_norm": 0.5825515985488892, + "learning_rate": 0.00021638278798744959, + "loss": 0.5595, + "step": 28093 + }, + { + "epoch": 1.5696566751403749, + "grad_norm": 0.3876208961009979, + "learning_rate": 0.00021635477364410576, + "loss": 0.3571, + "step": 28094 + }, + { + "epoch": 1.5697125457440566, + "grad_norm": 0.6045044660568237, + "learning_rate": 0.00021632675930076197, + "loss": 0.4796, + "step": 28095 + }, + { + "epoch": 1.5697684163477388, + "grad_norm": 0.3736931085586548, + "learning_rate": 0.0002162987449574182, + "loss": 0.4397, + "step": 28096 + }, + { + "epoch": 1.5698242869514205, + "grad_norm": 0.5125455856323242, + "learning_rate": 0.0002162707306140744, + "loss": 0.43, + "step": 28097 + }, + { + "epoch": 1.5698801575551022, + "grad_norm": 0.4251074492931366, + "learning_rate": 0.00021624271627073061, + "loss": 0.5542, + "step": 28098 + }, + { + "epoch": 1.5699360281587844, + "grad_norm": 0.40716055035591125, + "learning_rate": 0.00021621470192738682, + "loss": 0.4288, + "step": 28099 + }, + { + "epoch": 1.5699918987624661, + "grad_norm": 0.4488310217857361, + "learning_rate": 0.00021618668758404303, + "loss": 0.3314, + "step": 28100 + }, + { + "epoch": 1.5700477693661479, + "grad_norm": 0.5321462750434875, + "learning_rate": 0.00021615867324069923, + "loss": 0.3422, + "step": 28101 + }, + { + "epoch": 1.5701036399698298, + "grad_norm": 0.4727156162261963, + "learning_rate": 0.00021613065889735544, + "loss": 0.4366, + "step": 28102 + }, + { + "epoch": 1.5701595105735118, + "grad_norm": 0.4217137098312378, + "learning_rate": 0.00021610264455401164, + "loss": 0.4366, + "step": 28103 + }, + { + "epoch": 1.5702153811771935, + "grad_norm": 1.092404842376709, + "learning_rate": 0.00021607463021066788, + "loss": 0.5725, + "step": 28104 + }, + { + "epoch": 1.5702712517808755, + "grad_norm": 0.3653077185153961, + "learning_rate": 0.00021604661586732408, + "loss": 0.4615, + "step": 28105 + }, + { + "epoch": 1.5703271223845574, + "grad_norm": 0.4540639817714691, + "learning_rate": 0.0002160186015239803, + "loss": 0.3663, + "step": 28106 + }, + { + "epoch": 1.5703829929882391, + "grad_norm": 0.6854623556137085, + "learning_rate": 0.0002159905871806365, + "loss": 0.4717, + "step": 28107 + }, + { + "epoch": 1.570438863591921, + "grad_norm": 0.6545155644416809, + "learning_rate": 0.0002159625728372927, + "loss": 0.6106, + "step": 28108 + }, + { + "epoch": 1.570494734195603, + "grad_norm": 0.6117077469825745, + "learning_rate": 0.0002159345584939489, + "loss": 0.4179, + "step": 28109 + }, + { + "epoch": 1.5705506047992848, + "grad_norm": 0.44257852435112, + "learning_rate": 0.0002159065441506051, + "loss": 0.5304, + "step": 28110 + }, + { + "epoch": 1.5706064754029667, + "grad_norm": 0.4888863265514374, + "learning_rate": 0.00021587852980726134, + "loss": 0.3841, + "step": 28111 + }, + { + "epoch": 1.5706623460066487, + "grad_norm": 0.31002306938171387, + "learning_rate": 0.00021585051546391755, + "loss": 0.3334, + "step": 28112 + }, + { + "epoch": 1.5707182166103304, + "grad_norm": 0.3845265507698059, + "learning_rate": 0.00021582250112057376, + "loss": 0.4394, + "step": 28113 + }, + { + "epoch": 1.5707740872140123, + "grad_norm": 1.6208219528198242, + "learning_rate": 0.00021579448677722996, + "loss": 0.4591, + "step": 28114 + }, + { + "epoch": 1.5708299578176943, + "grad_norm": 0.46600204706192017, + "learning_rate": 0.00021576647243388614, + "loss": 0.4339, + "step": 28115 + }, + { + "epoch": 1.570885828421376, + "grad_norm": 0.8980640172958374, + "learning_rate": 0.00021573845809054235, + "loss": 0.5301, + "step": 28116 + }, + { + "epoch": 1.570941699025058, + "grad_norm": 1.4715347290039062, + "learning_rate": 0.00021571044374719855, + "loss": 0.3857, + "step": 28117 + }, + { + "epoch": 1.57099756962874, + "grad_norm": 0.7349167466163635, + "learning_rate": 0.00021568242940385476, + "loss": 0.3927, + "step": 28118 + }, + { + "epoch": 1.5710534402324217, + "grad_norm": 5.26799201965332, + "learning_rate": 0.000215654415060511, + "loss": 0.3578, + "step": 28119 + }, + { + "epoch": 1.5711093108361036, + "grad_norm": 2.0530219078063965, + "learning_rate": 0.0002156264007171672, + "loss": 0.4103, + "step": 28120 + }, + { + "epoch": 1.5711651814397856, + "grad_norm": 0.735892117023468, + "learning_rate": 0.0002155983863738234, + "loss": 0.4175, + "step": 28121 + }, + { + "epoch": 1.5712210520434673, + "grad_norm": 0.500457227230072, + "learning_rate": 0.0002155703720304796, + "loss": 0.4319, + "step": 28122 + }, + { + "epoch": 1.5712769226471492, + "grad_norm": 0.46319398283958435, + "learning_rate": 0.00021554235768713581, + "loss": 0.3147, + "step": 28123 + }, + { + "epoch": 1.5713327932508312, + "grad_norm": 0.3931727409362793, + "learning_rate": 0.00021551434334379202, + "loss": 0.3304, + "step": 28124 + }, + { + "epoch": 1.571388663854513, + "grad_norm": 0.4785924553871155, + "learning_rate": 0.00021548632900044823, + "loss": 0.4037, + "step": 28125 + }, + { + "epoch": 1.5714445344581949, + "grad_norm": 0.49654272198677063, + "learning_rate": 0.00021545831465710443, + "loss": 0.5006, + "step": 28126 + }, + { + "epoch": 1.5715004050618768, + "grad_norm": 0.4974437952041626, + "learning_rate": 0.00021543030031376066, + "loss": 0.4331, + "step": 28127 + }, + { + "epoch": 1.5715562756655586, + "grad_norm": 0.348550409078598, + "learning_rate": 0.00021540228597041687, + "loss": 0.4132, + "step": 28128 + }, + { + "epoch": 1.5716121462692403, + "grad_norm": 0.6056678891181946, + "learning_rate": 0.00021537427162707308, + "loss": 0.4018, + "step": 28129 + }, + { + "epoch": 1.5716680168729225, + "grad_norm": 0.4196085035800934, + "learning_rate": 0.00021534625728372928, + "loss": 0.5336, + "step": 28130 + }, + { + "epoch": 1.5717238874766042, + "grad_norm": 0.6143171787261963, + "learning_rate": 0.0002153182429403855, + "loss": 0.3592, + "step": 28131 + }, + { + "epoch": 1.571779758080286, + "grad_norm": 0.47294285893440247, + "learning_rate": 0.0002152902285970417, + "loss": 0.379, + "step": 28132 + }, + { + "epoch": 1.571835628683968, + "grad_norm": 0.6619939804077148, + "learning_rate": 0.0002152622142536979, + "loss": 0.5253, + "step": 28133 + }, + { + "epoch": 1.5718914992876498, + "grad_norm": 0.8860530257225037, + "learning_rate": 0.0002152341999103541, + "loss": 0.4255, + "step": 28134 + }, + { + "epoch": 1.5719473698913315, + "grad_norm": 0.5387077927589417, + "learning_rate": 0.00021520618556701034, + "loss": 0.4799, + "step": 28135 + }, + { + "epoch": 1.5720032404950135, + "grad_norm": 0.4190707504749298, + "learning_rate": 0.00021517817122366652, + "loss": 0.397, + "step": 28136 + }, + { + "epoch": 1.5720591110986954, + "grad_norm": 0.40049460530281067, + "learning_rate": 0.00021515015688032272, + "loss": 0.3555, + "step": 28137 + }, + { + "epoch": 1.5721149817023772, + "grad_norm": 0.4603278338909149, + "learning_rate": 0.00021512214253697893, + "loss": 0.3189, + "step": 28138 + }, + { + "epoch": 1.5721708523060591, + "grad_norm": 0.5537460446357727, + "learning_rate": 0.00021509412819363513, + "loss": 0.4516, + "step": 28139 + }, + { + "epoch": 1.572226722909741, + "grad_norm": 2.667372226715088, + "learning_rate": 0.00021506611385029134, + "loss": 0.3905, + "step": 28140 + }, + { + "epoch": 1.5722825935134228, + "grad_norm": 0.5092966556549072, + "learning_rate": 0.00021503809950694755, + "loss": 0.4167, + "step": 28141 + }, + { + "epoch": 1.5723384641171048, + "grad_norm": 0.42826539278030396, + "learning_rate": 0.00021501008516360375, + "loss": 0.3869, + "step": 28142 + }, + { + "epoch": 1.5723943347207867, + "grad_norm": 0.7356235384941101, + "learning_rate": 0.00021498207082025998, + "loss": 0.3152, + "step": 28143 + }, + { + "epoch": 1.5724502053244684, + "grad_norm": 0.4677817225456238, + "learning_rate": 0.0002149540564769162, + "loss": 0.3636, + "step": 28144 + }, + { + "epoch": 1.5725060759281504, + "grad_norm": 0.4099673926830292, + "learning_rate": 0.0002149260421335724, + "loss": 0.3344, + "step": 28145 + }, + { + "epoch": 1.5725619465318323, + "grad_norm": 0.5502830147743225, + "learning_rate": 0.0002148980277902286, + "loss": 0.4519, + "step": 28146 + }, + { + "epoch": 1.572617817135514, + "grad_norm": 0.33510497212409973, + "learning_rate": 0.0002148700134468848, + "loss": 0.3999, + "step": 28147 + }, + { + "epoch": 1.572673687739196, + "grad_norm": 0.9095263481140137, + "learning_rate": 0.00021484199910354101, + "loss": 0.4711, + "step": 28148 + }, + { + "epoch": 1.572729558342878, + "grad_norm": 0.49827268719673157, + "learning_rate": 0.00021481398476019722, + "loss": 0.3807, + "step": 28149 + }, + { + "epoch": 1.5727854289465597, + "grad_norm": 0.48631027340888977, + "learning_rate": 0.00021478597041685343, + "loss": 0.391, + "step": 28150 + }, + { + "epoch": 1.5728412995502417, + "grad_norm": 0.687689483165741, + "learning_rate": 0.00021475795607350966, + "loss": 0.3827, + "step": 28151 + }, + { + "epoch": 1.5728971701539236, + "grad_norm": 0.5495401620864868, + "learning_rate": 0.00021472994173016586, + "loss": 0.4486, + "step": 28152 + }, + { + "epoch": 1.5729530407576053, + "grad_norm": 0.43399354815483093, + "learning_rate": 0.00021470192738682207, + "loss": 0.4394, + "step": 28153 + }, + { + "epoch": 1.5730089113612873, + "grad_norm": 0.3161948621273041, + "learning_rate": 0.00021467391304347828, + "loss": 0.3397, + "step": 28154 + }, + { + "epoch": 1.5730647819649692, + "grad_norm": 0.42214182019233704, + "learning_rate": 0.00021464589870013448, + "loss": 0.4248, + "step": 28155 + }, + { + "epoch": 1.573120652568651, + "grad_norm": 0.6653305292129517, + "learning_rate": 0.0002146178843567907, + "loss": 0.4156, + "step": 28156 + }, + { + "epoch": 1.573176523172333, + "grad_norm": 0.3812558650970459, + "learning_rate": 0.0002145898700134469, + "loss": 0.4326, + "step": 28157 + }, + { + "epoch": 1.5732323937760149, + "grad_norm": 0.6561188697814941, + "learning_rate": 0.00021456185567010307, + "loss": 0.392, + "step": 28158 + }, + { + "epoch": 1.5732882643796966, + "grad_norm": 0.4629736840724945, + "learning_rate": 0.0002145338413267593, + "loss": 0.5804, + "step": 28159 + }, + { + "epoch": 1.5733441349833786, + "grad_norm": 0.9142897129058838, + "learning_rate": 0.0002145058269834155, + "loss": 0.4422, + "step": 28160 + }, + { + "epoch": 1.5734000055870605, + "grad_norm": 0.321818083524704, + "learning_rate": 0.00021447781264007172, + "loss": 0.3573, + "step": 28161 + }, + { + "epoch": 1.5734558761907422, + "grad_norm": 0.3831825852394104, + "learning_rate": 0.00021444979829672792, + "loss": 0.3985, + "step": 28162 + }, + { + "epoch": 1.573511746794424, + "grad_norm": 1.9458755254745483, + "learning_rate": 0.00021442178395338413, + "loss": 0.4099, + "step": 28163 + }, + { + "epoch": 1.5735676173981061, + "grad_norm": 0.4665021598339081, + "learning_rate": 0.00021439376961004033, + "loss": 0.5037, + "step": 28164 + }, + { + "epoch": 1.5736234880017879, + "grad_norm": 0.37502291798591614, + "learning_rate": 0.00021436575526669654, + "loss": 0.452, + "step": 28165 + }, + { + "epoch": 1.5736793586054696, + "grad_norm": 1.0980666875839233, + "learning_rate": 0.00021433774092335275, + "loss": 0.4353, + "step": 28166 + }, + { + "epoch": 1.5737352292091518, + "grad_norm": 0.44995665550231934, + "learning_rate": 0.00021430972658000898, + "loss": 0.4076, + "step": 28167 + }, + { + "epoch": 1.5737910998128335, + "grad_norm": 0.3927735686302185, + "learning_rate": 0.00021428171223666518, + "loss": 0.4218, + "step": 28168 + }, + { + "epoch": 1.5738469704165152, + "grad_norm": 0.30716532468795776, + "learning_rate": 0.0002142536978933214, + "loss": 0.3692, + "step": 28169 + }, + { + "epoch": 1.5739028410201972, + "grad_norm": 0.3031883239746094, + "learning_rate": 0.0002142256835499776, + "loss": 0.3226, + "step": 28170 + }, + { + "epoch": 1.5739587116238791, + "grad_norm": 0.4485800564289093, + "learning_rate": 0.0002141976692066338, + "loss": 0.4611, + "step": 28171 + }, + { + "epoch": 1.5740145822275609, + "grad_norm": 0.35489875078201294, + "learning_rate": 0.00021416965486329, + "loss": 0.403, + "step": 28172 + }, + { + "epoch": 1.5740704528312428, + "grad_norm": 0.6194885969161987, + "learning_rate": 0.0002141416405199462, + "loss": 0.4172, + "step": 28173 + }, + { + "epoch": 1.5741263234349248, + "grad_norm": 0.46086105704307556, + "learning_rate": 0.00021411362617660245, + "loss": 0.3947, + "step": 28174 + }, + { + "epoch": 1.5741821940386065, + "grad_norm": 2.794111967086792, + "learning_rate": 0.00021408561183325865, + "loss": 0.3618, + "step": 28175 + }, + { + "epoch": 1.5742380646422884, + "grad_norm": 0.546217143535614, + "learning_rate": 0.00021405759748991486, + "loss": 0.3495, + "step": 28176 + }, + { + "epoch": 1.5742939352459704, + "grad_norm": 1.4225374460220337, + "learning_rate": 0.00021402958314657106, + "loss": 0.4213, + "step": 28177 + }, + { + "epoch": 1.5743498058496521, + "grad_norm": 1.4326903820037842, + "learning_rate": 0.00021400156880322727, + "loss": 0.3919, + "step": 28178 + }, + { + "epoch": 1.574405676453334, + "grad_norm": 0.3825998604297638, + "learning_rate": 0.00021397355445988345, + "loss": 0.3616, + "step": 28179 + }, + { + "epoch": 1.574461547057016, + "grad_norm": 0.5123613476753235, + "learning_rate": 0.00021394554011653965, + "loss": 0.459, + "step": 28180 + }, + { + "epoch": 1.5745174176606977, + "grad_norm": 0.5077981352806091, + "learning_rate": 0.00021391752577319586, + "loss": 0.5484, + "step": 28181 + }, + { + "epoch": 1.5745732882643797, + "grad_norm": 0.3642744719982147, + "learning_rate": 0.0002138895114298521, + "loss": 0.4326, + "step": 28182 + }, + { + "epoch": 1.5746291588680617, + "grad_norm": 0.5717699527740479, + "learning_rate": 0.0002138614970865083, + "loss": 0.4179, + "step": 28183 + }, + { + "epoch": 1.5746850294717434, + "grad_norm": 0.3815142512321472, + "learning_rate": 0.0002138334827431645, + "loss": 0.3734, + "step": 28184 + }, + { + "epoch": 1.5747409000754253, + "grad_norm": 0.4506058096885681, + "learning_rate": 0.0002138054683998207, + "loss": 0.5724, + "step": 28185 + }, + { + "epoch": 1.5747967706791073, + "grad_norm": 0.7609546184539795, + "learning_rate": 0.00021377745405647692, + "loss": 0.6489, + "step": 28186 + }, + { + "epoch": 1.574852641282789, + "grad_norm": 0.3866475224494934, + "learning_rate": 0.00021374943971313312, + "loss": 0.3481, + "step": 28187 + }, + { + "epoch": 1.574908511886471, + "grad_norm": 0.6074244976043701, + "learning_rate": 0.00021372142536978933, + "loss": 0.3724, + "step": 28188 + }, + { + "epoch": 1.574964382490153, + "grad_norm": 0.4053474962711334, + "learning_rate": 0.00021369341102644553, + "loss": 0.501, + "step": 28189 + }, + { + "epoch": 1.5750202530938346, + "grad_norm": 0.35632234811782837, + "learning_rate": 0.00021366539668310177, + "loss": 0.3627, + "step": 28190 + }, + { + "epoch": 1.5750761236975166, + "grad_norm": 0.4881342947483063, + "learning_rate": 0.00021363738233975797, + "loss": 0.429, + "step": 28191 + }, + { + "epoch": 1.5751319943011985, + "grad_norm": 0.3992152512073517, + "learning_rate": 0.00021360936799641418, + "loss": 0.4008, + "step": 28192 + }, + { + "epoch": 1.5751878649048803, + "grad_norm": 0.40102341771125793, + "learning_rate": 0.00021358135365307038, + "loss": 0.4084, + "step": 28193 + }, + { + "epoch": 1.575243735508562, + "grad_norm": 0.9674350023269653, + "learning_rate": 0.0002135533393097266, + "loss": 0.2911, + "step": 28194 + }, + { + "epoch": 1.5752996061122442, + "grad_norm": 0.6276465654373169, + "learning_rate": 0.0002135253249663828, + "loss": 0.548, + "step": 28195 + }, + { + "epoch": 1.575355476715926, + "grad_norm": 0.820789098739624, + "learning_rate": 0.000213497310623039, + "loss": 0.4507, + "step": 28196 + }, + { + "epoch": 1.5754113473196076, + "grad_norm": 0.5481900572776794, + "learning_rate": 0.0002134692962796952, + "loss": 0.4673, + "step": 28197 + }, + { + "epoch": 1.5754672179232898, + "grad_norm": 0.4085002839565277, + "learning_rate": 0.00021344128193635144, + "loss": 0.388, + "step": 28198 + }, + { + "epoch": 1.5755230885269715, + "grad_norm": 0.4985540509223938, + "learning_rate": 0.00021341326759300765, + "loss": 0.4015, + "step": 28199 + }, + { + "epoch": 1.5755789591306533, + "grad_norm": 0.5445350408554077, + "learning_rate": 0.00021338525324966382, + "loss": 0.4974, + "step": 28200 + }, + { + "epoch": 1.5756348297343354, + "grad_norm": 0.39124229550361633, + "learning_rate": 0.00021335723890632003, + "loss": 0.375, + "step": 28201 + }, + { + "epoch": 1.5756907003380172, + "grad_norm": 0.547557532787323, + "learning_rate": 0.00021332922456297624, + "loss": 0.4187, + "step": 28202 + }, + { + "epoch": 1.575746570941699, + "grad_norm": 0.6620389819145203, + "learning_rate": 0.00021330121021963244, + "loss": 0.3178, + "step": 28203 + }, + { + "epoch": 1.5758024415453808, + "grad_norm": 0.4563360810279846, + "learning_rate": 0.00021327319587628865, + "loss": 0.4609, + "step": 28204 + }, + { + "epoch": 1.5758583121490628, + "grad_norm": 0.616050660610199, + "learning_rate": 0.00021324518153294485, + "loss": 0.4169, + "step": 28205 + }, + { + "epoch": 1.5759141827527445, + "grad_norm": 0.4062947928905487, + "learning_rate": 0.0002132171671896011, + "loss": 0.3907, + "step": 28206 + }, + { + "epoch": 1.5759700533564265, + "grad_norm": 0.44790685176849365, + "learning_rate": 0.0002131891528462573, + "loss": 0.4308, + "step": 28207 + }, + { + "epoch": 1.5760259239601084, + "grad_norm": 0.5094790458679199, + "learning_rate": 0.0002131611385029135, + "loss": 0.3344, + "step": 28208 + }, + { + "epoch": 1.5760817945637902, + "grad_norm": 0.6451693773269653, + "learning_rate": 0.0002131331241595697, + "loss": 0.5556, + "step": 28209 + }, + { + "epoch": 1.5761376651674721, + "grad_norm": 0.3814452588558197, + "learning_rate": 0.0002131051098162259, + "loss": 0.4436, + "step": 28210 + }, + { + "epoch": 1.576193535771154, + "grad_norm": 0.2926769554615021, + "learning_rate": 0.00021307709547288212, + "loss": 0.3549, + "step": 28211 + }, + { + "epoch": 1.5762494063748358, + "grad_norm": 0.5750218033790588, + "learning_rate": 0.00021304908112953832, + "loss": 0.3537, + "step": 28212 + }, + { + "epoch": 1.5763052769785177, + "grad_norm": 0.36473339796066284, + "learning_rate": 0.00021302106678619453, + "loss": 0.3708, + "step": 28213 + }, + { + "epoch": 1.5763611475821997, + "grad_norm": 0.7333546876907349, + "learning_rate": 0.00021299305244285076, + "loss": 0.3445, + "step": 28214 + }, + { + "epoch": 1.5764170181858814, + "grad_norm": 0.5160609483718872, + "learning_rate": 0.00021296503809950697, + "loss": 0.6063, + "step": 28215 + }, + { + "epoch": 1.5764728887895634, + "grad_norm": 0.5506553649902344, + "learning_rate": 0.00021293702375616317, + "loss": 0.6599, + "step": 28216 + }, + { + "epoch": 1.5765287593932453, + "grad_norm": 1.209640622138977, + "learning_rate": 0.00021290900941281938, + "loss": 0.3639, + "step": 28217 + }, + { + "epoch": 1.576584629996927, + "grad_norm": 0.43067964911460876, + "learning_rate": 0.00021288099506947558, + "loss": 0.4004, + "step": 28218 + }, + { + "epoch": 1.576640500600609, + "grad_norm": 0.39025232195854187, + "learning_rate": 0.0002128529807261318, + "loss": 0.4711, + "step": 28219 + }, + { + "epoch": 1.576696371204291, + "grad_norm": 0.440579891204834, + "learning_rate": 0.000212824966382788, + "loss": 0.3912, + "step": 28220 + }, + { + "epoch": 1.5767522418079727, + "grad_norm": 0.6970080137252808, + "learning_rate": 0.00021279695203944417, + "loss": 0.4028, + "step": 28221 + }, + { + "epoch": 1.5768081124116546, + "grad_norm": 1.493672251701355, + "learning_rate": 0.0002127689376961004, + "loss": 0.3691, + "step": 28222 + }, + { + "epoch": 1.5768639830153366, + "grad_norm": 0.39445891976356506, + "learning_rate": 0.0002127409233527566, + "loss": 0.4243, + "step": 28223 + }, + { + "epoch": 1.5769198536190183, + "grad_norm": 0.6736710071563721, + "learning_rate": 0.00021271290900941282, + "loss": 0.4413, + "step": 28224 + }, + { + "epoch": 1.5769757242227003, + "grad_norm": 0.5819727182388306, + "learning_rate": 0.00021268489466606902, + "loss": 0.455, + "step": 28225 + }, + { + "epoch": 1.5770315948263822, + "grad_norm": 0.6811338067054749, + "learning_rate": 0.00021265688032272523, + "loss": 0.406, + "step": 28226 + }, + { + "epoch": 1.577087465430064, + "grad_norm": 0.41075778007507324, + "learning_rate": 0.00021262886597938144, + "loss": 0.4901, + "step": 28227 + }, + { + "epoch": 1.5771433360337457, + "grad_norm": 0.573167622089386, + "learning_rate": 0.00021260085163603764, + "loss": 0.431, + "step": 28228 + }, + { + "epoch": 1.5771992066374279, + "grad_norm": 0.3206891119480133, + "learning_rate": 0.00021257283729269385, + "loss": 0.3594, + "step": 28229 + }, + { + "epoch": 1.5772550772411096, + "grad_norm": 0.4429229199886322, + "learning_rate": 0.00021254482294935008, + "loss": 0.3186, + "step": 28230 + }, + { + "epoch": 1.5773109478447913, + "grad_norm": 0.4750441908836365, + "learning_rate": 0.00021251680860600629, + "loss": 0.3361, + "step": 28231 + }, + { + "epoch": 1.5773668184484735, + "grad_norm": 0.4629569947719574, + "learning_rate": 0.0002124887942626625, + "loss": 0.4845, + "step": 28232 + }, + { + "epoch": 1.5774226890521552, + "grad_norm": 0.5034759044647217, + "learning_rate": 0.0002124607799193187, + "loss": 0.4588, + "step": 28233 + }, + { + "epoch": 1.577478559655837, + "grad_norm": 8.802860260009766, + "learning_rate": 0.0002124327655759749, + "loss": 0.3629, + "step": 28234 + }, + { + "epoch": 1.5775344302595191, + "grad_norm": 0.4473876357078552, + "learning_rate": 0.0002124047512326311, + "loss": 0.4914, + "step": 28235 + }, + { + "epoch": 1.5775903008632008, + "grad_norm": 0.6157010197639465, + "learning_rate": 0.00021237673688928732, + "loss": 0.4191, + "step": 28236 + }, + { + "epoch": 1.5776461714668826, + "grad_norm": 0.3118780255317688, + "learning_rate": 0.00021234872254594352, + "loss": 0.3577, + "step": 28237 + }, + { + "epoch": 1.5777020420705645, + "grad_norm": 0.3480415642261505, + "learning_rate": 0.00021232070820259975, + "loss": 0.4971, + "step": 28238 + }, + { + "epoch": 1.5777579126742465, + "grad_norm": 0.4943540096282959, + "learning_rate": 0.00021229269385925596, + "loss": 0.3841, + "step": 28239 + }, + { + "epoch": 1.5778137832779282, + "grad_norm": 0.589679479598999, + "learning_rate": 0.00021226467951591217, + "loss": 0.4322, + "step": 28240 + }, + { + "epoch": 1.5778696538816102, + "grad_norm": 0.7224860787391663, + "learning_rate": 0.00021223666517256837, + "loss": 0.4288, + "step": 28241 + }, + { + "epoch": 1.577925524485292, + "grad_norm": 0.3906976878643036, + "learning_rate": 0.00021220865082922458, + "loss": 0.4956, + "step": 28242 + }, + { + "epoch": 1.5779813950889738, + "grad_norm": 1.5035486221313477, + "learning_rate": 0.00021218063648588076, + "loss": 0.3742, + "step": 28243 + }, + { + "epoch": 1.5780372656926558, + "grad_norm": 0.5737444162368774, + "learning_rate": 0.00021215262214253696, + "loss": 0.5194, + "step": 28244 + }, + { + "epoch": 1.5780931362963377, + "grad_norm": 1.217347502708435, + "learning_rate": 0.00021212460779919317, + "loss": 0.4702, + "step": 28245 + }, + { + "epoch": 1.5781490069000195, + "grad_norm": 0.4599311351776123, + "learning_rate": 0.0002120965934558494, + "loss": 0.5428, + "step": 28246 + }, + { + "epoch": 1.5782048775037014, + "grad_norm": 0.7998025417327881, + "learning_rate": 0.0002120685791125056, + "loss": 0.3707, + "step": 28247 + }, + { + "epoch": 1.5782607481073834, + "grad_norm": 0.6811506748199463, + "learning_rate": 0.0002120405647691618, + "loss": 0.3911, + "step": 28248 + }, + { + "epoch": 1.578316618711065, + "grad_norm": 0.4647868275642395, + "learning_rate": 0.00021201255042581802, + "loss": 0.4402, + "step": 28249 + }, + { + "epoch": 1.578372489314747, + "grad_norm": 0.3372544050216675, + "learning_rate": 0.00021198453608247422, + "loss": 0.3413, + "step": 28250 + }, + { + "epoch": 1.578428359918429, + "grad_norm": 0.6712241172790527, + "learning_rate": 0.00021195652173913043, + "loss": 0.4403, + "step": 28251 + }, + { + "epoch": 1.5784842305221107, + "grad_norm": 0.40945756435394287, + "learning_rate": 0.00021192850739578664, + "loss": 0.4497, + "step": 28252 + }, + { + "epoch": 1.5785401011257927, + "grad_norm": 0.7644229531288147, + "learning_rate": 0.00021190049305244287, + "loss": 0.47, + "step": 28253 + }, + { + "epoch": 1.5785959717294746, + "grad_norm": 4.987579345703125, + "learning_rate": 0.00021187247870909907, + "loss": 0.4847, + "step": 28254 + }, + { + "epoch": 1.5786518423331564, + "grad_norm": 0.3165104389190674, + "learning_rate": 0.00021184446436575528, + "loss": 0.2994, + "step": 28255 + }, + { + "epoch": 1.5787077129368383, + "grad_norm": 0.3368356227874756, + "learning_rate": 0.00021181645002241149, + "loss": 0.3922, + "step": 28256 + }, + { + "epoch": 1.5787635835405203, + "grad_norm": 0.38384687900543213, + "learning_rate": 0.0002117884356790677, + "loss": 0.3998, + "step": 28257 + }, + { + "epoch": 1.578819454144202, + "grad_norm": 2.040461778640747, + "learning_rate": 0.0002117604213357239, + "loss": 0.4469, + "step": 28258 + }, + { + "epoch": 1.578875324747884, + "grad_norm": 0.47790706157684326, + "learning_rate": 0.0002117324069923801, + "loss": 0.4501, + "step": 28259 + }, + { + "epoch": 1.578931195351566, + "grad_norm": 0.33605432510375977, + "learning_rate": 0.0002117043926490363, + "loss": 0.4251, + "step": 28260 + }, + { + "epoch": 1.5789870659552476, + "grad_norm": 0.9405480623245239, + "learning_rate": 0.00021167637830569254, + "loss": 0.3501, + "step": 28261 + }, + { + "epoch": 1.5790429365589294, + "grad_norm": 0.3582788407802582, + "learning_rate": 0.00021164836396234875, + "loss": 0.4327, + "step": 28262 + }, + { + "epoch": 1.5790988071626115, + "grad_norm": 0.3270112872123718, + "learning_rate": 0.00021162034961900495, + "loss": 0.3435, + "step": 28263 + }, + { + "epoch": 1.5791546777662933, + "grad_norm": 0.6296301484107971, + "learning_rate": 0.00021159233527566113, + "loss": 0.5465, + "step": 28264 + }, + { + "epoch": 1.579210548369975, + "grad_norm": 0.3603270351886749, + "learning_rate": 0.00021156432093231734, + "loss": 0.3463, + "step": 28265 + }, + { + "epoch": 1.5792664189736572, + "grad_norm": 0.3797479271888733, + "learning_rate": 0.00021153630658897354, + "loss": 0.3945, + "step": 28266 + }, + { + "epoch": 1.579322289577339, + "grad_norm": 1.2341129779815674, + "learning_rate": 0.00021150829224562975, + "loss": 0.4397, + "step": 28267 + }, + { + "epoch": 1.5793781601810206, + "grad_norm": 0.44047683477401733, + "learning_rate": 0.00021148027790228596, + "loss": 0.4623, + "step": 28268 + }, + { + "epoch": 1.5794340307847028, + "grad_norm": 0.4998818039894104, + "learning_rate": 0.0002114522635589422, + "loss": 0.4699, + "step": 28269 + }, + { + "epoch": 1.5794899013883845, + "grad_norm": 0.46715494990348816, + "learning_rate": 0.0002114242492155984, + "loss": 0.3468, + "step": 28270 + }, + { + "epoch": 1.5795457719920662, + "grad_norm": 0.49215951561927795, + "learning_rate": 0.0002113962348722546, + "loss": 0.41, + "step": 28271 + }, + { + "epoch": 1.5796016425957482, + "grad_norm": 0.4337247908115387, + "learning_rate": 0.0002113682205289108, + "loss": 0.4871, + "step": 28272 + }, + { + "epoch": 1.5796575131994302, + "grad_norm": 0.4627307057380676, + "learning_rate": 0.000211340206185567, + "loss": 0.4565, + "step": 28273 + }, + { + "epoch": 1.5797133838031119, + "grad_norm": 1.3498188257217407, + "learning_rate": 0.00021131219184222322, + "loss": 0.4357, + "step": 28274 + }, + { + "epoch": 1.5797692544067938, + "grad_norm": 0.5351046919822693, + "learning_rate": 0.00021128417749887942, + "loss": 0.5417, + "step": 28275 + }, + { + "epoch": 1.5798251250104758, + "grad_norm": 0.43356969952583313, + "learning_rate": 0.00021125616315553563, + "loss": 0.4916, + "step": 28276 + }, + { + "epoch": 1.5798809956141575, + "grad_norm": 0.9617195725440979, + "learning_rate": 0.00021122814881219186, + "loss": 0.3848, + "step": 28277 + }, + { + "epoch": 1.5799368662178395, + "grad_norm": 0.3628755211830139, + "learning_rate": 0.00021120013446884807, + "loss": 0.4103, + "step": 28278 + }, + { + "epoch": 1.5799927368215214, + "grad_norm": 0.6416108012199402, + "learning_rate": 0.00021117212012550427, + "loss": 0.5, + "step": 28279 + }, + { + "epoch": 1.5800486074252031, + "grad_norm": 0.3608487546443939, + "learning_rate": 0.00021114410578216048, + "loss": 0.3828, + "step": 28280 + }, + { + "epoch": 1.580104478028885, + "grad_norm": 0.7511181235313416, + "learning_rate": 0.00021111609143881669, + "loss": 0.4084, + "step": 28281 + }, + { + "epoch": 1.580160348632567, + "grad_norm": 0.3016762137413025, + "learning_rate": 0.0002110880770954729, + "loss": 0.3085, + "step": 28282 + }, + { + "epoch": 1.5802162192362488, + "grad_norm": 0.48880109190940857, + "learning_rate": 0.0002110600627521291, + "loss": 0.4734, + "step": 28283 + }, + { + "epoch": 1.5802720898399307, + "grad_norm": 0.3376937508583069, + "learning_rate": 0.0002110320484087853, + "loss": 0.2989, + "step": 28284 + }, + { + "epoch": 1.5803279604436127, + "grad_norm": 0.5561743974685669, + "learning_rate": 0.0002110040340654415, + "loss": 0.4869, + "step": 28285 + }, + { + "epoch": 1.5803838310472944, + "grad_norm": 0.5629010200500488, + "learning_rate": 0.00021097601972209771, + "loss": 0.4406, + "step": 28286 + }, + { + "epoch": 1.5804397016509764, + "grad_norm": 0.3877045512199402, + "learning_rate": 0.00021094800537875392, + "loss": 0.4347, + "step": 28287 + }, + { + "epoch": 1.5804955722546583, + "grad_norm": 0.7314720749855042, + "learning_rate": 0.00021091999103541013, + "loss": 0.4449, + "step": 28288 + }, + { + "epoch": 1.58055144285834, + "grad_norm": 0.25733065605163574, + "learning_rate": 0.00021089197669206633, + "loss": 0.2933, + "step": 28289 + }, + { + "epoch": 1.580607313462022, + "grad_norm": 0.4414336681365967, + "learning_rate": 0.00021086396234872254, + "loss": 0.4277, + "step": 28290 + }, + { + "epoch": 1.580663184065704, + "grad_norm": 0.6067591905593872, + "learning_rate": 0.00021083594800537874, + "loss": 0.3806, + "step": 28291 + }, + { + "epoch": 1.5807190546693857, + "grad_norm": 0.4594418704509735, + "learning_rate": 0.00021080793366203495, + "loss": 0.454, + "step": 28292 + }, + { + "epoch": 1.5807749252730676, + "grad_norm": 3.917675256729126, + "learning_rate": 0.00021077991931869118, + "loss": 0.4541, + "step": 28293 + }, + { + "epoch": 1.5808307958767496, + "grad_norm": 0.36778414249420166, + "learning_rate": 0.0002107519049753474, + "loss": 0.4091, + "step": 28294 + }, + { + "epoch": 1.5808866664804313, + "grad_norm": 0.5169408917427063, + "learning_rate": 0.0002107238906320036, + "loss": 0.4806, + "step": 28295 + }, + { + "epoch": 1.580942537084113, + "grad_norm": 0.38963884115219116, + "learning_rate": 0.0002106958762886598, + "loss": 0.3699, + "step": 28296 + }, + { + "epoch": 1.5809984076877952, + "grad_norm": 1.6569398641586304, + "learning_rate": 0.000210667861945316, + "loss": 0.3395, + "step": 28297 + }, + { + "epoch": 1.581054278291477, + "grad_norm": 2.019322395324707, + "learning_rate": 0.0002106398476019722, + "loss": 0.4083, + "step": 28298 + }, + { + "epoch": 1.5811101488951587, + "grad_norm": 0.5371368527412415, + "learning_rate": 0.00021061183325862842, + "loss": 0.4373, + "step": 28299 + }, + { + "epoch": 1.5811660194988408, + "grad_norm": 1.141769289970398, + "learning_rate": 0.00021058381891528462, + "loss": 0.4014, + "step": 28300 + }, + { + "epoch": 1.5812218901025226, + "grad_norm": 0.4354463815689087, + "learning_rate": 0.00021055580457194086, + "loss": 0.4547, + "step": 28301 + }, + { + "epoch": 1.5812777607062043, + "grad_norm": 0.5762983560562134, + "learning_rate": 0.00021052779022859706, + "loss": 0.42, + "step": 28302 + }, + { + "epoch": 1.5813336313098865, + "grad_norm": 0.44170042872428894, + "learning_rate": 0.00021049977588525327, + "loss": 0.4507, + "step": 28303 + }, + { + "epoch": 1.5813895019135682, + "grad_norm": 0.7054184079170227, + "learning_rate": 0.00021047176154190947, + "loss": 0.2509, + "step": 28304 + }, + { + "epoch": 1.58144537251725, + "grad_norm": 0.4349862039089203, + "learning_rate": 0.00021044374719856568, + "loss": 0.4038, + "step": 28305 + }, + { + "epoch": 1.5815012431209319, + "grad_norm": 0.3897482454776764, + "learning_rate": 0.00021041573285522186, + "loss": 0.4339, + "step": 28306 + }, + { + "epoch": 1.5815571137246138, + "grad_norm": 3.5311121940612793, + "learning_rate": 0.00021038771851187806, + "loss": 0.456, + "step": 28307 + }, + { + "epoch": 1.5816129843282956, + "grad_norm": 0.3076651990413666, + "learning_rate": 0.00021035970416853427, + "loss": 0.331, + "step": 28308 + }, + { + "epoch": 1.5816688549319775, + "grad_norm": 0.36710137128829956, + "learning_rate": 0.0002103316898251905, + "loss": 0.3838, + "step": 28309 + }, + { + "epoch": 1.5817247255356595, + "grad_norm": 0.43721500039100647, + "learning_rate": 0.0002103036754818467, + "loss": 0.4715, + "step": 28310 + }, + { + "epoch": 1.5817805961393412, + "grad_norm": 0.3688920736312866, + "learning_rate": 0.00021027566113850291, + "loss": 0.3813, + "step": 28311 + }, + { + "epoch": 1.5818364667430231, + "grad_norm": 0.6711531281471252, + "learning_rate": 0.00021024764679515912, + "loss": 0.3761, + "step": 28312 + }, + { + "epoch": 1.581892337346705, + "grad_norm": 1.3390600681304932, + "learning_rate": 0.00021021963245181533, + "loss": 0.3795, + "step": 28313 + }, + { + "epoch": 1.5819482079503868, + "grad_norm": 0.26975658535957336, + "learning_rate": 0.00021019161810847153, + "loss": 0.2754, + "step": 28314 + }, + { + "epoch": 1.5820040785540688, + "grad_norm": 0.44443807005882263, + "learning_rate": 0.00021016360376512774, + "loss": 0.4295, + "step": 28315 + }, + { + "epoch": 1.5820599491577507, + "grad_norm": 0.5303740501403809, + "learning_rate": 0.00021013558942178397, + "loss": 0.5517, + "step": 28316 + }, + { + "epoch": 1.5821158197614325, + "grad_norm": 0.4007909595966339, + "learning_rate": 0.00021010757507844018, + "loss": 0.4746, + "step": 28317 + }, + { + "epoch": 1.5821716903651144, + "grad_norm": 0.3671022355556488, + "learning_rate": 0.00021007956073509638, + "loss": 0.4327, + "step": 28318 + }, + { + "epoch": 1.5822275609687964, + "grad_norm": 0.48101529479026794, + "learning_rate": 0.0002100515463917526, + "loss": 0.4996, + "step": 28319 + }, + { + "epoch": 1.582283431572478, + "grad_norm": 0.654026448726654, + "learning_rate": 0.0002100235320484088, + "loss": 0.7408, + "step": 28320 + }, + { + "epoch": 1.58233930217616, + "grad_norm": 0.4029216468334198, + "learning_rate": 0.000209995517705065, + "loss": 0.3636, + "step": 28321 + }, + { + "epoch": 1.582395172779842, + "grad_norm": 0.6122556924819946, + "learning_rate": 0.0002099675033617212, + "loss": 0.5129, + "step": 28322 + }, + { + "epoch": 1.5824510433835237, + "grad_norm": 0.31902095675468445, + "learning_rate": 0.0002099394890183774, + "loss": 0.2555, + "step": 28323 + }, + { + "epoch": 1.5825069139872057, + "grad_norm": 0.3561430871486664, + "learning_rate": 0.00020991147467503364, + "loss": 0.3722, + "step": 28324 + }, + { + "epoch": 1.5825627845908876, + "grad_norm": 0.3685482442378998, + "learning_rate": 0.00020988346033168985, + "loss": 0.4024, + "step": 28325 + }, + { + "epoch": 1.5826186551945693, + "grad_norm": 0.37082356214523315, + "learning_rate": 0.00020985544598834606, + "loss": 0.4307, + "step": 28326 + }, + { + "epoch": 1.5826745257982513, + "grad_norm": 0.5637924075126648, + "learning_rate": 0.00020982743164500226, + "loss": 0.3861, + "step": 28327 + }, + { + "epoch": 1.5827303964019332, + "grad_norm": 0.3987240493297577, + "learning_rate": 0.00020979941730165844, + "loss": 0.3585, + "step": 28328 + }, + { + "epoch": 1.582786267005615, + "grad_norm": 0.4540972113609314, + "learning_rate": 0.00020977140295831465, + "loss": 0.4642, + "step": 28329 + }, + { + "epoch": 1.5828421376092967, + "grad_norm": 0.47255703806877136, + "learning_rate": 0.00020974338861497085, + "loss": 0.4738, + "step": 28330 + }, + { + "epoch": 1.5828980082129789, + "grad_norm": 0.7741196751594543, + "learning_rate": 0.00020971537427162706, + "loss": 0.4152, + "step": 28331 + }, + { + "epoch": 1.5829538788166606, + "grad_norm": 0.5521988868713379, + "learning_rate": 0.0002096873599282833, + "loss": 0.395, + "step": 28332 + }, + { + "epoch": 1.5830097494203423, + "grad_norm": 7.124446392059326, + "learning_rate": 0.0002096593455849395, + "loss": 0.3814, + "step": 28333 + }, + { + "epoch": 1.5830656200240245, + "grad_norm": 0.8715516924858093, + "learning_rate": 0.0002096313312415957, + "loss": 0.4351, + "step": 28334 + }, + { + "epoch": 1.5831214906277062, + "grad_norm": 0.4199657142162323, + "learning_rate": 0.0002096033168982519, + "loss": 0.4782, + "step": 28335 + }, + { + "epoch": 1.583177361231388, + "grad_norm": 0.754910409450531, + "learning_rate": 0.00020957530255490811, + "loss": 0.3999, + "step": 28336 + }, + { + "epoch": 1.5832332318350701, + "grad_norm": 0.5513484477996826, + "learning_rate": 0.00020954728821156432, + "loss": 0.4413, + "step": 28337 + }, + { + "epoch": 1.5832891024387519, + "grad_norm": 0.3651186227798462, + "learning_rate": 0.00020951927386822053, + "loss": 0.4456, + "step": 28338 + }, + { + "epoch": 1.5833449730424336, + "grad_norm": 0.3884512186050415, + "learning_rate": 0.00020949125952487673, + "loss": 0.4465, + "step": 28339 + }, + { + "epoch": 1.5834008436461156, + "grad_norm": 0.39806893467903137, + "learning_rate": 0.00020946324518153296, + "loss": 0.4746, + "step": 28340 + }, + { + "epoch": 1.5834567142497975, + "grad_norm": 0.544995129108429, + "learning_rate": 0.00020943523083818917, + "loss": 0.4302, + "step": 28341 + }, + { + "epoch": 1.5835125848534792, + "grad_norm": 0.4282309114933014, + "learning_rate": 0.00020940721649484538, + "loss": 0.4377, + "step": 28342 + }, + { + "epoch": 1.5835684554571612, + "grad_norm": 1.3677754402160645, + "learning_rate": 0.00020937920215150158, + "loss": 0.5516, + "step": 28343 + }, + { + "epoch": 1.5836243260608431, + "grad_norm": 0.40314027667045593, + "learning_rate": 0.0002093511878081578, + "loss": 0.3912, + "step": 28344 + }, + { + "epoch": 1.5836801966645249, + "grad_norm": 0.6237137317657471, + "learning_rate": 0.000209323173464814, + "loss": 0.5036, + "step": 28345 + }, + { + "epoch": 1.5837360672682068, + "grad_norm": 0.33342254161834717, + "learning_rate": 0.0002092951591214702, + "loss": 0.3925, + "step": 28346 + }, + { + "epoch": 1.5837919378718888, + "grad_norm": 0.41057002544403076, + "learning_rate": 0.0002092671447781264, + "loss": 0.3638, + "step": 28347 + }, + { + "epoch": 1.5838478084755705, + "grad_norm": 0.5830161571502686, + "learning_rate": 0.00020923913043478264, + "loss": 0.3778, + "step": 28348 + }, + { + "epoch": 1.5839036790792524, + "grad_norm": 0.485390841960907, + "learning_rate": 0.00020921111609143882, + "loss": 0.4194, + "step": 28349 + }, + { + "epoch": 1.5839595496829344, + "grad_norm": 2.8775928020477295, + "learning_rate": 0.00020918310174809502, + "loss": 0.5559, + "step": 28350 + }, + { + "epoch": 1.5840154202866161, + "grad_norm": 0.3556153178215027, + "learning_rate": 0.00020915508740475123, + "loss": 0.3808, + "step": 28351 + }, + { + "epoch": 1.584071290890298, + "grad_norm": 1.977429747581482, + "learning_rate": 0.00020912707306140743, + "loss": 0.4873, + "step": 28352 + }, + { + "epoch": 1.58412716149398, + "grad_norm": 0.31522825360298157, + "learning_rate": 0.00020909905871806364, + "loss": 0.3748, + "step": 28353 + }, + { + "epoch": 1.5841830320976618, + "grad_norm": 0.4902561604976654, + "learning_rate": 0.00020907104437471985, + "loss": 0.3983, + "step": 28354 + }, + { + "epoch": 1.5842389027013437, + "grad_norm": 0.5374685525894165, + "learning_rate": 0.00020904303003137605, + "loss": 0.524, + "step": 28355 + }, + { + "epoch": 1.5842947733050257, + "grad_norm": 0.5085402131080627, + "learning_rate": 0.00020901501568803228, + "loss": 0.3823, + "step": 28356 + }, + { + "epoch": 1.5843506439087074, + "grad_norm": 0.43731579184532166, + "learning_rate": 0.0002089870013446885, + "loss": 0.5533, + "step": 28357 + }, + { + "epoch": 1.5844065145123893, + "grad_norm": 0.5213608741760254, + "learning_rate": 0.0002089589870013447, + "loss": 0.4896, + "step": 28358 + }, + { + "epoch": 1.5844623851160713, + "grad_norm": 0.839979887008667, + "learning_rate": 0.0002089309726580009, + "loss": 0.3234, + "step": 28359 + }, + { + "epoch": 1.584518255719753, + "grad_norm": 0.8957884907722473, + "learning_rate": 0.0002089029583146571, + "loss": 0.5803, + "step": 28360 + }, + { + "epoch": 1.584574126323435, + "grad_norm": 0.5538317561149597, + "learning_rate": 0.0002088749439713133, + "loss": 0.5967, + "step": 28361 + }, + { + "epoch": 1.584629996927117, + "grad_norm": 0.456176221370697, + "learning_rate": 0.00020884692962796952, + "loss": 0.4128, + "step": 28362 + }, + { + "epoch": 1.5846858675307987, + "grad_norm": 1.0844887495040894, + "learning_rate": 0.00020881891528462573, + "loss": 0.467, + "step": 28363 + }, + { + "epoch": 1.5847417381344804, + "grad_norm": 0.5383906960487366, + "learning_rate": 0.00020879090094128196, + "loss": 0.458, + "step": 28364 + }, + { + "epoch": 1.5847976087381626, + "grad_norm": 1.0213873386383057, + "learning_rate": 0.00020876288659793816, + "loss": 0.5359, + "step": 28365 + }, + { + "epoch": 1.5848534793418443, + "grad_norm": 0.4663717746734619, + "learning_rate": 0.00020873487225459437, + "loss": 0.4262, + "step": 28366 + }, + { + "epoch": 1.584909349945526, + "grad_norm": 0.6508409976959229, + "learning_rate": 0.00020870685791125058, + "loss": 0.4562, + "step": 28367 + }, + { + "epoch": 1.5849652205492082, + "grad_norm": 6.194186210632324, + "learning_rate": 0.00020867884356790678, + "loss": 0.5292, + "step": 28368 + }, + { + "epoch": 1.58502109115289, + "grad_norm": 0.38650378584861755, + "learning_rate": 0.000208650829224563, + "loss": 0.408, + "step": 28369 + }, + { + "epoch": 1.5850769617565716, + "grad_norm": 0.3779977262020111, + "learning_rate": 0.00020862281488121917, + "loss": 0.3586, + "step": 28370 + }, + { + "epoch": 1.5851328323602538, + "grad_norm": 0.46092382073402405, + "learning_rate": 0.00020859480053787537, + "loss": 0.4284, + "step": 28371 + }, + { + "epoch": 1.5851887029639355, + "grad_norm": 0.30271345376968384, + "learning_rate": 0.0002085667861945316, + "loss": 0.3227, + "step": 28372 + }, + { + "epoch": 1.5852445735676173, + "grad_norm": 0.5312746167182922, + "learning_rate": 0.0002085387718511878, + "loss": 0.419, + "step": 28373 + }, + { + "epoch": 1.5853004441712992, + "grad_norm": 0.8887419104576111, + "learning_rate": 0.00020851075750784402, + "loss": 0.3695, + "step": 28374 + }, + { + "epoch": 1.5853563147749812, + "grad_norm": 0.5847569704055786, + "learning_rate": 0.00020848274316450022, + "loss": 0.3441, + "step": 28375 + }, + { + "epoch": 1.585412185378663, + "grad_norm": 0.4023343324661255, + "learning_rate": 0.00020845472882115643, + "loss": 0.446, + "step": 28376 + }, + { + "epoch": 1.5854680559823449, + "grad_norm": 0.4482870399951935, + "learning_rate": 0.00020842671447781263, + "loss": 0.488, + "step": 28377 + }, + { + "epoch": 1.5855239265860268, + "grad_norm": 0.43114349246025085, + "learning_rate": 0.00020839870013446884, + "loss": 0.4687, + "step": 28378 + }, + { + "epoch": 1.5855797971897085, + "grad_norm": 0.45377570390701294, + "learning_rate": 0.00020837068579112505, + "loss": 0.524, + "step": 28379 + }, + { + "epoch": 1.5856356677933905, + "grad_norm": 0.4377611577510834, + "learning_rate": 0.00020834267144778128, + "loss": 0.4466, + "step": 28380 + }, + { + "epoch": 1.5856915383970724, + "grad_norm": 2.3102126121520996, + "learning_rate": 0.00020831465710443748, + "loss": 0.4259, + "step": 28381 + }, + { + "epoch": 1.5857474090007542, + "grad_norm": 0.35745948553085327, + "learning_rate": 0.0002082866427610937, + "loss": 0.3963, + "step": 28382 + }, + { + "epoch": 1.5858032796044361, + "grad_norm": 1.1795916557312012, + "learning_rate": 0.0002082586284177499, + "loss": 0.4371, + "step": 28383 + }, + { + "epoch": 1.585859150208118, + "grad_norm": 0.48512256145477295, + "learning_rate": 0.0002082306140744061, + "loss": 0.3699, + "step": 28384 + }, + { + "epoch": 1.5859150208117998, + "grad_norm": 0.3294944167137146, + "learning_rate": 0.0002082025997310623, + "loss": 0.3513, + "step": 28385 + }, + { + "epoch": 1.5859708914154818, + "grad_norm": 0.36258646845817566, + "learning_rate": 0.0002081745853877185, + "loss": 0.4272, + "step": 28386 + }, + { + "epoch": 1.5860267620191637, + "grad_norm": 0.4019715189933777, + "learning_rate": 0.00020814657104437475, + "loss": 0.4502, + "step": 28387 + }, + { + "epoch": 1.5860826326228454, + "grad_norm": 0.3195478320121765, + "learning_rate": 0.00020811855670103095, + "loss": 0.2876, + "step": 28388 + }, + { + "epoch": 1.5861385032265274, + "grad_norm": 0.4762617349624634, + "learning_rate": 0.00020809054235768716, + "loss": 0.4518, + "step": 28389 + }, + { + "epoch": 1.5861943738302093, + "grad_norm": 1.905957579612732, + "learning_rate": 0.00020806252801434336, + "loss": 0.4614, + "step": 28390 + }, + { + "epoch": 1.586250244433891, + "grad_norm": 1.285169005393982, + "learning_rate": 0.00020803451367099954, + "loss": 0.4233, + "step": 28391 + }, + { + "epoch": 1.586306115037573, + "grad_norm": 0.801956295967102, + "learning_rate": 0.00020800649932765575, + "loss": 0.4955, + "step": 28392 + }, + { + "epoch": 1.586361985641255, + "grad_norm": 0.36731424927711487, + "learning_rate": 0.00020797848498431195, + "loss": 0.4487, + "step": 28393 + }, + { + "epoch": 1.5864178562449367, + "grad_norm": 0.4659106433391571, + "learning_rate": 0.00020795047064096816, + "loss": 0.4759, + "step": 28394 + }, + { + "epoch": 1.5864737268486186, + "grad_norm": 0.6964191198348999, + "learning_rate": 0.0002079224562976244, + "loss": 0.3795, + "step": 28395 + }, + { + "epoch": 1.5865295974523006, + "grad_norm": 0.3927532136440277, + "learning_rate": 0.0002078944419542806, + "loss": 0.4374, + "step": 28396 + }, + { + "epoch": 1.5865854680559823, + "grad_norm": 0.6287574768066406, + "learning_rate": 0.0002078664276109368, + "loss": 0.3874, + "step": 28397 + }, + { + "epoch": 1.586641338659664, + "grad_norm": 0.35338643193244934, + "learning_rate": 0.000207838413267593, + "loss": 0.3517, + "step": 28398 + }, + { + "epoch": 1.5866972092633462, + "grad_norm": 0.26722195744514465, + "learning_rate": 0.00020781039892424922, + "loss": 0.2436, + "step": 28399 + }, + { + "epoch": 1.586753079867028, + "grad_norm": 0.7052076458930969, + "learning_rate": 0.00020778238458090542, + "loss": 0.6993, + "step": 28400 + }, + { + "epoch": 1.5868089504707097, + "grad_norm": 0.5226325988769531, + "learning_rate": 0.00020775437023756163, + "loss": 0.4359, + "step": 28401 + }, + { + "epoch": 1.5868648210743919, + "grad_norm": 0.3764232099056244, + "learning_rate": 0.00020772635589421783, + "loss": 0.3407, + "step": 28402 + }, + { + "epoch": 1.5869206916780736, + "grad_norm": 0.3927558660507202, + "learning_rate": 0.00020769834155087407, + "loss": 0.4206, + "step": 28403 + }, + { + "epoch": 1.5869765622817553, + "grad_norm": 0.530319333076477, + "learning_rate": 0.00020767032720753027, + "loss": 0.5366, + "step": 28404 + }, + { + "epoch": 1.5870324328854373, + "grad_norm": 0.37025198340415955, + "learning_rate": 0.00020764231286418648, + "loss": 0.4043, + "step": 28405 + }, + { + "epoch": 1.5870883034891192, + "grad_norm": 0.36379337310791016, + "learning_rate": 0.00020761429852084268, + "loss": 0.419, + "step": 28406 + }, + { + "epoch": 1.587144174092801, + "grad_norm": 0.4046192169189453, + "learning_rate": 0.0002075862841774989, + "loss": 0.4085, + "step": 28407 + }, + { + "epoch": 1.587200044696483, + "grad_norm": 0.37724193930625916, + "learning_rate": 0.0002075582698341551, + "loss": 0.3841, + "step": 28408 + }, + { + "epoch": 1.5872559153001649, + "grad_norm": 0.4258206784725189, + "learning_rate": 0.0002075302554908113, + "loss": 0.3645, + "step": 28409 + }, + { + "epoch": 1.5873117859038466, + "grad_norm": 0.36650481820106506, + "learning_rate": 0.0002075022411474675, + "loss": 0.3432, + "step": 28410 + }, + { + "epoch": 1.5873676565075285, + "grad_norm": 0.9074486494064331, + "learning_rate": 0.00020747422680412374, + "loss": 0.4453, + "step": 28411 + }, + { + "epoch": 1.5874235271112105, + "grad_norm": 5.7728376388549805, + "learning_rate": 0.00020744621246077995, + "loss": 0.4277, + "step": 28412 + }, + { + "epoch": 1.5874793977148922, + "grad_norm": 0.33252742886543274, + "learning_rate": 0.00020741819811743612, + "loss": 0.3397, + "step": 28413 + }, + { + "epoch": 1.5875352683185742, + "grad_norm": 0.391287237405777, + "learning_rate": 0.00020739018377409233, + "loss": 0.4819, + "step": 28414 + }, + { + "epoch": 1.5875911389222561, + "grad_norm": 0.3742211163043976, + "learning_rate": 0.00020736216943074854, + "loss": 0.3878, + "step": 28415 + }, + { + "epoch": 1.5876470095259378, + "grad_norm": 4.20877742767334, + "learning_rate": 0.00020733415508740474, + "loss": 0.3871, + "step": 28416 + }, + { + "epoch": 1.5877028801296198, + "grad_norm": 0.3995693624019623, + "learning_rate": 0.00020730614074406095, + "loss": 0.4623, + "step": 28417 + }, + { + "epoch": 1.5877587507333017, + "grad_norm": 0.4061707556247711, + "learning_rate": 0.00020727812640071715, + "loss": 0.3448, + "step": 28418 + }, + { + "epoch": 1.5878146213369835, + "grad_norm": 0.3501474857330322, + "learning_rate": 0.00020725011205737339, + "loss": 0.3703, + "step": 28419 + }, + { + "epoch": 1.5878704919406654, + "grad_norm": 0.6063116192817688, + "learning_rate": 0.0002072220977140296, + "loss": 0.4651, + "step": 28420 + }, + { + "epoch": 1.5879263625443474, + "grad_norm": 0.6480092406272888, + "learning_rate": 0.0002071940833706858, + "loss": 0.4718, + "step": 28421 + }, + { + "epoch": 1.587982233148029, + "grad_norm": 0.5341344475746155, + "learning_rate": 0.000207166069027342, + "loss": 0.5232, + "step": 28422 + }, + { + "epoch": 1.588038103751711, + "grad_norm": 0.4978792369365692, + "learning_rate": 0.0002071380546839982, + "loss": 0.544, + "step": 28423 + }, + { + "epoch": 1.588093974355393, + "grad_norm": 0.8562363386154175, + "learning_rate": 0.00020711004034065442, + "loss": 0.3634, + "step": 28424 + }, + { + "epoch": 1.5881498449590747, + "grad_norm": 0.44404494762420654, + "learning_rate": 0.00020708202599731062, + "loss": 0.4542, + "step": 28425 + }, + { + "epoch": 1.5882057155627567, + "grad_norm": 0.36559462547302246, + "learning_rate": 0.00020705401165396683, + "loss": 0.4441, + "step": 28426 + }, + { + "epoch": 1.5882615861664386, + "grad_norm": 0.37184667587280273, + "learning_rate": 0.00020702599731062306, + "loss": 0.3991, + "step": 28427 + }, + { + "epoch": 1.5883174567701204, + "grad_norm": 0.38162103295326233, + "learning_rate": 0.00020699798296727927, + "loss": 0.4135, + "step": 28428 + }, + { + "epoch": 1.5883733273738023, + "grad_norm": 0.4648858308792114, + "learning_rate": 0.00020696996862393547, + "loss": 0.4081, + "step": 28429 + }, + { + "epoch": 1.5884291979774843, + "grad_norm": 0.9606919288635254, + "learning_rate": 0.00020694195428059168, + "loss": 0.4819, + "step": 28430 + }, + { + "epoch": 1.588485068581166, + "grad_norm": 0.40452122688293457, + "learning_rate": 0.00020691393993724788, + "loss": 0.4195, + "step": 28431 + }, + { + "epoch": 1.5885409391848477, + "grad_norm": 0.3673346936702728, + "learning_rate": 0.0002068859255939041, + "loss": 0.4826, + "step": 28432 + }, + { + "epoch": 1.58859680978853, + "grad_norm": 1.7827627658843994, + "learning_rate": 0.0002068579112505603, + "loss": 0.4229, + "step": 28433 + }, + { + "epoch": 1.5886526803922116, + "grad_norm": 0.4609341025352478, + "learning_rate": 0.00020682989690721647, + "loss": 0.4591, + "step": 28434 + }, + { + "epoch": 1.5887085509958934, + "grad_norm": 3.1766765117645264, + "learning_rate": 0.0002068018825638727, + "loss": 0.4376, + "step": 28435 + }, + { + "epoch": 1.5887644215995755, + "grad_norm": 0.3801860213279724, + "learning_rate": 0.0002067738682205289, + "loss": 0.3734, + "step": 28436 + }, + { + "epoch": 1.5888202922032573, + "grad_norm": 0.43240949511528015, + "learning_rate": 0.00020674585387718512, + "loss": 0.305, + "step": 28437 + }, + { + "epoch": 1.588876162806939, + "grad_norm": 0.4245259761810303, + "learning_rate": 0.00020671783953384132, + "loss": 0.4388, + "step": 28438 + }, + { + "epoch": 1.588932033410621, + "grad_norm": 0.36631160974502563, + "learning_rate": 0.00020668982519049753, + "loss": 0.3775, + "step": 28439 + }, + { + "epoch": 1.588987904014303, + "grad_norm": 0.4306127727031708, + "learning_rate": 0.00020666181084715374, + "loss": 0.3984, + "step": 28440 + }, + { + "epoch": 1.5890437746179846, + "grad_norm": 0.6024297475814819, + "learning_rate": 0.00020663379650380994, + "loss": 0.4676, + "step": 28441 + }, + { + "epoch": 1.5890996452216666, + "grad_norm": 0.3633204698562622, + "learning_rate": 0.00020660578216046615, + "loss": 0.3882, + "step": 28442 + }, + { + "epoch": 1.5891555158253485, + "grad_norm": 0.38197168707847595, + "learning_rate": 0.00020657776781712238, + "loss": 0.4077, + "step": 28443 + }, + { + "epoch": 1.5892113864290303, + "grad_norm": 3.2415614128112793, + "learning_rate": 0.00020654975347377859, + "loss": 0.4399, + "step": 28444 + }, + { + "epoch": 1.5892672570327122, + "grad_norm": 0.3654162585735321, + "learning_rate": 0.0002065217391304348, + "loss": 0.4391, + "step": 28445 + }, + { + "epoch": 1.5893231276363942, + "grad_norm": 0.6748754978179932, + "learning_rate": 0.000206493724787091, + "loss": 0.4933, + "step": 28446 + }, + { + "epoch": 1.589378998240076, + "grad_norm": 0.38742518424987793, + "learning_rate": 0.0002064657104437472, + "loss": 0.4651, + "step": 28447 + }, + { + "epoch": 1.5894348688437578, + "grad_norm": 0.9854929447174072, + "learning_rate": 0.0002064376961004034, + "loss": 0.4087, + "step": 28448 + }, + { + "epoch": 1.5894907394474398, + "grad_norm": 0.613749086856842, + "learning_rate": 0.00020640968175705961, + "loss": 0.3725, + "step": 28449 + }, + { + "epoch": 1.5895466100511215, + "grad_norm": 0.4417479932308197, + "learning_rate": 0.00020638166741371585, + "loss": 0.6664, + "step": 28450 + }, + { + "epoch": 1.5896024806548035, + "grad_norm": 0.38430601358413696, + "learning_rate": 0.00020635365307037205, + "loss": 0.422, + "step": 28451 + }, + { + "epoch": 1.5896583512584854, + "grad_norm": 0.3550732433795929, + "learning_rate": 0.00020632563872702826, + "loss": 0.3909, + "step": 28452 + }, + { + "epoch": 1.5897142218621672, + "grad_norm": 0.3367191255092621, + "learning_rate": 0.00020629762438368447, + "loss": 0.3751, + "step": 28453 + }, + { + "epoch": 1.589770092465849, + "grad_norm": 0.4032030701637268, + "learning_rate": 0.00020626961004034067, + "loss": 0.4328, + "step": 28454 + }, + { + "epoch": 1.589825963069531, + "grad_norm": 0.3033214211463928, + "learning_rate": 0.00020624159569699685, + "loss": 0.3846, + "step": 28455 + }, + { + "epoch": 1.5898818336732128, + "grad_norm": 4.080590724945068, + "learning_rate": 0.00020621358135365306, + "loss": 0.4895, + "step": 28456 + }, + { + "epoch": 1.5899377042768947, + "grad_norm": 0.6514151692390442, + "learning_rate": 0.00020618556701030926, + "loss": 0.451, + "step": 28457 + }, + { + "epoch": 1.5899935748805767, + "grad_norm": 0.5351266860961914, + "learning_rate": 0.0002061575526669655, + "loss": 0.5313, + "step": 28458 + }, + { + "epoch": 1.5900494454842584, + "grad_norm": 0.35684484243392944, + "learning_rate": 0.0002061295383236217, + "loss": 0.4094, + "step": 28459 + }, + { + "epoch": 1.5901053160879404, + "grad_norm": 0.49076759815216064, + "learning_rate": 0.0002061015239802779, + "loss": 0.4703, + "step": 28460 + }, + { + "epoch": 1.5901611866916223, + "grad_norm": 1.0111610889434814, + "learning_rate": 0.0002060735096369341, + "loss": 0.499, + "step": 28461 + }, + { + "epoch": 1.590217057295304, + "grad_norm": 0.7928857207298279, + "learning_rate": 0.00020604549529359032, + "loss": 0.408, + "step": 28462 + }, + { + "epoch": 1.590272927898986, + "grad_norm": 0.5388689637184143, + "learning_rate": 0.00020601748095024652, + "loss": 0.3083, + "step": 28463 + }, + { + "epoch": 1.590328798502668, + "grad_norm": 0.4402880370616913, + "learning_rate": 0.00020598946660690273, + "loss": 0.4003, + "step": 28464 + }, + { + "epoch": 1.5903846691063497, + "grad_norm": 0.6005921363830566, + "learning_rate": 0.00020596145226355894, + "loss": 0.4778, + "step": 28465 + }, + { + "epoch": 1.5904405397100314, + "grad_norm": 0.47730425000190735, + "learning_rate": 0.00020593343792021517, + "loss": 0.3782, + "step": 28466 + }, + { + "epoch": 1.5904964103137136, + "grad_norm": 0.2932303845882416, + "learning_rate": 0.00020590542357687137, + "loss": 0.3429, + "step": 28467 + }, + { + "epoch": 1.5905522809173953, + "grad_norm": 1.5203990936279297, + "learning_rate": 0.00020587740923352758, + "loss": 0.5021, + "step": 28468 + }, + { + "epoch": 1.590608151521077, + "grad_norm": 0.4490324556827545, + "learning_rate": 0.00020584939489018379, + "loss": 0.5385, + "step": 28469 + }, + { + "epoch": 1.5906640221247592, + "grad_norm": 0.4864867925643921, + "learning_rate": 0.00020582138054684, + "loss": 0.6972, + "step": 28470 + }, + { + "epoch": 1.590719892728441, + "grad_norm": 0.40983685851097107, + "learning_rate": 0.0002057933662034962, + "loss": 0.412, + "step": 28471 + }, + { + "epoch": 1.5907757633321227, + "grad_norm": 0.4459088146686554, + "learning_rate": 0.0002057653518601524, + "loss": 0.3669, + "step": 28472 + }, + { + "epoch": 1.5908316339358046, + "grad_norm": 0.6453219652175903, + "learning_rate": 0.0002057373375168086, + "loss": 0.5701, + "step": 28473 + }, + { + "epoch": 1.5908875045394866, + "grad_norm": 0.4088418185710907, + "learning_rate": 0.00020570932317346484, + "loss": 0.4487, + "step": 28474 + }, + { + "epoch": 1.5909433751431683, + "grad_norm": 0.292519748210907, + "learning_rate": 0.00020568130883012105, + "loss": 0.3858, + "step": 28475 + }, + { + "epoch": 1.5909992457468503, + "grad_norm": 0.5257911086082458, + "learning_rate": 0.00020565329448677723, + "loss": 0.4388, + "step": 28476 + }, + { + "epoch": 1.5910551163505322, + "grad_norm": 3.6690287590026855, + "learning_rate": 0.00020562528014343343, + "loss": 0.5953, + "step": 28477 + }, + { + "epoch": 1.591110986954214, + "grad_norm": 0.5145134925842285, + "learning_rate": 0.00020559726580008964, + "loss": 0.4429, + "step": 28478 + }, + { + "epoch": 1.5911668575578959, + "grad_norm": 0.6980517506599426, + "learning_rate": 0.00020556925145674584, + "loss": 0.3922, + "step": 28479 + }, + { + "epoch": 1.5912227281615778, + "grad_norm": 2.2022335529327393, + "learning_rate": 0.00020554123711340205, + "loss": 0.4557, + "step": 28480 + }, + { + "epoch": 1.5912785987652596, + "grad_norm": 0.34151530265808105, + "learning_rate": 0.00020551322277005826, + "loss": 0.3853, + "step": 28481 + }, + { + "epoch": 1.5913344693689415, + "grad_norm": 0.7226732969284058, + "learning_rate": 0.0002054852084267145, + "loss": 0.3261, + "step": 28482 + }, + { + "epoch": 1.5913903399726235, + "grad_norm": 0.5028047561645508, + "learning_rate": 0.0002054571940833707, + "loss": 0.4142, + "step": 28483 + }, + { + "epoch": 1.5914462105763052, + "grad_norm": 5.9270405769348145, + "learning_rate": 0.0002054291797400269, + "loss": 0.8422, + "step": 28484 + }, + { + "epoch": 1.5915020811799871, + "grad_norm": 0.3801758587360382, + "learning_rate": 0.0002054011653966831, + "loss": 0.4702, + "step": 28485 + }, + { + "epoch": 1.591557951783669, + "grad_norm": 0.5499476194381714, + "learning_rate": 0.0002053731510533393, + "loss": 0.5635, + "step": 28486 + }, + { + "epoch": 1.5916138223873508, + "grad_norm": 1.244990587234497, + "learning_rate": 0.00020534513670999552, + "loss": 0.418, + "step": 28487 + }, + { + "epoch": 1.5916696929910328, + "grad_norm": 2.070640802383423, + "learning_rate": 0.00020531712236665172, + "loss": 0.4151, + "step": 28488 + }, + { + "epoch": 1.5917255635947147, + "grad_norm": 0.4576431214809418, + "learning_rate": 0.00020528910802330793, + "loss": 0.4307, + "step": 28489 + }, + { + "epoch": 1.5917814341983965, + "grad_norm": 0.5240599513053894, + "learning_rate": 0.00020526109367996416, + "loss": 0.4725, + "step": 28490 + }, + { + "epoch": 1.5918373048020784, + "grad_norm": 0.4631662368774414, + "learning_rate": 0.00020523307933662037, + "loss": 0.5211, + "step": 28491 + }, + { + "epoch": 1.5918931754057604, + "grad_norm": 0.6408889293670654, + "learning_rate": 0.00020520506499327657, + "loss": 0.3361, + "step": 28492 + }, + { + "epoch": 1.591949046009442, + "grad_norm": 1.257099986076355, + "learning_rate": 0.00020517705064993278, + "loss": 0.445, + "step": 28493 + }, + { + "epoch": 1.592004916613124, + "grad_norm": 0.3599033057689667, + "learning_rate": 0.00020514903630658898, + "loss": 0.38, + "step": 28494 + }, + { + "epoch": 1.592060787216806, + "grad_norm": 0.4304240345954895, + "learning_rate": 0.0002051210219632452, + "loss": 0.4319, + "step": 28495 + }, + { + "epoch": 1.5921166578204877, + "grad_norm": 0.4610536992549896, + "learning_rate": 0.0002050930076199014, + "loss": 0.3853, + "step": 28496 + }, + { + "epoch": 1.5921725284241697, + "grad_norm": 0.4942721426486969, + "learning_rate": 0.0002050649932765576, + "loss": 0.5019, + "step": 28497 + }, + { + "epoch": 1.5922283990278516, + "grad_norm": 0.4363175332546234, + "learning_rate": 0.0002050369789332138, + "loss": 0.4388, + "step": 28498 + }, + { + "epoch": 1.5922842696315334, + "grad_norm": 0.3305700421333313, + "learning_rate": 0.00020500896458987001, + "loss": 0.4424, + "step": 28499 + }, + { + "epoch": 1.592340140235215, + "grad_norm": 0.3972415626049042, + "learning_rate": 0.00020498095024652622, + "loss": 0.4924, + "step": 28500 + }, + { + "epoch": 1.592340140235215, + "eval_cer": 0.08397433799220974, + "eval_loss": 0.3148682713508606, + "eval_runtime": 56.1691, + "eval_samples_per_second": 80.792, + "eval_steps_per_second": 5.056, + "eval_wer": 0.33270020390491883, + "step": 28500 + }, + { + "epoch": 1.5923960108388973, + "grad_norm": 0.7251068949699402, + "learning_rate": 0.00020495293590318243, + "loss": 0.4968, + "step": 28501 + }, + { + "epoch": 1.592451881442579, + "grad_norm": 0.3594285249710083, + "learning_rate": 0.00020492492155983863, + "loss": 0.3999, + "step": 28502 + }, + { + "epoch": 1.5925077520462607, + "grad_norm": 0.40079790353775024, + "learning_rate": 0.00020489690721649484, + "loss": 0.3007, + "step": 28503 + }, + { + "epoch": 1.592563622649943, + "grad_norm": 0.4723038375377655, + "learning_rate": 0.00020486889287315104, + "loss": 0.4213, + "step": 28504 + }, + { + "epoch": 1.5926194932536246, + "grad_norm": 0.6704965829849243, + "learning_rate": 0.00020484087852980725, + "loss": 0.3443, + "step": 28505 + }, + { + "epoch": 1.5926753638573063, + "grad_norm": 0.4323621094226837, + "learning_rate": 0.00020481286418646348, + "loss": 0.4635, + "step": 28506 + }, + { + "epoch": 1.5927312344609883, + "grad_norm": 0.4980507791042328, + "learning_rate": 0.0002047848498431197, + "loss": 0.4046, + "step": 28507 + }, + { + "epoch": 1.5927871050646702, + "grad_norm": 0.41455724835395813, + "learning_rate": 0.0002047568354997759, + "loss": 0.3521, + "step": 28508 + }, + { + "epoch": 1.592842975668352, + "grad_norm": Infinity, + "learning_rate": 0.0002047568354997759, + "loss": 0.4986, + "step": 28509 + }, + { + "epoch": 1.592898846272034, + "grad_norm": 1.0442029237747192, + "learning_rate": 0.0002047288211564321, + "loss": 0.3753, + "step": 28510 + }, + { + "epoch": 1.5929547168757159, + "grad_norm": 0.4409755766391754, + "learning_rate": 0.0002047008068130883, + "loss": 0.4905, + "step": 28511 + }, + { + "epoch": 1.5930105874793976, + "grad_norm": 0.4737102687358856, + "learning_rate": 0.0002046727924697445, + "loss": 0.4254, + "step": 28512 + }, + { + "epoch": 1.5930664580830796, + "grad_norm": 0.36411890387535095, + "learning_rate": 0.00020464477812640072, + "loss": 0.4161, + "step": 28513 + }, + { + "epoch": 1.5931223286867615, + "grad_norm": 1.2122443914413452, + "learning_rate": 0.00020461676378305695, + "loss": 0.4105, + "step": 28514 + }, + { + "epoch": 1.5931781992904432, + "grad_norm": 2.6435773372650146, + "learning_rate": 0.00020458874943971316, + "loss": 0.3762, + "step": 28515 + }, + { + "epoch": 1.5932340698941252, + "grad_norm": 0.4090855121612549, + "learning_rate": 0.00020456073509636936, + "loss": 0.3984, + "step": 28516 + }, + { + "epoch": 1.5932899404978071, + "grad_norm": 0.36608344316482544, + "learning_rate": 0.00020453272075302557, + "loss": 0.3538, + "step": 28517 + }, + { + "epoch": 1.5933458111014889, + "grad_norm": 0.541986882686615, + "learning_rate": 0.00020450470640968177, + "loss": 0.3703, + "step": 28518 + }, + { + "epoch": 1.5934016817051708, + "grad_norm": 0.32156994938850403, + "learning_rate": 0.00020447669206633798, + "loss": 0.3125, + "step": 28519 + }, + { + "epoch": 1.5934575523088528, + "grad_norm": 0.4865342974662781, + "learning_rate": 0.00020444867772299416, + "loss": 0.4679, + "step": 28520 + }, + { + "epoch": 1.5935134229125345, + "grad_norm": 0.9771979451179504, + "learning_rate": 0.00020442066337965036, + "loss": 0.3287, + "step": 28521 + }, + { + "epoch": 1.5935692935162165, + "grad_norm": 4.459903717041016, + "learning_rate": 0.0002043926490363066, + "loss": 0.4269, + "step": 28522 + }, + { + "epoch": 1.5936251641198984, + "grad_norm": 0.6428641080856323, + "learning_rate": 0.0002043646346929628, + "loss": 0.3911, + "step": 28523 + }, + { + "epoch": 1.5936810347235801, + "grad_norm": 0.33119580149650574, + "learning_rate": 0.000204336620349619, + "loss": 0.4254, + "step": 28524 + }, + { + "epoch": 1.593736905327262, + "grad_norm": 0.5954486727714539, + "learning_rate": 0.00020430860600627521, + "loss": 0.485, + "step": 28525 + }, + { + "epoch": 1.593792775930944, + "grad_norm": 0.4911845326423645, + "learning_rate": 0.00020428059166293142, + "loss": 0.3052, + "step": 28526 + }, + { + "epoch": 1.5938486465346258, + "grad_norm": 0.7332399487495422, + "learning_rate": 0.00020425257731958763, + "loss": 0.4054, + "step": 28527 + }, + { + "epoch": 1.5939045171383077, + "grad_norm": 0.7590106129646301, + "learning_rate": 0.00020422456297624383, + "loss": 0.4789, + "step": 28528 + }, + { + "epoch": 1.5939603877419897, + "grad_norm": 0.5762163400650024, + "learning_rate": 0.00020419654863290004, + "loss": 0.498, + "step": 28529 + }, + { + "epoch": 1.5940162583456714, + "grad_norm": 0.41376015543937683, + "learning_rate": 0.00020416853428955627, + "loss": 0.5043, + "step": 28530 + }, + { + "epoch": 1.5940721289493534, + "grad_norm": 0.3446808159351349, + "learning_rate": 0.00020414051994621248, + "loss": 0.3343, + "step": 28531 + }, + { + "epoch": 1.5941279995530353, + "grad_norm": 0.37837228178977966, + "learning_rate": 0.00020411250560286868, + "loss": 0.4674, + "step": 28532 + }, + { + "epoch": 1.594183870156717, + "grad_norm": 0.418708860874176, + "learning_rate": 0.0002040844912595249, + "loss": 0.4295, + "step": 28533 + }, + { + "epoch": 1.5942397407603988, + "grad_norm": 0.5202136635780334, + "learning_rate": 0.0002040564769161811, + "loss": 0.3996, + "step": 28534 + }, + { + "epoch": 1.594295611364081, + "grad_norm": 0.9491950273513794, + "learning_rate": 0.0002040284625728373, + "loss": 0.4463, + "step": 28535 + }, + { + "epoch": 1.5943514819677627, + "grad_norm": 2.067586898803711, + "learning_rate": 0.0002040004482294935, + "loss": 0.3705, + "step": 28536 + }, + { + "epoch": 1.5944073525714444, + "grad_norm": 0.8121040463447571, + "learning_rate": 0.0002039724338861497, + "loss": 0.5674, + "step": 28537 + }, + { + "epoch": 1.5944632231751266, + "grad_norm": 1.050696611404419, + "learning_rate": 0.00020394441954280594, + "loss": 0.3465, + "step": 28538 + }, + { + "epoch": 1.5945190937788083, + "grad_norm": 0.5120842456817627, + "learning_rate": 0.00020391640519946215, + "loss": 0.47, + "step": 28539 + }, + { + "epoch": 1.59457496438249, + "grad_norm": 0.36109229922294617, + "learning_rate": 0.00020388839085611836, + "loss": 0.3807, + "step": 28540 + }, + { + "epoch": 1.594630834986172, + "grad_norm": 0.36865320801734924, + "learning_rate": 0.00020386037651277453, + "loss": 0.3875, + "step": 28541 + }, + { + "epoch": 1.594686705589854, + "grad_norm": 0.32283419370651245, + "learning_rate": 0.00020383236216943074, + "loss": 0.3975, + "step": 28542 + }, + { + "epoch": 1.5947425761935357, + "grad_norm": 0.37265846133232117, + "learning_rate": 0.00020380434782608695, + "loss": 0.3736, + "step": 28543 + }, + { + "epoch": 1.5947984467972176, + "grad_norm": 2.99717378616333, + "learning_rate": 0.00020377633348274315, + "loss": 0.4454, + "step": 28544 + }, + { + "epoch": 1.5948543174008996, + "grad_norm": 0.31100666522979736, + "learning_rate": 0.00020374831913939936, + "loss": 0.4017, + "step": 28545 + }, + { + "epoch": 1.5949101880045813, + "grad_norm": 1.4696553945541382, + "learning_rate": 0.0002037203047960556, + "loss": 0.4063, + "step": 28546 + }, + { + "epoch": 1.5949660586082632, + "grad_norm": 0.5834122896194458, + "learning_rate": 0.0002036922904527118, + "loss": 0.3374, + "step": 28547 + }, + { + "epoch": 1.5950219292119452, + "grad_norm": 0.4487350285053253, + "learning_rate": 0.000203664276109368, + "loss": 0.3027, + "step": 28548 + }, + { + "epoch": 1.595077799815627, + "grad_norm": 0.3489462733268738, + "learning_rate": 0.0002036362617660242, + "loss": 0.3501, + "step": 28549 + }, + { + "epoch": 1.5951336704193089, + "grad_norm": 0.4606949985027313, + "learning_rate": 0.0002036082474226804, + "loss": 0.4095, + "step": 28550 + }, + { + "epoch": 1.5951895410229908, + "grad_norm": 2.6225645542144775, + "learning_rate": 0.00020358023307933662, + "loss": 0.3143, + "step": 28551 + }, + { + "epoch": 1.5952454116266725, + "grad_norm": 0.36738845705986023, + "learning_rate": 0.00020355221873599282, + "loss": 0.3584, + "step": 28552 + }, + { + "epoch": 1.5953012822303545, + "grad_norm": 0.3963560163974762, + "learning_rate": 0.00020352420439264903, + "loss": 0.4643, + "step": 28553 + }, + { + "epoch": 1.5953571528340365, + "grad_norm": 0.48793280124664307, + "learning_rate": 0.00020349619004930526, + "loss": 0.5232, + "step": 28554 + }, + { + "epoch": 1.5954130234377182, + "grad_norm": 0.5002020597457886, + "learning_rate": 0.00020346817570596147, + "loss": 0.4281, + "step": 28555 + }, + { + "epoch": 1.5954688940414001, + "grad_norm": 0.5268816351890564, + "learning_rate": 0.00020344016136261768, + "loss": 0.5524, + "step": 28556 + }, + { + "epoch": 1.595524764645082, + "grad_norm": 0.6563624143600464, + "learning_rate": 0.00020341214701927388, + "loss": 0.3478, + "step": 28557 + }, + { + "epoch": 1.5955806352487638, + "grad_norm": 2.4309494495391846, + "learning_rate": 0.0002033841326759301, + "loss": 0.3328, + "step": 28558 + }, + { + "epoch": 1.5956365058524458, + "grad_norm": 0.3737892508506775, + "learning_rate": 0.0002033561183325863, + "loss": 0.3776, + "step": 28559 + }, + { + "epoch": 1.5956923764561277, + "grad_norm": 16.24112892150879, + "learning_rate": 0.0002033281039892425, + "loss": 0.3832, + "step": 28560 + }, + { + "epoch": 1.5957482470598094, + "grad_norm": 3.30867075920105, + "learning_rate": 0.0002033000896458987, + "loss": 0.4144, + "step": 28561 + }, + { + "epoch": 1.5958041176634914, + "grad_norm": 0.5005009770393372, + "learning_rate": 0.0002032720753025549, + "loss": 0.4453, + "step": 28562 + }, + { + "epoch": 1.5958599882671733, + "grad_norm": 0.8425697088241577, + "learning_rate": 0.00020324406095921112, + "loss": 0.4579, + "step": 28563 + }, + { + "epoch": 1.595915858870855, + "grad_norm": 0.5855544209480286, + "learning_rate": 0.00020321604661586732, + "loss": 0.357, + "step": 28564 + }, + { + "epoch": 1.595971729474537, + "grad_norm": 0.399544894695282, + "learning_rate": 0.00020318803227252353, + "loss": 0.3577, + "step": 28565 + }, + { + "epoch": 1.596027600078219, + "grad_norm": 0.3806580603122711, + "learning_rate": 0.00020316001792917973, + "loss": 0.3545, + "step": 28566 + }, + { + "epoch": 1.5960834706819007, + "grad_norm": 0.8482164144515991, + "learning_rate": 0.00020313200358583594, + "loss": 0.377, + "step": 28567 + }, + { + "epoch": 1.5961393412855824, + "grad_norm": 0.6015025973320007, + "learning_rate": 0.00020310398924249215, + "loss": 0.6905, + "step": 28568 + }, + { + "epoch": 1.5961952118892646, + "grad_norm": 0.3735581636428833, + "learning_rate": 0.00020307597489914835, + "loss": 0.38, + "step": 28569 + }, + { + "epoch": 1.5962510824929463, + "grad_norm": 0.39700189232826233, + "learning_rate": 0.00020304796055580458, + "loss": 0.3564, + "step": 28570 + }, + { + "epoch": 1.596306953096628, + "grad_norm": 0.2879824936389923, + "learning_rate": 0.0002030199462124608, + "loss": 0.3208, + "step": 28571 + }, + { + "epoch": 1.5963628237003102, + "grad_norm": 0.2763313055038452, + "learning_rate": 0.000202991931869117, + "loss": 0.3322, + "step": 28572 + }, + { + "epoch": 1.596418694303992, + "grad_norm": 0.7233096361160278, + "learning_rate": 0.0002029639175257732, + "loss": 0.4525, + "step": 28573 + }, + { + "epoch": 1.5964745649076737, + "grad_norm": 0.34245777130126953, + "learning_rate": 0.0002029359031824294, + "loss": 0.538, + "step": 28574 + }, + { + "epoch": 1.5965304355113557, + "grad_norm": 0.3082387447357178, + "learning_rate": 0.0002029078888390856, + "loss": 0.3496, + "step": 28575 + }, + { + "epoch": 1.5965863061150376, + "grad_norm": 0.3780938982963562, + "learning_rate": 0.00020287987449574182, + "loss": 0.3253, + "step": 28576 + }, + { + "epoch": 1.5966421767187193, + "grad_norm": 0.4373783469200134, + "learning_rate": 0.00020285186015239802, + "loss": 0.412, + "step": 28577 + }, + { + "epoch": 1.5966980473224013, + "grad_norm": 0.3267117738723755, + "learning_rate": 0.00020282384580905426, + "loss": 0.439, + "step": 28578 + }, + { + "epoch": 1.5967539179260832, + "grad_norm": 0.3893810510635376, + "learning_rate": 0.00020279583146571046, + "loss": 0.4494, + "step": 28579 + }, + { + "epoch": 1.596809788529765, + "grad_norm": 0.4511619508266449, + "learning_rate": 0.00020276781712236667, + "loss": 0.4653, + "step": 28580 + }, + { + "epoch": 1.596865659133447, + "grad_norm": 0.6710975170135498, + "learning_rate": 0.00020273980277902287, + "loss": 0.3745, + "step": 28581 + }, + { + "epoch": 1.5969215297371289, + "grad_norm": 0.3990357220172882, + "learning_rate": 0.00020271178843567908, + "loss": 0.3759, + "step": 28582 + }, + { + "epoch": 1.5969774003408106, + "grad_norm": 3.5916199684143066, + "learning_rate": 0.0002026837740923353, + "loss": 0.3855, + "step": 28583 + }, + { + "epoch": 1.5970332709444925, + "grad_norm": 0.4406909644603729, + "learning_rate": 0.00020265575974899147, + "loss": 0.3121, + "step": 28584 + }, + { + "epoch": 1.5970891415481745, + "grad_norm": 0.3910326063632965, + "learning_rate": 0.00020262774540564767, + "loss": 0.4825, + "step": 28585 + }, + { + "epoch": 1.5971450121518562, + "grad_norm": 0.3633323609828949, + "learning_rate": 0.0002025997310623039, + "loss": 0.342, + "step": 28586 + }, + { + "epoch": 1.5972008827555382, + "grad_norm": 0.4268137216567993, + "learning_rate": 0.0002025717167189601, + "loss": 0.4329, + "step": 28587 + }, + { + "epoch": 1.5972567533592201, + "grad_norm": 0.38238388299942017, + "learning_rate": 0.00020254370237561632, + "loss": 0.3158, + "step": 28588 + }, + { + "epoch": 1.5973126239629019, + "grad_norm": 0.749289333820343, + "learning_rate": 0.00020251568803227252, + "loss": 0.3532, + "step": 28589 + }, + { + "epoch": 1.5973684945665838, + "grad_norm": 0.641160786151886, + "learning_rate": 0.00020248767368892873, + "loss": 0.3623, + "step": 28590 + }, + { + "epoch": 1.5974243651702658, + "grad_norm": 0.6188558340072632, + "learning_rate": 0.00020245965934558493, + "loss": 0.4917, + "step": 28591 + }, + { + "epoch": 1.5974802357739475, + "grad_norm": 0.34409356117248535, + "learning_rate": 0.00020243164500224114, + "loss": 0.3617, + "step": 28592 + }, + { + "epoch": 1.5975361063776294, + "grad_norm": 0.29962196946144104, + "learning_rate": 0.00020240363065889737, + "loss": 0.3516, + "step": 28593 + }, + { + "epoch": 1.5975919769813114, + "grad_norm": 0.5986258387565613, + "learning_rate": 0.00020237561631555358, + "loss": 0.3436, + "step": 28594 + }, + { + "epoch": 1.5976478475849931, + "grad_norm": 0.35044100880622864, + "learning_rate": 0.00020234760197220978, + "loss": 0.3953, + "step": 28595 + }, + { + "epoch": 1.597703718188675, + "grad_norm": 0.3660466969013214, + "learning_rate": 0.000202319587628866, + "loss": 0.3795, + "step": 28596 + }, + { + "epoch": 1.597759588792357, + "grad_norm": 0.5011192560195923, + "learning_rate": 0.0002022915732855222, + "loss": 0.3251, + "step": 28597 + }, + { + "epoch": 1.5978154593960388, + "grad_norm": 0.604195773601532, + "learning_rate": 0.0002022635589421784, + "loss": 0.4869, + "step": 28598 + }, + { + "epoch": 1.5978713299997207, + "grad_norm": 0.3488648533821106, + "learning_rate": 0.0002022355445988346, + "loss": 0.3539, + "step": 28599 + }, + { + "epoch": 1.5979272006034027, + "grad_norm": 2.4790260791778564, + "learning_rate": 0.0002022075302554908, + "loss": 0.4651, + "step": 28600 + }, + { + "epoch": 1.5979830712070844, + "grad_norm": 0.3544570207595825, + "learning_rate": 0.00020217951591214705, + "loss": 0.4028, + "step": 28601 + }, + { + "epoch": 1.5980389418107661, + "grad_norm": 0.7251235246658325, + "learning_rate": 0.00020215150156880325, + "loss": 0.343, + "step": 28602 + }, + { + "epoch": 1.5980948124144483, + "grad_norm": 1.6546943187713623, + "learning_rate": 0.00020212348722545946, + "loss": 0.6209, + "step": 28603 + }, + { + "epoch": 1.59815068301813, + "grad_norm": 0.32817983627319336, + "learning_rate": 0.00020209547288211566, + "loss": 0.3423, + "step": 28604 + }, + { + "epoch": 1.5982065536218117, + "grad_norm": 1.5428200960159302, + "learning_rate": 0.00020206745853877184, + "loss": 0.3969, + "step": 28605 + }, + { + "epoch": 1.598262424225494, + "grad_norm": 0.33151426911354065, + "learning_rate": 0.00020203944419542805, + "loss": 0.3076, + "step": 28606 + }, + { + "epoch": 1.5983182948291756, + "grad_norm": 0.3598484694957733, + "learning_rate": 0.00020201142985208425, + "loss": 0.3367, + "step": 28607 + }, + { + "epoch": 1.5983741654328574, + "grad_norm": 0.41096845269203186, + "learning_rate": 0.00020198341550874046, + "loss": 0.4188, + "step": 28608 + }, + { + "epoch": 1.5984300360365393, + "grad_norm": 8.174324989318848, + "learning_rate": 0.0002019554011653967, + "loss": 0.5024, + "step": 28609 + }, + { + "epoch": 1.5984859066402213, + "grad_norm": 0.7011489272117615, + "learning_rate": 0.0002019273868220529, + "loss": 0.3408, + "step": 28610 + }, + { + "epoch": 1.598541777243903, + "grad_norm": 0.4033789336681366, + "learning_rate": 0.0002018993724787091, + "loss": 0.4135, + "step": 28611 + }, + { + "epoch": 1.598597647847585, + "grad_norm": 0.33754539489746094, + "learning_rate": 0.0002018713581353653, + "loss": 0.4694, + "step": 28612 + }, + { + "epoch": 1.598653518451267, + "grad_norm": 0.580730676651001, + "learning_rate": 0.00020184334379202152, + "loss": 0.4128, + "step": 28613 + }, + { + "epoch": 1.5987093890549486, + "grad_norm": 0.346299946308136, + "learning_rate": 0.00020181532944867772, + "loss": 0.342, + "step": 28614 + }, + { + "epoch": 1.5987652596586306, + "grad_norm": 0.491670697927475, + "learning_rate": 0.00020178731510533393, + "loss": 0.4814, + "step": 28615 + }, + { + "epoch": 1.5988211302623125, + "grad_norm": 0.4185490608215332, + "learning_rate": 0.00020175930076199013, + "loss": 0.3249, + "step": 28616 + }, + { + "epoch": 1.5988770008659943, + "grad_norm": 0.7029772996902466, + "learning_rate": 0.00020173128641864637, + "loss": 0.3791, + "step": 28617 + }, + { + "epoch": 1.5989328714696762, + "grad_norm": 0.4873393177986145, + "learning_rate": 0.00020170327207530257, + "loss": 0.4817, + "step": 28618 + }, + { + "epoch": 1.5989887420733582, + "grad_norm": 0.5960863828659058, + "learning_rate": 0.00020167525773195878, + "loss": 0.4342, + "step": 28619 + }, + { + "epoch": 1.59904461267704, + "grad_norm": 0.6539536118507385, + "learning_rate": 0.00020164724338861498, + "loss": 0.4373, + "step": 28620 + }, + { + "epoch": 1.5991004832807219, + "grad_norm": 0.40659525990486145, + "learning_rate": 0.0002016192290452712, + "loss": 0.3674, + "step": 28621 + }, + { + "epoch": 1.5991563538844038, + "grad_norm": 1.3437129259109497, + "learning_rate": 0.0002015912147019274, + "loss": 0.4584, + "step": 28622 + }, + { + "epoch": 1.5992122244880855, + "grad_norm": 0.3138907849788666, + "learning_rate": 0.0002015632003585836, + "loss": 0.4108, + "step": 28623 + }, + { + "epoch": 1.5992680950917675, + "grad_norm": 0.8362243175506592, + "learning_rate": 0.0002015351860152398, + "loss": 0.3283, + "step": 28624 + }, + { + "epoch": 1.5993239656954494, + "grad_norm": 1.0698989629745483, + "learning_rate": 0.00020150717167189604, + "loss": 0.4063, + "step": 28625 + }, + { + "epoch": 1.5993798362991312, + "grad_norm": 0.6119518876075745, + "learning_rate": 0.00020147915732855222, + "loss": 0.5768, + "step": 28626 + }, + { + "epoch": 1.5994357069028131, + "grad_norm": 0.3705936074256897, + "learning_rate": 0.00020145114298520842, + "loss": 0.3874, + "step": 28627 + }, + { + "epoch": 1.599491577506495, + "grad_norm": 0.7573403716087341, + "learning_rate": 0.00020142312864186463, + "loss": 0.4707, + "step": 28628 + }, + { + "epoch": 1.5995474481101768, + "grad_norm": 0.42760151624679565, + "learning_rate": 0.00020139511429852084, + "loss": 0.3679, + "step": 28629 + }, + { + "epoch": 1.5996033187138587, + "grad_norm": 0.399627149105072, + "learning_rate": 0.00020136709995517704, + "loss": 0.4663, + "step": 28630 + }, + { + "epoch": 1.5996591893175407, + "grad_norm": 0.5375787615776062, + "learning_rate": 0.00020133908561183325, + "loss": 0.4907, + "step": 28631 + }, + { + "epoch": 1.5997150599212224, + "grad_norm": 0.5105729103088379, + "learning_rate": 0.00020131107126848945, + "loss": 0.4496, + "step": 28632 + }, + { + "epoch": 1.5997709305249042, + "grad_norm": 0.2828086316585541, + "learning_rate": 0.00020128305692514569, + "loss": 0.2766, + "step": 28633 + }, + { + "epoch": 1.5998268011285863, + "grad_norm": 0.3374841511249542, + "learning_rate": 0.0002012550425818019, + "loss": 0.4018, + "step": 28634 + }, + { + "epoch": 1.599882671732268, + "grad_norm": 0.6078690886497498, + "learning_rate": 0.0002012270282384581, + "loss": 0.4006, + "step": 28635 + }, + { + "epoch": 1.5999385423359498, + "grad_norm": 0.48359084129333496, + "learning_rate": 0.0002011990138951143, + "loss": 0.3846, + "step": 28636 + }, + { + "epoch": 1.599994412939632, + "grad_norm": 0.3961501717567444, + "learning_rate": 0.0002011709995517705, + "loss": 0.3858, + "step": 28637 + }, + { + "epoch": 1.6000502835433137, + "grad_norm": 0.44056612253189087, + "learning_rate": 0.00020114298520842671, + "loss": 0.3037, + "step": 28638 + }, + { + "epoch": 1.6001061541469954, + "grad_norm": 0.33803245425224304, + "learning_rate": 0.00020111497086508292, + "loss": 0.2758, + "step": 28639 + }, + { + "epoch": 1.6001620247506776, + "grad_norm": 1.4871724843978882, + "learning_rate": 0.00020108695652173913, + "loss": 0.3061, + "step": 28640 + }, + { + "epoch": 1.6002178953543593, + "grad_norm": 0.7463836073875427, + "learning_rate": 0.00020105894217839536, + "loss": 0.3989, + "step": 28641 + }, + { + "epoch": 1.600273765958041, + "grad_norm": 0.4283444881439209, + "learning_rate": 0.00020103092783505157, + "loss": 0.3712, + "step": 28642 + }, + { + "epoch": 1.600329636561723, + "grad_norm": 0.4099065065383911, + "learning_rate": 0.00020100291349170777, + "loss": 0.3009, + "step": 28643 + }, + { + "epoch": 1.600385507165405, + "grad_norm": 0.4155885875225067, + "learning_rate": 0.00020097489914836398, + "loss": 0.5372, + "step": 28644 + }, + { + "epoch": 1.6004413777690867, + "grad_norm": 0.4700426757335663, + "learning_rate": 0.00020094688480502018, + "loss": 0.3171, + "step": 28645 + }, + { + "epoch": 1.6004972483727686, + "grad_norm": 0.6418483257293701, + "learning_rate": 0.0002009188704616764, + "loss": 0.4302, + "step": 28646 + }, + { + "epoch": 1.6005531189764506, + "grad_norm": 0.5006236433982849, + "learning_rate": 0.00020089085611833257, + "loss": 0.4003, + "step": 28647 + }, + { + "epoch": 1.6006089895801323, + "grad_norm": 0.3205326199531555, + "learning_rate": 0.00020086284177498877, + "loss": 0.3733, + "step": 28648 + }, + { + "epoch": 1.6006648601838143, + "grad_norm": 0.46191224455833435, + "learning_rate": 0.000200834827431645, + "loss": 0.3599, + "step": 28649 + }, + { + "epoch": 1.6007207307874962, + "grad_norm": 0.34830421209335327, + "learning_rate": 0.0002008068130883012, + "loss": 0.3675, + "step": 28650 + }, + { + "epoch": 1.600776601391178, + "grad_norm": 0.5353772044181824, + "learning_rate": 0.00020077879874495742, + "loss": 0.4448, + "step": 28651 + }, + { + "epoch": 1.60083247199486, + "grad_norm": 0.3906761407852173, + "learning_rate": 0.00020075078440161362, + "loss": 0.4262, + "step": 28652 + }, + { + "epoch": 1.6008883425985418, + "grad_norm": 0.4168716371059418, + "learning_rate": 0.00020072277005826983, + "loss": 0.4108, + "step": 28653 + }, + { + "epoch": 1.6009442132022236, + "grad_norm": 0.46180278062820435, + "learning_rate": 0.00020069475571492603, + "loss": 0.4801, + "step": 28654 + }, + { + "epoch": 1.6010000838059055, + "grad_norm": 0.40836748480796814, + "learning_rate": 0.00020066674137158224, + "loss": 0.389, + "step": 28655 + }, + { + "epoch": 1.6010559544095875, + "grad_norm": 4.842157363891602, + "learning_rate": 0.00020063872702823847, + "loss": 0.3329, + "step": 28656 + }, + { + "epoch": 1.6011118250132692, + "grad_norm": 0.6405971646308899, + "learning_rate": 0.00020061071268489468, + "loss": 0.4795, + "step": 28657 + }, + { + "epoch": 1.6011676956169512, + "grad_norm": 0.5848654508590698, + "learning_rate": 0.00020058269834155089, + "loss": 0.5187, + "step": 28658 + }, + { + "epoch": 1.601223566220633, + "grad_norm": 0.5424283146858215, + "learning_rate": 0.0002005546839982071, + "loss": 0.473, + "step": 28659 + }, + { + "epoch": 1.6012794368243148, + "grad_norm": 0.5851477384567261, + "learning_rate": 0.0002005266696548633, + "loss": 0.3835, + "step": 28660 + }, + { + "epoch": 1.6013353074279968, + "grad_norm": 0.9736050963401794, + "learning_rate": 0.0002004986553115195, + "loss": 0.4591, + "step": 28661 + }, + { + "epoch": 1.6013911780316787, + "grad_norm": 0.6866334676742554, + "learning_rate": 0.0002004706409681757, + "loss": 0.435, + "step": 28662 + }, + { + "epoch": 1.6014470486353605, + "grad_norm": 0.38719555735588074, + "learning_rate": 0.00020044262662483191, + "loss": 0.4244, + "step": 28663 + }, + { + "epoch": 1.6015029192390424, + "grad_norm": 0.32801032066345215, + "learning_rate": 0.00020041461228148815, + "loss": 0.3824, + "step": 28664 + }, + { + "epoch": 1.6015587898427244, + "grad_norm": 0.3810858428478241, + "learning_rate": 0.00020038659793814435, + "loss": 0.4594, + "step": 28665 + }, + { + "epoch": 1.601614660446406, + "grad_norm": 0.4254544675350189, + "learning_rate": 0.00020035858359480056, + "loss": 0.3597, + "step": 28666 + }, + { + "epoch": 1.6016705310500878, + "grad_norm": 0.7517873048782349, + "learning_rate": 0.00020033056925145676, + "loss": 0.6077, + "step": 28667 + }, + { + "epoch": 1.60172640165377, + "grad_norm": 0.4229961633682251, + "learning_rate": 0.00020030255490811297, + "loss": 0.4088, + "step": 28668 + }, + { + "epoch": 1.6017822722574517, + "grad_norm": 1.09153413772583, + "learning_rate": 0.00020027454056476915, + "loss": 0.3969, + "step": 28669 + }, + { + "epoch": 1.6018381428611335, + "grad_norm": 0.3394998013973236, + "learning_rate": 0.00020024652622142536, + "loss": 0.3208, + "step": 28670 + }, + { + "epoch": 1.6018940134648156, + "grad_norm": 0.45240527391433716, + "learning_rate": 0.00020021851187808156, + "loss": 0.3705, + "step": 28671 + }, + { + "epoch": 1.6019498840684974, + "grad_norm": 0.34440454840660095, + "learning_rate": 0.0002001904975347378, + "loss": 0.3621, + "step": 28672 + }, + { + "epoch": 1.602005754672179, + "grad_norm": 0.843297004699707, + "learning_rate": 0.000200162483191394, + "loss": 0.4082, + "step": 28673 + }, + { + "epoch": 1.6020616252758613, + "grad_norm": 0.4263294041156769, + "learning_rate": 0.0002001344688480502, + "loss": 0.5023, + "step": 28674 + }, + { + "epoch": 1.602117495879543, + "grad_norm": 0.43694639205932617, + "learning_rate": 0.0002001064545047064, + "loss": 0.4721, + "step": 28675 + }, + { + "epoch": 1.6021733664832247, + "grad_norm": 0.7046859860420227, + "learning_rate": 0.00020007844016136262, + "loss": 0.5286, + "step": 28676 + }, + { + "epoch": 1.6022292370869067, + "grad_norm": 0.35191425681114197, + "learning_rate": 0.00020005042581801882, + "loss": 0.4227, + "step": 28677 + }, + { + "epoch": 1.6022851076905886, + "grad_norm": 0.5319810509681702, + "learning_rate": 0.00020002241147467503, + "loss": 0.5211, + "step": 28678 + }, + { + "epoch": 1.6023409782942704, + "grad_norm": 0.4300205707550049, + "learning_rate": 0.00019999439713133123, + "loss": 0.4987, + "step": 28679 + }, + { + "epoch": 1.6023968488979523, + "grad_norm": 0.4378259479999542, + "learning_rate": 0.00019996638278798747, + "loss": 0.3955, + "step": 28680 + }, + { + "epoch": 1.6024527195016343, + "grad_norm": 0.36123067140579224, + "learning_rate": 0.00019993836844464367, + "loss": 0.3816, + "step": 28681 + }, + { + "epoch": 1.602508590105316, + "grad_norm": 3.8334157466888428, + "learning_rate": 0.00019991035410129988, + "loss": 0.3746, + "step": 28682 + }, + { + "epoch": 1.602564460708998, + "grad_norm": 2.868262529373169, + "learning_rate": 0.00019988233975795608, + "loss": 0.3486, + "step": 28683 + }, + { + "epoch": 1.60262033131268, + "grad_norm": 0.4505320191383362, + "learning_rate": 0.0001998543254146123, + "loss": 0.3862, + "step": 28684 + }, + { + "epoch": 1.6026762019163616, + "grad_norm": 0.3613250255584717, + "learning_rate": 0.0001998263110712685, + "loss": 0.5311, + "step": 28685 + }, + { + "epoch": 1.6027320725200436, + "grad_norm": 0.5061860680580139, + "learning_rate": 0.0001997982967279247, + "loss": 0.532, + "step": 28686 + }, + { + "epoch": 1.6027879431237255, + "grad_norm": 0.687621533870697, + "learning_rate": 0.0001997702823845809, + "loss": 0.5106, + "step": 28687 + }, + { + "epoch": 1.6028438137274073, + "grad_norm": 0.3400905430316925, + "learning_rate": 0.00019974226804123714, + "loss": 0.3434, + "step": 28688 + }, + { + "epoch": 1.6028996843310892, + "grad_norm": 1.7270747423171997, + "learning_rate": 0.00019971425369789335, + "loss": 0.5044, + "step": 28689 + }, + { + "epoch": 1.6029555549347712, + "grad_norm": 0.3383721113204956, + "learning_rate": 0.00019968623935454953, + "loss": 0.3722, + "step": 28690 + }, + { + "epoch": 1.6030114255384529, + "grad_norm": 0.9117429256439209, + "learning_rate": 0.00019965822501120573, + "loss": 0.4213, + "step": 28691 + }, + { + "epoch": 1.6030672961421348, + "grad_norm": 0.7816744446754456, + "learning_rate": 0.00019963021066786194, + "loss": 0.4458, + "step": 28692 + }, + { + "epoch": 1.6031231667458168, + "grad_norm": 0.4118538498878479, + "learning_rate": 0.00019960219632451814, + "loss": 0.3887, + "step": 28693 + }, + { + "epoch": 1.6031790373494985, + "grad_norm": 0.2953588664531708, + "learning_rate": 0.00019957418198117435, + "loss": 0.3517, + "step": 28694 + }, + { + "epoch": 1.6032349079531805, + "grad_norm": 2.1849567890167236, + "learning_rate": 0.00019954616763783055, + "loss": 0.7163, + "step": 28695 + }, + { + "epoch": 1.6032907785568624, + "grad_norm": 2.2881109714508057, + "learning_rate": 0.0001995181532944868, + "loss": 0.3927, + "step": 28696 + }, + { + "epoch": 1.6033466491605441, + "grad_norm": 0.3573470115661621, + "learning_rate": 0.000199490138951143, + "loss": 0.4347, + "step": 28697 + }, + { + "epoch": 1.603402519764226, + "grad_norm": 0.3495219349861145, + "learning_rate": 0.0001994621246077992, + "loss": 0.3577, + "step": 28698 + }, + { + "epoch": 1.603458390367908, + "grad_norm": 0.46802881360054016, + "learning_rate": 0.0001994341102644554, + "loss": 0.4425, + "step": 28699 + }, + { + "epoch": 1.6035142609715898, + "grad_norm": 0.5521624088287354, + "learning_rate": 0.0001994060959211116, + "loss": 0.4018, + "step": 28700 + }, + { + "epoch": 1.6035701315752715, + "grad_norm": 0.39590904116630554, + "learning_rate": 0.00019937808157776782, + "loss": 0.3908, + "step": 28701 + }, + { + "epoch": 1.6036260021789537, + "grad_norm": 0.6901087760925293, + "learning_rate": 0.00019935006723442402, + "loss": 0.445, + "step": 28702 + }, + { + "epoch": 1.6036818727826354, + "grad_norm": 0.5113982558250427, + "learning_rate": 0.00019932205289108023, + "loss": 0.4113, + "step": 28703 + }, + { + "epoch": 1.6037377433863171, + "grad_norm": 0.5954023599624634, + "learning_rate": 0.00019929403854773646, + "loss": 0.4139, + "step": 28704 + }, + { + "epoch": 1.6037936139899993, + "grad_norm": 0.3436248004436493, + "learning_rate": 0.00019926602420439267, + "loss": 0.4807, + "step": 28705 + }, + { + "epoch": 1.603849484593681, + "grad_norm": 1.376076579093933, + "learning_rate": 0.00019923800986104887, + "loss": 0.3834, + "step": 28706 + }, + { + "epoch": 1.6039053551973628, + "grad_norm": 0.5507269501686096, + "learning_rate": 0.00019920999551770508, + "loss": 0.4574, + "step": 28707 + }, + { + "epoch": 1.603961225801045, + "grad_norm": 0.3499031364917755, + "learning_rate": 0.00019918198117436128, + "loss": 0.3552, + "step": 28708 + }, + { + "epoch": 1.6040170964047267, + "grad_norm": 0.426722913980484, + "learning_rate": 0.0001991539668310175, + "loss": 0.4567, + "step": 28709 + }, + { + "epoch": 1.6040729670084084, + "grad_norm": 0.43283528089523315, + "learning_rate": 0.0001991259524876737, + "loss": 0.4136, + "step": 28710 + }, + { + "epoch": 1.6041288376120904, + "grad_norm": 2.2328059673309326, + "learning_rate": 0.00019909793814432987, + "loss": 0.5179, + "step": 28711 + }, + { + "epoch": 1.6041847082157723, + "grad_norm": 12.303603172302246, + "learning_rate": 0.0001990699238009861, + "loss": 0.3602, + "step": 28712 + }, + { + "epoch": 1.604240578819454, + "grad_norm": 0.3783418536186218, + "learning_rate": 0.0001990419094576423, + "loss": 0.3664, + "step": 28713 + }, + { + "epoch": 1.604296449423136, + "grad_norm": 0.8557462692260742, + "learning_rate": 0.00019901389511429852, + "loss": 0.4143, + "step": 28714 + }, + { + "epoch": 1.604352320026818, + "grad_norm": 0.3292660415172577, + "learning_rate": 0.00019898588077095473, + "loss": 0.4346, + "step": 28715 + }, + { + "epoch": 1.6044081906304997, + "grad_norm": 0.29588472843170166, + "learning_rate": 0.00019895786642761093, + "loss": 0.3388, + "step": 28716 + }, + { + "epoch": 1.6044640612341816, + "grad_norm": 0.4360923171043396, + "learning_rate": 0.00019892985208426714, + "loss": 0.4035, + "step": 28717 + }, + { + "epoch": 1.6045199318378636, + "grad_norm": 0.37490442395210266, + "learning_rate": 0.00019890183774092334, + "loss": 0.3516, + "step": 28718 + }, + { + "epoch": 1.6045758024415453, + "grad_norm": 0.9295598268508911, + "learning_rate": 0.00019887382339757955, + "loss": 0.4284, + "step": 28719 + }, + { + "epoch": 1.6046316730452272, + "grad_norm": 0.5408213138580322, + "learning_rate": 0.00019884580905423578, + "loss": 0.303, + "step": 28720 + }, + { + "epoch": 1.6046875436489092, + "grad_norm": 0.3997274339199066, + "learning_rate": 0.000198817794710892, + "loss": 0.4276, + "step": 28721 + }, + { + "epoch": 1.604743414252591, + "grad_norm": 0.39582696557044983, + "learning_rate": 0.0001987897803675482, + "loss": 0.3791, + "step": 28722 + }, + { + "epoch": 1.6047992848562729, + "grad_norm": 0.5424821972846985, + "learning_rate": 0.0001987617660242044, + "loss": 0.4868, + "step": 28723 + }, + { + "epoch": 1.6048551554599548, + "grad_norm": 0.5972121357917786, + "learning_rate": 0.0001987337516808606, + "loss": 0.4402, + "step": 28724 + }, + { + "epoch": 1.6049110260636366, + "grad_norm": 0.5731542110443115, + "learning_rate": 0.0001987057373375168, + "loss": 0.6307, + "step": 28725 + }, + { + "epoch": 1.6049668966673185, + "grad_norm": 0.4726253151893616, + "learning_rate": 0.00019867772299417302, + "loss": 0.4293, + "step": 28726 + }, + { + "epoch": 1.6050227672710005, + "grad_norm": 0.4797905683517456, + "learning_rate": 0.00019864970865082925, + "loss": 0.4549, + "step": 28727 + }, + { + "epoch": 1.6050786378746822, + "grad_norm": 0.3856934905052185, + "learning_rate": 0.00019862169430748545, + "loss": 0.3809, + "step": 28728 + }, + { + "epoch": 1.6051345084783641, + "grad_norm": 0.446283221244812, + "learning_rate": 0.00019859367996414166, + "loss": 0.4695, + "step": 28729 + }, + { + "epoch": 1.605190379082046, + "grad_norm": 1.5253382921218872, + "learning_rate": 0.00019856566562079787, + "loss": 0.3998, + "step": 28730 + }, + { + "epoch": 1.6052462496857278, + "grad_norm": 0.43350762128829956, + "learning_rate": 0.00019853765127745407, + "loss": 0.5411, + "step": 28731 + }, + { + "epoch": 1.6053021202894098, + "grad_norm": 0.441253125667572, + "learning_rate": 0.00019850963693411025, + "loss": 0.4005, + "step": 28732 + }, + { + "epoch": 1.6053579908930917, + "grad_norm": 0.6233014464378357, + "learning_rate": 0.00019848162259076646, + "loss": 0.5813, + "step": 28733 + }, + { + "epoch": 1.6054138614967735, + "grad_norm": 0.4225228726863861, + "learning_rate": 0.00019845360824742266, + "loss": 0.457, + "step": 28734 + }, + { + "epoch": 1.6054697321004552, + "grad_norm": 0.28695493936538696, + "learning_rate": 0.0001984255939040789, + "loss": 0.3231, + "step": 28735 + }, + { + "epoch": 1.6055256027041374, + "grad_norm": 0.36045578122138977, + "learning_rate": 0.0001983975795607351, + "loss": 0.3935, + "step": 28736 + }, + { + "epoch": 1.605581473307819, + "grad_norm": 0.42154762148857117, + "learning_rate": 0.0001983695652173913, + "loss": 0.5449, + "step": 28737 + }, + { + "epoch": 1.6056373439115008, + "grad_norm": 0.6147410869598389, + "learning_rate": 0.0001983415508740475, + "loss": 0.4429, + "step": 28738 + }, + { + "epoch": 1.605693214515183, + "grad_norm": 0.6214414238929749, + "learning_rate": 0.00019831353653070372, + "loss": 0.402, + "step": 28739 + }, + { + "epoch": 1.6057490851188647, + "grad_norm": 0.7707720398902893, + "learning_rate": 0.00019828552218735992, + "loss": 0.4881, + "step": 28740 + }, + { + "epoch": 1.6058049557225464, + "grad_norm": 0.3656427562236786, + "learning_rate": 0.00019825750784401613, + "loss": 0.3425, + "step": 28741 + }, + { + "epoch": 1.6058608263262286, + "grad_norm": 0.4152282178401947, + "learning_rate": 0.00019822949350067234, + "loss": 0.407, + "step": 28742 + }, + { + "epoch": 1.6059166969299103, + "grad_norm": 0.35979384183883667, + "learning_rate": 0.00019820147915732857, + "loss": 0.4334, + "step": 28743 + }, + { + "epoch": 1.605972567533592, + "grad_norm": 0.4132451117038727, + "learning_rate": 0.00019817346481398478, + "loss": 0.4396, + "step": 28744 + }, + { + "epoch": 1.606028438137274, + "grad_norm": 7.7991204261779785, + "learning_rate": 0.00019814545047064098, + "loss": 0.4359, + "step": 28745 + }, + { + "epoch": 1.606084308740956, + "grad_norm": 0.5665789246559143, + "learning_rate": 0.0001981174361272972, + "loss": 0.3709, + "step": 28746 + }, + { + "epoch": 1.6061401793446377, + "grad_norm": 0.30484914779663086, + "learning_rate": 0.0001980894217839534, + "loss": 0.3587, + "step": 28747 + }, + { + "epoch": 1.6061960499483197, + "grad_norm": 1.171858549118042, + "learning_rate": 0.0001980614074406096, + "loss": 0.5047, + "step": 28748 + }, + { + "epoch": 1.6062519205520016, + "grad_norm": 0.7311542630195618, + "learning_rate": 0.0001980333930972658, + "loss": 0.6483, + "step": 28749 + }, + { + "epoch": 1.6063077911556833, + "grad_norm": 0.4148288071155548, + "learning_rate": 0.000198005378753922, + "loss": 0.3375, + "step": 28750 + }, + { + "epoch": 1.6063636617593653, + "grad_norm": 0.45846837759017944, + "learning_rate": 0.00019797736441057824, + "loss": 0.3898, + "step": 28751 + }, + { + "epoch": 1.6064195323630472, + "grad_norm": 0.48151081800460815, + "learning_rate": 0.00019794935006723445, + "loss": 0.4056, + "step": 28752 + }, + { + "epoch": 1.606475402966729, + "grad_norm": 0.36502113938331604, + "learning_rate": 0.00019792133572389065, + "loss": 0.3639, + "step": 28753 + }, + { + "epoch": 1.606531273570411, + "grad_norm": 0.5420340299606323, + "learning_rate": 0.00019789332138054683, + "loss": 0.4749, + "step": 28754 + }, + { + "epoch": 1.6065871441740929, + "grad_norm": 0.43559780716896057, + "learning_rate": 0.00019786530703720304, + "loss": 0.4914, + "step": 28755 + }, + { + "epoch": 1.6066430147777746, + "grad_norm": 0.4007667601108551, + "learning_rate": 0.00019783729269385924, + "loss": 0.4589, + "step": 28756 + }, + { + "epoch": 1.6066988853814566, + "grad_norm": 0.5307853817939758, + "learning_rate": 0.00019780927835051545, + "loss": 0.4137, + "step": 28757 + }, + { + "epoch": 1.6067547559851385, + "grad_norm": 0.40152600407600403, + "learning_rate": 0.00019778126400717166, + "loss": 0.3416, + "step": 28758 + }, + { + "epoch": 1.6068106265888202, + "grad_norm": 0.31484514474868774, + "learning_rate": 0.0001977532496638279, + "loss": 0.3133, + "step": 28759 + }, + { + "epoch": 1.6068664971925022, + "grad_norm": 0.4041673243045807, + "learning_rate": 0.0001977252353204841, + "loss": 0.3727, + "step": 28760 + }, + { + "epoch": 1.6069223677961841, + "grad_norm": 0.39711034297943115, + "learning_rate": 0.0001976972209771403, + "loss": 0.3636, + "step": 28761 + }, + { + "epoch": 1.6069782383998659, + "grad_norm": 0.4299168288707733, + "learning_rate": 0.0001976692066337965, + "loss": 0.3971, + "step": 28762 + }, + { + "epoch": 1.6070341090035478, + "grad_norm": 0.5622867941856384, + "learning_rate": 0.0001976411922904527, + "loss": 0.3861, + "step": 28763 + }, + { + "epoch": 1.6070899796072298, + "grad_norm": 0.35016220808029175, + "learning_rate": 0.00019761317794710892, + "loss": 0.4132, + "step": 28764 + }, + { + "epoch": 1.6071458502109115, + "grad_norm": 0.3659639358520508, + "learning_rate": 0.00019758516360376512, + "loss": 0.4158, + "step": 28765 + }, + { + "epoch": 1.6072017208145934, + "grad_norm": 0.6631718873977661, + "learning_rate": 0.00019755714926042133, + "loss": 0.7616, + "step": 28766 + }, + { + "epoch": 1.6072575914182754, + "grad_norm": 0.9144445061683655, + "learning_rate": 0.00019752913491707756, + "loss": 0.4077, + "step": 28767 + }, + { + "epoch": 1.6073134620219571, + "grad_norm": 0.4183580279350281, + "learning_rate": 0.00019750112057373377, + "loss": 0.3796, + "step": 28768 + }, + { + "epoch": 1.6073693326256389, + "grad_norm": 0.43834707140922546, + "learning_rate": 0.00019747310623038997, + "loss": 0.4594, + "step": 28769 + }, + { + "epoch": 1.607425203229321, + "grad_norm": 0.35140225291252136, + "learning_rate": 0.00019744509188704618, + "loss": 0.4167, + "step": 28770 + }, + { + "epoch": 1.6074810738330028, + "grad_norm": 1.2335411310195923, + "learning_rate": 0.00019741707754370239, + "loss": 0.4424, + "step": 28771 + }, + { + "epoch": 1.6075369444366845, + "grad_norm": 0.4215202033519745, + "learning_rate": 0.0001973890632003586, + "loss": 0.4367, + "step": 28772 + }, + { + "epoch": 1.6075928150403667, + "grad_norm": 2.8595054149627686, + "learning_rate": 0.0001973610488570148, + "loss": 0.4673, + "step": 28773 + }, + { + "epoch": 1.6076486856440484, + "grad_norm": 0.42553412914276123, + "learning_rate": 0.000197333034513671, + "loss": 0.383, + "step": 28774 + }, + { + "epoch": 1.6077045562477301, + "grad_norm": 0.43161872029304504, + "learning_rate": 0.0001973050201703272, + "loss": 0.4451, + "step": 28775 + }, + { + "epoch": 1.6077604268514123, + "grad_norm": 0.4152374267578125, + "learning_rate": 0.00019727700582698342, + "loss": 0.3894, + "step": 28776 + }, + { + "epoch": 1.607816297455094, + "grad_norm": 0.3996700048446655, + "learning_rate": 0.00019724899148363962, + "loss": 0.3737, + "step": 28777 + }, + { + "epoch": 1.6078721680587758, + "grad_norm": 0.3583279550075531, + "learning_rate": 0.00019722097714029583, + "loss": 0.3592, + "step": 28778 + }, + { + "epoch": 1.6079280386624577, + "grad_norm": 0.5707963705062866, + "learning_rate": 0.00019719296279695203, + "loss": 0.3971, + "step": 28779 + }, + { + "epoch": 1.6079839092661397, + "grad_norm": 0.3832622468471527, + "learning_rate": 0.00019716494845360824, + "loss": 0.414, + "step": 28780 + }, + { + "epoch": 1.6080397798698214, + "grad_norm": 2.0887889862060547, + "learning_rate": 0.00019713693411026444, + "loss": 0.4145, + "step": 28781 + }, + { + "epoch": 1.6080956504735033, + "grad_norm": 1.8490314483642578, + "learning_rate": 0.00019710891976692065, + "loss": 0.5513, + "step": 28782 + }, + { + "epoch": 1.6081515210771853, + "grad_norm": 0.4015319049358368, + "learning_rate": 0.00019708090542357688, + "loss": 0.3282, + "step": 28783 + }, + { + "epoch": 1.608207391680867, + "grad_norm": 0.3587709963321686, + "learning_rate": 0.0001970528910802331, + "loss": 0.3524, + "step": 28784 + }, + { + "epoch": 1.608263262284549, + "grad_norm": 1.162678837776184, + "learning_rate": 0.0001970248767368893, + "loss": 0.4442, + "step": 28785 + }, + { + "epoch": 1.608319132888231, + "grad_norm": 0.39603307843208313, + "learning_rate": 0.0001969968623935455, + "loss": 0.4087, + "step": 28786 + }, + { + "epoch": 1.6083750034919126, + "grad_norm": 0.617207944393158, + "learning_rate": 0.0001969688480502017, + "loss": 0.5255, + "step": 28787 + }, + { + "epoch": 1.6084308740955946, + "grad_norm": 1.1622523069381714, + "learning_rate": 0.0001969408337068579, + "loss": 0.3394, + "step": 28788 + }, + { + "epoch": 1.6084867446992765, + "grad_norm": 0.4235650599002838, + "learning_rate": 0.00019691281936351412, + "loss": 0.5486, + "step": 28789 + }, + { + "epoch": 1.6085426153029583, + "grad_norm": 0.46392935514450073, + "learning_rate": 0.00019688480502017035, + "loss": 0.5345, + "step": 28790 + }, + { + "epoch": 1.6085984859066402, + "grad_norm": 0.43209442496299744, + "learning_rate": 0.00019685679067682656, + "loss": 0.388, + "step": 28791 + }, + { + "epoch": 1.6086543565103222, + "grad_norm": 0.3376099467277527, + "learning_rate": 0.00019682877633348276, + "loss": 0.3418, + "step": 28792 + }, + { + "epoch": 1.608710227114004, + "grad_norm": 0.5170657634735107, + "learning_rate": 0.00019680076199013897, + "loss": 0.4164, + "step": 28793 + }, + { + "epoch": 1.6087660977176859, + "grad_norm": 0.5494477152824402, + "learning_rate": 0.00019677274764679517, + "loss": 0.3719, + "step": 28794 + }, + { + "epoch": 1.6088219683213678, + "grad_norm": 0.3365176320075989, + "learning_rate": 0.00019674473330345138, + "loss": 0.3815, + "step": 28795 + }, + { + "epoch": 1.6088778389250495, + "grad_norm": 0.3685118854045868, + "learning_rate": 0.00019671671896010756, + "loss": 0.41, + "step": 28796 + }, + { + "epoch": 1.6089337095287315, + "grad_norm": 0.558035135269165, + "learning_rate": 0.00019668870461676376, + "loss": 0.4333, + "step": 28797 + }, + { + "epoch": 1.6089895801324134, + "grad_norm": 2.092637777328491, + "learning_rate": 0.00019666069027342, + "loss": 0.4083, + "step": 28798 + }, + { + "epoch": 1.6090454507360952, + "grad_norm": 1.2889069318771362, + "learning_rate": 0.0001966326759300762, + "loss": 0.403, + "step": 28799 + }, + { + "epoch": 1.6091013213397771, + "grad_norm": 4.863763809204102, + "learning_rate": 0.0001966046615867324, + "loss": 0.4379, + "step": 28800 + }, + { + "epoch": 1.609157191943459, + "grad_norm": 0.8111980557441711, + "learning_rate": 0.00019657664724338862, + "loss": 0.5779, + "step": 28801 + }, + { + "epoch": 1.6092130625471408, + "grad_norm": 0.5294747352600098, + "learning_rate": 0.00019654863290004482, + "loss": 0.5403, + "step": 28802 + }, + { + "epoch": 1.6092689331508225, + "grad_norm": 0.870937168598175, + "learning_rate": 0.00019652061855670103, + "loss": 0.4514, + "step": 28803 + }, + { + "epoch": 1.6093248037545047, + "grad_norm": 0.4144347906112671, + "learning_rate": 0.00019649260421335723, + "loss": 0.383, + "step": 28804 + }, + { + "epoch": 1.6093806743581864, + "grad_norm": 2.7209668159484863, + "learning_rate": 0.00019646458987001344, + "loss": 0.3943, + "step": 28805 + }, + { + "epoch": 1.6094365449618682, + "grad_norm": 0.3893928825855255, + "learning_rate": 0.00019643657552666967, + "loss": 0.4186, + "step": 28806 + }, + { + "epoch": 1.6094924155655503, + "grad_norm": 2.315631151199341, + "learning_rate": 0.00019640856118332588, + "loss": 0.2999, + "step": 28807 + }, + { + "epoch": 1.609548286169232, + "grad_norm": 0.4377990663051605, + "learning_rate": 0.00019638054683998208, + "loss": 0.5421, + "step": 28808 + }, + { + "epoch": 1.6096041567729138, + "grad_norm": 0.33692610263824463, + "learning_rate": 0.0001963525324966383, + "loss": 0.3607, + "step": 28809 + }, + { + "epoch": 1.609660027376596, + "grad_norm": 0.5235499739646912, + "learning_rate": 0.0001963245181532945, + "loss": 0.3075, + "step": 28810 + }, + { + "epoch": 1.6097158979802777, + "grad_norm": 0.6172971129417419, + "learning_rate": 0.0001962965038099507, + "loss": 0.2631, + "step": 28811 + }, + { + "epoch": 1.6097717685839594, + "grad_norm": 0.2817543148994446, + "learning_rate": 0.0001962684894666069, + "loss": 0.3213, + "step": 28812 + }, + { + "epoch": 1.6098276391876414, + "grad_norm": 0.3374658524990082, + "learning_rate": 0.0001962404751232631, + "loss": 0.4302, + "step": 28813 + }, + { + "epoch": 1.6098835097913233, + "grad_norm": 0.38355064392089844, + "learning_rate": 0.00019621246077991934, + "loss": 0.3254, + "step": 28814 + }, + { + "epoch": 1.609939380395005, + "grad_norm": 0.7155103087425232, + "learning_rate": 0.00019618444643657555, + "loss": 0.3708, + "step": 28815 + }, + { + "epoch": 1.609995250998687, + "grad_norm": 0.5160996317863464, + "learning_rate": 0.00019615643209323176, + "loss": 0.4288, + "step": 28816 + }, + { + "epoch": 1.610051121602369, + "grad_norm": 0.6387158632278442, + "learning_rate": 0.00019612841774988794, + "loss": 0.3855, + "step": 28817 + }, + { + "epoch": 1.6101069922060507, + "grad_norm": 0.5148544311523438, + "learning_rate": 0.00019610040340654414, + "loss": 0.5706, + "step": 28818 + }, + { + "epoch": 1.6101628628097326, + "grad_norm": 0.3091753125190735, + "learning_rate": 0.00019607238906320035, + "loss": 0.3987, + "step": 28819 + }, + { + "epoch": 1.6102187334134146, + "grad_norm": 0.5399022698402405, + "learning_rate": 0.00019604437471985655, + "loss": 0.4631, + "step": 28820 + }, + { + "epoch": 1.6102746040170963, + "grad_norm": 0.4364640414714813, + "learning_rate": 0.00019601636037651276, + "loss": 0.3772, + "step": 28821 + }, + { + "epoch": 1.6103304746207783, + "grad_norm": 4.294776916503906, + "learning_rate": 0.000195988346033169, + "loss": 0.4768, + "step": 28822 + }, + { + "epoch": 1.6103863452244602, + "grad_norm": 0.5319470763206482, + "learning_rate": 0.0001959603316898252, + "loss": 0.3704, + "step": 28823 + }, + { + "epoch": 1.610442215828142, + "grad_norm": 0.483573853969574, + "learning_rate": 0.0001959323173464814, + "loss": 0.5257, + "step": 28824 + }, + { + "epoch": 1.610498086431824, + "grad_norm": 0.45964521169662476, + "learning_rate": 0.0001959043030031376, + "loss": 0.3752, + "step": 28825 + }, + { + "epoch": 1.6105539570355059, + "grad_norm": 0.497001588344574, + "learning_rate": 0.00019587628865979381, + "loss": 0.4653, + "step": 28826 + }, + { + "epoch": 1.6106098276391876, + "grad_norm": 2.381817579269409, + "learning_rate": 0.00019584827431645002, + "loss": 0.4202, + "step": 28827 + }, + { + "epoch": 1.6106656982428695, + "grad_norm": 1.1309776306152344, + "learning_rate": 0.00019582025997310623, + "loss": 0.5322, + "step": 28828 + }, + { + "epoch": 1.6107215688465515, + "grad_norm": 2.6726245880126953, + "learning_rate": 0.00019579224562976243, + "loss": 0.4389, + "step": 28829 + }, + { + "epoch": 1.6107774394502332, + "grad_norm": 0.3720559775829315, + "learning_rate": 0.00019576423128641866, + "loss": 0.4351, + "step": 28830 + }, + { + "epoch": 1.6108333100539152, + "grad_norm": 0.3448653817176819, + "learning_rate": 0.00019573621694307487, + "loss": 0.346, + "step": 28831 + }, + { + "epoch": 1.6108891806575971, + "grad_norm": 0.47085830569267273, + "learning_rate": 0.00019570820259973108, + "loss": 0.4825, + "step": 28832 + }, + { + "epoch": 1.6109450512612788, + "grad_norm": 0.37720534205436707, + "learning_rate": 0.00019568018825638728, + "loss": 0.3184, + "step": 28833 + }, + { + "epoch": 1.6110009218649608, + "grad_norm": 0.518151044845581, + "learning_rate": 0.0001956521739130435, + "loss": 0.5024, + "step": 28834 + }, + { + "epoch": 1.6110567924686428, + "grad_norm": 0.8639337420463562, + "learning_rate": 0.0001956241595696997, + "loss": 0.6078, + "step": 28835 + }, + { + "epoch": 1.6111126630723245, + "grad_norm": 0.4167754054069519, + "learning_rate": 0.0001955961452263559, + "loss": 0.3315, + "step": 28836 + }, + { + "epoch": 1.6111685336760062, + "grad_norm": 0.42905136942863464, + "learning_rate": 0.0001955681308830121, + "loss": 0.3428, + "step": 28837 + }, + { + "epoch": 1.6112244042796884, + "grad_norm": 0.36737051606178284, + "learning_rate": 0.00019554011653966834, + "loss": 0.3948, + "step": 28838 + }, + { + "epoch": 1.6112802748833701, + "grad_norm": 0.41491442918777466, + "learning_rate": 0.00019551210219632452, + "loss": 0.3626, + "step": 28839 + }, + { + "epoch": 1.6113361454870518, + "grad_norm": 0.5571990013122559, + "learning_rate": 0.00019548408785298072, + "loss": 0.4819, + "step": 28840 + }, + { + "epoch": 1.611392016090734, + "grad_norm": 0.38166287541389465, + "learning_rate": 0.00019545607350963693, + "loss": 0.4306, + "step": 28841 + }, + { + "epoch": 1.6114478866944157, + "grad_norm": 0.38945916295051575, + "learning_rate": 0.00019542805916629313, + "loss": 0.3612, + "step": 28842 + }, + { + "epoch": 1.6115037572980975, + "grad_norm": 0.4157577157020569, + "learning_rate": 0.00019540004482294934, + "loss": 0.3065, + "step": 28843 + }, + { + "epoch": 1.6115596279017794, + "grad_norm": 1.4620037078857422, + "learning_rate": 0.00019537203047960555, + "loss": 0.3833, + "step": 28844 + }, + { + "epoch": 1.6116154985054614, + "grad_norm": 0.44660937786102295, + "learning_rate": 0.00019534401613626175, + "loss": 0.4258, + "step": 28845 + }, + { + "epoch": 1.611671369109143, + "grad_norm": 0.3902270793914795, + "learning_rate": 0.00019531600179291799, + "loss": 0.3852, + "step": 28846 + }, + { + "epoch": 1.611727239712825, + "grad_norm": 0.43008676171302795, + "learning_rate": 0.0001952879874495742, + "loss": 0.3673, + "step": 28847 + }, + { + "epoch": 1.611783110316507, + "grad_norm": 0.6640434265136719, + "learning_rate": 0.0001952599731062304, + "loss": 0.4564, + "step": 28848 + }, + { + "epoch": 1.6118389809201887, + "grad_norm": 0.31083589792251587, + "learning_rate": 0.0001952319587628866, + "loss": 0.3749, + "step": 28849 + }, + { + "epoch": 1.6118948515238707, + "grad_norm": 0.651404619216919, + "learning_rate": 0.0001952039444195428, + "loss": 0.4023, + "step": 28850 + }, + { + "epoch": 1.6119507221275526, + "grad_norm": 0.624173641204834, + "learning_rate": 0.00019517593007619901, + "loss": 0.4017, + "step": 28851 + }, + { + "epoch": 1.6120065927312344, + "grad_norm": 0.7161171436309814, + "learning_rate": 0.00019514791573285522, + "loss": 0.7091, + "step": 28852 + }, + { + "epoch": 1.6120624633349163, + "grad_norm": 0.29135382175445557, + "learning_rate": 0.00019511990138951145, + "loss": 0.3037, + "step": 28853 + }, + { + "epoch": 1.6121183339385983, + "grad_norm": 0.48160436749458313, + "learning_rate": 0.00019509188704616766, + "loss": 0.4683, + "step": 28854 + }, + { + "epoch": 1.61217420454228, + "grad_norm": 0.47729891538619995, + "learning_rate": 0.00019506387270282386, + "loss": 0.4206, + "step": 28855 + }, + { + "epoch": 1.612230075145962, + "grad_norm": 1.1560673713684082, + "learning_rate": 0.00019503585835948007, + "loss": 0.5159, + "step": 28856 + }, + { + "epoch": 1.612285945749644, + "grad_norm": 0.36014702916145325, + "learning_rate": 0.00019500784401613628, + "loss": 0.3987, + "step": 28857 + }, + { + "epoch": 1.6123418163533256, + "grad_norm": 0.48458313941955566, + "learning_rate": 0.00019497982967279248, + "loss": 0.4389, + "step": 28858 + }, + { + "epoch": 1.6123976869570076, + "grad_norm": 0.6005481481552124, + "learning_rate": 0.0001949518153294487, + "loss": 0.4796, + "step": 28859 + }, + { + "epoch": 1.6124535575606895, + "grad_norm": 0.7785438299179077, + "learning_rate": 0.00019492380098610487, + "loss": 0.2935, + "step": 28860 + }, + { + "epoch": 1.6125094281643713, + "grad_norm": 0.3716447353363037, + "learning_rate": 0.0001948957866427611, + "loss": 0.3859, + "step": 28861 + }, + { + "epoch": 1.6125652987680532, + "grad_norm": 0.4196670353412628, + "learning_rate": 0.0001948677722994173, + "loss": 0.3293, + "step": 28862 + }, + { + "epoch": 1.6126211693717352, + "grad_norm": 0.5710942149162292, + "learning_rate": 0.0001948397579560735, + "loss": 0.431, + "step": 28863 + }, + { + "epoch": 1.612677039975417, + "grad_norm": 1.5274738073349, + "learning_rate": 0.00019481174361272972, + "loss": 0.3732, + "step": 28864 + }, + { + "epoch": 1.6127329105790988, + "grad_norm": 0.5930102467536926, + "learning_rate": 0.00019478372926938592, + "loss": 0.5773, + "step": 28865 + }, + { + "epoch": 1.6127887811827808, + "grad_norm": 0.5934531092643738, + "learning_rate": 0.00019475571492604213, + "loss": 0.3831, + "step": 28866 + }, + { + "epoch": 1.6128446517864625, + "grad_norm": 0.40676429867744446, + "learning_rate": 0.00019472770058269833, + "loss": 0.3492, + "step": 28867 + }, + { + "epoch": 1.6129005223901445, + "grad_norm": 0.4480780363082886, + "learning_rate": 0.00019469968623935454, + "loss": 0.4275, + "step": 28868 + }, + { + "epoch": 1.6129563929938264, + "grad_norm": 0.5520662069320679, + "learning_rate": 0.00019467167189601077, + "loss": 0.3784, + "step": 28869 + }, + { + "epoch": 1.6130122635975082, + "grad_norm": 0.48484307527542114, + "learning_rate": 0.00019464365755266698, + "loss": 0.4088, + "step": 28870 + }, + { + "epoch": 1.6130681342011899, + "grad_norm": 0.5064995288848877, + "learning_rate": 0.00019461564320932318, + "loss": 0.5669, + "step": 28871 + }, + { + "epoch": 1.613124004804872, + "grad_norm": 0.8694159984588623, + "learning_rate": 0.0001945876288659794, + "loss": 0.4757, + "step": 28872 + }, + { + "epoch": 1.6131798754085538, + "grad_norm": 0.645535409450531, + "learning_rate": 0.0001945596145226356, + "loss": 0.4142, + "step": 28873 + }, + { + "epoch": 1.6132357460122355, + "grad_norm": 0.4037218987941742, + "learning_rate": 0.0001945316001792918, + "loss": 0.3752, + "step": 28874 + }, + { + "epoch": 1.6132916166159177, + "grad_norm": 0.36135080456733704, + "learning_rate": 0.000194503585835948, + "loss": 0.3648, + "step": 28875 + }, + { + "epoch": 1.6133474872195994, + "grad_norm": 0.5516688227653503, + "learning_rate": 0.00019447557149260421, + "loss": 0.4304, + "step": 28876 + }, + { + "epoch": 1.6134033578232811, + "grad_norm": 2.606621265411377, + "learning_rate": 0.00019444755714926045, + "loss": 0.4272, + "step": 28877 + }, + { + "epoch": 1.613459228426963, + "grad_norm": 1.8083966970443726, + "learning_rate": 0.00019441954280591665, + "loss": 0.455, + "step": 28878 + }, + { + "epoch": 1.613515099030645, + "grad_norm": 1.2677849531173706, + "learning_rate": 0.00019439152846257286, + "loss": 0.4018, + "step": 28879 + }, + { + "epoch": 1.6135709696343268, + "grad_norm": 0.5602576732635498, + "learning_rate": 0.00019436351411922906, + "loss": 0.5702, + "step": 28880 + }, + { + "epoch": 1.6136268402380087, + "grad_norm": 0.5750237703323364, + "learning_rate": 0.00019433549977588524, + "loss": 0.3655, + "step": 28881 + }, + { + "epoch": 1.6136827108416907, + "grad_norm": 3.1886632442474365, + "learning_rate": 0.00019430748543254145, + "loss": 0.45, + "step": 28882 + }, + { + "epoch": 1.6137385814453724, + "grad_norm": 0.5176052451133728, + "learning_rate": 0.00019427947108919765, + "loss": 0.4994, + "step": 28883 + }, + { + "epoch": 1.6137944520490544, + "grad_norm": 1.4371278285980225, + "learning_rate": 0.00019425145674585386, + "loss": 0.3196, + "step": 28884 + }, + { + "epoch": 1.6138503226527363, + "grad_norm": 0.45736944675445557, + "learning_rate": 0.0001942234424025101, + "loss": 0.4097, + "step": 28885 + }, + { + "epoch": 1.613906193256418, + "grad_norm": 0.41704675555229187, + "learning_rate": 0.0001941954280591663, + "loss": 0.3602, + "step": 28886 + }, + { + "epoch": 1.6139620638601, + "grad_norm": 0.5919277667999268, + "learning_rate": 0.0001941674137158225, + "loss": 0.3544, + "step": 28887 + }, + { + "epoch": 1.614017934463782, + "grad_norm": 0.36447733640670776, + "learning_rate": 0.0001941393993724787, + "loss": 0.5349, + "step": 28888 + }, + { + "epoch": 1.6140738050674637, + "grad_norm": 0.6985049843788147, + "learning_rate": 0.00019411138502913492, + "loss": 0.5862, + "step": 28889 + }, + { + "epoch": 1.6141296756711456, + "grad_norm": 0.4517659842967987, + "learning_rate": 0.00019408337068579112, + "loss": 0.4077, + "step": 28890 + }, + { + "epoch": 1.6141855462748276, + "grad_norm": 0.7785025835037231, + "learning_rate": 0.00019405535634244733, + "loss": 0.3658, + "step": 28891 + }, + { + "epoch": 1.6142414168785093, + "grad_norm": 0.6890230178833008, + "learning_rate": 0.00019402734199910353, + "loss": 0.4442, + "step": 28892 + }, + { + "epoch": 1.6142972874821913, + "grad_norm": 0.40094122290611267, + "learning_rate": 0.00019399932765575977, + "loss": 0.3777, + "step": 28893 + }, + { + "epoch": 1.6143531580858732, + "grad_norm": 0.4240477979183197, + "learning_rate": 0.00019397131331241597, + "loss": 0.4088, + "step": 28894 + }, + { + "epoch": 1.614409028689555, + "grad_norm": 0.4148198366165161, + "learning_rate": 0.00019394329896907218, + "loss": 0.3313, + "step": 28895 + }, + { + "epoch": 1.6144648992932369, + "grad_norm": 0.39509737491607666, + "learning_rate": 0.00019391528462572838, + "loss": 0.3738, + "step": 28896 + }, + { + "epoch": 1.6145207698969188, + "grad_norm": 0.34194692969322205, + "learning_rate": 0.0001938872702823846, + "loss": 0.3335, + "step": 28897 + }, + { + "epoch": 1.6145766405006006, + "grad_norm": 0.4104746878147125, + "learning_rate": 0.0001938592559390408, + "loss": 0.402, + "step": 28898 + }, + { + "epoch": 1.6146325111042825, + "grad_norm": 0.4632207155227661, + "learning_rate": 0.000193831241595697, + "loss": 0.5021, + "step": 28899 + }, + { + "epoch": 1.6146883817079645, + "grad_norm": 0.42736339569091797, + "learning_rate": 0.0001938032272523532, + "loss": 0.4185, + "step": 28900 + }, + { + "epoch": 1.6147442523116462, + "grad_norm": 1.5182862281799316, + "learning_rate": 0.00019377521290900944, + "loss": 0.7325, + "step": 28901 + }, + { + "epoch": 1.6148001229153282, + "grad_norm": 0.40726837515830994, + "learning_rate": 0.00019374719856566562, + "loss": 0.4075, + "step": 28902 + }, + { + "epoch": 1.61485599351901, + "grad_norm": 0.7182503938674927, + "learning_rate": 0.00019371918422232183, + "loss": 0.4556, + "step": 28903 + }, + { + "epoch": 1.6149118641226918, + "grad_norm": 0.5834602117538452, + "learning_rate": 0.00019369116987897803, + "loss": 0.4115, + "step": 28904 + }, + { + "epoch": 1.6149677347263736, + "grad_norm": 0.7229307889938354, + "learning_rate": 0.00019366315553563424, + "loss": 0.4655, + "step": 28905 + }, + { + "epoch": 1.6150236053300557, + "grad_norm": 0.47022557258605957, + "learning_rate": 0.00019363514119229044, + "loss": 0.5131, + "step": 28906 + }, + { + "epoch": 1.6150794759337375, + "grad_norm": 0.3347988724708557, + "learning_rate": 0.00019360712684894665, + "loss": 0.3708, + "step": 28907 + }, + { + "epoch": 1.6151353465374192, + "grad_norm": 0.402917742729187, + "learning_rate": 0.00019357911250560285, + "loss": 0.3844, + "step": 28908 + }, + { + "epoch": 1.6151912171411014, + "grad_norm": 0.38567399978637695, + "learning_rate": 0.0001935510981622591, + "loss": 0.2871, + "step": 28909 + }, + { + "epoch": 1.615247087744783, + "grad_norm": 0.3521167039871216, + "learning_rate": 0.0001935230838189153, + "loss": 0.4832, + "step": 28910 + }, + { + "epoch": 1.6153029583484648, + "grad_norm": 0.3422611653804779, + "learning_rate": 0.0001934950694755715, + "loss": 0.4222, + "step": 28911 + }, + { + "epoch": 1.6153588289521468, + "grad_norm": 0.40680545568466187, + "learning_rate": 0.0001934670551322277, + "loss": 0.3986, + "step": 28912 + }, + { + "epoch": 1.6154146995558287, + "grad_norm": 1.0352081060409546, + "learning_rate": 0.0001934390407888839, + "loss": 0.5292, + "step": 28913 + }, + { + "epoch": 1.6154705701595105, + "grad_norm": 0.5363004207611084, + "learning_rate": 0.00019341102644554012, + "loss": 0.44, + "step": 28914 + }, + { + "epoch": 1.6155264407631924, + "grad_norm": 0.4279995858669281, + "learning_rate": 0.00019338301210219632, + "loss": 0.4765, + "step": 28915 + }, + { + "epoch": 1.6155823113668744, + "grad_norm": 0.27902641892433167, + "learning_rate": 0.00019335499775885253, + "loss": 0.3299, + "step": 28916 + }, + { + "epoch": 1.615638181970556, + "grad_norm": 0.3852308392524719, + "learning_rate": 0.00019332698341550876, + "loss": 0.3283, + "step": 28917 + }, + { + "epoch": 1.615694052574238, + "grad_norm": 0.3710927665233612, + "learning_rate": 0.00019329896907216497, + "loss": 0.5746, + "step": 28918 + }, + { + "epoch": 1.61574992317792, + "grad_norm": 0.41754150390625, + "learning_rate": 0.00019327095472882117, + "loss": 0.3519, + "step": 28919 + }, + { + "epoch": 1.6158057937816017, + "grad_norm": 0.40322145819664, + "learning_rate": 0.00019324294038547738, + "loss": 0.4815, + "step": 28920 + }, + { + "epoch": 1.6158616643852837, + "grad_norm": 0.4592773914337158, + "learning_rate": 0.00019321492604213358, + "loss": 0.3873, + "step": 28921 + }, + { + "epoch": 1.6159175349889656, + "grad_norm": 0.31695568561553955, + "learning_rate": 0.0001931869116987898, + "loss": 0.345, + "step": 28922 + }, + { + "epoch": 1.6159734055926473, + "grad_norm": 0.36293289065361023, + "learning_rate": 0.000193158897355446, + "loss": 0.3982, + "step": 28923 + }, + { + "epoch": 1.6160292761963293, + "grad_norm": 0.33134570717811584, + "learning_rate": 0.00019313088301210217, + "loss": 0.3847, + "step": 28924 + }, + { + "epoch": 1.6160851468000113, + "grad_norm": 0.6136660575866699, + "learning_rate": 0.0001931028686687584, + "loss": 0.4696, + "step": 28925 + }, + { + "epoch": 1.616141017403693, + "grad_norm": 0.3707423210144043, + "learning_rate": 0.0001930748543254146, + "loss": 0.4454, + "step": 28926 + }, + { + "epoch": 1.616196888007375, + "grad_norm": 0.3817979097366333, + "learning_rate": 0.00019304683998207082, + "loss": 0.3655, + "step": 28927 + }, + { + "epoch": 1.6162527586110569, + "grad_norm": 0.4737897515296936, + "learning_rate": 0.00019301882563872702, + "loss": 0.4585, + "step": 28928 + }, + { + "epoch": 1.6163086292147386, + "grad_norm": 0.38984397053718567, + "learning_rate": 0.00019299081129538323, + "loss": 0.3945, + "step": 28929 + }, + { + "epoch": 1.6163644998184206, + "grad_norm": 0.7715592384338379, + "learning_rate": 0.00019296279695203944, + "loss": 0.4179, + "step": 28930 + }, + { + "epoch": 1.6164203704221025, + "grad_norm": 2.8515121936798096, + "learning_rate": 0.00019293478260869564, + "loss": 0.3068, + "step": 28931 + }, + { + "epoch": 1.6164762410257842, + "grad_norm": 0.4511353671550751, + "learning_rate": 0.00019290676826535187, + "loss": 0.3469, + "step": 28932 + }, + { + "epoch": 1.6165321116294662, + "grad_norm": 0.32708269357681274, + "learning_rate": 0.00019287875392200808, + "loss": 0.5196, + "step": 28933 + }, + { + "epoch": 1.6165879822331481, + "grad_norm": 0.32303598523139954, + "learning_rate": 0.0001928507395786643, + "loss": 0.3323, + "step": 28934 + }, + { + "epoch": 1.6166438528368299, + "grad_norm": 0.47634297609329224, + "learning_rate": 0.0001928227252353205, + "loss": 0.3809, + "step": 28935 + }, + { + "epoch": 1.6166997234405118, + "grad_norm": 0.3924175798892975, + "learning_rate": 0.0001927947108919767, + "loss": 0.3911, + "step": 28936 + }, + { + "epoch": 1.6167555940441938, + "grad_norm": 0.37447184324264526, + "learning_rate": 0.0001927666965486329, + "loss": 0.359, + "step": 28937 + }, + { + "epoch": 1.6168114646478755, + "grad_norm": 0.431489497423172, + "learning_rate": 0.0001927386822052891, + "loss": 0.5042, + "step": 28938 + }, + { + "epoch": 1.6168673352515572, + "grad_norm": 0.43894755840301514, + "learning_rate": 0.00019271066786194532, + "loss": 0.4674, + "step": 28939 + }, + { + "epoch": 1.6169232058552394, + "grad_norm": 0.44867202639579773, + "learning_rate": 0.00019268265351860155, + "loss": 0.4685, + "step": 28940 + }, + { + "epoch": 1.6169790764589211, + "grad_norm": 0.6363849639892578, + "learning_rate": 0.00019265463917525775, + "loss": 0.3212, + "step": 28941 + }, + { + "epoch": 1.6170349470626029, + "grad_norm": 0.4189720153808594, + "learning_rate": 0.00019262662483191396, + "loss": 0.3967, + "step": 28942 + }, + { + "epoch": 1.617090817666285, + "grad_norm": 0.8776332139968872, + "learning_rate": 0.00019259861048857017, + "loss": 0.4923, + "step": 28943 + }, + { + "epoch": 1.6171466882699668, + "grad_norm": 0.369213730096817, + "learning_rate": 0.00019257059614522637, + "loss": 0.4239, + "step": 28944 + }, + { + "epoch": 1.6172025588736485, + "grad_norm": 6.8771796226501465, + "learning_rate": 0.00019254258180188255, + "loss": 0.3345, + "step": 28945 + }, + { + "epoch": 1.6172584294773305, + "grad_norm": 0.5645265579223633, + "learning_rate": 0.00019251456745853876, + "loss": 0.4695, + "step": 28946 + }, + { + "epoch": 1.6173143000810124, + "grad_norm": 0.6873832941055298, + "learning_rate": 0.00019248655311519496, + "loss": 0.5101, + "step": 28947 + }, + { + "epoch": 1.6173701706846941, + "grad_norm": 0.806992769241333, + "learning_rate": 0.0001924585387718512, + "loss": 0.4552, + "step": 28948 + }, + { + "epoch": 1.617426041288376, + "grad_norm": 0.5613232851028442, + "learning_rate": 0.0001924305244285074, + "loss": 0.2996, + "step": 28949 + }, + { + "epoch": 1.617481911892058, + "grad_norm": 0.38307639956474304, + "learning_rate": 0.0001924025100851636, + "loss": 0.423, + "step": 28950 + }, + { + "epoch": 1.6175377824957398, + "grad_norm": 0.2788604199886322, + "learning_rate": 0.0001923744957418198, + "loss": 0.3252, + "step": 28951 + }, + { + "epoch": 1.6175936530994217, + "grad_norm": 0.3463982343673706, + "learning_rate": 0.00019234648139847602, + "loss": 0.3349, + "step": 28952 + }, + { + "epoch": 1.6176495237031037, + "grad_norm": 0.7745634913444519, + "learning_rate": 0.00019231846705513222, + "loss": 0.4414, + "step": 28953 + }, + { + "epoch": 1.6177053943067854, + "grad_norm": 1.77775239944458, + "learning_rate": 0.00019229045271178843, + "loss": 0.4369, + "step": 28954 + }, + { + "epoch": 1.6177612649104673, + "grad_norm": 0.6579762697219849, + "learning_rate": 0.00019226243836844464, + "loss": 0.4841, + "step": 28955 + }, + { + "epoch": 1.6178171355141493, + "grad_norm": 0.5737695693969727, + "learning_rate": 0.00019223442402510087, + "loss": 0.3009, + "step": 28956 + }, + { + "epoch": 1.617873006117831, + "grad_norm": 0.8787167072296143, + "learning_rate": 0.00019220640968175707, + "loss": 0.5005, + "step": 28957 + }, + { + "epoch": 1.617928876721513, + "grad_norm": 0.5313796997070312, + "learning_rate": 0.00019217839533841328, + "loss": 0.4458, + "step": 28958 + }, + { + "epoch": 1.617984747325195, + "grad_norm": 0.36800962686538696, + "learning_rate": 0.00019215038099506949, + "loss": 0.424, + "step": 28959 + }, + { + "epoch": 1.6180406179288767, + "grad_norm": 0.43226125836372375, + "learning_rate": 0.0001921223666517257, + "loss": 0.4857, + "step": 28960 + }, + { + "epoch": 1.6180964885325586, + "grad_norm": 0.5931415557861328, + "learning_rate": 0.0001920943523083819, + "loss": 0.464, + "step": 28961 + }, + { + "epoch": 1.6181523591362406, + "grad_norm": 0.5045907497406006, + "learning_rate": 0.0001920663379650381, + "loss": 0.6194, + "step": 28962 + }, + { + "epoch": 1.6182082297399223, + "grad_norm": 0.4569312334060669, + "learning_rate": 0.0001920383236216943, + "loss": 0.4456, + "step": 28963 + }, + { + "epoch": 1.6182641003436042, + "grad_norm": 0.5088441371917725, + "learning_rate": 0.00019201030927835054, + "loss": 0.3205, + "step": 28964 + }, + { + "epoch": 1.6183199709472862, + "grad_norm": 0.532366156578064, + "learning_rate": 0.00019198229493500675, + "loss": 0.3853, + "step": 28965 + }, + { + "epoch": 1.618375841550968, + "grad_norm": 0.42866605520248413, + "learning_rate": 0.00019195428059166293, + "loss": 0.4215, + "step": 28966 + }, + { + "epoch": 1.6184317121546499, + "grad_norm": 1.2583063840866089, + "learning_rate": 0.00019192626624831913, + "loss": 0.4065, + "step": 28967 + }, + { + "epoch": 1.6184875827583318, + "grad_norm": 0.4775109589099884, + "learning_rate": 0.00019189825190497534, + "loss": 0.4554, + "step": 28968 + }, + { + "epoch": 1.6185434533620136, + "grad_norm": 0.39643433690071106, + "learning_rate": 0.00019187023756163154, + "loss": 0.3577, + "step": 28969 + }, + { + "epoch": 1.6185993239656955, + "grad_norm": 0.41588670015335083, + "learning_rate": 0.00019184222321828775, + "loss": 0.4201, + "step": 28970 + }, + { + "epoch": 1.6186551945693775, + "grad_norm": 0.42774447798728943, + "learning_rate": 0.00019181420887494396, + "loss": 0.3333, + "step": 28971 + }, + { + "epoch": 1.6187110651730592, + "grad_norm": 2.0556397438049316, + "learning_rate": 0.0001917861945316002, + "loss": 0.3935, + "step": 28972 + }, + { + "epoch": 1.618766935776741, + "grad_norm": 0.3797617554664612, + "learning_rate": 0.0001917581801882564, + "loss": 0.4188, + "step": 28973 + }, + { + "epoch": 1.618822806380423, + "grad_norm": 0.4582855701446533, + "learning_rate": 0.0001917301658449126, + "loss": 0.3753, + "step": 28974 + }, + { + "epoch": 1.6188786769841048, + "grad_norm": 0.4728144407272339, + "learning_rate": 0.0001917021515015688, + "loss": 0.3744, + "step": 28975 + }, + { + "epoch": 1.6189345475877865, + "grad_norm": 1.1835529804229736, + "learning_rate": 0.000191674137158225, + "loss": 0.3586, + "step": 28976 + }, + { + "epoch": 1.6189904181914687, + "grad_norm": 0.4857272803783417, + "learning_rate": 0.00019164612281488122, + "loss": 0.6533, + "step": 28977 + }, + { + "epoch": 1.6190462887951504, + "grad_norm": 0.7025061249732971, + "learning_rate": 0.00019161810847153742, + "loss": 0.3019, + "step": 28978 + }, + { + "epoch": 1.6191021593988322, + "grad_norm": 0.33979469537734985, + "learning_rate": 0.00019159009412819363, + "loss": 0.3323, + "step": 28979 + }, + { + "epoch": 1.6191580300025141, + "grad_norm": 0.470816433429718, + "learning_rate": 0.00019156207978484986, + "loss": 0.4682, + "step": 28980 + }, + { + "epoch": 1.619213900606196, + "grad_norm": 0.389212429523468, + "learning_rate": 0.00019153406544150607, + "loss": 0.4354, + "step": 28981 + }, + { + "epoch": 1.6192697712098778, + "grad_norm": 0.6073698401451111, + "learning_rate": 0.00019150605109816227, + "loss": 0.4783, + "step": 28982 + }, + { + "epoch": 1.6193256418135598, + "grad_norm": 0.5260566473007202, + "learning_rate": 0.00019147803675481848, + "loss": 0.3702, + "step": 28983 + }, + { + "epoch": 1.6193815124172417, + "grad_norm": 0.3860325813293457, + "learning_rate": 0.00019145002241147469, + "loss": 0.406, + "step": 28984 + }, + { + "epoch": 1.6194373830209234, + "grad_norm": 0.770666778087616, + "learning_rate": 0.0001914220080681309, + "loss": 0.3839, + "step": 28985 + }, + { + "epoch": 1.6194932536246054, + "grad_norm": 0.5164949297904968, + "learning_rate": 0.0001913939937247871, + "loss": 0.4708, + "step": 28986 + }, + { + "epoch": 1.6195491242282873, + "grad_norm": 0.3648254871368408, + "learning_rate": 0.00019136597938144328, + "loss": 0.4162, + "step": 28987 + }, + { + "epoch": 1.619604994831969, + "grad_norm": 0.37670594453811646, + "learning_rate": 0.0001913379650380995, + "loss": 0.4659, + "step": 28988 + }, + { + "epoch": 1.619660865435651, + "grad_norm": 0.5360459089279175, + "learning_rate": 0.00019130995069475571, + "loss": 0.4364, + "step": 28989 + }, + { + "epoch": 1.619716736039333, + "grad_norm": 0.4553464949131012, + "learning_rate": 0.00019128193635141192, + "loss": 0.3642, + "step": 28990 + }, + { + "epoch": 1.6197726066430147, + "grad_norm": 0.4362434446811676, + "learning_rate": 0.00019125392200806813, + "loss": 0.4214, + "step": 28991 + }, + { + "epoch": 1.6198284772466967, + "grad_norm": 0.37001916766166687, + "learning_rate": 0.00019122590766472433, + "loss": 0.4058, + "step": 28992 + }, + { + "epoch": 1.6198843478503786, + "grad_norm": 0.5151397585868835, + "learning_rate": 0.00019119789332138054, + "loss": 0.4338, + "step": 28993 + }, + { + "epoch": 1.6199402184540603, + "grad_norm": 1.2276482582092285, + "learning_rate": 0.00019116987897803674, + "loss": 0.422, + "step": 28994 + }, + { + "epoch": 1.6199960890577423, + "grad_norm": 0.7181307673454285, + "learning_rate": 0.00019114186463469298, + "loss": 0.42, + "step": 28995 + }, + { + "epoch": 1.6200519596614242, + "grad_norm": 0.436294823884964, + "learning_rate": 0.00019111385029134918, + "loss": 0.4218, + "step": 28996 + }, + { + "epoch": 1.620107830265106, + "grad_norm": 6.956881999969482, + "learning_rate": 0.0001910858359480054, + "loss": 0.4251, + "step": 28997 + }, + { + "epoch": 1.620163700868788, + "grad_norm": 5.987883567810059, + "learning_rate": 0.0001910578216046616, + "loss": 0.467, + "step": 28998 + }, + { + "epoch": 1.6202195714724699, + "grad_norm": 0.6298765540122986, + "learning_rate": 0.0001910298072613178, + "loss": 0.4086, + "step": 28999 + }, + { + "epoch": 1.6202754420761516, + "grad_norm": 0.3344480097293854, + "learning_rate": 0.000191001792917974, + "loss": 0.3216, + "step": 29000 + }, + { + "epoch": 1.6202754420761516, + "eval_cer": 0.08495630257602042, + "eval_loss": 0.3173025846481323, + "eval_runtime": 56.335, + "eval_samples_per_second": 80.554, + "eval_steps_per_second": 5.041, + "eval_wer": 0.33945979162593226, + "step": 29000 + }, + { + "epoch": 1.6203313126798335, + "grad_norm": 0.5862093567848206, + "learning_rate": 0.0001909737785746302, + "loss": 0.4621, + "step": 29001 + }, + { + "epoch": 1.6203871832835155, + "grad_norm": 0.4187532961368561, + "learning_rate": 0.00019094576423128642, + "loss": 0.3913, + "step": 29002 + }, + { + "epoch": 1.6204430538871972, + "grad_norm": 0.34141266345977783, + "learning_rate": 0.00019091774988794265, + "loss": 0.495, + "step": 29003 + }, + { + "epoch": 1.6204989244908792, + "grad_norm": 0.37226349115371704, + "learning_rate": 0.00019088973554459886, + "loss": 0.4012, + "step": 29004 + }, + { + "epoch": 1.6205547950945611, + "grad_norm": 0.6666069030761719, + "learning_rate": 0.00019086172120125506, + "loss": 0.4955, + "step": 29005 + }, + { + "epoch": 1.6206106656982429, + "grad_norm": 0.6057596206665039, + "learning_rate": 0.00019083370685791127, + "loss": 0.3879, + "step": 29006 + }, + { + "epoch": 1.6206665363019246, + "grad_norm": 0.38734540343284607, + "learning_rate": 0.00019080569251456747, + "loss": 0.4495, + "step": 29007 + }, + { + "epoch": 1.6207224069056068, + "grad_norm": 4.32038688659668, + "learning_rate": 0.00019077767817122368, + "loss": 0.5492, + "step": 29008 + }, + { + "epoch": 1.6207782775092885, + "grad_norm": 0.49478745460510254, + "learning_rate": 0.00019074966382787986, + "loss": 0.4837, + "step": 29009 + }, + { + "epoch": 1.6208341481129702, + "grad_norm": 3.2617764472961426, + "learning_rate": 0.00019072164948453606, + "loss": 0.3312, + "step": 29010 + }, + { + "epoch": 1.6208900187166524, + "grad_norm": 0.8599520921707153, + "learning_rate": 0.0001906936351411923, + "loss": 0.4232, + "step": 29011 + }, + { + "epoch": 1.6209458893203341, + "grad_norm": 1.115667462348938, + "learning_rate": 0.0001906656207978485, + "loss": 0.395, + "step": 29012 + }, + { + "epoch": 1.6210017599240159, + "grad_norm": 0.42538002133369446, + "learning_rate": 0.0001906376064545047, + "loss": 0.5817, + "step": 29013 + }, + { + "epoch": 1.6210576305276978, + "grad_norm": 0.4347620904445648, + "learning_rate": 0.00019060959211116091, + "loss": 0.4738, + "step": 29014 + }, + { + "epoch": 1.6211135011313798, + "grad_norm": 0.3652384579181671, + "learning_rate": 0.00019058157776781712, + "loss": 0.3114, + "step": 29015 + }, + { + "epoch": 1.6211693717350615, + "grad_norm": 0.4073701500892639, + "learning_rate": 0.00019055356342447333, + "loss": 0.5994, + "step": 29016 + }, + { + "epoch": 1.6212252423387434, + "grad_norm": 0.5432225465774536, + "learning_rate": 0.00019052554908112953, + "loss": 0.5142, + "step": 29017 + }, + { + "epoch": 1.6212811129424254, + "grad_norm": 0.43200209736824036, + "learning_rate": 0.00019049753473778574, + "loss": 0.4402, + "step": 29018 + }, + { + "epoch": 1.6213369835461071, + "grad_norm": 0.40233197808265686, + "learning_rate": 0.00019046952039444197, + "loss": 0.2835, + "step": 29019 + }, + { + "epoch": 1.621392854149789, + "grad_norm": 0.7662056088447571, + "learning_rate": 0.00019044150605109818, + "loss": 0.393, + "step": 29020 + }, + { + "epoch": 1.621448724753471, + "grad_norm": 0.3821231424808502, + "learning_rate": 0.00019041349170775438, + "loss": 0.3919, + "step": 29021 + }, + { + "epoch": 1.6215045953571527, + "grad_norm": 0.526553750038147, + "learning_rate": 0.0001903854773644106, + "loss": 0.4629, + "step": 29022 + }, + { + "epoch": 1.6215604659608347, + "grad_norm": 0.46697860956192017, + "learning_rate": 0.0001903574630210668, + "loss": 0.3627, + "step": 29023 + }, + { + "epoch": 1.6216163365645166, + "grad_norm": 0.4348222017288208, + "learning_rate": 0.000190329448677723, + "loss": 0.3893, + "step": 29024 + }, + { + "epoch": 1.6216722071681984, + "grad_norm": 0.770270049571991, + "learning_rate": 0.0001903014343343792, + "loss": 0.3832, + "step": 29025 + }, + { + "epoch": 1.6217280777718803, + "grad_norm": 1.5083975791931152, + "learning_rate": 0.0001902734199910354, + "loss": 0.5042, + "step": 29026 + }, + { + "epoch": 1.6217839483755623, + "grad_norm": 0.32760536670684814, + "learning_rate": 0.00019024540564769164, + "loss": 0.3609, + "step": 29027 + }, + { + "epoch": 1.621839818979244, + "grad_norm": 0.5484350323677063, + "learning_rate": 0.00019021739130434785, + "loss": 0.3825, + "step": 29028 + }, + { + "epoch": 1.621895689582926, + "grad_norm": 1.3325083255767822, + "learning_rate": 0.00019018937696100406, + "loss": 0.371, + "step": 29029 + }, + { + "epoch": 1.621951560186608, + "grad_norm": 0.43472304940223694, + "learning_rate": 0.00019016136261766023, + "loss": 0.3757, + "step": 29030 + }, + { + "epoch": 1.6220074307902896, + "grad_norm": 0.5354563593864441, + "learning_rate": 0.00019013334827431644, + "loss": 0.3944, + "step": 29031 + }, + { + "epoch": 1.6220633013939716, + "grad_norm": 11.27468490600586, + "learning_rate": 0.00019010533393097265, + "loss": 0.4699, + "step": 29032 + }, + { + "epoch": 1.6221191719976535, + "grad_norm": 0.4705570340156555, + "learning_rate": 0.00019007731958762885, + "loss": 0.5482, + "step": 29033 + }, + { + "epoch": 1.6221750426013353, + "grad_norm": 0.3476674258708954, + "learning_rate": 0.00019004930524428506, + "loss": 0.3362, + "step": 29034 + }, + { + "epoch": 1.6222309132050172, + "grad_norm": 2.3470733165740967, + "learning_rate": 0.0001900212909009413, + "loss": 0.4212, + "step": 29035 + }, + { + "epoch": 1.6222867838086992, + "grad_norm": 0.3492170572280884, + "learning_rate": 0.0001899932765575975, + "loss": 0.4939, + "step": 29036 + }, + { + "epoch": 1.622342654412381, + "grad_norm": 1.0857932567596436, + "learning_rate": 0.0001899652622142537, + "loss": 0.372, + "step": 29037 + }, + { + "epoch": 1.6223985250160629, + "grad_norm": 1.1841415166854858, + "learning_rate": 0.0001899372478709099, + "loss": 0.409, + "step": 29038 + }, + { + "epoch": 1.6224543956197448, + "grad_norm": 0.693318247795105, + "learning_rate": 0.00018990923352756611, + "loss": 0.3861, + "step": 29039 + }, + { + "epoch": 1.6225102662234265, + "grad_norm": 0.47589409351348877, + "learning_rate": 0.00018988121918422232, + "loss": 0.3858, + "step": 29040 + }, + { + "epoch": 1.6225661368271083, + "grad_norm": 0.9934449195861816, + "learning_rate": 0.00018985320484087853, + "loss": 0.3926, + "step": 29041 + }, + { + "epoch": 1.6226220074307904, + "grad_norm": 1.1720799207687378, + "learning_rate": 0.00018982519049753473, + "loss": 0.4185, + "step": 29042 + }, + { + "epoch": 1.6226778780344722, + "grad_norm": 0.37640583515167236, + "learning_rate": 0.00018979717615419096, + "loss": 0.5022, + "step": 29043 + }, + { + "epoch": 1.622733748638154, + "grad_norm": 0.4142776131629944, + "learning_rate": 0.00018976916181084717, + "loss": 0.3618, + "step": 29044 + }, + { + "epoch": 1.622789619241836, + "grad_norm": 0.37940630316734314, + "learning_rate": 0.00018974114746750338, + "loss": 0.3762, + "step": 29045 + }, + { + "epoch": 1.6228454898455178, + "grad_norm": 0.7972807288169861, + "learning_rate": 0.00018971313312415958, + "loss": 0.6532, + "step": 29046 + }, + { + "epoch": 1.6229013604491995, + "grad_norm": 0.4559454321861267, + "learning_rate": 0.0001896851187808158, + "loss": 0.38, + "step": 29047 + }, + { + "epoch": 1.6229572310528815, + "grad_norm": 0.3247925341129303, + "learning_rate": 0.000189657104437472, + "loss": 0.3416, + "step": 29048 + }, + { + "epoch": 1.6230131016565634, + "grad_norm": 0.401992529630661, + "learning_rate": 0.0001896290900941282, + "loss": 0.3677, + "step": 29049 + }, + { + "epoch": 1.6230689722602452, + "grad_norm": 0.41555294394493103, + "learning_rate": 0.0001896010757507844, + "loss": 0.4156, + "step": 29050 + }, + { + "epoch": 1.623124842863927, + "grad_norm": 0.27338990569114685, + "learning_rate": 0.0001895730614074406, + "loss": 0.3465, + "step": 29051 + }, + { + "epoch": 1.623180713467609, + "grad_norm": 1.1475551128387451, + "learning_rate": 0.00018954504706409682, + "loss": 0.4193, + "step": 29052 + }, + { + "epoch": 1.6232365840712908, + "grad_norm": 0.4099474549293518, + "learning_rate": 0.00018951703272075302, + "loss": 0.498, + "step": 29053 + }, + { + "epoch": 1.6232924546749727, + "grad_norm": 0.613336443901062, + "learning_rate": 0.00018948901837740923, + "loss": 0.4419, + "step": 29054 + }, + { + "epoch": 1.6233483252786547, + "grad_norm": 0.3706388771533966, + "learning_rate": 0.00018946100403406543, + "loss": 0.3891, + "step": 29055 + }, + { + "epoch": 1.6234041958823364, + "grad_norm": 0.4672187268733978, + "learning_rate": 0.00018943298969072164, + "loss": 0.3892, + "step": 29056 + }, + { + "epoch": 1.6234600664860184, + "grad_norm": 0.36154693365097046, + "learning_rate": 0.00018940497534737785, + "loss": 0.3544, + "step": 29057 + }, + { + "epoch": 1.6235159370897003, + "grad_norm": 0.45953506231307983, + "learning_rate": 0.00018937696100403405, + "loss": 0.4561, + "step": 29058 + }, + { + "epoch": 1.623571807693382, + "grad_norm": 0.41996902227401733, + "learning_rate": 0.00018934894666069028, + "loss": 0.4767, + "step": 29059 + }, + { + "epoch": 1.623627678297064, + "grad_norm": 0.3814115524291992, + "learning_rate": 0.0001893209323173465, + "loss": 0.4042, + "step": 29060 + }, + { + "epoch": 1.623683548900746, + "grad_norm": 0.6099414229393005, + "learning_rate": 0.0001892929179740027, + "loss": 0.4292, + "step": 29061 + }, + { + "epoch": 1.6237394195044277, + "grad_norm": 1.994536280632019, + "learning_rate": 0.0001892649036306589, + "loss": 0.3798, + "step": 29062 + }, + { + "epoch": 1.6237952901081096, + "grad_norm": 0.574708878993988, + "learning_rate": 0.0001892368892873151, + "loss": 0.3413, + "step": 29063 + }, + { + "epoch": 1.6238511607117916, + "grad_norm": 0.6324139833450317, + "learning_rate": 0.00018920887494397131, + "loss": 0.5543, + "step": 29064 + }, + { + "epoch": 1.6239070313154733, + "grad_norm": 0.3866701126098633, + "learning_rate": 0.00018918086060062752, + "loss": 0.515, + "step": 29065 + }, + { + "epoch": 1.6239629019191553, + "grad_norm": 0.4464033544063568, + "learning_rate": 0.00018915284625728375, + "loss": 0.4143, + "step": 29066 + }, + { + "epoch": 1.6240187725228372, + "grad_norm": 0.3427176773548126, + "learning_rate": 0.00018912483191393996, + "loss": 0.4078, + "step": 29067 + }, + { + "epoch": 1.624074643126519, + "grad_norm": 0.8105063438415527, + "learning_rate": 0.00018909681757059616, + "loss": 0.4279, + "step": 29068 + }, + { + "epoch": 1.624130513730201, + "grad_norm": 0.9880606532096863, + "learning_rate": 0.00018906880322725237, + "loss": 0.5882, + "step": 29069 + }, + { + "epoch": 1.6241863843338828, + "grad_norm": 0.5149582624435425, + "learning_rate": 0.00018904078888390858, + "loss": 0.4407, + "step": 29070 + }, + { + "epoch": 1.6242422549375646, + "grad_norm": 0.5854864716529846, + "learning_rate": 0.00018901277454056478, + "loss": 0.3862, + "step": 29071 + }, + { + "epoch": 1.6242981255412463, + "grad_norm": 3.0507259368896484, + "learning_rate": 0.00018898476019722096, + "loss": 0.421, + "step": 29072 + }, + { + "epoch": 1.6243539961449285, + "grad_norm": 0.5647808909416199, + "learning_rate": 0.00018895674585387717, + "loss": 0.4198, + "step": 29073 + }, + { + "epoch": 1.6244098667486102, + "grad_norm": 0.30756986141204834, + "learning_rate": 0.0001889287315105334, + "loss": 0.3286, + "step": 29074 + }, + { + "epoch": 1.624465737352292, + "grad_norm": 0.35470321774482727, + "learning_rate": 0.0001889007171671896, + "loss": 0.4232, + "step": 29075 + }, + { + "epoch": 1.6245216079559741, + "grad_norm": 0.9073907136917114, + "learning_rate": 0.0001888727028238458, + "loss": 0.3839, + "step": 29076 + }, + { + "epoch": 1.6245774785596558, + "grad_norm": 0.4020170271396637, + "learning_rate": 0.00018884468848050202, + "loss": 0.5269, + "step": 29077 + }, + { + "epoch": 1.6246333491633376, + "grad_norm": 0.497201144695282, + "learning_rate": 0.00018881667413715822, + "loss": 0.4864, + "step": 29078 + }, + { + "epoch": 1.6246892197670197, + "grad_norm": 0.533413290977478, + "learning_rate": 0.00018878865979381443, + "loss": 0.3843, + "step": 29079 + }, + { + "epoch": 1.6247450903707015, + "grad_norm": 0.336916446685791, + "learning_rate": 0.00018876064545047063, + "loss": 0.3955, + "step": 29080 + }, + { + "epoch": 1.6248009609743832, + "grad_norm": 0.3718840479850769, + "learning_rate": 0.00018873263110712684, + "loss": 0.3886, + "step": 29081 + }, + { + "epoch": 1.6248568315780652, + "grad_norm": 0.4377793073654175, + "learning_rate": 0.00018870461676378307, + "loss": 0.3222, + "step": 29082 + }, + { + "epoch": 1.624912702181747, + "grad_norm": 0.4197322726249695, + "learning_rate": 0.00018867660242043928, + "loss": 0.5247, + "step": 29083 + }, + { + "epoch": 1.6249685727854288, + "grad_norm": 0.4377998411655426, + "learning_rate": 0.00018864858807709548, + "loss": 0.3406, + "step": 29084 + }, + { + "epoch": 1.6250244433891108, + "grad_norm": 0.4510357081890106, + "learning_rate": 0.0001886205737337517, + "loss": 0.3096, + "step": 29085 + }, + { + "epoch": 1.6250803139927927, + "grad_norm": 0.5091392397880554, + "learning_rate": 0.0001885925593904079, + "loss": 0.3993, + "step": 29086 + }, + { + "epoch": 1.6251361845964745, + "grad_norm": 0.30447709560394287, + "learning_rate": 0.0001885645450470641, + "loss": 0.4199, + "step": 29087 + }, + { + "epoch": 1.6251920552001564, + "grad_norm": 1.6939022541046143, + "learning_rate": 0.0001885365307037203, + "loss": 0.4409, + "step": 29088 + }, + { + "epoch": 1.6252479258038384, + "grad_norm": 0.7451399564743042, + "learning_rate": 0.0001885085163603765, + "loss": 0.6896, + "step": 29089 + }, + { + "epoch": 1.62530379640752, + "grad_norm": 0.36739498376846313, + "learning_rate": 0.00018848050201703275, + "loss": 0.356, + "step": 29090 + }, + { + "epoch": 1.625359667011202, + "grad_norm": 0.3722396790981293, + "learning_rate": 0.00018845248767368895, + "loss": 0.4008, + "step": 29091 + }, + { + "epoch": 1.625415537614884, + "grad_norm": 0.448074072599411, + "learning_rate": 0.00018842447333034516, + "loss": 0.5012, + "step": 29092 + }, + { + "epoch": 1.6254714082185657, + "grad_norm": 0.43938130140304565, + "learning_rate": 0.00018839645898700136, + "loss": 0.4023, + "step": 29093 + }, + { + "epoch": 1.6255272788222477, + "grad_norm": 0.3584452271461487, + "learning_rate": 0.00018836844464365754, + "loss": 0.3631, + "step": 29094 + }, + { + "epoch": 1.6255831494259296, + "grad_norm": 3.145817995071411, + "learning_rate": 0.00018834043030031375, + "loss": 0.3783, + "step": 29095 + }, + { + "epoch": 1.6256390200296114, + "grad_norm": 0.5375767350196838, + "learning_rate": 0.00018831241595696995, + "loss": 0.5244, + "step": 29096 + }, + { + "epoch": 1.6256948906332933, + "grad_norm": 0.41902562975883484, + "learning_rate": 0.00018828440161362616, + "loss": 0.4062, + "step": 29097 + }, + { + "epoch": 1.6257507612369753, + "grad_norm": 3.2113170623779297, + "learning_rate": 0.0001882563872702824, + "loss": 0.394, + "step": 29098 + }, + { + "epoch": 1.625806631840657, + "grad_norm": 0.5660697221755981, + "learning_rate": 0.0001882283729269386, + "loss": 0.4457, + "step": 29099 + }, + { + "epoch": 1.625862502444339, + "grad_norm": 1.538931131362915, + "learning_rate": 0.0001882003585835948, + "loss": 0.5902, + "step": 29100 + }, + { + "epoch": 1.625918373048021, + "grad_norm": 0.3221418261528015, + "learning_rate": 0.000188172344240251, + "loss": 0.3918, + "step": 29101 + }, + { + "epoch": 1.6259742436517026, + "grad_norm": 0.3567037284374237, + "learning_rate": 0.00018814432989690722, + "loss": 0.3765, + "step": 29102 + }, + { + "epoch": 1.6260301142553846, + "grad_norm": 0.4017467796802521, + "learning_rate": 0.00018811631555356342, + "loss": 0.5091, + "step": 29103 + }, + { + "epoch": 1.6260859848590665, + "grad_norm": 0.4483494758605957, + "learning_rate": 0.00018808830121021963, + "loss": 0.3305, + "step": 29104 + }, + { + "epoch": 1.6261418554627483, + "grad_norm": 0.5041773915290833, + "learning_rate": 0.00018806028686687583, + "loss": 0.4416, + "step": 29105 + }, + { + "epoch": 1.62619772606643, + "grad_norm": 1.090067982673645, + "learning_rate": 0.00018803227252353207, + "loss": 0.4472, + "step": 29106 + }, + { + "epoch": 1.6262535966701122, + "grad_norm": 0.4686819612979889, + "learning_rate": 0.00018800425818018827, + "loss": 0.4506, + "step": 29107 + }, + { + "epoch": 1.6263094672737939, + "grad_norm": 0.36258813738822937, + "learning_rate": 0.00018797624383684448, + "loss": 0.3455, + "step": 29108 + }, + { + "epoch": 1.6263653378774756, + "grad_norm": 2.8194022178649902, + "learning_rate": 0.00018794822949350068, + "loss": 0.5115, + "step": 29109 + }, + { + "epoch": 1.6264212084811578, + "grad_norm": 0.5728548765182495, + "learning_rate": 0.0001879202151501569, + "loss": 0.3668, + "step": 29110 + }, + { + "epoch": 1.6264770790848395, + "grad_norm": 0.7645208835601807, + "learning_rate": 0.0001878922008068131, + "loss": 0.5084, + "step": 29111 + }, + { + "epoch": 1.6265329496885212, + "grad_norm": 4.249688148498535, + "learning_rate": 0.0001878641864634693, + "loss": 0.3221, + "step": 29112 + }, + { + "epoch": 1.6265888202922034, + "grad_norm": 1.5899676084518433, + "learning_rate": 0.0001878361721201255, + "loss": 0.3624, + "step": 29113 + }, + { + "epoch": 1.6266446908958851, + "grad_norm": 1.21326744556427, + "learning_rate": 0.00018780815777678174, + "loss": 0.4477, + "step": 29114 + }, + { + "epoch": 1.6267005614995669, + "grad_norm": 0.48921626806259155, + "learning_rate": 0.00018778014343343792, + "loss": 0.4751, + "step": 29115 + }, + { + "epoch": 1.6267564321032488, + "grad_norm": 0.3591156005859375, + "learning_rate": 0.00018775212909009412, + "loss": 0.3705, + "step": 29116 + }, + { + "epoch": 1.6268123027069308, + "grad_norm": 1.740445852279663, + "learning_rate": 0.00018772411474675033, + "loss": 0.4774, + "step": 29117 + }, + { + "epoch": 1.6268681733106125, + "grad_norm": 0.39493051171302795, + "learning_rate": 0.00018769610040340654, + "loss": 0.3817, + "step": 29118 + }, + { + "epoch": 1.6269240439142945, + "grad_norm": 0.3751620948314667, + "learning_rate": 0.00018766808606006274, + "loss": 0.4428, + "step": 29119 + }, + { + "epoch": 1.6269799145179764, + "grad_norm": 0.8376654982566833, + "learning_rate": 0.00018764007171671895, + "loss": 0.3754, + "step": 29120 + }, + { + "epoch": 1.6270357851216581, + "grad_norm": 2.1397719383239746, + "learning_rate": 0.00018761205737337515, + "loss": 0.4253, + "step": 29121 + }, + { + "epoch": 1.62709165572534, + "grad_norm": 0.3143770098686218, + "learning_rate": 0.0001875840430300314, + "loss": 0.411, + "step": 29122 + }, + { + "epoch": 1.627147526329022, + "grad_norm": 0.49419060349464417, + "learning_rate": 0.0001875560286866876, + "loss": 0.4863, + "step": 29123 + }, + { + "epoch": 1.6272033969327038, + "grad_norm": 0.48055562376976013, + "learning_rate": 0.0001875280143433438, + "loss": 0.4772, + "step": 29124 + }, + { + "epoch": 1.6272592675363857, + "grad_norm": 0.3801068961620331, + "learning_rate": 0.0001875, + "loss": 0.4551, + "step": 29125 + }, + { + "epoch": 1.6273151381400677, + "grad_norm": 0.3562489151954651, + "learning_rate": 0.0001874719856566562, + "loss": 0.3399, + "step": 29126 + }, + { + "epoch": 1.6273710087437494, + "grad_norm": 1.089059829711914, + "learning_rate": 0.00018744397131331242, + "loss": 0.4857, + "step": 29127 + }, + { + "epoch": 1.6274268793474314, + "grad_norm": 0.4481457769870758, + "learning_rate": 0.00018741595696996862, + "loss": 0.3343, + "step": 29128 + }, + { + "epoch": 1.6274827499511133, + "grad_norm": 0.3144592046737671, + "learning_rate": 0.00018738794262662485, + "loss": 0.3677, + "step": 29129 + }, + { + "epoch": 1.627538620554795, + "grad_norm": 0.4143497049808502, + "learning_rate": 0.00018735992828328106, + "loss": 0.373, + "step": 29130 + }, + { + "epoch": 1.627594491158477, + "grad_norm": 0.4165688455104828, + "learning_rate": 0.00018733191393993727, + "loss": 0.5046, + "step": 29131 + }, + { + "epoch": 1.627650361762159, + "grad_norm": 0.5516923666000366, + "learning_rate": 0.00018730389959659347, + "loss": 0.4296, + "step": 29132 + }, + { + "epoch": 1.6277062323658407, + "grad_norm": 0.4630761742591858, + "learning_rate": 0.00018727588525324968, + "loss": 0.4567, + "step": 29133 + }, + { + "epoch": 1.6277621029695226, + "grad_norm": 0.3791816532611847, + "learning_rate": 0.00018724787090990588, + "loss": 0.3222, + "step": 29134 + }, + { + "epoch": 1.6278179735732046, + "grad_norm": 0.4517882168292999, + "learning_rate": 0.0001872198565665621, + "loss": 0.4953, + "step": 29135 + }, + { + "epoch": 1.6278738441768863, + "grad_norm": 0.2805659770965576, + "learning_rate": 0.00018719184222321827, + "loss": 0.332, + "step": 29136 + }, + { + "epoch": 1.6279297147805682, + "grad_norm": 0.4096216559410095, + "learning_rate": 0.0001871638278798745, + "loss": 0.4961, + "step": 29137 + }, + { + "epoch": 1.6279855853842502, + "grad_norm": 1.7119383811950684, + "learning_rate": 0.0001871358135365307, + "loss": 0.4027, + "step": 29138 + }, + { + "epoch": 1.628041455987932, + "grad_norm": 0.39507806301116943, + "learning_rate": 0.0001871077991931869, + "loss": 0.3915, + "step": 29139 + }, + { + "epoch": 1.6280973265916137, + "grad_norm": 0.5469672083854675, + "learning_rate": 0.00018707978484984312, + "loss": 0.3181, + "step": 29140 + }, + { + "epoch": 1.6281531971952958, + "grad_norm": 0.7469016909599304, + "learning_rate": 0.00018705177050649932, + "loss": 0.4918, + "step": 29141 + }, + { + "epoch": 1.6282090677989776, + "grad_norm": 0.485338419675827, + "learning_rate": 0.00018702375616315553, + "loss": 0.3805, + "step": 29142 + }, + { + "epoch": 1.6282649384026593, + "grad_norm": 0.4123501181602478, + "learning_rate": 0.00018699574181981174, + "loss": 0.4086, + "step": 29143 + }, + { + "epoch": 1.6283208090063415, + "grad_norm": 2.331000804901123, + "learning_rate": 0.00018696772747646794, + "loss": 0.4935, + "step": 29144 + }, + { + "epoch": 1.6283766796100232, + "grad_norm": 1.1814826726913452, + "learning_rate": 0.00018693971313312417, + "loss": 0.4454, + "step": 29145 + }, + { + "epoch": 1.628432550213705, + "grad_norm": 0.3842512369155884, + "learning_rate": 0.00018691169878978038, + "loss": 0.3894, + "step": 29146 + }, + { + "epoch": 1.628488420817387, + "grad_norm": 1.0463753938674927, + "learning_rate": 0.00018688368444643659, + "loss": 0.4559, + "step": 29147 + }, + { + "epoch": 1.6285442914210688, + "grad_norm": 0.571385383605957, + "learning_rate": 0.0001868556701030928, + "loss": 0.4517, + "step": 29148 + }, + { + "epoch": 1.6286001620247506, + "grad_norm": 0.3154419958591461, + "learning_rate": 0.000186827655759749, + "loss": 0.3933, + "step": 29149 + }, + { + "epoch": 1.6286560326284325, + "grad_norm": 0.4480230212211609, + "learning_rate": 0.0001867996414164052, + "loss": 0.413, + "step": 29150 + }, + { + "epoch": 1.6287119032321145, + "grad_norm": 1.8182703256607056, + "learning_rate": 0.0001867716270730614, + "loss": 0.3381, + "step": 29151 + }, + { + "epoch": 1.6287677738357962, + "grad_norm": 0.3597646951675415, + "learning_rate": 0.00018674361272971762, + "loss": 0.4697, + "step": 29152 + }, + { + "epoch": 1.6288236444394781, + "grad_norm": 0.46572667360305786, + "learning_rate": 0.00018671559838637385, + "loss": 0.3984, + "step": 29153 + }, + { + "epoch": 1.62887951504316, + "grad_norm": 3.682177782058716, + "learning_rate": 0.00018668758404303005, + "loss": 0.4455, + "step": 29154 + }, + { + "epoch": 1.6289353856468418, + "grad_norm": 0.40270692110061646, + "learning_rate": 0.00018665956969968626, + "loss": 0.4028, + "step": 29155 + }, + { + "epoch": 1.6289912562505238, + "grad_norm": 0.3367566764354706, + "learning_rate": 0.00018663155535634247, + "loss": 0.3484, + "step": 29156 + }, + { + "epoch": 1.6290471268542057, + "grad_norm": 0.5075157284736633, + "learning_rate": 0.00018660354101299864, + "loss": 0.4339, + "step": 29157 + }, + { + "epoch": 1.6291029974578874, + "grad_norm": 0.3882924020290375, + "learning_rate": 0.00018657552666965485, + "loss": 0.3016, + "step": 29158 + }, + { + "epoch": 1.6291588680615694, + "grad_norm": 1.0526796579360962, + "learning_rate": 0.00018654751232631106, + "loss": 0.4465, + "step": 29159 + }, + { + "epoch": 1.6292147386652513, + "grad_norm": 0.3864953815937042, + "learning_rate": 0.00018651949798296726, + "loss": 0.5053, + "step": 29160 + }, + { + "epoch": 1.629270609268933, + "grad_norm": 0.34376418590545654, + "learning_rate": 0.0001864914836396235, + "loss": 0.3842, + "step": 29161 + }, + { + "epoch": 1.629326479872615, + "grad_norm": 1.3977315425872803, + "learning_rate": 0.0001864634692962797, + "loss": 0.4985, + "step": 29162 + }, + { + "epoch": 1.629382350476297, + "grad_norm": 1.5405322313308716, + "learning_rate": 0.0001864354549529359, + "loss": 0.3878, + "step": 29163 + }, + { + "epoch": 1.6294382210799787, + "grad_norm": 0.8543402552604675, + "learning_rate": 0.0001864074406095921, + "loss": 0.6321, + "step": 29164 + }, + { + "epoch": 1.6294940916836607, + "grad_norm": 0.36266380548477173, + "learning_rate": 0.00018637942626624832, + "loss": 0.4498, + "step": 29165 + }, + { + "epoch": 1.6295499622873426, + "grad_norm": 1.6821208000183105, + "learning_rate": 0.00018635141192290452, + "loss": 0.3689, + "step": 29166 + }, + { + "epoch": 1.6296058328910243, + "grad_norm": 0.45794349908828735, + "learning_rate": 0.00018632339757956073, + "loss": 0.4702, + "step": 29167 + }, + { + "epoch": 1.6296617034947063, + "grad_norm": 0.3970561921596527, + "learning_rate": 0.00018629538323621694, + "loss": 0.3773, + "step": 29168 + }, + { + "epoch": 1.6297175740983882, + "grad_norm": 0.3978491425514221, + "learning_rate": 0.00018626736889287317, + "loss": 0.3679, + "step": 29169 + }, + { + "epoch": 1.62977344470207, + "grad_norm": 0.3567487895488739, + "learning_rate": 0.00018623935454952937, + "loss": 0.3331, + "step": 29170 + }, + { + "epoch": 1.629829315305752, + "grad_norm": 0.8144828081130981, + "learning_rate": 0.00018621134020618558, + "loss": 0.3439, + "step": 29171 + }, + { + "epoch": 1.6298851859094339, + "grad_norm": 0.5398668646812439, + "learning_rate": 0.00018618332586284179, + "loss": 0.3438, + "step": 29172 + }, + { + "epoch": 1.6299410565131156, + "grad_norm": 0.5956440567970276, + "learning_rate": 0.000186155311519498, + "loss": 0.365, + "step": 29173 + }, + { + "epoch": 1.6299969271167973, + "grad_norm": 0.33547723293304443, + "learning_rate": 0.0001861272971761542, + "loss": 0.3802, + "step": 29174 + }, + { + "epoch": 1.6300527977204795, + "grad_norm": 0.40047889947891235, + "learning_rate": 0.0001860992828328104, + "loss": 0.3985, + "step": 29175 + }, + { + "epoch": 1.6301086683241612, + "grad_norm": 0.3630659580230713, + "learning_rate": 0.0001860712684894666, + "loss": 0.4115, + "step": 29176 + }, + { + "epoch": 1.630164538927843, + "grad_norm": 0.35684677958488464, + "learning_rate": 0.00018604325414612284, + "loss": 0.3571, + "step": 29177 + }, + { + "epoch": 1.6302204095315251, + "grad_norm": 0.2935035824775696, + "learning_rate": 0.00018601523980277905, + "loss": 0.3799, + "step": 29178 + }, + { + "epoch": 1.6302762801352069, + "grad_norm": 0.5815011858940125, + "learning_rate": 0.00018598722545943523, + "loss": 0.4188, + "step": 29179 + }, + { + "epoch": 1.6303321507388886, + "grad_norm": 0.5131670832633972, + "learning_rate": 0.00018595921111609143, + "loss": 0.3605, + "step": 29180 + }, + { + "epoch": 1.6303880213425708, + "grad_norm": 0.46436333656311035, + "learning_rate": 0.00018593119677274764, + "loss": 0.4227, + "step": 29181 + }, + { + "epoch": 1.6304438919462525, + "grad_norm": 0.3833850026130676, + "learning_rate": 0.00018590318242940384, + "loss": 0.4036, + "step": 29182 + }, + { + "epoch": 1.6304997625499342, + "grad_norm": 0.5795506834983826, + "learning_rate": 0.00018587516808606005, + "loss": 0.3934, + "step": 29183 + }, + { + "epoch": 1.6305556331536162, + "grad_norm": 0.8751184344291687, + "learning_rate": 0.00018584715374271626, + "loss": 0.3779, + "step": 29184 + }, + { + "epoch": 1.6306115037572981, + "grad_norm": 2.3815085887908936, + "learning_rate": 0.0001858191393993725, + "loss": 0.3761, + "step": 29185 + }, + { + "epoch": 1.6306673743609799, + "grad_norm": 0.46388542652130127, + "learning_rate": 0.0001857911250560287, + "loss": 0.4166, + "step": 29186 + }, + { + "epoch": 1.6307232449646618, + "grad_norm": 0.3351474106311798, + "learning_rate": 0.0001857631107126849, + "loss": 0.4223, + "step": 29187 + }, + { + "epoch": 1.6307791155683438, + "grad_norm": 0.47510820627212524, + "learning_rate": 0.0001857350963693411, + "loss": 0.3513, + "step": 29188 + }, + { + "epoch": 1.6308349861720255, + "grad_norm": 0.4621051847934723, + "learning_rate": 0.0001857070820259973, + "loss": 0.4145, + "step": 29189 + }, + { + "epoch": 1.6308908567757074, + "grad_norm": 0.5859742164611816, + "learning_rate": 0.00018567906768265352, + "loss": 0.3647, + "step": 29190 + }, + { + "epoch": 1.6309467273793894, + "grad_norm": 0.4333902597427368, + "learning_rate": 0.00018565105333930972, + "loss": 0.3671, + "step": 29191 + }, + { + "epoch": 1.6310025979830711, + "grad_norm": 0.530207097530365, + "learning_rate": 0.00018562303899596596, + "loss": 0.3764, + "step": 29192 + }, + { + "epoch": 1.631058468586753, + "grad_norm": 0.611619770526886, + "learning_rate": 0.00018559502465262216, + "loss": 0.3231, + "step": 29193 + }, + { + "epoch": 1.631114339190435, + "grad_norm": 0.622769296169281, + "learning_rate": 0.00018556701030927837, + "loss": 0.3048, + "step": 29194 + }, + { + "epoch": 1.6311702097941168, + "grad_norm": 0.8167303800582886, + "learning_rate": 0.00018553899596593457, + "loss": 0.4826, + "step": 29195 + }, + { + "epoch": 1.6312260803977987, + "grad_norm": 0.48709428310394287, + "learning_rate": 0.00018551098162259078, + "loss": 0.4007, + "step": 29196 + }, + { + "epoch": 1.6312819510014807, + "grad_norm": 0.29478511214256287, + "learning_rate": 0.00018548296727924699, + "loss": 0.3253, + "step": 29197 + }, + { + "epoch": 1.6313378216051624, + "grad_norm": 0.5046358108520508, + "learning_rate": 0.0001854549529359032, + "loss": 0.4375, + "step": 29198 + }, + { + "epoch": 1.6313936922088443, + "grad_norm": 0.34884706139564514, + "learning_rate": 0.0001854269385925594, + "loss": 0.3884, + "step": 29199 + }, + { + "epoch": 1.6314495628125263, + "grad_norm": 0.3672378957271576, + "learning_rate": 0.0001853989242492156, + "loss": 0.4259, + "step": 29200 + }, + { + "epoch": 1.631505433416208, + "grad_norm": 0.49024975299835205, + "learning_rate": 0.0001853709099058718, + "loss": 0.3877, + "step": 29201 + }, + { + "epoch": 1.63156130401989, + "grad_norm": 0.4428389072418213, + "learning_rate": 0.00018534289556252801, + "loss": 0.3705, + "step": 29202 + }, + { + "epoch": 1.631617174623572, + "grad_norm": 0.3306107521057129, + "learning_rate": 0.00018531488121918422, + "loss": 0.3697, + "step": 29203 + }, + { + "epoch": 1.6316730452272536, + "grad_norm": 0.34503409266471863, + "learning_rate": 0.00018528686687584043, + "loss": 0.4424, + "step": 29204 + }, + { + "epoch": 1.6317289158309356, + "grad_norm": 0.37711238861083984, + "learning_rate": 0.00018525885253249663, + "loss": 0.3449, + "step": 29205 + }, + { + "epoch": 1.6317847864346176, + "grad_norm": 0.30938485264778137, + "learning_rate": 0.00018523083818915284, + "loss": 0.3355, + "step": 29206 + }, + { + "epoch": 1.6318406570382993, + "grad_norm": 2.2270593643188477, + "learning_rate": 0.00018520282384580904, + "loss": 0.3449, + "step": 29207 + }, + { + "epoch": 1.631896527641981, + "grad_norm": 0.734830379486084, + "learning_rate": 0.00018517480950246528, + "loss": 0.4101, + "step": 29208 + }, + { + "epoch": 1.6319523982456632, + "grad_norm": 0.5324979424476624, + "learning_rate": 0.00018514679515912148, + "loss": 0.4497, + "step": 29209 + }, + { + "epoch": 1.632008268849345, + "grad_norm": 0.415814071893692, + "learning_rate": 0.0001851187808157777, + "loss": 0.4128, + "step": 29210 + }, + { + "epoch": 1.6320641394530266, + "grad_norm": 0.3530224561691284, + "learning_rate": 0.0001850907664724339, + "loss": 0.3791, + "step": 29211 + }, + { + "epoch": 1.6321200100567088, + "grad_norm": 0.40856608748435974, + "learning_rate": 0.0001850627521290901, + "loss": 0.4147, + "step": 29212 + }, + { + "epoch": 1.6321758806603905, + "grad_norm": 0.31545501947402954, + "learning_rate": 0.0001850347377857463, + "loss": 0.3706, + "step": 29213 + }, + { + "epoch": 1.6322317512640723, + "grad_norm": 0.3422616720199585, + "learning_rate": 0.0001850067234424025, + "loss": 0.3426, + "step": 29214 + }, + { + "epoch": 1.6322876218677544, + "grad_norm": 3.184561252593994, + "learning_rate": 0.00018497870909905872, + "loss": 0.4572, + "step": 29215 + }, + { + "epoch": 1.6323434924714362, + "grad_norm": 0.41577744483947754, + "learning_rate": 0.00018495069475571495, + "loss": 0.4036, + "step": 29216 + }, + { + "epoch": 1.632399363075118, + "grad_norm": 0.5036173462867737, + "learning_rate": 0.00018492268041237116, + "loss": 0.5847, + "step": 29217 + }, + { + "epoch": 1.6324552336787999, + "grad_norm": 0.4565834701061249, + "learning_rate": 0.00018489466606902736, + "loss": 0.5421, + "step": 29218 + }, + { + "epoch": 1.6325111042824818, + "grad_norm": 2.433122158050537, + "learning_rate": 0.00018486665172568357, + "loss": 0.3665, + "step": 29219 + }, + { + "epoch": 1.6325669748861635, + "grad_norm": 0.40151044726371765, + "learning_rate": 0.00018483863738233977, + "loss": 0.4371, + "step": 29220 + }, + { + "epoch": 1.6326228454898455, + "grad_norm": 0.607201337814331, + "learning_rate": 0.00018481062303899595, + "loss": 0.4126, + "step": 29221 + }, + { + "epoch": 1.6326787160935274, + "grad_norm": 0.4270040690898895, + "learning_rate": 0.00018478260869565216, + "loss": 0.3868, + "step": 29222 + }, + { + "epoch": 1.6327345866972092, + "grad_norm": 0.559946596622467, + "learning_rate": 0.00018475459435230836, + "loss": 0.3833, + "step": 29223 + }, + { + "epoch": 1.6327904573008911, + "grad_norm": 0.4090614318847656, + "learning_rate": 0.0001847265800089646, + "loss": 0.3671, + "step": 29224 + }, + { + "epoch": 1.632846327904573, + "grad_norm": 0.499477356672287, + "learning_rate": 0.0001846985656656208, + "loss": 0.373, + "step": 29225 + }, + { + "epoch": 1.6329021985082548, + "grad_norm": 0.34790605306625366, + "learning_rate": 0.000184670551322277, + "loss": 0.4073, + "step": 29226 + }, + { + "epoch": 1.6329580691119367, + "grad_norm": 1.0183199644088745, + "learning_rate": 0.00018464253697893321, + "loss": 0.7106, + "step": 29227 + }, + { + "epoch": 1.6330139397156187, + "grad_norm": 0.6618139743804932, + "learning_rate": 0.00018461452263558942, + "loss": 0.4437, + "step": 29228 + }, + { + "epoch": 1.6330698103193004, + "grad_norm": 0.3497260510921478, + "learning_rate": 0.00018458650829224563, + "loss": 0.3335, + "step": 29229 + }, + { + "epoch": 1.6331256809229824, + "grad_norm": 0.4358561635017395, + "learning_rate": 0.00018455849394890183, + "loss": 0.5019, + "step": 29230 + }, + { + "epoch": 1.6331815515266643, + "grad_norm": 0.3786011040210724, + "learning_rate": 0.00018453047960555804, + "loss": 0.4372, + "step": 29231 + }, + { + "epoch": 1.633237422130346, + "grad_norm": 0.43879348039627075, + "learning_rate": 0.00018450246526221427, + "loss": 0.4089, + "step": 29232 + }, + { + "epoch": 1.633293292734028, + "grad_norm": 0.7118018269538879, + "learning_rate": 0.00018447445091887048, + "loss": 0.5027, + "step": 29233 + }, + { + "epoch": 1.63334916333771, + "grad_norm": 0.5894063115119934, + "learning_rate": 0.00018444643657552668, + "loss": 0.4282, + "step": 29234 + }, + { + "epoch": 1.6334050339413917, + "grad_norm": 0.31348809599876404, + "learning_rate": 0.0001844184222321829, + "loss": 0.3985, + "step": 29235 + }, + { + "epoch": 1.6334609045450736, + "grad_norm": 0.37185120582580566, + "learning_rate": 0.0001843904078888391, + "loss": 0.4108, + "step": 29236 + }, + { + "epoch": 1.6335167751487556, + "grad_norm": 0.43564724922180176, + "learning_rate": 0.0001843623935454953, + "loss": 0.3521, + "step": 29237 + }, + { + "epoch": 1.6335726457524373, + "grad_norm": 0.40893688797950745, + "learning_rate": 0.0001843343792021515, + "loss": 0.4183, + "step": 29238 + }, + { + "epoch": 1.6336285163561193, + "grad_norm": 0.4843941032886505, + "learning_rate": 0.0001843063648588077, + "loss": 0.3779, + "step": 29239 + }, + { + "epoch": 1.6336843869598012, + "grad_norm": 0.5074400901794434, + "learning_rate": 0.00018427835051546394, + "loss": 0.434, + "step": 29240 + }, + { + "epoch": 1.633740257563483, + "grad_norm": 0.3398393988609314, + "learning_rate": 0.00018425033617212015, + "loss": 0.3942, + "step": 29241 + }, + { + "epoch": 1.6337961281671647, + "grad_norm": 0.4717387557029724, + "learning_rate": 0.00018422232182877633, + "loss": 0.4937, + "step": 29242 + }, + { + "epoch": 1.6338519987708469, + "grad_norm": 1.7406190633773804, + "learning_rate": 0.00018419430748543253, + "loss": 0.3096, + "step": 29243 + }, + { + "epoch": 1.6339078693745286, + "grad_norm": 0.3402120769023895, + "learning_rate": 0.00018416629314208874, + "loss": 0.4102, + "step": 29244 + }, + { + "epoch": 1.6339637399782103, + "grad_norm": 0.49881744384765625, + "learning_rate": 0.00018413827879874495, + "loss": 0.4589, + "step": 29245 + }, + { + "epoch": 1.6340196105818925, + "grad_norm": 1.2168529033660889, + "learning_rate": 0.00018411026445540115, + "loss": 0.3672, + "step": 29246 + }, + { + "epoch": 1.6340754811855742, + "grad_norm": 0.322975754737854, + "learning_rate": 0.00018408225011205736, + "loss": 0.4128, + "step": 29247 + }, + { + "epoch": 1.634131351789256, + "grad_norm": 0.3541584312915802, + "learning_rate": 0.0001840542357687136, + "loss": 0.4663, + "step": 29248 + }, + { + "epoch": 1.6341872223929381, + "grad_norm": 0.49382495880126953, + "learning_rate": 0.0001840262214253698, + "loss": 0.5818, + "step": 29249 + }, + { + "epoch": 1.6342430929966199, + "grad_norm": 0.7170819640159607, + "learning_rate": 0.000183998207082026, + "loss": 0.3871, + "step": 29250 + }, + { + "epoch": 1.6342989636003016, + "grad_norm": 0.54529869556427, + "learning_rate": 0.0001839701927386822, + "loss": 0.4106, + "step": 29251 + }, + { + "epoch": 1.6343548342039835, + "grad_norm": 1.314254641532898, + "learning_rate": 0.0001839421783953384, + "loss": 0.4585, + "step": 29252 + }, + { + "epoch": 1.6344107048076655, + "grad_norm": 0.3377823531627655, + "learning_rate": 0.00018391416405199462, + "loss": 0.3194, + "step": 29253 + }, + { + "epoch": 1.6344665754113472, + "grad_norm": 0.3551650941371918, + "learning_rate": 0.00018388614970865083, + "loss": 0.3762, + "step": 29254 + }, + { + "epoch": 1.6345224460150292, + "grad_norm": 0.5924152135848999, + "learning_rate": 0.00018385813536530703, + "loss": 0.4689, + "step": 29255 + }, + { + "epoch": 1.6345783166187111, + "grad_norm": 0.371899276971817, + "learning_rate": 0.00018383012102196326, + "loss": 0.4632, + "step": 29256 + }, + { + "epoch": 1.6346341872223928, + "grad_norm": 0.45119333267211914, + "learning_rate": 0.00018380210667861947, + "loss": 0.3219, + "step": 29257 + }, + { + "epoch": 1.6346900578260748, + "grad_norm": 1.1972532272338867, + "learning_rate": 0.00018377409233527568, + "loss": 0.4943, + "step": 29258 + }, + { + "epoch": 1.6347459284297567, + "grad_norm": 0.3719431459903717, + "learning_rate": 0.00018374607799193188, + "loss": 0.3245, + "step": 29259 + }, + { + "epoch": 1.6348017990334385, + "grad_norm": 0.3531859815120697, + "learning_rate": 0.0001837180636485881, + "loss": 0.3023, + "step": 29260 + }, + { + "epoch": 1.6348576696371204, + "grad_norm": 0.4970380663871765, + "learning_rate": 0.0001836900493052443, + "loss": 0.5762, + "step": 29261 + }, + { + "epoch": 1.6349135402408024, + "grad_norm": 0.4517674446105957, + "learning_rate": 0.0001836620349619005, + "loss": 0.4035, + "step": 29262 + }, + { + "epoch": 1.634969410844484, + "grad_norm": 1.9035968780517578, + "learning_rate": 0.00018363402061855673, + "loss": 0.423, + "step": 29263 + }, + { + "epoch": 1.635025281448166, + "grad_norm": 1.099867343902588, + "learning_rate": 0.0001836060062752129, + "loss": 0.3586, + "step": 29264 + }, + { + "epoch": 1.635081152051848, + "grad_norm": 0.5966991782188416, + "learning_rate": 0.00018357799193186912, + "loss": 0.4391, + "step": 29265 + }, + { + "epoch": 1.6351370226555297, + "grad_norm": 0.3062775135040283, + "learning_rate": 0.00018354997758852532, + "loss": 0.3691, + "step": 29266 + }, + { + "epoch": 1.6351928932592117, + "grad_norm": 0.3915121555328369, + "learning_rate": 0.00018352196324518153, + "loss": 0.3272, + "step": 29267 + }, + { + "epoch": 1.6352487638628936, + "grad_norm": 0.44633808732032776, + "learning_rate": 0.00018349394890183773, + "loss": 0.3606, + "step": 29268 + }, + { + "epoch": 1.6353046344665754, + "grad_norm": 0.4576810598373413, + "learning_rate": 0.00018346593455849394, + "loss": 0.4491, + "step": 29269 + }, + { + "epoch": 1.6353605050702573, + "grad_norm": 0.731796383857727, + "learning_rate": 0.00018343792021515015, + "loss": 0.3772, + "step": 29270 + }, + { + "epoch": 1.6354163756739393, + "grad_norm": 2.514845609664917, + "learning_rate": 0.00018340990587180638, + "loss": 0.4004, + "step": 29271 + }, + { + "epoch": 1.635472246277621, + "grad_norm": 0.3555956780910492, + "learning_rate": 0.00018338189152846258, + "loss": 0.4517, + "step": 29272 + }, + { + "epoch": 1.635528116881303, + "grad_norm": 0.9722654223442078, + "learning_rate": 0.0001833538771851188, + "loss": 0.4166, + "step": 29273 + }, + { + "epoch": 1.635583987484985, + "grad_norm": 0.46338140964508057, + "learning_rate": 0.000183325862841775, + "loss": 0.5977, + "step": 29274 + }, + { + "epoch": 1.6356398580886666, + "grad_norm": 0.9396249651908875, + "learning_rate": 0.0001832978484984312, + "loss": 0.3953, + "step": 29275 + }, + { + "epoch": 1.6356957286923484, + "grad_norm": 0.4007190763950348, + "learning_rate": 0.0001832698341550874, + "loss": 0.4608, + "step": 29276 + }, + { + "epoch": 1.6357515992960305, + "grad_norm": 0.6286928057670593, + "learning_rate": 0.0001832418198117436, + "loss": 0.4187, + "step": 29277 + }, + { + "epoch": 1.6358074698997123, + "grad_norm": 0.34600234031677246, + "learning_rate": 0.00018321380546839982, + "loss": 0.4194, + "step": 29278 + }, + { + "epoch": 1.635863340503394, + "grad_norm": 0.38533663749694824, + "learning_rate": 0.00018318579112505605, + "loss": 0.4102, + "step": 29279 + }, + { + "epoch": 1.6359192111070762, + "grad_norm": 1.0186856985092163, + "learning_rate": 0.00018315777678171226, + "loss": 0.4613, + "step": 29280 + }, + { + "epoch": 1.635975081710758, + "grad_norm": 0.9034581184387207, + "learning_rate": 0.00018312976243836846, + "loss": 0.4256, + "step": 29281 + }, + { + "epoch": 1.6360309523144396, + "grad_norm": 0.6304842233657837, + "learning_rate": 0.00018310174809502467, + "loss": 0.367, + "step": 29282 + }, + { + "epoch": 1.6360868229181216, + "grad_norm": 0.36873507499694824, + "learning_rate": 0.00018307373375168088, + "loss": 0.3694, + "step": 29283 + }, + { + "epoch": 1.6361426935218035, + "grad_norm": 0.41742953658103943, + "learning_rate": 0.00018304571940833708, + "loss": 0.4232, + "step": 29284 + }, + { + "epoch": 1.6361985641254853, + "grad_norm": 0.8145022988319397, + "learning_rate": 0.00018301770506499326, + "loss": 0.4373, + "step": 29285 + }, + { + "epoch": 1.6362544347291672, + "grad_norm": 0.5330902338027954, + "learning_rate": 0.00018298969072164947, + "loss": 0.3093, + "step": 29286 + }, + { + "epoch": 1.6363103053328492, + "grad_norm": 0.724966824054718, + "learning_rate": 0.0001829616763783057, + "loss": 0.4331, + "step": 29287 + }, + { + "epoch": 1.6363661759365309, + "grad_norm": 1.506028413772583, + "learning_rate": 0.0001829336620349619, + "loss": 0.3944, + "step": 29288 + }, + { + "epoch": 1.6364220465402128, + "grad_norm": 0.5254178047180176, + "learning_rate": 0.0001829056476916181, + "loss": 0.3783, + "step": 29289 + }, + { + "epoch": 1.6364779171438948, + "grad_norm": 0.40688833594322205, + "learning_rate": 0.00018287763334827432, + "loss": 0.4325, + "step": 29290 + }, + { + "epoch": 1.6365337877475765, + "grad_norm": 0.8358919620513916, + "learning_rate": 0.00018284961900493052, + "loss": 0.5433, + "step": 29291 + }, + { + "epoch": 1.6365896583512585, + "grad_norm": 1.0199066400527954, + "learning_rate": 0.00018282160466158673, + "loss": 0.4239, + "step": 29292 + }, + { + "epoch": 1.6366455289549404, + "grad_norm": 0.3730703890323639, + "learning_rate": 0.00018279359031824293, + "loss": 0.4449, + "step": 29293 + }, + { + "epoch": 1.6367013995586221, + "grad_norm": 0.4167048931121826, + "learning_rate": 0.00018276557597489914, + "loss": 0.4365, + "step": 29294 + }, + { + "epoch": 1.636757270162304, + "grad_norm": 0.4971751868724823, + "learning_rate": 0.00018273756163155537, + "loss": 0.3305, + "step": 29295 + }, + { + "epoch": 1.636813140765986, + "grad_norm": 0.4633796513080597, + "learning_rate": 0.00018270954728821158, + "loss": 0.4968, + "step": 29296 + }, + { + "epoch": 1.6368690113696678, + "grad_norm": 0.5281404852867126, + "learning_rate": 0.00018268153294486778, + "loss": 0.6098, + "step": 29297 + }, + { + "epoch": 1.6369248819733497, + "grad_norm": 2.589146375656128, + "learning_rate": 0.000182653518601524, + "loss": 0.6458, + "step": 29298 + }, + { + "epoch": 1.6369807525770317, + "grad_norm": 0.34816205501556396, + "learning_rate": 0.0001826255042581802, + "loss": 0.4908, + "step": 29299 + }, + { + "epoch": 1.6370366231807134, + "grad_norm": 0.36424165964126587, + "learning_rate": 0.0001825974899148364, + "loss": 0.3773, + "step": 29300 + }, + { + "epoch": 1.6370924937843954, + "grad_norm": 1.281782865524292, + "learning_rate": 0.0001825694755714926, + "loss": 0.3726, + "step": 29301 + }, + { + "epoch": 1.6371483643880773, + "grad_norm": 0.4086896479129791, + "learning_rate": 0.0001825414612281488, + "loss": 0.3328, + "step": 29302 + }, + { + "epoch": 1.637204234991759, + "grad_norm": 0.4264826774597168, + "learning_rate": 0.00018251344688480505, + "loss": 0.4445, + "step": 29303 + }, + { + "epoch": 1.637260105595441, + "grad_norm": 0.7112791538238525, + "learning_rate": 0.00018248543254146125, + "loss": 0.4401, + "step": 29304 + }, + { + "epoch": 1.637315976199123, + "grad_norm": 0.6692409515380859, + "learning_rate": 0.00018245741819811746, + "loss": 0.3955, + "step": 29305 + }, + { + "epoch": 1.6373718468028047, + "grad_norm": 0.3731234669685364, + "learning_rate": 0.00018242940385477364, + "loss": 0.3, + "step": 29306 + }, + { + "epoch": 1.6374277174064866, + "grad_norm": 0.6166295409202576, + "learning_rate": 0.00018240138951142984, + "loss": 0.3313, + "step": 29307 + }, + { + "epoch": 1.6374835880101686, + "grad_norm": 1.37251877784729, + "learning_rate": 0.00018237337516808605, + "loss": 0.3663, + "step": 29308 + }, + { + "epoch": 1.6375394586138503, + "grad_norm": 0.36504191160202026, + "learning_rate": 0.00018234536082474225, + "loss": 0.438, + "step": 29309 + }, + { + "epoch": 1.637595329217532, + "grad_norm": 0.3658874034881592, + "learning_rate": 0.00018231734648139846, + "loss": 0.4046, + "step": 29310 + }, + { + "epoch": 1.6376511998212142, + "grad_norm": 0.6458540558815002, + "learning_rate": 0.0001822893321380547, + "loss": 0.4797, + "step": 29311 + }, + { + "epoch": 1.637707070424896, + "grad_norm": 0.5589548349380493, + "learning_rate": 0.0001822613177947109, + "loss": 0.3766, + "step": 29312 + }, + { + "epoch": 1.6377629410285777, + "grad_norm": 1.5217376947402954, + "learning_rate": 0.0001822333034513671, + "loss": 0.5239, + "step": 29313 + }, + { + "epoch": 1.6378188116322598, + "grad_norm": 0.39382869005203247, + "learning_rate": 0.0001822052891080233, + "loss": 0.442, + "step": 29314 + }, + { + "epoch": 1.6378746822359416, + "grad_norm": 0.37567567825317383, + "learning_rate": 0.00018217727476467952, + "loss": 0.4402, + "step": 29315 + }, + { + "epoch": 1.6379305528396233, + "grad_norm": 0.35686302185058594, + "learning_rate": 0.00018214926042133572, + "loss": 0.487, + "step": 29316 + }, + { + "epoch": 1.6379864234433053, + "grad_norm": 2.473233461380005, + "learning_rate": 0.00018212124607799193, + "loss": 0.365, + "step": 29317 + }, + { + "epoch": 1.6380422940469872, + "grad_norm": 0.3518773317337036, + "learning_rate": 0.00018209323173464813, + "loss": 0.4404, + "step": 29318 + }, + { + "epoch": 1.638098164650669, + "grad_norm": 0.2599669396877289, + "learning_rate": 0.00018206521739130437, + "loss": 0.4088, + "step": 29319 + }, + { + "epoch": 1.6381540352543509, + "grad_norm": 0.40139132738113403, + "learning_rate": 0.00018203720304796057, + "loss": 0.4279, + "step": 29320 + }, + { + "epoch": 1.6382099058580328, + "grad_norm": 0.661675751209259, + "learning_rate": 0.00018200918870461678, + "loss": 0.5267, + "step": 29321 + }, + { + "epoch": 1.6382657764617146, + "grad_norm": 0.38198935985565186, + "learning_rate": 0.00018198117436127298, + "loss": 0.3392, + "step": 29322 + }, + { + "epoch": 1.6383216470653965, + "grad_norm": 0.361878365278244, + "learning_rate": 0.0001819531600179292, + "loss": 0.3842, + "step": 29323 + }, + { + "epoch": 1.6383775176690785, + "grad_norm": 0.5194540619850159, + "learning_rate": 0.0001819251456745854, + "loss": 0.4057, + "step": 29324 + }, + { + "epoch": 1.6384333882727602, + "grad_norm": 0.44670119881629944, + "learning_rate": 0.0001818971313312416, + "loss": 0.3881, + "step": 29325 + }, + { + "epoch": 1.6384892588764421, + "grad_norm": 0.35704919695854187, + "learning_rate": 0.00018186911698789783, + "loss": 0.3676, + "step": 29326 + }, + { + "epoch": 1.638545129480124, + "grad_norm": 0.40708136558532715, + "learning_rate": 0.000181841102644554, + "loss": 0.4208, + "step": 29327 + }, + { + "epoch": 1.6386010000838058, + "grad_norm": 0.4815828800201416, + "learning_rate": 0.00018181308830121022, + "loss": 0.5919, + "step": 29328 + }, + { + "epoch": 1.6386568706874878, + "grad_norm": 0.42747384309768677, + "learning_rate": 0.00018178507395786642, + "loss": 0.5302, + "step": 29329 + }, + { + "epoch": 1.6387127412911697, + "grad_norm": 0.3636431097984314, + "learning_rate": 0.00018175705961452263, + "loss": 0.424, + "step": 29330 + }, + { + "epoch": 1.6387686118948515, + "grad_norm": 0.3725290298461914, + "learning_rate": 0.00018172904527117884, + "loss": 0.4022, + "step": 29331 + }, + { + "epoch": 1.6388244824985334, + "grad_norm": 1.2307779788970947, + "learning_rate": 0.00018170103092783504, + "loss": 0.3276, + "step": 29332 + }, + { + "epoch": 1.6388803531022154, + "grad_norm": 0.5818896889686584, + "learning_rate": 0.00018167301658449125, + "loss": 0.3674, + "step": 29333 + }, + { + "epoch": 1.638936223705897, + "grad_norm": 0.4074983596801758, + "learning_rate": 0.00018164500224114748, + "loss": 0.4144, + "step": 29334 + }, + { + "epoch": 1.638992094309579, + "grad_norm": 0.5735787749290466, + "learning_rate": 0.00018161698789780369, + "loss": 0.4627, + "step": 29335 + }, + { + "epoch": 1.639047964913261, + "grad_norm": 0.5763615369796753, + "learning_rate": 0.0001815889735544599, + "loss": 0.5184, + "step": 29336 + }, + { + "epoch": 1.6391038355169427, + "grad_norm": 0.5542707443237305, + "learning_rate": 0.0001815609592111161, + "loss": 0.4027, + "step": 29337 + }, + { + "epoch": 1.6391597061206247, + "grad_norm": 0.3271011412143707, + "learning_rate": 0.0001815329448677723, + "loss": 0.3394, + "step": 29338 + }, + { + "epoch": 1.6392155767243066, + "grad_norm": 0.562401533126831, + "learning_rate": 0.0001815049305244285, + "loss": 0.3676, + "step": 29339 + }, + { + "epoch": 1.6392714473279884, + "grad_norm": 0.40803542733192444, + "learning_rate": 0.00018147691618108471, + "loss": 0.4188, + "step": 29340 + }, + { + "epoch": 1.6393273179316703, + "grad_norm": 0.5699936151504517, + "learning_rate": 0.00018144890183774092, + "loss": 0.5095, + "step": 29341 + }, + { + "epoch": 1.6393831885353523, + "grad_norm": 0.8795342445373535, + "learning_rate": 0.00018142088749439715, + "loss": 0.4051, + "step": 29342 + }, + { + "epoch": 1.639439059139034, + "grad_norm": 0.4199928641319275, + "learning_rate": 0.00018139287315105336, + "loss": 0.4271, + "step": 29343 + }, + { + "epoch": 1.6394949297427157, + "grad_norm": 1.00046706199646, + "learning_rate": 0.00018136485880770957, + "loss": 0.3911, + "step": 29344 + }, + { + "epoch": 1.6395508003463979, + "grad_norm": 0.42309531569480896, + "learning_rate": 0.00018133684446436577, + "loss": 0.5204, + "step": 29345 + }, + { + "epoch": 1.6396066709500796, + "grad_norm": 0.3681308925151825, + "learning_rate": 0.00018130883012102198, + "loss": 0.3927, + "step": 29346 + }, + { + "epoch": 1.6396625415537613, + "grad_norm": 0.5185977220535278, + "learning_rate": 0.00018128081577767818, + "loss": 0.364, + "step": 29347 + }, + { + "epoch": 1.6397184121574435, + "grad_norm": 0.4999077618122101, + "learning_rate": 0.0001812528014343344, + "loss": 0.3778, + "step": 29348 + }, + { + "epoch": 1.6397742827611252, + "grad_norm": 0.38692784309387207, + "learning_rate": 0.00018122478709099057, + "loss": 0.504, + "step": 29349 + }, + { + "epoch": 1.639830153364807, + "grad_norm": 1.5437451601028442, + "learning_rate": 0.0001811967727476468, + "loss": 0.5678, + "step": 29350 + }, + { + "epoch": 1.639886023968489, + "grad_norm": 0.4678039848804474, + "learning_rate": 0.000181168758404303, + "loss": 0.5543, + "step": 29351 + }, + { + "epoch": 1.6399418945721709, + "grad_norm": 0.43021664023399353, + "learning_rate": 0.0001811407440609592, + "loss": 0.4685, + "step": 29352 + }, + { + "epoch": 1.6399977651758526, + "grad_norm": 0.7541439533233643, + "learning_rate": 0.00018111272971761542, + "loss": 0.3283, + "step": 29353 + }, + { + "epoch": 1.6400536357795346, + "grad_norm": 0.4683889150619507, + "learning_rate": 0.00018108471537427162, + "loss": 0.5503, + "step": 29354 + }, + { + "epoch": 1.6401095063832165, + "grad_norm": 0.3144313097000122, + "learning_rate": 0.00018105670103092783, + "loss": 0.3354, + "step": 29355 + }, + { + "epoch": 1.6401653769868982, + "grad_norm": 0.3347582519054413, + "learning_rate": 0.00018102868668758404, + "loss": 0.3604, + "step": 29356 + }, + { + "epoch": 1.6402212475905802, + "grad_norm": 0.8558241724967957, + "learning_rate": 0.00018100067234424024, + "loss": 0.4275, + "step": 29357 + }, + { + "epoch": 1.6402771181942621, + "grad_norm": 1.953269600868225, + "learning_rate": 0.00018097265800089647, + "loss": 0.4099, + "step": 29358 + }, + { + "epoch": 1.6403329887979439, + "grad_norm": 0.4280686378479004, + "learning_rate": 0.00018094464365755268, + "loss": 0.3919, + "step": 29359 + }, + { + "epoch": 1.6403888594016258, + "grad_norm": 0.42514467239379883, + "learning_rate": 0.00018091662931420889, + "loss": 0.3992, + "step": 29360 + }, + { + "epoch": 1.6404447300053078, + "grad_norm": 0.7849116921424866, + "learning_rate": 0.0001808886149708651, + "loss": 0.3455, + "step": 29361 + }, + { + "epoch": 1.6405006006089895, + "grad_norm": 0.706970751285553, + "learning_rate": 0.0001808606006275213, + "loss": 0.4125, + "step": 29362 + }, + { + "epoch": 1.6405564712126715, + "grad_norm": 0.3311614692211151, + "learning_rate": 0.0001808325862841775, + "loss": 0.2814, + "step": 29363 + }, + { + "epoch": 1.6406123418163534, + "grad_norm": 0.4256269335746765, + "learning_rate": 0.0001808045719408337, + "loss": 0.4322, + "step": 29364 + }, + { + "epoch": 1.6406682124200351, + "grad_norm": 0.5214030146598816, + "learning_rate": 0.00018077655759748991, + "loss": 0.5249, + "step": 29365 + }, + { + "epoch": 1.640724083023717, + "grad_norm": 0.325651615858078, + "learning_rate": 0.00018074854325414615, + "loss": 0.4684, + "step": 29366 + }, + { + "epoch": 1.640779953627399, + "grad_norm": 0.4036239981651306, + "learning_rate": 0.00018072052891080235, + "loss": 0.3571, + "step": 29367 + }, + { + "epoch": 1.6408358242310808, + "grad_norm": 0.5693024396896362, + "learning_rate": 0.00018069251456745856, + "loss": 0.5513, + "step": 29368 + }, + { + "epoch": 1.6408916948347627, + "grad_norm": 0.44439199566841125, + "learning_rate": 0.00018066450022411476, + "loss": 0.394, + "step": 29369 + }, + { + "epoch": 1.6409475654384447, + "grad_norm": 1.2056490182876587, + "learning_rate": 0.00018063648588077094, + "loss": 0.5215, + "step": 29370 + }, + { + "epoch": 1.6410034360421264, + "grad_norm": 0.317516028881073, + "learning_rate": 0.00018060847153742715, + "loss": 0.438, + "step": 29371 + }, + { + "epoch": 1.6410593066458083, + "grad_norm": 0.4930955767631531, + "learning_rate": 0.00018058045719408336, + "loss": 0.3679, + "step": 29372 + }, + { + "epoch": 1.6411151772494903, + "grad_norm": 0.4183426797389984, + "learning_rate": 0.00018055244285073956, + "loss": 0.3759, + "step": 29373 + }, + { + "epoch": 1.641171047853172, + "grad_norm": 0.5650453567504883, + "learning_rate": 0.0001805244285073958, + "loss": 0.3807, + "step": 29374 + }, + { + "epoch": 1.641226918456854, + "grad_norm": 0.5744853019714355, + "learning_rate": 0.000180496414164052, + "loss": 0.4439, + "step": 29375 + }, + { + "epoch": 1.641282789060536, + "grad_norm": 0.484046995639801, + "learning_rate": 0.0001804683998207082, + "loss": 0.4044, + "step": 29376 + }, + { + "epoch": 1.6413386596642177, + "grad_norm": 0.7272024154663086, + "learning_rate": 0.0001804403854773644, + "loss": 0.3641, + "step": 29377 + }, + { + "epoch": 1.6413945302678994, + "grad_norm": 0.4428675174713135, + "learning_rate": 0.00018041237113402062, + "loss": 0.3931, + "step": 29378 + }, + { + "epoch": 1.6414504008715816, + "grad_norm": 0.4952127933502197, + "learning_rate": 0.00018038435679067682, + "loss": 0.4677, + "step": 29379 + }, + { + "epoch": 1.6415062714752633, + "grad_norm": 0.40221571922302246, + "learning_rate": 0.00018035634244733303, + "loss": 0.4351, + "step": 29380 + }, + { + "epoch": 1.641562142078945, + "grad_norm": 0.3523838520050049, + "learning_rate": 0.00018032832810398923, + "loss": 0.4001, + "step": 29381 + }, + { + "epoch": 1.6416180126826272, + "grad_norm": 1.6651785373687744, + "learning_rate": 0.00018030031376064547, + "loss": 0.3438, + "step": 29382 + }, + { + "epoch": 1.641673883286309, + "grad_norm": 0.44451847672462463, + "learning_rate": 0.00018027229941730167, + "loss": 0.3888, + "step": 29383 + }, + { + "epoch": 1.6417297538899907, + "grad_norm": 0.38926029205322266, + "learning_rate": 0.00018024428507395788, + "loss": 0.4052, + "step": 29384 + }, + { + "epoch": 1.6417856244936726, + "grad_norm": 0.5073692202568054, + "learning_rate": 0.00018021627073061409, + "loss": 0.3873, + "step": 29385 + }, + { + "epoch": 1.6418414950973546, + "grad_norm": 0.6689884066581726, + "learning_rate": 0.0001801882563872703, + "loss": 0.3704, + "step": 29386 + }, + { + "epoch": 1.6418973657010363, + "grad_norm": 0.5181166529655457, + "learning_rate": 0.0001801602420439265, + "loss": 0.3667, + "step": 29387 + }, + { + "epoch": 1.6419532363047182, + "grad_norm": 1.0174305438995361, + "learning_rate": 0.0001801322277005827, + "loss": 0.4031, + "step": 29388 + }, + { + "epoch": 1.6420091069084002, + "grad_norm": 0.38965722918510437, + "learning_rate": 0.0001801042133572389, + "loss": 0.4473, + "step": 29389 + }, + { + "epoch": 1.642064977512082, + "grad_norm": 0.5069434642791748, + "learning_rate": 0.00018007619901389514, + "loss": 0.4031, + "step": 29390 + }, + { + "epoch": 1.6421208481157639, + "grad_norm": 0.6144258975982666, + "learning_rate": 0.00018004818467055132, + "loss": 0.3508, + "step": 29391 + }, + { + "epoch": 1.6421767187194458, + "grad_norm": 0.47068360447883606, + "learning_rate": 0.00018002017032720753, + "loss": 0.366, + "step": 29392 + }, + { + "epoch": 1.6422325893231275, + "grad_norm": 0.5826690793037415, + "learning_rate": 0.00017999215598386373, + "loss": 0.4698, + "step": 29393 + }, + { + "epoch": 1.6422884599268095, + "grad_norm": 0.43528175354003906, + "learning_rate": 0.00017996414164051994, + "loss": 0.3905, + "step": 29394 + }, + { + "epoch": 1.6423443305304914, + "grad_norm": 0.988699734210968, + "learning_rate": 0.00017993612729717614, + "loss": 0.5333, + "step": 29395 + }, + { + "epoch": 1.6424002011341732, + "grad_norm": 0.8102937936782837, + "learning_rate": 0.00017990811295383235, + "loss": 0.3681, + "step": 29396 + }, + { + "epoch": 1.6424560717378551, + "grad_norm": 0.45325860381126404, + "learning_rate": 0.00017988009861048855, + "loss": 0.355, + "step": 29397 + }, + { + "epoch": 1.642511942341537, + "grad_norm": 0.372859388589859, + "learning_rate": 0.0001798520842671448, + "loss": 0.3945, + "step": 29398 + }, + { + "epoch": 1.6425678129452188, + "grad_norm": 1.6728265285491943, + "learning_rate": 0.000179824069923801, + "loss": 0.4825, + "step": 29399 + }, + { + "epoch": 1.6426236835489008, + "grad_norm": 3.0412561893463135, + "learning_rate": 0.0001797960555804572, + "loss": 0.3436, + "step": 29400 + }, + { + "epoch": 1.6426795541525827, + "grad_norm": 0.3512091636657715, + "learning_rate": 0.0001797680412371134, + "loss": 0.3122, + "step": 29401 + }, + { + "epoch": 1.6427354247562644, + "grad_norm": 0.3797951936721802, + "learning_rate": 0.0001797400268937696, + "loss": 0.4251, + "step": 29402 + }, + { + "epoch": 1.6427912953599464, + "grad_norm": 0.530592679977417, + "learning_rate": 0.00017971201255042582, + "loss": 0.4853, + "step": 29403 + }, + { + "epoch": 1.6428471659636283, + "grad_norm": 0.4089137613773346, + "learning_rate": 0.00017968399820708202, + "loss": 0.5181, + "step": 29404 + }, + { + "epoch": 1.64290303656731, + "grad_norm": 0.3308374285697937, + "learning_rate": 0.00017965598386373826, + "loss": 0.388, + "step": 29405 + }, + { + "epoch": 1.642958907170992, + "grad_norm": 0.5353536605834961, + "learning_rate": 0.00017962796952039446, + "loss": 0.3489, + "step": 29406 + }, + { + "epoch": 1.643014777774674, + "grad_norm": 0.9565948247909546, + "learning_rate": 0.00017959995517705067, + "loss": 0.5259, + "step": 29407 + }, + { + "epoch": 1.6430706483783557, + "grad_norm": 6.514998912811279, + "learning_rate": 0.00017957194083370687, + "loss": 0.4242, + "step": 29408 + }, + { + "epoch": 1.6431265189820377, + "grad_norm": 0.3473621904850006, + "learning_rate": 0.00017954392649036308, + "loss": 0.4301, + "step": 29409 + }, + { + "epoch": 1.6431823895857196, + "grad_norm": 0.5044518113136292, + "learning_rate": 0.00017951591214701928, + "loss": 0.3779, + "step": 29410 + }, + { + "epoch": 1.6432382601894013, + "grad_norm": 0.3817472755908966, + "learning_rate": 0.0001794878978036755, + "loss": 0.3451, + "step": 29411 + }, + { + "epoch": 1.643294130793083, + "grad_norm": 0.30775901675224304, + "learning_rate": 0.00017945988346033167, + "loss": 0.3464, + "step": 29412 + }, + { + "epoch": 1.6433500013967652, + "grad_norm": 0.4094618856906891, + "learning_rate": 0.0001794318691169879, + "loss": 0.38, + "step": 29413 + }, + { + "epoch": 1.643405872000447, + "grad_norm": 0.5258957743644714, + "learning_rate": 0.0001794038547736441, + "loss": 0.37, + "step": 29414 + }, + { + "epoch": 1.6434617426041287, + "grad_norm": 0.41012853384017944, + "learning_rate": 0.00017937584043030031, + "loss": 0.3921, + "step": 29415 + }, + { + "epoch": 1.6435176132078109, + "grad_norm": 0.3527962267398834, + "learning_rate": 0.00017934782608695652, + "loss": 0.3528, + "step": 29416 + }, + { + "epoch": 1.6435734838114926, + "grad_norm": 0.4570799171924591, + "learning_rate": 0.00017931981174361273, + "loss": 0.3876, + "step": 29417 + }, + { + "epoch": 1.6436293544151743, + "grad_norm": 0.3778747022151947, + "learning_rate": 0.00017929179740026893, + "loss": 0.5002, + "step": 29418 + }, + { + "epoch": 1.6436852250188563, + "grad_norm": 0.5191072225570679, + "learning_rate": 0.00017926378305692514, + "loss": 0.43, + "step": 29419 + }, + { + "epoch": 1.6437410956225382, + "grad_norm": 0.6262335181236267, + "learning_rate": 0.00017923576871358134, + "loss": 0.3314, + "step": 29420 + }, + { + "epoch": 1.64379696622622, + "grad_norm": 0.5505505800247192, + "learning_rate": 0.00017920775437023758, + "loss": 0.2968, + "step": 29421 + }, + { + "epoch": 1.643852836829902, + "grad_norm": 0.3294391334056854, + "learning_rate": 0.00017917974002689378, + "loss": 0.3499, + "step": 29422 + }, + { + "epoch": 1.6439087074335839, + "grad_norm": 0.5159341096878052, + "learning_rate": 0.00017915172568355, + "loss": 0.4825, + "step": 29423 + }, + { + "epoch": 1.6439645780372656, + "grad_norm": 0.5870281457901001, + "learning_rate": 0.0001791237113402062, + "loss": 0.4305, + "step": 29424 + }, + { + "epoch": 1.6440204486409475, + "grad_norm": 0.49357616901397705, + "learning_rate": 0.0001790956969968624, + "loss": 0.4638, + "step": 29425 + }, + { + "epoch": 1.6440763192446295, + "grad_norm": 0.5186004042625427, + "learning_rate": 0.0001790676826535186, + "loss": 0.4617, + "step": 29426 + }, + { + "epoch": 1.6441321898483112, + "grad_norm": 0.4869844615459442, + "learning_rate": 0.0001790396683101748, + "loss": 0.4046, + "step": 29427 + }, + { + "epoch": 1.6441880604519932, + "grad_norm": 0.41091302037239075, + "learning_rate": 0.00017901165396683102, + "loss": 0.4072, + "step": 29428 + }, + { + "epoch": 1.6442439310556751, + "grad_norm": 0.40250277519226074, + "learning_rate": 0.00017898363962348725, + "loss": 0.5018, + "step": 29429 + }, + { + "epoch": 1.6442998016593569, + "grad_norm": 0.7879865765571594, + "learning_rate": 0.00017895562528014346, + "loss": 0.4684, + "step": 29430 + }, + { + "epoch": 1.6443556722630388, + "grad_norm": 0.4209537208080292, + "learning_rate": 0.00017892761093679966, + "loss": 0.5378, + "step": 29431 + }, + { + "epoch": 1.6444115428667208, + "grad_norm": 5.159926414489746, + "learning_rate": 0.00017889959659345587, + "loss": 0.3282, + "step": 29432 + }, + { + "epoch": 1.6444674134704025, + "grad_norm": 1.1920785903930664, + "learning_rate": 0.00017887158225011207, + "loss": 0.4024, + "step": 29433 + }, + { + "epoch": 1.6445232840740844, + "grad_norm": 0.39883360266685486, + "learning_rate": 0.00017884356790676825, + "loss": 0.481, + "step": 29434 + }, + { + "epoch": 1.6445791546777664, + "grad_norm": 0.4993338882923126, + "learning_rate": 0.00017881555356342446, + "loss": 0.4242, + "step": 29435 + }, + { + "epoch": 1.6446350252814481, + "grad_norm": 0.40072357654571533, + "learning_rate": 0.00017878753922008066, + "loss": 0.3896, + "step": 29436 + }, + { + "epoch": 1.64469089588513, + "grad_norm": 0.32099419832229614, + "learning_rate": 0.0001787595248767369, + "loss": 0.3794, + "step": 29437 + }, + { + "epoch": 1.644746766488812, + "grad_norm": 0.4493448734283447, + "learning_rate": 0.0001787315105333931, + "loss": 0.6266, + "step": 29438 + }, + { + "epoch": 1.6448026370924937, + "grad_norm": 0.48728325963020325, + "learning_rate": 0.0001787034961900493, + "loss": 0.4173, + "step": 29439 + }, + { + "epoch": 1.6448585076961757, + "grad_norm": 0.43236497044563293, + "learning_rate": 0.0001786754818467055, + "loss": 0.44, + "step": 29440 + }, + { + "epoch": 1.6449143782998576, + "grad_norm": 0.4454898238182068, + "learning_rate": 0.00017864746750336172, + "loss": 0.4007, + "step": 29441 + }, + { + "epoch": 1.6449702489035394, + "grad_norm": 2.5817341804504395, + "learning_rate": 0.00017861945316001792, + "loss": 0.4298, + "step": 29442 + }, + { + "epoch": 1.6450261195072213, + "grad_norm": 0.26529213786125183, + "learning_rate": 0.00017859143881667413, + "loss": 0.2614, + "step": 29443 + }, + { + "epoch": 1.6450819901109033, + "grad_norm": 0.3442313075065613, + "learning_rate": 0.00017856342447333034, + "loss": 0.375, + "step": 29444 + }, + { + "epoch": 1.645137860714585, + "grad_norm": 0.5232978463172913, + "learning_rate": 0.00017853541012998657, + "loss": 0.433, + "step": 29445 + }, + { + "epoch": 1.6451937313182667, + "grad_norm": 0.44334837794303894, + "learning_rate": 0.00017850739578664278, + "loss": 0.3654, + "step": 29446 + }, + { + "epoch": 1.645249601921949, + "grad_norm": 0.4412645399570465, + "learning_rate": 0.00017847938144329898, + "loss": 0.3708, + "step": 29447 + }, + { + "epoch": 1.6453054725256306, + "grad_norm": 4.874611854553223, + "learning_rate": 0.0001784513670999552, + "loss": 0.4427, + "step": 29448 + }, + { + "epoch": 1.6453613431293124, + "grad_norm": 0.4772338271141052, + "learning_rate": 0.0001784233527566114, + "loss": 0.402, + "step": 29449 + }, + { + "epoch": 1.6454172137329945, + "grad_norm": 0.45342037081718445, + "learning_rate": 0.0001783953384132676, + "loss": 0.5365, + "step": 29450 + }, + { + "epoch": 1.6454730843366763, + "grad_norm": 0.47301825881004333, + "learning_rate": 0.0001783673240699238, + "loss": 0.3266, + "step": 29451 + }, + { + "epoch": 1.645528954940358, + "grad_norm": 0.3991262912750244, + "learning_rate": 0.00017833930972658, + "loss": 0.4289, + "step": 29452 + }, + { + "epoch": 1.64558482554404, + "grad_norm": 1.3393480777740479, + "learning_rate": 0.00017831129538323624, + "loss": 0.3314, + "step": 29453 + }, + { + "epoch": 1.645640696147722, + "grad_norm": 0.501535952091217, + "learning_rate": 0.00017828328103989245, + "loss": 0.4602, + "step": 29454 + }, + { + "epoch": 1.6456965667514036, + "grad_norm": 2.6239168643951416, + "learning_rate": 0.00017825526669654863, + "loss": 0.4242, + "step": 29455 + }, + { + "epoch": 1.6457524373550856, + "grad_norm": 0.5644170641899109, + "learning_rate": 0.00017822725235320483, + "loss": 0.3263, + "step": 29456 + }, + { + "epoch": 1.6458083079587675, + "grad_norm": 0.3042939007282257, + "learning_rate": 0.00017819923800986104, + "loss": 0.3467, + "step": 29457 + }, + { + "epoch": 1.6458641785624493, + "grad_norm": 0.4470001459121704, + "learning_rate": 0.00017817122366651725, + "loss": 0.3402, + "step": 29458 + }, + { + "epoch": 1.6459200491661312, + "grad_norm": 0.44680604338645935, + "learning_rate": 0.00017814320932317345, + "loss": 0.4969, + "step": 29459 + }, + { + "epoch": 1.6459759197698132, + "grad_norm": 0.5270786881446838, + "learning_rate": 0.00017811519497982966, + "loss": 0.3452, + "step": 29460 + }, + { + "epoch": 1.646031790373495, + "grad_norm": 0.5101353526115417, + "learning_rate": 0.0001780871806364859, + "loss": 0.5697, + "step": 29461 + }, + { + "epoch": 1.6460876609771768, + "grad_norm": 0.3807760775089264, + "learning_rate": 0.0001780591662931421, + "loss": 0.404, + "step": 29462 + }, + { + "epoch": 1.6461435315808588, + "grad_norm": 0.33062267303466797, + "learning_rate": 0.0001780311519497983, + "loss": 0.394, + "step": 29463 + }, + { + "epoch": 1.6461994021845405, + "grad_norm": 0.3067866861820221, + "learning_rate": 0.0001780031376064545, + "loss": 0.3959, + "step": 29464 + }, + { + "epoch": 1.6462552727882225, + "grad_norm": 1.1357241868972778, + "learning_rate": 0.0001779751232631107, + "loss": 0.336, + "step": 29465 + }, + { + "epoch": 1.6463111433919044, + "grad_norm": 0.4152924418449402, + "learning_rate": 0.00017794710891976692, + "loss": 0.4353, + "step": 29466 + }, + { + "epoch": 1.6463670139955862, + "grad_norm": 0.3456061780452728, + "learning_rate": 0.00017791909457642312, + "loss": 0.4261, + "step": 29467 + }, + { + "epoch": 1.646422884599268, + "grad_norm": 0.4656878411769867, + "learning_rate": 0.00017789108023307936, + "loss": 0.4232, + "step": 29468 + }, + { + "epoch": 1.64647875520295, + "grad_norm": 0.4445704221725464, + "learning_rate": 0.00017786306588973556, + "loss": 0.3668, + "step": 29469 + }, + { + "epoch": 1.6465346258066318, + "grad_norm": 0.34444254636764526, + "learning_rate": 0.00017783505154639177, + "loss": 0.3532, + "step": 29470 + }, + { + "epoch": 1.6465904964103137, + "grad_norm": 0.6209419369697571, + "learning_rate": 0.00017780703720304797, + "loss": 0.5134, + "step": 29471 + }, + { + "epoch": 1.6466463670139957, + "grad_norm": 0.7002026438713074, + "learning_rate": 0.00017777902285970418, + "loss": 0.3883, + "step": 29472 + }, + { + "epoch": 1.6467022376176774, + "grad_norm": 0.47945863008499146, + "learning_rate": 0.0001777510085163604, + "loss": 0.4308, + "step": 29473 + }, + { + "epoch": 1.6467581082213594, + "grad_norm": 0.33870729804039, + "learning_rate": 0.0001777229941730166, + "loss": 0.4211, + "step": 29474 + }, + { + "epoch": 1.6468139788250413, + "grad_norm": 1.363106608390808, + "learning_rate": 0.0001776949798296728, + "loss": 0.5113, + "step": 29475 + }, + { + "epoch": 1.646869849428723, + "grad_norm": 0.4069608449935913, + "learning_rate": 0.000177666965486329, + "loss": 0.3084, + "step": 29476 + }, + { + "epoch": 1.646925720032405, + "grad_norm": 0.4343098998069763, + "learning_rate": 0.0001776389511429852, + "loss": 0.4638, + "step": 29477 + }, + { + "epoch": 1.646981590636087, + "grad_norm": 0.5515615344047546, + "learning_rate": 0.00017761093679964142, + "loss": 0.4462, + "step": 29478 + }, + { + "epoch": 1.6470374612397687, + "grad_norm": 0.3476084768772125, + "learning_rate": 0.00017758292245629762, + "loss": 0.4109, + "step": 29479 + }, + { + "epoch": 1.6470933318434504, + "grad_norm": 1.468677043914795, + "learning_rate": 0.00017755490811295383, + "loss": 0.429, + "step": 29480 + }, + { + "epoch": 1.6471492024471326, + "grad_norm": 0.49224695563316345, + "learning_rate": 0.00017752689376961003, + "loss": 0.608, + "step": 29481 + }, + { + "epoch": 1.6472050730508143, + "grad_norm": 0.6136031150817871, + "learning_rate": 0.00017749887942626624, + "loss": 0.3837, + "step": 29482 + }, + { + "epoch": 1.647260943654496, + "grad_norm": 0.6294516324996948, + "learning_rate": 0.00017747086508292244, + "loss": 0.4411, + "step": 29483 + }, + { + "epoch": 1.6473168142581782, + "grad_norm": 0.4464815557003021, + "learning_rate": 0.00017744285073957868, + "loss": 0.3698, + "step": 29484 + }, + { + "epoch": 1.64737268486186, + "grad_norm": 0.3515922427177429, + "learning_rate": 0.00017741483639623488, + "loss": 0.3795, + "step": 29485 + }, + { + "epoch": 1.6474285554655417, + "grad_norm": 0.45129260420799255, + "learning_rate": 0.0001773868220528911, + "loss": 0.5078, + "step": 29486 + }, + { + "epoch": 1.6474844260692236, + "grad_norm": 0.3410910964012146, + "learning_rate": 0.0001773588077095473, + "loss": 0.3438, + "step": 29487 + }, + { + "epoch": 1.6475402966729056, + "grad_norm": 0.33243268728256226, + "learning_rate": 0.0001773307933662035, + "loss": 0.3968, + "step": 29488 + }, + { + "epoch": 1.6475961672765873, + "grad_norm": 0.5091156363487244, + "learning_rate": 0.0001773027790228597, + "loss": 0.3314, + "step": 29489 + }, + { + "epoch": 1.6476520378802693, + "grad_norm": 0.359224796295166, + "learning_rate": 0.0001772747646795159, + "loss": 0.4154, + "step": 29490 + }, + { + "epoch": 1.6477079084839512, + "grad_norm": 0.3854990005493164, + "learning_rate": 0.00017724675033617212, + "loss": 0.4286, + "step": 29491 + }, + { + "epoch": 1.647763779087633, + "grad_norm": 0.37226295471191406, + "learning_rate": 0.00017721873599282835, + "loss": 0.4677, + "step": 29492 + }, + { + "epoch": 1.647819649691315, + "grad_norm": 0.2877601385116577, + "learning_rate": 0.00017719072164948456, + "loss": 0.3103, + "step": 29493 + }, + { + "epoch": 1.6478755202949968, + "grad_norm": 0.35536742210388184, + "learning_rate": 0.00017716270730614076, + "loss": 0.4447, + "step": 29494 + }, + { + "epoch": 1.6479313908986786, + "grad_norm": 0.49771493673324585, + "learning_rate": 0.00017713469296279697, + "loss": 0.5822, + "step": 29495 + }, + { + "epoch": 1.6479872615023605, + "grad_norm": 0.3516300618648529, + "learning_rate": 0.00017710667861945317, + "loss": 0.4012, + "step": 29496 + }, + { + "epoch": 1.6480431321060425, + "grad_norm": 0.5138104557991028, + "learning_rate": 0.00017707866427610935, + "loss": 0.4285, + "step": 29497 + }, + { + "epoch": 1.6480990027097242, + "grad_norm": 0.4005357325077057, + "learning_rate": 0.00017705064993276556, + "loss": 0.4783, + "step": 29498 + }, + { + "epoch": 1.6481548733134062, + "grad_norm": 0.36508211493492126, + "learning_rate": 0.00017702263558942176, + "loss": 0.34, + "step": 29499 + }, + { + "epoch": 1.648210743917088, + "grad_norm": 0.39163750410079956, + "learning_rate": 0.000176994621246078, + "loss": 0.4593, + "step": 29500 + }, + { + "epoch": 1.648210743917088, + "eval_cer": 0.0833360610127328, + "eval_loss": 0.3155766725540161, + "eval_runtime": 55.9347, + "eval_samples_per_second": 81.13, + "eval_steps_per_second": 5.077, + "eval_wer": 0.3296555962123963, + "step": 29500 + }, + { + "epoch": 1.6482666145207698, + "grad_norm": 0.9616982936859131, + "learning_rate": 0.0001769666069027342, + "loss": 0.3455, + "step": 29501 + }, + { + "epoch": 1.6483224851244518, + "grad_norm": 0.6456069946289062, + "learning_rate": 0.0001769385925593904, + "loss": 0.285, + "step": 29502 + }, + { + "epoch": 1.6483783557281337, + "grad_norm": 0.37016576528549194, + "learning_rate": 0.00017691057821604662, + "loss": 0.3866, + "step": 29503 + }, + { + "epoch": 1.6484342263318155, + "grad_norm": 0.44239330291748047, + "learning_rate": 0.00017688256387270282, + "loss": 0.4533, + "step": 29504 + }, + { + "epoch": 1.6484900969354974, + "grad_norm": 0.7211301922798157, + "learning_rate": 0.00017685454952935903, + "loss": 0.5974, + "step": 29505 + }, + { + "epoch": 1.6485459675391794, + "grad_norm": 0.4356355667114258, + "learning_rate": 0.00017682653518601523, + "loss": 0.3446, + "step": 29506 + }, + { + "epoch": 1.648601838142861, + "grad_norm": 0.32927894592285156, + "learning_rate": 0.00017679852084267144, + "loss": 0.3849, + "step": 29507 + }, + { + "epoch": 1.648657708746543, + "grad_norm": 0.5317347049713135, + "learning_rate": 0.00017677050649932767, + "loss": 0.3435, + "step": 29508 + }, + { + "epoch": 1.648713579350225, + "grad_norm": 0.5605387091636658, + "learning_rate": 0.00017674249215598388, + "loss": 0.3577, + "step": 29509 + }, + { + "epoch": 1.6487694499539067, + "grad_norm": 0.4979918301105499, + "learning_rate": 0.00017671447781264008, + "loss": 0.3642, + "step": 29510 + }, + { + "epoch": 1.6488253205575885, + "grad_norm": 1.5686691999435425, + "learning_rate": 0.0001766864634692963, + "loss": 0.3946, + "step": 29511 + }, + { + "epoch": 1.6488811911612706, + "grad_norm": 0.3596523404121399, + "learning_rate": 0.0001766584491259525, + "loss": 0.4423, + "step": 29512 + }, + { + "epoch": 1.6489370617649524, + "grad_norm": 0.9280366897583008, + "learning_rate": 0.0001766304347826087, + "loss": 0.582, + "step": 29513 + }, + { + "epoch": 1.648992932368634, + "grad_norm": 0.36422407627105713, + "learning_rate": 0.0001766024204392649, + "loss": 0.3893, + "step": 29514 + }, + { + "epoch": 1.6490488029723163, + "grad_norm": 0.6316453218460083, + "learning_rate": 0.0001765744060959211, + "loss": 0.3748, + "step": 29515 + }, + { + "epoch": 1.649104673575998, + "grad_norm": 0.5709185600280762, + "learning_rate": 0.00017654639175257735, + "loss": 0.4668, + "step": 29516 + }, + { + "epoch": 1.6491605441796797, + "grad_norm": 0.5684497952461243, + "learning_rate": 0.00017651837740923355, + "loss": 0.6557, + "step": 29517 + }, + { + "epoch": 1.649216414783362, + "grad_norm": 0.502833366394043, + "learning_rate": 0.00017649036306588976, + "loss": 0.3825, + "step": 29518 + }, + { + "epoch": 1.6492722853870436, + "grad_norm": 0.6219872832298279, + "learning_rate": 0.00017646234872254594, + "loss": 0.3577, + "step": 29519 + }, + { + "epoch": 1.6493281559907254, + "grad_norm": 0.9092152714729309, + "learning_rate": 0.00017643433437920214, + "loss": 0.4408, + "step": 29520 + }, + { + "epoch": 1.6493840265944073, + "grad_norm": 0.34662118554115295, + "learning_rate": 0.00017640632003585835, + "loss": 0.4148, + "step": 29521 + }, + { + "epoch": 1.6494398971980893, + "grad_norm": 0.5025126338005066, + "learning_rate": 0.00017637830569251455, + "loss": 0.3782, + "step": 29522 + }, + { + "epoch": 1.649495767801771, + "grad_norm": 0.3486143946647644, + "learning_rate": 0.00017635029134917076, + "loss": 0.3208, + "step": 29523 + }, + { + "epoch": 1.649551638405453, + "grad_norm": 0.370086133480072, + "learning_rate": 0.000176322277005827, + "loss": 0.3942, + "step": 29524 + }, + { + "epoch": 1.6496075090091349, + "grad_norm": 0.3233477473258972, + "learning_rate": 0.0001762942626624832, + "loss": 0.4932, + "step": 29525 + }, + { + "epoch": 1.6496633796128166, + "grad_norm": 0.4315149188041687, + "learning_rate": 0.0001762662483191394, + "loss": 0.3824, + "step": 29526 + }, + { + "epoch": 1.6497192502164986, + "grad_norm": 0.5425122380256653, + "learning_rate": 0.0001762382339757956, + "loss": 0.5378, + "step": 29527 + }, + { + "epoch": 1.6497751208201805, + "grad_norm": 0.5359933972358704, + "learning_rate": 0.00017621021963245181, + "loss": 0.3983, + "step": 29528 + }, + { + "epoch": 1.6498309914238622, + "grad_norm": 0.4055759608745575, + "learning_rate": 0.00017618220528910802, + "loss": 0.3313, + "step": 29529 + }, + { + "epoch": 1.6498868620275442, + "grad_norm": 0.3110582232475281, + "learning_rate": 0.00017615419094576423, + "loss": 0.3375, + "step": 29530 + }, + { + "epoch": 1.6499427326312261, + "grad_norm": 0.3107266426086426, + "learning_rate": 0.00017612617660242046, + "loss": 0.3892, + "step": 29531 + }, + { + "epoch": 1.6499986032349079, + "grad_norm": 5.252355575561523, + "learning_rate": 0.00017609816225907667, + "loss": 0.3828, + "step": 29532 + }, + { + "epoch": 1.6500544738385898, + "grad_norm": 0.3774723410606384, + "learning_rate": 0.00017607014791573287, + "loss": 0.3777, + "step": 29533 + }, + { + "epoch": 1.6501103444422718, + "grad_norm": 0.41686883568763733, + "learning_rate": 0.00017604213357238908, + "loss": 0.6192, + "step": 29534 + }, + { + "epoch": 1.6501662150459535, + "grad_norm": 1.5524789094924927, + "learning_rate": 0.00017601411922904528, + "loss": 0.5134, + "step": 29535 + }, + { + "epoch": 1.6502220856496355, + "grad_norm": 0.7990249395370483, + "learning_rate": 0.0001759861048857015, + "loss": 0.3344, + "step": 29536 + }, + { + "epoch": 1.6502779562533174, + "grad_norm": 0.7944322824478149, + "learning_rate": 0.0001759580905423577, + "loss": 0.4384, + "step": 29537 + }, + { + "epoch": 1.6503338268569991, + "grad_norm": 0.44066572189331055, + "learning_rate": 0.0001759300761990139, + "loss": 0.3802, + "step": 29538 + }, + { + "epoch": 1.650389697460681, + "grad_norm": 0.47501516342163086, + "learning_rate": 0.00017590206185567013, + "loss": 0.5078, + "step": 29539 + }, + { + "epoch": 1.650445568064363, + "grad_norm": 0.51439368724823, + "learning_rate": 0.0001758740475123263, + "loss": 0.3858, + "step": 29540 + }, + { + "epoch": 1.6505014386680448, + "grad_norm": 0.49234285950660706, + "learning_rate": 0.00017584603316898252, + "loss": 0.4135, + "step": 29541 + }, + { + "epoch": 1.6505573092717267, + "grad_norm": 0.4654615521430969, + "learning_rate": 0.00017581801882563872, + "loss": 0.3708, + "step": 29542 + }, + { + "epoch": 1.6506131798754087, + "grad_norm": 0.5127266049385071, + "learning_rate": 0.00017579000448229493, + "loss": 0.396, + "step": 29543 + }, + { + "epoch": 1.6506690504790904, + "grad_norm": 0.3597017824649811, + "learning_rate": 0.00017576199013895113, + "loss": 0.4503, + "step": 29544 + }, + { + "epoch": 1.6507249210827721, + "grad_norm": 0.7909939289093018, + "learning_rate": 0.00017573397579560734, + "loss": 0.3969, + "step": 29545 + }, + { + "epoch": 1.6507807916864543, + "grad_norm": 0.5018071532249451, + "learning_rate": 0.00017570596145226355, + "loss": 0.389, + "step": 29546 + }, + { + "epoch": 1.650836662290136, + "grad_norm": 0.30465468764305115, + "learning_rate": 0.00017567794710891978, + "loss": 0.271, + "step": 29547 + }, + { + "epoch": 1.6508925328938178, + "grad_norm": 0.5007274746894836, + "learning_rate": 0.00017564993276557599, + "loss": 0.5014, + "step": 29548 + }, + { + "epoch": 1.6509484034975, + "grad_norm": 0.4741688072681427, + "learning_rate": 0.0001756219184222322, + "loss": 0.3581, + "step": 29549 + }, + { + "epoch": 1.6510042741011817, + "grad_norm": 0.4175487458705902, + "learning_rate": 0.0001755939040788884, + "loss": 0.4061, + "step": 29550 + }, + { + "epoch": 1.6510601447048634, + "grad_norm": 0.7375360727310181, + "learning_rate": 0.0001755658897355446, + "loss": 0.4501, + "step": 29551 + }, + { + "epoch": 1.6511160153085456, + "grad_norm": 14.76992416381836, + "learning_rate": 0.0001755378753922008, + "loss": 0.4532, + "step": 29552 + }, + { + "epoch": 1.6511718859122273, + "grad_norm": 0.3528423607349396, + "learning_rate": 0.00017550986104885701, + "loss": 0.5212, + "step": 29553 + }, + { + "epoch": 1.651227756515909, + "grad_norm": 0.27369561791419983, + "learning_rate": 0.00017548184670551322, + "loss": 0.2786, + "step": 29554 + }, + { + "epoch": 1.651283627119591, + "grad_norm": 0.5752304196357727, + "learning_rate": 0.00017545383236216945, + "loss": 0.5091, + "step": 29555 + }, + { + "epoch": 1.651339497723273, + "grad_norm": 2.424661874771118, + "learning_rate": 0.00017542581801882566, + "loss": 0.4469, + "step": 29556 + }, + { + "epoch": 1.6513953683269547, + "grad_norm": 0.4833153486251831, + "learning_rate": 0.00017539780367548186, + "loss": 0.4127, + "step": 29557 + }, + { + "epoch": 1.6514512389306366, + "grad_norm": 0.6627767086029053, + "learning_rate": 0.00017536978933213807, + "loss": 0.4982, + "step": 29558 + }, + { + "epoch": 1.6515071095343186, + "grad_norm": 0.520148754119873, + "learning_rate": 0.00017534177498879428, + "loss": 0.3921, + "step": 29559 + }, + { + "epoch": 1.6515629801380003, + "grad_norm": 0.40704280138015747, + "learning_rate": 0.00017531376064545048, + "loss": 0.4834, + "step": 29560 + }, + { + "epoch": 1.6516188507416822, + "grad_norm": 0.4440261721611023, + "learning_rate": 0.00017528574630210666, + "loss": 0.4281, + "step": 29561 + }, + { + "epoch": 1.6516747213453642, + "grad_norm": 0.38233059644699097, + "learning_rate": 0.00017525773195876287, + "loss": 0.4138, + "step": 29562 + }, + { + "epoch": 1.651730591949046, + "grad_norm": 0.3998877704143524, + "learning_rate": 0.0001752297176154191, + "loss": 0.507, + "step": 29563 + }, + { + "epoch": 1.6517864625527279, + "grad_norm": 0.3247964382171631, + "learning_rate": 0.0001752017032720753, + "loss": 0.4771, + "step": 29564 + }, + { + "epoch": 1.6518423331564098, + "grad_norm": 0.31343504786491394, + "learning_rate": 0.0001751736889287315, + "loss": 0.3569, + "step": 29565 + }, + { + "epoch": 1.6518982037600916, + "grad_norm": 0.7066351175308228, + "learning_rate": 0.00017514567458538772, + "loss": 0.4207, + "step": 29566 + }, + { + "epoch": 1.6519540743637735, + "grad_norm": 0.3343973159790039, + "learning_rate": 0.00017511766024204392, + "loss": 0.4263, + "step": 29567 + }, + { + "epoch": 1.6520099449674555, + "grad_norm": 0.35526180267333984, + "learning_rate": 0.00017508964589870013, + "loss": 0.3952, + "step": 29568 + }, + { + "epoch": 1.6520658155711372, + "grad_norm": 0.39467543363571167, + "learning_rate": 0.00017506163155535633, + "loss": 0.5368, + "step": 29569 + }, + { + "epoch": 1.6521216861748191, + "grad_norm": 0.5589849948883057, + "learning_rate": 0.00017503361721201254, + "loss": 0.4142, + "step": 29570 + }, + { + "epoch": 1.652177556778501, + "grad_norm": 0.5162833333015442, + "learning_rate": 0.00017500560286866877, + "loss": 0.484, + "step": 29571 + }, + { + "epoch": 1.6522334273821828, + "grad_norm": 0.3809964060783386, + "learning_rate": 0.00017497758852532498, + "loss": 0.3648, + "step": 29572 + }, + { + "epoch": 1.6522892979858648, + "grad_norm": 0.31073349714279175, + "learning_rate": 0.00017494957418198118, + "loss": 0.3897, + "step": 29573 + }, + { + "epoch": 1.6523451685895467, + "grad_norm": 1.2107857465744019, + "learning_rate": 0.0001749215598386374, + "loss": 0.6161, + "step": 29574 + }, + { + "epoch": 1.6524010391932284, + "grad_norm": 0.584354043006897, + "learning_rate": 0.0001748935454952936, + "loss": 0.3913, + "step": 29575 + }, + { + "epoch": 1.6524569097969104, + "grad_norm": 0.449170857667923, + "learning_rate": 0.0001748655311519498, + "loss": 0.5007, + "step": 29576 + }, + { + "epoch": 1.6525127804005924, + "grad_norm": 1.6764369010925293, + "learning_rate": 0.000174837516808606, + "loss": 0.2764, + "step": 29577 + }, + { + "epoch": 1.652568651004274, + "grad_norm": 0.38138076663017273, + "learning_rate": 0.00017480950246526221, + "loss": 0.4743, + "step": 29578 + }, + { + "epoch": 1.6526245216079558, + "grad_norm": 0.42716434597969055, + "learning_rate": 0.00017478148812191845, + "loss": 0.3848, + "step": 29579 + }, + { + "epoch": 1.652680392211638, + "grad_norm": 0.46961086988449097, + "learning_rate": 0.00017475347377857465, + "loss": 0.644, + "step": 29580 + }, + { + "epoch": 1.6527362628153197, + "grad_norm": 0.5986471772193909, + "learning_rate": 0.00017472545943523086, + "loss": 0.4671, + "step": 29581 + }, + { + "epoch": 1.6527921334190014, + "grad_norm": 0.605772078037262, + "learning_rate": 0.00017469744509188704, + "loss": 0.4558, + "step": 29582 + }, + { + "epoch": 1.6528480040226836, + "grad_norm": 0.7085616588592529, + "learning_rate": 0.00017466943074854324, + "loss": 0.4328, + "step": 29583 + }, + { + "epoch": 1.6529038746263653, + "grad_norm": 1.1341873407363892, + "learning_rate": 0.00017464141640519945, + "loss": 0.4901, + "step": 29584 + }, + { + "epoch": 1.652959745230047, + "grad_norm": 1.7485127449035645, + "learning_rate": 0.00017461340206185565, + "loss": 0.4449, + "step": 29585 + }, + { + "epoch": 1.6530156158337292, + "grad_norm": 1.733856439590454, + "learning_rate": 0.00017458538771851186, + "loss": 0.3933, + "step": 29586 + }, + { + "epoch": 1.653071486437411, + "grad_norm": 0.8750406503677368, + "learning_rate": 0.0001745573733751681, + "loss": 0.3974, + "step": 29587 + }, + { + "epoch": 1.6531273570410927, + "grad_norm": 0.8132467269897461, + "learning_rate": 0.0001745293590318243, + "loss": 0.3983, + "step": 29588 + }, + { + "epoch": 1.6531832276447747, + "grad_norm": 0.5491474270820618, + "learning_rate": 0.0001745013446884805, + "loss": 0.5496, + "step": 29589 + }, + { + "epoch": 1.6532390982484566, + "grad_norm": 2.9213225841522217, + "learning_rate": 0.0001744733303451367, + "loss": 0.371, + "step": 29590 + }, + { + "epoch": 1.6532949688521383, + "grad_norm": 0.3491731286048889, + "learning_rate": 0.00017444531600179292, + "loss": 0.3324, + "step": 29591 + }, + { + "epoch": 1.6533508394558203, + "grad_norm": 0.35528624057769775, + "learning_rate": 0.00017441730165844912, + "loss": 0.3312, + "step": 29592 + }, + { + "epoch": 1.6534067100595022, + "grad_norm": 0.4034348428249359, + "learning_rate": 0.00017438928731510533, + "loss": 0.4868, + "step": 29593 + }, + { + "epoch": 1.653462580663184, + "grad_norm": 0.5990042090415955, + "learning_rate": 0.00017436127297176153, + "loss": 0.4108, + "step": 29594 + }, + { + "epoch": 1.653518451266866, + "grad_norm": 0.5437791347503662, + "learning_rate": 0.00017433325862841777, + "loss": 0.331, + "step": 29595 + }, + { + "epoch": 1.6535743218705479, + "grad_norm": 0.4028449058532715, + "learning_rate": 0.00017430524428507397, + "loss": 0.3952, + "step": 29596 + }, + { + "epoch": 1.6536301924742296, + "grad_norm": 0.39909428358078003, + "learning_rate": 0.00017427722994173018, + "loss": 0.4352, + "step": 29597 + }, + { + "epoch": 1.6536860630779115, + "grad_norm": 1.603016972541809, + "learning_rate": 0.00017424921559838638, + "loss": 0.3897, + "step": 29598 + }, + { + "epoch": 1.6537419336815935, + "grad_norm": 0.37601426243782043, + "learning_rate": 0.0001742212012550426, + "loss": 0.4286, + "step": 29599 + }, + { + "epoch": 1.6537978042852752, + "grad_norm": 0.4634120464324951, + "learning_rate": 0.0001741931869116988, + "loss": 0.3749, + "step": 29600 + }, + { + "epoch": 1.6538536748889572, + "grad_norm": 0.5439640283584595, + "learning_rate": 0.000174165172568355, + "loss": 0.3787, + "step": 29601 + }, + { + "epoch": 1.6539095454926391, + "grad_norm": 0.4161684811115265, + "learning_rate": 0.00017413715822501123, + "loss": 0.4889, + "step": 29602 + }, + { + "epoch": 1.6539654160963209, + "grad_norm": 0.5095134973526001, + "learning_rate": 0.00017410914388166744, + "loss": 0.5127, + "step": 29603 + }, + { + "epoch": 1.6540212867000028, + "grad_norm": 1.1261053085327148, + "learning_rate": 0.00017408112953832362, + "loss": 0.5647, + "step": 29604 + }, + { + "epoch": 1.6540771573036848, + "grad_norm": 0.512252688407898, + "learning_rate": 0.00017405311519497983, + "loss": 0.3351, + "step": 29605 + }, + { + "epoch": 1.6541330279073665, + "grad_norm": 0.4015052318572998, + "learning_rate": 0.00017402510085163603, + "loss": 0.4688, + "step": 29606 + }, + { + "epoch": 1.6541888985110484, + "grad_norm": 6.5422563552856445, + "learning_rate": 0.00017399708650829224, + "loss": 0.4377, + "step": 29607 + }, + { + "epoch": 1.6542447691147304, + "grad_norm": 0.4901329278945923, + "learning_rate": 0.00017396907216494844, + "loss": 0.3655, + "step": 29608 + }, + { + "epoch": 1.6543006397184121, + "grad_norm": 0.6122576594352722, + "learning_rate": 0.00017394105782160465, + "loss": 0.3756, + "step": 29609 + }, + { + "epoch": 1.654356510322094, + "grad_norm": 0.37459659576416016, + "learning_rate": 0.00017391304347826088, + "loss": 0.5269, + "step": 29610 + }, + { + "epoch": 1.654412380925776, + "grad_norm": 0.4953157603740692, + "learning_rate": 0.0001738850291349171, + "loss": 0.4229, + "step": 29611 + }, + { + "epoch": 1.6544682515294578, + "grad_norm": 0.5163589715957642, + "learning_rate": 0.0001738570147915733, + "loss": 0.503, + "step": 29612 + }, + { + "epoch": 1.6545241221331395, + "grad_norm": 1.9828916788101196, + "learning_rate": 0.0001738290004482295, + "loss": 0.4834, + "step": 29613 + }, + { + "epoch": 1.6545799927368217, + "grad_norm": 0.5088005065917969, + "learning_rate": 0.0001738009861048857, + "loss": 0.478, + "step": 29614 + }, + { + "epoch": 1.6546358633405034, + "grad_norm": 0.44601884484291077, + "learning_rate": 0.0001737729717615419, + "loss": 0.5186, + "step": 29615 + }, + { + "epoch": 1.6546917339441851, + "grad_norm": 0.46084651350975037, + "learning_rate": 0.00017374495741819812, + "loss": 0.4517, + "step": 29616 + }, + { + "epoch": 1.6547476045478673, + "grad_norm": 0.332288920879364, + "learning_rate": 0.00017371694307485432, + "loss": 0.3911, + "step": 29617 + }, + { + "epoch": 1.654803475151549, + "grad_norm": 0.3677213788032532, + "learning_rate": 0.00017368892873151056, + "loss": 0.4607, + "step": 29618 + }, + { + "epoch": 1.6548593457552307, + "grad_norm": 0.6433641910552979, + "learning_rate": 0.00017366091438816676, + "loss": 0.4947, + "step": 29619 + }, + { + "epoch": 1.654915216358913, + "grad_norm": 1.9724445343017578, + "learning_rate": 0.00017363290004482297, + "loss": 0.3464, + "step": 29620 + }, + { + "epoch": 1.6549710869625947, + "grad_norm": 0.5876967906951904, + "learning_rate": 0.00017360488570147917, + "loss": 0.3904, + "step": 29621 + }, + { + "epoch": 1.6550269575662764, + "grad_norm": 0.919748842716217, + "learning_rate": 0.00017357687135813538, + "loss": 0.3707, + "step": 29622 + }, + { + "epoch": 1.6550828281699583, + "grad_norm": 0.6037687063217163, + "learning_rate": 0.00017354885701479158, + "loss": 0.461, + "step": 29623 + }, + { + "epoch": 1.6551386987736403, + "grad_norm": 0.36658015847206116, + "learning_rate": 0.0001735208426714478, + "loss": 0.3616, + "step": 29624 + }, + { + "epoch": 1.655194569377322, + "grad_norm": 0.3246379792690277, + "learning_rate": 0.00017349282832810397, + "loss": 0.4367, + "step": 29625 + }, + { + "epoch": 1.655250439981004, + "grad_norm": 0.3765772879123688, + "learning_rate": 0.0001734648139847602, + "loss": 0.5099, + "step": 29626 + }, + { + "epoch": 1.655306310584686, + "grad_norm": 0.41098013520240784, + "learning_rate": 0.0001734367996414164, + "loss": 0.4809, + "step": 29627 + }, + { + "epoch": 1.6553621811883676, + "grad_norm": 0.2702462673187256, + "learning_rate": 0.0001734087852980726, + "loss": 0.3076, + "step": 29628 + }, + { + "epoch": 1.6554180517920496, + "grad_norm": 0.5772530436515808, + "learning_rate": 0.00017338077095472882, + "loss": 0.369, + "step": 29629 + }, + { + "epoch": 1.6554739223957315, + "grad_norm": 0.4693993926048279, + "learning_rate": 0.00017335275661138502, + "loss": 0.312, + "step": 29630 + }, + { + "epoch": 1.6555297929994133, + "grad_norm": 0.3965218961238861, + "learning_rate": 0.00017332474226804123, + "loss": 0.42, + "step": 29631 + }, + { + "epoch": 1.6555856636030952, + "grad_norm": 0.35702383518218994, + "learning_rate": 0.00017329672792469744, + "loss": 0.3384, + "step": 29632 + }, + { + "epoch": 1.6556415342067772, + "grad_norm": 0.7239370346069336, + "learning_rate": 0.00017326871358135364, + "loss": 0.366, + "step": 29633 + }, + { + "epoch": 1.655697404810459, + "grad_norm": 0.41086140275001526, + "learning_rate": 0.00017324069923800988, + "loss": 0.349, + "step": 29634 + }, + { + "epoch": 1.6557532754141409, + "grad_norm": 0.6463062167167664, + "learning_rate": 0.00017321268489466608, + "loss": 0.3194, + "step": 29635 + }, + { + "epoch": 1.6558091460178228, + "grad_norm": 0.34553205966949463, + "learning_rate": 0.0001731846705513223, + "loss": 0.4122, + "step": 29636 + }, + { + "epoch": 1.6558650166215045, + "grad_norm": 1.6378722190856934, + "learning_rate": 0.0001731566562079785, + "loss": 0.3843, + "step": 29637 + }, + { + "epoch": 1.6559208872251865, + "grad_norm": 0.4892752468585968, + "learning_rate": 0.0001731286418646347, + "loss": 0.498, + "step": 29638 + }, + { + "epoch": 1.6559767578288684, + "grad_norm": 0.4020635485649109, + "learning_rate": 0.0001731006275212909, + "loss": 0.4106, + "step": 29639 + }, + { + "epoch": 1.6560326284325502, + "grad_norm": 0.46823590993881226, + "learning_rate": 0.0001730726131779471, + "loss": 0.3755, + "step": 29640 + }, + { + "epoch": 1.6560884990362321, + "grad_norm": 0.40388908982276917, + "learning_rate": 0.00017304459883460332, + "loss": 0.3897, + "step": 29641 + }, + { + "epoch": 1.656144369639914, + "grad_norm": 0.287341445684433, + "learning_rate": 0.00017301658449125955, + "loss": 0.3432, + "step": 29642 + }, + { + "epoch": 1.6562002402435958, + "grad_norm": 0.4296491742134094, + "learning_rate": 0.00017298857014791575, + "loss": 0.5629, + "step": 29643 + }, + { + "epoch": 1.6562561108472778, + "grad_norm": 0.464245468378067, + "learning_rate": 0.00017296055580457196, + "loss": 0.5212, + "step": 29644 + }, + { + "epoch": 1.6563119814509597, + "grad_norm": 2.113445520401001, + "learning_rate": 0.00017293254146122817, + "loss": 0.3514, + "step": 29645 + }, + { + "epoch": 1.6563678520546414, + "grad_norm": 0.39052560925483704, + "learning_rate": 0.00017290452711788434, + "loss": 0.4334, + "step": 29646 + }, + { + "epoch": 1.6564237226583232, + "grad_norm": 0.5711157321929932, + "learning_rate": 0.00017287651277454055, + "loss": 0.5064, + "step": 29647 + }, + { + "epoch": 1.6564795932620053, + "grad_norm": 3.669241428375244, + "learning_rate": 0.00017284849843119676, + "loss": 0.4613, + "step": 29648 + }, + { + "epoch": 1.656535463865687, + "grad_norm": 0.4314804673194885, + "learning_rate": 0.00017282048408785296, + "loss": 0.4398, + "step": 29649 + }, + { + "epoch": 1.6565913344693688, + "grad_norm": 8.041597366333008, + "learning_rate": 0.0001727924697445092, + "loss": 0.4466, + "step": 29650 + }, + { + "epoch": 1.656647205073051, + "grad_norm": 0.3956640660762787, + "learning_rate": 0.0001727644554011654, + "loss": 0.3958, + "step": 29651 + }, + { + "epoch": 1.6567030756767327, + "grad_norm": 0.6186176538467407, + "learning_rate": 0.0001727364410578216, + "loss": 0.5358, + "step": 29652 + }, + { + "epoch": 1.6567589462804144, + "grad_norm": 1.9805856943130493, + "learning_rate": 0.0001727084267144778, + "loss": 0.4828, + "step": 29653 + }, + { + "epoch": 1.6568148168840966, + "grad_norm": 0.4194923937320709, + "learning_rate": 0.00017268041237113402, + "loss": 0.3322, + "step": 29654 + }, + { + "epoch": 1.6568706874877783, + "grad_norm": 1.241825819015503, + "learning_rate": 0.00017265239802779022, + "loss": 0.39, + "step": 29655 + }, + { + "epoch": 1.65692655809146, + "grad_norm": 0.6833893656730652, + "learning_rate": 0.00017262438368444643, + "loss": 0.5239, + "step": 29656 + }, + { + "epoch": 1.656982428695142, + "grad_norm": 0.46133604645729065, + "learning_rate": 0.00017259636934110264, + "loss": 0.4905, + "step": 29657 + }, + { + "epoch": 1.657038299298824, + "grad_norm": 0.4763346314430237, + "learning_rate": 0.00017256835499775887, + "loss": 0.3128, + "step": 29658 + }, + { + "epoch": 1.6570941699025057, + "grad_norm": 0.4272424280643463, + "learning_rate": 0.00017254034065441507, + "loss": 0.3314, + "step": 29659 + }, + { + "epoch": 1.6571500405061876, + "grad_norm": 0.8696469664573669, + "learning_rate": 0.00017251232631107128, + "loss": 0.5531, + "step": 29660 + }, + { + "epoch": 1.6572059111098696, + "grad_norm": 0.4014219343662262, + "learning_rate": 0.00017248431196772749, + "loss": 0.385, + "step": 29661 + }, + { + "epoch": 1.6572617817135513, + "grad_norm": 0.8652154803276062, + "learning_rate": 0.0001724562976243837, + "loss": 0.3205, + "step": 29662 + }, + { + "epoch": 1.6573176523172333, + "grad_norm": 0.4658631682395935, + "learning_rate": 0.0001724282832810399, + "loss": 0.5293, + "step": 29663 + }, + { + "epoch": 1.6573735229209152, + "grad_norm": 0.42468100786209106, + "learning_rate": 0.0001724002689376961, + "loss": 0.428, + "step": 29664 + }, + { + "epoch": 1.657429393524597, + "grad_norm": 0.3479073643684387, + "learning_rate": 0.00017237225459435234, + "loss": 0.3368, + "step": 29665 + }, + { + "epoch": 1.657485264128279, + "grad_norm": 0.4618076980113983, + "learning_rate": 0.00017234424025100854, + "loss": 0.4075, + "step": 29666 + }, + { + "epoch": 1.6575411347319609, + "grad_norm": 1.4602408409118652, + "learning_rate": 0.00017231622590766472, + "loss": 0.4489, + "step": 29667 + }, + { + "epoch": 1.6575970053356426, + "grad_norm": 0.3412254750728607, + "learning_rate": 0.00017228821156432093, + "loss": 0.2853, + "step": 29668 + }, + { + "epoch": 1.6576528759393245, + "grad_norm": 0.44837772846221924, + "learning_rate": 0.00017226019722097713, + "loss": 0.3682, + "step": 29669 + }, + { + "epoch": 1.6577087465430065, + "grad_norm": 0.43669605255126953, + "learning_rate": 0.00017223218287763334, + "loss": 0.3421, + "step": 29670 + }, + { + "epoch": 1.6577646171466882, + "grad_norm": 7.586999893188477, + "learning_rate": 0.00017220416853428954, + "loss": 0.385, + "step": 29671 + }, + { + "epoch": 1.6578204877503702, + "grad_norm": 0.4427558481693268, + "learning_rate": 0.00017217615419094575, + "loss": 0.2671, + "step": 29672 + }, + { + "epoch": 1.6578763583540521, + "grad_norm": 0.3591931164264679, + "learning_rate": 0.00017214813984760198, + "loss": 0.3984, + "step": 29673 + }, + { + "epoch": 1.6579322289577338, + "grad_norm": 0.4666741192340851, + "learning_rate": 0.0001721201255042582, + "loss": 0.5011, + "step": 29674 + }, + { + "epoch": 1.6579880995614158, + "grad_norm": 0.4040425419807434, + "learning_rate": 0.0001720921111609144, + "loss": 0.448, + "step": 29675 + }, + { + "epoch": 1.6580439701650977, + "grad_norm": Infinity, + "learning_rate": 0.0001720921111609144, + "loss": 0.4171, + "step": 29676 + }, + { + "epoch": 1.6580998407687795, + "grad_norm": 0.519250750541687, + "learning_rate": 0.0001720640968175706, + "loss": 0.5526, + "step": 29677 + }, + { + "epoch": 1.6581557113724614, + "grad_norm": 0.3171892762184143, + "learning_rate": 0.0001720360824742268, + "loss": 0.376, + "step": 29678 + }, + { + "epoch": 1.6582115819761434, + "grad_norm": 0.5358031392097473, + "learning_rate": 0.000172008068130883, + "loss": 0.5147, + "step": 29679 + }, + { + "epoch": 1.658267452579825, + "grad_norm": 0.41859114170074463, + "learning_rate": 0.00017198005378753922, + "loss": 0.4685, + "step": 29680 + }, + { + "epoch": 1.6583233231835068, + "grad_norm": 0.6724753975868225, + "learning_rate": 0.00017195203944419542, + "loss": 0.4889, + "step": 29681 + }, + { + "epoch": 1.658379193787189, + "grad_norm": 0.5223425030708313, + "learning_rate": 0.00017192402510085166, + "loss": 0.388, + "step": 29682 + }, + { + "epoch": 1.6584350643908707, + "grad_norm": 0.37334591150283813, + "learning_rate": 0.00017189601075750786, + "loss": 0.3887, + "step": 29683 + }, + { + "epoch": 1.6584909349945525, + "grad_norm": 0.5400261878967285, + "learning_rate": 0.00017186799641416407, + "loss": 0.4941, + "step": 29684 + }, + { + "epoch": 1.6585468055982346, + "grad_norm": 1.331895351409912, + "learning_rate": 0.00017183998207082027, + "loss": 0.4439, + "step": 29685 + }, + { + "epoch": 1.6586026762019164, + "grad_norm": 0.33494430780410767, + "learning_rate": 0.00017181196772747648, + "loss": 0.3336, + "step": 29686 + }, + { + "epoch": 1.658658546805598, + "grad_norm": 0.4340001046657562, + "learning_rate": 0.00017178395338413269, + "loss": 0.3958, + "step": 29687 + }, + { + "epoch": 1.6587144174092803, + "grad_norm": 0.38525089621543884, + "learning_rate": 0.0001717559390407889, + "loss": 0.3578, + "step": 29688 + }, + { + "epoch": 1.658770288012962, + "grad_norm": 0.43614518642425537, + "learning_rate": 0.0001717279246974451, + "loss": 0.5358, + "step": 29689 + }, + { + "epoch": 1.6588261586166437, + "grad_norm": 0.43653586506843567, + "learning_rate": 0.0001716999103541013, + "loss": 0.3582, + "step": 29690 + }, + { + "epoch": 1.6588820292203257, + "grad_norm": 0.32407665252685547, + "learning_rate": 0.0001716718960107575, + "loss": 0.4101, + "step": 29691 + }, + { + "epoch": 1.6589378998240076, + "grad_norm": 0.32920271158218384, + "learning_rate": 0.00017164388166741372, + "loss": 0.4107, + "step": 29692 + }, + { + "epoch": 1.6589937704276894, + "grad_norm": 2.0672807693481445, + "learning_rate": 0.00017161586732406992, + "loss": 0.4946, + "step": 29693 + }, + { + "epoch": 1.6590496410313713, + "grad_norm": 0.32498669624328613, + "learning_rate": 0.00017158785298072613, + "loss": 0.3865, + "step": 29694 + }, + { + "epoch": 1.6591055116350533, + "grad_norm": 0.676099419593811, + "learning_rate": 0.00017155983863738233, + "loss": 0.4973, + "step": 29695 + }, + { + "epoch": 1.659161382238735, + "grad_norm": 0.795201301574707, + "learning_rate": 0.00017153182429403854, + "loss": 0.388, + "step": 29696 + }, + { + "epoch": 1.659217252842417, + "grad_norm": 0.49018239974975586, + "learning_rate": 0.00017150380995069474, + "loss": 0.5289, + "step": 29697 + }, + { + "epoch": 1.659273123446099, + "grad_norm": 0.6014469861984253, + "learning_rate": 0.00017147579560735098, + "loss": 0.4669, + "step": 29698 + }, + { + "epoch": 1.6593289940497806, + "grad_norm": 0.4372512698173523, + "learning_rate": 0.00017144778126400718, + "loss": 0.465, + "step": 29699 + }, + { + "epoch": 1.6593848646534626, + "grad_norm": 0.40268072485923767, + "learning_rate": 0.0001714197669206634, + "loss": 0.4029, + "step": 29700 + }, + { + "epoch": 1.6594407352571445, + "grad_norm": 1.1268835067749023, + "learning_rate": 0.0001713917525773196, + "loss": 0.373, + "step": 29701 + }, + { + "epoch": 1.6594966058608263, + "grad_norm": 0.7779936194419861, + "learning_rate": 0.0001713637382339758, + "loss": 0.5446, + "step": 29702 + }, + { + "epoch": 1.6595524764645082, + "grad_norm": 0.38646844029426575, + "learning_rate": 0.000171335723890632, + "loss": 0.3826, + "step": 29703 + }, + { + "epoch": 1.6596083470681902, + "grad_norm": 0.38101986050605774, + "learning_rate": 0.0001713077095472882, + "loss": 0.3549, + "step": 29704 + }, + { + "epoch": 1.6596642176718719, + "grad_norm": 0.377729207277298, + "learning_rate": 0.00017127969520394442, + "loss": 0.2486, + "step": 29705 + }, + { + "epoch": 1.6597200882755538, + "grad_norm": 0.7993308901786804, + "learning_rate": 0.00017125168086060065, + "loss": 0.4366, + "step": 29706 + }, + { + "epoch": 1.6597759588792358, + "grad_norm": 0.3403959572315216, + "learning_rate": 0.00017122366651725686, + "loss": 0.433, + "step": 29707 + }, + { + "epoch": 1.6598318294829175, + "grad_norm": 0.4236249029636383, + "learning_rate": 0.00017119565217391306, + "loss": 0.5415, + "step": 29708 + }, + { + "epoch": 1.6598877000865995, + "grad_norm": 0.31745433807373047, + "learning_rate": 0.00017116763783056927, + "loss": 0.3388, + "step": 29709 + }, + { + "epoch": 1.6599435706902814, + "grad_norm": 0.35935965180397034, + "learning_rate": 0.00017113962348722547, + "loss": 0.4459, + "step": 29710 + }, + { + "epoch": 1.6599994412939632, + "grad_norm": 5.64904260635376, + "learning_rate": 0.00017111160914388165, + "loss": 0.4276, + "step": 29711 + }, + { + "epoch": 1.660055311897645, + "grad_norm": 1.6740602254867554, + "learning_rate": 0.00017108359480053786, + "loss": 0.384, + "step": 29712 + }, + { + "epoch": 1.660111182501327, + "grad_norm": 0.3830045759677887, + "learning_rate": 0.00017105558045719406, + "loss": 0.3892, + "step": 29713 + }, + { + "epoch": 1.6601670531050088, + "grad_norm": 0.451186865568161, + "learning_rate": 0.0001710275661138503, + "loss": 0.4212, + "step": 29714 + }, + { + "epoch": 1.6602229237086905, + "grad_norm": 0.34664157032966614, + "learning_rate": 0.0001709995517705065, + "loss": 0.4421, + "step": 29715 + }, + { + "epoch": 1.6602787943123727, + "grad_norm": 2.9181747436523438, + "learning_rate": 0.0001709715374271627, + "loss": 0.3646, + "step": 29716 + }, + { + "epoch": 1.6603346649160544, + "grad_norm": 0.3887886703014374, + "learning_rate": 0.00017094352308381891, + "loss": 0.4941, + "step": 29717 + }, + { + "epoch": 1.6603905355197361, + "grad_norm": 2.60402774810791, + "learning_rate": 0.00017091550874047512, + "loss": 0.4104, + "step": 29718 + }, + { + "epoch": 1.6604464061234183, + "grad_norm": 0.2936863899230957, + "learning_rate": 0.00017088749439713133, + "loss": 0.3425, + "step": 29719 + }, + { + "epoch": 1.6605022767271, + "grad_norm": 0.6493451595306396, + "learning_rate": 0.00017085948005378753, + "loss": 0.376, + "step": 29720 + }, + { + "epoch": 1.6605581473307818, + "grad_norm": 0.4209655225276947, + "learning_rate": 0.00017083146571044374, + "loss": 0.3505, + "step": 29721 + }, + { + "epoch": 1.6606140179344637, + "grad_norm": 0.2878193259239197, + "learning_rate": 0.00017080345136709997, + "loss": 0.3521, + "step": 29722 + }, + { + "epoch": 1.6606698885381457, + "grad_norm": 0.3682857155799866, + "learning_rate": 0.00017077543702375618, + "loss": 0.3884, + "step": 29723 + }, + { + "epoch": 1.6607257591418274, + "grad_norm": 0.3572355806827545, + "learning_rate": 0.00017074742268041238, + "loss": 0.3377, + "step": 29724 + }, + { + "epoch": 1.6607816297455094, + "grad_norm": 0.8825226426124573, + "learning_rate": 0.0001707194083370686, + "loss": 0.5631, + "step": 29725 + }, + { + "epoch": 1.6608375003491913, + "grad_norm": 0.46770256757736206, + "learning_rate": 0.0001706913939937248, + "loss": 0.3149, + "step": 29726 + }, + { + "epoch": 1.660893370952873, + "grad_norm": 4.173201084136963, + "learning_rate": 0.000170663379650381, + "loss": 0.4361, + "step": 29727 + }, + { + "epoch": 1.660949241556555, + "grad_norm": 0.3088817596435547, + "learning_rate": 0.0001706353653070372, + "loss": 0.3494, + "step": 29728 + }, + { + "epoch": 1.661005112160237, + "grad_norm": 0.3188847303390503, + "learning_rate": 0.0001706073509636934, + "loss": 0.3397, + "step": 29729 + }, + { + "epoch": 1.6610609827639187, + "grad_norm": 0.3601762354373932, + "learning_rate": 0.00017057933662034964, + "loss": 0.4567, + "step": 29730 + }, + { + "epoch": 1.6611168533676006, + "grad_norm": 0.5876928567886353, + "learning_rate": 0.00017055132227700585, + "loss": 0.3916, + "step": 29731 + }, + { + "epoch": 1.6611727239712826, + "grad_norm": 0.33451613783836365, + "learning_rate": 0.00017052330793366203, + "loss": 0.3421, + "step": 29732 + }, + { + "epoch": 1.6612285945749643, + "grad_norm": 0.5814723968505859, + "learning_rate": 0.00017049529359031823, + "loss": 0.4907, + "step": 29733 + }, + { + "epoch": 1.6612844651786463, + "grad_norm": 0.44731688499450684, + "learning_rate": 0.00017046727924697444, + "loss": 0.3956, + "step": 29734 + }, + { + "epoch": 1.6613403357823282, + "grad_norm": 1.4553395509719849, + "learning_rate": 0.00017043926490363065, + "loss": 0.4752, + "step": 29735 + }, + { + "epoch": 1.66139620638601, + "grad_norm": 0.4736156463623047, + "learning_rate": 0.00017041125056028685, + "loss": 0.4282, + "step": 29736 + }, + { + "epoch": 1.6614520769896919, + "grad_norm": 0.3612074553966522, + "learning_rate": 0.00017038323621694306, + "loss": 0.4775, + "step": 29737 + }, + { + "epoch": 1.6615079475933738, + "grad_norm": 0.5161358118057251, + "learning_rate": 0.0001703552218735993, + "loss": 0.3675, + "step": 29738 + }, + { + "epoch": 1.6615638181970556, + "grad_norm": 0.5883616805076599, + "learning_rate": 0.0001703272075302555, + "loss": 0.3877, + "step": 29739 + }, + { + "epoch": 1.6616196888007375, + "grad_norm": 0.5061833262443542, + "learning_rate": 0.0001702991931869117, + "loss": 0.4252, + "step": 29740 + }, + { + "epoch": 1.6616755594044195, + "grad_norm": 0.45306697487831116, + "learning_rate": 0.0001702711788435679, + "loss": 0.4119, + "step": 29741 + }, + { + "epoch": 1.6617314300081012, + "grad_norm": 0.401340126991272, + "learning_rate": 0.00017024316450022411, + "loss": 0.3649, + "step": 29742 + }, + { + "epoch": 1.6617873006117831, + "grad_norm": 0.4948071241378784, + "learning_rate": 0.00017021515015688032, + "loss": 0.4333, + "step": 29743 + }, + { + "epoch": 1.661843171215465, + "grad_norm": 1.3292759656906128, + "learning_rate": 0.00017018713581353653, + "loss": 0.5442, + "step": 29744 + }, + { + "epoch": 1.6618990418191468, + "grad_norm": 0.41872087121009827, + "learning_rate": 0.00017015912147019276, + "loss": 0.3367, + "step": 29745 + }, + { + "epoch": 1.6619549124228288, + "grad_norm": 0.489482581615448, + "learning_rate": 0.00017013110712684896, + "loss": 0.5132, + "step": 29746 + }, + { + "epoch": 1.6620107830265107, + "grad_norm": 0.39306581020355225, + "learning_rate": 0.00017010309278350517, + "loss": 0.419, + "step": 29747 + }, + { + "epoch": 1.6620666536301925, + "grad_norm": 2.2455837726593018, + "learning_rate": 0.00017007507844016138, + "loss": 0.4114, + "step": 29748 + }, + { + "epoch": 1.6621225242338742, + "grad_norm": 0.5352036356925964, + "learning_rate": 0.00017004706409681758, + "loss": 0.3738, + "step": 29749 + }, + { + "epoch": 1.6621783948375564, + "grad_norm": 0.4298640787601471, + "learning_rate": 0.0001700190497534738, + "loss": 0.4078, + "step": 29750 + }, + { + "epoch": 1.662234265441238, + "grad_norm": 0.3225787878036499, + "learning_rate": 0.00016999103541013, + "loss": 0.3518, + "step": 29751 + }, + { + "epoch": 1.6622901360449198, + "grad_norm": 0.6485644578933716, + "learning_rate": 0.0001699630210667862, + "loss": 0.336, + "step": 29752 + }, + { + "epoch": 1.662346006648602, + "grad_norm": 0.6428395509719849, + "learning_rate": 0.0001699350067234424, + "loss": 0.4063, + "step": 29753 + }, + { + "epoch": 1.6624018772522837, + "grad_norm": 1.0260977745056152, + "learning_rate": 0.0001699069923800986, + "loss": 0.4802, + "step": 29754 + }, + { + "epoch": 1.6624577478559655, + "grad_norm": 0.5294249653816223, + "learning_rate": 0.00016987897803675482, + "loss": 0.5927, + "step": 29755 + }, + { + "epoch": 1.6625136184596474, + "grad_norm": 0.46583372354507446, + "learning_rate": 0.00016985096369341102, + "loss": 0.4781, + "step": 29756 + }, + { + "epoch": 1.6625694890633294, + "grad_norm": 0.621117115020752, + "learning_rate": 0.00016982294935006723, + "loss": 0.6322, + "step": 29757 + }, + { + "epoch": 1.662625359667011, + "grad_norm": 0.4945772886276245, + "learning_rate": 0.00016979493500672343, + "loss": 0.359, + "step": 29758 + }, + { + "epoch": 1.662681230270693, + "grad_norm": 1.0559890270233154, + "learning_rate": 0.00016976692066337964, + "loss": 0.4808, + "step": 29759 + }, + { + "epoch": 1.662737100874375, + "grad_norm": 0.34783807396888733, + "learning_rate": 0.00016973890632003585, + "loss": 0.3996, + "step": 29760 + }, + { + "epoch": 1.6627929714780567, + "grad_norm": 0.38074836134910583, + "learning_rate": 0.00016971089197669208, + "loss": 0.3426, + "step": 29761 + }, + { + "epoch": 1.6628488420817387, + "grad_norm": 2.2297723293304443, + "learning_rate": 0.00016968287763334828, + "loss": 0.4492, + "step": 29762 + }, + { + "epoch": 1.6629047126854206, + "grad_norm": 0.6494765281677246, + "learning_rate": 0.0001696548632900045, + "loss": 0.4275, + "step": 29763 + }, + { + "epoch": 1.6629605832891023, + "grad_norm": 5.295227527618408, + "learning_rate": 0.0001696268489466607, + "loss": 0.3451, + "step": 29764 + }, + { + "epoch": 1.6630164538927843, + "grad_norm": 0.3377383053302765, + "learning_rate": 0.0001695988346033169, + "loss": 0.2871, + "step": 29765 + }, + { + "epoch": 1.6630723244964662, + "grad_norm": 0.38544559478759766, + "learning_rate": 0.0001695708202599731, + "loss": 0.4139, + "step": 29766 + }, + { + "epoch": 1.663128195100148, + "grad_norm": 0.6502152681350708, + "learning_rate": 0.00016954280591662931, + "loss": 0.412, + "step": 29767 + }, + { + "epoch": 1.66318406570383, + "grad_norm": 0.5054818987846375, + "learning_rate": 0.00016951479157328552, + "loss": 0.5691, + "step": 29768 + }, + { + "epoch": 1.6632399363075119, + "grad_norm": 3.2096383571624756, + "learning_rate": 0.00016948677722994175, + "loss": 0.4425, + "step": 29769 + }, + { + "epoch": 1.6632958069111936, + "grad_norm": 0.3663831055164337, + "learning_rate": 0.00016945876288659796, + "loss": 0.4525, + "step": 29770 + }, + { + "epoch": 1.6633516775148756, + "grad_norm": 0.8327823877334595, + "learning_rate": 0.00016943074854325416, + "loss": 0.4091, + "step": 29771 + }, + { + "epoch": 1.6634075481185575, + "grad_norm": 0.5940095782279968, + "learning_rate": 0.00016940273419991037, + "loss": 0.3881, + "step": 29772 + }, + { + "epoch": 1.6634634187222392, + "grad_norm": 0.3357160985469818, + "learning_rate": 0.00016937471985656658, + "loss": 0.4036, + "step": 29773 + }, + { + "epoch": 1.6635192893259212, + "grad_norm": 0.5264624953269958, + "learning_rate": 0.00016934670551322278, + "loss": 0.4851, + "step": 29774 + }, + { + "epoch": 1.6635751599296031, + "grad_norm": 1.9073652029037476, + "learning_rate": 0.00016931869116987896, + "loss": 0.4266, + "step": 29775 + }, + { + "epoch": 1.6636310305332849, + "grad_norm": 1.0733391046524048, + "learning_rate": 0.00016929067682653517, + "loss": 0.4448, + "step": 29776 + }, + { + "epoch": 1.6636869011369668, + "grad_norm": 0.654220461845398, + "learning_rate": 0.0001692626624831914, + "loss": 0.4036, + "step": 29777 + }, + { + "epoch": 1.6637427717406488, + "grad_norm": 0.4923449456691742, + "learning_rate": 0.0001692346481398476, + "loss": 0.3118, + "step": 29778 + }, + { + "epoch": 1.6637986423443305, + "grad_norm": 0.7525801658630371, + "learning_rate": 0.0001692066337965038, + "loss": 0.4376, + "step": 29779 + }, + { + "epoch": 1.6638545129480125, + "grad_norm": 0.3487260043621063, + "learning_rate": 0.00016917861945316002, + "loss": 0.4682, + "step": 29780 + }, + { + "epoch": 1.6639103835516944, + "grad_norm": 0.46829113364219666, + "learning_rate": 0.00016915060510981622, + "loss": 0.3907, + "step": 29781 + }, + { + "epoch": 1.6639662541553761, + "grad_norm": 0.3122499883174896, + "learning_rate": 0.00016912259076647243, + "loss": 0.4361, + "step": 29782 + }, + { + "epoch": 1.6640221247590579, + "grad_norm": 1.9126548767089844, + "learning_rate": 0.00016909457642312863, + "loss": 0.3005, + "step": 29783 + }, + { + "epoch": 1.66407799536274, + "grad_norm": 0.585543155670166, + "learning_rate": 0.00016906656207978484, + "loss": 0.561, + "step": 29784 + }, + { + "epoch": 1.6641338659664218, + "grad_norm": 0.7441182732582092, + "learning_rate": 0.00016903854773644107, + "loss": 0.394, + "step": 29785 + }, + { + "epoch": 1.6641897365701035, + "grad_norm": 0.6624036431312561, + "learning_rate": 0.00016901053339309728, + "loss": 0.3327, + "step": 29786 + }, + { + "epoch": 1.6642456071737857, + "grad_norm": 0.8063470125198364, + "learning_rate": 0.00016898251904975348, + "loss": 0.4762, + "step": 29787 + }, + { + "epoch": 1.6643014777774674, + "grad_norm": 0.37058350443840027, + "learning_rate": 0.0001689545047064097, + "loss": 0.5399, + "step": 29788 + }, + { + "epoch": 1.6643573483811491, + "grad_norm": 0.4138842225074768, + "learning_rate": 0.0001689264903630659, + "loss": 0.4227, + "step": 29789 + }, + { + "epoch": 1.664413218984831, + "grad_norm": 0.44484594464302063, + "learning_rate": 0.0001688984760197221, + "loss": 0.4767, + "step": 29790 + }, + { + "epoch": 1.664469089588513, + "grad_norm": 0.7208504676818848, + "learning_rate": 0.0001688704616763783, + "loss": 0.3958, + "step": 29791 + }, + { + "epoch": 1.6645249601921948, + "grad_norm": 0.360935777425766, + "learning_rate": 0.0001688424473330345, + "loss": 0.3907, + "step": 29792 + }, + { + "epoch": 1.6645808307958767, + "grad_norm": 0.36392712593078613, + "learning_rate": 0.00016881443298969075, + "loss": 0.3929, + "step": 29793 + }, + { + "epoch": 1.6646367013995587, + "grad_norm": 0.5816735625267029, + "learning_rate": 0.00016878641864634695, + "loss": 0.413, + "step": 29794 + }, + { + "epoch": 1.6646925720032404, + "grad_norm": 0.40097108483314514, + "learning_rate": 0.00016875840430300316, + "loss": 0.4231, + "step": 29795 + }, + { + "epoch": 1.6647484426069223, + "grad_norm": 0.48747989535331726, + "learning_rate": 0.00016873038995965934, + "loss": 0.3613, + "step": 29796 + }, + { + "epoch": 1.6648043132106043, + "grad_norm": 0.4961816668510437, + "learning_rate": 0.00016870237561631554, + "loss": 0.5623, + "step": 29797 + }, + { + "epoch": 1.664860183814286, + "grad_norm": 0.4412253499031067, + "learning_rate": 0.00016867436127297175, + "loss": 0.4377, + "step": 29798 + }, + { + "epoch": 1.664916054417968, + "grad_norm": 0.4648142457008362, + "learning_rate": 0.00016864634692962795, + "loss": 0.4358, + "step": 29799 + }, + { + "epoch": 1.66497192502165, + "grad_norm": 0.39813944697380066, + "learning_rate": 0.00016861833258628416, + "loss": 0.4135, + "step": 29800 + }, + { + "epoch": 1.6650277956253317, + "grad_norm": 0.9384295344352722, + "learning_rate": 0.0001685903182429404, + "loss": 0.4281, + "step": 29801 + }, + { + "epoch": 1.6650836662290136, + "grad_norm": 0.2779136598110199, + "learning_rate": 0.0001685623038995966, + "loss": 0.3206, + "step": 29802 + }, + { + "epoch": 1.6651395368326956, + "grad_norm": 0.3116244077682495, + "learning_rate": 0.0001685342895562528, + "loss": 0.3483, + "step": 29803 + }, + { + "epoch": 1.6651954074363773, + "grad_norm": 0.7258368730545044, + "learning_rate": 0.000168506275212909, + "loss": 0.3875, + "step": 29804 + }, + { + "epoch": 1.6652512780400592, + "grad_norm": 3.1863796710968018, + "learning_rate": 0.00016847826086956522, + "loss": 0.4575, + "step": 29805 + }, + { + "epoch": 1.6653071486437412, + "grad_norm": 0.37752005457878113, + "learning_rate": 0.00016845024652622142, + "loss": 0.4627, + "step": 29806 + }, + { + "epoch": 1.665363019247423, + "grad_norm": 0.44098466634750366, + "learning_rate": 0.00016842223218287763, + "loss": 0.4233, + "step": 29807 + }, + { + "epoch": 1.6654188898511049, + "grad_norm": 0.3915548026561737, + "learning_rate": 0.00016839421783953386, + "loss": 0.4069, + "step": 29808 + }, + { + "epoch": 1.6654747604547868, + "grad_norm": 0.5226084589958191, + "learning_rate": 0.00016836620349619007, + "loss": 0.6639, + "step": 29809 + }, + { + "epoch": 1.6655306310584685, + "grad_norm": 1.6725003719329834, + "learning_rate": 0.00016833818915284627, + "loss": 0.4074, + "step": 29810 + }, + { + "epoch": 1.6655865016621505, + "grad_norm": 0.3920358419418335, + "learning_rate": 0.00016831017480950248, + "loss": 0.3411, + "step": 29811 + }, + { + "epoch": 1.6656423722658324, + "grad_norm": 0.43541935086250305, + "learning_rate": 0.00016828216046615868, + "loss": 0.419, + "step": 29812 + }, + { + "epoch": 1.6656982428695142, + "grad_norm": 0.3648369312286377, + "learning_rate": 0.0001682541461228149, + "loss": 0.4302, + "step": 29813 + }, + { + "epoch": 1.6657541134731961, + "grad_norm": 0.4065341055393219, + "learning_rate": 0.0001682261317794711, + "loss": 0.3868, + "step": 29814 + }, + { + "epoch": 1.665809984076878, + "grad_norm": 0.44095566868782043, + "learning_rate": 0.0001681981174361273, + "loss": 0.4625, + "step": 29815 + }, + { + "epoch": 1.6658658546805598, + "grad_norm": 0.3592407703399658, + "learning_rate": 0.00016817010309278353, + "loss": 0.3322, + "step": 29816 + }, + { + "epoch": 1.6659217252842415, + "grad_norm": 0.2679884731769562, + "learning_rate": 0.0001681420887494397, + "loss": 0.3759, + "step": 29817 + }, + { + "epoch": 1.6659775958879237, + "grad_norm": 0.49301132559776306, + "learning_rate": 0.00016811407440609592, + "loss": 0.4929, + "step": 29818 + }, + { + "epoch": 1.6660334664916054, + "grad_norm": 0.36028581857681274, + "learning_rate": 0.00016808606006275212, + "loss": 0.3977, + "step": 29819 + }, + { + "epoch": 1.6660893370952872, + "grad_norm": 0.39861491322517395, + "learning_rate": 0.00016805804571940833, + "loss": 0.421, + "step": 29820 + }, + { + "epoch": 1.6661452076989693, + "grad_norm": 1.316748857498169, + "learning_rate": 0.00016803003137606454, + "loss": 0.5688, + "step": 29821 + }, + { + "epoch": 1.666201078302651, + "grad_norm": 0.5234755873680115, + "learning_rate": 0.00016800201703272074, + "loss": 0.4446, + "step": 29822 + }, + { + "epoch": 1.6662569489063328, + "grad_norm": 0.620009183883667, + "learning_rate": 0.00016797400268937695, + "loss": 0.3873, + "step": 29823 + }, + { + "epoch": 1.6663128195100148, + "grad_norm": 0.3222336173057556, + "learning_rate": 0.00016794598834603318, + "loss": 0.3835, + "step": 29824 + }, + { + "epoch": 1.6663686901136967, + "grad_norm": 0.4034559428691864, + "learning_rate": 0.0001679179740026894, + "loss": 0.4171, + "step": 29825 + }, + { + "epoch": 1.6664245607173784, + "grad_norm": 0.6009373664855957, + "learning_rate": 0.0001678899596593456, + "loss": 0.5603, + "step": 29826 + }, + { + "epoch": 1.6664804313210604, + "grad_norm": 0.5236926674842834, + "learning_rate": 0.0001678619453160018, + "loss": 0.4495, + "step": 29827 + }, + { + "epoch": 1.6665363019247423, + "grad_norm": 5.308396339416504, + "learning_rate": 0.000167833930972658, + "loss": 0.3502, + "step": 29828 + }, + { + "epoch": 1.666592172528424, + "grad_norm": 0.40253135561943054, + "learning_rate": 0.0001678059166293142, + "loss": 0.4546, + "step": 29829 + }, + { + "epoch": 1.666648043132106, + "grad_norm": 0.5771504044532776, + "learning_rate": 0.00016777790228597042, + "loss": 0.4511, + "step": 29830 + }, + { + "epoch": 1.666703913735788, + "grad_norm": 0.6295286417007446, + "learning_rate": 0.00016774988794262662, + "loss": 0.4307, + "step": 29831 + }, + { + "epoch": 1.6667597843394697, + "grad_norm": 0.4901268184185028, + "learning_rate": 0.00016772187359928285, + "loss": 0.4214, + "step": 29832 + }, + { + "epoch": 1.6668156549431516, + "grad_norm": 0.5727595090866089, + "learning_rate": 0.00016769385925593906, + "loss": 0.6018, + "step": 29833 + }, + { + "epoch": 1.6668715255468336, + "grad_norm": 0.36286449432373047, + "learning_rate": 0.00016766584491259527, + "loss": 0.3643, + "step": 29834 + }, + { + "epoch": 1.6669273961505153, + "grad_norm": 0.3921189308166504, + "learning_rate": 0.00016763783056925147, + "loss": 0.3934, + "step": 29835 + }, + { + "epoch": 1.6669832667541973, + "grad_norm": 3.444425106048584, + "learning_rate": 0.00016760981622590768, + "loss": 0.3341, + "step": 29836 + }, + { + "epoch": 1.6670391373578792, + "grad_norm": 0.5778743028640747, + "learning_rate": 0.00016758180188256388, + "loss": 0.4377, + "step": 29837 + }, + { + "epoch": 1.667095007961561, + "grad_norm": 0.9298853874206543, + "learning_rate": 0.00016755378753922006, + "loss": 0.4488, + "step": 29838 + }, + { + "epoch": 1.667150878565243, + "grad_norm": 0.4583759307861328, + "learning_rate": 0.00016752577319587627, + "loss": 0.3342, + "step": 29839 + }, + { + "epoch": 1.6672067491689249, + "grad_norm": 1.4509589672088623, + "learning_rate": 0.0001674977588525325, + "loss": 0.4022, + "step": 29840 + }, + { + "epoch": 1.6672626197726066, + "grad_norm": 0.41632694005966187, + "learning_rate": 0.0001674697445091887, + "loss": 0.398, + "step": 29841 + }, + { + "epoch": 1.6673184903762885, + "grad_norm": 0.4170199930667877, + "learning_rate": 0.0001674417301658449, + "loss": 0.582, + "step": 29842 + }, + { + "epoch": 1.6673743609799705, + "grad_norm": 0.4380181133747101, + "learning_rate": 0.00016741371582250112, + "loss": 0.4881, + "step": 29843 + }, + { + "epoch": 1.6674302315836522, + "grad_norm": 0.5100034475326538, + "learning_rate": 0.00016738570147915732, + "loss": 0.5065, + "step": 29844 + }, + { + "epoch": 1.6674861021873342, + "grad_norm": 0.39115938544273376, + "learning_rate": 0.00016735768713581353, + "loss": 0.4167, + "step": 29845 + }, + { + "epoch": 1.6675419727910161, + "grad_norm": 0.32250288128852844, + "learning_rate": 0.00016732967279246974, + "loss": 0.3309, + "step": 29846 + }, + { + "epoch": 1.6675978433946979, + "grad_norm": 0.4725017249584198, + "learning_rate": 0.00016730165844912594, + "loss": 0.4654, + "step": 29847 + }, + { + "epoch": 1.6676537139983798, + "grad_norm": 0.5079727172851562, + "learning_rate": 0.00016727364410578217, + "loss": 0.3893, + "step": 29848 + }, + { + "epoch": 1.6677095846020618, + "grad_norm": 0.6738384962081909, + "learning_rate": 0.00016724562976243838, + "loss": 0.4372, + "step": 29849 + }, + { + "epoch": 1.6677654552057435, + "grad_norm": 0.3765528202056885, + "learning_rate": 0.00016721761541909459, + "loss": 0.4768, + "step": 29850 + }, + { + "epoch": 1.6678213258094252, + "grad_norm": 0.35573437809944153, + "learning_rate": 0.0001671896010757508, + "loss": 0.3978, + "step": 29851 + }, + { + "epoch": 1.6678771964131074, + "grad_norm": 0.43889376521110535, + "learning_rate": 0.000167161586732407, + "loss": 0.4822, + "step": 29852 + }, + { + "epoch": 1.6679330670167891, + "grad_norm": 0.8612982034683228, + "learning_rate": 0.0001671335723890632, + "loss": 0.4936, + "step": 29853 + }, + { + "epoch": 1.6679889376204708, + "grad_norm": 20.467693328857422, + "learning_rate": 0.0001671055580457194, + "loss": 0.3623, + "step": 29854 + }, + { + "epoch": 1.668044808224153, + "grad_norm": 0.2838985323905945, + "learning_rate": 0.00016707754370237562, + "loss": 0.3159, + "step": 29855 + }, + { + "epoch": 1.6681006788278347, + "grad_norm": 1.2278333902359009, + "learning_rate": 0.00016704952935903185, + "loss": 0.5118, + "step": 29856 + }, + { + "epoch": 1.6681565494315165, + "grad_norm": 0.3853053152561188, + "learning_rate": 0.00016702151501568805, + "loss": 0.5256, + "step": 29857 + }, + { + "epoch": 1.6682124200351984, + "grad_norm": 0.4127325415611267, + "learning_rate": 0.00016699350067234426, + "loss": 0.4738, + "step": 29858 + }, + { + "epoch": 1.6682682906388804, + "grad_norm": 0.46460580825805664, + "learning_rate": 0.00016696548632900047, + "loss": 0.4811, + "step": 29859 + }, + { + "epoch": 1.668324161242562, + "grad_norm": 1.7564409971237183, + "learning_rate": 0.00016693747198565664, + "loss": 0.4166, + "step": 29860 + }, + { + "epoch": 1.668380031846244, + "grad_norm": 0.3510027527809143, + "learning_rate": 0.00016690945764231285, + "loss": 0.4398, + "step": 29861 + }, + { + "epoch": 1.668435902449926, + "grad_norm": 0.46118468046188354, + "learning_rate": 0.00016688144329896906, + "loss": 0.4134, + "step": 29862 + }, + { + "epoch": 1.6684917730536077, + "grad_norm": 0.32564809918403625, + "learning_rate": 0.00016685342895562526, + "loss": 0.3754, + "step": 29863 + }, + { + "epoch": 1.6685476436572897, + "grad_norm": 0.42629656195640564, + "learning_rate": 0.0001668254146122815, + "loss": 0.3909, + "step": 29864 + }, + { + "epoch": 1.6686035142609716, + "grad_norm": 0.39932700991630554, + "learning_rate": 0.0001667974002689377, + "loss": 0.3825, + "step": 29865 + }, + { + "epoch": 1.6686593848646534, + "grad_norm": 0.32985901832580566, + "learning_rate": 0.0001667693859255939, + "loss": 0.4015, + "step": 29866 + }, + { + "epoch": 1.6687152554683353, + "grad_norm": 0.4276598393917084, + "learning_rate": 0.0001667413715822501, + "loss": 0.4671, + "step": 29867 + }, + { + "epoch": 1.6687711260720173, + "grad_norm": 0.3606202304363251, + "learning_rate": 0.00016671335723890632, + "loss": 0.3723, + "step": 29868 + }, + { + "epoch": 1.668826996675699, + "grad_norm": 0.34828808903694153, + "learning_rate": 0.00016668534289556252, + "loss": 0.3101, + "step": 29869 + }, + { + "epoch": 1.668882867279381, + "grad_norm": 0.6217843294143677, + "learning_rate": 0.00016665732855221873, + "loss": 0.3989, + "step": 29870 + }, + { + "epoch": 1.668938737883063, + "grad_norm": 0.5365795493125916, + "learning_rate": 0.00016662931420887496, + "loss": 0.4647, + "step": 29871 + }, + { + "epoch": 1.6689946084867446, + "grad_norm": 0.3779529631137848, + "learning_rate": 0.00016660129986553117, + "loss": 0.4643, + "step": 29872 + }, + { + "epoch": 1.6690504790904266, + "grad_norm": 0.3622815012931824, + "learning_rate": 0.00016657328552218737, + "loss": 0.3288, + "step": 29873 + }, + { + "epoch": 1.6691063496941085, + "grad_norm": 0.7964171767234802, + "learning_rate": 0.00016654527117884358, + "loss": 0.525, + "step": 29874 + }, + { + "epoch": 1.6691622202977903, + "grad_norm": 0.41272619366645813, + "learning_rate": 0.00016651725683549979, + "loss": 0.3547, + "step": 29875 + }, + { + "epoch": 1.6692180909014722, + "grad_norm": 0.803050696849823, + "learning_rate": 0.000166489242492156, + "loss": 0.3838, + "step": 29876 + }, + { + "epoch": 1.6692739615051542, + "grad_norm": 0.3083300292491913, + "learning_rate": 0.0001664612281488122, + "loss": 0.2663, + "step": 29877 + }, + { + "epoch": 1.669329832108836, + "grad_norm": 0.36086568236351013, + "learning_rate": 0.0001664332138054684, + "loss": 0.4069, + "step": 29878 + }, + { + "epoch": 1.6693857027125178, + "grad_norm": 0.3922925293445587, + "learning_rate": 0.00016640519946212464, + "loss": 0.4129, + "step": 29879 + }, + { + "epoch": 1.6694415733161998, + "grad_norm": 0.3143612742424011, + "learning_rate": 0.00016637718511878084, + "loss": 0.4567, + "step": 29880 + }, + { + "epoch": 1.6694974439198815, + "grad_norm": 1.1678028106689453, + "learning_rate": 0.00016634917077543702, + "loss": 0.4156, + "step": 29881 + }, + { + "epoch": 1.6695533145235635, + "grad_norm": 0.41272830963134766, + "learning_rate": 0.00016632115643209323, + "loss": 0.5067, + "step": 29882 + }, + { + "epoch": 1.6696091851272454, + "grad_norm": 0.3628973662853241, + "learning_rate": 0.00016629314208874943, + "loss": 0.3444, + "step": 29883 + }, + { + "epoch": 1.6696650557309272, + "grad_norm": 0.534956693649292, + "learning_rate": 0.00016626512774540564, + "loss": 0.4172, + "step": 29884 + }, + { + "epoch": 1.669720926334609, + "grad_norm": 0.4220615327358246, + "learning_rate": 0.00016623711340206184, + "loss": 0.4067, + "step": 29885 + }, + { + "epoch": 1.669776796938291, + "grad_norm": 1.4387588500976562, + "learning_rate": 0.00016620909905871805, + "loss": 0.3895, + "step": 29886 + }, + { + "epoch": 1.6698326675419728, + "grad_norm": 0.5689263343811035, + "learning_rate": 0.00016618108471537428, + "loss": 0.524, + "step": 29887 + }, + { + "epoch": 1.6698885381456545, + "grad_norm": 0.5870437026023865, + "learning_rate": 0.0001661530703720305, + "loss": 0.4437, + "step": 29888 + }, + { + "epoch": 1.6699444087493367, + "grad_norm": 0.4506867229938507, + "learning_rate": 0.0001661250560286867, + "loss": 0.4452, + "step": 29889 + }, + { + "epoch": 1.6700002793530184, + "grad_norm": 0.5265975594520569, + "learning_rate": 0.0001660970416853429, + "loss": 0.4591, + "step": 29890 + }, + { + "epoch": 1.6700561499567002, + "grad_norm": 0.428691565990448, + "learning_rate": 0.0001660690273419991, + "loss": 0.5088, + "step": 29891 + }, + { + "epoch": 1.670112020560382, + "grad_norm": 0.3318386971950531, + "learning_rate": 0.0001660410129986553, + "loss": 0.3744, + "step": 29892 + }, + { + "epoch": 1.670167891164064, + "grad_norm": 0.39207902550697327, + "learning_rate": 0.00016601299865531152, + "loss": 0.4052, + "step": 29893 + }, + { + "epoch": 1.6702237617677458, + "grad_norm": 0.35096412897109985, + "learning_rate": 0.00016598498431196772, + "loss": 0.3517, + "step": 29894 + }, + { + "epoch": 1.6702796323714277, + "grad_norm": 0.7950496673583984, + "learning_rate": 0.00016595696996862396, + "loss": 0.4729, + "step": 29895 + }, + { + "epoch": 1.6703355029751097, + "grad_norm": 0.6897532939910889, + "learning_rate": 0.00016592895562528016, + "loss": 0.3914, + "step": 29896 + }, + { + "epoch": 1.6703913735787914, + "grad_norm": 0.536783754825592, + "learning_rate": 0.00016590094128193637, + "loss": 0.6028, + "step": 29897 + }, + { + "epoch": 1.6704472441824734, + "grad_norm": 0.4886324107646942, + "learning_rate": 0.00016587292693859257, + "loss": 0.4502, + "step": 29898 + }, + { + "epoch": 1.6705031147861553, + "grad_norm": 0.8128216862678528, + "learning_rate": 0.00016584491259524878, + "loss": 0.4157, + "step": 29899 + }, + { + "epoch": 1.670558985389837, + "grad_norm": 0.3609892725944519, + "learning_rate": 0.00016581689825190499, + "loss": 0.3643, + "step": 29900 + }, + { + "epoch": 1.670614855993519, + "grad_norm": 0.39763495326042175, + "learning_rate": 0.0001657888839085612, + "loss": 0.3838, + "step": 29901 + }, + { + "epoch": 1.670670726597201, + "grad_norm": 0.4032725691795349, + "learning_rate": 0.00016576086956521737, + "loss": 0.3747, + "step": 29902 + }, + { + "epoch": 1.6707265972008827, + "grad_norm": 2.0421082973480225, + "learning_rate": 0.0001657328552218736, + "loss": 0.4078, + "step": 29903 + }, + { + "epoch": 1.6707824678045646, + "grad_norm": 0.39454197883605957, + "learning_rate": 0.0001657048408785298, + "loss": 0.478, + "step": 29904 + }, + { + "epoch": 1.6708383384082466, + "grad_norm": 3.296860933303833, + "learning_rate": 0.00016567682653518601, + "loss": 0.4405, + "step": 29905 + }, + { + "epoch": 1.6708942090119283, + "grad_norm": 0.8457250595092773, + "learning_rate": 0.00016564881219184222, + "loss": 0.3897, + "step": 29906 + }, + { + "epoch": 1.6709500796156103, + "grad_norm": 0.5366543531417847, + "learning_rate": 0.00016562079784849843, + "loss": 0.5131, + "step": 29907 + }, + { + "epoch": 1.6710059502192922, + "grad_norm": 0.321329802274704, + "learning_rate": 0.00016559278350515463, + "loss": 0.3907, + "step": 29908 + }, + { + "epoch": 1.671061820822974, + "grad_norm": 0.5575001835823059, + "learning_rate": 0.00016556476916181084, + "loss": 0.472, + "step": 29909 + }, + { + "epoch": 1.671117691426656, + "grad_norm": 0.45006683468818665, + "learning_rate": 0.00016553675481846704, + "loss": 0.3829, + "step": 29910 + }, + { + "epoch": 1.6711735620303378, + "grad_norm": 0.51767897605896, + "learning_rate": 0.00016550874047512328, + "loss": 0.5867, + "step": 29911 + }, + { + "epoch": 1.6712294326340196, + "grad_norm": 0.3953908681869507, + "learning_rate": 0.00016548072613177948, + "loss": 0.3477, + "step": 29912 + }, + { + "epoch": 1.6712853032377015, + "grad_norm": 0.47824767231941223, + "learning_rate": 0.0001654527117884357, + "loss": 0.4534, + "step": 29913 + }, + { + "epoch": 1.6713411738413835, + "grad_norm": 0.7125014662742615, + "learning_rate": 0.0001654246974450919, + "loss": 0.3761, + "step": 29914 + }, + { + "epoch": 1.6713970444450652, + "grad_norm": 0.33100396394729614, + "learning_rate": 0.0001653966831017481, + "loss": 0.3556, + "step": 29915 + }, + { + "epoch": 1.6714529150487472, + "grad_norm": 0.6456719636917114, + "learning_rate": 0.0001653686687584043, + "loss": 0.4195, + "step": 29916 + }, + { + "epoch": 1.671508785652429, + "grad_norm": 0.4848509430885315, + "learning_rate": 0.0001653406544150605, + "loss": 0.4539, + "step": 29917 + }, + { + "epoch": 1.6715646562561108, + "grad_norm": 0.36635783314704895, + "learning_rate": 0.00016531264007171672, + "loss": 0.3705, + "step": 29918 + }, + { + "epoch": 1.6716205268597926, + "grad_norm": 0.42966383695602417, + "learning_rate": 0.00016528462572837295, + "loss": 0.4155, + "step": 29919 + }, + { + "epoch": 1.6716763974634747, + "grad_norm": 0.47318190336227417, + "learning_rate": 0.00016525661138502916, + "loss": 0.4402, + "step": 29920 + }, + { + "epoch": 1.6717322680671565, + "grad_norm": 0.3200848698616028, + "learning_rate": 0.00016522859704168536, + "loss": 0.4083, + "step": 29921 + }, + { + "epoch": 1.6717881386708382, + "grad_norm": 0.44341257214546204, + "learning_rate": 0.00016520058269834157, + "loss": 0.4502, + "step": 29922 + }, + { + "epoch": 1.6718440092745204, + "grad_norm": 0.42217567563056946, + "learning_rate": 0.00016517256835499775, + "loss": 0.5168, + "step": 29923 + }, + { + "epoch": 1.671899879878202, + "grad_norm": 0.28444015979766846, + "learning_rate": 0.00016514455401165395, + "loss": 0.3804, + "step": 29924 + }, + { + "epoch": 1.6719557504818838, + "grad_norm": 0.45139336585998535, + "learning_rate": 0.00016511653966831016, + "loss": 0.4628, + "step": 29925 + }, + { + "epoch": 1.6720116210855658, + "grad_norm": 4.137104511260986, + "learning_rate": 0.00016508852532496636, + "loss": 0.522, + "step": 29926 + }, + { + "epoch": 1.6720674916892477, + "grad_norm": 0.3626900911331177, + "learning_rate": 0.0001650605109816226, + "loss": 0.4089, + "step": 29927 + }, + { + "epoch": 1.6721233622929295, + "grad_norm": 0.8155657649040222, + "learning_rate": 0.0001650324966382788, + "loss": 0.3599, + "step": 29928 + }, + { + "epoch": 1.6721792328966114, + "grad_norm": 0.49313661456108093, + "learning_rate": 0.000165004482294935, + "loss": 0.3778, + "step": 29929 + }, + { + "epoch": 1.6722351035002934, + "grad_norm": 0.4643367528915405, + "learning_rate": 0.00016497646795159121, + "loss": 0.5021, + "step": 29930 + }, + { + "epoch": 1.672290974103975, + "grad_norm": 0.33304062485694885, + "learning_rate": 0.00016494845360824742, + "loss": 0.4446, + "step": 29931 + }, + { + "epoch": 1.672346844707657, + "grad_norm": 0.4168662428855896, + "learning_rate": 0.00016492043926490363, + "loss": 0.4648, + "step": 29932 + }, + { + "epoch": 1.672402715311339, + "grad_norm": 0.7121668457984924, + "learning_rate": 0.00016489242492155983, + "loss": 0.4386, + "step": 29933 + }, + { + "epoch": 1.6724585859150207, + "grad_norm": 0.40346023440361023, + "learning_rate": 0.00016486441057821604, + "loss": 0.4045, + "step": 29934 + }, + { + "epoch": 1.6725144565187027, + "grad_norm": 0.43436598777770996, + "learning_rate": 0.00016483639623487227, + "loss": 0.379, + "step": 29935 + }, + { + "epoch": 1.6725703271223846, + "grad_norm": 0.338418185710907, + "learning_rate": 0.00016480838189152848, + "loss": 0.5292, + "step": 29936 + }, + { + "epoch": 1.6726261977260664, + "grad_norm": 0.9615235924720764, + "learning_rate": 0.00016478036754818468, + "loss": 0.3222, + "step": 29937 + }, + { + "epoch": 1.6726820683297483, + "grad_norm": 0.31537505984306335, + "learning_rate": 0.0001647523532048409, + "loss": 0.3918, + "step": 29938 + }, + { + "epoch": 1.6727379389334303, + "grad_norm": 0.3329072594642639, + "learning_rate": 0.0001647243388614971, + "loss": 0.3483, + "step": 29939 + }, + { + "epoch": 1.672793809537112, + "grad_norm": 0.3856417238712311, + "learning_rate": 0.0001646963245181533, + "loss": 0.3221, + "step": 29940 + }, + { + "epoch": 1.672849680140794, + "grad_norm": 0.6112145185470581, + "learning_rate": 0.0001646683101748095, + "loss": 0.4567, + "step": 29941 + }, + { + "epoch": 1.6729055507444759, + "grad_norm": 0.322662353515625, + "learning_rate": 0.00016464029583146574, + "loss": 0.3129, + "step": 29942 + }, + { + "epoch": 1.6729614213481576, + "grad_norm": 0.4568310081958771, + "learning_rate": 0.00016461228148812194, + "loss": 0.4642, + "step": 29943 + }, + { + "epoch": 1.6730172919518396, + "grad_norm": 0.5933864712715149, + "learning_rate": 0.00016458426714477815, + "loss": 0.4739, + "step": 29944 + }, + { + "epoch": 1.6730731625555215, + "grad_norm": 0.49865177273750305, + "learning_rate": 0.00016455625280143433, + "loss": 0.444, + "step": 29945 + }, + { + "epoch": 1.6731290331592032, + "grad_norm": 0.39923369884490967, + "learning_rate": 0.00016452823845809053, + "loss": 0.3826, + "step": 29946 + }, + { + "epoch": 1.6731849037628852, + "grad_norm": 0.715002715587616, + "learning_rate": 0.00016450022411474674, + "loss": 0.5101, + "step": 29947 + }, + { + "epoch": 1.6732407743665672, + "grad_norm": 0.33209577202796936, + "learning_rate": 0.00016447220977140295, + "loss": 0.3512, + "step": 29948 + }, + { + "epoch": 1.6732966449702489, + "grad_norm": 0.3379441201686859, + "learning_rate": 0.00016444419542805915, + "loss": 0.4617, + "step": 29949 + }, + { + "epoch": 1.6733525155739306, + "grad_norm": 1.612223505973816, + "learning_rate": 0.00016441618108471538, + "loss": 0.3909, + "step": 29950 + }, + { + "epoch": 1.6734083861776128, + "grad_norm": 0.4238894283771515, + "learning_rate": 0.0001643881667413716, + "loss": 0.4854, + "step": 29951 + }, + { + "epoch": 1.6734642567812945, + "grad_norm": 0.3589964807033539, + "learning_rate": 0.0001643601523980278, + "loss": 0.484, + "step": 29952 + }, + { + "epoch": 1.6735201273849762, + "grad_norm": 0.3458848297595978, + "learning_rate": 0.000164332138054684, + "loss": 0.3949, + "step": 29953 + }, + { + "epoch": 1.6735759979886584, + "grad_norm": 0.3170410990715027, + "learning_rate": 0.0001643041237113402, + "loss": 0.4243, + "step": 29954 + }, + { + "epoch": 1.6736318685923401, + "grad_norm": 0.4131665825843811, + "learning_rate": 0.00016427610936799641, + "loss": 0.4252, + "step": 29955 + }, + { + "epoch": 1.6736877391960219, + "grad_norm": 0.465355783700943, + "learning_rate": 0.00016424809502465262, + "loss": 0.4386, + "step": 29956 + }, + { + "epoch": 1.673743609799704, + "grad_norm": 2.652679204940796, + "learning_rate": 0.00016422008068130883, + "loss": 0.4273, + "step": 29957 + }, + { + "epoch": 1.6737994804033858, + "grad_norm": 0.2560397684574127, + "learning_rate": 0.00016419206633796506, + "loss": 0.3561, + "step": 29958 + }, + { + "epoch": 1.6738553510070675, + "grad_norm": 0.4447183609008789, + "learning_rate": 0.00016416405199462126, + "loss": 0.3942, + "step": 29959 + }, + { + "epoch": 1.6739112216107495, + "grad_norm": 0.30650806427001953, + "learning_rate": 0.00016413603765127747, + "loss": 0.467, + "step": 29960 + }, + { + "epoch": 1.6739670922144314, + "grad_norm": 0.5068030953407288, + "learning_rate": 0.00016410802330793368, + "loss": 0.5368, + "step": 29961 + }, + { + "epoch": 1.6740229628181131, + "grad_norm": 0.30687418580055237, + "learning_rate": 0.00016408000896458988, + "loss": 0.3156, + "step": 29962 + }, + { + "epoch": 1.674078833421795, + "grad_norm": 0.46979594230651855, + "learning_rate": 0.0001640519946212461, + "loss": 0.4747, + "step": 29963 + }, + { + "epoch": 1.674134704025477, + "grad_norm": 0.5191566944122314, + "learning_rate": 0.0001640239802779023, + "loss": 0.5248, + "step": 29964 + }, + { + "epoch": 1.6741905746291588, + "grad_norm": 0.5987371206283569, + "learning_rate": 0.0001639959659345585, + "loss": 0.4499, + "step": 29965 + }, + { + "epoch": 1.6742464452328407, + "grad_norm": 0.4016363024711609, + "learning_rate": 0.0001639679515912147, + "loss": 0.3869, + "step": 29966 + }, + { + "epoch": 1.6743023158365227, + "grad_norm": 0.6122843027114868, + "learning_rate": 0.0001639399372478709, + "loss": 0.4554, + "step": 29967 + }, + { + "epoch": 1.6743581864402044, + "grad_norm": 0.48036032915115356, + "learning_rate": 0.00016391192290452712, + "loss": 0.5081, + "step": 29968 + }, + { + "epoch": 1.6744140570438864, + "grad_norm": 0.7246362566947937, + "learning_rate": 0.00016388390856118332, + "loss": 0.4088, + "step": 29969 + }, + { + "epoch": 1.6744699276475683, + "grad_norm": 0.9466614723205566, + "learning_rate": 0.00016385589421783953, + "loss": 0.4195, + "step": 29970 + }, + { + "epoch": 1.67452579825125, + "grad_norm": 0.3397181034088135, + "learning_rate": 0.00016382787987449573, + "loss": 0.4432, + "step": 29971 + }, + { + "epoch": 1.674581668854932, + "grad_norm": 0.3503602147102356, + "learning_rate": 0.00016379986553115194, + "loss": 0.3965, + "step": 29972 + }, + { + "epoch": 1.674637539458614, + "grad_norm": 0.5033782720565796, + "learning_rate": 0.00016377185118780815, + "loss": 0.4212, + "step": 29973 + }, + { + "epoch": 1.6746934100622957, + "grad_norm": 0.45386025309562683, + "learning_rate": 0.00016374383684446438, + "loss": 0.3908, + "step": 29974 + }, + { + "epoch": 1.6747492806659776, + "grad_norm": 0.3782189190387726, + "learning_rate": 0.00016371582250112058, + "loss": 0.4333, + "step": 29975 + }, + { + "epoch": 1.6748051512696596, + "grad_norm": 0.407453328371048, + "learning_rate": 0.0001636878081577768, + "loss": 0.4692, + "step": 29976 + }, + { + "epoch": 1.6748610218733413, + "grad_norm": 0.3924352526664734, + "learning_rate": 0.000163659793814433, + "loss": 0.4804, + "step": 29977 + }, + { + "epoch": 1.6749168924770232, + "grad_norm": 0.39628732204437256, + "learning_rate": 0.0001636317794710892, + "loss": 0.392, + "step": 29978 + }, + { + "epoch": 1.6749727630807052, + "grad_norm": 0.43830356001853943, + "learning_rate": 0.0001636037651277454, + "loss": 0.4582, + "step": 29979 + }, + { + "epoch": 1.675028633684387, + "grad_norm": 6.244492053985596, + "learning_rate": 0.0001635757507844016, + "loss": 0.4318, + "step": 29980 + }, + { + "epoch": 1.6750845042880689, + "grad_norm": 1.4900885820388794, + "learning_rate": 0.00016354773644105782, + "loss": 0.5776, + "step": 29981 + }, + { + "epoch": 1.6751403748917508, + "grad_norm": 0.3975076377391815, + "learning_rate": 0.00016351972209771405, + "loss": 0.5809, + "step": 29982 + }, + { + "epoch": 1.6751962454954326, + "grad_norm": 0.4281860589981079, + "learning_rate": 0.00016349170775437026, + "loss": 0.4292, + "step": 29983 + }, + { + "epoch": 1.6752521160991143, + "grad_norm": 0.31917524337768555, + "learning_rate": 0.00016346369341102646, + "loss": 0.4246, + "step": 29984 + }, + { + "epoch": 1.6753079867027965, + "grad_norm": 0.6659488677978516, + "learning_rate": 0.00016343567906768267, + "loss": 0.5585, + "step": 29985 + }, + { + "epoch": 1.6753638573064782, + "grad_norm": 0.40955162048339844, + "learning_rate": 0.00016340766472433888, + "loss": 0.4988, + "step": 29986 + }, + { + "epoch": 1.67541972791016, + "grad_norm": 0.6151166558265686, + "learning_rate": 0.00016337965038099505, + "loss": 0.4572, + "step": 29987 + }, + { + "epoch": 1.675475598513842, + "grad_norm": 0.5556128621101379, + "learning_rate": 0.00016335163603765126, + "loss": 0.5267, + "step": 29988 + }, + { + "epoch": 1.6755314691175238, + "grad_norm": 0.45335882902145386, + "learning_rate": 0.00016332362169430747, + "loss": 0.5922, + "step": 29989 + }, + { + "epoch": 1.6755873397212055, + "grad_norm": 0.5651610493659973, + "learning_rate": 0.0001632956073509637, + "loss": 0.377, + "step": 29990 + }, + { + "epoch": 1.6756432103248877, + "grad_norm": 0.28539711236953735, + "learning_rate": 0.0001632675930076199, + "loss": 0.4067, + "step": 29991 + }, + { + "epoch": 1.6756990809285695, + "grad_norm": 0.39051586389541626, + "learning_rate": 0.0001632395786642761, + "loss": 0.395, + "step": 29992 + }, + { + "epoch": 1.6757549515322512, + "grad_norm": 1.0865070819854736, + "learning_rate": 0.00016321156432093232, + "loss": 0.4757, + "step": 29993 + }, + { + "epoch": 1.6758108221359331, + "grad_norm": 0.6878522038459778, + "learning_rate": 0.00016318354997758852, + "loss": 0.4174, + "step": 29994 + }, + { + "epoch": 1.675866692739615, + "grad_norm": 1.094403624534607, + "learning_rate": 0.00016315553563424473, + "loss": 0.3815, + "step": 29995 + }, + { + "epoch": 1.6759225633432968, + "grad_norm": 0.4044061005115509, + "learning_rate": 0.00016312752129090093, + "loss": 0.4091, + "step": 29996 + }, + { + "epoch": 1.6759784339469788, + "grad_norm": 1.6676186323165894, + "learning_rate": 0.00016309950694755714, + "loss": 0.4037, + "step": 29997 + }, + { + "epoch": 1.6760343045506607, + "grad_norm": 0.6995496153831482, + "learning_rate": 0.00016307149260421337, + "loss": 0.342, + "step": 29998 + }, + { + "epoch": 1.6760901751543424, + "grad_norm": 0.5357198119163513, + "learning_rate": 0.00016304347826086958, + "loss": 0.3401, + "step": 29999 + }, + { + "epoch": 1.6761460457580244, + "grad_norm": 0.8350773453712463, + "learning_rate": 0.00016301546391752578, + "loss": 0.3163, + "step": 30000 + }, + { + "epoch": 1.6761460457580244, + "eval_cer": 0.0842743827261519, + "eval_loss": 0.314140647649765, + "eval_runtime": 56.0544, + "eval_samples_per_second": 80.957, + "eval_steps_per_second": 5.067, + "eval_wer": 0.33546548979078794, + "step": 30000 + }, + { + "epoch": 1.6762019163617063, + "grad_norm": 0.341926634311676, + "learning_rate": 0.000162987449574182, + "loss": 0.4304, + "step": 30001 + }, + { + "epoch": 1.676257786965388, + "grad_norm": 0.33519047498703003, + "learning_rate": 0.0001629594352308382, + "loss": 0.3765, + "step": 30002 + }, + { + "epoch": 1.67631365756907, + "grad_norm": 0.7132281064987183, + "learning_rate": 0.0001629314208874944, + "loss": 0.2987, + "step": 30003 + }, + { + "epoch": 1.676369528172752, + "grad_norm": 0.3844422698020935, + "learning_rate": 0.0001629034065441506, + "loss": 0.4675, + "step": 30004 + }, + { + "epoch": 1.6764253987764337, + "grad_norm": 0.4154326319694519, + "learning_rate": 0.00016287539220080684, + "loss": 0.4174, + "step": 30005 + }, + { + "epoch": 1.6764812693801157, + "grad_norm": 0.3121461272239685, + "learning_rate": 0.00016284737785746305, + "loss": 0.3414, + "step": 30006 + }, + { + "epoch": 1.6765371399837976, + "grad_norm": 0.7821503281593323, + "learning_rate": 0.00016281936351411925, + "loss": 0.3143, + "step": 30007 + }, + { + "epoch": 1.6765930105874793, + "grad_norm": 0.32892221212387085, + "learning_rate": 0.00016279134917077543, + "loss": 0.4012, + "step": 30008 + }, + { + "epoch": 1.6766488811911613, + "grad_norm": 0.48699328303337097, + "learning_rate": 0.00016276333482743164, + "loss": 0.5269, + "step": 30009 + }, + { + "epoch": 1.6767047517948432, + "grad_norm": 0.4691578447818756, + "learning_rate": 0.00016273532048408784, + "loss": 0.3817, + "step": 30010 + }, + { + "epoch": 1.676760622398525, + "grad_norm": 0.5530269742012024, + "learning_rate": 0.00016270730614074405, + "loss": 0.3894, + "step": 30011 + }, + { + "epoch": 1.676816493002207, + "grad_norm": 0.5813290476799011, + "learning_rate": 0.00016267929179740025, + "loss": 0.3918, + "step": 30012 + }, + { + "epoch": 1.6768723636058889, + "grad_norm": 0.4310387372970581, + "learning_rate": 0.0001626512774540565, + "loss": 0.4418, + "step": 30013 + }, + { + "epoch": 1.6769282342095706, + "grad_norm": 0.4852006137371063, + "learning_rate": 0.0001626232631107127, + "loss": 0.4255, + "step": 30014 + }, + { + "epoch": 1.6769841048132526, + "grad_norm": 1.2806484699249268, + "learning_rate": 0.0001625952487673689, + "loss": 0.5601, + "step": 30015 + }, + { + "epoch": 1.6770399754169345, + "grad_norm": 1.3292486667633057, + "learning_rate": 0.0001625672344240251, + "loss": 0.3814, + "step": 30016 + }, + { + "epoch": 1.6770958460206162, + "grad_norm": 1.9037526845932007, + "learning_rate": 0.0001625392200806813, + "loss": 0.4847, + "step": 30017 + }, + { + "epoch": 1.677151716624298, + "grad_norm": 0.5368911623954773, + "learning_rate": 0.00016251120573733752, + "loss": 0.4748, + "step": 30018 + }, + { + "epoch": 1.6772075872279801, + "grad_norm": 0.35629573464393616, + "learning_rate": 0.00016248319139399372, + "loss": 0.4055, + "step": 30019 + }, + { + "epoch": 1.6772634578316619, + "grad_norm": 0.5409963726997375, + "learning_rate": 0.00016245517705064993, + "loss": 0.4897, + "step": 30020 + }, + { + "epoch": 1.6773193284353436, + "grad_norm": 0.5151257514953613, + "learning_rate": 0.00016242716270730616, + "loss": 0.3984, + "step": 30021 + }, + { + "epoch": 1.6773751990390258, + "grad_norm": 1.0092713832855225, + "learning_rate": 0.00016239914836396237, + "loss": 0.4995, + "step": 30022 + }, + { + "epoch": 1.6774310696427075, + "grad_norm": 0.9104849100112915, + "learning_rate": 0.00016237113402061857, + "loss": 0.4708, + "step": 30023 + }, + { + "epoch": 1.6774869402463892, + "grad_norm": 0.4419928789138794, + "learning_rate": 0.00016234311967727478, + "loss": 0.5032, + "step": 30024 + }, + { + "epoch": 1.6775428108500714, + "grad_norm": 0.5856849551200867, + "learning_rate": 0.00016231510533393098, + "loss": 0.4464, + "step": 30025 + }, + { + "epoch": 1.6775986814537531, + "grad_norm": 2.7327959537506104, + "learning_rate": 0.0001622870909905872, + "loss": 0.4897, + "step": 30026 + }, + { + "epoch": 1.6776545520574349, + "grad_norm": 0.3942338526248932, + "learning_rate": 0.0001622590766472434, + "loss": 0.3052, + "step": 30027 + }, + { + "epoch": 1.6777104226611168, + "grad_norm": 0.33608710765838623, + "learning_rate": 0.0001622310623038996, + "loss": 0.3789, + "step": 30028 + }, + { + "epoch": 1.6777662932647988, + "grad_norm": 0.31452107429504395, + "learning_rate": 0.00016220304796055583, + "loss": 0.3952, + "step": 30029 + }, + { + "epoch": 1.6778221638684805, + "grad_norm": 0.4670000970363617, + "learning_rate": 0.000162175033617212, + "loss": 0.3575, + "step": 30030 + }, + { + "epoch": 1.6778780344721624, + "grad_norm": 0.5151067972183228, + "learning_rate": 0.00016214701927386822, + "loss": 0.3899, + "step": 30031 + }, + { + "epoch": 1.6779339050758444, + "grad_norm": 0.48114150762557983, + "learning_rate": 0.00016211900493052442, + "loss": 0.3262, + "step": 30032 + }, + { + "epoch": 1.6779897756795261, + "grad_norm": 1.1126295328140259, + "learning_rate": 0.00016209099058718063, + "loss": 0.3239, + "step": 30033 + }, + { + "epoch": 1.678045646283208, + "grad_norm": 0.8318020105361938, + "learning_rate": 0.00016206297624383684, + "loss": 0.4565, + "step": 30034 + }, + { + "epoch": 1.67810151688689, + "grad_norm": 0.42366307973861694, + "learning_rate": 0.00016203496190049304, + "loss": 0.4109, + "step": 30035 + }, + { + "epoch": 1.6781573874905718, + "grad_norm": 1.3030784130096436, + "learning_rate": 0.00016200694755714925, + "loss": 0.6105, + "step": 30036 + }, + { + "epoch": 1.6782132580942537, + "grad_norm": 0.6854676604270935, + "learning_rate": 0.00016197893321380548, + "loss": 0.3626, + "step": 30037 + }, + { + "epoch": 1.6782691286979357, + "grad_norm": 0.6634908318519592, + "learning_rate": 0.00016195091887046169, + "loss": 0.584, + "step": 30038 + }, + { + "epoch": 1.6783249993016174, + "grad_norm": 0.3645319640636444, + "learning_rate": 0.0001619229045271179, + "loss": 0.472, + "step": 30039 + }, + { + "epoch": 1.6783808699052993, + "grad_norm": 0.37251582741737366, + "learning_rate": 0.0001618948901837741, + "loss": 0.4151, + "step": 30040 + }, + { + "epoch": 1.6784367405089813, + "grad_norm": 0.3400768041610718, + "learning_rate": 0.0001618668758404303, + "loss": 0.4037, + "step": 30041 + }, + { + "epoch": 1.678492611112663, + "grad_norm": 2.7261600494384766, + "learning_rate": 0.0001618388614970865, + "loss": 0.5721, + "step": 30042 + }, + { + "epoch": 1.678548481716345, + "grad_norm": 0.635309100151062, + "learning_rate": 0.00016181084715374272, + "loss": 0.363, + "step": 30043 + }, + { + "epoch": 1.678604352320027, + "grad_norm": 0.553363025188446, + "learning_rate": 0.00016178283281039892, + "loss": 0.4743, + "step": 30044 + }, + { + "epoch": 1.6786602229237086, + "grad_norm": 0.4215012192726135, + "learning_rate": 0.00016175481846705515, + "loss": 0.4207, + "step": 30045 + }, + { + "epoch": 1.6787160935273906, + "grad_norm": 0.3544439971446991, + "learning_rate": 0.00016172680412371136, + "loss": 0.4272, + "step": 30046 + }, + { + "epoch": 1.6787719641310725, + "grad_norm": 0.28211313486099243, + "learning_rate": 0.00016169878978036757, + "loss": 0.3397, + "step": 30047 + }, + { + "epoch": 1.6788278347347543, + "grad_norm": 1.8518626689910889, + "learning_rate": 0.00016167077543702377, + "loss": 0.58, + "step": 30048 + }, + { + "epoch": 1.6788837053384362, + "grad_norm": 0.46959230303764343, + "learning_rate": 0.00016164276109367998, + "loss": 0.3655, + "step": 30049 + }, + { + "epoch": 1.6789395759421182, + "grad_norm": 0.4014168083667755, + "learning_rate": 0.00016161474675033618, + "loss": 0.4394, + "step": 30050 + }, + { + "epoch": 1.6789954465458, + "grad_norm": 0.343919038772583, + "learning_rate": 0.00016158673240699236, + "loss": 0.2939, + "step": 30051 + }, + { + "epoch": 1.6790513171494816, + "grad_norm": 0.5768270492553711, + "learning_rate": 0.00016155871806364857, + "loss": 0.4923, + "step": 30052 + }, + { + "epoch": 1.6791071877531638, + "grad_norm": 0.5438682436943054, + "learning_rate": 0.0001615307037203048, + "loss": 0.3961, + "step": 30053 + }, + { + "epoch": 1.6791630583568455, + "grad_norm": 4.823269367218018, + "learning_rate": 0.000161502689376961, + "loss": 0.4442, + "step": 30054 + }, + { + "epoch": 1.6792189289605273, + "grad_norm": 0.48959240317344666, + "learning_rate": 0.0001614746750336172, + "loss": 0.3959, + "step": 30055 + }, + { + "epoch": 1.6792747995642094, + "grad_norm": 0.4731843173503876, + "learning_rate": 0.00016144666069027342, + "loss": 0.4256, + "step": 30056 + }, + { + "epoch": 1.6793306701678912, + "grad_norm": 0.4499104917049408, + "learning_rate": 0.00016141864634692962, + "loss": 0.3815, + "step": 30057 + }, + { + "epoch": 1.679386540771573, + "grad_norm": 0.44841429591178894, + "learning_rate": 0.00016139063200358583, + "loss": 0.5107, + "step": 30058 + }, + { + "epoch": 1.679442411375255, + "grad_norm": 0.4282918870449066, + "learning_rate": 0.00016136261766024204, + "loss": 0.4079, + "step": 30059 + }, + { + "epoch": 1.6794982819789368, + "grad_norm": 0.5978766679763794, + "learning_rate": 0.00016133460331689824, + "loss": 0.3379, + "step": 30060 + }, + { + "epoch": 1.6795541525826185, + "grad_norm": 0.36962342262268066, + "learning_rate": 0.00016130658897355447, + "loss": 0.3935, + "step": 30061 + }, + { + "epoch": 1.6796100231863005, + "grad_norm": 3.900712490081787, + "learning_rate": 0.00016127857463021068, + "loss": 0.4622, + "step": 30062 + }, + { + "epoch": 1.6796658937899824, + "grad_norm": 1.2052878141403198, + "learning_rate": 0.00016125056028686689, + "loss": 0.3499, + "step": 30063 + }, + { + "epoch": 1.6797217643936642, + "grad_norm": 1.3032920360565186, + "learning_rate": 0.0001612225459435231, + "loss": 0.4624, + "step": 30064 + }, + { + "epoch": 1.6797776349973461, + "grad_norm": 0.6119353175163269, + "learning_rate": 0.0001611945316001793, + "loss": 0.339, + "step": 30065 + }, + { + "epoch": 1.679833505601028, + "grad_norm": 0.29979604482650757, + "learning_rate": 0.0001611665172568355, + "loss": 0.3782, + "step": 30066 + }, + { + "epoch": 1.6798893762047098, + "grad_norm": 1.2274726629257202, + "learning_rate": 0.0001611385029134917, + "loss": 0.3876, + "step": 30067 + }, + { + "epoch": 1.6799452468083917, + "grad_norm": 0.3400784730911255, + "learning_rate": 0.00016111048857014791, + "loss": 0.4631, + "step": 30068 + }, + { + "epoch": 1.6800011174120737, + "grad_norm": 0.34761157631874084, + "learning_rate": 0.00016108247422680415, + "loss": 0.4038, + "step": 30069 + }, + { + "epoch": 1.6800569880157554, + "grad_norm": 0.5591105818748474, + "learning_rate": 0.00016105445988346035, + "loss": 0.4631, + "step": 30070 + }, + { + "epoch": 1.6801128586194374, + "grad_norm": 0.3214731514453888, + "learning_rate": 0.00016102644554011656, + "loss": 0.4007, + "step": 30071 + }, + { + "epoch": 1.6801687292231193, + "grad_norm": 0.43304964900016785, + "learning_rate": 0.00016099843119677274, + "loss": 0.3553, + "step": 30072 + }, + { + "epoch": 1.680224599826801, + "grad_norm": 0.40976378321647644, + "learning_rate": 0.00016097041685342894, + "loss": 0.3712, + "step": 30073 + }, + { + "epoch": 1.680280470430483, + "grad_norm": 0.4497562646865845, + "learning_rate": 0.00016094240251008515, + "loss": 0.5379, + "step": 30074 + }, + { + "epoch": 1.680336341034165, + "grad_norm": 1.2380796670913696, + "learning_rate": 0.00016091438816674136, + "loss": 0.4188, + "step": 30075 + }, + { + "epoch": 1.6803922116378467, + "grad_norm": 0.4099409282207489, + "learning_rate": 0.00016088637382339756, + "loss": 0.4065, + "step": 30076 + }, + { + "epoch": 1.6804480822415286, + "grad_norm": 0.4936923384666443, + "learning_rate": 0.0001608583594800538, + "loss": 0.5737, + "step": 30077 + }, + { + "epoch": 1.6805039528452106, + "grad_norm": 0.45339471101760864, + "learning_rate": 0.00016083034513671, + "loss": 0.3141, + "step": 30078 + }, + { + "epoch": 1.6805598234488923, + "grad_norm": 0.3451547920703888, + "learning_rate": 0.0001608023307933662, + "loss": 0.3473, + "step": 30079 + }, + { + "epoch": 1.6806156940525743, + "grad_norm": 0.441608726978302, + "learning_rate": 0.0001607743164500224, + "loss": 0.4361, + "step": 30080 + }, + { + "epoch": 1.6806715646562562, + "grad_norm": 0.4267643392086029, + "learning_rate": 0.00016074630210667862, + "loss": 0.4841, + "step": 30081 + }, + { + "epoch": 1.680727435259938, + "grad_norm": 0.5080280303955078, + "learning_rate": 0.00016071828776333482, + "loss": 0.5879, + "step": 30082 + }, + { + "epoch": 1.68078330586362, + "grad_norm": 2.2676947116851807, + "learning_rate": 0.00016069027341999103, + "loss": 0.3077, + "step": 30083 + }, + { + "epoch": 1.6808391764673019, + "grad_norm": 0.3243895471096039, + "learning_rate": 0.00016066225907664726, + "loss": 0.3471, + "step": 30084 + }, + { + "epoch": 1.6808950470709836, + "grad_norm": 1.2029838562011719, + "learning_rate": 0.00016063424473330347, + "loss": 0.3804, + "step": 30085 + }, + { + "epoch": 1.6809509176746653, + "grad_norm": 0.35763776302337646, + "learning_rate": 0.00016060623038995967, + "loss": 0.3762, + "step": 30086 + }, + { + "epoch": 1.6810067882783475, + "grad_norm": 0.38882148265838623, + "learning_rate": 0.00016057821604661588, + "loss": 0.3866, + "step": 30087 + }, + { + "epoch": 1.6810626588820292, + "grad_norm": 1.182518720626831, + "learning_rate": 0.00016055020170327209, + "loss": 0.6845, + "step": 30088 + }, + { + "epoch": 1.681118529485711, + "grad_norm": 0.6423459649085999, + "learning_rate": 0.0001605221873599283, + "loss": 0.4832, + "step": 30089 + }, + { + "epoch": 1.6811744000893931, + "grad_norm": 0.42083680629730225, + "learning_rate": 0.0001604941730165845, + "loss": 0.3529, + "step": 30090 + }, + { + "epoch": 1.6812302706930748, + "grad_norm": 3.3482511043548584, + "learning_rate": 0.0001604661586732407, + "loss": 0.4921, + "step": 30091 + }, + { + "epoch": 1.6812861412967566, + "grad_norm": 0.3115469515323639, + "learning_rate": 0.00016043814432989694, + "loss": 0.3891, + "step": 30092 + }, + { + "epoch": 1.6813420119004387, + "grad_norm": 0.40387195348739624, + "learning_rate": 0.00016041012998655311, + "loss": 0.3626, + "step": 30093 + }, + { + "epoch": 1.6813978825041205, + "grad_norm": 2.9431092739105225, + "learning_rate": 0.00016038211564320932, + "loss": 0.6168, + "step": 30094 + }, + { + "epoch": 1.6814537531078022, + "grad_norm": 0.3116304576396942, + "learning_rate": 0.00016035410129986553, + "loss": 0.351, + "step": 30095 + }, + { + "epoch": 1.6815096237114842, + "grad_norm": 0.616200864315033, + "learning_rate": 0.00016032608695652173, + "loss": 0.4914, + "step": 30096 + }, + { + "epoch": 1.681565494315166, + "grad_norm": 1.4757602214813232, + "learning_rate": 0.00016029807261317794, + "loss": 0.394, + "step": 30097 + }, + { + "epoch": 1.6816213649188478, + "grad_norm": 1.4180912971496582, + "learning_rate": 0.00016027005826983414, + "loss": 0.5374, + "step": 30098 + }, + { + "epoch": 1.6816772355225298, + "grad_norm": 0.435249000787735, + "learning_rate": 0.00016024204392649035, + "loss": 0.3592, + "step": 30099 + }, + { + "epoch": 1.6817331061262117, + "grad_norm": 0.553627073764801, + "learning_rate": 0.00016021402958314658, + "loss": 0.4679, + "step": 30100 + }, + { + "epoch": 1.6817889767298935, + "grad_norm": 0.32963597774505615, + "learning_rate": 0.0001601860152398028, + "loss": 0.4081, + "step": 30101 + }, + { + "epoch": 1.6818448473335754, + "grad_norm": 0.4416026175022125, + "learning_rate": 0.000160158000896459, + "loss": 0.4669, + "step": 30102 + }, + { + "epoch": 1.6819007179372574, + "grad_norm": 0.4470868706703186, + "learning_rate": 0.0001601299865531152, + "loss": 0.3258, + "step": 30103 + }, + { + "epoch": 1.681956588540939, + "grad_norm": 0.33534109592437744, + "learning_rate": 0.0001601019722097714, + "loss": 0.4088, + "step": 30104 + }, + { + "epoch": 1.682012459144621, + "grad_norm": 0.3367442488670349, + "learning_rate": 0.0001600739578664276, + "loss": 0.4035, + "step": 30105 + }, + { + "epoch": 1.682068329748303, + "grad_norm": 0.43190693855285645, + "learning_rate": 0.00016004594352308382, + "loss": 0.4575, + "step": 30106 + }, + { + "epoch": 1.6821242003519847, + "grad_norm": 0.9889886379241943, + "learning_rate": 0.00016001792917974002, + "loss": 0.3749, + "step": 30107 + }, + { + "epoch": 1.6821800709556667, + "grad_norm": 0.3867762088775635, + "learning_rate": 0.00015998991483639626, + "loss": 0.4017, + "step": 30108 + }, + { + "epoch": 1.6822359415593486, + "grad_norm": 0.3811776041984558, + "learning_rate": 0.00015996190049305246, + "loss": 0.4059, + "step": 30109 + }, + { + "epoch": 1.6822918121630304, + "grad_norm": 0.4606453776359558, + "learning_rate": 0.00015993388614970867, + "loss": 0.3399, + "step": 30110 + }, + { + "epoch": 1.6823476827667123, + "grad_norm": 0.4108893573284149, + "learning_rate": 0.00015990587180636487, + "loss": 0.4139, + "step": 30111 + }, + { + "epoch": 1.6824035533703943, + "grad_norm": 0.4552747309207916, + "learning_rate": 0.00015987785746302108, + "loss": 0.3607, + "step": 30112 + }, + { + "epoch": 1.682459423974076, + "grad_norm": 0.6710257530212402, + "learning_rate": 0.00015984984311967728, + "loss": 0.622, + "step": 30113 + }, + { + "epoch": 1.682515294577758, + "grad_norm": 0.5581334233283997, + "learning_rate": 0.0001598218287763335, + "loss": 0.51, + "step": 30114 + }, + { + "epoch": 1.68257116518144, + "grad_norm": 7.404072284698486, + "learning_rate": 0.00015979381443298967, + "loss": 0.4519, + "step": 30115 + }, + { + "epoch": 1.6826270357851216, + "grad_norm": 0.36691343784332275, + "learning_rate": 0.0001597658000896459, + "loss": 0.4466, + "step": 30116 + }, + { + "epoch": 1.6826829063888036, + "grad_norm": 0.40177538990974426, + "learning_rate": 0.0001597377857463021, + "loss": 0.4354, + "step": 30117 + }, + { + "epoch": 1.6827387769924855, + "grad_norm": 1.0449219942092896, + "learning_rate": 0.00015970977140295831, + "loss": 0.3665, + "step": 30118 + }, + { + "epoch": 1.6827946475961673, + "grad_norm": 2.5912110805511475, + "learning_rate": 0.00015968175705961452, + "loss": 0.3589, + "step": 30119 + }, + { + "epoch": 1.682850518199849, + "grad_norm": 0.3633488416671753, + "learning_rate": 0.00015965374271627073, + "loss": 0.3876, + "step": 30120 + }, + { + "epoch": 1.6829063888035312, + "grad_norm": 0.46985429525375366, + "learning_rate": 0.00015962572837292693, + "loss": 0.4234, + "step": 30121 + }, + { + "epoch": 1.682962259407213, + "grad_norm": 0.3124782145023346, + "learning_rate": 0.00015959771402958314, + "loss": 0.3294, + "step": 30122 + }, + { + "epoch": 1.6830181300108946, + "grad_norm": 0.5186640024185181, + "learning_rate": 0.00015956969968623934, + "loss": 0.4532, + "step": 30123 + }, + { + "epoch": 1.6830740006145768, + "grad_norm": 5.502918720245361, + "learning_rate": 0.00015954168534289558, + "loss": 0.4003, + "step": 30124 + }, + { + "epoch": 1.6831298712182585, + "grad_norm": 0.3300841450691223, + "learning_rate": 0.00015951367099955178, + "loss": 0.3474, + "step": 30125 + }, + { + "epoch": 1.6831857418219403, + "grad_norm": 3.8792989253997803, + "learning_rate": 0.000159485656656208, + "loss": 0.6345, + "step": 30126 + }, + { + "epoch": 1.6832416124256224, + "grad_norm": 0.48740652203559875, + "learning_rate": 0.0001594576423128642, + "loss": 0.3442, + "step": 30127 + }, + { + "epoch": 1.6832974830293042, + "grad_norm": 0.454963356256485, + "learning_rate": 0.0001594296279695204, + "loss": 0.5802, + "step": 30128 + }, + { + "epoch": 1.6833533536329859, + "grad_norm": 0.48376980423927307, + "learning_rate": 0.0001594016136261766, + "loss": 0.3725, + "step": 30129 + }, + { + "epoch": 1.6834092242366678, + "grad_norm": 2.7381551265716553, + "learning_rate": 0.0001593735992828328, + "loss": 0.3931, + "step": 30130 + }, + { + "epoch": 1.6834650948403498, + "grad_norm": 1.5162526369094849, + "learning_rate": 0.00015934558493948902, + "loss": 0.3966, + "step": 30131 + }, + { + "epoch": 1.6835209654440315, + "grad_norm": 0.33189284801483154, + "learning_rate": 0.00015931757059614525, + "loss": 0.3794, + "step": 30132 + }, + { + "epoch": 1.6835768360477135, + "grad_norm": 1.408392310142517, + "learning_rate": 0.00015928955625280146, + "loss": 0.4667, + "step": 30133 + }, + { + "epoch": 1.6836327066513954, + "grad_norm": 1.1687359809875488, + "learning_rate": 0.00015926154190945766, + "loss": 0.4752, + "step": 30134 + }, + { + "epoch": 1.6836885772550771, + "grad_norm": 1.223551869392395, + "learning_rate": 0.00015923352756611387, + "loss": 0.3649, + "step": 30135 + }, + { + "epoch": 1.683744447858759, + "grad_norm": 0.6989678740501404, + "learning_rate": 0.00015920551322277005, + "loss": 0.3874, + "step": 30136 + }, + { + "epoch": 1.683800318462441, + "grad_norm": 0.5217908024787903, + "learning_rate": 0.00015917749887942625, + "loss": 0.386, + "step": 30137 + }, + { + "epoch": 1.6838561890661228, + "grad_norm": 0.44992607831954956, + "learning_rate": 0.00015914948453608246, + "loss": 0.462, + "step": 30138 + }, + { + "epoch": 1.6839120596698047, + "grad_norm": 0.49941450357437134, + "learning_rate": 0.00015912147019273866, + "loss": 0.3398, + "step": 30139 + }, + { + "epoch": 1.6839679302734867, + "grad_norm": 0.49447745084762573, + "learning_rate": 0.0001590934558493949, + "loss": 0.4337, + "step": 30140 + }, + { + "epoch": 1.6840238008771684, + "grad_norm": 4.309815406799316, + "learning_rate": 0.0001590654415060511, + "loss": 0.3713, + "step": 30141 + }, + { + "epoch": 1.6840796714808504, + "grad_norm": 0.6340694427490234, + "learning_rate": 0.0001590374271627073, + "loss": 0.479, + "step": 30142 + }, + { + "epoch": 1.6841355420845323, + "grad_norm": 0.48523402214050293, + "learning_rate": 0.00015900941281936351, + "loss": 0.5181, + "step": 30143 + }, + { + "epoch": 1.684191412688214, + "grad_norm": 0.3502611815929413, + "learning_rate": 0.00015898139847601972, + "loss": 0.3315, + "step": 30144 + }, + { + "epoch": 1.684247283291896, + "grad_norm": 0.4534909427165985, + "learning_rate": 0.00015895338413267593, + "loss": 0.3601, + "step": 30145 + }, + { + "epoch": 1.684303153895578, + "grad_norm": 0.5051678419113159, + "learning_rate": 0.00015892536978933213, + "loss": 0.4441, + "step": 30146 + }, + { + "epoch": 1.6843590244992597, + "grad_norm": 0.4045621454715729, + "learning_rate": 0.00015889735544598836, + "loss": 0.4296, + "step": 30147 + }, + { + "epoch": 1.6844148951029416, + "grad_norm": 0.7078512907028198, + "learning_rate": 0.00015886934110264457, + "loss": 0.3471, + "step": 30148 + }, + { + "epoch": 1.6844707657066236, + "grad_norm": 0.3259676694869995, + "learning_rate": 0.00015884132675930078, + "loss": 0.4505, + "step": 30149 + }, + { + "epoch": 1.6845266363103053, + "grad_norm": 0.8367729187011719, + "learning_rate": 0.00015881331241595698, + "loss": 0.4131, + "step": 30150 + }, + { + "epoch": 1.6845825069139873, + "grad_norm": 0.41864755749702454, + "learning_rate": 0.0001587852980726132, + "loss": 0.4898, + "step": 30151 + }, + { + "epoch": 1.6846383775176692, + "grad_norm": 0.3664648234844208, + "learning_rate": 0.0001587572837292694, + "loss": 0.322, + "step": 30152 + }, + { + "epoch": 1.684694248121351, + "grad_norm": 0.33165407180786133, + "learning_rate": 0.0001587292693859256, + "loss": 0.4007, + "step": 30153 + }, + { + "epoch": 1.6847501187250327, + "grad_norm": 0.7153699994087219, + "learning_rate": 0.0001587012550425818, + "loss": 0.5555, + "step": 30154 + }, + { + "epoch": 1.6848059893287148, + "grad_norm": 0.7323513627052307, + "learning_rate": 0.00015867324069923804, + "loss": 0.6968, + "step": 30155 + }, + { + "epoch": 1.6848618599323966, + "grad_norm": 0.5393884778022766, + "learning_rate": 0.00015864522635589424, + "loss": 0.4484, + "step": 30156 + }, + { + "epoch": 1.6849177305360783, + "grad_norm": 0.4356802701950073, + "learning_rate": 0.00015861721201255042, + "loss": 0.5019, + "step": 30157 + }, + { + "epoch": 1.6849736011397605, + "grad_norm": 0.30113834142684937, + "learning_rate": 0.00015858919766920663, + "loss": 0.3834, + "step": 30158 + }, + { + "epoch": 1.6850294717434422, + "grad_norm": 0.4506736099720001, + "learning_rate": 0.00015856118332586283, + "loss": 0.3781, + "step": 30159 + }, + { + "epoch": 1.685085342347124, + "grad_norm": 0.6174049973487854, + "learning_rate": 0.00015853316898251904, + "loss": 0.3418, + "step": 30160 + }, + { + "epoch": 1.685141212950806, + "grad_norm": 0.36893320083618164, + "learning_rate": 0.00015850515463917525, + "loss": 0.4168, + "step": 30161 + }, + { + "epoch": 1.6851970835544878, + "grad_norm": 0.31441783905029297, + "learning_rate": 0.00015847714029583145, + "loss": 0.3917, + "step": 30162 + }, + { + "epoch": 1.6852529541581696, + "grad_norm": 0.5016137957572937, + "learning_rate": 0.00015844912595248768, + "loss": 0.3558, + "step": 30163 + }, + { + "epoch": 1.6853088247618515, + "grad_norm": 0.3178475499153137, + "learning_rate": 0.0001584211116091439, + "loss": 0.4507, + "step": 30164 + }, + { + "epoch": 1.6853646953655335, + "grad_norm": 0.35701173543930054, + "learning_rate": 0.0001583930972658001, + "loss": 0.4477, + "step": 30165 + }, + { + "epoch": 1.6854205659692152, + "grad_norm": 0.45029038190841675, + "learning_rate": 0.0001583650829224563, + "loss": 0.4093, + "step": 30166 + }, + { + "epoch": 1.6854764365728971, + "grad_norm": 0.41538235545158386, + "learning_rate": 0.0001583370685791125, + "loss": 0.4328, + "step": 30167 + }, + { + "epoch": 1.685532307176579, + "grad_norm": 0.4577021300792694, + "learning_rate": 0.0001583090542357687, + "loss": 0.4528, + "step": 30168 + }, + { + "epoch": 1.6855881777802608, + "grad_norm": 0.3815114200115204, + "learning_rate": 0.00015828103989242492, + "loss": 0.4484, + "step": 30169 + }, + { + "epoch": 1.6856440483839428, + "grad_norm": 0.6583699584007263, + "learning_rate": 0.00015825302554908112, + "loss": 0.4065, + "step": 30170 + }, + { + "epoch": 1.6856999189876247, + "grad_norm": 0.3124699890613556, + "learning_rate": 0.00015822501120573736, + "loss": 0.2659, + "step": 30171 + }, + { + "epoch": 1.6857557895913065, + "grad_norm": 2.203935146331787, + "learning_rate": 0.00015819699686239356, + "loss": 0.3021, + "step": 30172 + }, + { + "epoch": 1.6858116601949884, + "grad_norm": 0.6214574575424194, + "learning_rate": 0.00015816898251904977, + "loss": 0.5099, + "step": 30173 + }, + { + "epoch": 1.6858675307986704, + "grad_norm": 1.5360230207443237, + "learning_rate": 0.00015814096817570598, + "loss": 0.5917, + "step": 30174 + }, + { + "epoch": 1.685923401402352, + "grad_norm": 0.6321947574615479, + "learning_rate": 0.00015811295383236218, + "loss": 0.428, + "step": 30175 + }, + { + "epoch": 1.685979272006034, + "grad_norm": 0.48554691672325134, + "learning_rate": 0.0001580849394890184, + "loss": 0.5214, + "step": 30176 + }, + { + "epoch": 1.686035142609716, + "grad_norm": 1.138164758682251, + "learning_rate": 0.0001580569251456746, + "loss": 0.4085, + "step": 30177 + }, + { + "epoch": 1.6860910132133977, + "grad_norm": 0.3850244879722595, + "learning_rate": 0.00015802891080233077, + "loss": 0.4077, + "step": 30178 + }, + { + "epoch": 1.6861468838170797, + "grad_norm": 0.4234500229358673, + "learning_rate": 0.000158000896458987, + "loss": 0.4555, + "step": 30179 + }, + { + "epoch": 1.6862027544207616, + "grad_norm": 3.141160488128662, + "learning_rate": 0.0001579728821156432, + "loss": 0.4177, + "step": 30180 + }, + { + "epoch": 1.6862586250244433, + "grad_norm": 0.6602550745010376, + "learning_rate": 0.00015794486777229942, + "loss": 0.492, + "step": 30181 + }, + { + "epoch": 1.6863144956281253, + "grad_norm": 0.3352707326412201, + "learning_rate": 0.00015791685342895562, + "loss": 0.4072, + "step": 30182 + }, + { + "epoch": 1.6863703662318072, + "grad_norm": 0.7491292357444763, + "learning_rate": 0.00015788883908561183, + "loss": 0.4383, + "step": 30183 + }, + { + "epoch": 1.686426236835489, + "grad_norm": 0.3579082190990448, + "learning_rate": 0.00015786082474226803, + "loss": 0.4296, + "step": 30184 + }, + { + "epoch": 1.686482107439171, + "grad_norm": 0.4141029417514801, + "learning_rate": 0.00015783281039892424, + "loss": 0.3826, + "step": 30185 + }, + { + "epoch": 1.6865379780428529, + "grad_norm": 0.502348005771637, + "learning_rate": 0.00015780479605558044, + "loss": 0.4335, + "step": 30186 + }, + { + "epoch": 1.6865938486465346, + "grad_norm": 0.47037452459335327, + "learning_rate": 0.00015777678171223668, + "loss": 0.5529, + "step": 30187 + }, + { + "epoch": 1.6866497192502163, + "grad_norm": 0.420052707195282, + "learning_rate": 0.00015774876736889288, + "loss": 0.4763, + "step": 30188 + }, + { + "epoch": 1.6867055898538985, + "grad_norm": 0.28092461824417114, + "learning_rate": 0.0001577207530255491, + "loss": 0.2449, + "step": 30189 + }, + { + "epoch": 1.6867614604575802, + "grad_norm": 0.6392021179199219, + "learning_rate": 0.0001576927386822053, + "loss": 0.4689, + "step": 30190 + }, + { + "epoch": 1.686817331061262, + "grad_norm": 3.071178674697876, + "learning_rate": 0.0001576647243388615, + "loss": 0.4967, + "step": 30191 + }, + { + "epoch": 1.6868732016649441, + "grad_norm": 0.480973482131958, + "learning_rate": 0.0001576367099955177, + "loss": 0.3683, + "step": 30192 + }, + { + "epoch": 1.6869290722686259, + "grad_norm": 0.4318854510784149, + "learning_rate": 0.0001576086956521739, + "loss": 0.3649, + "step": 30193 + }, + { + "epoch": 1.6869849428723076, + "grad_norm": 0.3329899311065674, + "learning_rate": 0.00015758068130883012, + "loss": 0.4063, + "step": 30194 + }, + { + "epoch": 1.6870408134759896, + "grad_norm": 0.3459647297859192, + "learning_rate": 0.00015755266696548635, + "loss": 0.3515, + "step": 30195 + }, + { + "epoch": 1.6870966840796715, + "grad_norm": 0.5969052314758301, + "learning_rate": 0.00015752465262214256, + "loss": 0.4327, + "step": 30196 + }, + { + "epoch": 1.6871525546833532, + "grad_norm": 4.065917491912842, + "learning_rate": 0.00015749663827879876, + "loss": 0.5746, + "step": 30197 + }, + { + "epoch": 1.6872084252870352, + "grad_norm": 0.5102059245109558, + "learning_rate": 0.00015746862393545497, + "loss": 0.4013, + "step": 30198 + }, + { + "epoch": 1.6872642958907171, + "grad_norm": 0.45781630277633667, + "learning_rate": 0.00015744060959211117, + "loss": 0.574, + "step": 30199 + }, + { + "epoch": 1.6873201664943989, + "grad_norm": 0.7383933663368225, + "learning_rate": 0.00015741259524876735, + "loss": 0.4179, + "step": 30200 + }, + { + "epoch": 1.6873760370980808, + "grad_norm": 0.3133249580860138, + "learning_rate": 0.00015738458090542356, + "loss": 0.4487, + "step": 30201 + }, + { + "epoch": 1.6874319077017628, + "grad_norm": 2.6708264350891113, + "learning_rate": 0.00015735656656207977, + "loss": 0.3472, + "step": 30202 + }, + { + "epoch": 1.6874877783054445, + "grad_norm": 0.7077803015708923, + "learning_rate": 0.000157328552218736, + "loss": 0.4258, + "step": 30203 + }, + { + "epoch": 1.6875436489091264, + "grad_norm": 0.3432350158691406, + "learning_rate": 0.0001573005378753922, + "loss": 0.3246, + "step": 30204 + }, + { + "epoch": 1.6875995195128084, + "grad_norm": 0.37171170115470886, + "learning_rate": 0.0001572725235320484, + "loss": 0.385, + "step": 30205 + }, + { + "epoch": 1.6876553901164901, + "grad_norm": 0.42717355489730835, + "learning_rate": 0.00015724450918870462, + "loss": 0.4136, + "step": 30206 + }, + { + "epoch": 1.687711260720172, + "grad_norm": 0.40296289324760437, + "learning_rate": 0.00015721649484536082, + "loss": 0.4912, + "step": 30207 + }, + { + "epoch": 1.687767131323854, + "grad_norm": 0.6106090545654297, + "learning_rate": 0.00015718848050201703, + "loss": 0.4447, + "step": 30208 + }, + { + "epoch": 1.6878230019275358, + "grad_norm": 0.37995588779449463, + "learning_rate": 0.00015716046615867323, + "loss": 0.4821, + "step": 30209 + }, + { + "epoch": 1.6878788725312177, + "grad_norm": 0.4144570231437683, + "learning_rate": 0.00015713245181532947, + "loss": 0.3239, + "step": 30210 + }, + { + "epoch": 1.6879347431348997, + "grad_norm": 0.47190067172050476, + "learning_rate": 0.00015710443747198567, + "loss": 0.4309, + "step": 30211 + }, + { + "epoch": 1.6879906137385814, + "grad_norm": 0.38136452436447144, + "learning_rate": 0.00015707642312864188, + "loss": 0.407, + "step": 30212 + }, + { + "epoch": 1.6880464843422633, + "grad_norm": 0.453768789768219, + "learning_rate": 0.00015704840878529808, + "loss": 0.3997, + "step": 30213 + }, + { + "epoch": 1.6881023549459453, + "grad_norm": 0.6643487811088562, + "learning_rate": 0.0001570203944419543, + "loss": 0.5151, + "step": 30214 + }, + { + "epoch": 1.688158225549627, + "grad_norm": 0.4300307631492615, + "learning_rate": 0.0001569923800986105, + "loss": 0.4844, + "step": 30215 + }, + { + "epoch": 1.688214096153309, + "grad_norm": 0.6053546071052551, + "learning_rate": 0.0001569643657552667, + "loss": 0.4732, + "step": 30216 + }, + { + "epoch": 1.688269966756991, + "grad_norm": 0.8524609804153442, + "learning_rate": 0.0001569363514119229, + "loss": 0.6851, + "step": 30217 + }, + { + "epoch": 1.6883258373606727, + "grad_norm": 0.5949439406394958, + "learning_rate": 0.00015690833706857914, + "loss": 0.4275, + "step": 30218 + }, + { + "epoch": 1.6883817079643546, + "grad_norm": 0.428824782371521, + "learning_rate": 0.00015688032272523535, + "loss": 0.5064, + "step": 30219 + }, + { + "epoch": 1.6884375785680366, + "grad_norm": 0.4236562252044678, + "learning_rate": 0.00015685230838189155, + "loss": 0.3917, + "step": 30220 + }, + { + "epoch": 1.6884934491717183, + "grad_norm": 0.4491504728794098, + "learning_rate": 0.00015682429403854773, + "loss": 0.3487, + "step": 30221 + }, + { + "epoch": 1.6885493197754, + "grad_norm": 2.290701389312744, + "learning_rate": 0.00015679627969520394, + "loss": 0.3783, + "step": 30222 + }, + { + "epoch": 1.6886051903790822, + "grad_norm": 0.5816791653633118, + "learning_rate": 0.00015676826535186014, + "loss": 0.507, + "step": 30223 + }, + { + "epoch": 1.688661060982764, + "grad_norm": 0.3967653512954712, + "learning_rate": 0.00015674025100851635, + "loss": 0.4422, + "step": 30224 + }, + { + "epoch": 1.6887169315864456, + "grad_norm": 0.40484490990638733, + "learning_rate": 0.00015671223666517255, + "loss": 0.3943, + "step": 30225 + }, + { + "epoch": 1.6887728021901278, + "grad_norm": 0.39252543449401855, + "learning_rate": 0.00015668422232182879, + "loss": 0.4024, + "step": 30226 + }, + { + "epoch": 1.6888286727938095, + "grad_norm": 1.1866919994354248, + "learning_rate": 0.000156656207978485, + "loss": 0.3982, + "step": 30227 + }, + { + "epoch": 1.6888845433974913, + "grad_norm": 0.686292290687561, + "learning_rate": 0.0001566281936351412, + "loss": 0.5164, + "step": 30228 + }, + { + "epoch": 1.6889404140011732, + "grad_norm": 0.4042664170265198, + "learning_rate": 0.0001566001792917974, + "loss": 0.3991, + "step": 30229 + }, + { + "epoch": 1.6889962846048552, + "grad_norm": 0.47741296887397766, + "learning_rate": 0.0001565721649484536, + "loss": 0.3771, + "step": 30230 + }, + { + "epoch": 1.689052155208537, + "grad_norm": 0.3356856405735016, + "learning_rate": 0.00015654415060510982, + "loss": 0.4449, + "step": 30231 + }, + { + "epoch": 1.6891080258122189, + "grad_norm": 0.5157215595245361, + "learning_rate": 0.00015651613626176602, + "loss": 0.4694, + "step": 30232 + }, + { + "epoch": 1.6891638964159008, + "grad_norm": 0.6531292200088501, + "learning_rate": 0.00015648812191842223, + "loss": 0.4519, + "step": 30233 + }, + { + "epoch": 1.6892197670195825, + "grad_norm": 0.3897513747215271, + "learning_rate": 0.00015646010757507846, + "loss": 0.4214, + "step": 30234 + }, + { + "epoch": 1.6892756376232645, + "grad_norm": 0.45963841676712036, + "learning_rate": 0.00015643209323173467, + "loss": 0.5282, + "step": 30235 + }, + { + "epoch": 1.6893315082269464, + "grad_norm": 0.4753164052963257, + "learning_rate": 0.00015640407888839087, + "loss": 0.312, + "step": 30236 + }, + { + "epoch": 1.6893873788306282, + "grad_norm": 0.42709264159202576, + "learning_rate": 0.00015637606454504708, + "loss": 0.4249, + "step": 30237 + }, + { + "epoch": 1.6894432494343101, + "grad_norm": 0.4875389337539673, + "learning_rate": 0.00015634805020170328, + "loss": 0.5617, + "step": 30238 + }, + { + "epoch": 1.689499120037992, + "grad_norm": 0.392577201128006, + "learning_rate": 0.0001563200358583595, + "loss": 0.4593, + "step": 30239 + }, + { + "epoch": 1.6895549906416738, + "grad_norm": 0.5369172692298889, + "learning_rate": 0.0001562920215150157, + "loss": 0.4028, + "step": 30240 + }, + { + "epoch": 1.6896108612453558, + "grad_norm": 3.987722873687744, + "learning_rate": 0.0001562640071716719, + "loss": 0.4554, + "step": 30241 + }, + { + "epoch": 1.6896667318490377, + "grad_norm": 1.0069315433502197, + "learning_rate": 0.0001562359928283281, + "loss": 0.4395, + "step": 30242 + }, + { + "epoch": 1.6897226024527194, + "grad_norm": 0.6417668461799622, + "learning_rate": 0.0001562079784849843, + "loss": 0.463, + "step": 30243 + }, + { + "epoch": 1.6897784730564014, + "grad_norm": 0.45640093088150024, + "learning_rate": 0.00015617996414164052, + "loss": 0.4471, + "step": 30244 + }, + { + "epoch": 1.6898343436600833, + "grad_norm": 0.3611442446708679, + "learning_rate": 0.00015615194979829672, + "loss": 0.4468, + "step": 30245 + }, + { + "epoch": 1.689890214263765, + "grad_norm": 0.47053074836730957, + "learning_rate": 0.00015612393545495293, + "loss": 0.5378, + "step": 30246 + }, + { + "epoch": 1.689946084867447, + "grad_norm": 0.41998031735420227, + "learning_rate": 0.00015609592111160914, + "loss": 0.364, + "step": 30247 + }, + { + "epoch": 1.690001955471129, + "grad_norm": 0.4169637858867645, + "learning_rate": 0.00015606790676826534, + "loss": 0.4613, + "step": 30248 + }, + { + "epoch": 1.6900578260748107, + "grad_norm": 0.39180687069892883, + "learning_rate": 0.00015603989242492155, + "loss": 0.4449, + "step": 30249 + }, + { + "epoch": 1.6901136966784926, + "grad_norm": 0.39590683579444885, + "learning_rate": 0.00015601187808157778, + "loss": 0.3961, + "step": 30250 + }, + { + "epoch": 1.6901695672821746, + "grad_norm": 0.566484272480011, + "learning_rate": 0.00015598386373823399, + "loss": 0.6415, + "step": 30251 + }, + { + "epoch": 1.6902254378858563, + "grad_norm": 0.44671911001205444, + "learning_rate": 0.0001559558493948902, + "loss": 0.3394, + "step": 30252 + }, + { + "epoch": 1.6902813084895383, + "grad_norm": 0.4811137020587921, + "learning_rate": 0.0001559278350515464, + "loss": 0.3819, + "step": 30253 + }, + { + "epoch": 1.6903371790932202, + "grad_norm": 0.361793577671051, + "learning_rate": 0.0001558998207082026, + "loss": 0.4348, + "step": 30254 + }, + { + "epoch": 1.690393049696902, + "grad_norm": 0.398275226354599, + "learning_rate": 0.0001558718063648588, + "loss": 0.4029, + "step": 30255 + }, + { + "epoch": 1.6904489203005837, + "grad_norm": 0.44092047214508057, + "learning_rate": 0.00015584379202151501, + "loss": 0.5031, + "step": 30256 + }, + { + "epoch": 1.6905047909042659, + "grad_norm": 0.38654351234436035, + "learning_rate": 0.00015581577767817122, + "loss": 0.4264, + "step": 30257 + }, + { + "epoch": 1.6905606615079476, + "grad_norm": 1.020134449005127, + "learning_rate": 0.00015578776333482745, + "loss": 0.355, + "step": 30258 + }, + { + "epoch": 1.6906165321116293, + "grad_norm": 0.4322343170642853, + "learning_rate": 0.00015575974899148366, + "loss": 0.4676, + "step": 30259 + }, + { + "epoch": 1.6906724027153115, + "grad_norm": 0.52605140209198, + "learning_rate": 0.00015573173464813987, + "loss": 0.4089, + "step": 30260 + }, + { + "epoch": 1.6907282733189932, + "grad_norm": 0.3850537836551666, + "learning_rate": 0.00015570372030479607, + "loss": 0.3785, + "step": 30261 + }, + { + "epoch": 1.690784143922675, + "grad_norm": 0.46837764978408813, + "learning_rate": 0.00015567570596145228, + "loss": 0.5016, + "step": 30262 + }, + { + "epoch": 1.690840014526357, + "grad_norm": 0.35848599672317505, + "learning_rate": 0.00015564769161810846, + "loss": 0.3843, + "step": 30263 + }, + { + "epoch": 1.6908958851300389, + "grad_norm": 1.8292760848999023, + "learning_rate": 0.00015561967727476466, + "loss": 0.466, + "step": 30264 + }, + { + "epoch": 1.6909517557337206, + "grad_norm": 0.5751210451126099, + "learning_rate": 0.00015559166293142087, + "loss": 0.406, + "step": 30265 + }, + { + "epoch": 1.6910076263374025, + "grad_norm": 0.456853985786438, + "learning_rate": 0.0001555636485880771, + "loss": 0.3704, + "step": 30266 + }, + { + "epoch": 1.6910634969410845, + "grad_norm": 0.46132275462150574, + "learning_rate": 0.0001555356342447333, + "loss": 0.4421, + "step": 30267 + }, + { + "epoch": 1.6911193675447662, + "grad_norm": 0.6188775897026062, + "learning_rate": 0.0001555076199013895, + "loss": 0.4517, + "step": 30268 + }, + { + "epoch": 1.6911752381484482, + "grad_norm": 0.34200361371040344, + "learning_rate": 0.00015547960555804572, + "loss": 0.4033, + "step": 30269 + }, + { + "epoch": 1.6912311087521301, + "grad_norm": 0.43404293060302734, + "learning_rate": 0.00015545159121470192, + "loss": 0.444, + "step": 30270 + }, + { + "epoch": 1.6912869793558118, + "grad_norm": 0.34541091322898865, + "learning_rate": 0.00015542357687135813, + "loss": 0.3776, + "step": 30271 + }, + { + "epoch": 1.6913428499594938, + "grad_norm": 0.36243051290512085, + "learning_rate": 0.00015539556252801433, + "loss": 0.3957, + "step": 30272 + }, + { + "epoch": 1.6913987205631758, + "grad_norm": 0.3820360004901886, + "learning_rate": 0.00015536754818467054, + "loss": 0.3644, + "step": 30273 + }, + { + "epoch": 1.6914545911668575, + "grad_norm": 0.5619180798530579, + "learning_rate": 0.00015533953384132677, + "loss": 0.4056, + "step": 30274 + }, + { + "epoch": 1.6915104617705394, + "grad_norm": 0.6304979920387268, + "learning_rate": 0.00015531151949798298, + "loss": 0.5154, + "step": 30275 + }, + { + "epoch": 1.6915663323742214, + "grad_norm": 0.4196334481239319, + "learning_rate": 0.00015528350515463919, + "loss": 0.4084, + "step": 30276 + }, + { + "epoch": 1.6916222029779031, + "grad_norm": 0.3852550983428955, + "learning_rate": 0.0001552554908112954, + "loss": 0.434, + "step": 30277 + }, + { + "epoch": 1.691678073581585, + "grad_norm": 0.33765220642089844, + "learning_rate": 0.0001552274764679516, + "loss": 0.4316, + "step": 30278 + }, + { + "epoch": 1.691733944185267, + "grad_norm": 0.41830042004585266, + "learning_rate": 0.0001551994621246078, + "loss": 0.422, + "step": 30279 + }, + { + "epoch": 1.6917898147889487, + "grad_norm": 0.4867834448814392, + "learning_rate": 0.000155171447781264, + "loss": 0.3553, + "step": 30280 + }, + { + "epoch": 1.6918456853926307, + "grad_norm": 0.3846835196018219, + "learning_rate": 0.00015514343343792024, + "loss": 0.3377, + "step": 30281 + }, + { + "epoch": 1.6919015559963126, + "grad_norm": 0.8832949995994568, + "learning_rate": 0.00015511541909457645, + "loss": 0.552, + "step": 30282 + }, + { + "epoch": 1.6919574265999944, + "grad_norm": 0.3461758494377136, + "learning_rate": 0.00015508740475123265, + "loss": 0.4371, + "step": 30283 + }, + { + "epoch": 1.6920132972036763, + "grad_norm": 0.7287971377372742, + "learning_rate": 0.00015505939040788883, + "loss": 0.4863, + "step": 30284 + }, + { + "epoch": 1.6920691678073583, + "grad_norm": 0.37126436829566956, + "learning_rate": 0.00015503137606454504, + "loss": 0.4239, + "step": 30285 + }, + { + "epoch": 1.69212503841104, + "grad_norm": 0.7217692732810974, + "learning_rate": 0.00015500336172120124, + "loss": 0.3929, + "step": 30286 + }, + { + "epoch": 1.692180909014722, + "grad_norm": 1.2728424072265625, + "learning_rate": 0.00015497534737785745, + "loss": 0.4784, + "step": 30287 + }, + { + "epoch": 1.692236779618404, + "grad_norm": 0.4901426136493683, + "learning_rate": 0.00015494733303451365, + "loss": 0.4629, + "step": 30288 + }, + { + "epoch": 1.6922926502220856, + "grad_norm": 0.5242897868156433, + "learning_rate": 0.0001549193186911699, + "loss": 0.4937, + "step": 30289 + }, + { + "epoch": 1.6923485208257674, + "grad_norm": 0.4266435205936432, + "learning_rate": 0.0001548913043478261, + "loss": 0.451, + "step": 30290 + }, + { + "epoch": 1.6924043914294495, + "grad_norm": 0.3182283937931061, + "learning_rate": 0.0001548632900044823, + "loss": 0.3609, + "step": 30291 + }, + { + "epoch": 1.6924602620331313, + "grad_norm": 0.31490975618362427, + "learning_rate": 0.0001548352756611385, + "loss": 0.4474, + "step": 30292 + }, + { + "epoch": 1.692516132636813, + "grad_norm": 0.9428167343139648, + "learning_rate": 0.0001548072613177947, + "loss": 0.4105, + "step": 30293 + }, + { + "epoch": 1.6925720032404952, + "grad_norm": 0.5608308911323547, + "learning_rate": 0.00015477924697445092, + "loss": 0.3529, + "step": 30294 + }, + { + "epoch": 1.692627873844177, + "grad_norm": 0.5104441046714783, + "learning_rate": 0.00015475123263110712, + "loss": 0.5067, + "step": 30295 + }, + { + "epoch": 1.6926837444478586, + "grad_norm": 0.4238840341567993, + "learning_rate": 0.00015472321828776333, + "loss": 0.3505, + "step": 30296 + }, + { + "epoch": 1.6927396150515406, + "grad_norm": 0.35687586665153503, + "learning_rate": 0.00015469520394441956, + "loss": 0.4488, + "step": 30297 + }, + { + "epoch": 1.6927954856552225, + "grad_norm": 0.6044889688491821, + "learning_rate": 0.00015466718960107577, + "loss": 0.494, + "step": 30298 + }, + { + "epoch": 1.6928513562589043, + "grad_norm": 0.5433734655380249, + "learning_rate": 0.00015463917525773197, + "loss": 0.455, + "step": 30299 + }, + { + "epoch": 1.6929072268625862, + "grad_norm": 0.592708945274353, + "learning_rate": 0.00015461116091438818, + "loss": 0.4193, + "step": 30300 + }, + { + "epoch": 1.6929630974662682, + "grad_norm": 0.3940798342227936, + "learning_rate": 0.00015458314657104438, + "loss": 0.4002, + "step": 30301 + }, + { + "epoch": 1.69301896806995, + "grad_norm": 0.5098593831062317, + "learning_rate": 0.0001545551322277006, + "loss": 0.4452, + "step": 30302 + }, + { + "epoch": 1.6930748386736318, + "grad_norm": 0.4587629437446594, + "learning_rate": 0.0001545271178843568, + "loss": 0.391, + "step": 30303 + }, + { + "epoch": 1.6931307092773138, + "grad_norm": 0.3587439954280853, + "learning_rate": 0.000154499103541013, + "loss": 0.4496, + "step": 30304 + }, + { + "epoch": 1.6931865798809955, + "grad_norm": 0.4716564416885376, + "learning_rate": 0.00015447108919766924, + "loss": 0.3829, + "step": 30305 + }, + { + "epoch": 1.6932424504846775, + "grad_norm": 0.7030230760574341, + "learning_rate": 0.00015444307485432541, + "loss": 0.4142, + "step": 30306 + }, + { + "epoch": 1.6932983210883594, + "grad_norm": 0.36654675006866455, + "learning_rate": 0.00015441506051098162, + "loss": 0.3614, + "step": 30307 + }, + { + "epoch": 1.6933541916920412, + "grad_norm": 10.2604398727417, + "learning_rate": 0.00015438704616763783, + "loss": 0.4263, + "step": 30308 + }, + { + "epoch": 1.693410062295723, + "grad_norm": 0.5825942754745483, + "learning_rate": 0.00015435903182429403, + "loss": 0.4205, + "step": 30309 + }, + { + "epoch": 1.693465932899405, + "grad_norm": 0.49981844425201416, + "learning_rate": 0.00015433101748095024, + "loss": 0.4909, + "step": 30310 + }, + { + "epoch": 1.6935218035030868, + "grad_norm": 5.270467758178711, + "learning_rate": 0.00015430300313760644, + "loss": 0.4002, + "step": 30311 + }, + { + "epoch": 1.6935776741067687, + "grad_norm": 0.3440028727054596, + "learning_rate": 0.00015427498879426265, + "loss": 0.3837, + "step": 30312 + }, + { + "epoch": 1.6936335447104507, + "grad_norm": 0.5879929661750793, + "learning_rate": 0.00015424697445091888, + "loss": 0.3239, + "step": 30313 + }, + { + "epoch": 1.6936894153141324, + "grad_norm": 0.3420152962207794, + "learning_rate": 0.0001542189601075751, + "loss": 0.3169, + "step": 30314 + }, + { + "epoch": 1.6937452859178144, + "grad_norm": 0.39380455017089844, + "learning_rate": 0.0001541909457642313, + "loss": 0.3473, + "step": 30315 + }, + { + "epoch": 1.6938011565214963, + "grad_norm": 0.3376863896846771, + "learning_rate": 0.0001541629314208875, + "loss": 0.4217, + "step": 30316 + }, + { + "epoch": 1.693857027125178, + "grad_norm": 0.4390277564525604, + "learning_rate": 0.0001541349170775437, + "loss": 0.4034, + "step": 30317 + }, + { + "epoch": 1.69391289772886, + "grad_norm": 0.31182995438575745, + "learning_rate": 0.0001541069027341999, + "loss": 0.2869, + "step": 30318 + }, + { + "epoch": 1.693968768332542, + "grad_norm": 0.6744735240936279, + "learning_rate": 0.00015407888839085612, + "loss": 0.4214, + "step": 30319 + }, + { + "epoch": 1.6940246389362237, + "grad_norm": 0.4996505081653595, + "learning_rate": 0.00015405087404751232, + "loss": 0.4942, + "step": 30320 + }, + { + "epoch": 1.6940805095399056, + "grad_norm": 0.6289443969726562, + "learning_rate": 0.00015402285970416856, + "loss": 0.4351, + "step": 30321 + }, + { + "epoch": 1.6941363801435876, + "grad_norm": 0.3807026743888855, + "learning_rate": 0.00015399484536082476, + "loss": 0.3219, + "step": 30322 + }, + { + "epoch": 1.6941922507472693, + "grad_norm": 0.5738570690155029, + "learning_rate": 0.00015396683101748097, + "loss": 0.3724, + "step": 30323 + }, + { + "epoch": 1.694248121350951, + "grad_norm": 2.0354628562927246, + "learning_rate": 0.00015393881667413717, + "loss": 0.4623, + "step": 30324 + }, + { + "epoch": 1.6943039919546332, + "grad_norm": 0.37172195315361023, + "learning_rate": 0.00015391080233079338, + "loss": 0.4732, + "step": 30325 + }, + { + "epoch": 1.694359862558315, + "grad_norm": 0.3741808235645294, + "learning_rate": 0.00015388278798744958, + "loss": 0.3708, + "step": 30326 + }, + { + "epoch": 1.6944157331619967, + "grad_norm": 0.35733866691589355, + "learning_rate": 0.00015385477364410576, + "loss": 0.3909, + "step": 30327 + }, + { + "epoch": 1.6944716037656788, + "grad_norm": 0.5491318702697754, + "learning_rate": 0.00015382675930076197, + "loss": 0.452, + "step": 30328 + }, + { + "epoch": 1.6945274743693606, + "grad_norm": 0.8121123313903809, + "learning_rate": 0.0001537987449574182, + "loss": 0.3329, + "step": 30329 + }, + { + "epoch": 1.6945833449730423, + "grad_norm": 5.43250846862793, + "learning_rate": 0.0001537707306140744, + "loss": 0.4496, + "step": 30330 + }, + { + "epoch": 1.6946392155767243, + "grad_norm": 0.39266565442085266, + "learning_rate": 0.0001537427162707306, + "loss": 0.387, + "step": 30331 + }, + { + "epoch": 1.6946950861804062, + "grad_norm": 0.36667248606681824, + "learning_rate": 0.00015371470192738682, + "loss": 0.416, + "step": 30332 + }, + { + "epoch": 1.694750956784088, + "grad_norm": 0.2878878712654114, + "learning_rate": 0.00015368668758404303, + "loss": 0.3675, + "step": 30333 + }, + { + "epoch": 1.6948068273877699, + "grad_norm": 0.43217939138412476, + "learning_rate": 0.00015365867324069923, + "loss": 0.3646, + "step": 30334 + }, + { + "epoch": 1.6948626979914518, + "grad_norm": 3.3187100887298584, + "learning_rate": 0.00015363065889735544, + "loss": 0.4077, + "step": 30335 + }, + { + "epoch": 1.6949185685951336, + "grad_norm": 1.0079652070999146, + "learning_rate": 0.00015360264455401164, + "loss": 0.3813, + "step": 30336 + }, + { + "epoch": 1.6949744391988155, + "grad_norm": 5.863954544067383, + "learning_rate": 0.00015357463021066788, + "loss": 0.4365, + "step": 30337 + }, + { + "epoch": 1.6950303098024975, + "grad_norm": 0.3252738416194916, + "learning_rate": 0.00015354661586732408, + "loss": 0.3862, + "step": 30338 + }, + { + "epoch": 1.6950861804061792, + "grad_norm": 0.9243289232254028, + "learning_rate": 0.0001535186015239803, + "loss": 0.5267, + "step": 30339 + }, + { + "epoch": 1.6951420510098612, + "grad_norm": 0.3849754333496094, + "learning_rate": 0.0001534905871806365, + "loss": 0.3245, + "step": 30340 + }, + { + "epoch": 1.695197921613543, + "grad_norm": 0.6685423851013184, + "learning_rate": 0.0001534625728372927, + "loss": 0.4256, + "step": 30341 + }, + { + "epoch": 1.6952537922172248, + "grad_norm": 0.5088969469070435, + "learning_rate": 0.0001534345584939489, + "loss": 0.4927, + "step": 30342 + }, + { + "epoch": 1.6953096628209068, + "grad_norm": 0.4149470925331116, + "learning_rate": 0.0001534065441506051, + "loss": 0.358, + "step": 30343 + }, + { + "epoch": 1.6953655334245887, + "grad_norm": 1.01585054397583, + "learning_rate": 0.00015337852980726134, + "loss": 0.4743, + "step": 30344 + }, + { + "epoch": 1.6954214040282705, + "grad_norm": 0.3265150189399719, + "learning_rate": 0.00015335051546391755, + "loss": 0.4491, + "step": 30345 + }, + { + "epoch": 1.6954772746319524, + "grad_norm": 0.7967347502708435, + "learning_rate": 0.00015332250112057375, + "loss": 0.3839, + "step": 30346 + }, + { + "epoch": 1.6955331452356344, + "grad_norm": 0.3756546676158905, + "learning_rate": 0.00015329448677722996, + "loss": 0.4484, + "step": 30347 + }, + { + "epoch": 1.695589015839316, + "grad_norm": 1.9909336566925049, + "learning_rate": 0.00015326647243388614, + "loss": 0.4055, + "step": 30348 + }, + { + "epoch": 1.695644886442998, + "grad_norm": 0.45014262199401855, + "learning_rate": 0.00015323845809054235, + "loss": 0.3752, + "step": 30349 + }, + { + "epoch": 1.69570075704668, + "grad_norm": 0.36154812574386597, + "learning_rate": 0.00015321044374719855, + "loss": 0.4364, + "step": 30350 + }, + { + "epoch": 1.6957566276503617, + "grad_norm": 0.4323868155479431, + "learning_rate": 0.00015318242940385476, + "loss": 0.4314, + "step": 30351 + }, + { + "epoch": 1.6958124982540437, + "grad_norm": 0.35654497146606445, + "learning_rate": 0.000153154415060511, + "loss": 0.3038, + "step": 30352 + }, + { + "epoch": 1.6958683688577256, + "grad_norm": 0.4926120340824127, + "learning_rate": 0.0001531264007171672, + "loss": 0.3615, + "step": 30353 + }, + { + "epoch": 1.6959242394614074, + "grad_norm": 1.612534999847412, + "learning_rate": 0.0001530983863738234, + "loss": 0.4129, + "step": 30354 + }, + { + "epoch": 1.6959801100650893, + "grad_norm": 0.5498555302619934, + "learning_rate": 0.0001530703720304796, + "loss": 0.362, + "step": 30355 + }, + { + "epoch": 1.6960359806687713, + "grad_norm": 0.5435191988945007, + "learning_rate": 0.0001530423576871358, + "loss": 0.4618, + "step": 30356 + }, + { + "epoch": 1.696091851272453, + "grad_norm": 1.8126715421676636, + "learning_rate": 0.00015301434334379202, + "loss": 0.4416, + "step": 30357 + }, + { + "epoch": 1.6961477218761347, + "grad_norm": 1.462843894958496, + "learning_rate": 0.00015298632900044822, + "loss": 0.5393, + "step": 30358 + }, + { + "epoch": 1.696203592479817, + "grad_norm": 0.406692773103714, + "learning_rate": 0.00015295831465710443, + "loss": 0.3479, + "step": 30359 + }, + { + "epoch": 1.6962594630834986, + "grad_norm": 0.825909435749054, + "learning_rate": 0.00015293030031376066, + "loss": 0.3517, + "step": 30360 + }, + { + "epoch": 1.6963153336871803, + "grad_norm": 1.6710307598114014, + "learning_rate": 0.00015290228597041687, + "loss": 0.3334, + "step": 30361 + }, + { + "epoch": 1.6963712042908625, + "grad_norm": 0.42490291595458984, + "learning_rate": 0.00015287427162707308, + "loss": 0.4306, + "step": 30362 + }, + { + "epoch": 1.6964270748945443, + "grad_norm": 0.6935012340545654, + "learning_rate": 0.00015284625728372928, + "loss": 0.5152, + "step": 30363 + }, + { + "epoch": 1.696482945498226, + "grad_norm": 0.38830575346946716, + "learning_rate": 0.0001528182429403855, + "loss": 0.4628, + "step": 30364 + }, + { + "epoch": 1.696538816101908, + "grad_norm": 8.830348014831543, + "learning_rate": 0.0001527902285970417, + "loss": 0.4329, + "step": 30365 + }, + { + "epoch": 1.6965946867055899, + "grad_norm": 0.6677039265632629, + "learning_rate": 0.0001527622142536979, + "loss": 0.4174, + "step": 30366 + }, + { + "epoch": 1.6966505573092716, + "grad_norm": 0.6529161334037781, + "learning_rate": 0.0001527341999103541, + "loss": 0.4631, + "step": 30367 + }, + { + "epoch": 1.6967064279129536, + "grad_norm": 0.37171506881713867, + "learning_rate": 0.00015270618556701034, + "loss": 0.2912, + "step": 30368 + }, + { + "epoch": 1.6967622985166355, + "grad_norm": 1.7440308332443237, + "learning_rate": 0.00015267817122366652, + "loss": 0.4765, + "step": 30369 + }, + { + "epoch": 1.6968181691203172, + "grad_norm": 0.43052539229393005, + "learning_rate": 0.00015265015688032272, + "loss": 0.3973, + "step": 30370 + }, + { + "epoch": 1.6968740397239992, + "grad_norm": 0.35471266508102417, + "learning_rate": 0.00015262214253697893, + "loss": 0.3659, + "step": 30371 + }, + { + "epoch": 1.6969299103276811, + "grad_norm": 0.4098668694496155, + "learning_rate": 0.00015259412819363513, + "loss": 0.3713, + "step": 30372 + }, + { + "epoch": 1.6969857809313629, + "grad_norm": 0.9831579327583313, + "learning_rate": 0.00015256611385029134, + "loss": 0.3769, + "step": 30373 + }, + { + "epoch": 1.6970416515350448, + "grad_norm": 0.39818623661994934, + "learning_rate": 0.00015253809950694754, + "loss": 0.332, + "step": 30374 + }, + { + "epoch": 1.6970975221387268, + "grad_norm": 0.6117801666259766, + "learning_rate": 0.00015251008516360375, + "loss": 0.4578, + "step": 30375 + }, + { + "epoch": 1.6971533927424085, + "grad_norm": 0.4084908664226532, + "learning_rate": 0.00015248207082025998, + "loss": 0.5496, + "step": 30376 + }, + { + "epoch": 1.6972092633460905, + "grad_norm": 0.41958922147750854, + "learning_rate": 0.0001524540564769162, + "loss": 0.4086, + "step": 30377 + }, + { + "epoch": 1.6972651339497724, + "grad_norm": 1.5312836170196533, + "learning_rate": 0.0001524260421335724, + "loss": 0.4985, + "step": 30378 + }, + { + "epoch": 1.6973210045534541, + "grad_norm": 0.33895301818847656, + "learning_rate": 0.0001523980277902286, + "loss": 0.3971, + "step": 30379 + }, + { + "epoch": 1.697376875157136, + "grad_norm": 0.587413489818573, + "learning_rate": 0.0001523700134468848, + "loss": 0.4675, + "step": 30380 + }, + { + "epoch": 1.697432745760818, + "grad_norm": 0.5109254717826843, + "learning_rate": 0.000152341999103541, + "loss": 0.4243, + "step": 30381 + }, + { + "epoch": 1.6974886163644998, + "grad_norm": 0.49110350012779236, + "learning_rate": 0.00015231398476019722, + "loss": 0.533, + "step": 30382 + }, + { + "epoch": 1.6975444869681817, + "grad_norm": 0.7376829385757446, + "learning_rate": 0.00015228597041685342, + "loss": 0.6992, + "step": 30383 + }, + { + "epoch": 1.6976003575718637, + "grad_norm": 0.326322078704834, + "learning_rate": 0.00015225795607350966, + "loss": 0.3779, + "step": 30384 + }, + { + "epoch": 1.6976562281755454, + "grad_norm": 0.3276446759700775, + "learning_rate": 0.00015222994173016586, + "loss": 0.387, + "step": 30385 + }, + { + "epoch": 1.6977120987792274, + "grad_norm": 0.35885992646217346, + "learning_rate": 0.00015220192738682207, + "loss": 0.3715, + "step": 30386 + }, + { + "epoch": 1.6977679693829093, + "grad_norm": 0.40151727199554443, + "learning_rate": 0.00015217391304347827, + "loss": 0.5373, + "step": 30387 + }, + { + "epoch": 1.697823839986591, + "grad_norm": 0.3430540859699249, + "learning_rate": 0.00015214589870013448, + "loss": 0.3238, + "step": 30388 + }, + { + "epoch": 1.6978797105902728, + "grad_norm": 1.0789892673492432, + "learning_rate": 0.00015211788435679069, + "loss": 0.4302, + "step": 30389 + }, + { + "epoch": 1.697935581193955, + "grad_norm": 0.409359335899353, + "learning_rate": 0.0001520898700134469, + "loss": 0.4199, + "step": 30390 + }, + { + "epoch": 1.6979914517976367, + "grad_norm": 0.49697235226631165, + "learning_rate": 0.00015206185567010307, + "loss": 0.4963, + "step": 30391 + }, + { + "epoch": 1.6980473224013184, + "grad_norm": 0.3443050980567932, + "learning_rate": 0.0001520338413267593, + "loss": 0.3079, + "step": 30392 + }, + { + "epoch": 1.6981031930050006, + "grad_norm": 0.3537832200527191, + "learning_rate": 0.0001520058269834155, + "loss": 0.3375, + "step": 30393 + }, + { + "epoch": 1.6981590636086823, + "grad_norm": 0.43615561723709106, + "learning_rate": 0.00015197781264007172, + "loss": 0.3735, + "step": 30394 + }, + { + "epoch": 1.698214934212364, + "grad_norm": 0.35180726647377014, + "learning_rate": 0.00015194979829672792, + "loss": 0.3007, + "step": 30395 + }, + { + "epoch": 1.6982708048160462, + "grad_norm": 0.40769463777542114, + "learning_rate": 0.00015192178395338413, + "loss": 0.4616, + "step": 30396 + }, + { + "epoch": 1.698326675419728, + "grad_norm": 0.37299633026123047, + "learning_rate": 0.00015189376961004033, + "loss": 0.3428, + "step": 30397 + }, + { + "epoch": 1.6983825460234097, + "grad_norm": 0.3832649290561676, + "learning_rate": 0.00015186575526669654, + "loss": 0.3916, + "step": 30398 + }, + { + "epoch": 1.6984384166270916, + "grad_norm": 0.4310757517814636, + "learning_rate": 0.00015183774092335274, + "loss": 0.3459, + "step": 30399 + }, + { + "epoch": 1.6984942872307736, + "grad_norm": 0.6562264561653137, + "learning_rate": 0.00015180972658000898, + "loss": 0.3348, + "step": 30400 + }, + { + "epoch": 1.6985501578344553, + "grad_norm": 0.3869490623474121, + "learning_rate": 0.00015178171223666518, + "loss": 0.4021, + "step": 30401 + }, + { + "epoch": 1.6986060284381372, + "grad_norm": 0.6189430356025696, + "learning_rate": 0.0001517536978933214, + "loss": 0.458, + "step": 30402 + }, + { + "epoch": 1.6986618990418192, + "grad_norm": 0.46955323219299316, + "learning_rate": 0.0001517256835499776, + "loss": 0.4565, + "step": 30403 + }, + { + "epoch": 1.698717769645501, + "grad_norm": 0.7729200124740601, + "learning_rate": 0.0001516976692066338, + "loss": 0.3433, + "step": 30404 + }, + { + "epoch": 1.6987736402491829, + "grad_norm": 0.3853016793727875, + "learning_rate": 0.00015166965486329, + "loss": 0.4464, + "step": 30405 + }, + { + "epoch": 1.6988295108528648, + "grad_norm": 0.3913209140300751, + "learning_rate": 0.0001516416405199462, + "loss": 0.5322, + "step": 30406 + }, + { + "epoch": 1.6988853814565466, + "grad_norm": 0.5638818740844727, + "learning_rate": 0.00015161362617660245, + "loss": 0.4116, + "step": 30407 + }, + { + "epoch": 1.6989412520602285, + "grad_norm": 0.4787980020046234, + "learning_rate": 0.00015158561183325865, + "loss": 0.4008, + "step": 30408 + }, + { + "epoch": 1.6989971226639105, + "grad_norm": 0.35803067684173584, + "learning_rate": 0.00015155759748991486, + "loss": 0.4751, + "step": 30409 + }, + { + "epoch": 1.6990529932675922, + "grad_norm": 0.5117636919021606, + "learning_rate": 0.00015152958314657106, + "loss": 0.5116, + "step": 30410 + }, + { + "epoch": 1.6991088638712741, + "grad_norm": 0.66380375623703, + "learning_rate": 0.00015150156880322727, + "loss": 0.5001, + "step": 30411 + }, + { + "epoch": 1.699164734474956, + "grad_norm": 0.36482787132263184, + "learning_rate": 0.00015147355445988345, + "loss": 0.3498, + "step": 30412 + }, + { + "epoch": 1.6992206050786378, + "grad_norm": 0.3723302483558655, + "learning_rate": 0.00015144554011653965, + "loss": 0.3688, + "step": 30413 + }, + { + "epoch": 1.6992764756823198, + "grad_norm": 0.600544273853302, + "learning_rate": 0.00015141752577319586, + "loss": 0.4664, + "step": 30414 + }, + { + "epoch": 1.6993323462860017, + "grad_norm": 0.8671201467514038, + "learning_rate": 0.0001513895114298521, + "loss": 0.5089, + "step": 30415 + }, + { + "epoch": 1.6993882168896834, + "grad_norm": 0.5396139025688171, + "learning_rate": 0.0001513614970865083, + "loss": 0.4514, + "step": 30416 + }, + { + "epoch": 1.6994440874933654, + "grad_norm": 2.279364824295044, + "learning_rate": 0.0001513334827431645, + "loss": 0.4216, + "step": 30417 + }, + { + "epoch": 1.6994999580970473, + "grad_norm": 0.392316997051239, + "learning_rate": 0.0001513054683998207, + "loss": 0.4079, + "step": 30418 + }, + { + "epoch": 1.699555828700729, + "grad_norm": 0.3511005938053131, + "learning_rate": 0.00015127745405647691, + "loss": 0.403, + "step": 30419 + }, + { + "epoch": 1.699611699304411, + "grad_norm": 0.6613070368766785, + "learning_rate": 0.00015124943971313312, + "loss": 0.4302, + "step": 30420 + }, + { + "epoch": 1.699667569908093, + "grad_norm": 0.4035854637622833, + "learning_rate": 0.00015122142536978933, + "loss": 0.5283, + "step": 30421 + }, + { + "epoch": 1.6997234405117747, + "grad_norm": 0.4780261218547821, + "learning_rate": 0.00015119341102644553, + "loss": 0.4357, + "step": 30422 + }, + { + "epoch": 1.6997793111154564, + "grad_norm": 0.6800615191459656, + "learning_rate": 0.00015116539668310177, + "loss": 0.4393, + "step": 30423 + }, + { + "epoch": 1.6998351817191386, + "grad_norm": 0.5540453791618347, + "learning_rate": 0.00015113738233975797, + "loss": 0.3778, + "step": 30424 + }, + { + "epoch": 1.6998910523228203, + "grad_norm": 0.5310530662536621, + "learning_rate": 0.00015110936799641418, + "loss": 0.3842, + "step": 30425 + }, + { + "epoch": 1.699946922926502, + "grad_norm": 4.628725528717041, + "learning_rate": 0.00015108135365307038, + "loss": 0.3627, + "step": 30426 + }, + { + "epoch": 1.7000027935301842, + "grad_norm": 0.8188226819038391, + "learning_rate": 0.0001510533393097266, + "loss": 0.4201, + "step": 30427 + }, + { + "epoch": 1.700058664133866, + "grad_norm": 0.3502201437950134, + "learning_rate": 0.0001510253249663828, + "loss": 0.4676, + "step": 30428 + }, + { + "epoch": 1.7001145347375477, + "grad_norm": 0.4763847291469574, + "learning_rate": 0.000150997310623039, + "loss": 0.3467, + "step": 30429 + }, + { + "epoch": 1.7001704053412299, + "grad_norm": 0.40232449769973755, + "learning_rate": 0.0001509692962796952, + "loss": 0.3584, + "step": 30430 + }, + { + "epoch": 1.7002262759449116, + "grad_norm": 0.5496158003807068, + "learning_rate": 0.00015094128193635144, + "loss": 0.3946, + "step": 30431 + }, + { + "epoch": 1.7002821465485933, + "grad_norm": 0.5943741798400879, + "learning_rate": 0.00015091326759300764, + "loss": 0.3991, + "step": 30432 + }, + { + "epoch": 1.7003380171522753, + "grad_norm": 0.4094848930835724, + "learning_rate": 0.00015088525324966382, + "loss": 0.4323, + "step": 30433 + }, + { + "epoch": 1.7003938877559572, + "grad_norm": 0.37243586778640747, + "learning_rate": 0.00015085723890632003, + "loss": 0.3677, + "step": 30434 + }, + { + "epoch": 1.700449758359639, + "grad_norm": 0.342081755399704, + "learning_rate": 0.00015082922456297624, + "loss": 0.3086, + "step": 30435 + }, + { + "epoch": 1.700505628963321, + "grad_norm": 0.377538800239563, + "learning_rate": 0.00015080121021963244, + "loss": 0.3541, + "step": 30436 + }, + { + "epoch": 1.7005614995670029, + "grad_norm": 0.44950351119041443, + "learning_rate": 0.00015077319587628865, + "loss": 0.35, + "step": 30437 + }, + { + "epoch": 1.7006173701706846, + "grad_norm": 0.39687466621398926, + "learning_rate": 0.00015074518153294485, + "loss": 0.4183, + "step": 30438 + }, + { + "epoch": 1.7006732407743665, + "grad_norm": 2.5767247676849365, + "learning_rate": 0.00015071716718960109, + "loss": 0.4114, + "step": 30439 + }, + { + "epoch": 1.7007291113780485, + "grad_norm": 0.40671706199645996, + "learning_rate": 0.0001506891528462573, + "loss": 0.3717, + "step": 30440 + }, + { + "epoch": 1.7007849819817302, + "grad_norm": 0.3835434317588806, + "learning_rate": 0.0001506611385029135, + "loss": 0.4525, + "step": 30441 + }, + { + "epoch": 1.7008408525854122, + "grad_norm": 0.6575836539268494, + "learning_rate": 0.0001506331241595697, + "loss": 0.3572, + "step": 30442 + }, + { + "epoch": 1.7008967231890941, + "grad_norm": 0.3787059187889099, + "learning_rate": 0.0001506051098162259, + "loss": 0.4236, + "step": 30443 + }, + { + "epoch": 1.7009525937927759, + "grad_norm": 1.42276930809021, + "learning_rate": 0.00015057709547288211, + "loss": 0.4514, + "step": 30444 + }, + { + "epoch": 1.7010084643964578, + "grad_norm": 0.4651338458061218, + "learning_rate": 0.00015054908112953832, + "loss": 0.3915, + "step": 30445 + }, + { + "epoch": 1.7010643350001398, + "grad_norm": 0.35057494044303894, + "learning_rate": 0.00015052106678619453, + "loss": 0.3131, + "step": 30446 + }, + { + "epoch": 1.7011202056038215, + "grad_norm": 0.5275206565856934, + "learning_rate": 0.00015049305244285076, + "loss": 0.5102, + "step": 30447 + }, + { + "epoch": 1.7011760762075034, + "grad_norm": 0.3157897889614105, + "learning_rate": 0.00015046503809950696, + "loss": 0.3676, + "step": 30448 + }, + { + "epoch": 1.7012319468111854, + "grad_norm": 0.33730223774909973, + "learning_rate": 0.00015043702375616317, + "loss": 0.3567, + "step": 30449 + }, + { + "epoch": 1.7012878174148671, + "grad_norm": 0.3158342242240906, + "learning_rate": 0.00015040900941281938, + "loss": 0.3307, + "step": 30450 + }, + { + "epoch": 1.701343688018549, + "grad_norm": 0.29490238428115845, + "learning_rate": 0.00015038099506947558, + "loss": 0.44, + "step": 30451 + }, + { + "epoch": 1.701399558622231, + "grad_norm": 0.3417869508266449, + "learning_rate": 0.0001503529807261318, + "loss": 0.3274, + "step": 30452 + }, + { + "epoch": 1.7014554292259128, + "grad_norm": 0.3958129286766052, + "learning_rate": 0.000150324966382788, + "loss": 0.45, + "step": 30453 + }, + { + "epoch": 1.7015112998295947, + "grad_norm": 0.34467557072639465, + "learning_rate": 0.00015029695203944417, + "loss": 0.4978, + "step": 30454 + }, + { + "epoch": 1.7015671704332767, + "grad_norm": 1.6116797924041748, + "learning_rate": 0.0001502689376961004, + "loss": 0.3879, + "step": 30455 + }, + { + "epoch": 1.7016230410369584, + "grad_norm": 7.370880126953125, + "learning_rate": 0.0001502409233527566, + "loss": 0.3335, + "step": 30456 + }, + { + "epoch": 1.7016789116406401, + "grad_norm": 4.147916316986084, + "learning_rate": 0.00015021290900941282, + "loss": 0.3407, + "step": 30457 + }, + { + "epoch": 1.7017347822443223, + "grad_norm": 0.4637536108493805, + "learning_rate": 0.00015018489466606902, + "loss": 0.3901, + "step": 30458 + }, + { + "epoch": 1.701790652848004, + "grad_norm": 0.38868477940559387, + "learning_rate": 0.00015015688032272523, + "loss": 0.4973, + "step": 30459 + }, + { + "epoch": 1.7018465234516857, + "grad_norm": 0.32605910301208496, + "learning_rate": 0.00015012886597938143, + "loss": 0.3698, + "step": 30460 + }, + { + "epoch": 1.701902394055368, + "grad_norm": 0.7017481923103333, + "learning_rate": 0.00015010085163603764, + "loss": 0.5037, + "step": 30461 + }, + { + "epoch": 1.7019582646590496, + "grad_norm": 0.566625714302063, + "learning_rate": 0.00015007283729269385, + "loss": 0.3092, + "step": 30462 + }, + { + "epoch": 1.7020141352627314, + "grad_norm": 0.5380405187606812, + "learning_rate": 0.00015004482294935008, + "loss": 0.4895, + "step": 30463 + }, + { + "epoch": 1.7020700058664135, + "grad_norm": 0.5867582559585571, + "learning_rate": 0.00015001680860600629, + "loss": 0.3085, + "step": 30464 + }, + { + "epoch": 1.7021258764700953, + "grad_norm": 0.6122711300849915, + "learning_rate": 0.0001499887942626625, + "loss": 0.3867, + "step": 30465 + }, + { + "epoch": 1.702181747073777, + "grad_norm": 0.5232241749763489, + "learning_rate": 0.0001499607799193187, + "loss": 0.4396, + "step": 30466 + }, + { + "epoch": 1.702237617677459, + "grad_norm": 0.5926331877708435, + "learning_rate": 0.0001499327655759749, + "loss": 0.4608, + "step": 30467 + }, + { + "epoch": 1.702293488281141, + "grad_norm": 0.9176285266876221, + "learning_rate": 0.0001499047512326311, + "loss": 0.3669, + "step": 30468 + }, + { + "epoch": 1.7023493588848226, + "grad_norm": 0.5388416647911072, + "learning_rate": 0.00014987673688928731, + "loss": 0.3892, + "step": 30469 + }, + { + "epoch": 1.7024052294885046, + "grad_norm": 1.4141514301300049, + "learning_rate": 0.00014984872254594352, + "loss": 0.5033, + "step": 30470 + }, + { + "epoch": 1.7024611000921865, + "grad_norm": 0.3582724332809448, + "learning_rate": 0.00014982070820259975, + "loss": 0.4547, + "step": 30471 + }, + { + "epoch": 1.7025169706958683, + "grad_norm": 0.8496718406677246, + "learning_rate": 0.00014979269385925596, + "loss": 0.4458, + "step": 30472 + }, + { + "epoch": 1.7025728412995502, + "grad_norm": 0.35656848549842834, + "learning_rate": 0.00014976467951591216, + "loss": 0.3512, + "step": 30473 + }, + { + "epoch": 1.7026287119032322, + "grad_norm": 0.6273221969604492, + "learning_rate": 0.00014973666517256837, + "loss": 0.5536, + "step": 30474 + }, + { + "epoch": 1.702684582506914, + "grad_norm": 0.45268386602401733, + "learning_rate": 0.00014970865082922458, + "loss": 0.3515, + "step": 30475 + }, + { + "epoch": 1.7027404531105959, + "grad_norm": 0.3475625813007355, + "learning_rate": 0.00014968063648588075, + "loss": 0.439, + "step": 30476 + }, + { + "epoch": 1.7027963237142778, + "grad_norm": 3.1492807865142822, + "learning_rate": 0.00014965262214253696, + "loss": 0.4456, + "step": 30477 + }, + { + "epoch": 1.7028521943179595, + "grad_norm": 0.3940126299858093, + "learning_rate": 0.00014962460779919317, + "loss": 0.2957, + "step": 30478 + }, + { + "epoch": 1.7029080649216415, + "grad_norm": 0.3103237450122833, + "learning_rate": 0.0001495965934558494, + "loss": 0.4566, + "step": 30479 + }, + { + "epoch": 1.7029639355253234, + "grad_norm": 0.6683543920516968, + "learning_rate": 0.0001495685791125056, + "loss": 0.5789, + "step": 30480 + }, + { + "epoch": 1.7030198061290052, + "grad_norm": 1.5138847827911377, + "learning_rate": 0.0001495405647691618, + "loss": 0.3875, + "step": 30481 + }, + { + "epoch": 1.7030756767326871, + "grad_norm": 1.0969135761260986, + "learning_rate": 0.00014951255042581802, + "loss": 0.3101, + "step": 30482 + }, + { + "epoch": 1.703131547336369, + "grad_norm": 0.4217999577522278, + "learning_rate": 0.00014948453608247422, + "loss": 0.5697, + "step": 30483 + }, + { + "epoch": 1.7031874179400508, + "grad_norm": 0.27032989263534546, + "learning_rate": 0.00014945652173913043, + "loss": 0.3419, + "step": 30484 + }, + { + "epoch": 1.7032432885437327, + "grad_norm": 0.4077684283256531, + "learning_rate": 0.00014942850739578663, + "loss": 0.4503, + "step": 30485 + }, + { + "epoch": 1.7032991591474147, + "grad_norm": 0.95347660779953, + "learning_rate": 0.00014940049305244287, + "loss": 0.4118, + "step": 30486 + }, + { + "epoch": 1.7033550297510964, + "grad_norm": 0.3571563959121704, + "learning_rate": 0.00014937247870909907, + "loss": 0.4218, + "step": 30487 + }, + { + "epoch": 1.7034109003547784, + "grad_norm": 1.1797372102737427, + "learning_rate": 0.00014934446436575528, + "loss": 0.4256, + "step": 30488 + }, + { + "epoch": 1.7034667709584603, + "grad_norm": 0.37864595651626587, + "learning_rate": 0.00014931645002241148, + "loss": 0.3878, + "step": 30489 + }, + { + "epoch": 1.703522641562142, + "grad_norm": 4.663007736206055, + "learning_rate": 0.0001492884356790677, + "loss": 0.4093, + "step": 30490 + }, + { + "epoch": 1.7035785121658238, + "grad_norm": 0.6239718794822693, + "learning_rate": 0.0001492604213357239, + "loss": 0.3895, + "step": 30491 + }, + { + "epoch": 1.703634382769506, + "grad_norm": 0.37927716970443726, + "learning_rate": 0.0001492324069923801, + "loss": 0.3647, + "step": 30492 + }, + { + "epoch": 1.7036902533731877, + "grad_norm": 0.4442436993122101, + "learning_rate": 0.0001492043926490363, + "loss": 0.3883, + "step": 30493 + }, + { + "epoch": 1.7037461239768694, + "grad_norm": 0.33610448241233826, + "learning_rate": 0.00014917637830569254, + "loss": 0.4064, + "step": 30494 + }, + { + "epoch": 1.7038019945805516, + "grad_norm": 0.42020702362060547, + "learning_rate": 0.00014914836396234875, + "loss": 0.4646, + "step": 30495 + }, + { + "epoch": 1.7038578651842333, + "grad_norm": 0.3652844727039337, + "learning_rate": 0.00014912034961900495, + "loss": 0.4081, + "step": 30496 + }, + { + "epoch": 1.703913735787915, + "grad_norm": 0.6557400226593018, + "learning_rate": 0.00014909233527566113, + "loss": 0.4276, + "step": 30497 + }, + { + "epoch": 1.7039696063915972, + "grad_norm": 0.4769832491874695, + "learning_rate": 0.00014906432093231734, + "loss": 0.3767, + "step": 30498 + }, + { + "epoch": 1.704025476995279, + "grad_norm": 0.4585017263889313, + "learning_rate": 0.00014903630658897354, + "loss": 0.4124, + "step": 30499 + }, + { + "epoch": 1.7040813475989607, + "grad_norm": 0.7475141286849976, + "learning_rate": 0.00014900829224562975, + "loss": 0.3492, + "step": 30500 + }, + { + "epoch": 1.7040813475989607, + "eval_cer": 0.08397433799220974, + "eval_loss": 0.3146584928035736, + "eval_runtime": 56.2037, + "eval_samples_per_second": 80.742, + "eval_steps_per_second": 5.053, + "eval_wer": 0.3337336945895366, + "step": 30500 + }, + { + "epoch": 1.7041372182026426, + "grad_norm": 0.5668635368347168, + "learning_rate": 0.00014898027790228595, + "loss": 0.3935, + "step": 30501 + }, + { + "epoch": 1.7041930888063246, + "grad_norm": 0.5362403988838196, + "learning_rate": 0.0001489522635589422, + "loss": 0.461, + "step": 30502 + }, + { + "epoch": 1.7042489594100063, + "grad_norm": 0.27607500553131104, + "learning_rate": 0.0001489242492155984, + "loss": 0.3029, + "step": 30503 + }, + { + "epoch": 1.7043048300136883, + "grad_norm": 0.4116414785385132, + "learning_rate": 0.0001488962348722546, + "loss": 0.3971, + "step": 30504 + }, + { + "epoch": 1.7043607006173702, + "grad_norm": 0.45032498240470886, + "learning_rate": 0.0001488682205289108, + "loss": 0.4387, + "step": 30505 + }, + { + "epoch": 1.704416571221052, + "grad_norm": 0.43898943066596985, + "learning_rate": 0.000148840206185567, + "loss": 0.3919, + "step": 30506 + }, + { + "epoch": 1.704472441824734, + "grad_norm": 0.41493335366249084, + "learning_rate": 0.00014881219184222322, + "loss": 0.4366, + "step": 30507 + }, + { + "epoch": 1.7045283124284158, + "grad_norm": 0.3902187943458557, + "learning_rate": 0.00014878417749887942, + "loss": 0.3712, + "step": 30508 + }, + { + "epoch": 1.7045841830320976, + "grad_norm": 0.4263065755367279, + "learning_rate": 0.00014875616315553563, + "loss": 0.5514, + "step": 30509 + }, + { + "epoch": 1.7046400536357795, + "grad_norm": 0.32173290848731995, + "learning_rate": 0.00014872814881219186, + "loss": 0.3741, + "step": 30510 + }, + { + "epoch": 1.7046959242394615, + "grad_norm": 0.3143075406551361, + "learning_rate": 0.00014870013446884807, + "loss": 0.3663, + "step": 30511 + }, + { + "epoch": 1.7047517948431432, + "grad_norm": 0.5591047406196594, + "learning_rate": 0.00014867212012550427, + "loss": 0.3953, + "step": 30512 + }, + { + "epoch": 1.7048076654468252, + "grad_norm": 0.4502909183502197, + "learning_rate": 0.00014864410578216048, + "loss": 0.3763, + "step": 30513 + }, + { + "epoch": 1.7048635360505071, + "grad_norm": 0.6000884175300598, + "learning_rate": 0.00014861609143881668, + "loss": 0.4512, + "step": 30514 + }, + { + "epoch": 1.7049194066541888, + "grad_norm": 0.4062427580356598, + "learning_rate": 0.0001485880770954729, + "loss": 0.5079, + "step": 30515 + }, + { + "epoch": 1.7049752772578708, + "grad_norm": 0.8211089968681335, + "learning_rate": 0.0001485600627521291, + "loss": 0.5394, + "step": 30516 + }, + { + "epoch": 1.7050311478615527, + "grad_norm": 0.35768818855285645, + "learning_rate": 0.0001485320484087853, + "loss": 0.4533, + "step": 30517 + }, + { + "epoch": 1.7050870184652345, + "grad_norm": 0.3686259388923645, + "learning_rate": 0.0001485040340654415, + "loss": 0.2903, + "step": 30518 + }, + { + "epoch": 1.7051428890689164, + "grad_norm": 0.2993147373199463, + "learning_rate": 0.0001484760197220977, + "loss": 0.2917, + "step": 30519 + }, + { + "epoch": 1.7051987596725984, + "grad_norm": 0.43942153453826904, + "learning_rate": 0.00014844800537875392, + "loss": 0.433, + "step": 30520 + }, + { + "epoch": 1.70525463027628, + "grad_norm": 0.5125871300697327, + "learning_rate": 0.00014841999103541012, + "loss": 0.4065, + "step": 30521 + }, + { + "epoch": 1.705310500879962, + "grad_norm": 0.4130726754665375, + "learning_rate": 0.00014839197669206633, + "loss": 0.4068, + "step": 30522 + }, + { + "epoch": 1.705366371483644, + "grad_norm": 0.4576214551925659, + "learning_rate": 0.00014836396234872254, + "loss": 0.2913, + "step": 30523 + }, + { + "epoch": 1.7054222420873257, + "grad_norm": 0.36292463541030884, + "learning_rate": 0.00014833594800537874, + "loss": 0.473, + "step": 30524 + }, + { + "epoch": 1.7054781126910075, + "grad_norm": 0.34474611282348633, + "learning_rate": 0.00014830793366203495, + "loss": 0.4382, + "step": 30525 + }, + { + "epoch": 1.7055339832946896, + "grad_norm": 0.3495814800262451, + "learning_rate": 0.00014827991931869118, + "loss": 0.4553, + "step": 30526 + }, + { + "epoch": 1.7055898538983714, + "grad_norm": 0.4228653907775879, + "learning_rate": 0.0001482519049753474, + "loss": 0.3889, + "step": 30527 + }, + { + "epoch": 1.705645724502053, + "grad_norm": 0.3421931564807892, + "learning_rate": 0.0001482238906320036, + "loss": 0.3637, + "step": 30528 + }, + { + "epoch": 1.7057015951057353, + "grad_norm": 1.1861724853515625, + "learning_rate": 0.0001481958762886598, + "loss": 0.3983, + "step": 30529 + }, + { + "epoch": 1.705757465709417, + "grad_norm": 0.3716696798801422, + "learning_rate": 0.000148167861945316, + "loss": 0.4183, + "step": 30530 + }, + { + "epoch": 1.7058133363130987, + "grad_norm": 0.4056450128555298, + "learning_rate": 0.0001481398476019722, + "loss": 0.3814, + "step": 30531 + }, + { + "epoch": 1.705869206916781, + "grad_norm": 0.5590280294418335, + "learning_rate": 0.00014811183325862842, + "loss": 0.4639, + "step": 30532 + }, + { + "epoch": 1.7059250775204626, + "grad_norm": 0.3773571848869324, + "learning_rate": 0.00014808381891528462, + "loss": 0.4071, + "step": 30533 + }, + { + "epoch": 1.7059809481241444, + "grad_norm": 0.4802830219268799, + "learning_rate": 0.00014805580457194085, + "loss": 0.4423, + "step": 30534 + }, + { + "epoch": 1.7060368187278263, + "grad_norm": 0.38279202580451965, + "learning_rate": 0.00014802779022859706, + "loss": 0.4557, + "step": 30535 + }, + { + "epoch": 1.7060926893315083, + "grad_norm": 0.995572030544281, + "learning_rate": 0.00014799977588525327, + "loss": 0.5092, + "step": 30536 + }, + { + "epoch": 1.70614855993519, + "grad_norm": 1.519018292427063, + "learning_rate": 0.00014797176154190947, + "loss": 0.4748, + "step": 30537 + }, + { + "epoch": 1.706204430538872, + "grad_norm": 0.6286140084266663, + "learning_rate": 0.00014794374719856568, + "loss": 0.476, + "step": 30538 + }, + { + "epoch": 1.706260301142554, + "grad_norm": 0.4391116797924042, + "learning_rate": 0.00014791573285522186, + "loss": 0.3952, + "step": 30539 + }, + { + "epoch": 1.7063161717462356, + "grad_norm": 0.4601932466030121, + "learning_rate": 0.00014788771851187806, + "loss": 0.3021, + "step": 30540 + }, + { + "epoch": 1.7063720423499176, + "grad_norm": 0.452872633934021, + "learning_rate": 0.00014785970416853427, + "loss": 0.3529, + "step": 30541 + }, + { + "epoch": 1.7064279129535995, + "grad_norm": 7.471056938171387, + "learning_rate": 0.0001478316898251905, + "loss": 0.3706, + "step": 30542 + }, + { + "epoch": 1.7064837835572813, + "grad_norm": 0.43222057819366455, + "learning_rate": 0.0001478036754818467, + "loss": 0.4087, + "step": 30543 + }, + { + "epoch": 1.7065396541609632, + "grad_norm": 1.8818633556365967, + "learning_rate": 0.0001477756611385029, + "loss": 0.3758, + "step": 30544 + }, + { + "epoch": 1.7065955247646452, + "grad_norm": 0.37979865074157715, + "learning_rate": 0.00014774764679515912, + "loss": 0.4002, + "step": 30545 + }, + { + "epoch": 1.7066513953683269, + "grad_norm": 0.38990581035614014, + "learning_rate": 0.00014771963245181532, + "loss": 0.3943, + "step": 30546 + }, + { + "epoch": 1.7067072659720088, + "grad_norm": 0.8522530794143677, + "learning_rate": 0.00014769161810847153, + "loss": 0.3414, + "step": 30547 + }, + { + "epoch": 1.7067631365756908, + "grad_norm": 0.48590147495269775, + "learning_rate": 0.00014766360376512774, + "loss": 0.4146, + "step": 30548 + }, + { + "epoch": 1.7068190071793725, + "grad_norm": 0.41926881670951843, + "learning_rate": 0.00014763558942178397, + "loss": 0.4427, + "step": 30549 + }, + { + "epoch": 1.7068748777830545, + "grad_norm": 0.5050707459449768, + "learning_rate": 0.00014760757507844017, + "loss": 0.3889, + "step": 30550 + }, + { + "epoch": 1.7069307483867364, + "grad_norm": 0.46023350954055786, + "learning_rate": 0.00014757956073509638, + "loss": 0.4144, + "step": 30551 + }, + { + "epoch": 1.7069866189904181, + "grad_norm": 0.395370751619339, + "learning_rate": 0.0001475515463917526, + "loss": 0.3565, + "step": 30552 + }, + { + "epoch": 1.7070424895941, + "grad_norm": 0.35656455159187317, + "learning_rate": 0.0001475235320484088, + "loss": 0.4111, + "step": 30553 + }, + { + "epoch": 1.707098360197782, + "grad_norm": 0.40690016746520996, + "learning_rate": 0.000147495517705065, + "loss": 0.4383, + "step": 30554 + }, + { + "epoch": 1.7071542308014638, + "grad_norm": 0.4190275967121124, + "learning_rate": 0.0001474675033617212, + "loss": 0.3001, + "step": 30555 + }, + { + "epoch": 1.7072101014051457, + "grad_norm": 0.31275880336761475, + "learning_rate": 0.0001474394890183774, + "loss": 0.3595, + "step": 30556 + }, + { + "epoch": 1.7072659720088277, + "grad_norm": 0.35446789860725403, + "learning_rate": 0.00014741147467503364, + "loss": 0.3529, + "step": 30557 + }, + { + "epoch": 1.7073218426125094, + "grad_norm": 0.36850690841674805, + "learning_rate": 0.00014738346033168985, + "loss": 0.4513, + "step": 30558 + }, + { + "epoch": 1.7073777132161911, + "grad_norm": 0.4821092486381531, + "learning_rate": 0.00014735544598834605, + "loss": 0.3807, + "step": 30559 + }, + { + "epoch": 1.7074335838198733, + "grad_norm": 0.33954039216041565, + "learning_rate": 0.00014732743164500226, + "loss": 0.3807, + "step": 30560 + }, + { + "epoch": 1.707489454423555, + "grad_norm": 0.5959884524345398, + "learning_rate": 0.00014729941730165844, + "loss": 0.4948, + "step": 30561 + }, + { + "epoch": 1.7075453250272368, + "grad_norm": 0.38748136162757874, + "learning_rate": 0.00014727140295831464, + "loss": 0.4741, + "step": 30562 + }, + { + "epoch": 1.707601195630919, + "grad_norm": 0.34462687373161316, + "learning_rate": 0.00014724338861497085, + "loss": 0.4357, + "step": 30563 + }, + { + "epoch": 1.7076570662346007, + "grad_norm": 0.3861338794231415, + "learning_rate": 0.00014721537427162706, + "loss": 0.5418, + "step": 30564 + }, + { + "epoch": 1.7077129368382824, + "grad_norm": 0.519545316696167, + "learning_rate": 0.0001471873599282833, + "loss": 0.4444, + "step": 30565 + }, + { + "epoch": 1.7077688074419646, + "grad_norm": 0.3673523962497711, + "learning_rate": 0.0001471593455849395, + "loss": 0.4051, + "step": 30566 + }, + { + "epoch": 1.7078246780456463, + "grad_norm": 0.4627644419670105, + "learning_rate": 0.0001471313312415957, + "loss": 0.3376, + "step": 30567 + }, + { + "epoch": 1.707880548649328, + "grad_norm": 0.520488440990448, + "learning_rate": 0.0001471033168982519, + "loss": 0.484, + "step": 30568 + }, + { + "epoch": 1.70793641925301, + "grad_norm": 0.43407854437828064, + "learning_rate": 0.0001470753025549081, + "loss": 0.3071, + "step": 30569 + }, + { + "epoch": 1.707992289856692, + "grad_norm": 0.44450098276138306, + "learning_rate": 0.00014704728821156432, + "loss": 0.4363, + "step": 30570 + }, + { + "epoch": 1.7080481604603737, + "grad_norm": 0.4119108319282532, + "learning_rate": 0.00014701927386822052, + "loss": 0.4424, + "step": 30571 + }, + { + "epoch": 1.7081040310640556, + "grad_norm": 0.35606449842453003, + "learning_rate": 0.00014699125952487673, + "loss": 0.4148, + "step": 30572 + }, + { + "epoch": 1.7081599016677376, + "grad_norm": 0.4433010220527649, + "learning_rate": 0.00014696324518153296, + "loss": 0.3804, + "step": 30573 + }, + { + "epoch": 1.7082157722714193, + "grad_norm": 0.2945558428764343, + "learning_rate": 0.00014693523083818917, + "loss": 0.3202, + "step": 30574 + }, + { + "epoch": 1.7082716428751012, + "grad_norm": 0.39888861775398254, + "learning_rate": 0.00014690721649484537, + "loss": 0.4033, + "step": 30575 + }, + { + "epoch": 1.7083275134787832, + "grad_norm": 0.46856188774108887, + "learning_rate": 0.00014687920215150158, + "loss": 0.4044, + "step": 30576 + }, + { + "epoch": 1.708383384082465, + "grad_norm": 0.9765827655792236, + "learning_rate": 0.00014685118780815779, + "loss": 0.4009, + "step": 30577 + }, + { + "epoch": 1.7084392546861469, + "grad_norm": 0.42767906188964844, + "learning_rate": 0.000146823173464814, + "loss": 0.3285, + "step": 30578 + }, + { + "epoch": 1.7084951252898288, + "grad_norm": 1.5245567560195923, + "learning_rate": 0.0001467951591214702, + "loss": 0.3286, + "step": 30579 + }, + { + "epoch": 1.7085509958935106, + "grad_norm": 0.801918625831604, + "learning_rate": 0.0001467671447781264, + "loss": 0.3976, + "step": 30580 + }, + { + "epoch": 1.7086068664971925, + "grad_norm": 0.48995915055274963, + "learning_rate": 0.00014673913043478264, + "loss": 0.4504, + "step": 30581 + }, + { + "epoch": 1.7086627371008745, + "grad_norm": 0.4215024411678314, + "learning_rate": 0.00014671111609143882, + "loss": 0.4418, + "step": 30582 + }, + { + "epoch": 1.7087186077045562, + "grad_norm": 0.7760986685752869, + "learning_rate": 0.00014668310174809502, + "loss": 0.4543, + "step": 30583 + }, + { + "epoch": 1.7087744783082381, + "grad_norm": 0.6719567179679871, + "learning_rate": 0.00014665508740475123, + "loss": 0.4117, + "step": 30584 + }, + { + "epoch": 1.70883034891192, + "grad_norm": 0.4239557981491089, + "learning_rate": 0.00014662707306140743, + "loss": 0.3142, + "step": 30585 + }, + { + "epoch": 1.7088862195156018, + "grad_norm": 0.3861655294895172, + "learning_rate": 0.00014659905871806364, + "loss": 0.4069, + "step": 30586 + }, + { + "epoch": 1.7089420901192838, + "grad_norm": 0.40243324637413025, + "learning_rate": 0.00014657104437471984, + "loss": 0.385, + "step": 30587 + }, + { + "epoch": 1.7089979607229657, + "grad_norm": 0.5162674188613892, + "learning_rate": 0.00014654303003137605, + "loss": 0.381, + "step": 30588 + }, + { + "epoch": 1.7090538313266475, + "grad_norm": 0.40552425384521484, + "learning_rate": 0.00014651501568803228, + "loss": 0.3401, + "step": 30589 + }, + { + "epoch": 1.7091097019303294, + "grad_norm": 4.5283708572387695, + "learning_rate": 0.0001464870013446885, + "loss": 0.4458, + "step": 30590 + }, + { + "epoch": 1.7091655725340114, + "grad_norm": 0.3572257161140442, + "learning_rate": 0.0001464589870013447, + "loss": 0.412, + "step": 30591 + }, + { + "epoch": 1.709221443137693, + "grad_norm": 2.0288946628570557, + "learning_rate": 0.0001464309726580009, + "loss": 0.465, + "step": 30592 + }, + { + "epoch": 1.7092773137413748, + "grad_norm": 0.5283505916595459, + "learning_rate": 0.0001464029583146571, + "loss": 0.5054, + "step": 30593 + }, + { + "epoch": 1.709333184345057, + "grad_norm": 0.3967544734477997, + "learning_rate": 0.0001463749439713133, + "loss": 0.4086, + "step": 30594 + }, + { + "epoch": 1.7093890549487387, + "grad_norm": 0.7167336344718933, + "learning_rate": 0.00014634692962796952, + "loss": 0.384, + "step": 30595 + }, + { + "epoch": 1.7094449255524204, + "grad_norm": 0.507317066192627, + "learning_rate": 0.00014631891528462572, + "loss": 0.3898, + "step": 30596 + }, + { + "epoch": 1.7095007961561026, + "grad_norm": 0.6309771537780762, + "learning_rate": 0.00014629090094128196, + "loss": 0.4004, + "step": 30597 + }, + { + "epoch": 1.7095566667597843, + "grad_norm": 0.5982019305229187, + "learning_rate": 0.00014626288659793816, + "loss": 0.4996, + "step": 30598 + }, + { + "epoch": 1.709612537363466, + "grad_norm": 0.3908885717391968, + "learning_rate": 0.00014623487225459437, + "loss": 0.3171, + "step": 30599 + }, + { + "epoch": 1.7096684079671483, + "grad_norm": 0.6392151117324829, + "learning_rate": 0.00014620685791125057, + "loss": 0.4997, + "step": 30600 + }, + { + "epoch": 1.70972427857083, + "grad_norm": 0.4874250292778015, + "learning_rate": 0.00014617884356790678, + "loss": 0.3692, + "step": 30601 + }, + { + "epoch": 1.7097801491745117, + "grad_norm": 0.5144428014755249, + "learning_rate": 0.00014615082922456299, + "loss": 0.5488, + "step": 30602 + }, + { + "epoch": 1.7098360197781937, + "grad_norm": 0.4256872236728668, + "learning_rate": 0.00014612281488121916, + "loss": 0.4458, + "step": 30603 + }, + { + "epoch": 1.7098918903818756, + "grad_norm": 0.4359513819217682, + "learning_rate": 0.00014609480053787537, + "loss": 0.4648, + "step": 30604 + }, + { + "epoch": 1.7099477609855573, + "grad_norm": 0.8781453371047974, + "learning_rate": 0.0001460667861945316, + "loss": 0.3868, + "step": 30605 + }, + { + "epoch": 1.7100036315892393, + "grad_norm": 0.40374594926834106, + "learning_rate": 0.0001460387718511878, + "loss": 0.3859, + "step": 30606 + }, + { + "epoch": 1.7100595021929212, + "grad_norm": 0.6223090887069702, + "learning_rate": 0.00014601075750784401, + "loss": 0.3954, + "step": 30607 + }, + { + "epoch": 1.710115372796603, + "grad_norm": 0.34077078104019165, + "learning_rate": 0.00014598274316450022, + "loss": 0.4096, + "step": 30608 + }, + { + "epoch": 1.710171243400285, + "grad_norm": 0.661268413066864, + "learning_rate": 0.00014595472882115643, + "loss": 0.3879, + "step": 30609 + }, + { + "epoch": 1.7102271140039669, + "grad_norm": 0.5809363722801208, + "learning_rate": 0.00014592671447781263, + "loss": 0.3794, + "step": 30610 + }, + { + "epoch": 1.7102829846076486, + "grad_norm": 0.4693959951400757, + "learning_rate": 0.00014589870013446884, + "loss": 0.4848, + "step": 30611 + }, + { + "epoch": 1.7103388552113306, + "grad_norm": 1.158374547958374, + "learning_rate": 0.00014587068579112504, + "loss": 0.3676, + "step": 30612 + }, + { + "epoch": 1.7103947258150125, + "grad_norm": 0.344074010848999, + "learning_rate": 0.00014584267144778128, + "loss": 0.3413, + "step": 30613 + }, + { + "epoch": 1.7104505964186942, + "grad_norm": 0.5351225137710571, + "learning_rate": 0.00014581465710443748, + "loss": 0.4591, + "step": 30614 + }, + { + "epoch": 1.7105064670223762, + "grad_norm": 0.4679498076438904, + "learning_rate": 0.0001457866427610937, + "loss": 0.5192, + "step": 30615 + }, + { + "epoch": 1.7105623376260581, + "grad_norm": 6.35480260848999, + "learning_rate": 0.0001457586284177499, + "loss": 0.3516, + "step": 30616 + }, + { + "epoch": 1.7106182082297399, + "grad_norm": 0.32662802934646606, + "learning_rate": 0.0001457306140744061, + "loss": 0.3662, + "step": 30617 + }, + { + "epoch": 1.7106740788334218, + "grad_norm": 0.4751461446285248, + "learning_rate": 0.0001457025997310623, + "loss": 0.4176, + "step": 30618 + }, + { + "epoch": 1.7107299494371038, + "grad_norm": 0.3665195107460022, + "learning_rate": 0.0001456745853877185, + "loss": 0.389, + "step": 30619 + }, + { + "epoch": 1.7107858200407855, + "grad_norm": 0.869369626045227, + "learning_rate": 0.00014564657104437474, + "loss": 0.6101, + "step": 30620 + }, + { + "epoch": 1.7108416906444674, + "grad_norm": 0.43274015188217163, + "learning_rate": 0.00014561855670103095, + "loss": 0.4428, + "step": 30621 + }, + { + "epoch": 1.7108975612481494, + "grad_norm": 0.4127933084964752, + "learning_rate": 0.00014559054235768716, + "loss": 0.333, + "step": 30622 + }, + { + "epoch": 1.7109534318518311, + "grad_norm": 0.37868499755859375, + "learning_rate": 0.00014556252801434336, + "loss": 0.3455, + "step": 30623 + }, + { + "epoch": 1.711009302455513, + "grad_norm": 0.5200012922286987, + "learning_rate": 0.00014553451367099954, + "loss": 0.3735, + "step": 30624 + }, + { + "epoch": 1.711065173059195, + "grad_norm": 0.3790600299835205, + "learning_rate": 0.00014550649932765575, + "loss": 0.3818, + "step": 30625 + }, + { + "epoch": 1.7111210436628768, + "grad_norm": 0.4481135606765747, + "learning_rate": 0.00014547848498431195, + "loss": 0.3231, + "step": 30626 + }, + { + "epoch": 1.7111769142665585, + "grad_norm": 0.2895562946796417, + "learning_rate": 0.00014545047064096816, + "loss": 0.3498, + "step": 30627 + }, + { + "epoch": 1.7112327848702407, + "grad_norm": 0.39768603444099426, + "learning_rate": 0.0001454224562976244, + "loss": 0.3936, + "step": 30628 + }, + { + "epoch": 1.7112886554739224, + "grad_norm": 0.5682694911956787, + "learning_rate": 0.0001453944419542806, + "loss": 0.4441, + "step": 30629 + }, + { + "epoch": 1.7113445260776041, + "grad_norm": 0.5307745337486267, + "learning_rate": 0.0001453664276109368, + "loss": 0.6057, + "step": 30630 + }, + { + "epoch": 1.7114003966812863, + "grad_norm": 0.37152111530303955, + "learning_rate": 0.000145338413267593, + "loss": 0.4091, + "step": 30631 + }, + { + "epoch": 1.711456267284968, + "grad_norm": 0.6040124893188477, + "learning_rate": 0.00014531039892424921, + "loss": 0.3679, + "step": 30632 + }, + { + "epoch": 1.7115121378886498, + "grad_norm": 0.33733105659484863, + "learning_rate": 0.00014528238458090542, + "loss": 0.3814, + "step": 30633 + }, + { + "epoch": 1.7115680084923317, + "grad_norm": 0.5076028108596802, + "learning_rate": 0.00014525437023756163, + "loss": 0.5183, + "step": 30634 + }, + { + "epoch": 1.7116238790960137, + "grad_norm": 0.48804327845573425, + "learning_rate": 0.00014522635589421783, + "loss": 0.3461, + "step": 30635 + }, + { + "epoch": 1.7116797496996954, + "grad_norm": 0.4476529657840729, + "learning_rate": 0.00014519834155087406, + "loss": 0.4613, + "step": 30636 + }, + { + "epoch": 1.7117356203033773, + "grad_norm": 0.4314819574356079, + "learning_rate": 0.00014517032720753027, + "loss": 0.5522, + "step": 30637 + }, + { + "epoch": 1.7117914909070593, + "grad_norm": 0.5225551724433899, + "learning_rate": 0.00014514231286418648, + "loss": 0.4056, + "step": 30638 + }, + { + "epoch": 1.711847361510741, + "grad_norm": 0.9684009552001953, + "learning_rate": 0.00014511429852084268, + "loss": 0.4232, + "step": 30639 + }, + { + "epoch": 1.711903232114423, + "grad_norm": 0.49850761890411377, + "learning_rate": 0.0001450862841774989, + "loss": 0.3558, + "step": 30640 + }, + { + "epoch": 1.711959102718105, + "grad_norm": 0.8073302507400513, + "learning_rate": 0.0001450582698341551, + "loss": 0.4492, + "step": 30641 + }, + { + "epoch": 1.7120149733217866, + "grad_norm": 0.3547211289405823, + "learning_rate": 0.0001450302554908113, + "loss": 0.3884, + "step": 30642 + }, + { + "epoch": 1.7120708439254686, + "grad_norm": 0.4157077670097351, + "learning_rate": 0.0001450022411474675, + "loss": 0.5392, + "step": 30643 + }, + { + "epoch": 1.7121267145291506, + "grad_norm": 0.6524147391319275, + "learning_rate": 0.00014497422680412374, + "loss": 0.4227, + "step": 30644 + }, + { + "epoch": 1.7121825851328323, + "grad_norm": 0.8691736459732056, + "learning_rate": 0.00014494621246077994, + "loss": 0.511, + "step": 30645 + }, + { + "epoch": 1.7122384557365142, + "grad_norm": 0.3357708752155304, + "learning_rate": 0.00014491819811743612, + "loss": 0.4478, + "step": 30646 + }, + { + "epoch": 1.7122943263401962, + "grad_norm": 0.45874178409576416, + "learning_rate": 0.00014489018377409233, + "loss": 0.4457, + "step": 30647 + }, + { + "epoch": 1.712350196943878, + "grad_norm": 0.37108439207077026, + "learning_rate": 0.00014486216943074853, + "loss": 0.3814, + "step": 30648 + }, + { + "epoch": 1.7124060675475599, + "grad_norm": 0.6671695113182068, + "learning_rate": 0.00014483415508740474, + "loss": 0.3818, + "step": 30649 + }, + { + "epoch": 1.7124619381512418, + "grad_norm": 0.402911901473999, + "learning_rate": 0.00014480614074406095, + "loss": 0.3928, + "step": 30650 + }, + { + "epoch": 1.7125178087549235, + "grad_norm": 0.5784667134284973, + "learning_rate": 0.00014477812640071715, + "loss": 0.4754, + "step": 30651 + }, + { + "epoch": 1.7125736793586055, + "grad_norm": 0.30292391777038574, + "learning_rate": 0.00014475011205737338, + "loss": 0.4066, + "step": 30652 + }, + { + "epoch": 1.7126295499622874, + "grad_norm": 0.4820835590362549, + "learning_rate": 0.0001447220977140296, + "loss": 0.3922, + "step": 30653 + }, + { + "epoch": 1.7126854205659692, + "grad_norm": 0.4045208990573883, + "learning_rate": 0.0001446940833706858, + "loss": 0.45, + "step": 30654 + }, + { + "epoch": 1.7127412911696511, + "grad_norm": 0.38747739791870117, + "learning_rate": 0.000144666069027342, + "loss": 0.3552, + "step": 30655 + }, + { + "epoch": 1.712797161773333, + "grad_norm": 0.8550559878349304, + "learning_rate": 0.0001446380546839982, + "loss": 0.3656, + "step": 30656 + }, + { + "epoch": 1.7128530323770148, + "grad_norm": 0.4291459918022156, + "learning_rate": 0.00014461004034065441, + "loss": 0.4384, + "step": 30657 + }, + { + "epoch": 1.7129089029806968, + "grad_norm": 0.6547760963439941, + "learning_rate": 0.00014458202599731062, + "loss": 0.4971, + "step": 30658 + }, + { + "epoch": 1.7129647735843787, + "grad_norm": 0.8351494669914246, + "learning_rate": 0.00014455401165396683, + "loss": 0.3645, + "step": 30659 + }, + { + "epoch": 1.7130206441880604, + "grad_norm": 0.420560359954834, + "learning_rate": 0.00014452599731062306, + "loss": 0.3994, + "step": 30660 + }, + { + "epoch": 1.7130765147917422, + "grad_norm": 0.676470935344696, + "learning_rate": 0.00014449798296727926, + "loss": 0.3275, + "step": 30661 + }, + { + "epoch": 1.7131323853954243, + "grad_norm": 0.4459214508533478, + "learning_rate": 0.00014446996862393547, + "loss": 0.3824, + "step": 30662 + }, + { + "epoch": 1.713188255999106, + "grad_norm": 1.1754850149154663, + "learning_rate": 0.00014444195428059168, + "loss": 0.3339, + "step": 30663 + }, + { + "epoch": 1.7132441266027878, + "grad_norm": 0.466948926448822, + "learning_rate": 0.00014441393993724788, + "loss": 0.3308, + "step": 30664 + }, + { + "epoch": 1.71329999720647, + "grad_norm": 0.40723398327827454, + "learning_rate": 0.0001443859255939041, + "loss": 0.3993, + "step": 30665 + }, + { + "epoch": 1.7133558678101517, + "grad_norm": 0.34504321217536926, + "learning_rate": 0.0001443579112505603, + "loss": 0.384, + "step": 30666 + }, + { + "epoch": 1.7134117384138334, + "grad_norm": 0.34408634901046753, + "learning_rate": 0.00014432989690721647, + "loss": 0.4279, + "step": 30667 + }, + { + "epoch": 1.7134676090175154, + "grad_norm": 0.7432114481925964, + "learning_rate": 0.0001443018825638727, + "loss": 0.4747, + "step": 30668 + }, + { + "epoch": 1.7135234796211973, + "grad_norm": 0.44978711009025574, + "learning_rate": 0.0001442738682205289, + "loss": 0.3826, + "step": 30669 + }, + { + "epoch": 1.713579350224879, + "grad_norm": 0.6121969223022461, + "learning_rate": 0.00014424585387718512, + "loss": 0.3571, + "step": 30670 + }, + { + "epoch": 1.713635220828561, + "grad_norm": 0.45951077342033386, + "learning_rate": 0.00014421783953384132, + "loss": 0.497, + "step": 30671 + }, + { + "epoch": 1.713691091432243, + "grad_norm": 0.5166139006614685, + "learning_rate": 0.00014418982519049753, + "loss": 0.4846, + "step": 30672 + }, + { + "epoch": 1.7137469620359247, + "grad_norm": 0.48379600048065186, + "learning_rate": 0.00014416181084715373, + "loss": 0.4294, + "step": 30673 + }, + { + "epoch": 1.7138028326396066, + "grad_norm": 0.49933382868766785, + "learning_rate": 0.00014413379650380994, + "loss": 0.6286, + "step": 30674 + }, + { + "epoch": 1.7138587032432886, + "grad_norm": 0.3561755418777466, + "learning_rate": 0.00014410578216046615, + "loss": 0.3546, + "step": 30675 + }, + { + "epoch": 1.7139145738469703, + "grad_norm": 0.4431092441082001, + "learning_rate": 0.00014407776781712238, + "loss": 0.4357, + "step": 30676 + }, + { + "epoch": 1.7139704444506523, + "grad_norm": 0.4996799826622009, + "learning_rate": 0.00014404975347377858, + "loss": 0.4685, + "step": 30677 + }, + { + "epoch": 1.7140263150543342, + "grad_norm": 0.5957464575767517, + "learning_rate": 0.0001440217391304348, + "loss": 0.3524, + "step": 30678 + }, + { + "epoch": 1.714082185658016, + "grad_norm": 5.897800922393799, + "learning_rate": 0.000143993724787091, + "loss": 0.4096, + "step": 30679 + }, + { + "epoch": 1.714138056261698, + "grad_norm": 1.9972535371780396, + "learning_rate": 0.0001439657104437472, + "loss": 0.3634, + "step": 30680 + }, + { + "epoch": 1.7141939268653799, + "grad_norm": 0.6284046769142151, + "learning_rate": 0.0001439376961004034, + "loss": 0.5314, + "step": 30681 + }, + { + "epoch": 1.7142497974690616, + "grad_norm": 0.41958650946617126, + "learning_rate": 0.00014390968175705961, + "loss": 0.4843, + "step": 30682 + }, + { + "epoch": 1.7143056680727435, + "grad_norm": 0.5784603953361511, + "learning_rate": 0.00014388166741371585, + "loss": 0.3704, + "step": 30683 + }, + { + "epoch": 1.7143615386764255, + "grad_norm": 0.3076043128967285, + "learning_rate": 0.00014385365307037205, + "loss": 0.3341, + "step": 30684 + }, + { + "epoch": 1.7144174092801072, + "grad_norm": 1.6652888059616089, + "learning_rate": 0.00014382563872702826, + "loss": 0.3337, + "step": 30685 + }, + { + "epoch": 1.7144732798837892, + "grad_norm": 0.44697266817092896, + "learning_rate": 0.00014379762438368446, + "loss": 0.3917, + "step": 30686 + }, + { + "epoch": 1.7145291504874711, + "grad_norm": 0.45807453989982605, + "learning_rate": 0.00014376961004034067, + "loss": 0.398, + "step": 30687 + }, + { + "epoch": 1.7145850210911528, + "grad_norm": 0.40788188576698303, + "learning_rate": 0.00014374159569699685, + "loss": 0.365, + "step": 30688 + }, + { + "epoch": 1.7146408916948348, + "grad_norm": 0.3256407380104065, + "learning_rate": 0.00014371358135365305, + "loss": 0.3362, + "step": 30689 + }, + { + "epoch": 1.7146967622985168, + "grad_norm": 0.6050518155097961, + "learning_rate": 0.00014368556701030926, + "loss": 0.4869, + "step": 30690 + }, + { + "epoch": 1.7147526329021985, + "grad_norm": 0.3107644021511078, + "learning_rate": 0.0001436575526669655, + "loss": 0.3795, + "step": 30691 + }, + { + "epoch": 1.7148085035058804, + "grad_norm": 0.44845959544181824, + "learning_rate": 0.0001436295383236217, + "loss": 0.311, + "step": 30692 + }, + { + "epoch": 1.7148643741095624, + "grad_norm": 0.43196094036102295, + "learning_rate": 0.0001436015239802779, + "loss": 0.4669, + "step": 30693 + }, + { + "epoch": 1.7149202447132441, + "grad_norm": 0.3726122975349426, + "learning_rate": 0.0001435735096369341, + "loss": 0.3635, + "step": 30694 + }, + { + "epoch": 1.7149761153169258, + "grad_norm": 0.43379831314086914, + "learning_rate": 0.00014354549529359032, + "loss": 0.4687, + "step": 30695 + }, + { + "epoch": 1.715031985920608, + "grad_norm": 0.4158792793750763, + "learning_rate": 0.00014351748095024652, + "loss": 0.518, + "step": 30696 + }, + { + "epoch": 1.7150878565242897, + "grad_norm": 0.32245123386383057, + "learning_rate": 0.00014348946660690273, + "loss": 0.3789, + "step": 30697 + }, + { + "epoch": 1.7151437271279715, + "grad_norm": 0.5850815176963806, + "learning_rate": 0.00014346145226355893, + "loss": 0.462, + "step": 30698 + }, + { + "epoch": 1.7151995977316536, + "grad_norm": 0.328779399394989, + "learning_rate": 0.00014343343792021517, + "loss": 0.3484, + "step": 30699 + }, + { + "epoch": 1.7152554683353354, + "grad_norm": 0.3373471200466156, + "learning_rate": 0.00014340542357687137, + "loss": 0.319, + "step": 30700 + }, + { + "epoch": 1.715311338939017, + "grad_norm": 0.48123735189437866, + "learning_rate": 0.00014337740923352758, + "loss": 0.4445, + "step": 30701 + }, + { + "epoch": 1.715367209542699, + "grad_norm": 2.3665945529937744, + "learning_rate": 0.00014334939489018378, + "loss": 0.4872, + "step": 30702 + }, + { + "epoch": 1.715423080146381, + "grad_norm": 0.5160138010978699, + "learning_rate": 0.00014332138054684, + "loss": 0.3747, + "step": 30703 + }, + { + "epoch": 1.7154789507500627, + "grad_norm": 0.38431018590927124, + "learning_rate": 0.0001432933662034962, + "loss": 0.4995, + "step": 30704 + }, + { + "epoch": 1.7155348213537447, + "grad_norm": 0.6203874945640564, + "learning_rate": 0.0001432653518601524, + "loss": 0.4617, + "step": 30705 + }, + { + "epoch": 1.7155906919574266, + "grad_norm": 0.933181881904602, + "learning_rate": 0.0001432373375168086, + "loss": 0.7218, + "step": 30706 + }, + { + "epoch": 1.7156465625611084, + "grad_norm": 0.5386576056480408, + "learning_rate": 0.00014320932317346484, + "loss": 0.4024, + "step": 30707 + }, + { + "epoch": 1.7157024331647903, + "grad_norm": 0.4249606132507324, + "learning_rate": 0.00014318130883012105, + "loss": 0.4134, + "step": 30708 + }, + { + "epoch": 1.7157583037684723, + "grad_norm": 0.36014866828918457, + "learning_rate": 0.00014315329448677722, + "loss": 0.3454, + "step": 30709 + }, + { + "epoch": 1.715814174372154, + "grad_norm": 0.31706538796424866, + "learning_rate": 0.00014312528014343343, + "loss": 0.3097, + "step": 30710 + }, + { + "epoch": 1.715870044975836, + "grad_norm": 0.7902326583862305, + "learning_rate": 0.00014309726580008964, + "loss": 0.4109, + "step": 30711 + }, + { + "epoch": 1.715925915579518, + "grad_norm": 0.3125472366809845, + "learning_rate": 0.00014306925145674584, + "loss": 0.3856, + "step": 30712 + }, + { + "epoch": 1.7159817861831996, + "grad_norm": 0.6254535913467407, + "learning_rate": 0.00014304123711340205, + "loss": 0.3864, + "step": 30713 + }, + { + "epoch": 1.7160376567868816, + "grad_norm": 0.7114830613136292, + "learning_rate": 0.00014301322277005825, + "loss": 0.4909, + "step": 30714 + }, + { + "epoch": 1.7160935273905635, + "grad_norm": 0.4611778259277344, + "learning_rate": 0.0001429852084267145, + "loss": 0.4566, + "step": 30715 + }, + { + "epoch": 1.7161493979942453, + "grad_norm": 0.7011179327964783, + "learning_rate": 0.0001429571940833707, + "loss": 0.4333, + "step": 30716 + }, + { + "epoch": 1.7162052685979272, + "grad_norm": 0.4943983554840088, + "learning_rate": 0.0001429291797400269, + "loss": 0.5741, + "step": 30717 + }, + { + "epoch": 1.7162611392016092, + "grad_norm": 0.5365913510322571, + "learning_rate": 0.0001429011653966831, + "loss": 0.5116, + "step": 30718 + }, + { + "epoch": 1.716317009805291, + "grad_norm": 0.5292895436286926, + "learning_rate": 0.0001428731510533393, + "loss": 0.3793, + "step": 30719 + }, + { + "epoch": 1.7163728804089728, + "grad_norm": 0.5583752989768982, + "learning_rate": 0.00014284513670999552, + "loss": 0.5389, + "step": 30720 + }, + { + "epoch": 1.7164287510126548, + "grad_norm": 0.4588935971260071, + "learning_rate": 0.00014281712236665172, + "loss": 0.4152, + "step": 30721 + }, + { + "epoch": 1.7164846216163365, + "grad_norm": 0.7028584480285645, + "learning_rate": 0.00014278910802330793, + "loss": 0.4487, + "step": 30722 + }, + { + "epoch": 1.7165404922200185, + "grad_norm": 2.0682826042175293, + "learning_rate": 0.00014276109367996416, + "loss": 0.3204, + "step": 30723 + }, + { + "epoch": 1.7165963628237004, + "grad_norm": 4.035478591918945, + "learning_rate": 0.00014273307933662037, + "loss": 0.6319, + "step": 30724 + }, + { + "epoch": 1.7166522334273822, + "grad_norm": 0.5271751880645752, + "learning_rate": 0.00014270506499327657, + "loss": 0.338, + "step": 30725 + }, + { + "epoch": 1.716708104031064, + "grad_norm": 2.193533182144165, + "learning_rate": 0.00014267705064993278, + "loss": 0.5902, + "step": 30726 + }, + { + "epoch": 1.716763974634746, + "grad_norm": 0.4736722409725189, + "learning_rate": 0.00014264903630658898, + "loss": 0.3851, + "step": 30727 + }, + { + "epoch": 1.7168198452384278, + "grad_norm": 0.6147834658622742, + "learning_rate": 0.0001426210219632452, + "loss": 0.3611, + "step": 30728 + }, + { + "epoch": 1.7168757158421095, + "grad_norm": 0.48225679993629456, + "learning_rate": 0.0001425930076199014, + "loss": 0.4608, + "step": 30729 + }, + { + "epoch": 1.7169315864457917, + "grad_norm": 0.353139728307724, + "learning_rate": 0.0001425649932765576, + "loss": 0.4158, + "step": 30730 + }, + { + "epoch": 1.7169874570494734, + "grad_norm": 0.4405866265296936, + "learning_rate": 0.0001425369789332138, + "loss": 0.4799, + "step": 30731 + }, + { + "epoch": 1.7170433276531551, + "grad_norm": 0.8318396806716919, + "learning_rate": 0.00014250896458987, + "loss": 0.455, + "step": 30732 + }, + { + "epoch": 1.7170991982568373, + "grad_norm": 0.378823846578598, + "learning_rate": 0.00014248095024652622, + "loss": 0.3376, + "step": 30733 + }, + { + "epoch": 1.717155068860519, + "grad_norm": 0.3413693308830261, + "learning_rate": 0.00014245293590318242, + "loss": 0.3533, + "step": 30734 + }, + { + "epoch": 1.7172109394642008, + "grad_norm": 1.2063742876052856, + "learning_rate": 0.00014242492155983863, + "loss": 0.5299, + "step": 30735 + }, + { + "epoch": 1.7172668100678827, + "grad_norm": 0.3664615750312805, + "learning_rate": 0.00014239690721649484, + "loss": 0.3176, + "step": 30736 + }, + { + "epoch": 1.7173226806715647, + "grad_norm": 0.7711934447288513, + "learning_rate": 0.00014236889287315104, + "loss": 0.3821, + "step": 30737 + }, + { + "epoch": 1.7173785512752464, + "grad_norm": 0.5427992939949036, + "learning_rate": 0.00014234087852980725, + "loss": 0.3568, + "step": 30738 + }, + { + "epoch": 1.7174344218789284, + "grad_norm": 0.45950162410736084, + "learning_rate": 0.00014231286418646348, + "loss": 0.5019, + "step": 30739 + }, + { + "epoch": 1.7174902924826103, + "grad_norm": Infinity, + "learning_rate": 0.00014231286418646348, + "loss": 0.3465, + "step": 30740 + }, + { + "epoch": 1.717546163086292, + "grad_norm": 0.43167179822921753, + "learning_rate": 0.00014228484984311969, + "loss": 0.4629, + "step": 30741 + }, + { + "epoch": 1.717602033689974, + "grad_norm": 0.5284801721572876, + "learning_rate": 0.0001422568354997759, + "loss": 0.5777, + "step": 30742 + }, + { + "epoch": 1.717657904293656, + "grad_norm": 0.4491134583950043, + "learning_rate": 0.0001422288211564321, + "loss": 0.4899, + "step": 30743 + }, + { + "epoch": 1.7177137748973377, + "grad_norm": 0.4831450283527374, + "learning_rate": 0.0001422008068130883, + "loss": 0.4163, + "step": 30744 + }, + { + "epoch": 1.7177696455010196, + "grad_norm": 0.4084014296531677, + "learning_rate": 0.0001421727924697445, + "loss": 0.4065, + "step": 30745 + }, + { + "epoch": 1.7178255161047016, + "grad_norm": 0.47961270809173584, + "learning_rate": 0.00014214477812640072, + "loss": 0.5287, + "step": 30746 + }, + { + "epoch": 1.7178813867083833, + "grad_norm": 0.30024224519729614, + "learning_rate": 0.00014211676378305695, + "loss": 0.3659, + "step": 30747 + }, + { + "epoch": 1.7179372573120653, + "grad_norm": 0.4610266089439392, + "learning_rate": 0.00014208874943971315, + "loss": 0.4789, + "step": 30748 + }, + { + "epoch": 1.7179931279157472, + "grad_norm": 0.5164749026298523, + "learning_rate": 0.00014206073509636936, + "loss": 0.5728, + "step": 30749 + }, + { + "epoch": 1.718048998519429, + "grad_norm": 0.69679856300354, + "learning_rate": 0.00014203272075302557, + "loss": 0.459, + "step": 30750 + }, + { + "epoch": 1.718104869123111, + "grad_norm": 6.0981645584106445, + "learning_rate": 0.00014200470640968177, + "loss": 0.4727, + "step": 30751 + }, + { + "epoch": 1.7181607397267928, + "grad_norm": 3.030626058578491, + "learning_rate": 0.00014197669206633798, + "loss": 0.6431, + "step": 30752 + }, + { + "epoch": 1.7182166103304746, + "grad_norm": 0.31454887986183167, + "learning_rate": 0.00014194867772299416, + "loss": 0.3011, + "step": 30753 + }, + { + "epoch": 1.7182724809341565, + "grad_norm": 0.44892627000808716, + "learning_rate": 0.00014192066337965036, + "loss": 0.3736, + "step": 30754 + }, + { + "epoch": 1.7183283515378385, + "grad_norm": 0.4625583589076996, + "learning_rate": 0.0001418926490363066, + "loss": 0.413, + "step": 30755 + }, + { + "epoch": 1.7183842221415202, + "grad_norm": 0.3432040810585022, + "learning_rate": 0.0001418646346929628, + "loss": 0.3433, + "step": 30756 + }, + { + "epoch": 1.7184400927452022, + "grad_norm": 0.415463924407959, + "learning_rate": 0.000141836620349619, + "loss": 0.4071, + "step": 30757 + }, + { + "epoch": 1.718495963348884, + "grad_norm": 3.3497939109802246, + "learning_rate": 0.0001418086060062752, + "loss": 0.5618, + "step": 30758 + }, + { + "epoch": 1.7185518339525658, + "grad_norm": 0.709139883518219, + "learning_rate": 0.00014178059166293142, + "loss": 0.4682, + "step": 30759 + }, + { + "epoch": 1.7186077045562478, + "grad_norm": 0.5509200096130371, + "learning_rate": 0.00014175257731958762, + "loss": 0.4266, + "step": 30760 + }, + { + "epoch": 1.7186635751599297, + "grad_norm": 0.30759021639823914, + "learning_rate": 0.00014172456297624383, + "loss": 0.3303, + "step": 30761 + }, + { + "epoch": 1.7187194457636115, + "grad_norm": 0.5793008804321289, + "learning_rate": 0.00014169654863290004, + "loss": 0.5838, + "step": 30762 + }, + { + "epoch": 1.7187753163672932, + "grad_norm": 0.39193233847618103, + "learning_rate": 0.00014166853428955627, + "loss": 0.3784, + "step": 30763 + }, + { + "epoch": 1.7188311869709754, + "grad_norm": 4.884180068969727, + "learning_rate": 0.00014164051994621247, + "loss": 0.3991, + "step": 30764 + }, + { + "epoch": 1.718887057574657, + "grad_norm": 1.218113660812378, + "learning_rate": 0.00014161250560286868, + "loss": 0.4076, + "step": 30765 + }, + { + "epoch": 1.7189429281783388, + "grad_norm": 0.3732626140117645, + "learning_rate": 0.00014158449125952489, + "loss": 0.3976, + "step": 30766 + }, + { + "epoch": 1.718998798782021, + "grad_norm": 0.45404407382011414, + "learning_rate": 0.0001415564769161811, + "loss": 0.3144, + "step": 30767 + }, + { + "epoch": 1.7190546693857027, + "grad_norm": 0.7103826999664307, + "learning_rate": 0.0001415284625728373, + "loss": 0.4656, + "step": 30768 + }, + { + "epoch": 1.7191105399893845, + "grad_norm": 0.42585858702659607, + "learning_rate": 0.0001415004482294935, + "loss": 0.4157, + "step": 30769 + }, + { + "epoch": 1.7191664105930664, + "grad_norm": 0.36051109433174133, + "learning_rate": 0.0001414724338861497, + "loss": 0.3804, + "step": 30770 + }, + { + "epoch": 1.7192222811967484, + "grad_norm": 0.33666688203811646, + "learning_rate": 0.00014144441954280594, + "loss": 0.4229, + "step": 30771 + }, + { + "epoch": 1.71927815180043, + "grad_norm": 0.3141992688179016, + "learning_rate": 0.00014141640519946215, + "loss": 0.4116, + "step": 30772 + }, + { + "epoch": 1.719334022404112, + "grad_norm": 0.3925444781780243, + "learning_rate": 0.00014138839085611835, + "loss": 0.4048, + "step": 30773 + }, + { + "epoch": 1.719389893007794, + "grad_norm": 0.4360697269439697, + "learning_rate": 0.00014136037651277453, + "loss": 0.4401, + "step": 30774 + }, + { + "epoch": 1.7194457636114757, + "grad_norm": 4.688111305236816, + "learning_rate": 0.00014133236216943074, + "loss": 0.4414, + "step": 30775 + }, + { + "epoch": 1.7195016342151577, + "grad_norm": 0.3539465367794037, + "learning_rate": 0.00014130434782608694, + "loss": 0.3584, + "step": 30776 + }, + { + "epoch": 1.7195575048188396, + "grad_norm": 0.3794735372066498, + "learning_rate": 0.00014127633348274315, + "loss": 0.4469, + "step": 30777 + }, + { + "epoch": 1.7196133754225214, + "grad_norm": 0.5291370153427124, + "learning_rate": 0.00014124831913939936, + "loss": 0.3508, + "step": 30778 + }, + { + "epoch": 1.7196692460262033, + "grad_norm": 0.38332903385162354, + "learning_rate": 0.0001412203047960556, + "loss": 0.4649, + "step": 30779 + }, + { + "epoch": 1.7197251166298853, + "grad_norm": 0.4677514433860779, + "learning_rate": 0.0001411922904527118, + "loss": 0.5562, + "step": 30780 + }, + { + "epoch": 1.719780987233567, + "grad_norm": 0.39827296137809753, + "learning_rate": 0.000141164276109368, + "loss": 0.4046, + "step": 30781 + }, + { + "epoch": 1.719836857837249, + "grad_norm": 0.40781623125076294, + "learning_rate": 0.0001411362617660242, + "loss": 0.4276, + "step": 30782 + }, + { + "epoch": 1.7198927284409309, + "grad_norm": 0.5950707793235779, + "learning_rate": 0.0001411082474226804, + "loss": 0.527, + "step": 30783 + }, + { + "epoch": 1.7199485990446126, + "grad_norm": 0.4052385091781616, + "learning_rate": 0.00014108023307933662, + "loss": 0.4977, + "step": 30784 + }, + { + "epoch": 1.7200044696482946, + "grad_norm": 0.7275767922401428, + "learning_rate": 0.00014105221873599282, + "loss": 0.4407, + "step": 30785 + }, + { + "epoch": 1.7200603402519765, + "grad_norm": 0.7537901401519775, + "learning_rate": 0.00014102420439264903, + "loss": 0.4851, + "step": 30786 + }, + { + "epoch": 1.7201162108556582, + "grad_norm": 0.38314247131347656, + "learning_rate": 0.00014099619004930526, + "loss": 0.4057, + "step": 30787 + }, + { + "epoch": 1.7201720814593402, + "grad_norm": 1.6418708562850952, + "learning_rate": 0.00014096817570596147, + "loss": 0.3531, + "step": 30788 + }, + { + "epoch": 1.7202279520630221, + "grad_norm": 0.3909919857978821, + "learning_rate": 0.00014094016136261767, + "loss": 0.3937, + "step": 30789 + }, + { + "epoch": 1.7202838226667039, + "grad_norm": 0.47911226749420166, + "learning_rate": 0.00014091214701927388, + "loss": 0.4742, + "step": 30790 + }, + { + "epoch": 1.7203396932703858, + "grad_norm": 0.6012067794799805, + "learning_rate": 0.00014088413267593009, + "loss": 0.4312, + "step": 30791 + }, + { + "epoch": 1.7203955638740678, + "grad_norm": 2.321133613586426, + "learning_rate": 0.0001408561183325863, + "loss": 0.3444, + "step": 30792 + }, + { + "epoch": 1.7204514344777495, + "grad_norm": 0.6866324543952942, + "learning_rate": 0.0001408281039892425, + "loss": 0.3631, + "step": 30793 + }, + { + "epoch": 1.7205073050814315, + "grad_norm": 0.5176250338554382, + "learning_rate": 0.0001408000896458987, + "loss": 0.4434, + "step": 30794 + }, + { + "epoch": 1.7205631756851134, + "grad_norm": 0.5597522854804993, + "learning_rate": 0.0001407720753025549, + "loss": 0.5146, + "step": 30795 + }, + { + "epoch": 1.7206190462887951, + "grad_norm": 0.4462542235851288, + "learning_rate": 0.00014074406095921111, + "loss": 0.5301, + "step": 30796 + }, + { + "epoch": 1.7206749168924769, + "grad_norm": 0.3655821979045868, + "learning_rate": 0.00014071604661586732, + "loss": 0.422, + "step": 30797 + }, + { + "epoch": 1.720730787496159, + "grad_norm": 0.3696763813495636, + "learning_rate": 0.00014068803227252353, + "loss": 0.4317, + "step": 30798 + }, + { + "epoch": 1.7207866580998408, + "grad_norm": 0.35785093903541565, + "learning_rate": 0.00014066001792917973, + "loss": 0.4952, + "step": 30799 + }, + { + "epoch": 1.7208425287035225, + "grad_norm": 1.632727026939392, + "learning_rate": 0.00014063200358583594, + "loss": 0.3678, + "step": 30800 + }, + { + "epoch": 1.7208983993072047, + "grad_norm": 1.3761862516403198, + "learning_rate": 0.00014060398924249214, + "loss": 0.3279, + "step": 30801 + }, + { + "epoch": 1.7209542699108864, + "grad_norm": 0.41796550154685974, + "learning_rate": 0.00014057597489914835, + "loss": 0.3519, + "step": 30802 + }, + { + "epoch": 1.7210101405145681, + "grad_norm": 0.37769827246665955, + "learning_rate": 0.00014054796055580458, + "loss": 0.3627, + "step": 30803 + }, + { + "epoch": 1.72106601111825, + "grad_norm": 0.36994844675064087, + "learning_rate": 0.0001405199462124608, + "loss": 0.4902, + "step": 30804 + }, + { + "epoch": 1.721121881721932, + "grad_norm": 0.6351805329322815, + "learning_rate": 0.000140491931869117, + "loss": 0.5333, + "step": 30805 + }, + { + "epoch": 1.7211777523256138, + "grad_norm": 0.6821958422660828, + "learning_rate": 0.0001404639175257732, + "loss": 0.4234, + "step": 30806 + }, + { + "epoch": 1.7212336229292957, + "grad_norm": 0.4127097725868225, + "learning_rate": 0.0001404359031824294, + "loss": 0.3339, + "step": 30807 + }, + { + "epoch": 1.7212894935329777, + "grad_norm": 0.689221203327179, + "learning_rate": 0.0001404078888390856, + "loss": 0.3584, + "step": 30808 + }, + { + "epoch": 1.7213453641366594, + "grad_norm": 0.41434380412101746, + "learning_rate": 0.00014037987449574182, + "loss": 0.4044, + "step": 30809 + }, + { + "epoch": 1.7214012347403413, + "grad_norm": 1.7010984420776367, + "learning_rate": 0.00014035186015239802, + "loss": 0.5077, + "step": 30810 + }, + { + "epoch": 1.7214571053440233, + "grad_norm": 0.6548634767532349, + "learning_rate": 0.00014032384580905426, + "loss": 0.5025, + "step": 30811 + }, + { + "epoch": 1.721512975947705, + "grad_norm": 0.46429455280303955, + "learning_rate": 0.00014029583146571046, + "loss": 0.4301, + "step": 30812 + }, + { + "epoch": 1.721568846551387, + "grad_norm": 0.4504738450050354, + "learning_rate": 0.00014026781712236667, + "loss": 0.4055, + "step": 30813 + }, + { + "epoch": 1.721624717155069, + "grad_norm": 0.3519321382045746, + "learning_rate": 0.00014023980277902287, + "loss": 0.4554, + "step": 30814 + }, + { + "epoch": 1.7216805877587507, + "grad_norm": 0.9727960228919983, + "learning_rate": 0.00014021178843567908, + "loss": 0.4782, + "step": 30815 + }, + { + "epoch": 1.7217364583624326, + "grad_norm": 0.8193076848983765, + "learning_rate": 0.00014018377409233529, + "loss": 0.5238, + "step": 30816 + }, + { + "epoch": 1.7217923289661146, + "grad_norm": 0.3894229829311371, + "learning_rate": 0.00014015575974899146, + "loss": 0.3758, + "step": 30817 + }, + { + "epoch": 1.7218481995697963, + "grad_norm": 0.9303464889526367, + "learning_rate": 0.00014012774540564767, + "loss": 0.439, + "step": 30818 + }, + { + "epoch": 1.7219040701734782, + "grad_norm": 0.42207977175712585, + "learning_rate": 0.0001400997310623039, + "loss": 0.4069, + "step": 30819 + }, + { + "epoch": 1.7219599407771602, + "grad_norm": 0.9970376491546631, + "learning_rate": 0.0001400717167189601, + "loss": 0.3783, + "step": 30820 + }, + { + "epoch": 1.722015811380842, + "grad_norm": 0.8721408247947693, + "learning_rate": 0.00014004370237561631, + "loss": 0.4375, + "step": 30821 + }, + { + "epoch": 1.7220716819845239, + "grad_norm": 0.6531175971031189, + "learning_rate": 0.00014001568803227252, + "loss": 0.4925, + "step": 30822 + }, + { + "epoch": 1.7221275525882058, + "grad_norm": 0.40029194951057434, + "learning_rate": 0.00013998767368892873, + "loss": 0.3941, + "step": 30823 + }, + { + "epoch": 1.7221834231918876, + "grad_norm": 0.6868239641189575, + "learning_rate": 0.00013995965934558493, + "loss": 0.3877, + "step": 30824 + }, + { + "epoch": 1.7222392937955695, + "grad_norm": 0.4266306161880493, + "learning_rate": 0.00013993164500224114, + "loss": 0.497, + "step": 30825 + }, + { + "epoch": 1.7222951643992515, + "grad_norm": 0.4810905456542969, + "learning_rate": 0.00013990363065889737, + "loss": 0.3661, + "step": 30826 + }, + { + "epoch": 1.7223510350029332, + "grad_norm": 0.3092972934246063, + "learning_rate": 0.00013987561631555358, + "loss": 0.3794, + "step": 30827 + }, + { + "epoch": 1.722406905606615, + "grad_norm": 2.5940773487091064, + "learning_rate": 0.00013984760197220978, + "loss": 0.4476, + "step": 30828 + }, + { + "epoch": 1.722462776210297, + "grad_norm": 0.36158525943756104, + "learning_rate": 0.000139819587628866, + "loss": 0.4388, + "step": 30829 + }, + { + "epoch": 1.7225186468139788, + "grad_norm": 0.54632568359375, + "learning_rate": 0.0001397915732855222, + "loss": 0.5447, + "step": 30830 + }, + { + "epoch": 1.7225745174176605, + "grad_norm": 0.3922041058540344, + "learning_rate": 0.0001397635589421784, + "loss": 0.3117, + "step": 30831 + }, + { + "epoch": 1.7226303880213427, + "grad_norm": 0.9918121695518494, + "learning_rate": 0.0001397355445988346, + "loss": 0.3405, + "step": 30832 + }, + { + "epoch": 1.7226862586250244, + "grad_norm": 0.6213135123252869, + "learning_rate": 0.0001397075302554908, + "loss": 0.5151, + "step": 30833 + }, + { + "epoch": 1.7227421292287062, + "grad_norm": 0.47469401359558105, + "learning_rate": 0.00013967951591214704, + "loss": 0.4956, + "step": 30834 + }, + { + "epoch": 1.7227979998323883, + "grad_norm": 0.8857958912849426, + "learning_rate": 0.00013965150156880325, + "loss": 0.4632, + "step": 30835 + }, + { + "epoch": 1.72285387043607, + "grad_norm": 0.4289613366127014, + "learning_rate": 0.00013962348722545946, + "loss": 0.417, + "step": 30836 + }, + { + "epoch": 1.7229097410397518, + "grad_norm": 0.5412551164627075, + "learning_rate": 0.00013959547288211566, + "loss": 0.4769, + "step": 30837 + }, + { + "epoch": 1.7229656116434338, + "grad_norm": 1.1388918161392212, + "learning_rate": 0.00013956745853877184, + "loss": 0.436, + "step": 30838 + }, + { + "epoch": 1.7230214822471157, + "grad_norm": 0.46407800912857056, + "learning_rate": 0.00013953944419542805, + "loss": 0.6423, + "step": 30839 + }, + { + "epoch": 1.7230773528507974, + "grad_norm": 0.9493117332458496, + "learning_rate": 0.00013951142985208425, + "loss": 0.3987, + "step": 30840 + }, + { + "epoch": 1.7231332234544794, + "grad_norm": 0.34686654806137085, + "learning_rate": 0.00013948341550874046, + "loss": 0.4096, + "step": 30841 + }, + { + "epoch": 1.7231890940581613, + "grad_norm": 0.5337194800376892, + "learning_rate": 0.0001394554011653967, + "loss": 0.3803, + "step": 30842 + }, + { + "epoch": 1.723244964661843, + "grad_norm": 0.4785051643848419, + "learning_rate": 0.0001394273868220529, + "loss": 0.4145, + "step": 30843 + }, + { + "epoch": 1.723300835265525, + "grad_norm": 0.5894016623497009, + "learning_rate": 0.0001393993724787091, + "loss": 0.3978, + "step": 30844 + }, + { + "epoch": 1.723356705869207, + "grad_norm": 0.5752236843109131, + "learning_rate": 0.0001393713581353653, + "loss": 0.466, + "step": 30845 + }, + { + "epoch": 1.7234125764728887, + "grad_norm": 1.152589201927185, + "learning_rate": 0.00013934334379202151, + "loss": 0.4154, + "step": 30846 + }, + { + "epoch": 1.7234684470765707, + "grad_norm": 0.78350430727005, + "learning_rate": 0.00013931532944867772, + "loss": 0.3748, + "step": 30847 + }, + { + "epoch": 1.7235243176802526, + "grad_norm": 0.32780173420906067, + "learning_rate": 0.00013928731510533393, + "loss": 0.3346, + "step": 30848 + }, + { + "epoch": 1.7235801882839343, + "grad_norm": 0.3582516014575958, + "learning_rate": 0.00013925930076199013, + "loss": 0.3831, + "step": 30849 + }, + { + "epoch": 1.7236360588876163, + "grad_norm": 0.5567457675933838, + "learning_rate": 0.00013923128641864636, + "loss": 0.5034, + "step": 30850 + }, + { + "epoch": 1.7236919294912982, + "grad_norm": 0.8015682697296143, + "learning_rate": 0.00013920327207530257, + "loss": 0.5972, + "step": 30851 + }, + { + "epoch": 1.72374780009498, + "grad_norm": 0.4361380338668823, + "learning_rate": 0.00013917525773195878, + "loss": 0.309, + "step": 30852 + }, + { + "epoch": 1.723803670698662, + "grad_norm": 0.488269567489624, + "learning_rate": 0.00013914724338861498, + "loss": 0.4579, + "step": 30853 + }, + { + "epoch": 1.7238595413023439, + "grad_norm": 0.4576774537563324, + "learning_rate": 0.0001391192290452712, + "loss": 0.5043, + "step": 30854 + }, + { + "epoch": 1.7239154119060256, + "grad_norm": 0.47118616104125977, + "learning_rate": 0.0001390912147019274, + "loss": 0.3763, + "step": 30855 + }, + { + "epoch": 1.7239712825097075, + "grad_norm": 0.4747130274772644, + "learning_rate": 0.0001390632003585836, + "loss": 0.3787, + "step": 30856 + }, + { + "epoch": 1.7240271531133895, + "grad_norm": 0.564357340335846, + "learning_rate": 0.0001390351860152398, + "loss": 0.423, + "step": 30857 + }, + { + "epoch": 1.7240830237170712, + "grad_norm": 0.34422171115875244, + "learning_rate": 0.00013900717167189604, + "loss": 0.3695, + "step": 30858 + }, + { + "epoch": 1.7241388943207532, + "grad_norm": 0.3865453004837036, + "learning_rate": 0.00013897915732855222, + "loss": 0.4242, + "step": 30859 + }, + { + "epoch": 1.7241947649244351, + "grad_norm": 0.485160231590271, + "learning_rate": 0.00013895114298520842, + "loss": 0.3579, + "step": 30860 + }, + { + "epoch": 1.7242506355281169, + "grad_norm": 0.4213552474975586, + "learning_rate": 0.00013892312864186463, + "loss": 0.4234, + "step": 30861 + }, + { + "epoch": 1.7243065061317986, + "grad_norm": 0.48094627261161804, + "learning_rate": 0.00013889511429852083, + "loss": 0.472, + "step": 30862 + }, + { + "epoch": 1.7243623767354808, + "grad_norm": 0.3565804362297058, + "learning_rate": 0.00013886709995517704, + "loss": 0.3259, + "step": 30863 + }, + { + "epoch": 1.7244182473391625, + "grad_norm": 0.7316330671310425, + "learning_rate": 0.00013883908561183325, + "loss": 0.3053, + "step": 30864 + }, + { + "epoch": 1.7244741179428442, + "grad_norm": 0.4632572829723358, + "learning_rate": 0.00013881107126848945, + "loss": 0.4508, + "step": 30865 + }, + { + "epoch": 1.7245299885465264, + "grad_norm": 0.34974128007888794, + "learning_rate": 0.00013878305692514568, + "loss": 0.4035, + "step": 30866 + }, + { + "epoch": 1.7245858591502081, + "grad_norm": 0.44938987493515015, + "learning_rate": 0.0001387550425818019, + "loss": 0.3446, + "step": 30867 + }, + { + "epoch": 1.7246417297538899, + "grad_norm": 0.4907263219356537, + "learning_rate": 0.0001387270282384581, + "loss": 0.3764, + "step": 30868 + }, + { + "epoch": 1.724697600357572, + "grad_norm": 0.43287193775177, + "learning_rate": 0.0001386990138951143, + "loss": 0.4639, + "step": 30869 + }, + { + "epoch": 1.7247534709612538, + "grad_norm": 6.047272205352783, + "learning_rate": 0.0001386709995517705, + "loss": 0.4152, + "step": 30870 + }, + { + "epoch": 1.7248093415649355, + "grad_norm": 0.6085650324821472, + "learning_rate": 0.0001386429852084267, + "loss": 0.4071, + "step": 30871 + }, + { + "epoch": 1.7248652121686174, + "grad_norm": 0.3445946276187897, + "learning_rate": 0.00013861497086508292, + "loss": 0.3539, + "step": 30872 + }, + { + "epoch": 1.7249210827722994, + "grad_norm": 0.4239356219768524, + "learning_rate": 0.00013858695652173913, + "loss": 0.4174, + "step": 30873 + }, + { + "epoch": 1.7249769533759811, + "grad_norm": 0.34440648555755615, + "learning_rate": 0.00013855894217839536, + "loss": 0.4619, + "step": 30874 + }, + { + "epoch": 1.725032823979663, + "grad_norm": 0.37555131316185, + "learning_rate": 0.00013853092783505156, + "loss": 0.3877, + "step": 30875 + }, + { + "epoch": 1.725088694583345, + "grad_norm": 0.38794195652008057, + "learning_rate": 0.00013850291349170777, + "loss": 0.3272, + "step": 30876 + }, + { + "epoch": 1.7251445651870267, + "grad_norm": 0.3447839319705963, + "learning_rate": 0.00013847489914836398, + "loss": 0.4443, + "step": 30877 + }, + { + "epoch": 1.7252004357907087, + "grad_norm": 0.3847557306289673, + "learning_rate": 0.00013844688480502018, + "loss": 0.4428, + "step": 30878 + }, + { + "epoch": 1.7252563063943906, + "grad_norm": 0.48653173446655273, + "learning_rate": 0.0001384188704616764, + "loss": 0.4974, + "step": 30879 + }, + { + "epoch": 1.7253121769980724, + "grad_norm": 0.5674911737442017, + "learning_rate": 0.00013839085611833257, + "loss": 0.4509, + "step": 30880 + }, + { + "epoch": 1.7253680476017543, + "grad_norm": 1.258640170097351, + "learning_rate": 0.00013836284177498877, + "loss": 0.4688, + "step": 30881 + }, + { + "epoch": 1.7254239182054363, + "grad_norm": 0.42033275961875916, + "learning_rate": 0.000138334827431645, + "loss": 0.378, + "step": 30882 + }, + { + "epoch": 1.725479788809118, + "grad_norm": 0.4599853754043579, + "learning_rate": 0.0001383068130883012, + "loss": 0.4236, + "step": 30883 + }, + { + "epoch": 1.7255356594128, + "grad_norm": 0.3661853075027466, + "learning_rate": 0.00013827879874495742, + "loss": 0.4179, + "step": 30884 + }, + { + "epoch": 1.725591530016482, + "grad_norm": 0.3727536201477051, + "learning_rate": 0.00013825078440161362, + "loss": 0.376, + "step": 30885 + }, + { + "epoch": 1.7256474006201636, + "grad_norm": 1.9128004312515259, + "learning_rate": 0.00013822277005826983, + "loss": 0.4387, + "step": 30886 + }, + { + "epoch": 1.7257032712238456, + "grad_norm": 0.4011991620063782, + "learning_rate": 0.00013819475571492603, + "loss": 0.5265, + "step": 30887 + }, + { + "epoch": 1.7257591418275275, + "grad_norm": 0.49436208605766296, + "learning_rate": 0.00013816674137158224, + "loss": 0.3419, + "step": 30888 + }, + { + "epoch": 1.7258150124312093, + "grad_norm": 0.8173214793205261, + "learning_rate": 0.00013813872702823847, + "loss": 0.5739, + "step": 30889 + }, + { + "epoch": 1.7258708830348912, + "grad_norm": 0.3572717010974884, + "learning_rate": 0.00013811071268489468, + "loss": 0.4064, + "step": 30890 + }, + { + "epoch": 1.7259267536385732, + "grad_norm": 0.31569305062294006, + "learning_rate": 0.00013808269834155088, + "loss": 0.3041, + "step": 30891 + }, + { + "epoch": 1.725982624242255, + "grad_norm": 1.1018387079238892, + "learning_rate": 0.0001380546839982071, + "loss": 0.5679, + "step": 30892 + }, + { + "epoch": 1.7260384948459369, + "grad_norm": 0.612281858921051, + "learning_rate": 0.0001380266696548633, + "loss": 0.4347, + "step": 30893 + }, + { + "epoch": 1.7260943654496188, + "grad_norm": 0.28658416867256165, + "learning_rate": 0.0001379986553115195, + "loss": 0.3782, + "step": 30894 + }, + { + "epoch": 1.7261502360533005, + "grad_norm": 0.3344983160495758, + "learning_rate": 0.0001379706409681757, + "loss": 0.3379, + "step": 30895 + }, + { + "epoch": 1.7262061066569823, + "grad_norm": 0.5388743281364441, + "learning_rate": 0.0001379426266248319, + "loss": 0.4575, + "step": 30896 + }, + { + "epoch": 1.7262619772606644, + "grad_norm": 0.8676743507385254, + "learning_rate": 0.00013791461228148815, + "loss": 0.4354, + "step": 30897 + }, + { + "epoch": 1.7263178478643462, + "grad_norm": 0.4207872450351715, + "learning_rate": 0.00013788659793814435, + "loss": 0.407, + "step": 30898 + }, + { + "epoch": 1.726373718468028, + "grad_norm": 0.44370052218437195, + "learning_rate": 0.00013785858359480056, + "loss": 0.3638, + "step": 30899 + }, + { + "epoch": 1.72642958907171, + "grad_norm": 2.1019575595855713, + "learning_rate": 0.00013783056925145676, + "loss": 0.4475, + "step": 30900 + }, + { + "epoch": 1.7264854596753918, + "grad_norm": 0.46570083498954773, + "learning_rate": 0.00013780255490811297, + "loss": 0.3457, + "step": 30901 + }, + { + "epoch": 1.7265413302790735, + "grad_norm": 0.4783811569213867, + "learning_rate": 0.00013777454056476915, + "loss": 0.423, + "step": 30902 + }, + { + "epoch": 1.7265972008827557, + "grad_norm": 0.48212575912475586, + "learning_rate": 0.00013774652622142535, + "loss": 0.436, + "step": 30903 + }, + { + "epoch": 1.7266530714864374, + "grad_norm": 0.6289594173431396, + "learning_rate": 0.00013771851187808156, + "loss": 0.6832, + "step": 30904 + }, + { + "epoch": 1.7267089420901192, + "grad_norm": 0.47125673294067383, + "learning_rate": 0.0001376904975347378, + "loss": 0.318, + "step": 30905 + }, + { + "epoch": 1.726764812693801, + "grad_norm": 0.44966134428977966, + "learning_rate": 0.000137662483191394, + "loss": 0.384, + "step": 30906 + }, + { + "epoch": 1.726820683297483, + "grad_norm": 0.6160604953765869, + "learning_rate": 0.0001376344688480502, + "loss": 0.4143, + "step": 30907 + }, + { + "epoch": 1.7268765539011648, + "grad_norm": 0.6094693541526794, + "learning_rate": 0.0001376064545047064, + "loss": 0.3369, + "step": 30908 + }, + { + "epoch": 1.7269324245048467, + "grad_norm": 0.5544642806053162, + "learning_rate": 0.00013757844016136262, + "loss": 0.5541, + "step": 30909 + }, + { + "epoch": 1.7269882951085287, + "grad_norm": 0.6102953553199768, + "learning_rate": 0.00013755042581801882, + "loss": 0.4398, + "step": 30910 + }, + { + "epoch": 1.7270441657122104, + "grad_norm": 0.45523369312286377, + "learning_rate": 0.00013752241147467503, + "loss": 0.424, + "step": 30911 + }, + { + "epoch": 1.7271000363158924, + "grad_norm": 0.5463778376579285, + "learning_rate": 0.00013749439713133123, + "loss": 0.415, + "step": 30912 + }, + { + "epoch": 1.7271559069195743, + "grad_norm": 0.34927499294281006, + "learning_rate": 0.00013746638278798747, + "loss": 0.4506, + "step": 30913 + }, + { + "epoch": 1.727211777523256, + "grad_norm": 0.4607028663158417, + "learning_rate": 0.00013743836844464367, + "loss": 0.5401, + "step": 30914 + }, + { + "epoch": 1.727267648126938, + "grad_norm": 0.3479473888874054, + "learning_rate": 0.00013741035410129988, + "loss": 0.3894, + "step": 30915 + }, + { + "epoch": 1.72732351873062, + "grad_norm": 0.38097524642944336, + "learning_rate": 0.00013738233975795608, + "loss": 0.3063, + "step": 30916 + }, + { + "epoch": 1.7273793893343017, + "grad_norm": 0.391735702753067, + "learning_rate": 0.0001373543254146123, + "loss": 0.3868, + "step": 30917 + }, + { + "epoch": 1.7274352599379836, + "grad_norm": 0.44545358419418335, + "learning_rate": 0.0001373263110712685, + "loss": 0.3682, + "step": 30918 + }, + { + "epoch": 1.7274911305416656, + "grad_norm": 0.7714049220085144, + "learning_rate": 0.0001372982967279247, + "loss": 0.4465, + "step": 30919 + }, + { + "epoch": 1.7275470011453473, + "grad_norm": 0.33565858006477356, + "learning_rate": 0.0001372702823845809, + "loss": 0.4347, + "step": 30920 + }, + { + "epoch": 1.7276028717490293, + "grad_norm": 0.29093244671821594, + "learning_rate": 0.00013724226804123714, + "loss": 0.387, + "step": 30921 + }, + { + "epoch": 1.7276587423527112, + "grad_norm": 0.429360568523407, + "learning_rate": 0.00013721425369789335, + "loss": 0.4753, + "step": 30922 + }, + { + "epoch": 1.727714612956393, + "grad_norm": 0.45652133226394653, + "learning_rate": 0.00013718623935454952, + "loss": 0.4418, + "step": 30923 + }, + { + "epoch": 1.727770483560075, + "grad_norm": 0.4369628131389618, + "learning_rate": 0.00013715822501120573, + "loss": 0.462, + "step": 30924 + }, + { + "epoch": 1.7278263541637569, + "grad_norm": 0.37248721718788147, + "learning_rate": 0.00013713021066786194, + "loss": 0.4041, + "step": 30925 + }, + { + "epoch": 1.7278822247674386, + "grad_norm": 0.486312597990036, + "learning_rate": 0.00013710219632451814, + "loss": 0.436, + "step": 30926 + }, + { + "epoch": 1.7279380953711205, + "grad_norm": 0.5225654244422913, + "learning_rate": 0.00013707418198117435, + "loss": 0.4806, + "step": 30927 + }, + { + "epoch": 1.7279939659748025, + "grad_norm": 0.48998144268989563, + "learning_rate": 0.00013704616763783055, + "loss": 0.4372, + "step": 30928 + }, + { + "epoch": 1.7280498365784842, + "grad_norm": 0.9414536952972412, + "learning_rate": 0.00013701815329448679, + "loss": 0.4714, + "step": 30929 + }, + { + "epoch": 1.728105707182166, + "grad_norm": 0.5916061401367188, + "learning_rate": 0.000136990138951143, + "loss": 0.5599, + "step": 30930 + }, + { + "epoch": 1.7281615777858481, + "grad_norm": 1.0068168640136719, + "learning_rate": 0.0001369621246077992, + "loss": 0.3793, + "step": 30931 + }, + { + "epoch": 1.7282174483895298, + "grad_norm": 0.602673351764679, + "learning_rate": 0.0001369341102644554, + "loss": 0.4623, + "step": 30932 + }, + { + "epoch": 1.7282733189932116, + "grad_norm": 0.35290247201919556, + "learning_rate": 0.0001369060959211116, + "loss": 0.3964, + "step": 30933 + }, + { + "epoch": 1.7283291895968937, + "grad_norm": 0.5064780116081238, + "learning_rate": 0.00013687808157776782, + "loss": 0.4763, + "step": 30934 + }, + { + "epoch": 1.7283850602005755, + "grad_norm": 1.841292142868042, + "learning_rate": 0.00013685006723442402, + "loss": 0.397, + "step": 30935 + }, + { + "epoch": 1.7284409308042572, + "grad_norm": 0.39145392179489136, + "learning_rate": 0.00013682205289108023, + "loss": 0.5057, + "step": 30936 + }, + { + "epoch": 1.7284968014079394, + "grad_norm": 0.3069743514060974, + "learning_rate": 0.00013679403854773646, + "loss": 0.3621, + "step": 30937 + }, + { + "epoch": 1.728552672011621, + "grad_norm": 0.6159598231315613, + "learning_rate": 0.00013676602420439267, + "loss": 0.5006, + "step": 30938 + }, + { + "epoch": 1.7286085426153028, + "grad_norm": 0.3779817223548889, + "learning_rate": 0.00013673800986104887, + "loss": 0.4499, + "step": 30939 + }, + { + "epoch": 1.7286644132189848, + "grad_norm": 0.5143275260925293, + "learning_rate": 0.00013670999551770508, + "loss": 0.353, + "step": 30940 + }, + { + "epoch": 1.7287202838226667, + "grad_norm": 0.6018110513687134, + "learning_rate": 0.00013668198117436128, + "loss": 0.4287, + "step": 30941 + }, + { + "epoch": 1.7287761544263485, + "grad_norm": 0.44096651673316956, + "learning_rate": 0.0001366539668310175, + "loss": 0.4394, + "step": 30942 + }, + { + "epoch": 1.7288320250300304, + "grad_norm": 0.6867247223854065, + "learning_rate": 0.0001366259524876737, + "loss": 0.3884, + "step": 30943 + }, + { + "epoch": 1.7288878956337124, + "grad_norm": 0.39995449781417847, + "learning_rate": 0.00013659793814432987, + "loss": 0.3902, + "step": 30944 + }, + { + "epoch": 1.728943766237394, + "grad_norm": 0.5000678896903992, + "learning_rate": 0.0001365699238009861, + "loss": 0.4832, + "step": 30945 + }, + { + "epoch": 1.728999636841076, + "grad_norm": 0.3055984377861023, + "learning_rate": 0.0001365419094576423, + "loss": 0.3005, + "step": 30946 + }, + { + "epoch": 1.729055507444758, + "grad_norm": 0.49959883093833923, + "learning_rate": 0.00013651389511429852, + "loss": 0.5195, + "step": 30947 + }, + { + "epoch": 1.7291113780484397, + "grad_norm": 0.5434229373931885, + "learning_rate": 0.00013648588077095472, + "loss": 0.3618, + "step": 30948 + }, + { + "epoch": 1.7291672486521217, + "grad_norm": 0.5860532522201538, + "learning_rate": 0.00013645786642761093, + "loss": 0.3831, + "step": 30949 + }, + { + "epoch": 1.7292231192558036, + "grad_norm": 0.41584864258766174, + "learning_rate": 0.00013642985208426714, + "loss": 0.3456, + "step": 30950 + }, + { + "epoch": 1.7292789898594854, + "grad_norm": 0.4646633267402649, + "learning_rate": 0.00013640183774092334, + "loss": 0.4094, + "step": 30951 + }, + { + "epoch": 1.7293348604631673, + "grad_norm": 0.8869543075561523, + "learning_rate": 0.00013637382339757955, + "loss": 0.5275, + "step": 30952 + }, + { + "epoch": 1.7293907310668493, + "grad_norm": 1.075769305229187, + "learning_rate": 0.00013634580905423578, + "loss": 0.4205, + "step": 30953 + }, + { + "epoch": 1.729446601670531, + "grad_norm": 0.5374137759208679, + "learning_rate": 0.00013631779471089199, + "loss": 0.4187, + "step": 30954 + }, + { + "epoch": 1.729502472274213, + "grad_norm": 0.48667609691619873, + "learning_rate": 0.0001362897803675482, + "loss": 0.3304, + "step": 30955 + }, + { + "epoch": 1.729558342877895, + "grad_norm": 1.9421006441116333, + "learning_rate": 0.0001362617660242044, + "loss": 0.3989, + "step": 30956 + }, + { + "epoch": 1.7296142134815766, + "grad_norm": 0.5002951622009277, + "learning_rate": 0.0001362337516808606, + "loss": 0.4023, + "step": 30957 + }, + { + "epoch": 1.7296700840852586, + "grad_norm": 0.49374693632125854, + "learning_rate": 0.0001362057373375168, + "loss": 0.4466, + "step": 30958 + }, + { + "epoch": 1.7297259546889405, + "grad_norm": 0.6086283922195435, + "learning_rate": 0.00013617772299417301, + "loss": 0.3937, + "step": 30959 + }, + { + "epoch": 1.7297818252926223, + "grad_norm": 0.31939271092414856, + "learning_rate": 0.00013614970865082925, + "loss": 0.3432, + "step": 30960 + }, + { + "epoch": 1.7298376958963042, + "grad_norm": 0.4441719651222229, + "learning_rate": 0.00013612169430748545, + "loss": 0.4009, + "step": 30961 + }, + { + "epoch": 1.7298935664999862, + "grad_norm": 0.3687482476234436, + "learning_rate": 0.00013609367996414166, + "loss": 0.4245, + "step": 30962 + }, + { + "epoch": 1.7299494371036679, + "grad_norm": 0.6150199770927429, + "learning_rate": 0.00013606566562079787, + "loss": 0.5144, + "step": 30963 + }, + { + "epoch": 1.7300053077073496, + "grad_norm": 0.46657538414001465, + "learning_rate": 0.00013603765127745407, + "loss": 0.4487, + "step": 30964 + }, + { + "epoch": 1.7300611783110318, + "grad_norm": 0.49157729744911194, + "learning_rate": 0.00013600963693411025, + "loss": 0.5644, + "step": 30965 + }, + { + "epoch": 1.7301170489147135, + "grad_norm": 0.3859769403934479, + "learning_rate": 0.00013598162259076646, + "loss": 0.4136, + "step": 30966 + }, + { + "epoch": 1.7301729195183952, + "grad_norm": 0.4696807563304901, + "learning_rate": 0.00013595360824742266, + "loss": 0.3552, + "step": 30967 + }, + { + "epoch": 1.7302287901220774, + "grad_norm": 0.6753506064414978, + "learning_rate": 0.0001359255939040789, + "loss": 0.3544, + "step": 30968 + }, + { + "epoch": 1.7302846607257591, + "grad_norm": 4.071242809295654, + "learning_rate": 0.0001358975795607351, + "loss": 0.4234, + "step": 30969 + }, + { + "epoch": 1.7303405313294409, + "grad_norm": 0.44031891226768494, + "learning_rate": 0.0001358695652173913, + "loss": 0.3777, + "step": 30970 + }, + { + "epoch": 1.730396401933123, + "grad_norm": 0.3388434946537018, + "learning_rate": 0.0001358415508740475, + "loss": 0.3177, + "step": 30971 + }, + { + "epoch": 1.7304522725368048, + "grad_norm": 0.4157068133354187, + "learning_rate": 0.00013581353653070372, + "loss": 0.4294, + "step": 30972 + }, + { + "epoch": 1.7305081431404865, + "grad_norm": 0.3614494204521179, + "learning_rate": 0.00013578552218735992, + "loss": 0.43, + "step": 30973 + }, + { + "epoch": 1.7305640137441685, + "grad_norm": 0.4893716275691986, + "learning_rate": 0.00013575750784401613, + "loss": 0.394, + "step": 30974 + }, + { + "epoch": 1.7306198843478504, + "grad_norm": 0.3842774033546448, + "learning_rate": 0.00013572949350067234, + "loss": 0.3812, + "step": 30975 + }, + { + "epoch": 1.7306757549515321, + "grad_norm": 1.7047598361968994, + "learning_rate": 0.00013570147915732857, + "loss": 0.3967, + "step": 30976 + }, + { + "epoch": 1.730731625555214, + "grad_norm": 0.6465632915496826, + "learning_rate": 0.00013567346481398477, + "loss": 0.6735, + "step": 30977 + }, + { + "epoch": 1.730787496158896, + "grad_norm": 0.4547126591205597, + "learning_rate": 0.00013564545047064098, + "loss": 0.4895, + "step": 30978 + }, + { + "epoch": 1.7308433667625778, + "grad_norm": 0.3288690149784088, + "learning_rate": 0.00013561743612729719, + "loss": 0.3883, + "step": 30979 + }, + { + "epoch": 1.7308992373662597, + "grad_norm": 0.4935647249221802, + "learning_rate": 0.0001355894217839534, + "loss": 0.3732, + "step": 30980 + }, + { + "epoch": 1.7309551079699417, + "grad_norm": 0.4119088351726532, + "learning_rate": 0.0001355614074406096, + "loss": 0.3872, + "step": 30981 + }, + { + "epoch": 1.7310109785736234, + "grad_norm": 0.7846087217330933, + "learning_rate": 0.0001355333930972658, + "loss": 0.5661, + "step": 30982 + }, + { + "epoch": 1.7310668491773054, + "grad_norm": 0.4246925711631775, + "learning_rate": 0.000135505378753922, + "loss": 0.384, + "step": 30983 + }, + { + "epoch": 1.7311227197809873, + "grad_norm": 0.33998212218284607, + "learning_rate": 0.00013547736441057824, + "loss": 0.3939, + "step": 30984 + }, + { + "epoch": 1.731178590384669, + "grad_norm": 0.45100274682044983, + "learning_rate": 0.00013544935006723445, + "loss": 0.3334, + "step": 30985 + }, + { + "epoch": 1.731234460988351, + "grad_norm": 0.5178055763244629, + "learning_rate": 0.00013542133572389065, + "loss": 0.4588, + "step": 30986 + }, + { + "epoch": 1.731290331592033, + "grad_norm": 0.5051098465919495, + "learning_rate": 0.00013539332138054683, + "loss": 0.3576, + "step": 30987 + }, + { + "epoch": 1.7313462021957147, + "grad_norm": 0.4301167130470276, + "learning_rate": 0.00013536530703720304, + "loss": 0.4192, + "step": 30988 + }, + { + "epoch": 1.7314020727993966, + "grad_norm": 0.3328537940979004, + "learning_rate": 0.00013533729269385924, + "loss": 0.3849, + "step": 30989 + }, + { + "epoch": 1.7314579434030786, + "grad_norm": 0.42320185899734497, + "learning_rate": 0.00013530927835051545, + "loss": 0.4957, + "step": 30990 + }, + { + "epoch": 1.7315138140067603, + "grad_norm": 0.3554709553718567, + "learning_rate": 0.00013528126400717166, + "loss": 0.4493, + "step": 30991 + }, + { + "epoch": 1.7315696846104423, + "grad_norm": 0.9468709826469421, + "learning_rate": 0.0001352532496638279, + "loss": 0.2971, + "step": 30992 + }, + { + "epoch": 1.7316255552141242, + "grad_norm": 0.31953486800193787, + "learning_rate": 0.0001352252353204841, + "loss": 0.3015, + "step": 30993 + }, + { + "epoch": 1.731681425817806, + "grad_norm": 0.4190433621406555, + "learning_rate": 0.0001351972209771403, + "loss": 0.4368, + "step": 30994 + }, + { + "epoch": 1.7317372964214879, + "grad_norm": 0.48685023188591003, + "learning_rate": 0.0001351692066337965, + "loss": 0.588, + "step": 30995 + }, + { + "epoch": 1.7317931670251698, + "grad_norm": 0.39189139008522034, + "learning_rate": 0.0001351411922904527, + "loss": 0.3907, + "step": 30996 + }, + { + "epoch": 1.7318490376288516, + "grad_norm": 0.43896716833114624, + "learning_rate": 0.00013511317794710892, + "loss": 0.4514, + "step": 30997 + }, + { + "epoch": 1.7319049082325333, + "grad_norm": 0.4212152361869812, + "learning_rate": 0.00013508516360376512, + "loss": 0.3937, + "step": 30998 + }, + { + "epoch": 1.7319607788362155, + "grad_norm": 0.36237862706184387, + "learning_rate": 0.00013505714926042133, + "loss": 0.4126, + "step": 30999 + }, + { + "epoch": 1.7320166494398972, + "grad_norm": 0.3610214591026306, + "learning_rate": 0.00013502913491707756, + "loss": 0.4529, + "step": 31000 + }, + { + "epoch": 1.7320166494398972, + "eval_cer": 0.08433439167294032, + "eval_loss": 0.31499484181404114, + "eval_runtime": 56.4308, + "eval_samples_per_second": 80.417, + "eval_steps_per_second": 5.033, + "eval_wer": 0.3363034552107483, + "step": 31000 + }, + { + "epoch": 1.732072520043579, + "grad_norm": 0.8275609612464905, + "learning_rate": 0.00013500112057373377, + "loss": 0.466, + "step": 31001 + }, + { + "epoch": 1.732128390647261, + "grad_norm": 0.5324826240539551, + "learning_rate": 0.00013497310623038997, + "loss": 0.5049, + "step": 31002 + }, + { + "epoch": 1.7321842612509428, + "grad_norm": 0.42438164353370667, + "learning_rate": 0.00013494509188704618, + "loss": 0.4252, + "step": 31003 + }, + { + "epoch": 1.7322401318546246, + "grad_norm": 0.3826936185359955, + "learning_rate": 0.00013491707754370239, + "loss": 0.4402, + "step": 31004 + }, + { + "epoch": 1.7322960024583067, + "grad_norm": 0.4463058114051819, + "learning_rate": 0.0001348890632003586, + "loss": 0.4426, + "step": 31005 + }, + { + "epoch": 1.7323518730619885, + "grad_norm": 0.5524913668632507, + "learning_rate": 0.0001348610488570148, + "loss": 0.4398, + "step": 31006 + }, + { + "epoch": 1.7324077436656702, + "grad_norm": 0.4002235531806946, + "learning_rate": 0.000134833034513671, + "loss": 0.4653, + "step": 31007 + }, + { + "epoch": 1.7324636142693521, + "grad_norm": 1.0289735794067383, + "learning_rate": 0.0001348050201703272, + "loss": 0.4583, + "step": 31008 + }, + { + "epoch": 1.732519484873034, + "grad_norm": 0.43383651971817017, + "learning_rate": 0.00013477700582698341, + "loss": 0.3854, + "step": 31009 + }, + { + "epoch": 1.7325753554767158, + "grad_norm": 0.3606805205345154, + "learning_rate": 0.00013474899148363962, + "loss": 0.4158, + "step": 31010 + }, + { + "epoch": 1.7326312260803978, + "grad_norm": 0.38262665271759033, + "learning_rate": 0.00013472097714029583, + "loss": 0.4153, + "step": 31011 + }, + { + "epoch": 1.7326870966840797, + "grad_norm": 0.4235116243362427, + "learning_rate": 0.00013469296279695203, + "loss": 0.5207, + "step": 31012 + }, + { + "epoch": 1.7327429672877614, + "grad_norm": 0.7679775357246399, + "learning_rate": 0.00013466494845360824, + "loss": 0.3986, + "step": 31013 + }, + { + "epoch": 1.7327988378914434, + "grad_norm": 0.38258063793182373, + "learning_rate": 0.00013463693411026444, + "loss": 0.4005, + "step": 31014 + }, + { + "epoch": 1.7328547084951254, + "grad_norm": 0.33409059047698975, + "learning_rate": 0.00013460891976692065, + "loss": 0.3687, + "step": 31015 + }, + { + "epoch": 1.732910579098807, + "grad_norm": 0.36102962493896484, + "learning_rate": 0.00013458090542357688, + "loss": 0.4612, + "step": 31016 + }, + { + "epoch": 1.732966449702489, + "grad_norm": 0.4020494222640991, + "learning_rate": 0.0001345528910802331, + "loss": 0.4384, + "step": 31017 + }, + { + "epoch": 1.733022320306171, + "grad_norm": 0.5964098572731018, + "learning_rate": 0.0001345248767368893, + "loss": 0.5069, + "step": 31018 + }, + { + "epoch": 1.7330781909098527, + "grad_norm": 0.5249307751655579, + "learning_rate": 0.0001344968623935455, + "loss": 0.3632, + "step": 31019 + }, + { + "epoch": 1.7331340615135347, + "grad_norm": 0.4831807315349579, + "learning_rate": 0.0001344688480502017, + "loss": 0.4561, + "step": 31020 + }, + { + "epoch": 1.7331899321172166, + "grad_norm": 0.4131432771682739, + "learning_rate": 0.0001344408337068579, + "loss": 0.4577, + "step": 31021 + }, + { + "epoch": 1.7332458027208983, + "grad_norm": 0.33139026165008545, + "learning_rate": 0.00013441281936351412, + "loss": 0.3674, + "step": 31022 + }, + { + "epoch": 1.7333016733245803, + "grad_norm": 1.054318904876709, + "learning_rate": 0.00013438480502017035, + "loss": 0.6235, + "step": 31023 + }, + { + "epoch": 1.7333575439282622, + "grad_norm": 0.4084649682044983, + "learning_rate": 0.00013435679067682656, + "loss": 0.4004, + "step": 31024 + }, + { + "epoch": 1.733413414531944, + "grad_norm": 0.3677787780761719, + "learning_rate": 0.00013432877633348276, + "loss": 0.5457, + "step": 31025 + }, + { + "epoch": 1.733469285135626, + "grad_norm": 0.2937262952327728, + "learning_rate": 0.00013430076199013897, + "loss": 0.2831, + "step": 31026 + }, + { + "epoch": 1.7335251557393079, + "grad_norm": 0.3788090646266937, + "learning_rate": 0.00013427274764679517, + "loss": 0.4445, + "step": 31027 + }, + { + "epoch": 1.7335810263429896, + "grad_norm": 0.5207070708274841, + "learning_rate": 0.00013424473330345138, + "loss": 0.4303, + "step": 31028 + }, + { + "epoch": 1.7336368969466716, + "grad_norm": 5.33277702331543, + "learning_rate": 0.00013421671896010756, + "loss": 0.4285, + "step": 31029 + }, + { + "epoch": 1.7336927675503535, + "grad_norm": 0.5074042677879333, + "learning_rate": 0.00013418870461676376, + "loss": 0.5062, + "step": 31030 + }, + { + "epoch": 1.7337486381540352, + "grad_norm": 0.49021464586257935, + "learning_rate": 0.00013416069027342, + "loss": 0.4596, + "step": 31031 + }, + { + "epoch": 1.733804508757717, + "grad_norm": 0.7205792665481567, + "learning_rate": 0.0001341326759300762, + "loss": 0.5197, + "step": 31032 + }, + { + "epoch": 1.7338603793613991, + "grad_norm": 0.4970436990261078, + "learning_rate": 0.0001341046615867324, + "loss": 0.4356, + "step": 31033 + }, + { + "epoch": 1.7339162499650809, + "grad_norm": 0.40369269251823425, + "learning_rate": 0.00013407664724338861, + "loss": 0.4377, + "step": 31034 + }, + { + "epoch": 1.7339721205687626, + "grad_norm": 4.842221736907959, + "learning_rate": 0.00013404863290004482, + "loss": 0.4725, + "step": 31035 + }, + { + "epoch": 1.7340279911724448, + "grad_norm": 0.5586792230606079, + "learning_rate": 0.00013402061855670103, + "loss": 0.6103, + "step": 31036 + }, + { + "epoch": 1.7340838617761265, + "grad_norm": 0.3925815522670746, + "learning_rate": 0.00013399260421335723, + "loss": 0.3258, + "step": 31037 + }, + { + "epoch": 1.7341397323798082, + "grad_norm": 0.4245203137397766, + "learning_rate": 0.00013396458987001344, + "loss": 0.3293, + "step": 31038 + }, + { + "epoch": 1.7341956029834904, + "grad_norm": 1.1292836666107178, + "learning_rate": 0.00013393657552666967, + "loss": 0.3285, + "step": 31039 + }, + { + "epoch": 1.7342514735871721, + "grad_norm": 0.3371758759021759, + "learning_rate": 0.00013390856118332588, + "loss": 0.3666, + "step": 31040 + }, + { + "epoch": 1.7343073441908539, + "grad_norm": 0.6947349309921265, + "learning_rate": 0.00013388054683998208, + "loss": 0.3945, + "step": 31041 + }, + { + "epoch": 1.7343632147945358, + "grad_norm": 0.9562290906906128, + "learning_rate": 0.0001338525324966383, + "loss": 0.4255, + "step": 31042 + }, + { + "epoch": 1.7344190853982178, + "grad_norm": 0.34689250588417053, + "learning_rate": 0.0001338245181532945, + "loss": 0.5038, + "step": 31043 + }, + { + "epoch": 1.7344749560018995, + "grad_norm": 0.681873619556427, + "learning_rate": 0.0001337965038099507, + "loss": 0.415, + "step": 31044 + }, + { + "epoch": 1.7345308266055814, + "grad_norm": 0.38084664940834045, + "learning_rate": 0.0001337684894666069, + "loss": 0.4516, + "step": 31045 + }, + { + "epoch": 1.7345866972092634, + "grad_norm": 0.43279528617858887, + "learning_rate": 0.0001337404751232631, + "loss": 0.3892, + "step": 31046 + }, + { + "epoch": 1.7346425678129451, + "grad_norm": 0.6411974430084229, + "learning_rate": 0.00013371246077991934, + "loss": 0.4627, + "step": 31047 + }, + { + "epoch": 1.734698438416627, + "grad_norm": 0.43288424611091614, + "learning_rate": 0.00013368444643657555, + "loss": 0.3386, + "step": 31048 + }, + { + "epoch": 1.734754309020309, + "grad_norm": 0.39559227228164673, + "learning_rate": 0.00013365643209323176, + "loss": 0.4312, + "step": 31049 + }, + { + "epoch": 1.7348101796239908, + "grad_norm": 0.39527738094329834, + "learning_rate": 0.00013362841774988793, + "loss": 0.3777, + "step": 31050 + }, + { + "epoch": 1.7348660502276727, + "grad_norm": 0.5587428212165833, + "learning_rate": 0.00013360040340654414, + "loss": 0.4533, + "step": 31051 + }, + { + "epoch": 1.7349219208313547, + "grad_norm": 6.552061557769775, + "learning_rate": 0.00013357238906320035, + "loss": 0.4406, + "step": 31052 + }, + { + "epoch": 1.7349777914350364, + "grad_norm": 1.0962845087051392, + "learning_rate": 0.00013354437471985655, + "loss": 0.575, + "step": 31053 + }, + { + "epoch": 1.7350336620387183, + "grad_norm": 0.7598211765289307, + "learning_rate": 0.00013351636037651276, + "loss": 0.4739, + "step": 31054 + }, + { + "epoch": 1.7350895326424003, + "grad_norm": 0.5359531044960022, + "learning_rate": 0.000133488346033169, + "loss": 0.4347, + "step": 31055 + }, + { + "epoch": 1.735145403246082, + "grad_norm": 1.1544464826583862, + "learning_rate": 0.0001334603316898252, + "loss": 0.5086, + "step": 31056 + }, + { + "epoch": 1.735201273849764, + "grad_norm": 2.5602879524230957, + "learning_rate": 0.0001334323173464814, + "loss": 0.3609, + "step": 31057 + }, + { + "epoch": 1.735257144453446, + "grad_norm": 0.40687260031700134, + "learning_rate": 0.0001334043030031376, + "loss": 0.4373, + "step": 31058 + }, + { + "epoch": 1.7353130150571277, + "grad_norm": 0.35096877813339233, + "learning_rate": 0.0001333762886597938, + "loss": 0.2576, + "step": 31059 + }, + { + "epoch": 1.7353688856608096, + "grad_norm": 0.341682106256485, + "learning_rate": 0.00013334827431645002, + "loss": 0.369, + "step": 31060 + }, + { + "epoch": 1.7354247562644916, + "grad_norm": 1.056057333946228, + "learning_rate": 0.00013332025997310622, + "loss": 0.5712, + "step": 31061 + }, + { + "epoch": 1.7354806268681733, + "grad_norm": 0.94325190782547, + "learning_rate": 0.00013329224562976243, + "loss": 0.3473, + "step": 31062 + }, + { + "epoch": 1.7355364974718552, + "grad_norm": 0.5166191458702087, + "learning_rate": 0.00013326423128641866, + "loss": 0.4344, + "step": 31063 + }, + { + "epoch": 1.7355923680755372, + "grad_norm": 0.464069664478302, + "learning_rate": 0.00013323621694307487, + "loss": 0.4725, + "step": 31064 + }, + { + "epoch": 1.735648238679219, + "grad_norm": 0.35138222575187683, + "learning_rate": 0.00013320820259973108, + "loss": 0.3263, + "step": 31065 + }, + { + "epoch": 1.7357041092829006, + "grad_norm": 0.42598477005958557, + "learning_rate": 0.00013318018825638728, + "loss": 0.3696, + "step": 31066 + }, + { + "epoch": 1.7357599798865828, + "grad_norm": 0.5460772514343262, + "learning_rate": 0.0001331521739130435, + "loss": 0.3558, + "step": 31067 + }, + { + "epoch": 1.7358158504902645, + "grad_norm": 0.5022573471069336, + "learning_rate": 0.0001331241595696997, + "loss": 0.4095, + "step": 31068 + }, + { + "epoch": 1.7358717210939463, + "grad_norm": 0.41019928455352783, + "learning_rate": 0.0001330961452263559, + "loss": 0.4373, + "step": 31069 + }, + { + "epoch": 1.7359275916976284, + "grad_norm": 0.4570836126804352, + "learning_rate": 0.0001330681308830121, + "loss": 0.3723, + "step": 31070 + }, + { + "epoch": 1.7359834623013102, + "grad_norm": 0.3171563744544983, + "learning_rate": 0.00013304011653966834, + "loss": 0.3461, + "step": 31071 + }, + { + "epoch": 1.736039332904992, + "grad_norm": 0.3483501076698303, + "learning_rate": 0.00013301210219632452, + "loss": 0.3362, + "step": 31072 + }, + { + "epoch": 1.7360952035086739, + "grad_norm": 0.5755283832550049, + "learning_rate": 0.00013298408785298072, + "loss": 0.4321, + "step": 31073 + }, + { + "epoch": 1.7361510741123558, + "grad_norm": 0.6494706273078918, + "learning_rate": 0.00013295607350963693, + "loss": 0.4395, + "step": 31074 + }, + { + "epoch": 1.7362069447160375, + "grad_norm": 0.3674270510673523, + "learning_rate": 0.00013292805916629313, + "loss": 0.3922, + "step": 31075 + }, + { + "epoch": 1.7362628153197195, + "grad_norm": 0.3999558687210083, + "learning_rate": 0.00013290004482294934, + "loss": 0.43, + "step": 31076 + }, + { + "epoch": 1.7363186859234014, + "grad_norm": 0.46602579951286316, + "learning_rate": 0.00013287203047960555, + "loss": 0.3757, + "step": 31077 + }, + { + "epoch": 1.7363745565270832, + "grad_norm": 1.4486316442489624, + "learning_rate": 0.00013284401613626175, + "loss": 0.406, + "step": 31078 + }, + { + "epoch": 1.7364304271307651, + "grad_norm": 0.5638400316238403, + "learning_rate": 0.00013281600179291798, + "loss": 0.3925, + "step": 31079 + }, + { + "epoch": 1.736486297734447, + "grad_norm": 0.48983442783355713, + "learning_rate": 0.0001327879874495742, + "loss": 0.5884, + "step": 31080 + }, + { + "epoch": 1.7365421683381288, + "grad_norm": 0.9937781691551208, + "learning_rate": 0.0001327599731062304, + "loss": 0.3286, + "step": 31081 + }, + { + "epoch": 1.7365980389418108, + "grad_norm": 0.3489297926425934, + "learning_rate": 0.0001327319587628866, + "loss": 0.3762, + "step": 31082 + }, + { + "epoch": 1.7366539095454927, + "grad_norm": 0.3730258643627167, + "learning_rate": 0.0001327039444195428, + "loss": 0.4174, + "step": 31083 + }, + { + "epoch": 1.7367097801491744, + "grad_norm": 1.901500940322876, + "learning_rate": 0.000132675930076199, + "loss": 0.4321, + "step": 31084 + }, + { + "epoch": 1.7367656507528564, + "grad_norm": 2.028062343597412, + "learning_rate": 0.00013264791573285522, + "loss": 0.5007, + "step": 31085 + }, + { + "epoch": 1.7368215213565383, + "grad_norm": 0.32920533418655396, + "learning_rate": 0.00013261990138951145, + "loss": 0.3354, + "step": 31086 + }, + { + "epoch": 1.73687739196022, + "grad_norm": 0.7415532469749451, + "learning_rate": 0.00013259188704616766, + "loss": 0.3918, + "step": 31087 + }, + { + "epoch": 1.736933262563902, + "grad_norm": 11.84549331665039, + "learning_rate": 0.00013256387270282386, + "loss": 0.4402, + "step": 31088 + }, + { + "epoch": 1.736989133167584, + "grad_norm": 1.6561448574066162, + "learning_rate": 0.00013253585835948007, + "loss": 0.436, + "step": 31089 + }, + { + "epoch": 1.7370450037712657, + "grad_norm": 0.4627458453178406, + "learning_rate": 0.00013250784401613627, + "loss": 0.4812, + "step": 31090 + }, + { + "epoch": 1.7371008743749476, + "grad_norm": 0.3902866244316101, + "learning_rate": 0.00013247982967279248, + "loss": 0.3419, + "step": 31091 + }, + { + "epoch": 1.7371567449786296, + "grad_norm": 0.5707401633262634, + "learning_rate": 0.0001324518153294487, + "loss": 0.4318, + "step": 31092 + }, + { + "epoch": 1.7372126155823113, + "grad_norm": 0.3462696075439453, + "learning_rate": 0.00013242380098610487, + "loss": 0.4009, + "step": 31093 + }, + { + "epoch": 1.7372684861859933, + "grad_norm": 0.5518768429756165, + "learning_rate": 0.0001323957866427611, + "loss": 0.4279, + "step": 31094 + }, + { + "epoch": 1.7373243567896752, + "grad_norm": 0.3877796530723572, + "learning_rate": 0.0001323677722994173, + "loss": 0.49, + "step": 31095 + }, + { + "epoch": 1.737380227393357, + "grad_norm": 0.5585708618164062, + "learning_rate": 0.0001323397579560735, + "loss": 0.6133, + "step": 31096 + }, + { + "epoch": 1.737436097997039, + "grad_norm": 0.55604487657547, + "learning_rate": 0.00013231174361272972, + "loss": 0.4001, + "step": 31097 + }, + { + "epoch": 1.7374919686007209, + "grad_norm": 2.1964614391326904, + "learning_rate": 0.00013228372926938592, + "loss": 0.4654, + "step": 31098 + }, + { + "epoch": 1.7375478392044026, + "grad_norm": 0.36311227083206177, + "learning_rate": 0.00013225571492604213, + "loss": 0.3284, + "step": 31099 + }, + { + "epoch": 1.7376037098080843, + "grad_norm": 0.36788326501846313, + "learning_rate": 0.00013222770058269833, + "loss": 0.3241, + "step": 31100 + }, + { + "epoch": 1.7376595804117665, + "grad_norm": 0.39193814992904663, + "learning_rate": 0.00013219968623935454, + "loss": 0.3909, + "step": 31101 + }, + { + "epoch": 1.7377154510154482, + "grad_norm": 0.4619545340538025, + "learning_rate": 0.00013217167189601077, + "loss": 0.4604, + "step": 31102 + }, + { + "epoch": 1.73777132161913, + "grad_norm": 0.4453307092189789, + "learning_rate": 0.00013214365755266698, + "loss": 0.3574, + "step": 31103 + }, + { + "epoch": 1.7378271922228121, + "grad_norm": 0.4608398377895355, + "learning_rate": 0.00013211564320932318, + "loss": 0.5159, + "step": 31104 + }, + { + "epoch": 1.7378830628264939, + "grad_norm": 0.529766857624054, + "learning_rate": 0.0001320876288659794, + "loss": 0.3636, + "step": 31105 + }, + { + "epoch": 1.7379389334301756, + "grad_norm": 0.4175630509853363, + "learning_rate": 0.0001320596145226356, + "loss": 0.6785, + "step": 31106 + }, + { + "epoch": 1.7379948040338575, + "grad_norm": 1.5267714262008667, + "learning_rate": 0.0001320316001792918, + "loss": 0.614, + "step": 31107 + }, + { + "epoch": 1.7380506746375395, + "grad_norm": 0.40172114968299866, + "learning_rate": 0.000132003585835948, + "loss": 0.4306, + "step": 31108 + }, + { + "epoch": 1.7381065452412212, + "grad_norm": 0.3514440655708313, + "learning_rate": 0.0001319755714926042, + "loss": 0.3682, + "step": 31109 + }, + { + "epoch": 1.7381624158449032, + "grad_norm": 0.4898416996002197, + "learning_rate": 0.00013194755714926045, + "loss": 0.3903, + "step": 31110 + }, + { + "epoch": 1.7382182864485851, + "grad_norm": 0.3821069002151489, + "learning_rate": 0.00013191954280591665, + "loss": 0.4357, + "step": 31111 + }, + { + "epoch": 1.7382741570522668, + "grad_norm": 0.35410749912261963, + "learning_rate": 0.00013189152846257286, + "loss": 0.4774, + "step": 31112 + }, + { + "epoch": 1.7383300276559488, + "grad_norm": 0.5774253606796265, + "learning_rate": 0.00013186351411922906, + "loss": 0.4493, + "step": 31113 + }, + { + "epoch": 1.7383858982596307, + "grad_norm": 1.091744303703308, + "learning_rate": 0.00013183549977588524, + "loss": 0.4006, + "step": 31114 + }, + { + "epoch": 1.7384417688633125, + "grad_norm": 0.5915024876594543, + "learning_rate": 0.00013180748543254145, + "loss": 0.5602, + "step": 31115 + }, + { + "epoch": 1.7384976394669944, + "grad_norm": 0.4901253879070282, + "learning_rate": 0.00013177947108919765, + "loss": 0.4972, + "step": 31116 + }, + { + "epoch": 1.7385535100706764, + "grad_norm": 0.4264788031578064, + "learning_rate": 0.00013175145674585386, + "loss": 0.3884, + "step": 31117 + }, + { + "epoch": 1.738609380674358, + "grad_norm": 0.3826189935207367, + "learning_rate": 0.0001317234424025101, + "loss": 0.3777, + "step": 31118 + }, + { + "epoch": 1.73866525127804, + "grad_norm": 0.32247528433799744, + "learning_rate": 0.0001316954280591663, + "loss": 0.3991, + "step": 31119 + }, + { + "epoch": 1.738721121881722, + "grad_norm": 2.321282386779785, + "learning_rate": 0.0001316674137158225, + "loss": 0.4335, + "step": 31120 + }, + { + "epoch": 1.7387769924854037, + "grad_norm": 4.4318366050720215, + "learning_rate": 0.0001316393993724787, + "loss": 0.3405, + "step": 31121 + }, + { + "epoch": 1.7388328630890857, + "grad_norm": 0.30644845962524414, + "learning_rate": 0.00013161138502913492, + "loss": 0.3215, + "step": 31122 + }, + { + "epoch": 1.7388887336927676, + "grad_norm": 0.685539186000824, + "learning_rate": 0.00013158337068579112, + "loss": 0.4276, + "step": 31123 + }, + { + "epoch": 1.7389446042964494, + "grad_norm": 0.48543602228164673, + "learning_rate": 0.00013155535634244733, + "loss": 0.3576, + "step": 31124 + }, + { + "epoch": 1.7390004749001313, + "grad_norm": 0.35359370708465576, + "learning_rate": 0.00013152734199910353, + "loss": 0.4002, + "step": 31125 + }, + { + "epoch": 1.7390563455038133, + "grad_norm": 0.45465564727783203, + "learning_rate": 0.00013149932765575977, + "loss": 0.5082, + "step": 31126 + }, + { + "epoch": 1.739112216107495, + "grad_norm": 0.8146563768386841, + "learning_rate": 0.00013147131331241597, + "loss": 0.3685, + "step": 31127 + }, + { + "epoch": 1.739168086711177, + "grad_norm": 1.345721960067749, + "learning_rate": 0.00013144329896907218, + "loss": 0.5254, + "step": 31128 + }, + { + "epoch": 1.739223957314859, + "grad_norm": 0.38614270091056824, + "learning_rate": 0.00013141528462572838, + "loss": 0.4118, + "step": 31129 + }, + { + "epoch": 1.7392798279185406, + "grad_norm": 0.3294007182121277, + "learning_rate": 0.0001313872702823846, + "loss": 0.3604, + "step": 31130 + }, + { + "epoch": 1.7393356985222226, + "grad_norm": 0.46251899003982544, + "learning_rate": 0.0001313592559390408, + "loss": 0.3604, + "step": 31131 + }, + { + "epoch": 1.7393915691259045, + "grad_norm": 0.690337598323822, + "learning_rate": 0.000131331241595697, + "loss": 0.431, + "step": 31132 + }, + { + "epoch": 1.7394474397295863, + "grad_norm": 0.3818475604057312, + "learning_rate": 0.0001313032272523532, + "loss": 0.479, + "step": 31133 + }, + { + "epoch": 1.739503310333268, + "grad_norm": 1.5260276794433594, + "learning_rate": 0.00013127521290900944, + "loss": 0.4644, + "step": 31134 + }, + { + "epoch": 1.7395591809369502, + "grad_norm": 0.44675418734550476, + "learning_rate": 0.00013124719856566562, + "loss": 0.4033, + "step": 31135 + }, + { + "epoch": 1.739615051540632, + "grad_norm": 0.4340117275714874, + "learning_rate": 0.00013121918422232182, + "loss": 0.4461, + "step": 31136 + }, + { + "epoch": 1.7396709221443136, + "grad_norm": 1.0260770320892334, + "learning_rate": 0.00013119116987897803, + "loss": 0.4228, + "step": 31137 + }, + { + "epoch": 1.7397267927479958, + "grad_norm": 0.39352819323539734, + "learning_rate": 0.00013116315553563424, + "loss": 0.3461, + "step": 31138 + }, + { + "epoch": 1.7397826633516775, + "grad_norm": 0.4749276340007782, + "learning_rate": 0.00013113514119229044, + "loss": 0.4948, + "step": 31139 + }, + { + "epoch": 1.7398385339553593, + "grad_norm": 0.38440626859664917, + "learning_rate": 0.00013110712684894665, + "loss": 0.3946, + "step": 31140 + }, + { + "epoch": 1.7398944045590412, + "grad_norm": 0.6924148201942444, + "learning_rate": 0.00013107911250560285, + "loss": 0.3772, + "step": 31141 + }, + { + "epoch": 1.7399502751627232, + "grad_norm": 0.3660707473754883, + "learning_rate": 0.00013105109816225909, + "loss": 0.3859, + "step": 31142 + }, + { + "epoch": 1.7400061457664049, + "grad_norm": 0.559795081615448, + "learning_rate": 0.0001310230838189153, + "loss": 0.4442, + "step": 31143 + }, + { + "epoch": 1.7400620163700868, + "grad_norm": 0.32152992486953735, + "learning_rate": 0.0001309950694755715, + "loss": 0.3572, + "step": 31144 + }, + { + "epoch": 1.7401178869737688, + "grad_norm": 0.5707135200500488, + "learning_rate": 0.0001309670551322277, + "loss": 0.4949, + "step": 31145 + }, + { + "epoch": 1.7401737575774505, + "grad_norm": 0.40879449248313904, + "learning_rate": 0.0001309390407888839, + "loss": 0.4025, + "step": 31146 + }, + { + "epoch": 1.7402296281811325, + "grad_norm": 0.5341518521308899, + "learning_rate": 0.00013091102644554011, + "loss": 0.4449, + "step": 31147 + }, + { + "epoch": 1.7402854987848144, + "grad_norm": 0.4640384316444397, + "learning_rate": 0.00013088301210219632, + "loss": 0.3798, + "step": 31148 + }, + { + "epoch": 1.7403413693884962, + "grad_norm": 0.6597774028778076, + "learning_rate": 0.00013085499775885253, + "loss": 0.433, + "step": 31149 + }, + { + "epoch": 1.740397239992178, + "grad_norm": 0.38673707842826843, + "learning_rate": 0.00013082698341550876, + "loss": 0.4002, + "step": 31150 + }, + { + "epoch": 1.74045311059586, + "grad_norm": 0.4313528835773468, + "learning_rate": 0.00013079896907216497, + "loss": 0.3398, + "step": 31151 + }, + { + "epoch": 1.7405089811995418, + "grad_norm": 0.4668337404727936, + "learning_rate": 0.00013077095472882117, + "loss": 0.3854, + "step": 31152 + }, + { + "epoch": 1.7405648518032237, + "grad_norm": 0.33477941155433655, + "learning_rate": 0.00013074294038547738, + "loss": 0.4047, + "step": 31153 + }, + { + "epoch": 1.7406207224069057, + "grad_norm": 0.4626413583755493, + "learning_rate": 0.00013071492604213358, + "loss": 0.432, + "step": 31154 + }, + { + "epoch": 1.7406765930105874, + "grad_norm": 1.9920982122421265, + "learning_rate": 0.0001306869116987898, + "loss": 0.3641, + "step": 31155 + }, + { + "epoch": 1.7407324636142694, + "grad_norm": 0.6873046159744263, + "learning_rate": 0.000130658897355446, + "loss": 0.4403, + "step": 31156 + }, + { + "epoch": 1.7407883342179513, + "grad_norm": 0.5061044692993164, + "learning_rate": 0.00013063088301210217, + "loss": 0.2663, + "step": 31157 + }, + { + "epoch": 1.740844204821633, + "grad_norm": 0.6709385514259338, + "learning_rate": 0.0001306028686687584, + "loss": 0.3396, + "step": 31158 + }, + { + "epoch": 1.740900075425315, + "grad_norm": 1.0350782871246338, + "learning_rate": 0.0001305748543254146, + "loss": 0.4781, + "step": 31159 + }, + { + "epoch": 1.740955946028997, + "grad_norm": 0.41251835227012634, + "learning_rate": 0.00013054683998207082, + "loss": 0.4103, + "step": 31160 + }, + { + "epoch": 1.7410118166326787, + "grad_norm": 0.3356921374797821, + "learning_rate": 0.00013051882563872702, + "loss": 0.3633, + "step": 31161 + }, + { + "epoch": 1.7410676872363606, + "grad_norm": 0.5109714865684509, + "learning_rate": 0.00013049081129538323, + "loss": 0.3775, + "step": 31162 + }, + { + "epoch": 1.7411235578400426, + "grad_norm": 0.4001379609107971, + "learning_rate": 0.00013046279695203943, + "loss": 0.3857, + "step": 31163 + }, + { + "epoch": 1.7411794284437243, + "grad_norm": 0.45383673906326294, + "learning_rate": 0.00013043478260869564, + "loss": 0.5371, + "step": 31164 + }, + { + "epoch": 1.7412352990474063, + "grad_norm": 0.428623229265213, + "learning_rate": 0.00013040676826535187, + "loss": 0.3603, + "step": 31165 + }, + { + "epoch": 1.7412911696510882, + "grad_norm": 0.4985663592815399, + "learning_rate": 0.00013037875392200808, + "loss": 0.4816, + "step": 31166 + }, + { + "epoch": 1.74134704025477, + "grad_norm": 0.416911780834198, + "learning_rate": 0.00013035073957866429, + "loss": 0.3568, + "step": 31167 + }, + { + "epoch": 1.7414029108584517, + "grad_norm": 1.7392046451568604, + "learning_rate": 0.0001303227252353205, + "loss": 0.3936, + "step": 31168 + }, + { + "epoch": 1.7414587814621338, + "grad_norm": 0.39077895879745483, + "learning_rate": 0.0001302947108919767, + "loss": 0.5163, + "step": 31169 + }, + { + "epoch": 1.7415146520658156, + "grad_norm": 0.5163328647613525, + "learning_rate": 0.0001302666965486329, + "loss": 0.4602, + "step": 31170 + }, + { + "epoch": 1.7415705226694973, + "grad_norm": 0.3572220504283905, + "learning_rate": 0.0001302386822052891, + "loss": 0.363, + "step": 31171 + }, + { + "epoch": 1.7416263932731795, + "grad_norm": 0.7358342409133911, + "learning_rate": 0.00013021066786194531, + "loss": 0.4787, + "step": 31172 + }, + { + "epoch": 1.7416822638768612, + "grad_norm": 0.5003299713134766, + "learning_rate": 0.00013018265351860155, + "loss": 0.3681, + "step": 31173 + }, + { + "epoch": 1.741738134480543, + "grad_norm": 0.2896220088005066, + "learning_rate": 0.00013015463917525775, + "loss": 0.3385, + "step": 31174 + }, + { + "epoch": 1.7417940050842249, + "grad_norm": 0.5096705555915833, + "learning_rate": 0.00013012662483191396, + "loss": 0.5511, + "step": 31175 + }, + { + "epoch": 1.7418498756879068, + "grad_norm": 0.35172808170318604, + "learning_rate": 0.00013009861048857016, + "loss": 0.3723, + "step": 31176 + }, + { + "epoch": 1.7419057462915886, + "grad_norm": 0.44913217425346375, + "learning_rate": 0.00013007059614522637, + "loss": 0.2835, + "step": 31177 + }, + { + "epoch": 1.7419616168952705, + "grad_norm": 0.3475596606731415, + "learning_rate": 0.00013004258180188255, + "loss": 0.4547, + "step": 31178 + }, + { + "epoch": 1.7420174874989525, + "grad_norm": 0.4682878255844116, + "learning_rate": 0.00013001456745853876, + "loss": 0.719, + "step": 31179 + }, + { + "epoch": 1.7420733581026342, + "grad_norm": 0.3653579652309418, + "learning_rate": 0.00012998655311519496, + "loss": 0.4178, + "step": 31180 + }, + { + "epoch": 1.7421292287063161, + "grad_norm": 1.8445196151733398, + "learning_rate": 0.0001299585387718512, + "loss": 0.3015, + "step": 31181 + }, + { + "epoch": 1.742185099309998, + "grad_norm": 2.2353734970092773, + "learning_rate": 0.0001299305244285074, + "loss": 0.467, + "step": 31182 + }, + { + "epoch": 1.7422409699136798, + "grad_norm": 0.42098742723464966, + "learning_rate": 0.0001299025100851636, + "loss": 0.4321, + "step": 31183 + }, + { + "epoch": 1.7422968405173618, + "grad_norm": 4.976806640625, + "learning_rate": 0.0001298744957418198, + "loss": 0.4104, + "step": 31184 + }, + { + "epoch": 1.7423527111210437, + "grad_norm": 0.7332518696784973, + "learning_rate": 0.00012984648139847602, + "loss": 0.5031, + "step": 31185 + }, + { + "epoch": 1.7424085817247255, + "grad_norm": 0.4772554934024811, + "learning_rate": 0.00012981846705513222, + "loss": 0.43, + "step": 31186 + }, + { + "epoch": 1.7424644523284074, + "grad_norm": 0.480916827917099, + "learning_rate": 0.00012979045271178843, + "loss": 0.4761, + "step": 31187 + }, + { + "epoch": 1.7425203229320894, + "grad_norm": 0.41030916571617126, + "learning_rate": 0.00012976243836844463, + "loss": 0.5726, + "step": 31188 + }, + { + "epoch": 1.742576193535771, + "grad_norm": 0.5316970944404602, + "learning_rate": 0.00012973442402510087, + "loss": 0.4499, + "step": 31189 + }, + { + "epoch": 1.742632064139453, + "grad_norm": 0.8016154170036316, + "learning_rate": 0.00012970640968175707, + "loss": 0.4797, + "step": 31190 + }, + { + "epoch": 1.742687934743135, + "grad_norm": 0.43317854404449463, + "learning_rate": 0.00012967839533841328, + "loss": 0.4003, + "step": 31191 + }, + { + "epoch": 1.7427438053468167, + "grad_norm": 0.5338134169578552, + "learning_rate": 0.00012965038099506948, + "loss": 0.4223, + "step": 31192 + }, + { + "epoch": 1.7427996759504987, + "grad_norm": 1.276560664176941, + "learning_rate": 0.0001296223666517257, + "loss": 0.4469, + "step": 31193 + }, + { + "epoch": 1.7428555465541806, + "grad_norm": 0.5013613700866699, + "learning_rate": 0.0001295943523083819, + "loss": 0.3739, + "step": 31194 + }, + { + "epoch": 1.7429114171578624, + "grad_norm": 0.4506305456161499, + "learning_rate": 0.0001295663379650381, + "loss": 0.4715, + "step": 31195 + }, + { + "epoch": 1.7429672877615443, + "grad_norm": 0.4603016972541809, + "learning_rate": 0.0001295383236216943, + "loss": 0.5012, + "step": 31196 + }, + { + "epoch": 1.7430231583652263, + "grad_norm": 0.7695583701133728, + "learning_rate": 0.00012951030927835054, + "loss": 0.4815, + "step": 31197 + }, + { + "epoch": 1.743079028968908, + "grad_norm": 0.47002530097961426, + "learning_rate": 0.00012948229493500675, + "loss": 0.4069, + "step": 31198 + }, + { + "epoch": 1.74313489957259, + "grad_norm": 0.3172306716442108, + "learning_rate": 0.00012945428059166293, + "loss": 0.3047, + "step": 31199 + }, + { + "epoch": 1.7431907701762719, + "grad_norm": 0.3726547062397003, + "learning_rate": 0.00012942626624831913, + "loss": 0.3846, + "step": 31200 + }, + { + "epoch": 1.7432466407799536, + "grad_norm": 0.42492127418518066, + "learning_rate": 0.00012939825190497534, + "loss": 0.3962, + "step": 31201 + }, + { + "epoch": 1.7433025113836353, + "grad_norm": 1.712377667427063, + "learning_rate": 0.00012937023756163154, + "loss": 0.4173, + "step": 31202 + }, + { + "epoch": 1.7433583819873175, + "grad_norm": 0.5007895231246948, + "learning_rate": 0.00012934222321828775, + "loss": 0.4465, + "step": 31203 + }, + { + "epoch": 1.7434142525909992, + "grad_norm": 0.401985228061676, + "learning_rate": 0.00012931420887494395, + "loss": 0.4597, + "step": 31204 + }, + { + "epoch": 1.743470123194681, + "grad_norm": 0.3622889220714569, + "learning_rate": 0.0001292861945316002, + "loss": 0.4303, + "step": 31205 + }, + { + "epoch": 1.7435259937983631, + "grad_norm": 4.857147216796875, + "learning_rate": 0.0001292581801882564, + "loss": 0.3754, + "step": 31206 + }, + { + "epoch": 1.7435818644020449, + "grad_norm": 0.5972156524658203, + "learning_rate": 0.0001292301658449126, + "loss": 0.4782, + "step": 31207 + }, + { + "epoch": 1.7436377350057266, + "grad_norm": 0.5768707990646362, + "learning_rate": 0.0001292021515015688, + "loss": 0.3498, + "step": 31208 + }, + { + "epoch": 1.7436936056094086, + "grad_norm": 0.47824519872665405, + "learning_rate": 0.000129174137158225, + "loss": 0.4343, + "step": 31209 + }, + { + "epoch": 1.7437494762130905, + "grad_norm": 0.40338096022605896, + "learning_rate": 0.00012914612281488122, + "loss": 0.3781, + "step": 31210 + }, + { + "epoch": 1.7438053468167722, + "grad_norm": 0.6340446472167969, + "learning_rate": 0.00012911810847153742, + "loss": 0.3878, + "step": 31211 + }, + { + "epoch": 1.7438612174204542, + "grad_norm": 0.6239457130432129, + "learning_rate": 0.00012909009412819363, + "loss": 0.477, + "step": 31212 + }, + { + "epoch": 1.7439170880241361, + "grad_norm": 0.4354584217071533, + "learning_rate": 0.00012906207978484986, + "loss": 0.4546, + "step": 31213 + }, + { + "epoch": 1.7439729586278179, + "grad_norm": 0.49141690135002136, + "learning_rate": 0.00012903406544150607, + "loss": 0.4197, + "step": 31214 + }, + { + "epoch": 1.7440288292314998, + "grad_norm": 0.4138517677783966, + "learning_rate": 0.00012900605109816227, + "loss": 0.3498, + "step": 31215 + }, + { + "epoch": 1.7440846998351818, + "grad_norm": 0.33398836851119995, + "learning_rate": 0.00012897803675481848, + "loss": 0.3381, + "step": 31216 + }, + { + "epoch": 1.7441405704388635, + "grad_norm": 1.0167628526687622, + "learning_rate": 0.00012895002241147468, + "loss": 0.535, + "step": 31217 + }, + { + "epoch": 1.7441964410425455, + "grad_norm": 0.6730229258537292, + "learning_rate": 0.0001289220080681309, + "loss": 0.4322, + "step": 31218 + }, + { + "epoch": 1.7442523116462274, + "grad_norm": 0.4470657408237457, + "learning_rate": 0.0001288939937247871, + "loss": 0.4488, + "step": 31219 + }, + { + "epoch": 1.7443081822499091, + "grad_norm": 0.9331328272819519, + "learning_rate": 0.00012886597938144327, + "loss": 0.4143, + "step": 31220 + }, + { + "epoch": 1.744364052853591, + "grad_norm": 3.547139883041382, + "learning_rate": 0.0001288379650380995, + "loss": 0.3864, + "step": 31221 + }, + { + "epoch": 1.744419923457273, + "grad_norm": 0.3587549030780792, + "learning_rate": 0.00012880995069475571, + "loss": 0.4238, + "step": 31222 + }, + { + "epoch": 1.7444757940609548, + "grad_norm": 0.38893625140190125, + "learning_rate": 0.00012878193635141192, + "loss": 0.3377, + "step": 31223 + }, + { + "epoch": 1.7445316646646367, + "grad_norm": 0.804758608341217, + "learning_rate": 0.00012875392200806813, + "loss": 0.6112, + "step": 31224 + }, + { + "epoch": 1.7445875352683187, + "grad_norm": 0.3939048647880554, + "learning_rate": 0.00012872590766472433, + "loss": 0.4435, + "step": 31225 + }, + { + "epoch": 1.7446434058720004, + "grad_norm": 0.6092199087142944, + "learning_rate": 0.00012869789332138054, + "loss": 0.4828, + "step": 31226 + }, + { + "epoch": 1.7446992764756823, + "grad_norm": 0.5609297752380371, + "learning_rate": 0.00012866987897803674, + "loss": 0.2441, + "step": 31227 + }, + { + "epoch": 1.7447551470793643, + "grad_norm": 3.579486131668091, + "learning_rate": 0.00012864186463469298, + "loss": 0.4365, + "step": 31228 + }, + { + "epoch": 1.744811017683046, + "grad_norm": 0.3343612253665924, + "learning_rate": 0.00012861385029134918, + "loss": 0.4495, + "step": 31229 + }, + { + "epoch": 1.744866888286728, + "grad_norm": 0.43801915645599365, + "learning_rate": 0.0001285858359480054, + "loss": 0.3966, + "step": 31230 + }, + { + "epoch": 1.74492275889041, + "grad_norm": 0.4452526569366455, + "learning_rate": 0.0001285578216046616, + "loss": 0.5286, + "step": 31231 + }, + { + "epoch": 1.7449786294940917, + "grad_norm": 0.5314300060272217, + "learning_rate": 0.0001285298072613178, + "loss": 0.3916, + "step": 31232 + }, + { + "epoch": 1.7450345000977736, + "grad_norm": 0.3830265402793884, + "learning_rate": 0.000128501792917974, + "loss": 0.4222, + "step": 31233 + }, + { + "epoch": 1.7450903707014556, + "grad_norm": 0.34788185358047485, + "learning_rate": 0.0001284737785746302, + "loss": 0.3469, + "step": 31234 + }, + { + "epoch": 1.7451462413051373, + "grad_norm": 0.39871472120285034, + "learning_rate": 0.00012844576423128642, + "loss": 0.4999, + "step": 31235 + }, + { + "epoch": 1.745202111908819, + "grad_norm": 0.6346787810325623, + "learning_rate": 0.00012841774988794265, + "loss": 0.3812, + "step": 31236 + }, + { + "epoch": 1.7452579825125012, + "grad_norm": 0.3409854769706726, + "learning_rate": 0.00012838973554459886, + "loss": 0.3174, + "step": 31237 + }, + { + "epoch": 1.745313853116183, + "grad_norm": 0.8499722480773926, + "learning_rate": 0.00012836172120125506, + "loss": 0.4872, + "step": 31238 + }, + { + "epoch": 1.7453697237198647, + "grad_norm": 0.5274030566215515, + "learning_rate": 0.00012833370685791127, + "loss": 0.3618, + "step": 31239 + }, + { + "epoch": 1.7454255943235468, + "grad_norm": 0.3221942186355591, + "learning_rate": 0.00012830569251456747, + "loss": 0.4139, + "step": 31240 + }, + { + "epoch": 1.7454814649272286, + "grad_norm": 0.41445526480674744, + "learning_rate": 0.00012827767817122368, + "loss": 0.4942, + "step": 31241 + }, + { + "epoch": 1.7455373355309103, + "grad_norm": 0.6372455954551697, + "learning_rate": 0.00012824966382787986, + "loss": 0.4873, + "step": 31242 + }, + { + "epoch": 1.7455932061345922, + "grad_norm": 0.4294585883617401, + "learning_rate": 0.00012822164948453606, + "loss": 0.4575, + "step": 31243 + }, + { + "epoch": 1.7456490767382742, + "grad_norm": 0.28811511397361755, + "learning_rate": 0.0001281936351411923, + "loss": 0.3553, + "step": 31244 + }, + { + "epoch": 1.745704947341956, + "grad_norm": 0.7790253162384033, + "learning_rate": 0.0001281656207978485, + "loss": 0.5062, + "step": 31245 + }, + { + "epoch": 1.7457608179456379, + "grad_norm": 1.2328262329101562, + "learning_rate": 0.0001281376064545047, + "loss": 0.4819, + "step": 31246 + }, + { + "epoch": 1.7458166885493198, + "grad_norm": 0.2959745526313782, + "learning_rate": 0.0001281095921111609, + "loss": 0.272, + "step": 31247 + }, + { + "epoch": 1.7458725591530015, + "grad_norm": 0.5296034216880798, + "learning_rate": 0.00012808157776781712, + "loss": 0.454, + "step": 31248 + }, + { + "epoch": 1.7459284297566835, + "grad_norm": 4.480146884918213, + "learning_rate": 0.00012805356342447332, + "loss": 0.3722, + "step": 31249 + }, + { + "epoch": 1.7459843003603654, + "grad_norm": 1.7296568155288696, + "learning_rate": 0.00012802554908112953, + "loss": 0.466, + "step": 31250 + }, + { + "epoch": 1.7460401709640472, + "grad_norm": 0.3940076231956482, + "learning_rate": 0.00012799753473778574, + "loss": 0.4443, + "step": 31251 + }, + { + "epoch": 1.7460960415677291, + "grad_norm": 0.4822855591773987, + "learning_rate": 0.00012796952039444197, + "loss": 0.5078, + "step": 31252 + }, + { + "epoch": 1.746151912171411, + "grad_norm": 0.6858016848564148, + "learning_rate": 0.00012794150605109818, + "loss": 0.3153, + "step": 31253 + }, + { + "epoch": 1.7462077827750928, + "grad_norm": 0.5276129245758057, + "learning_rate": 0.00012791349170775438, + "loss": 0.4957, + "step": 31254 + }, + { + "epoch": 1.7462636533787748, + "grad_norm": 0.42512813210487366, + "learning_rate": 0.0001278854773644106, + "loss": 0.4718, + "step": 31255 + }, + { + "epoch": 1.7463195239824567, + "grad_norm": 0.42061278223991394, + "learning_rate": 0.0001278574630210668, + "loss": 0.3981, + "step": 31256 + }, + { + "epoch": 1.7463753945861384, + "grad_norm": 0.43183907866477966, + "learning_rate": 0.000127829448677723, + "loss": 0.4815, + "step": 31257 + }, + { + "epoch": 1.7464312651898204, + "grad_norm": 5.569624423980713, + "learning_rate": 0.0001278014343343792, + "loss": 0.4503, + "step": 31258 + }, + { + "epoch": 1.7464871357935023, + "grad_norm": 0.37962019443511963, + "learning_rate": 0.0001277734199910354, + "loss": 0.328, + "step": 31259 + }, + { + "epoch": 1.746543006397184, + "grad_norm": 0.42047399282455444, + "learning_rate": 0.00012774540564769164, + "loss": 0.3974, + "step": 31260 + }, + { + "epoch": 1.746598877000866, + "grad_norm": 0.3568616211414337, + "learning_rate": 0.00012771739130434785, + "loss": 0.4631, + "step": 31261 + }, + { + "epoch": 1.746654747604548, + "grad_norm": 0.4689883291721344, + "learning_rate": 0.00012768937696100405, + "loss": 0.3801, + "step": 31262 + }, + { + "epoch": 1.7467106182082297, + "grad_norm": 1.1998344659805298, + "learning_rate": 0.00012766136261766023, + "loss": 0.4314, + "step": 31263 + }, + { + "epoch": 1.7467664888119117, + "grad_norm": 0.30534154176712036, + "learning_rate": 0.00012763334827431644, + "loss": 0.3934, + "step": 31264 + }, + { + "epoch": 1.7468223594155936, + "grad_norm": 0.45771950483322144, + "learning_rate": 0.00012760533393097264, + "loss": 0.3725, + "step": 31265 + }, + { + "epoch": 1.7468782300192753, + "grad_norm": 0.9560877084732056, + "learning_rate": 0.00012757731958762885, + "loss": 0.381, + "step": 31266 + }, + { + "epoch": 1.746934100622957, + "grad_norm": 0.3766952455043793, + "learning_rate": 0.00012754930524428506, + "loss": 0.3976, + "step": 31267 + }, + { + "epoch": 1.7469899712266392, + "grad_norm": 0.4804477095603943, + "learning_rate": 0.0001275212909009413, + "loss": 0.4384, + "step": 31268 + }, + { + "epoch": 1.747045841830321, + "grad_norm": 0.4133807122707367, + "learning_rate": 0.0001274932765575975, + "loss": 0.3283, + "step": 31269 + }, + { + "epoch": 1.7471017124340027, + "grad_norm": 0.5677363276481628, + "learning_rate": 0.0001274652622142537, + "loss": 0.4994, + "step": 31270 + }, + { + "epoch": 1.7471575830376849, + "grad_norm": 0.39508482813835144, + "learning_rate": 0.0001274372478709099, + "loss": 0.5456, + "step": 31271 + }, + { + "epoch": 1.7472134536413666, + "grad_norm": 0.4255269765853882, + "learning_rate": 0.0001274092335275661, + "loss": 0.3855, + "step": 31272 + }, + { + "epoch": 1.7472693242450483, + "grad_norm": 0.6297833323478699, + "learning_rate": 0.00012738121918422232, + "loss": 0.4763, + "step": 31273 + }, + { + "epoch": 1.7473251948487305, + "grad_norm": 0.46863245964050293, + "learning_rate": 0.00012735320484087852, + "loss": 0.4737, + "step": 31274 + }, + { + "epoch": 1.7473810654524122, + "grad_norm": 0.5092259049415588, + "learning_rate": 0.00012732519049753473, + "loss": 0.4482, + "step": 31275 + }, + { + "epoch": 1.747436936056094, + "grad_norm": 1.759561538696289, + "learning_rate": 0.00012729717615419096, + "loss": 0.3358, + "step": 31276 + }, + { + "epoch": 1.747492806659776, + "grad_norm": 0.382722407579422, + "learning_rate": 0.00012726916181084717, + "loss": 0.3712, + "step": 31277 + }, + { + "epoch": 1.7475486772634579, + "grad_norm": 1.4334303140640259, + "learning_rate": 0.00012724114746750337, + "loss": 0.4577, + "step": 31278 + }, + { + "epoch": 1.7476045478671396, + "grad_norm": 0.3748898506164551, + "learning_rate": 0.00012721313312415958, + "loss": 0.4243, + "step": 31279 + }, + { + "epoch": 1.7476604184708215, + "grad_norm": 3.637577772140503, + "learning_rate": 0.00012718511878081579, + "loss": 0.3624, + "step": 31280 + }, + { + "epoch": 1.7477162890745035, + "grad_norm": 1.0109738111495972, + "learning_rate": 0.000127157104437472, + "loss": 0.4326, + "step": 31281 + }, + { + "epoch": 1.7477721596781852, + "grad_norm": 0.5634437799453735, + "learning_rate": 0.0001271290900941282, + "loss": 0.4118, + "step": 31282 + }, + { + "epoch": 1.7478280302818672, + "grad_norm": 0.739297091960907, + "learning_rate": 0.0001271010757507844, + "loss": 0.4302, + "step": 31283 + }, + { + "epoch": 1.7478839008855491, + "grad_norm": 1.7515170574188232, + "learning_rate": 0.0001270730614074406, + "loss": 0.3924, + "step": 31284 + }, + { + "epoch": 1.7479397714892309, + "grad_norm": 0.550614058971405, + "learning_rate": 0.00012704504706409682, + "loss": 0.3865, + "step": 31285 + }, + { + "epoch": 1.7479956420929128, + "grad_norm": 0.3462202548980713, + "learning_rate": 0.00012701703272075302, + "loss": 0.4157, + "step": 31286 + }, + { + "epoch": 1.7480515126965948, + "grad_norm": 0.38661304116249084, + "learning_rate": 0.00012698901837740923, + "loss": 0.5093, + "step": 31287 + }, + { + "epoch": 1.7481073833002765, + "grad_norm": 0.5759018063545227, + "learning_rate": 0.00012696100403406543, + "loss": 0.4459, + "step": 31288 + }, + { + "epoch": 1.7481632539039584, + "grad_norm": 0.5075191259384155, + "learning_rate": 0.00012693298969072164, + "loss": 0.4638, + "step": 31289 + }, + { + "epoch": 1.7482191245076404, + "grad_norm": 0.4130185544490814, + "learning_rate": 0.00012690497534737784, + "loss": 0.4524, + "step": 31290 + }, + { + "epoch": 1.7482749951113221, + "grad_norm": 0.7464207410812378, + "learning_rate": 0.00012687696100403405, + "loss": 0.424, + "step": 31291 + }, + { + "epoch": 1.748330865715004, + "grad_norm": 0.5354061722755432, + "learning_rate": 0.00012684894666069028, + "loss": 0.3385, + "step": 31292 + }, + { + "epoch": 1.748386736318686, + "grad_norm": 0.4634702205657959, + "learning_rate": 0.0001268209323173465, + "loss": 0.3759, + "step": 31293 + }, + { + "epoch": 1.7484426069223677, + "grad_norm": 0.9561828374862671, + "learning_rate": 0.0001267929179740027, + "loss": 0.3648, + "step": 31294 + }, + { + "epoch": 1.7484984775260497, + "grad_norm": 1.186489462852478, + "learning_rate": 0.0001267649036306589, + "loss": 0.3963, + "step": 31295 + }, + { + "epoch": 1.7485543481297317, + "grad_norm": 0.48054152727127075, + "learning_rate": 0.0001267368892873151, + "loss": 0.4467, + "step": 31296 + }, + { + "epoch": 1.7486102187334134, + "grad_norm": 0.3965090811252594, + "learning_rate": 0.0001267088749439713, + "loss": 0.4851, + "step": 31297 + }, + { + "epoch": 1.7486660893370953, + "grad_norm": 0.3959169387817383, + "learning_rate": 0.00012668086060062752, + "loss": 0.3479, + "step": 31298 + }, + { + "epoch": 1.7487219599407773, + "grad_norm": 0.38237854838371277, + "learning_rate": 0.00012665284625728375, + "loss": 0.4442, + "step": 31299 + }, + { + "epoch": 1.748777830544459, + "grad_norm": 0.5586090683937073, + "learning_rate": 0.00012662483191393996, + "loss": 0.3838, + "step": 31300 + }, + { + "epoch": 1.7488337011481407, + "grad_norm": 0.3320568799972534, + "learning_rate": 0.00012659681757059616, + "loss": 0.4137, + "step": 31301 + }, + { + "epoch": 1.748889571751823, + "grad_norm": 0.39212700724601746, + "learning_rate": 0.00012656880322725237, + "loss": 0.3931, + "step": 31302 + }, + { + "epoch": 1.7489454423555046, + "grad_norm": 0.3936762511730194, + "learning_rate": 0.00012654078888390857, + "loss": 0.4519, + "step": 31303 + }, + { + "epoch": 1.7490013129591864, + "grad_norm": 5.579193592071533, + "learning_rate": 0.00012651277454056478, + "loss": 0.419, + "step": 31304 + }, + { + "epoch": 1.7490571835628685, + "grad_norm": 0.34686776995658875, + "learning_rate": 0.00012648476019722096, + "loss": 0.4549, + "step": 31305 + }, + { + "epoch": 1.7491130541665503, + "grad_norm": 0.5135602951049805, + "learning_rate": 0.00012645674585387716, + "loss": 0.4057, + "step": 31306 + }, + { + "epoch": 1.749168924770232, + "grad_norm": 0.31180867552757263, + "learning_rate": 0.0001264287315105334, + "loss": 0.3792, + "step": 31307 + }, + { + "epoch": 1.7492247953739142, + "grad_norm": 1.7146997451782227, + "learning_rate": 0.0001264007171671896, + "loss": 0.3616, + "step": 31308 + }, + { + "epoch": 1.749280665977596, + "grad_norm": 0.41948702931404114, + "learning_rate": 0.0001263727028238458, + "loss": 0.3683, + "step": 31309 + }, + { + "epoch": 1.7493365365812776, + "grad_norm": 1.0877147912979126, + "learning_rate": 0.00012634468848050202, + "loss": 0.465, + "step": 31310 + }, + { + "epoch": 1.7493924071849596, + "grad_norm": 1.100466251373291, + "learning_rate": 0.00012631667413715822, + "loss": 0.3786, + "step": 31311 + }, + { + "epoch": 1.7494482777886415, + "grad_norm": 0.35975411534309387, + "learning_rate": 0.00012628865979381443, + "loss": 0.3912, + "step": 31312 + }, + { + "epoch": 1.7495041483923233, + "grad_norm": 0.3273284435272217, + "learning_rate": 0.00012626064545047063, + "loss": 0.444, + "step": 31313 + }, + { + "epoch": 1.7495600189960052, + "grad_norm": 0.8426318764686584, + "learning_rate": 0.00012623263110712684, + "loss": 0.3747, + "step": 31314 + }, + { + "epoch": 1.7496158895996872, + "grad_norm": 0.41057780385017395, + "learning_rate": 0.00012620461676378307, + "loss": 0.4015, + "step": 31315 + }, + { + "epoch": 1.749671760203369, + "grad_norm": 0.5653418302536011, + "learning_rate": 0.00012617660242043928, + "loss": 0.4331, + "step": 31316 + }, + { + "epoch": 1.7497276308070508, + "grad_norm": 0.9671621918678284, + "learning_rate": 0.00012614858807709548, + "loss": 0.4711, + "step": 31317 + }, + { + "epoch": 1.7497835014107328, + "grad_norm": 2.0721724033355713, + "learning_rate": 0.0001261205737337517, + "loss": 0.4632, + "step": 31318 + }, + { + "epoch": 1.7498393720144145, + "grad_norm": 0.5119854211807251, + "learning_rate": 0.0001260925593904079, + "loss": 0.4408, + "step": 31319 + }, + { + "epoch": 1.7498952426180965, + "grad_norm": 0.605751097202301, + "learning_rate": 0.0001260645450470641, + "loss": 0.5169, + "step": 31320 + }, + { + "epoch": 1.7499511132217784, + "grad_norm": 0.6162915825843811, + "learning_rate": 0.0001260365307037203, + "loss": 0.3937, + "step": 31321 + }, + { + "epoch": 1.7500069838254602, + "grad_norm": 0.34666863083839417, + "learning_rate": 0.0001260085163603765, + "loss": 0.3856, + "step": 31322 + }, + { + "epoch": 1.7500628544291421, + "grad_norm": 0.6243050694465637, + "learning_rate": 0.00012598050201703274, + "loss": 0.5132, + "step": 31323 + }, + { + "epoch": 1.750118725032824, + "grad_norm": 0.334626704454422, + "learning_rate": 0.00012595248767368895, + "loss": 0.4654, + "step": 31324 + }, + { + "epoch": 1.7501745956365058, + "grad_norm": 0.5638274550437927, + "learning_rate": 0.00012592447333034516, + "loss": 0.3944, + "step": 31325 + }, + { + "epoch": 1.7502304662401877, + "grad_norm": 0.9821891784667969, + "learning_rate": 0.00012589645898700136, + "loss": 0.7847, + "step": 31326 + }, + { + "epoch": 1.7502863368438697, + "grad_norm": 0.6849529147148132, + "learning_rate": 0.00012586844464365754, + "loss": 0.3641, + "step": 31327 + }, + { + "epoch": 1.7503422074475514, + "grad_norm": 0.42225927114486694, + "learning_rate": 0.00012584043030031375, + "loss": 0.4054, + "step": 31328 + }, + { + "epoch": 1.7503980780512334, + "grad_norm": 1.19082772731781, + "learning_rate": 0.00012581241595696995, + "loss": 0.5241, + "step": 31329 + }, + { + "epoch": 1.7504539486549153, + "grad_norm": 0.8396801352500916, + "learning_rate": 0.00012578440161362616, + "loss": 0.4462, + "step": 31330 + }, + { + "epoch": 1.750509819258597, + "grad_norm": 0.8305972814559937, + "learning_rate": 0.0001257563872702824, + "loss": 0.4477, + "step": 31331 + }, + { + "epoch": 1.750565689862279, + "grad_norm": 0.7337629199028015, + "learning_rate": 0.0001257283729269386, + "loss": 0.3846, + "step": 31332 + }, + { + "epoch": 1.750621560465961, + "grad_norm": 0.6431648135185242, + "learning_rate": 0.0001257003585835948, + "loss": 0.5584, + "step": 31333 + }, + { + "epoch": 1.7506774310696427, + "grad_norm": 1.1626883745193481, + "learning_rate": 0.000125672344240251, + "loss": 0.4842, + "step": 31334 + }, + { + "epoch": 1.7507333016733244, + "grad_norm": 0.40064769983291626, + "learning_rate": 0.00012564432989690721, + "loss": 0.3838, + "step": 31335 + }, + { + "epoch": 1.7507891722770066, + "grad_norm": 0.46472862362861633, + "learning_rate": 0.00012561631555356342, + "loss": 0.3898, + "step": 31336 + }, + { + "epoch": 1.7508450428806883, + "grad_norm": 0.7737009525299072, + "learning_rate": 0.00012558830121021963, + "loss": 0.3402, + "step": 31337 + }, + { + "epoch": 1.75090091348437, + "grad_norm": 1.0172200202941895, + "learning_rate": 0.00012556028686687583, + "loss": 0.3891, + "step": 31338 + }, + { + "epoch": 1.7509567840880522, + "grad_norm": 0.6011019945144653, + "learning_rate": 0.00012553227252353207, + "loss": 0.3023, + "step": 31339 + }, + { + "epoch": 1.751012654691734, + "grad_norm": 0.5666103363037109, + "learning_rate": 0.00012550425818018827, + "loss": 0.4106, + "step": 31340 + }, + { + "epoch": 1.7510685252954157, + "grad_norm": 0.5907880663871765, + "learning_rate": 0.00012547624383684448, + "loss": 0.5942, + "step": 31341 + }, + { + "epoch": 1.7511243958990979, + "grad_norm": 0.635559618473053, + "learning_rate": 0.00012544822949350068, + "loss": 0.455, + "step": 31342 + }, + { + "epoch": 1.7511802665027796, + "grad_norm": 0.41541096568107605, + "learning_rate": 0.0001254202151501569, + "loss": 0.3034, + "step": 31343 + }, + { + "epoch": 1.7512361371064613, + "grad_norm": 0.7629638314247131, + "learning_rate": 0.0001253922008068131, + "loss": 0.368, + "step": 31344 + }, + { + "epoch": 1.7512920077101433, + "grad_norm": 0.41068896651268005, + "learning_rate": 0.0001253641864634693, + "loss": 0.4372, + "step": 31345 + }, + { + "epoch": 1.7513478783138252, + "grad_norm": 2.673581600189209, + "learning_rate": 0.0001253361721201255, + "loss": 0.4261, + "step": 31346 + }, + { + "epoch": 1.751403748917507, + "grad_norm": 0.5006468892097473, + "learning_rate": 0.00012530815777678174, + "loss": 0.4835, + "step": 31347 + }, + { + "epoch": 1.751459619521189, + "grad_norm": 0.38321611285209656, + "learning_rate": 0.00012528014343343792, + "loss": 0.4219, + "step": 31348 + }, + { + "epoch": 1.7515154901248708, + "grad_norm": 1.8104742765426636, + "learning_rate": 0.00012525212909009412, + "loss": 0.4048, + "step": 31349 + }, + { + "epoch": 1.7515713607285526, + "grad_norm": 0.6693049073219299, + "learning_rate": 0.00012522411474675033, + "loss": 0.3761, + "step": 31350 + }, + { + "epoch": 1.7516272313322345, + "grad_norm": 0.39136919379234314, + "learning_rate": 0.00012519610040340653, + "loss": 0.4832, + "step": 31351 + }, + { + "epoch": 1.7516831019359165, + "grad_norm": 0.35790103673934937, + "learning_rate": 0.00012516808606006274, + "loss": 0.4579, + "step": 31352 + }, + { + "epoch": 1.7517389725395982, + "grad_norm": 0.38915112614631653, + "learning_rate": 0.00012514007171671895, + "loss": 0.3737, + "step": 31353 + }, + { + "epoch": 1.7517948431432802, + "grad_norm": 2.045891761779785, + "learning_rate": 0.00012511205737337515, + "loss": 0.4762, + "step": 31354 + }, + { + "epoch": 1.751850713746962, + "grad_norm": 2.661770820617676, + "learning_rate": 0.00012508404303003139, + "loss": 0.3363, + "step": 31355 + }, + { + "epoch": 1.7519065843506438, + "grad_norm": 0.3763403296470642, + "learning_rate": 0.0001250560286866876, + "loss": 0.4421, + "step": 31356 + }, + { + "epoch": 1.7519624549543258, + "grad_norm": 0.3612740635871887, + "learning_rate": 0.0001250280143433438, + "loss": 0.3155, + "step": 31357 + }, + { + "epoch": 1.7520183255580077, + "grad_norm": 0.4920313358306885, + "learning_rate": 0.000125, + "loss": 0.4819, + "step": 31358 + }, + { + "epoch": 1.7520741961616895, + "grad_norm": 0.39983078837394714, + "learning_rate": 0.0001249719856566562, + "loss": 0.4714, + "step": 31359 + }, + { + "epoch": 1.7521300667653714, + "grad_norm": 0.6096280217170715, + "learning_rate": 0.00012494397131331241, + "loss": 0.5215, + "step": 31360 + }, + { + "epoch": 1.7521859373690534, + "grad_norm": 0.3092813789844513, + "learning_rate": 0.00012491595696996862, + "loss": 0.3403, + "step": 31361 + }, + { + "epoch": 1.752241807972735, + "grad_norm": 0.386059045791626, + "learning_rate": 0.00012488794262662485, + "loss": 0.4253, + "step": 31362 + }, + { + "epoch": 1.752297678576417, + "grad_norm": 0.5418751239776611, + "learning_rate": 0.00012485992828328106, + "loss": 0.3223, + "step": 31363 + }, + { + "epoch": 1.752353549180099, + "grad_norm": 0.37375012040138245, + "learning_rate": 0.00012483191393993724, + "loss": 0.3879, + "step": 31364 + }, + { + "epoch": 1.7524094197837807, + "grad_norm": 0.3591737449169159, + "learning_rate": 0.00012480389959659344, + "loss": 0.4086, + "step": 31365 + }, + { + "epoch": 1.7524652903874627, + "grad_norm": 1.1689034700393677, + "learning_rate": 0.00012477588525324968, + "loss": 0.3739, + "step": 31366 + }, + { + "epoch": 1.7525211609911446, + "grad_norm": 0.41008126735687256, + "learning_rate": 0.00012474787090990588, + "loss": 0.3981, + "step": 31367 + }, + { + "epoch": 1.7525770315948264, + "grad_norm": 0.369357705116272, + "learning_rate": 0.0001247198565665621, + "loss": 0.4669, + "step": 31368 + }, + { + "epoch": 1.752632902198508, + "grad_norm": 0.46661531925201416, + "learning_rate": 0.0001246918422232183, + "loss": 0.4162, + "step": 31369 + }, + { + "epoch": 1.7526887728021903, + "grad_norm": 0.47589367628097534, + "learning_rate": 0.0001246638278798745, + "loss": 0.3957, + "step": 31370 + }, + { + "epoch": 1.752744643405872, + "grad_norm": 0.3204948902130127, + "learning_rate": 0.0001246358135365307, + "loss": 0.3428, + "step": 31371 + }, + { + "epoch": 1.7528005140095537, + "grad_norm": 0.757867157459259, + "learning_rate": 0.0001246077991931869, + "loss": 0.4527, + "step": 31372 + }, + { + "epoch": 1.752856384613236, + "grad_norm": 0.4026520252227783, + "learning_rate": 0.00012457978484984312, + "loss": 0.4547, + "step": 31373 + }, + { + "epoch": 1.7529122552169176, + "grad_norm": 0.4246653616428375, + "learning_rate": 0.00012455177050649932, + "loss": 0.3569, + "step": 31374 + }, + { + "epoch": 1.7529681258205994, + "grad_norm": 1.1703836917877197, + "learning_rate": 0.00012452375616315553, + "loss": 0.5347, + "step": 31375 + }, + { + "epoch": 1.7530239964242815, + "grad_norm": 0.452390193939209, + "learning_rate": 0.00012449574181981173, + "loss": 0.2944, + "step": 31376 + }, + { + "epoch": 1.7530798670279633, + "grad_norm": 0.5550970435142517, + "learning_rate": 0.00012446772747646794, + "loss": 0.479, + "step": 31377 + }, + { + "epoch": 1.753135737631645, + "grad_norm": 0.4839167296886444, + "learning_rate": 0.00012443971313312417, + "loss": 0.4123, + "step": 31378 + }, + { + "epoch": 1.753191608235327, + "grad_norm": 0.4861021339893341, + "learning_rate": 0.00012441169878978038, + "loss": 0.5258, + "step": 31379 + }, + { + "epoch": 1.7532474788390089, + "grad_norm": 0.4091218411922455, + "learning_rate": 0.00012438368444643658, + "loss": 0.4313, + "step": 31380 + }, + { + "epoch": 1.7533033494426906, + "grad_norm": 0.35173991322517395, + "learning_rate": 0.0001243556701030928, + "loss": 0.3991, + "step": 31381 + }, + { + "epoch": 1.7533592200463726, + "grad_norm": 0.5447974801063538, + "learning_rate": 0.000124327655759749, + "loss": 0.4003, + "step": 31382 + }, + { + "epoch": 1.7534150906500545, + "grad_norm": 2.7502005100250244, + "learning_rate": 0.0001242996414164052, + "loss": 0.4888, + "step": 31383 + }, + { + "epoch": 1.7534709612537362, + "grad_norm": 0.39555394649505615, + "learning_rate": 0.0001242716270730614, + "loss": 0.4603, + "step": 31384 + }, + { + "epoch": 1.7535268318574182, + "grad_norm": 0.38376179337501526, + "learning_rate": 0.00012424361272971761, + "loss": 0.3636, + "step": 31385 + }, + { + "epoch": 1.7535827024611002, + "grad_norm": 0.5939490795135498, + "learning_rate": 0.00012421559838637382, + "loss": 0.4677, + "step": 31386 + }, + { + "epoch": 1.7536385730647819, + "grad_norm": 0.3833315968513489, + "learning_rate": 0.00012418758404303003, + "loss": 0.3877, + "step": 31387 + }, + { + "epoch": 1.7536944436684638, + "grad_norm": 0.4035737216472626, + "learning_rate": 0.00012415956969968623, + "loss": 0.4972, + "step": 31388 + }, + { + "epoch": 1.7537503142721458, + "grad_norm": 0.4411192238330841, + "learning_rate": 0.00012413155535634244, + "loss": 0.3799, + "step": 31389 + }, + { + "epoch": 1.7538061848758275, + "grad_norm": 0.42506977915763855, + "learning_rate": 0.00012410354101299867, + "loss": 0.4102, + "step": 31390 + }, + { + "epoch": 1.7538620554795095, + "grad_norm": 0.49266090989112854, + "learning_rate": 0.00012407552666965488, + "loss": 0.4844, + "step": 31391 + }, + { + "epoch": 1.7539179260831914, + "grad_norm": 0.3642345070838928, + "learning_rate": 0.00012404751232631108, + "loss": 0.3456, + "step": 31392 + }, + { + "epoch": 1.7539737966868731, + "grad_norm": 3.7962639331817627, + "learning_rate": 0.0001240194979829673, + "loss": 0.6801, + "step": 31393 + }, + { + "epoch": 1.754029667290555, + "grad_norm": 0.39019107818603516, + "learning_rate": 0.0001239914836396235, + "loss": 0.3679, + "step": 31394 + }, + { + "epoch": 1.754085537894237, + "grad_norm": 0.545020341873169, + "learning_rate": 0.0001239634692962797, + "loss": 0.52, + "step": 31395 + }, + { + "epoch": 1.7541414084979188, + "grad_norm": 0.4172825515270233, + "learning_rate": 0.0001239354549529359, + "loss": 0.3597, + "step": 31396 + }, + { + "epoch": 1.7541972791016007, + "grad_norm": 0.33268558979034424, + "learning_rate": 0.0001239074406095921, + "loss": 0.2988, + "step": 31397 + }, + { + "epoch": 1.7542531497052827, + "grad_norm": 0.40797334909439087, + "learning_rate": 0.00012387942626624832, + "loss": 0.3256, + "step": 31398 + }, + { + "epoch": 1.7543090203089644, + "grad_norm": 0.40263238549232483, + "learning_rate": 0.00012385141192290452, + "loss": 0.3692, + "step": 31399 + }, + { + "epoch": 1.7543648909126464, + "grad_norm": 0.7185498476028442, + "learning_rate": 0.00012382339757956073, + "loss": 0.3596, + "step": 31400 + }, + { + "epoch": 1.7544207615163283, + "grad_norm": 2.1330411434173584, + "learning_rate": 0.00012379538323621693, + "loss": 0.4238, + "step": 31401 + }, + { + "epoch": 1.75447663212001, + "grad_norm": 0.5110379457473755, + "learning_rate": 0.00012376736889287317, + "loss": 0.3963, + "step": 31402 + }, + { + "epoch": 1.7545325027236918, + "grad_norm": 0.4585443437099457, + "learning_rate": 0.00012373935454952937, + "loss": 0.2896, + "step": 31403 + }, + { + "epoch": 1.754588373327374, + "grad_norm": 0.4048948287963867, + "learning_rate": 0.00012371134020618558, + "loss": 0.3104, + "step": 31404 + }, + { + "epoch": 1.7546442439310557, + "grad_norm": 0.5663459300994873, + "learning_rate": 0.00012368332586284178, + "loss": 0.4883, + "step": 31405 + }, + { + "epoch": 1.7547001145347374, + "grad_norm": 0.44051504135131836, + "learning_rate": 0.000123655311519498, + "loss": 0.324, + "step": 31406 + }, + { + "epoch": 1.7547559851384196, + "grad_norm": 0.44249066710472107, + "learning_rate": 0.0001236272971761542, + "loss": 0.4256, + "step": 31407 + }, + { + "epoch": 1.7548118557421013, + "grad_norm": 0.7662802338600159, + "learning_rate": 0.0001235992828328104, + "loss": 0.3889, + "step": 31408 + }, + { + "epoch": 1.754867726345783, + "grad_norm": 0.273681104183197, + "learning_rate": 0.0001235712684894666, + "loss": 0.3548, + "step": 31409 + }, + { + "epoch": 1.7549235969494652, + "grad_norm": 0.375072717666626, + "learning_rate": 0.0001235432541461228, + "loss": 0.3824, + "step": 31410 + }, + { + "epoch": 1.754979467553147, + "grad_norm": 0.6171584725379944, + "learning_rate": 0.00012351523980277902, + "loss": 0.5619, + "step": 31411 + }, + { + "epoch": 1.7550353381568287, + "grad_norm": 0.6331568360328674, + "learning_rate": 0.00012348722545943523, + "loss": 0.4701, + "step": 31412 + }, + { + "epoch": 1.7550912087605106, + "grad_norm": 0.41955673694610596, + "learning_rate": 0.00012345921111609143, + "loss": 0.4592, + "step": 31413 + }, + { + "epoch": 1.7551470793641926, + "grad_norm": 1.1965564489364624, + "learning_rate": 0.00012343119677274766, + "loss": 0.3563, + "step": 31414 + }, + { + "epoch": 1.7552029499678743, + "grad_norm": 0.6681771278381348, + "learning_rate": 0.00012340318242940387, + "loss": 0.4841, + "step": 31415 + }, + { + "epoch": 1.7552588205715562, + "grad_norm": 0.6761191487312317, + "learning_rate": 0.00012337516808606008, + "loss": 0.5051, + "step": 31416 + }, + { + "epoch": 1.7553146911752382, + "grad_norm": 0.53896164894104, + "learning_rate": 0.00012334715374271625, + "loss": 0.4135, + "step": 31417 + }, + { + "epoch": 1.75537056177892, + "grad_norm": 0.3669707477092743, + "learning_rate": 0.0001233191393993725, + "loss": 0.3907, + "step": 31418 + }, + { + "epoch": 1.7554264323826019, + "grad_norm": 1.081284523010254, + "learning_rate": 0.0001232911250560287, + "loss": 0.4472, + "step": 31419 + }, + { + "epoch": 1.7554823029862838, + "grad_norm": 0.3810840845108032, + "learning_rate": 0.0001232631107126849, + "loss": 0.3477, + "step": 31420 + }, + { + "epoch": 1.7555381735899656, + "grad_norm": 0.4341717064380646, + "learning_rate": 0.0001232350963693411, + "loss": 0.4146, + "step": 31421 + }, + { + "epoch": 1.7555940441936475, + "grad_norm": 0.32478398084640503, + "learning_rate": 0.0001232070820259973, + "loss": 0.3788, + "step": 31422 + }, + { + "epoch": 1.7556499147973295, + "grad_norm": 0.4222337603569031, + "learning_rate": 0.00012317906768265352, + "loss": 0.5618, + "step": 31423 + }, + { + "epoch": 1.7557057854010112, + "grad_norm": 0.4498474895954132, + "learning_rate": 0.00012315105333930972, + "loss": 0.4607, + "step": 31424 + }, + { + "epoch": 1.7557616560046931, + "grad_norm": 0.3477969169616699, + "learning_rate": 0.00012312303899596595, + "loss": 0.3983, + "step": 31425 + }, + { + "epoch": 1.755817526608375, + "grad_norm": 0.41522419452667236, + "learning_rate": 0.00012309502465262216, + "loss": 0.4523, + "step": 31426 + }, + { + "epoch": 1.7558733972120568, + "grad_norm": 0.8076208233833313, + "learning_rate": 0.00012306701030927837, + "loss": 0.4147, + "step": 31427 + }, + { + "epoch": 1.7559292678157388, + "grad_norm": 0.6446994543075562, + "learning_rate": 0.00012303899596593455, + "loss": 0.4362, + "step": 31428 + }, + { + "epoch": 1.7559851384194207, + "grad_norm": 0.5339528322219849, + "learning_rate": 0.00012301098162259078, + "loss": 0.5658, + "step": 31429 + }, + { + "epoch": 1.7560410090231025, + "grad_norm": 0.8069213628768921, + "learning_rate": 0.00012298296727924698, + "loss": 0.4351, + "step": 31430 + }, + { + "epoch": 1.7560968796267844, + "grad_norm": 0.3333665132522583, + "learning_rate": 0.0001229549529359032, + "loss": 0.3797, + "step": 31431 + }, + { + "epoch": 1.7561527502304664, + "grad_norm": 0.39262381196022034, + "learning_rate": 0.0001229269385925594, + "loss": 0.3857, + "step": 31432 + }, + { + "epoch": 1.756208620834148, + "grad_norm": 0.349821537733078, + "learning_rate": 0.0001228989242492156, + "loss": 0.3414, + "step": 31433 + }, + { + "epoch": 1.75626449143783, + "grad_norm": 0.33532819151878357, + "learning_rate": 0.0001228709099058718, + "loss": 0.3209, + "step": 31434 + }, + { + "epoch": 1.756320362041512, + "grad_norm": 0.48810330033302307, + "learning_rate": 0.000122842895562528, + "loss": 0.4197, + "step": 31435 + }, + { + "epoch": 1.7563762326451937, + "grad_norm": 0.4800662398338318, + "learning_rate": 0.00012281488121918422, + "loss": 0.3884, + "step": 31436 + }, + { + "epoch": 1.7564321032488754, + "grad_norm": 3.3232462406158447, + "learning_rate": 0.00012278686687584045, + "loss": 0.3835, + "step": 31437 + }, + { + "epoch": 1.7564879738525576, + "grad_norm": 0.4626041054725647, + "learning_rate": 0.00012275885253249663, + "loss": 0.4742, + "step": 31438 + }, + { + "epoch": 1.7565438444562393, + "grad_norm": 0.6672350168228149, + "learning_rate": 0.00012273083818915284, + "loss": 0.4525, + "step": 31439 + }, + { + "epoch": 1.756599715059921, + "grad_norm": 0.3713214099407196, + "learning_rate": 0.00012270282384580904, + "loss": 0.4358, + "step": 31440 + }, + { + "epoch": 1.7566555856636032, + "grad_norm": 0.3164076805114746, + "learning_rate": 0.00012267480950246528, + "loss": 0.3366, + "step": 31441 + }, + { + "epoch": 1.756711456267285, + "grad_norm": 0.4663833975791931, + "learning_rate": 0.00012264679515912148, + "loss": 0.4108, + "step": 31442 + }, + { + "epoch": 1.7567673268709667, + "grad_norm": 4.052028179168701, + "learning_rate": 0.0001226187808157777, + "loss": 0.4034, + "step": 31443 + }, + { + "epoch": 1.7568231974746489, + "grad_norm": 0.5298634171485901, + "learning_rate": 0.0001225907664724339, + "loss": 0.3711, + "step": 31444 + }, + { + "epoch": 1.7568790680783306, + "grad_norm": 0.382443904876709, + "learning_rate": 0.0001225627521290901, + "loss": 0.4295, + "step": 31445 + }, + { + "epoch": 1.7569349386820123, + "grad_norm": 0.3142300546169281, + "learning_rate": 0.0001225347377857463, + "loss": 0.3506, + "step": 31446 + }, + { + "epoch": 1.7569908092856943, + "grad_norm": 0.6227512955665588, + "learning_rate": 0.0001225067234424025, + "loss": 0.462, + "step": 31447 + }, + { + "epoch": 1.7570466798893762, + "grad_norm": 0.5238151550292969, + "learning_rate": 0.00012247870909905872, + "loss": 0.4016, + "step": 31448 + }, + { + "epoch": 1.757102550493058, + "grad_norm": 0.5064757466316223, + "learning_rate": 0.00012245069475571492, + "loss": 0.3992, + "step": 31449 + }, + { + "epoch": 1.75715842109674, + "grad_norm": 3.603010654449463, + "learning_rate": 0.00012242268041237113, + "loss": 0.2829, + "step": 31450 + }, + { + "epoch": 1.7572142917004219, + "grad_norm": 0.3939096927642822, + "learning_rate": 0.00012239466606902733, + "loss": 0.3872, + "step": 31451 + }, + { + "epoch": 1.7572701623041036, + "grad_norm": 0.44635218381881714, + "learning_rate": 0.00012236665172568354, + "loss": 0.4002, + "step": 31452 + }, + { + "epoch": 1.7573260329077856, + "grad_norm": 1.8847746849060059, + "learning_rate": 0.00012233863738233977, + "loss": 0.462, + "step": 31453 + }, + { + "epoch": 1.7573819035114675, + "grad_norm": 0.3544970750808716, + "learning_rate": 0.00012231062303899598, + "loss": 0.3411, + "step": 31454 + }, + { + "epoch": 1.7574377741151492, + "grad_norm": 0.38795173168182373, + "learning_rate": 0.00012228260869565218, + "loss": 0.4144, + "step": 31455 + }, + { + "epoch": 1.7574936447188312, + "grad_norm": 0.48959407210350037, + "learning_rate": 0.0001222545943523084, + "loss": 0.4272, + "step": 31456 + }, + { + "epoch": 1.7575495153225131, + "grad_norm": 1.325864315032959, + "learning_rate": 0.0001222265800089646, + "loss": 0.4481, + "step": 31457 + }, + { + "epoch": 1.7576053859261949, + "grad_norm": 0.3535580337047577, + "learning_rate": 0.0001221985656656208, + "loss": 0.3807, + "step": 31458 + }, + { + "epoch": 1.7576612565298768, + "grad_norm": 7.016220569610596, + "learning_rate": 0.000122170551322277, + "loss": 0.4554, + "step": 31459 + }, + { + "epoch": 1.7577171271335588, + "grad_norm": 0.43126529455184937, + "learning_rate": 0.0001221425369789332, + "loss": 0.3326, + "step": 31460 + }, + { + "epoch": 1.7577729977372405, + "grad_norm": 0.39147746562957764, + "learning_rate": 0.00012211452263558942, + "loss": 0.4209, + "step": 31461 + }, + { + "epoch": 1.7578288683409224, + "grad_norm": 1.1565487384796143, + "learning_rate": 0.00012208650829224562, + "loss": 0.3676, + "step": 31462 + }, + { + "epoch": 1.7578847389446044, + "grad_norm": 0.44772419333457947, + "learning_rate": 0.00012205849394890184, + "loss": 0.4974, + "step": 31463 + }, + { + "epoch": 1.7579406095482861, + "grad_norm": 0.39101263880729675, + "learning_rate": 0.00012203047960555805, + "loss": 0.3132, + "step": 31464 + }, + { + "epoch": 1.757996480151968, + "grad_norm": 0.4585406482219696, + "learning_rate": 0.00012200246526221426, + "loss": 0.4795, + "step": 31465 + }, + { + "epoch": 1.75805235075565, + "grad_norm": 0.38598203659057617, + "learning_rate": 0.00012197445091887046, + "loss": 0.3188, + "step": 31466 + }, + { + "epoch": 1.7581082213593318, + "grad_norm": 2.107585906982422, + "learning_rate": 0.00012194643657552668, + "loss": 0.4935, + "step": 31467 + }, + { + "epoch": 1.7581640919630137, + "grad_norm": 2.095492124557495, + "learning_rate": 0.00012191842223218289, + "loss": 0.5158, + "step": 31468 + }, + { + "epoch": 1.7582199625666957, + "grad_norm": 0.35979893803596497, + "learning_rate": 0.00012189040788883909, + "loss": 0.373, + "step": 31469 + }, + { + "epoch": 1.7582758331703774, + "grad_norm": 0.5900230407714844, + "learning_rate": 0.00012186239354549528, + "loss": 0.3818, + "step": 31470 + }, + { + "epoch": 1.7583317037740591, + "grad_norm": 0.36403343081474304, + "learning_rate": 0.0001218343792021515, + "loss": 0.4204, + "step": 31471 + }, + { + "epoch": 1.7583875743777413, + "grad_norm": 0.3551936149597168, + "learning_rate": 0.00012180636485880771, + "loss": 0.3908, + "step": 31472 + }, + { + "epoch": 1.758443444981423, + "grad_norm": 0.3379869759082794, + "learning_rate": 0.00012177835051546392, + "loss": 0.3817, + "step": 31473 + }, + { + "epoch": 1.7584993155851047, + "grad_norm": 0.3977554738521576, + "learning_rate": 0.00012175033617212012, + "loss": 0.2962, + "step": 31474 + }, + { + "epoch": 1.758555186188787, + "grad_norm": 0.28804969787597656, + "learning_rate": 0.00012172232182877634, + "loss": 0.3497, + "step": 31475 + }, + { + "epoch": 1.7586110567924687, + "grad_norm": 0.4650086462497711, + "learning_rate": 0.00012169430748543255, + "loss": 0.3857, + "step": 31476 + }, + { + "epoch": 1.7586669273961504, + "grad_norm": 0.444038987159729, + "learning_rate": 0.00012166629314208875, + "loss": 0.3428, + "step": 31477 + }, + { + "epoch": 1.7587227979998326, + "grad_norm": 0.9166068434715271, + "learning_rate": 0.00012163827879874496, + "loss": 0.3909, + "step": 31478 + }, + { + "epoch": 1.7587786686035143, + "grad_norm": 0.48273229598999023, + "learning_rate": 0.00012161026445540118, + "loss": 0.356, + "step": 31479 + }, + { + "epoch": 1.758834539207196, + "grad_norm": 2.6992063522338867, + "learning_rate": 0.00012158225011205738, + "loss": 0.5582, + "step": 31480 + }, + { + "epoch": 1.758890409810878, + "grad_norm": 0.3630242347717285, + "learning_rate": 0.00012155423576871358, + "loss": 0.377, + "step": 31481 + }, + { + "epoch": 1.75894628041456, + "grad_norm": 0.9979323148727417, + "learning_rate": 0.00012152622142536978, + "loss": 0.5908, + "step": 31482 + }, + { + "epoch": 1.7590021510182416, + "grad_norm": 0.7352799773216248, + "learning_rate": 0.000121498207082026, + "loss": 0.4718, + "step": 31483 + }, + { + "epoch": 1.7590580216219236, + "grad_norm": 0.5147838592529297, + "learning_rate": 0.0001214701927386822, + "loss": 0.4091, + "step": 31484 + }, + { + "epoch": 1.7591138922256055, + "grad_norm": 0.4023350775241852, + "learning_rate": 0.00012144217839533841, + "loss": 0.4261, + "step": 31485 + }, + { + "epoch": 1.7591697628292873, + "grad_norm": 0.7242633104324341, + "learning_rate": 0.00012141416405199462, + "loss": 0.5186, + "step": 31486 + }, + { + "epoch": 1.7592256334329692, + "grad_norm": 0.3084433972835541, + "learning_rate": 0.00012138614970865084, + "loss": 0.401, + "step": 31487 + }, + { + "epoch": 1.7592815040366512, + "grad_norm": 0.32003360986709595, + "learning_rate": 0.00012135813536530704, + "loss": 0.4042, + "step": 31488 + }, + { + "epoch": 1.759337374640333, + "grad_norm": 0.43064117431640625, + "learning_rate": 0.00012133012102196325, + "loss": 0.4625, + "step": 31489 + }, + { + "epoch": 1.7593932452440149, + "grad_norm": 0.32716819643974304, + "learning_rate": 0.00012130210667861945, + "loss": 0.39, + "step": 31490 + }, + { + "epoch": 1.7594491158476968, + "grad_norm": 0.5281455516815186, + "learning_rate": 0.00012127409233527566, + "loss": 0.4055, + "step": 31491 + }, + { + "epoch": 1.7595049864513785, + "grad_norm": 0.3444617986679077, + "learning_rate": 0.00012124607799193187, + "loss": 0.3913, + "step": 31492 + }, + { + "epoch": 1.7595608570550605, + "grad_norm": 0.3697357177734375, + "learning_rate": 0.00012121806364858807, + "loss": 0.3487, + "step": 31493 + }, + { + "epoch": 1.7596167276587424, + "grad_norm": 0.5472471714019775, + "learning_rate": 0.00012119004930524428, + "loss": 0.6208, + "step": 31494 + }, + { + "epoch": 1.7596725982624242, + "grad_norm": 0.3853340744972229, + "learning_rate": 0.0001211620349619005, + "loss": 0.4292, + "step": 31495 + }, + { + "epoch": 1.7597284688661061, + "grad_norm": 0.315312922000885, + "learning_rate": 0.0001211340206185567, + "loss": 0.2909, + "step": 31496 + }, + { + "epoch": 1.759784339469788, + "grad_norm": 0.48537901043891907, + "learning_rate": 0.00012110600627521291, + "loss": 0.6975, + "step": 31497 + }, + { + "epoch": 1.7598402100734698, + "grad_norm": 0.42781999707221985, + "learning_rate": 0.00012107799193186913, + "loss": 0.4196, + "step": 31498 + }, + { + "epoch": 1.7598960806771518, + "grad_norm": 5.43608283996582, + "learning_rate": 0.00012104997758852533, + "loss": 0.3643, + "step": 31499 + }, + { + "epoch": 1.7599519512808337, + "grad_norm": 1.1934536695480347, + "learning_rate": 0.00012102196324518154, + "loss": 0.2976, + "step": 31500 + }, + { + "epoch": 1.7599519512808337, + "eval_cer": 0.08383249866343709, + "eval_loss": 0.31493711471557617, + "eval_runtime": 56.3251, + "eval_samples_per_second": 80.568, + "eval_steps_per_second": 5.042, + "eval_wer": 0.3319739672076199, + "step": 31500 + }, + { + "epoch": 1.7600078218845154, + "grad_norm": 0.5558372139930725, + "learning_rate": 0.00012099394890183775, + "loss": 0.364, + "step": 31501 + }, + { + "epoch": 1.7600636924881974, + "grad_norm": 1.278913140296936, + "learning_rate": 0.00012096593455849395, + "loss": 0.4034, + "step": 31502 + }, + { + "epoch": 1.7601195630918793, + "grad_norm": 0.41196900606155396, + "learning_rate": 0.00012093792021515016, + "loss": 0.2758, + "step": 31503 + }, + { + "epoch": 1.760175433695561, + "grad_norm": 0.3962152302265167, + "learning_rate": 0.00012090990587180636, + "loss": 0.3332, + "step": 31504 + }, + { + "epoch": 1.7602313042992428, + "grad_norm": 0.411489337682724, + "learning_rate": 0.00012088189152846257, + "loss": 0.3987, + "step": 31505 + }, + { + "epoch": 1.760287174902925, + "grad_norm": 0.4008850157260895, + "learning_rate": 0.00012085387718511879, + "loss": 0.3555, + "step": 31506 + }, + { + "epoch": 1.7603430455066067, + "grad_norm": 1.1089891195297241, + "learning_rate": 0.000120825862841775, + "loss": 0.3555, + "step": 31507 + }, + { + "epoch": 1.7603989161102884, + "grad_norm": 1.1776628494262695, + "learning_rate": 0.0001207978484984312, + "loss": 0.4351, + "step": 31508 + }, + { + "epoch": 1.7604547867139706, + "grad_norm": 0.4481344521045685, + "learning_rate": 0.0001207698341550874, + "loss": 0.5024, + "step": 31509 + }, + { + "epoch": 1.7605106573176523, + "grad_norm": 0.350281685590744, + "learning_rate": 0.00012074181981174363, + "loss": 0.4148, + "step": 31510 + }, + { + "epoch": 1.760566527921334, + "grad_norm": 0.3083142042160034, + "learning_rate": 0.00012071380546839983, + "loss": 0.2702, + "step": 31511 + }, + { + "epoch": 1.760622398525016, + "grad_norm": 0.44117748737335205, + "learning_rate": 0.00012068579112505604, + "loss": 0.4643, + "step": 31512 + }, + { + "epoch": 1.760678269128698, + "grad_norm": 0.4092458486557007, + "learning_rate": 0.00012065777678171223, + "loss": 0.4364, + "step": 31513 + }, + { + "epoch": 1.7607341397323797, + "grad_norm": 0.32180267572402954, + "learning_rate": 0.00012062976243836845, + "loss": 0.4014, + "step": 31514 + }, + { + "epoch": 1.7607900103360616, + "grad_norm": 0.5481683611869812, + "learning_rate": 0.00012060174809502465, + "loss": 0.4337, + "step": 31515 + }, + { + "epoch": 1.7608458809397436, + "grad_norm": 0.3762452006340027, + "learning_rate": 0.00012057373375168086, + "loss": 0.3774, + "step": 31516 + }, + { + "epoch": 1.7609017515434253, + "grad_norm": 0.5563046336174011, + "learning_rate": 0.00012054571940833707, + "loss": 0.4507, + "step": 31517 + }, + { + "epoch": 1.7609576221471073, + "grad_norm": 0.40352487564086914, + "learning_rate": 0.00012051770506499329, + "loss": 0.3603, + "step": 31518 + }, + { + "epoch": 1.7610134927507892, + "grad_norm": 0.39949604868888855, + "learning_rate": 0.00012048969072164949, + "loss": 0.4657, + "step": 31519 + }, + { + "epoch": 1.761069363354471, + "grad_norm": 0.5385668873786926, + "learning_rate": 0.0001204616763783057, + "loss": 0.4974, + "step": 31520 + }, + { + "epoch": 1.761125233958153, + "grad_norm": 5.900906085968018, + "learning_rate": 0.0001204336620349619, + "loss": 0.5243, + "step": 31521 + }, + { + "epoch": 1.7611811045618349, + "grad_norm": 3.832994222640991, + "learning_rate": 0.00012040564769161812, + "loss": 0.4825, + "step": 31522 + }, + { + "epoch": 1.7612369751655166, + "grad_norm": 0.42186546325683594, + "learning_rate": 0.00012037763334827431, + "loss": 0.4283, + "step": 31523 + }, + { + "epoch": 1.7612928457691985, + "grad_norm": 0.6257389187812805, + "learning_rate": 0.00012034961900493052, + "loss": 0.5235, + "step": 31524 + }, + { + "epoch": 1.7613487163728805, + "grad_norm": 0.39626872539520264, + "learning_rate": 0.00012032160466158673, + "loss": 0.4292, + "step": 31525 + }, + { + "epoch": 1.7614045869765622, + "grad_norm": 0.7104735374450684, + "learning_rate": 0.00012029359031824295, + "loss": 0.4359, + "step": 31526 + }, + { + "epoch": 1.7614604575802442, + "grad_norm": 0.5279558300971985, + "learning_rate": 0.00012026557597489915, + "loss": 0.45, + "step": 31527 + }, + { + "epoch": 1.7615163281839261, + "grad_norm": 0.9202028512954712, + "learning_rate": 0.00012023756163155536, + "loss": 0.3983, + "step": 31528 + }, + { + "epoch": 1.7615721987876078, + "grad_norm": 0.8643003702163696, + "learning_rate": 0.00012020954728821156, + "loss": 0.4352, + "step": 31529 + }, + { + "epoch": 1.7616280693912898, + "grad_norm": 0.48417964577674866, + "learning_rate": 0.00012018153294486778, + "loss": 0.3973, + "step": 31530 + }, + { + "epoch": 1.7616839399949717, + "grad_norm": 0.39703482389450073, + "learning_rate": 0.00012015351860152399, + "loss": 0.4164, + "step": 31531 + }, + { + "epoch": 1.7617398105986535, + "grad_norm": 0.49188485741615295, + "learning_rate": 0.0001201255042581802, + "loss": 0.4072, + "step": 31532 + }, + { + "epoch": 1.7617956812023354, + "grad_norm": 0.5528683066368103, + "learning_rate": 0.0001200974899148364, + "loss": 0.437, + "step": 31533 + }, + { + "epoch": 1.7618515518060174, + "grad_norm": 0.4306075870990753, + "learning_rate": 0.0001200694755714926, + "loss": 0.4486, + "step": 31534 + }, + { + "epoch": 1.761907422409699, + "grad_norm": 0.42050901055336, + "learning_rate": 0.00012004146122814881, + "loss": 0.4462, + "step": 31535 + }, + { + "epoch": 1.761963293013381, + "grad_norm": 0.39860159158706665, + "learning_rate": 0.00012001344688480502, + "loss": 0.4316, + "step": 31536 + }, + { + "epoch": 1.762019163617063, + "grad_norm": 0.4456721544265747, + "learning_rate": 0.00011998543254146122, + "loss": 0.4479, + "step": 31537 + }, + { + "epoch": 1.7620750342207447, + "grad_norm": 0.6237293481826782, + "learning_rate": 0.00011995741819811744, + "loss": 0.4371, + "step": 31538 + }, + { + "epoch": 1.7621309048244265, + "grad_norm": 2.244490623474121, + "learning_rate": 0.00011992940385477365, + "loss": 0.4934, + "step": 31539 + }, + { + "epoch": 1.7621867754281086, + "grad_norm": 0.6747792363166809, + "learning_rate": 0.00011990138951142985, + "loss": 0.2834, + "step": 31540 + }, + { + "epoch": 1.7622426460317904, + "grad_norm": 0.4262988269329071, + "learning_rate": 0.00011987337516808606, + "loss": 0.6048, + "step": 31541 + }, + { + "epoch": 1.762298516635472, + "grad_norm": 0.4557267725467682, + "learning_rate": 0.00011984536082474228, + "loss": 0.4242, + "step": 31542 + }, + { + "epoch": 1.7623543872391543, + "grad_norm": 0.46239355206489563, + "learning_rate": 0.00011981734648139849, + "loss": 0.3379, + "step": 31543 + }, + { + "epoch": 1.762410257842836, + "grad_norm": 0.8817043304443359, + "learning_rate": 0.00011978933213805468, + "loss": 0.3817, + "step": 31544 + }, + { + "epoch": 1.7624661284465177, + "grad_norm": 0.7488983869552612, + "learning_rate": 0.00011976131779471088, + "loss": 0.4963, + "step": 31545 + }, + { + "epoch": 1.7625219990501997, + "grad_norm": 0.5229520201683044, + "learning_rate": 0.0001197333034513671, + "loss": 0.4791, + "step": 31546 + }, + { + "epoch": 1.7625778696538816, + "grad_norm": 0.37521252036094666, + "learning_rate": 0.00011970528910802331, + "loss": 0.3833, + "step": 31547 + }, + { + "epoch": 1.7626337402575634, + "grad_norm": 0.3682505190372467, + "learning_rate": 0.00011967727476467951, + "loss": 0.3541, + "step": 31548 + }, + { + "epoch": 1.7626896108612453, + "grad_norm": 0.4292196035385132, + "learning_rate": 0.00011964926042133572, + "loss": 0.4482, + "step": 31549 + }, + { + "epoch": 1.7627454814649273, + "grad_norm": 0.5987961292266846, + "learning_rate": 0.00011962124607799194, + "loss": 0.4226, + "step": 31550 + }, + { + "epoch": 1.762801352068609, + "grad_norm": 0.3961001932621002, + "learning_rate": 0.00011959323173464815, + "loss": 0.4059, + "step": 31551 + }, + { + "epoch": 1.762857222672291, + "grad_norm": 0.46439629793167114, + "learning_rate": 0.00011956521739130435, + "loss": 0.5076, + "step": 31552 + }, + { + "epoch": 1.762913093275973, + "grad_norm": 0.4724852442741394, + "learning_rate": 0.00011953720304796056, + "loss": 0.3443, + "step": 31553 + }, + { + "epoch": 1.7629689638796546, + "grad_norm": 0.43590590357780457, + "learning_rate": 0.00011950918870461678, + "loss": 0.4203, + "step": 31554 + }, + { + "epoch": 1.7630248344833366, + "grad_norm": 0.6805886626243591, + "learning_rate": 0.00011948117436127297, + "loss": 0.415, + "step": 31555 + }, + { + "epoch": 1.7630807050870185, + "grad_norm": 0.5985432267189026, + "learning_rate": 0.00011945316001792917, + "loss": 0.4301, + "step": 31556 + }, + { + "epoch": 1.7631365756907003, + "grad_norm": 0.44912055134773254, + "learning_rate": 0.00011942514567458538, + "loss": 0.3617, + "step": 31557 + }, + { + "epoch": 1.7631924462943822, + "grad_norm": 0.3540697991847992, + "learning_rate": 0.0001193971313312416, + "loss": 0.3696, + "step": 31558 + }, + { + "epoch": 1.7632483168980642, + "grad_norm": 0.3842289447784424, + "learning_rate": 0.0001193691169878978, + "loss": 0.409, + "step": 31559 + }, + { + "epoch": 1.763304187501746, + "grad_norm": 0.4344330430030823, + "learning_rate": 0.00011934110264455401, + "loss": 0.4689, + "step": 31560 + }, + { + "epoch": 1.7633600581054278, + "grad_norm": 0.4588421583175659, + "learning_rate": 0.00011931308830121023, + "loss": 0.4386, + "step": 31561 + }, + { + "epoch": 1.7634159287091098, + "grad_norm": 1.1858634948730469, + "learning_rate": 0.00011928507395786644, + "loss": 0.4763, + "step": 31562 + }, + { + "epoch": 1.7634717993127915, + "grad_norm": 0.4952344000339508, + "learning_rate": 0.00011925705961452264, + "loss": 0.53, + "step": 31563 + }, + { + "epoch": 1.7635276699164735, + "grad_norm": 0.40796175599098206, + "learning_rate": 0.00011922904527117885, + "loss": 0.4292, + "step": 31564 + }, + { + "epoch": 1.7635835405201554, + "grad_norm": 0.721276581287384, + "learning_rate": 0.00011920103092783507, + "loss": 0.5023, + "step": 31565 + }, + { + "epoch": 1.7636394111238372, + "grad_norm": 1.5314428806304932, + "learning_rate": 0.00011917301658449126, + "loss": 0.5289, + "step": 31566 + }, + { + "epoch": 1.763695281727519, + "grad_norm": 0.39037832617759705, + "learning_rate": 0.00011914500224114747, + "loss": 0.2857, + "step": 31567 + }, + { + "epoch": 1.763751152331201, + "grad_norm": 0.9632468223571777, + "learning_rate": 0.00011911698789780367, + "loss": 0.4665, + "step": 31568 + }, + { + "epoch": 1.7638070229348828, + "grad_norm": 5.240898609161377, + "learning_rate": 0.00011908897355445989, + "loss": 0.4271, + "step": 31569 + }, + { + "epoch": 1.7638628935385647, + "grad_norm": 0.8189048171043396, + "learning_rate": 0.0001190609592111161, + "loss": 0.3273, + "step": 31570 + }, + { + "epoch": 1.7639187641422467, + "grad_norm": 0.5738352537155151, + "learning_rate": 0.0001190329448677723, + "loss": 0.4747, + "step": 31571 + }, + { + "epoch": 1.7639746347459284, + "grad_norm": 0.6016944646835327, + "learning_rate": 0.00011900493052442851, + "loss": 0.5372, + "step": 31572 + }, + { + "epoch": 1.7640305053496101, + "grad_norm": 1.218602180480957, + "learning_rate": 0.00011897691618108473, + "loss": 0.4984, + "step": 31573 + }, + { + "epoch": 1.7640863759532923, + "grad_norm": 0.3576766848564148, + "learning_rate": 0.00011894890183774093, + "loss": 0.3891, + "step": 31574 + }, + { + "epoch": 1.764142246556974, + "grad_norm": 0.4251437485218048, + "learning_rate": 0.00011892088749439714, + "loss": 0.4082, + "step": 31575 + }, + { + "epoch": 1.7641981171606558, + "grad_norm": 0.4354020059108734, + "learning_rate": 0.00011889287315105333, + "loss": 0.4077, + "step": 31576 + }, + { + "epoch": 1.764253987764338, + "grad_norm": 0.4071834087371826, + "learning_rate": 0.00011886485880770955, + "loss": 0.382, + "step": 31577 + }, + { + "epoch": 1.7643098583680197, + "grad_norm": 0.35758858919143677, + "learning_rate": 0.00011883684446436576, + "loss": 0.3037, + "step": 31578 + }, + { + "epoch": 1.7643657289717014, + "grad_norm": 0.6078523993492126, + "learning_rate": 0.00011880883012102196, + "loss": 0.4317, + "step": 31579 + }, + { + "epoch": 1.7644215995753834, + "grad_norm": 3.2302329540252686, + "learning_rate": 0.00011878081577767817, + "loss": 0.4135, + "step": 31580 + }, + { + "epoch": 1.7644774701790653, + "grad_norm": 0.595986008644104, + "learning_rate": 0.00011875280143433439, + "loss": 0.4999, + "step": 31581 + }, + { + "epoch": 1.764533340782747, + "grad_norm": 0.729296863079071, + "learning_rate": 0.00011872478709099059, + "loss": 0.3415, + "step": 31582 + }, + { + "epoch": 1.764589211386429, + "grad_norm": 0.37851956486701965, + "learning_rate": 0.0001186967727476468, + "loss": 0.4087, + "step": 31583 + }, + { + "epoch": 1.764645081990111, + "grad_norm": 0.47566068172454834, + "learning_rate": 0.000118668758404303, + "loss": 0.2999, + "step": 31584 + }, + { + "epoch": 1.7647009525937927, + "grad_norm": 0.6420467495918274, + "learning_rate": 0.00011864074406095922, + "loss": 0.3564, + "step": 31585 + }, + { + "epoch": 1.7647568231974746, + "grad_norm": 1.754269003868103, + "learning_rate": 0.00011861272971761543, + "loss": 0.5785, + "step": 31586 + }, + { + "epoch": 1.7648126938011566, + "grad_norm": 0.4304639995098114, + "learning_rate": 0.00011858471537427162, + "loss": 0.4311, + "step": 31587 + }, + { + "epoch": 1.7648685644048383, + "grad_norm": 0.8775295615196228, + "learning_rate": 0.00011855670103092783, + "loss": 0.376, + "step": 31588 + }, + { + "epoch": 1.7649244350085203, + "grad_norm": 0.480243057012558, + "learning_rate": 0.00011852868668758405, + "loss": 0.3369, + "step": 31589 + }, + { + "epoch": 1.7649803056122022, + "grad_norm": 0.5063501000404358, + "learning_rate": 0.00011850067234424025, + "loss": 0.627, + "step": 31590 + }, + { + "epoch": 1.765036176215884, + "grad_norm": 0.4874529540538788, + "learning_rate": 0.00011847265800089646, + "loss": 0.5847, + "step": 31591 + }, + { + "epoch": 1.7650920468195659, + "grad_norm": 0.5997948050498962, + "learning_rate": 0.00011844464365755266, + "loss": 0.3806, + "step": 31592 + }, + { + "epoch": 1.7651479174232478, + "grad_norm": 0.30644479393959045, + "learning_rate": 0.00011841662931420888, + "loss": 0.3936, + "step": 31593 + }, + { + "epoch": 1.7652037880269296, + "grad_norm": 0.9245882630348206, + "learning_rate": 0.00011838861497086509, + "loss": 0.3615, + "step": 31594 + }, + { + "epoch": 1.7652596586306115, + "grad_norm": 0.4444141983985901, + "learning_rate": 0.0001183606006275213, + "loss": 0.4662, + "step": 31595 + }, + { + "epoch": 1.7653155292342935, + "grad_norm": 0.381323903799057, + "learning_rate": 0.0001183325862841775, + "loss": 0.4044, + "step": 31596 + }, + { + "epoch": 1.7653713998379752, + "grad_norm": 2.2271456718444824, + "learning_rate": 0.00011830457194083372, + "loss": 0.4, + "step": 31597 + }, + { + "epoch": 1.7654272704416571, + "grad_norm": 0.3505699038505554, + "learning_rate": 0.00011827655759748991, + "loss": 0.3888, + "step": 31598 + }, + { + "epoch": 1.765483141045339, + "grad_norm": 0.6509354710578918, + "learning_rate": 0.00011824854325414612, + "loss": 0.4145, + "step": 31599 + }, + { + "epoch": 1.7655390116490208, + "grad_norm": 0.35409003496170044, + "learning_rate": 0.00011822052891080232, + "loss": 0.4168, + "step": 31600 + }, + { + "epoch": 1.7655948822527028, + "grad_norm": 0.5262879133224487, + "learning_rate": 0.00011819251456745854, + "loss": 0.4074, + "step": 31601 + }, + { + "epoch": 1.7656507528563847, + "grad_norm": 0.7615727186203003, + "learning_rate": 0.00011816450022411475, + "loss": 0.7441, + "step": 31602 + }, + { + "epoch": 1.7657066234600665, + "grad_norm": 0.4267369210720062, + "learning_rate": 0.00011813648588077096, + "loss": 0.3455, + "step": 31603 + }, + { + "epoch": 1.7657624940637484, + "grad_norm": 0.6536475419998169, + "learning_rate": 0.00011810847153742716, + "loss": 0.4314, + "step": 31604 + }, + { + "epoch": 1.7658183646674304, + "grad_norm": 0.42824140191078186, + "learning_rate": 0.00011808045719408338, + "loss": 0.4295, + "step": 31605 + }, + { + "epoch": 1.765874235271112, + "grad_norm": 0.5583625435829163, + "learning_rate": 0.00011805244285073959, + "loss": 0.4043, + "step": 31606 + }, + { + "epoch": 1.7659301058747938, + "grad_norm": 0.3020534813404083, + "learning_rate": 0.00011802442850739579, + "loss": 0.3154, + "step": 31607 + }, + { + "epoch": 1.765985976478476, + "grad_norm": 0.898160457611084, + "learning_rate": 0.00011799641416405198, + "loss": 0.37, + "step": 31608 + }, + { + "epoch": 1.7660418470821577, + "grad_norm": 0.5278001427650452, + "learning_rate": 0.0001179683998207082, + "loss": 0.4297, + "step": 31609 + }, + { + "epoch": 1.7660977176858395, + "grad_norm": 0.3115200996398926, + "learning_rate": 0.00011794038547736441, + "loss": 0.3463, + "step": 31610 + }, + { + "epoch": 1.7661535882895216, + "grad_norm": 0.3590901792049408, + "learning_rate": 0.00011791237113402062, + "loss": 0.3843, + "step": 31611 + }, + { + "epoch": 1.7662094588932034, + "grad_norm": 0.4310905337333679, + "learning_rate": 0.00011788435679067682, + "loss": 0.4145, + "step": 31612 + }, + { + "epoch": 1.766265329496885, + "grad_norm": 0.5809847712516785, + "learning_rate": 0.00011785634244733304, + "loss": 0.3936, + "step": 31613 + }, + { + "epoch": 1.766321200100567, + "grad_norm": 0.33099648356437683, + "learning_rate": 0.00011782832810398925, + "loss": 0.404, + "step": 31614 + }, + { + "epoch": 1.766377070704249, + "grad_norm": 0.6290991306304932, + "learning_rate": 0.00011780031376064545, + "loss": 0.3981, + "step": 31615 + }, + { + "epoch": 1.7664329413079307, + "grad_norm": 1.9333230257034302, + "learning_rate": 0.00011777229941730166, + "loss": 0.3963, + "step": 31616 + }, + { + "epoch": 1.7664888119116127, + "grad_norm": 0.482319712638855, + "learning_rate": 0.00011774428507395788, + "loss": 0.4071, + "step": 31617 + }, + { + "epoch": 1.7665446825152946, + "grad_norm": 0.46068984270095825, + "learning_rate": 0.00011771627073061408, + "loss": 0.4654, + "step": 31618 + }, + { + "epoch": 1.7666005531189763, + "grad_norm": 1.1653836965560913, + "learning_rate": 0.00011768825638727028, + "loss": 0.3138, + "step": 31619 + }, + { + "epoch": 1.7666564237226583, + "grad_norm": 0.5078619718551636, + "learning_rate": 0.00011766024204392648, + "loss": 0.4383, + "step": 31620 + }, + { + "epoch": 1.7667122943263402, + "grad_norm": 0.3683968782424927, + "learning_rate": 0.0001176322277005827, + "loss": 0.3333, + "step": 31621 + }, + { + "epoch": 1.766768164930022, + "grad_norm": 0.43304312229156494, + "learning_rate": 0.00011760421335723891, + "loss": 0.388, + "step": 31622 + }, + { + "epoch": 1.766824035533704, + "grad_norm": 1.2372068166732788, + "learning_rate": 0.00011757619901389511, + "loss": 0.4471, + "step": 31623 + }, + { + "epoch": 1.7668799061373859, + "grad_norm": 0.8130107522010803, + "learning_rate": 0.00011754818467055132, + "loss": 0.6681, + "step": 31624 + }, + { + "epoch": 1.7669357767410676, + "grad_norm": 3.5076584815979004, + "learning_rate": 0.00011752017032720754, + "loss": 0.3303, + "step": 31625 + }, + { + "epoch": 1.7669916473447496, + "grad_norm": 0.9620332717895508, + "learning_rate": 0.00011749215598386374, + "loss": 0.5887, + "step": 31626 + }, + { + "epoch": 1.7670475179484315, + "grad_norm": 1.0733455419540405, + "learning_rate": 0.00011746414164051995, + "loss": 0.6, + "step": 31627 + }, + { + "epoch": 1.7671033885521132, + "grad_norm": 1.4906930923461914, + "learning_rate": 0.00011743612729717617, + "loss": 0.4728, + "step": 31628 + }, + { + "epoch": 1.7671592591557952, + "grad_norm": 0.41727039217948914, + "learning_rate": 0.00011740811295383236, + "loss": 0.3412, + "step": 31629 + }, + { + "epoch": 1.7672151297594771, + "grad_norm": 1.2480818033218384, + "learning_rate": 0.00011738009861048857, + "loss": 0.3496, + "step": 31630 + }, + { + "epoch": 1.7672710003631589, + "grad_norm": 1.3756752014160156, + "learning_rate": 0.00011735208426714477, + "loss": 0.4085, + "step": 31631 + }, + { + "epoch": 1.7673268709668408, + "grad_norm": 0.4326397776603699, + "learning_rate": 0.00011732406992380099, + "loss": 0.3421, + "step": 31632 + }, + { + "epoch": 1.7673827415705228, + "grad_norm": 0.6361571550369263, + "learning_rate": 0.0001172960555804572, + "loss": 0.3873, + "step": 31633 + }, + { + "epoch": 1.7674386121742045, + "grad_norm": 0.780151903629303, + "learning_rate": 0.0001172680412371134, + "loss": 0.4881, + "step": 31634 + }, + { + "epoch": 1.7674944827778865, + "grad_norm": 0.27587035298347473, + "learning_rate": 0.00011724002689376961, + "loss": 0.3074, + "step": 31635 + }, + { + "epoch": 1.7675503533815684, + "grad_norm": 0.5354198217391968, + "learning_rate": 0.00011721201255042583, + "loss": 0.3754, + "step": 31636 + }, + { + "epoch": 1.7676062239852501, + "grad_norm": 0.4902915060520172, + "learning_rate": 0.00011718399820708203, + "loss": 0.4931, + "step": 31637 + }, + { + "epoch": 1.767662094588932, + "grad_norm": 0.9258067607879639, + "learning_rate": 0.00011715598386373824, + "loss": 0.5579, + "step": 31638 + }, + { + "epoch": 1.767717965192614, + "grad_norm": 0.42149725556373596, + "learning_rate": 0.00011712796952039445, + "loss": 0.4949, + "step": 31639 + }, + { + "epoch": 1.7677738357962958, + "grad_norm": 0.6292809844017029, + "learning_rate": 0.00011709995517705065, + "loss": 0.4895, + "step": 31640 + }, + { + "epoch": 1.7678297063999775, + "grad_norm": 0.4497605860233307, + "learning_rate": 0.00011707194083370686, + "loss": 0.4408, + "step": 31641 + }, + { + "epoch": 1.7678855770036597, + "grad_norm": 1.8561701774597168, + "learning_rate": 0.00011704392649036306, + "loss": 0.5363, + "step": 31642 + }, + { + "epoch": 1.7679414476073414, + "grad_norm": 0.31612053513526917, + "learning_rate": 0.00011701591214701927, + "loss": 0.3653, + "step": 31643 + }, + { + "epoch": 1.7679973182110231, + "grad_norm": 0.38826221227645874, + "learning_rate": 0.00011698789780367549, + "loss": 0.3504, + "step": 31644 + }, + { + "epoch": 1.7680531888147053, + "grad_norm": 0.4031623601913452, + "learning_rate": 0.0001169598834603317, + "loss": 0.4082, + "step": 31645 + }, + { + "epoch": 1.768109059418387, + "grad_norm": 0.34372478723526, + "learning_rate": 0.0001169318691169879, + "loss": 0.3877, + "step": 31646 + }, + { + "epoch": 1.7681649300220688, + "grad_norm": 1.0988538265228271, + "learning_rate": 0.0001169038547736441, + "loss": 0.3837, + "step": 31647 + }, + { + "epoch": 1.7682208006257507, + "grad_norm": 0.5155931115150452, + "learning_rate": 0.00011687584043030033, + "loss": 0.45, + "step": 31648 + }, + { + "epoch": 1.7682766712294327, + "grad_norm": 3.690976142883301, + "learning_rate": 0.00011684782608695653, + "loss": 0.368, + "step": 31649 + }, + { + "epoch": 1.7683325418331144, + "grad_norm": 0.6752966642379761, + "learning_rate": 0.00011681981174361274, + "loss": 0.4197, + "step": 31650 + }, + { + "epoch": 1.7683884124367963, + "grad_norm": 0.713964581489563, + "learning_rate": 0.00011679179740026893, + "loss": 0.6105, + "step": 31651 + }, + { + "epoch": 1.7684442830404783, + "grad_norm": 0.4691053330898285, + "learning_rate": 0.00011676378305692515, + "loss": 0.3107, + "step": 31652 + }, + { + "epoch": 1.76850015364416, + "grad_norm": 0.36323463916778564, + "learning_rate": 0.00011673576871358136, + "loss": 0.4946, + "step": 31653 + }, + { + "epoch": 1.768556024247842, + "grad_norm": 0.4147511422634125, + "learning_rate": 0.00011670775437023756, + "loss": 0.3624, + "step": 31654 + }, + { + "epoch": 1.768611894851524, + "grad_norm": 0.5434030294418335, + "learning_rate": 0.00011667974002689377, + "loss": 0.5555, + "step": 31655 + }, + { + "epoch": 1.7686677654552057, + "grad_norm": 0.4437359869480133, + "learning_rate": 0.00011665172568354999, + "loss": 0.4891, + "step": 31656 + }, + { + "epoch": 1.7687236360588876, + "grad_norm": 0.45628130435943604, + "learning_rate": 0.00011662371134020619, + "loss": 0.3378, + "step": 31657 + }, + { + "epoch": 1.7687795066625696, + "grad_norm": 0.33154773712158203, + "learning_rate": 0.0001165956969968624, + "loss": 0.369, + "step": 31658 + }, + { + "epoch": 1.7688353772662513, + "grad_norm": 1.5385966300964355, + "learning_rate": 0.0001165676826535186, + "loss": 0.4098, + "step": 31659 + }, + { + "epoch": 1.7688912478699332, + "grad_norm": 0.5450655817985535, + "learning_rate": 0.00011653966831017482, + "loss": 0.4955, + "step": 31660 + }, + { + "epoch": 1.7689471184736152, + "grad_norm": 0.41523271799087524, + "learning_rate": 0.00011651165396683102, + "loss": 0.4023, + "step": 31661 + }, + { + "epoch": 1.769002989077297, + "grad_norm": 0.5887847542762756, + "learning_rate": 0.00011648363962348722, + "loss": 0.5195, + "step": 31662 + }, + { + "epoch": 1.7690588596809789, + "grad_norm": 0.38504961133003235, + "learning_rate": 0.00011645562528014343, + "loss": 0.3959, + "step": 31663 + }, + { + "epoch": 1.7691147302846608, + "grad_norm": 0.47216662764549255, + "learning_rate": 0.00011642761093679965, + "loss": 0.469, + "step": 31664 + }, + { + "epoch": 1.7691706008883425, + "grad_norm": 0.44270193576812744, + "learning_rate": 0.00011639959659345585, + "loss": 0.4102, + "step": 31665 + }, + { + "epoch": 1.7692264714920245, + "grad_norm": 0.46783700585365295, + "learning_rate": 0.00011637158225011206, + "loss": 0.3439, + "step": 31666 + }, + { + "epoch": 1.7692823420957065, + "grad_norm": 0.6868758797645569, + "learning_rate": 0.00011634356790676826, + "loss": 0.4512, + "step": 31667 + }, + { + "epoch": 1.7693382126993882, + "grad_norm": 2.1134414672851562, + "learning_rate": 0.00011631555356342448, + "loss": 0.3244, + "step": 31668 + }, + { + "epoch": 1.7693940833030701, + "grad_norm": 0.4616192579269409, + "learning_rate": 0.00011628753922008069, + "loss": 0.4073, + "step": 31669 + }, + { + "epoch": 1.769449953906752, + "grad_norm": 0.45286285877227783, + "learning_rate": 0.0001162595248767369, + "loss": 0.4357, + "step": 31670 + }, + { + "epoch": 1.7695058245104338, + "grad_norm": 0.5563863515853882, + "learning_rate": 0.0001162315105333931, + "loss": 0.452, + "step": 31671 + }, + { + "epoch": 1.7695616951141158, + "grad_norm": 0.6672083139419556, + "learning_rate": 0.0001162034961900493, + "loss": 0.4276, + "step": 31672 + }, + { + "epoch": 1.7696175657177977, + "grad_norm": 0.7815179228782654, + "learning_rate": 0.00011617548184670551, + "loss": 0.4317, + "step": 31673 + }, + { + "epoch": 1.7696734363214794, + "grad_norm": 0.4349629282951355, + "learning_rate": 0.00011614746750336172, + "loss": 0.3372, + "step": 31674 + }, + { + "epoch": 1.7697293069251612, + "grad_norm": 0.4026866555213928, + "learning_rate": 0.00011611945316001792, + "loss": 0.3893, + "step": 31675 + }, + { + "epoch": 1.7697851775288433, + "grad_norm": 0.4619576036930084, + "learning_rate": 0.00011609143881667414, + "loss": 0.5622, + "step": 31676 + }, + { + "epoch": 1.769841048132525, + "grad_norm": 1.0549875497817993, + "learning_rate": 0.00011606342447333035, + "loss": 0.3237, + "step": 31677 + }, + { + "epoch": 1.7698969187362068, + "grad_norm": 0.5571673512458801, + "learning_rate": 0.00011603541012998655, + "loss": 0.361, + "step": 31678 + }, + { + "epoch": 1.769952789339889, + "grad_norm": 0.5260737538337708, + "learning_rate": 0.00011600739578664276, + "loss": 0.3753, + "step": 31679 + }, + { + "epoch": 1.7700086599435707, + "grad_norm": 0.31217706203460693, + "learning_rate": 0.00011597938144329898, + "loss": 0.3432, + "step": 31680 + }, + { + "epoch": 1.7700645305472524, + "grad_norm": 0.8029001355171204, + "learning_rate": 0.00011595136709995519, + "loss": 0.435, + "step": 31681 + }, + { + "epoch": 1.7701204011509344, + "grad_norm": 1.212152123451233, + "learning_rate": 0.00011592335275661139, + "loss": 0.4879, + "step": 31682 + }, + { + "epoch": 1.7701762717546163, + "grad_norm": 0.4707568883895874, + "learning_rate": 0.00011589533841326758, + "loss": 0.4238, + "step": 31683 + }, + { + "epoch": 1.770232142358298, + "grad_norm": 0.6399535536766052, + "learning_rate": 0.0001158673240699238, + "loss": 0.5455, + "step": 31684 + }, + { + "epoch": 1.77028801296198, + "grad_norm": 0.407356321811676, + "learning_rate": 0.00011583930972658001, + "loss": 0.4276, + "step": 31685 + }, + { + "epoch": 1.770343883565662, + "grad_norm": 0.4608050584793091, + "learning_rate": 0.00011581129538323621, + "loss": 0.4227, + "step": 31686 + }, + { + "epoch": 1.7703997541693437, + "grad_norm": 0.4570327699184418, + "learning_rate": 0.00011578328103989242, + "loss": 0.3362, + "step": 31687 + }, + { + "epoch": 1.7704556247730256, + "grad_norm": 0.279826283454895, + "learning_rate": 0.00011575526669654864, + "loss": 0.3067, + "step": 31688 + }, + { + "epoch": 1.7705114953767076, + "grad_norm": 1.4555858373641968, + "learning_rate": 0.00011572725235320485, + "loss": 0.5207, + "step": 31689 + }, + { + "epoch": 1.7705673659803893, + "grad_norm": 0.4358065128326416, + "learning_rate": 0.00011569923800986105, + "loss": 0.4665, + "step": 31690 + }, + { + "epoch": 1.7706232365840713, + "grad_norm": 0.921259880065918, + "learning_rate": 0.00011567122366651726, + "loss": 0.3478, + "step": 31691 + }, + { + "epoch": 1.7706791071877532, + "grad_norm": 0.3696548342704773, + "learning_rate": 0.00011564320932317348, + "loss": 0.4425, + "step": 31692 + }, + { + "epoch": 1.770734977791435, + "grad_norm": 0.32940465211868286, + "learning_rate": 0.00011561519497982967, + "loss": 0.3135, + "step": 31693 + }, + { + "epoch": 1.770790848395117, + "grad_norm": 0.805751383304596, + "learning_rate": 0.00011558718063648587, + "loss": 0.3964, + "step": 31694 + }, + { + "epoch": 1.7708467189987989, + "grad_norm": 0.5005905032157898, + "learning_rate": 0.00011555916629314208, + "loss": 0.461, + "step": 31695 + }, + { + "epoch": 1.7709025896024806, + "grad_norm": 3.4680659770965576, + "learning_rate": 0.0001155311519497983, + "loss": 0.4824, + "step": 31696 + }, + { + "epoch": 1.7709584602061625, + "grad_norm": 0.31225523352622986, + "learning_rate": 0.0001155031376064545, + "loss": 0.4044, + "step": 31697 + }, + { + "epoch": 1.7710143308098445, + "grad_norm": 0.486028254032135, + "learning_rate": 0.00011547512326311071, + "loss": 0.345, + "step": 31698 + }, + { + "epoch": 1.7710702014135262, + "grad_norm": 0.6044283509254456, + "learning_rate": 0.00011544710891976693, + "loss": 0.4409, + "step": 31699 + }, + { + "epoch": 1.7711260720172082, + "grad_norm": 0.3143458366394043, + "learning_rate": 0.00011541909457642314, + "loss": 0.3523, + "step": 31700 + }, + { + "epoch": 1.7711819426208901, + "grad_norm": 0.5576046109199524, + "learning_rate": 0.00011539108023307934, + "loss": 0.4211, + "step": 31701 + }, + { + "epoch": 1.7712378132245719, + "grad_norm": 0.4496707022190094, + "learning_rate": 0.00011536306588973555, + "loss": 0.4116, + "step": 31702 + }, + { + "epoch": 1.7712936838282538, + "grad_norm": 0.4633626341819763, + "learning_rate": 0.00011533505154639177, + "loss": 0.4848, + "step": 31703 + }, + { + "epoch": 1.7713495544319358, + "grad_norm": 0.41117846965789795, + "learning_rate": 0.00011530703720304796, + "loss": 0.404, + "step": 31704 + }, + { + "epoch": 1.7714054250356175, + "grad_norm": 0.510601818561554, + "learning_rate": 0.00011527902285970417, + "loss": 0.5408, + "step": 31705 + }, + { + "epoch": 1.7714612956392994, + "grad_norm": 0.5351837873458862, + "learning_rate": 0.00011525100851636037, + "loss": 0.4529, + "step": 31706 + }, + { + "epoch": 1.7715171662429814, + "grad_norm": 0.4461905062198639, + "learning_rate": 0.00011522299417301659, + "loss": 0.3129, + "step": 31707 + }, + { + "epoch": 1.7715730368466631, + "grad_norm": 2.8043935298919678, + "learning_rate": 0.0001151949798296728, + "loss": 0.4484, + "step": 31708 + }, + { + "epoch": 1.7716289074503448, + "grad_norm": 0.3609589636325836, + "learning_rate": 0.000115166965486329, + "loss": 0.2817, + "step": 31709 + }, + { + "epoch": 1.771684778054027, + "grad_norm": 0.46025413274765015, + "learning_rate": 0.00011513895114298521, + "loss": 0.4172, + "step": 31710 + }, + { + "epoch": 1.7717406486577087, + "grad_norm": 0.4100381135940552, + "learning_rate": 0.00011511093679964143, + "loss": 0.4715, + "step": 31711 + }, + { + "epoch": 1.7717965192613905, + "grad_norm": 0.6361172795295715, + "learning_rate": 0.00011508292245629763, + "loss": 0.4467, + "step": 31712 + }, + { + "epoch": 1.7718523898650727, + "grad_norm": 0.4444752037525177, + "learning_rate": 0.00011505490811295384, + "loss": 0.3934, + "step": 31713 + }, + { + "epoch": 1.7719082604687544, + "grad_norm": 1.1015440225601196, + "learning_rate": 0.00011502689376961003, + "loss": 0.4237, + "step": 31714 + }, + { + "epoch": 1.771964131072436, + "grad_norm": 1.3951539993286133, + "learning_rate": 0.00011499887942626625, + "loss": 0.4829, + "step": 31715 + }, + { + "epoch": 1.772020001676118, + "grad_norm": 0.5899978876113892, + "learning_rate": 0.00011497086508292246, + "loss": 0.3923, + "step": 31716 + }, + { + "epoch": 1.7720758722798, + "grad_norm": 0.5508697032928467, + "learning_rate": 0.00011494285073957866, + "loss": 0.4144, + "step": 31717 + }, + { + "epoch": 1.7721317428834817, + "grad_norm": 0.3666797876358032, + "learning_rate": 0.00011491483639623487, + "loss": 0.4049, + "step": 31718 + }, + { + "epoch": 1.7721876134871637, + "grad_norm": 0.8316260576248169, + "learning_rate": 0.00011488682205289109, + "loss": 0.3643, + "step": 31719 + }, + { + "epoch": 1.7722434840908456, + "grad_norm": 0.36223816871643066, + "learning_rate": 0.0001148588077095473, + "loss": 0.3656, + "step": 31720 + }, + { + "epoch": 1.7722993546945274, + "grad_norm": 0.6196144819259644, + "learning_rate": 0.0001148307933662035, + "loss": 0.5852, + "step": 31721 + }, + { + "epoch": 1.7723552252982093, + "grad_norm": 0.3742961883544922, + "learning_rate": 0.0001148027790228597, + "loss": 0.4474, + "step": 31722 + }, + { + "epoch": 1.7724110959018913, + "grad_norm": 0.4354715049266815, + "learning_rate": 0.00011477476467951592, + "loss": 0.4076, + "step": 31723 + }, + { + "epoch": 1.772466966505573, + "grad_norm": 0.49297526478767395, + "learning_rate": 0.00011474675033617213, + "loss": 0.3848, + "step": 31724 + }, + { + "epoch": 1.772522837109255, + "grad_norm": 0.5727415680885315, + "learning_rate": 0.00011471873599282832, + "loss": 0.3568, + "step": 31725 + }, + { + "epoch": 1.772578707712937, + "grad_norm": 0.8529189825057983, + "learning_rate": 0.00011469072164948453, + "loss": 0.5323, + "step": 31726 + }, + { + "epoch": 1.7726345783166186, + "grad_norm": 0.43312886357307434, + "learning_rate": 0.00011466270730614075, + "loss": 0.4996, + "step": 31727 + }, + { + "epoch": 1.7726904489203006, + "grad_norm": 0.34950128197669983, + "learning_rate": 0.00011463469296279695, + "loss": 0.3846, + "step": 31728 + }, + { + "epoch": 1.7727463195239825, + "grad_norm": 0.6204403042793274, + "learning_rate": 0.00011460667861945316, + "loss": 0.4104, + "step": 31729 + }, + { + "epoch": 1.7728021901276643, + "grad_norm": 3.0391130447387695, + "learning_rate": 0.00011457866427610937, + "loss": 0.4321, + "step": 31730 + }, + { + "epoch": 1.7728580607313462, + "grad_norm": 0.3903975486755371, + "learning_rate": 0.00011455064993276558, + "loss": 0.3676, + "step": 31731 + }, + { + "epoch": 1.7729139313350282, + "grad_norm": 0.3486458659172058, + "learning_rate": 0.00011452263558942179, + "loss": 0.3895, + "step": 31732 + }, + { + "epoch": 1.77296980193871, + "grad_norm": 0.3841603398323059, + "learning_rate": 0.000114494621246078, + "loss": 0.3861, + "step": 31733 + }, + { + "epoch": 1.7730256725423919, + "grad_norm": 0.43673115968704224, + "learning_rate": 0.0001144666069027342, + "loss": 0.4216, + "step": 31734 + }, + { + "epoch": 1.7730815431460738, + "grad_norm": 0.3656087815761566, + "learning_rate": 0.00011443859255939042, + "loss": 0.3809, + "step": 31735 + }, + { + "epoch": 1.7731374137497555, + "grad_norm": 0.7435001730918884, + "learning_rate": 0.00011441057821604661, + "loss": 0.4685, + "step": 31736 + }, + { + "epoch": 1.7731932843534375, + "grad_norm": 0.5908941030502319, + "learning_rate": 0.00011438256387270282, + "loss": 0.417, + "step": 31737 + }, + { + "epoch": 1.7732491549571194, + "grad_norm": 0.41013091802597046, + "learning_rate": 0.00011435454952935903, + "loss": 0.4808, + "step": 31738 + }, + { + "epoch": 1.7733050255608012, + "grad_norm": 0.4739099144935608, + "learning_rate": 0.00011432653518601524, + "loss": 0.4787, + "step": 31739 + }, + { + "epoch": 1.773360896164483, + "grad_norm": 0.3674868941307068, + "learning_rate": 0.00011429852084267145, + "loss": 0.3513, + "step": 31740 + }, + { + "epoch": 1.773416766768165, + "grad_norm": 0.41431766748428345, + "learning_rate": 0.00011427050649932766, + "loss": 0.3655, + "step": 31741 + }, + { + "epoch": 1.7734726373718468, + "grad_norm": 0.488567978143692, + "learning_rate": 0.00011424249215598386, + "loss": 0.3666, + "step": 31742 + }, + { + "epoch": 1.7735285079755285, + "grad_norm": 0.36891400814056396, + "learning_rate": 0.00011421447781264008, + "loss": 0.4173, + "step": 31743 + }, + { + "epoch": 1.7735843785792107, + "grad_norm": 0.6898561120033264, + "learning_rate": 0.00011418646346929629, + "loss": 0.478, + "step": 31744 + }, + { + "epoch": 1.7736402491828924, + "grad_norm": 0.40063974261283875, + "learning_rate": 0.0001141584491259525, + "loss": 0.276, + "step": 31745 + }, + { + "epoch": 1.7736961197865742, + "grad_norm": 0.9469269514083862, + "learning_rate": 0.00011413043478260869, + "loss": 0.432, + "step": 31746 + }, + { + "epoch": 1.7737519903902563, + "grad_norm": 0.3833176791667938, + "learning_rate": 0.0001141024204392649, + "loss": 0.3208, + "step": 31747 + }, + { + "epoch": 1.773807860993938, + "grad_norm": 0.46304255723953247, + "learning_rate": 0.00011407440609592111, + "loss": 0.439, + "step": 31748 + }, + { + "epoch": 1.7738637315976198, + "grad_norm": 1.2286407947540283, + "learning_rate": 0.00011404639175257732, + "loss": 0.5221, + "step": 31749 + }, + { + "epoch": 1.7739196022013017, + "grad_norm": 0.5485319495201111, + "learning_rate": 0.00011401837740923352, + "loss": 0.4638, + "step": 31750 + }, + { + "epoch": 1.7739754728049837, + "grad_norm": 0.48378005623817444, + "learning_rate": 0.00011399036306588974, + "loss": 0.2925, + "step": 31751 + }, + { + "epoch": 1.7740313434086654, + "grad_norm": 0.5113338828086853, + "learning_rate": 0.00011396234872254595, + "loss": 0.4697, + "step": 31752 + }, + { + "epoch": 1.7740872140123474, + "grad_norm": 1.3017367124557495, + "learning_rate": 0.00011393433437920215, + "loss": 0.5613, + "step": 31753 + }, + { + "epoch": 1.7741430846160293, + "grad_norm": 0.433677077293396, + "learning_rate": 0.00011390632003585836, + "loss": 0.4595, + "step": 31754 + }, + { + "epoch": 1.774198955219711, + "grad_norm": 0.41856107115745544, + "learning_rate": 0.00011387830569251458, + "loss": 0.491, + "step": 31755 + }, + { + "epoch": 1.774254825823393, + "grad_norm": 0.4441746473312378, + "learning_rate": 0.00011385029134917078, + "loss": 0.398, + "step": 31756 + }, + { + "epoch": 1.774310696427075, + "grad_norm": 1.9961515665054321, + "learning_rate": 0.00011382227700582698, + "loss": 0.3581, + "step": 31757 + }, + { + "epoch": 1.7743665670307567, + "grad_norm": 0.6169465184211731, + "learning_rate": 0.00011379426266248318, + "loss": 0.5452, + "step": 31758 + }, + { + "epoch": 1.7744224376344386, + "grad_norm": 1.0194287300109863, + "learning_rate": 0.0001137662483191394, + "loss": 0.498, + "step": 31759 + }, + { + "epoch": 1.7744783082381206, + "grad_norm": 5.451159954071045, + "learning_rate": 0.00011373823397579561, + "loss": 1.6422, + "step": 31760 + }, + { + "epoch": 1.7745341788418023, + "grad_norm": 0.42186596989631653, + "learning_rate": 0.00011371021963245181, + "loss": 0.3604, + "step": 31761 + }, + { + "epoch": 1.7745900494454843, + "grad_norm": 0.31005948781967163, + "learning_rate": 0.00011368220528910802, + "loss": 0.2938, + "step": 31762 + }, + { + "epoch": 1.7746459200491662, + "grad_norm": 0.6896096467971802, + "learning_rate": 0.00011365419094576424, + "loss": 0.4405, + "step": 31763 + }, + { + "epoch": 1.774701790652848, + "grad_norm": 0.3732941746711731, + "learning_rate": 0.00011362617660242044, + "loss": 0.4355, + "step": 31764 + }, + { + "epoch": 1.77475766125653, + "grad_norm": 0.37632232904434204, + "learning_rate": 0.00011359816225907665, + "loss": 0.3517, + "step": 31765 + }, + { + "epoch": 1.7748135318602118, + "grad_norm": 1.8993695974349976, + "learning_rate": 0.00011357014791573287, + "loss": 0.4017, + "step": 31766 + }, + { + "epoch": 1.7748694024638936, + "grad_norm": 0.7706785202026367, + "learning_rate": 0.00011354213357238908, + "loss": 0.3388, + "step": 31767 + }, + { + "epoch": 1.7749252730675755, + "grad_norm": 0.9398930668830872, + "learning_rate": 0.00011351411922904527, + "loss": 0.3429, + "step": 31768 + }, + { + "epoch": 1.7749811436712575, + "grad_norm": 0.48284587264060974, + "learning_rate": 0.00011348610488570147, + "loss": 0.4009, + "step": 31769 + }, + { + "epoch": 1.7750370142749392, + "grad_norm": 0.5172451734542847, + "learning_rate": 0.00011345809054235769, + "loss": 0.5027, + "step": 31770 + }, + { + "epoch": 1.7750928848786212, + "grad_norm": 0.5998033881187439, + "learning_rate": 0.0001134300761990139, + "loss": 0.3943, + "step": 31771 + }, + { + "epoch": 1.775148755482303, + "grad_norm": 0.7096021175384521, + "learning_rate": 0.0001134020618556701, + "loss": 0.4203, + "step": 31772 + }, + { + "epoch": 1.7752046260859848, + "grad_norm": 0.5440870523452759, + "learning_rate": 0.00011337404751232631, + "loss": 0.4497, + "step": 31773 + }, + { + "epoch": 1.7752604966896666, + "grad_norm": 0.3627096116542816, + "learning_rate": 0.00011334603316898253, + "loss": 0.395, + "step": 31774 + }, + { + "epoch": 1.7753163672933487, + "grad_norm": 0.3982057571411133, + "learning_rate": 0.00011331801882563874, + "loss": 0.453, + "step": 31775 + }, + { + "epoch": 1.7753722378970305, + "grad_norm": 0.8349748849868774, + "learning_rate": 0.00011329000448229494, + "loss": 0.4439, + "step": 31776 + }, + { + "epoch": 1.7754281085007122, + "grad_norm": 0.5437659025192261, + "learning_rate": 0.00011326199013895115, + "loss": 0.4669, + "step": 31777 + }, + { + "epoch": 1.7754839791043944, + "grad_norm": 3.8684706687927246, + "learning_rate": 0.00011323397579560735, + "loss": 0.3741, + "step": 31778 + }, + { + "epoch": 1.775539849708076, + "grad_norm": 0.8502748608589172, + "learning_rate": 0.00011320596145226356, + "loss": 0.4865, + "step": 31779 + }, + { + "epoch": 1.7755957203117578, + "grad_norm": 0.4395105242729187, + "learning_rate": 0.00011317794710891976, + "loss": 0.4815, + "step": 31780 + }, + { + "epoch": 1.77565159091544, + "grad_norm": 0.5489229559898376, + "learning_rate": 0.00011314993276557597, + "loss": 0.6621, + "step": 31781 + }, + { + "epoch": 1.7757074615191217, + "grad_norm": 0.4233723282814026, + "learning_rate": 0.00011312191842223219, + "loss": 0.4837, + "step": 31782 + }, + { + "epoch": 1.7757633321228035, + "grad_norm": 3.61214542388916, + "learning_rate": 0.0001130939040788884, + "loss": 0.4646, + "step": 31783 + }, + { + "epoch": 1.7758192027264854, + "grad_norm": 0.4576355516910553, + "learning_rate": 0.0001130658897355446, + "loss": 0.4345, + "step": 31784 + }, + { + "epoch": 1.7758750733301674, + "grad_norm": 0.5184494256973267, + "learning_rate": 0.00011303787539220081, + "loss": 0.3829, + "step": 31785 + }, + { + "epoch": 1.775930943933849, + "grad_norm": 0.40468496084213257, + "learning_rate": 0.00011300986104885703, + "loss": 0.3149, + "step": 31786 + }, + { + "epoch": 1.775986814537531, + "grad_norm": 0.29050880670547485, + "learning_rate": 0.00011298184670551323, + "loss": 0.3144, + "step": 31787 + }, + { + "epoch": 1.776042685141213, + "grad_norm": 0.4931797981262207, + "learning_rate": 0.00011295383236216944, + "loss": 0.3741, + "step": 31788 + }, + { + "epoch": 1.7760985557448947, + "grad_norm": 0.38349631428718567, + "learning_rate": 0.00011292581801882563, + "loss": 0.3357, + "step": 31789 + }, + { + "epoch": 1.7761544263485767, + "grad_norm": 2.4331724643707275, + "learning_rate": 0.00011289780367548185, + "loss": 0.398, + "step": 31790 + }, + { + "epoch": 1.7762102969522586, + "grad_norm": 0.3520265221595764, + "learning_rate": 0.00011286978933213806, + "loss": 0.4619, + "step": 31791 + }, + { + "epoch": 1.7762661675559404, + "grad_norm": 0.45104074478149414, + "learning_rate": 0.00011284177498879426, + "loss": 0.3323, + "step": 31792 + }, + { + "epoch": 1.7763220381596223, + "grad_norm": 0.6557282209396362, + "learning_rate": 0.00011281376064545047, + "loss": 0.6422, + "step": 31793 + }, + { + "epoch": 1.7763779087633043, + "grad_norm": 0.3189018964767456, + "learning_rate": 0.00011278574630210669, + "loss": 0.3287, + "step": 31794 + }, + { + "epoch": 1.776433779366986, + "grad_norm": 0.5483031868934631, + "learning_rate": 0.00011275773195876289, + "loss": 0.4712, + "step": 31795 + }, + { + "epoch": 1.776489649970668, + "grad_norm": 0.4054074287414551, + "learning_rate": 0.0001127297176154191, + "loss": 0.6704, + "step": 31796 + }, + { + "epoch": 1.77654552057435, + "grad_norm": 0.6087058782577515, + "learning_rate": 0.0001127017032720753, + "loss": 0.5131, + "step": 31797 + }, + { + "epoch": 1.7766013911780316, + "grad_norm": 0.7223394513130188, + "learning_rate": 0.00011267368892873152, + "loss": 0.4268, + "step": 31798 + }, + { + "epoch": 1.7766572617817136, + "grad_norm": 0.4823629856109619, + "learning_rate": 0.00011264567458538772, + "loss": 0.4586, + "step": 31799 + }, + { + "epoch": 1.7767131323853955, + "grad_norm": 0.7232621312141418, + "learning_rate": 0.00011261766024204392, + "loss": 0.4368, + "step": 31800 + }, + { + "epoch": 1.7767690029890773, + "grad_norm": 0.5657926797866821, + "learning_rate": 0.00011258964589870013, + "loss": 0.3455, + "step": 31801 + }, + { + "epoch": 1.7768248735927592, + "grad_norm": 0.5282644033432007, + "learning_rate": 0.00011256163155535635, + "loss": 0.408, + "step": 31802 + }, + { + "epoch": 1.7768807441964412, + "grad_norm": 0.5660207867622375, + "learning_rate": 0.00011253361721201255, + "loss": 0.276, + "step": 31803 + }, + { + "epoch": 1.7769366148001229, + "grad_norm": 0.6525065898895264, + "learning_rate": 0.00011250560286866876, + "loss": 0.3142, + "step": 31804 + }, + { + "epoch": 1.7769924854038048, + "grad_norm": 0.3083634376525879, + "learning_rate": 0.00011247758852532496, + "loss": 0.3919, + "step": 31805 + }, + { + "epoch": 1.7770483560074868, + "grad_norm": 2.630861520767212, + "learning_rate": 0.00011244957418198118, + "loss": 0.4277, + "step": 31806 + }, + { + "epoch": 1.7771042266111685, + "grad_norm": 0.4761141240596771, + "learning_rate": 0.00011242155983863739, + "loss": 0.4614, + "step": 31807 + }, + { + "epoch": 1.7771600972148502, + "grad_norm": 0.30322960019111633, + "learning_rate": 0.0001123935454952936, + "loss": 0.3944, + "step": 31808 + }, + { + "epoch": 1.7772159678185324, + "grad_norm": 0.5334864854812622, + "learning_rate": 0.0001123655311519498, + "loss": 0.4049, + "step": 31809 + }, + { + "epoch": 1.7772718384222141, + "grad_norm": 0.3090932071208954, + "learning_rate": 0.00011233751680860601, + "loss": 0.3536, + "step": 31810 + }, + { + "epoch": 1.7773277090258959, + "grad_norm": 0.48051324486732483, + "learning_rate": 0.00011230950246526221, + "loss": 0.3751, + "step": 31811 + }, + { + "epoch": 1.777383579629578, + "grad_norm": 0.6207534670829773, + "learning_rate": 0.00011228148812191842, + "loss": 0.4748, + "step": 31812 + }, + { + "epoch": 1.7774394502332598, + "grad_norm": 0.48022517561912537, + "learning_rate": 0.00011225347377857462, + "loss": 0.4974, + "step": 31813 + }, + { + "epoch": 1.7774953208369415, + "grad_norm": 0.5169448256492615, + "learning_rate": 0.00011222545943523084, + "loss": 0.383, + "step": 31814 + }, + { + "epoch": 1.7775511914406237, + "grad_norm": 0.4114236533641815, + "learning_rate": 0.00011219744509188705, + "loss": 0.4115, + "step": 31815 + }, + { + "epoch": 1.7776070620443054, + "grad_norm": 0.392217755317688, + "learning_rate": 0.00011216943074854326, + "loss": 0.3699, + "step": 31816 + }, + { + "epoch": 1.7776629326479871, + "grad_norm": 0.4209884703159332, + "learning_rate": 0.00011214141640519946, + "loss": 0.3779, + "step": 31817 + }, + { + "epoch": 1.777718803251669, + "grad_norm": 0.4301619529724121, + "learning_rate": 0.00011211340206185568, + "loss": 0.4715, + "step": 31818 + }, + { + "epoch": 1.777774673855351, + "grad_norm": 0.5358754992485046, + "learning_rate": 0.00011208538771851189, + "loss": 0.3786, + "step": 31819 + }, + { + "epoch": 1.7778305444590328, + "grad_norm": 0.5606170892715454, + "learning_rate": 0.00011205737337516809, + "loss": 0.3298, + "step": 31820 + }, + { + "epoch": 1.7778864150627147, + "grad_norm": 0.7630932927131653, + "learning_rate": 0.00011202935903182428, + "loss": 0.532, + "step": 31821 + }, + { + "epoch": 1.7779422856663967, + "grad_norm": 0.6160228252410889, + "learning_rate": 0.0001120013446884805, + "loss": 0.3518, + "step": 31822 + }, + { + "epoch": 1.7779981562700784, + "grad_norm": 0.37923577427864075, + "learning_rate": 0.00011197333034513671, + "loss": 0.4097, + "step": 31823 + }, + { + "epoch": 1.7780540268737604, + "grad_norm": 0.33824437856674194, + "learning_rate": 0.00011194531600179292, + "loss": 0.402, + "step": 31824 + }, + { + "epoch": 1.7781098974774423, + "grad_norm": 0.27244317531585693, + "learning_rate": 0.00011191730165844912, + "loss": 0.4287, + "step": 31825 + }, + { + "epoch": 1.778165768081124, + "grad_norm": 0.6458238959312439, + "learning_rate": 0.00011188928731510534, + "loss": 0.4503, + "step": 31826 + }, + { + "epoch": 1.778221638684806, + "grad_norm": 3.7619190216064453, + "learning_rate": 0.00011186127297176155, + "loss": 0.3353, + "step": 31827 + }, + { + "epoch": 1.778277509288488, + "grad_norm": 0.6125068664550781, + "learning_rate": 0.00011183325862841775, + "loss": 0.4437, + "step": 31828 + }, + { + "epoch": 1.7783333798921697, + "grad_norm": 0.3175153434276581, + "learning_rate": 0.00011180524428507396, + "loss": 0.4122, + "step": 31829 + }, + { + "epoch": 1.7783892504958516, + "grad_norm": 0.4227522015571594, + "learning_rate": 0.00011177722994173018, + "loss": 0.4306, + "step": 31830 + }, + { + "epoch": 1.7784451210995336, + "grad_norm": 0.5177589654922485, + "learning_rate": 0.00011174921559838637, + "loss": 0.4335, + "step": 31831 + }, + { + "epoch": 1.7785009917032153, + "grad_norm": 0.4541335999965668, + "learning_rate": 0.00011172120125504258, + "loss": 0.4418, + "step": 31832 + }, + { + "epoch": 1.7785568623068972, + "grad_norm": 0.3651807904243469, + "learning_rate": 0.00011169318691169878, + "loss": 0.3423, + "step": 31833 + }, + { + "epoch": 1.7786127329105792, + "grad_norm": 1.404164433479309, + "learning_rate": 0.000111665172568355, + "loss": 0.414, + "step": 31834 + }, + { + "epoch": 1.778668603514261, + "grad_norm": 0.37288472056388855, + "learning_rate": 0.0001116371582250112, + "loss": 0.3899, + "step": 31835 + }, + { + "epoch": 1.7787244741179429, + "grad_norm": 0.6991906762123108, + "learning_rate": 0.00011160914388166741, + "loss": 0.5764, + "step": 31836 + }, + { + "epoch": 1.7787803447216248, + "grad_norm": 0.4090649485588074, + "learning_rate": 0.00011158112953832363, + "loss": 0.5185, + "step": 31837 + }, + { + "epoch": 1.7788362153253066, + "grad_norm": 0.5197237730026245, + "learning_rate": 0.00011155311519497984, + "loss": 0.4562, + "step": 31838 + }, + { + "epoch": 1.7788920859289885, + "grad_norm": 0.4392740726470947, + "learning_rate": 0.00011152510085163604, + "loss": 0.3543, + "step": 31839 + }, + { + "epoch": 1.7789479565326705, + "grad_norm": 0.3860919773578644, + "learning_rate": 0.00011149708650829225, + "loss": 0.4031, + "step": 31840 + }, + { + "epoch": 1.7790038271363522, + "grad_norm": 0.5061669945716858, + "learning_rate": 0.00011146907216494847, + "loss": 0.5155, + "step": 31841 + }, + { + "epoch": 1.779059697740034, + "grad_norm": 0.41858598589897156, + "learning_rate": 0.00011144105782160466, + "loss": 0.3851, + "step": 31842 + }, + { + "epoch": 1.779115568343716, + "grad_norm": 0.7321102619171143, + "learning_rate": 0.00011141304347826087, + "loss": 0.3797, + "step": 31843 + }, + { + "epoch": 1.7791714389473978, + "grad_norm": 0.34044957160949707, + "learning_rate": 0.00011138502913491707, + "loss": 0.3659, + "step": 31844 + }, + { + "epoch": 1.7792273095510795, + "grad_norm": 0.5520982146263123, + "learning_rate": 0.00011135701479157329, + "loss": 0.4347, + "step": 31845 + }, + { + "epoch": 1.7792831801547617, + "grad_norm": 0.416167676448822, + "learning_rate": 0.0001113290004482295, + "loss": 0.3599, + "step": 31846 + }, + { + "epoch": 1.7793390507584435, + "grad_norm": 0.3887068033218384, + "learning_rate": 0.0001113009861048857, + "loss": 0.4553, + "step": 31847 + }, + { + "epoch": 1.7793949213621252, + "grad_norm": 4.541413307189941, + "learning_rate": 0.00011127297176154191, + "loss": 0.3656, + "step": 31848 + }, + { + "epoch": 1.7794507919658074, + "grad_norm": 0.5508867502212524, + "learning_rate": 0.00011124495741819813, + "loss": 0.5065, + "step": 31849 + }, + { + "epoch": 1.779506662569489, + "grad_norm": 0.5339690446853638, + "learning_rate": 0.00011121694307485433, + "loss": 0.4105, + "step": 31850 + }, + { + "epoch": 1.7795625331731708, + "grad_norm": 0.3903704881668091, + "learning_rate": 0.00011118892873151054, + "loss": 0.4118, + "step": 31851 + }, + { + "epoch": 1.7796184037768528, + "grad_norm": 0.39759427309036255, + "learning_rate": 0.00011116091438816675, + "loss": 0.38, + "step": 31852 + }, + { + "epoch": 1.7796742743805347, + "grad_norm": 0.3081064522266388, + "learning_rate": 0.00011113290004482295, + "loss": 0.2944, + "step": 31853 + }, + { + "epoch": 1.7797301449842164, + "grad_norm": 4.836397171020508, + "learning_rate": 0.00011110488570147916, + "loss": 0.3265, + "step": 31854 + }, + { + "epoch": 1.7797860155878984, + "grad_norm": 0.3572268486022949, + "learning_rate": 0.00011107687135813536, + "loss": 0.3863, + "step": 31855 + }, + { + "epoch": 1.7798418861915803, + "grad_norm": 0.5670017004013062, + "learning_rate": 0.00011104885701479157, + "loss": 0.7114, + "step": 31856 + }, + { + "epoch": 1.779897756795262, + "grad_norm": 0.6199503540992737, + "learning_rate": 0.00011102084267144779, + "loss": 0.3975, + "step": 31857 + }, + { + "epoch": 1.779953627398944, + "grad_norm": 0.4152108132839203, + "learning_rate": 0.000110992828328104, + "loss": 0.395, + "step": 31858 + }, + { + "epoch": 1.780009498002626, + "grad_norm": 3.387744903564453, + "learning_rate": 0.0001109648139847602, + "loss": 0.5186, + "step": 31859 + }, + { + "epoch": 1.7800653686063077, + "grad_norm": 0.4037277400493622, + "learning_rate": 0.0001109367996414164, + "loss": 0.4381, + "step": 31860 + }, + { + "epoch": 1.7801212392099897, + "grad_norm": 0.3446878492832184, + "learning_rate": 0.00011090878529807263, + "loss": 0.4367, + "step": 31861 + }, + { + "epoch": 1.7801771098136716, + "grad_norm": 0.3700981140136719, + "learning_rate": 0.00011088077095472883, + "loss": 0.4143, + "step": 31862 + }, + { + "epoch": 1.7802329804173533, + "grad_norm": 0.9800706505775452, + "learning_rate": 0.00011085275661138502, + "loss": 0.3551, + "step": 31863 + }, + { + "epoch": 1.7802888510210353, + "grad_norm": 1.3831342458724976, + "learning_rate": 0.00011082474226804123, + "loss": 0.4117, + "step": 31864 + }, + { + "epoch": 1.7803447216247172, + "grad_norm": 0.49850234389305115, + "learning_rate": 0.00011079672792469745, + "loss": 0.4533, + "step": 31865 + }, + { + "epoch": 1.780400592228399, + "grad_norm": 0.41767504811286926, + "learning_rate": 0.00011076871358135365, + "loss": 0.4797, + "step": 31866 + }, + { + "epoch": 1.780456462832081, + "grad_norm": 0.4146014451980591, + "learning_rate": 0.00011074069923800986, + "loss": 0.5318, + "step": 31867 + }, + { + "epoch": 1.7805123334357629, + "grad_norm": 0.43716514110565186, + "learning_rate": 0.00011071268489466607, + "loss": 0.4115, + "step": 31868 + }, + { + "epoch": 1.7805682040394446, + "grad_norm": 0.9697539210319519, + "learning_rate": 0.00011068467055132229, + "loss": 0.3663, + "step": 31869 + }, + { + "epoch": 1.7806240746431266, + "grad_norm": 0.45710644125938416, + "learning_rate": 0.00011065665620797849, + "loss": 0.3589, + "step": 31870 + }, + { + "epoch": 1.7806799452468085, + "grad_norm": 0.6186583638191223, + "learning_rate": 0.0001106286418646347, + "loss": 0.3504, + "step": 31871 + }, + { + "epoch": 1.7807358158504902, + "grad_norm": 0.573801577091217, + "learning_rate": 0.0001106006275212909, + "loss": 0.4212, + "step": 31872 + }, + { + "epoch": 1.7807916864541722, + "grad_norm": 1.252007007598877, + "learning_rate": 0.00011057261317794712, + "loss": 0.3612, + "step": 31873 + }, + { + "epoch": 1.7808475570578541, + "grad_norm": 0.39715322852134705, + "learning_rate": 0.00011054459883460331, + "loss": 0.48, + "step": 31874 + }, + { + "epoch": 1.7809034276615359, + "grad_norm": 0.41652989387512207, + "learning_rate": 0.00011051658449125952, + "loss": 0.361, + "step": 31875 + }, + { + "epoch": 1.7809592982652176, + "grad_norm": 0.5939876437187195, + "learning_rate": 0.00011048857014791573, + "loss": 0.3131, + "step": 31876 + }, + { + "epoch": 1.7810151688688998, + "grad_norm": 0.6449661254882812, + "learning_rate": 0.00011046055580457195, + "loss": 0.4788, + "step": 31877 + }, + { + "epoch": 1.7810710394725815, + "grad_norm": 2.8064770698547363, + "learning_rate": 0.00011043254146122815, + "loss": 0.4475, + "step": 31878 + }, + { + "epoch": 1.7811269100762632, + "grad_norm": 0.34014955163002014, + "learning_rate": 0.00011040452711788436, + "loss": 0.4172, + "step": 31879 + }, + { + "epoch": 1.7811827806799454, + "grad_norm": 0.4893435537815094, + "learning_rate": 0.00011037651277454056, + "loss": 0.4252, + "step": 31880 + }, + { + "epoch": 1.7812386512836271, + "grad_norm": 0.5105757713317871, + "learning_rate": 0.00011034849843119678, + "loss": 0.3357, + "step": 31881 + }, + { + "epoch": 1.7812945218873089, + "grad_norm": 0.45656949281692505, + "learning_rate": 0.00011032048408785299, + "loss": 0.3931, + "step": 31882 + }, + { + "epoch": 1.781350392490991, + "grad_norm": 0.35271233320236206, + "learning_rate": 0.0001102924697445092, + "loss": 0.3414, + "step": 31883 + }, + { + "epoch": 1.7814062630946728, + "grad_norm": 0.6800322532653809, + "learning_rate": 0.00011026445540116539, + "loss": 0.5084, + "step": 31884 + }, + { + "epoch": 1.7814621336983545, + "grad_norm": 2.627056360244751, + "learning_rate": 0.0001102364410578216, + "loss": 0.414, + "step": 31885 + }, + { + "epoch": 1.7815180043020364, + "grad_norm": 0.6791313290596008, + "learning_rate": 0.00011020842671447781, + "loss": 0.5242, + "step": 31886 + }, + { + "epoch": 1.7815738749057184, + "grad_norm": 0.8669683933258057, + "learning_rate": 0.00011018041237113402, + "loss": 0.3357, + "step": 31887 + }, + { + "epoch": 1.7816297455094001, + "grad_norm": 0.4763660728931427, + "learning_rate": 0.00011015239802779022, + "loss": 0.3609, + "step": 31888 + }, + { + "epoch": 1.781685616113082, + "grad_norm": 0.3987993001937866, + "learning_rate": 0.00011012438368444644, + "loss": 0.3495, + "step": 31889 + }, + { + "epoch": 1.781741486716764, + "grad_norm": 0.5251256227493286, + "learning_rate": 0.00011009636934110265, + "loss": 0.4629, + "step": 31890 + }, + { + "epoch": 1.7817973573204458, + "grad_norm": 0.8872851729393005, + "learning_rate": 0.00011006835499775885, + "loss": 0.3924, + "step": 31891 + }, + { + "epoch": 1.7818532279241277, + "grad_norm": 0.38813599944114685, + "learning_rate": 0.00011004034065441506, + "loss": 0.3513, + "step": 31892 + }, + { + "epoch": 1.7819090985278097, + "grad_norm": 0.48886018991470337, + "learning_rate": 0.00011001232631107128, + "loss": 0.5518, + "step": 31893 + }, + { + "epoch": 1.7819649691314914, + "grad_norm": 0.6169275045394897, + "learning_rate": 0.00010998431196772749, + "loss": 0.4553, + "step": 31894 + }, + { + "epoch": 1.7820208397351733, + "grad_norm": 0.5943176746368408, + "learning_rate": 0.00010995629762438368, + "loss": 0.5127, + "step": 31895 + }, + { + "epoch": 1.7820767103388553, + "grad_norm": 0.3096387982368469, + "learning_rate": 0.00010992828328103988, + "loss": 0.3577, + "step": 31896 + }, + { + "epoch": 1.782132580942537, + "grad_norm": 0.44368621706962585, + "learning_rate": 0.0001099002689376961, + "loss": 0.3, + "step": 31897 + }, + { + "epoch": 1.782188451546219, + "grad_norm": 0.524898111820221, + "learning_rate": 0.00010987225459435231, + "loss": 0.4707, + "step": 31898 + }, + { + "epoch": 1.782244322149901, + "grad_norm": 0.46815821528434753, + "learning_rate": 0.00010984424025100851, + "loss": 0.5243, + "step": 31899 + }, + { + "epoch": 1.7823001927535826, + "grad_norm": 0.43404456973075867, + "learning_rate": 0.00010981622590766473, + "loss": 0.5452, + "step": 31900 + }, + { + "epoch": 1.7823560633572646, + "grad_norm": 0.3207572102546692, + "learning_rate": 0.00010978821156432094, + "loss": 0.3603, + "step": 31901 + }, + { + "epoch": 1.7824119339609465, + "grad_norm": 0.4603421092033386, + "learning_rate": 0.00010976019722097715, + "loss": 0.6707, + "step": 31902 + }, + { + "epoch": 1.7824678045646283, + "grad_norm": 0.2889627516269684, + "learning_rate": 0.00010973218287763335, + "loss": 0.421, + "step": 31903 + }, + { + "epoch": 1.7825236751683102, + "grad_norm": 0.7290984392166138, + "learning_rate": 0.00010970416853428957, + "loss": 0.4342, + "step": 31904 + }, + { + "epoch": 1.7825795457719922, + "grad_norm": 0.4319707453250885, + "learning_rate": 0.00010967615419094578, + "loss": 0.3939, + "step": 31905 + }, + { + "epoch": 1.782635416375674, + "grad_norm": 0.6146321296691895, + "learning_rate": 0.00010964813984760197, + "loss": 0.4338, + "step": 31906 + }, + { + "epoch": 1.7826912869793559, + "grad_norm": 0.3184428811073303, + "learning_rate": 0.00010962012550425817, + "loss": 0.3611, + "step": 31907 + }, + { + "epoch": 1.7827471575830378, + "grad_norm": 0.40346407890319824, + "learning_rate": 0.0001095921111609144, + "loss": 0.4407, + "step": 31908 + }, + { + "epoch": 1.7828030281867195, + "grad_norm": 1.2555514574050903, + "learning_rate": 0.0001095640968175706, + "loss": 0.4166, + "step": 31909 + }, + { + "epoch": 1.7828588987904013, + "grad_norm": 0.33813804388046265, + "learning_rate": 0.0001095360824742268, + "loss": 0.356, + "step": 31910 + }, + { + "epoch": 1.7829147693940834, + "grad_norm": 0.387018084526062, + "learning_rate": 0.00010950806813088301, + "loss": 0.4036, + "step": 31911 + }, + { + "epoch": 1.7829706399977652, + "grad_norm": 0.5024418234825134, + "learning_rate": 0.00010948005378753923, + "loss": 0.4198, + "step": 31912 + }, + { + "epoch": 1.783026510601447, + "grad_norm": 0.49607551097869873, + "learning_rate": 0.00010945203944419544, + "loss": 0.4705, + "step": 31913 + }, + { + "epoch": 1.783082381205129, + "grad_norm": 0.39560651779174805, + "learning_rate": 0.00010942402510085164, + "loss": 0.3902, + "step": 31914 + }, + { + "epoch": 1.7831382518088108, + "grad_norm": 0.38461801409721375, + "learning_rate": 0.00010939601075750785, + "loss": 0.3311, + "step": 31915 + }, + { + "epoch": 1.7831941224124925, + "grad_norm": 0.3885178864002228, + "learning_rate": 0.00010936799641416405, + "loss": 0.4819, + "step": 31916 + }, + { + "epoch": 1.7832499930161747, + "grad_norm": 0.9761415123939514, + "learning_rate": 0.00010933998207082026, + "loss": 0.3359, + "step": 31917 + }, + { + "epoch": 1.7833058636198564, + "grad_norm": 0.4004380404949188, + "learning_rate": 0.00010931196772747647, + "loss": 0.3842, + "step": 31918 + }, + { + "epoch": 1.7833617342235382, + "grad_norm": 0.306804895401001, + "learning_rate": 0.00010928395338413267, + "loss": 0.2987, + "step": 31919 + }, + { + "epoch": 1.7834176048272201, + "grad_norm": 0.35403475165367126, + "learning_rate": 0.00010925593904078889, + "loss": 0.4374, + "step": 31920 + }, + { + "epoch": 1.783473475430902, + "grad_norm": 0.5072659254074097, + "learning_rate": 0.0001092279246974451, + "loss": 0.3047, + "step": 31921 + }, + { + "epoch": 1.7835293460345838, + "grad_norm": 0.46396780014038086, + "learning_rate": 0.0001091999103541013, + "loss": 0.3607, + "step": 31922 + }, + { + "epoch": 1.7835852166382657, + "grad_norm": 0.44501185417175293, + "learning_rate": 0.00010917189601075751, + "loss": 0.4047, + "step": 31923 + }, + { + "epoch": 1.7836410872419477, + "grad_norm": 0.3247006833553314, + "learning_rate": 0.00010914388166741373, + "loss": 0.2961, + "step": 31924 + }, + { + "epoch": 1.7836969578456294, + "grad_norm": 2.9180445671081543, + "learning_rate": 0.00010911586732406993, + "loss": 0.3593, + "step": 31925 + }, + { + "epoch": 1.7837528284493114, + "grad_norm": 0.49537232518196106, + "learning_rate": 0.00010908785298072614, + "loss": 0.3635, + "step": 31926 + }, + { + "epoch": 1.7838086990529933, + "grad_norm": 0.4129500091075897, + "learning_rate": 0.00010905983863738233, + "loss": 0.3506, + "step": 31927 + }, + { + "epoch": 1.783864569656675, + "grad_norm": 0.5484176874160767, + "learning_rate": 0.00010903182429403855, + "loss": 0.3792, + "step": 31928 + }, + { + "epoch": 1.783920440260357, + "grad_norm": 0.3755978047847748, + "learning_rate": 0.00010900380995069476, + "loss": 0.4207, + "step": 31929 + }, + { + "epoch": 1.783976310864039, + "grad_norm": 0.7262332439422607, + "learning_rate": 0.00010897579560735096, + "loss": 0.5145, + "step": 31930 + }, + { + "epoch": 1.7840321814677207, + "grad_norm": 0.3516658842563629, + "learning_rate": 0.00010894778126400717, + "loss": 0.404, + "step": 31931 + }, + { + "epoch": 1.7840880520714026, + "grad_norm": 0.43998685479164124, + "learning_rate": 0.00010891976692066339, + "loss": 0.4885, + "step": 31932 + }, + { + "epoch": 1.7841439226750846, + "grad_norm": 0.5354444980621338, + "learning_rate": 0.0001088917525773196, + "loss": 0.3996, + "step": 31933 + }, + { + "epoch": 1.7841997932787663, + "grad_norm": 0.6322029829025269, + "learning_rate": 0.0001088637382339758, + "loss": 0.4746, + "step": 31934 + }, + { + "epoch": 1.7842556638824483, + "grad_norm": 1.400020956993103, + "learning_rate": 0.000108835723890632, + "loss": 0.4321, + "step": 31935 + }, + { + "epoch": 1.7843115344861302, + "grad_norm": 0.40189826488494873, + "learning_rate": 0.00010880770954728822, + "loss": 0.4331, + "step": 31936 + }, + { + "epoch": 1.784367405089812, + "grad_norm": 0.36170342564582825, + "learning_rate": 0.00010877969520394442, + "loss": 0.4297, + "step": 31937 + }, + { + "epoch": 1.784423275693494, + "grad_norm": 1.6235872507095337, + "learning_rate": 0.00010875168086060062, + "loss": 0.6184, + "step": 31938 + }, + { + "epoch": 1.7844791462971759, + "grad_norm": 0.570926308631897, + "learning_rate": 0.00010872366651725683, + "loss": 0.4028, + "step": 31939 + }, + { + "epoch": 1.7845350169008576, + "grad_norm": 1.5796923637390137, + "learning_rate": 0.00010869565217391305, + "loss": 0.431, + "step": 31940 + }, + { + "epoch": 1.7845908875045395, + "grad_norm": 0.5853471159934998, + "learning_rate": 0.00010866763783056925, + "loss": 0.4079, + "step": 31941 + }, + { + "epoch": 1.7846467581082215, + "grad_norm": 0.9625852108001709, + "learning_rate": 0.00010863962348722546, + "loss": 0.4201, + "step": 31942 + }, + { + "epoch": 1.7847026287119032, + "grad_norm": 0.5667279362678528, + "learning_rate": 0.00010861160914388166, + "loss": 0.6244, + "step": 31943 + }, + { + "epoch": 1.784758499315585, + "grad_norm": 0.538985550403595, + "learning_rate": 0.00010858359480053788, + "loss": 0.3877, + "step": 31944 + }, + { + "epoch": 1.7848143699192671, + "grad_norm": 0.4323638081550598, + "learning_rate": 0.00010855558045719409, + "loss": 0.3807, + "step": 31945 + }, + { + "epoch": 1.7848702405229488, + "grad_norm": 0.3603339195251465, + "learning_rate": 0.0001085275661138503, + "loss": 0.3785, + "step": 31946 + }, + { + "epoch": 1.7849261111266306, + "grad_norm": 0.41169390082359314, + "learning_rate": 0.0001084995517705065, + "loss": 0.382, + "step": 31947 + }, + { + "epoch": 1.7849819817303128, + "grad_norm": 0.4309064447879791, + "learning_rate": 0.00010847153742716271, + "loss": 0.4425, + "step": 31948 + }, + { + "epoch": 1.7850378523339945, + "grad_norm": 0.4490073621273041, + "learning_rate": 0.00010844352308381891, + "loss": 0.4376, + "step": 31949 + }, + { + "epoch": 1.7850937229376762, + "grad_norm": 0.4443066716194153, + "learning_rate": 0.00010841550874047512, + "loss": 0.3961, + "step": 31950 + }, + { + "epoch": 1.7851495935413582, + "grad_norm": 0.4093339145183563, + "learning_rate": 0.00010838749439713133, + "loss": 0.419, + "step": 31951 + }, + { + "epoch": 1.78520546414504, + "grad_norm": 0.37975025177001953, + "learning_rate": 0.00010835948005378754, + "loss": 0.338, + "step": 31952 + }, + { + "epoch": 1.7852613347487218, + "grad_norm": 0.29440149664878845, + "learning_rate": 0.00010833146571044375, + "loss": 0.3381, + "step": 31953 + }, + { + "epoch": 1.7853172053524038, + "grad_norm": 2.0091588497161865, + "learning_rate": 0.00010830345136709996, + "loss": 0.5988, + "step": 31954 + }, + { + "epoch": 1.7853730759560857, + "grad_norm": 0.6659116744995117, + "learning_rate": 0.00010827543702375616, + "loss": 0.3968, + "step": 31955 + }, + { + "epoch": 1.7854289465597675, + "grad_norm": 2.1289467811584473, + "learning_rate": 0.00010824742268041238, + "loss": 0.4925, + "step": 31956 + }, + { + "epoch": 1.7854848171634494, + "grad_norm": 0.40623611211776733, + "learning_rate": 0.00010821940833706859, + "loss": 0.3802, + "step": 31957 + }, + { + "epoch": 1.7855406877671314, + "grad_norm": 0.3584878444671631, + "learning_rate": 0.00010819139399372479, + "loss": 0.3405, + "step": 31958 + }, + { + "epoch": 1.785596558370813, + "grad_norm": 0.4514928460121155, + "learning_rate": 0.00010816337965038099, + "loss": 0.4857, + "step": 31959 + }, + { + "epoch": 1.785652428974495, + "grad_norm": 0.3766014277935028, + "learning_rate": 0.0001081353653070372, + "loss": 0.4584, + "step": 31960 + }, + { + "epoch": 1.785708299578177, + "grad_norm": 0.46605050563812256, + "learning_rate": 0.00010810735096369341, + "loss": 0.4468, + "step": 31961 + }, + { + "epoch": 1.7857641701818587, + "grad_norm": 0.43640050292015076, + "learning_rate": 0.00010807933662034962, + "loss": 0.3934, + "step": 31962 + }, + { + "epoch": 1.7858200407855407, + "grad_norm": 0.33457300066947937, + "learning_rate": 0.00010805132227700582, + "loss": 0.4611, + "step": 31963 + }, + { + "epoch": 1.7858759113892226, + "grad_norm": 0.552971601486206, + "learning_rate": 0.00010802330793366204, + "loss": 0.3836, + "step": 31964 + }, + { + "epoch": 1.7859317819929044, + "grad_norm": 0.4639107584953308, + "learning_rate": 0.00010799529359031825, + "loss": 0.3771, + "step": 31965 + }, + { + "epoch": 1.7859876525965863, + "grad_norm": 0.41205456852912903, + "learning_rate": 0.00010796727924697445, + "loss": 0.4223, + "step": 31966 + }, + { + "epoch": 1.7860435232002683, + "grad_norm": 0.7880639433860779, + "learning_rate": 0.00010793926490363067, + "loss": 0.3785, + "step": 31967 + }, + { + "epoch": 1.78609939380395, + "grad_norm": 0.4735778570175171, + "learning_rate": 0.00010791125056028688, + "loss": 0.4127, + "step": 31968 + }, + { + "epoch": 1.786155264407632, + "grad_norm": 0.628206729888916, + "learning_rate": 0.00010788323621694307, + "loss": 0.3383, + "step": 31969 + }, + { + "epoch": 1.786211135011314, + "grad_norm": 2.772184133529663, + "learning_rate": 0.00010785522187359928, + "loss": 0.4015, + "step": 31970 + }, + { + "epoch": 1.7862670056149956, + "grad_norm": 0.2842792570590973, + "learning_rate": 0.0001078272075302555, + "loss": 0.3204, + "step": 31971 + }, + { + "epoch": 1.7863228762186776, + "grad_norm": 0.42456793785095215, + "learning_rate": 0.0001077991931869117, + "loss": 0.543, + "step": 31972 + }, + { + "epoch": 1.7863787468223595, + "grad_norm": 0.6058681011199951, + "learning_rate": 0.00010777117884356791, + "loss": 0.3932, + "step": 31973 + }, + { + "epoch": 1.7864346174260413, + "grad_norm": 0.5790477991104126, + "learning_rate": 0.00010774316450022411, + "loss": 0.4286, + "step": 31974 + }, + { + "epoch": 1.7864904880297232, + "grad_norm": 0.8476322889328003, + "learning_rate": 0.00010771515015688033, + "loss": 0.4949, + "step": 31975 + }, + { + "epoch": 1.7865463586334052, + "grad_norm": 0.48624637722969055, + "learning_rate": 0.00010768713581353654, + "loss": 0.4296, + "step": 31976 + }, + { + "epoch": 1.786602229237087, + "grad_norm": 0.48026835918426514, + "learning_rate": 0.00010765912147019274, + "loss": 0.4299, + "step": 31977 + }, + { + "epoch": 1.7866580998407686, + "grad_norm": 0.5867374539375305, + "learning_rate": 0.00010763110712684895, + "loss": 0.3952, + "step": 31978 + }, + { + "epoch": 1.7867139704444508, + "grad_norm": 0.34508803486824036, + "learning_rate": 0.00010760309278350517, + "loss": 0.4221, + "step": 31979 + }, + { + "epoch": 1.7867698410481325, + "grad_norm": 0.45410504937171936, + "learning_rate": 0.00010757507844016136, + "loss": 0.442, + "step": 31980 + }, + { + "epoch": 1.7868257116518143, + "grad_norm": 0.48941510915756226, + "learning_rate": 0.00010754706409681757, + "loss": 0.4175, + "step": 31981 + }, + { + "epoch": 1.7868815822554964, + "grad_norm": 0.5052236914634705, + "learning_rate": 0.00010751904975347377, + "loss": 0.4594, + "step": 31982 + }, + { + "epoch": 1.7869374528591782, + "grad_norm": 0.40317535400390625, + "learning_rate": 0.00010749103541012999, + "loss": 0.3589, + "step": 31983 + }, + { + "epoch": 1.7869933234628599, + "grad_norm": 0.35177990794181824, + "learning_rate": 0.0001074630210667862, + "loss": 0.3895, + "step": 31984 + }, + { + "epoch": 1.7870491940665418, + "grad_norm": 0.36725524067878723, + "learning_rate": 0.0001074350067234424, + "loss": 0.4687, + "step": 31985 + }, + { + "epoch": 1.7871050646702238, + "grad_norm": 0.44265973567962646, + "learning_rate": 0.00010740699238009861, + "loss": 0.4738, + "step": 31986 + }, + { + "epoch": 1.7871609352739055, + "grad_norm": 0.38976311683654785, + "learning_rate": 0.00010737897803675483, + "loss": 0.3851, + "step": 31987 + }, + { + "epoch": 1.7872168058775875, + "grad_norm": 0.35776421427726746, + "learning_rate": 0.00010735096369341104, + "loss": 0.3716, + "step": 31988 + }, + { + "epoch": 1.7872726764812694, + "grad_norm": 0.4909263551235199, + "learning_rate": 0.00010732294935006724, + "loss": 0.426, + "step": 31989 + }, + { + "epoch": 1.7873285470849511, + "grad_norm": 4.2328009605407715, + "learning_rate": 0.00010729493500672345, + "loss": 0.48, + "step": 31990 + }, + { + "epoch": 1.787384417688633, + "grad_norm": 0.6384636163711548, + "learning_rate": 0.00010726692066337965, + "loss": 0.4611, + "step": 31991 + }, + { + "epoch": 1.787440288292315, + "grad_norm": 0.3296359181404114, + "learning_rate": 0.00010723890632003586, + "loss": 0.3244, + "step": 31992 + }, + { + "epoch": 1.7874961588959968, + "grad_norm": 0.4441543519496918, + "learning_rate": 0.00010721089197669206, + "loss": 0.4648, + "step": 31993 + }, + { + "epoch": 1.7875520294996787, + "grad_norm": 0.9662633538246155, + "learning_rate": 0.00010718287763334827, + "loss": 0.3945, + "step": 31994 + }, + { + "epoch": 1.7876079001033607, + "grad_norm": 0.432964950799942, + "learning_rate": 0.00010715486329000449, + "loss": 0.4387, + "step": 31995 + }, + { + "epoch": 1.7876637707070424, + "grad_norm": 0.8183473944664001, + "learning_rate": 0.0001071268489466607, + "loss": 0.4265, + "step": 31996 + }, + { + "epoch": 1.7877196413107244, + "grad_norm": 23.03973960876465, + "learning_rate": 0.0001070988346033169, + "loss": 0.4131, + "step": 31997 + }, + { + "epoch": 1.7877755119144063, + "grad_norm": 0.39433589577674866, + "learning_rate": 0.0001070708202599731, + "loss": 0.3522, + "step": 31998 + }, + { + "epoch": 1.787831382518088, + "grad_norm": 11.72885799407959, + "learning_rate": 0.00010704280591662933, + "loss": 0.3681, + "step": 31999 + }, + { + "epoch": 1.78788725312177, + "grad_norm": 0.4352031350135803, + "learning_rate": 0.00010701479157328553, + "loss": 0.3691, + "step": 32000 + }, + { + "epoch": 1.78788725312177, + "eval_cer": 0.08355973072348968, + "eval_loss": 0.3140832483768463, + "eval_runtime": 56.2552, + "eval_samples_per_second": 80.668, + "eval_steps_per_second": 5.048, + "eval_wer": 0.3313035948716516, + "step": 32000 + }, + { + "epoch": 1.787943123725452, + "grad_norm": 0.5188249349594116, + "learning_rate": 0.00010698677722994172, + "loss": 0.3544, + "step": 32001 + }, + { + "epoch": 1.7879989943291337, + "grad_norm": 1.071302056312561, + "learning_rate": 0.00010695876288659793, + "loss": 0.5094, + "step": 32002 + }, + { + "epoch": 1.7880548649328156, + "grad_norm": 1.2806874513626099, + "learning_rate": 0.00010693074854325415, + "loss": 0.4327, + "step": 32003 + }, + { + "epoch": 1.7881107355364976, + "grad_norm": 2.129901647567749, + "learning_rate": 0.00010690273419991036, + "loss": 0.4327, + "step": 32004 + }, + { + "epoch": 1.7881666061401793, + "grad_norm": 0.39618897438049316, + "learning_rate": 0.00010687471985656656, + "loss": 0.5232, + "step": 32005 + }, + { + "epoch": 1.7882224767438613, + "grad_norm": 0.4538940191268921, + "learning_rate": 0.00010684670551322277, + "loss": 0.4115, + "step": 32006 + }, + { + "epoch": 1.7882783473475432, + "grad_norm": 0.47318851947784424, + "learning_rate": 0.00010681869116987899, + "loss": 0.3495, + "step": 32007 + }, + { + "epoch": 1.788334217951225, + "grad_norm": 1.1491549015045166, + "learning_rate": 0.00010679067682653519, + "loss": 0.5011, + "step": 32008 + }, + { + "epoch": 1.7883900885549069, + "grad_norm": 0.5283597707748413, + "learning_rate": 0.0001067626624831914, + "loss": 0.357, + "step": 32009 + }, + { + "epoch": 1.7884459591585888, + "grad_norm": 0.3514658212661743, + "learning_rate": 0.0001067346481398476, + "loss": 0.408, + "step": 32010 + }, + { + "epoch": 1.7885018297622706, + "grad_norm": 0.48292815685272217, + "learning_rate": 0.00010670663379650382, + "loss": 0.4306, + "step": 32011 + }, + { + "epoch": 1.7885577003659523, + "grad_norm": 0.4136585295200348, + "learning_rate": 0.00010667861945316002, + "loss": 0.4866, + "step": 32012 + }, + { + "epoch": 1.7886135709696345, + "grad_norm": 0.49022817611694336, + "learning_rate": 0.00010665060510981622, + "loss": 0.3996, + "step": 32013 + }, + { + "epoch": 1.7886694415733162, + "grad_norm": 2.2896268367767334, + "learning_rate": 0.00010662259076647243, + "loss": 0.4856, + "step": 32014 + }, + { + "epoch": 1.788725312176998, + "grad_norm": 0.36924824118614197, + "learning_rate": 0.00010659457642312865, + "loss": 0.434, + "step": 32015 + }, + { + "epoch": 1.78878118278068, + "grad_norm": 0.439743310213089, + "learning_rate": 0.00010656656207978485, + "loss": 0.3588, + "step": 32016 + }, + { + "epoch": 1.7888370533843618, + "grad_norm": 0.4140082895755768, + "learning_rate": 0.00010653854773644106, + "loss": 0.4452, + "step": 32017 + }, + { + "epoch": 1.7888929239880436, + "grad_norm": 0.3958495855331421, + "learning_rate": 0.00010651053339309726, + "loss": 0.4391, + "step": 32018 + }, + { + "epoch": 1.7889487945917255, + "grad_norm": 1.7106246948242188, + "learning_rate": 0.00010648251904975348, + "loss": 0.4709, + "step": 32019 + }, + { + "epoch": 1.7890046651954075, + "grad_norm": 1.3069161176681519, + "learning_rate": 0.00010645450470640969, + "loss": 0.4145, + "step": 32020 + }, + { + "epoch": 1.7890605357990892, + "grad_norm": 0.3758501708507538, + "learning_rate": 0.0001064264903630659, + "loss": 0.4193, + "step": 32021 + }, + { + "epoch": 1.7891164064027711, + "grad_norm": 0.3443920314311981, + "learning_rate": 0.00010639847601972209, + "loss": 0.4111, + "step": 32022 + }, + { + "epoch": 1.789172277006453, + "grad_norm": 0.416570246219635, + "learning_rate": 0.0001063704616763783, + "loss": 0.5136, + "step": 32023 + }, + { + "epoch": 1.7892281476101348, + "grad_norm": 0.35609856247901917, + "learning_rate": 0.00010634244733303451, + "loss": 0.4922, + "step": 32024 + }, + { + "epoch": 1.7892840182138168, + "grad_norm": 0.5004066228866577, + "learning_rate": 0.00010631443298969072, + "loss": 0.4988, + "step": 32025 + }, + { + "epoch": 1.7893398888174987, + "grad_norm": 0.3453820049762726, + "learning_rate": 0.00010628641864634692, + "loss": 0.3181, + "step": 32026 + }, + { + "epoch": 1.7893957594211805, + "grad_norm": 0.46624672412872314, + "learning_rate": 0.00010625840430300314, + "loss": 0.3723, + "step": 32027 + }, + { + "epoch": 1.7894516300248624, + "grad_norm": 0.43824300169944763, + "learning_rate": 0.00010623038995965935, + "loss": 0.4224, + "step": 32028 + }, + { + "epoch": 1.7895075006285444, + "grad_norm": 0.3842627704143524, + "learning_rate": 0.00010620237561631555, + "loss": 0.3276, + "step": 32029 + }, + { + "epoch": 1.789563371232226, + "grad_norm": 0.316035658121109, + "learning_rate": 0.00010617436127297176, + "loss": 0.3301, + "step": 32030 + }, + { + "epoch": 1.789619241835908, + "grad_norm": 0.3984113931655884, + "learning_rate": 0.00010614634692962798, + "loss": 0.4121, + "step": 32031 + }, + { + "epoch": 1.78967511243959, + "grad_norm": 0.4016181230545044, + "learning_rate": 0.00010611833258628419, + "loss": 0.4346, + "step": 32032 + }, + { + "epoch": 1.7897309830432717, + "grad_norm": 0.7390085458755493, + "learning_rate": 0.00010609031824294038, + "loss": 0.387, + "step": 32033 + }, + { + "epoch": 1.7897868536469537, + "grad_norm": 0.3375400900840759, + "learning_rate": 0.00010606230389959658, + "loss": 0.3546, + "step": 32034 + }, + { + "epoch": 1.7898427242506356, + "grad_norm": 0.3591294586658478, + "learning_rate": 0.0001060342895562528, + "loss": 0.3325, + "step": 32035 + }, + { + "epoch": 1.7898985948543173, + "grad_norm": 1.579742193222046, + "learning_rate": 0.00010600627521290901, + "loss": 0.2597, + "step": 32036 + }, + { + "epoch": 1.7899544654579993, + "grad_norm": 0.5706319808959961, + "learning_rate": 0.00010597826086956521, + "loss": 0.4517, + "step": 32037 + }, + { + "epoch": 1.7900103360616813, + "grad_norm": 0.7318527102470398, + "learning_rate": 0.00010595024652622143, + "loss": 0.4848, + "step": 32038 + }, + { + "epoch": 1.790066206665363, + "grad_norm": 0.6391046643257141, + "learning_rate": 0.00010592223218287764, + "loss": 0.4199, + "step": 32039 + }, + { + "epoch": 1.790122077269045, + "grad_norm": 0.34860000014305115, + "learning_rate": 0.00010589421783953385, + "loss": 0.3451, + "step": 32040 + }, + { + "epoch": 1.7901779478727269, + "grad_norm": 0.4133601486682892, + "learning_rate": 0.00010586620349619005, + "loss": 0.3639, + "step": 32041 + }, + { + "epoch": 1.7902338184764086, + "grad_norm": 0.35462266206741333, + "learning_rate": 0.00010583818915284627, + "loss": 0.4335, + "step": 32042 + }, + { + "epoch": 1.7902896890800906, + "grad_norm": 0.37933477759361267, + "learning_rate": 0.00010581017480950248, + "loss": 0.3344, + "step": 32043 + }, + { + "epoch": 1.7903455596837725, + "grad_norm": 0.4388585388660431, + "learning_rate": 0.00010578216046615867, + "loss": 0.4106, + "step": 32044 + }, + { + "epoch": 1.7904014302874542, + "grad_norm": 0.329399973154068, + "learning_rate": 0.00010575414612281487, + "loss": 0.375, + "step": 32045 + }, + { + "epoch": 1.790457300891136, + "grad_norm": 1.9976669549942017, + "learning_rate": 0.0001057261317794711, + "loss": 0.4452, + "step": 32046 + }, + { + "epoch": 1.7905131714948181, + "grad_norm": 0.800493061542511, + "learning_rate": 0.0001056981174361273, + "loss": 0.694, + "step": 32047 + }, + { + "epoch": 1.7905690420984999, + "grad_norm": 0.4921495020389557, + "learning_rate": 0.0001056701030927835, + "loss": 0.4547, + "step": 32048 + }, + { + "epoch": 1.7906249127021816, + "grad_norm": 0.4007548987865448, + "learning_rate": 0.00010564208874943971, + "loss": 0.4312, + "step": 32049 + }, + { + "epoch": 1.7906807833058638, + "grad_norm": 5.218757152557373, + "learning_rate": 0.00010561407440609593, + "loss": 0.4136, + "step": 32050 + }, + { + "epoch": 1.7907366539095455, + "grad_norm": 0.35816115140914917, + "learning_rate": 0.00010558606006275214, + "loss": 0.3151, + "step": 32051 + }, + { + "epoch": 1.7907925245132272, + "grad_norm": 0.3768213391304016, + "learning_rate": 0.00010555804571940834, + "loss": 0.4334, + "step": 32052 + }, + { + "epoch": 1.7908483951169092, + "grad_norm": 3.3188090324401855, + "learning_rate": 0.00010553003137606455, + "loss": 0.6309, + "step": 32053 + }, + { + "epoch": 1.7909042657205911, + "grad_norm": 0.26915431022644043, + "learning_rate": 0.00010550201703272075, + "loss": 0.3616, + "step": 32054 + }, + { + "epoch": 1.7909601363242729, + "grad_norm": 0.48466262221336365, + "learning_rate": 0.00010547400268937696, + "loss": 0.4734, + "step": 32055 + }, + { + "epoch": 1.7910160069279548, + "grad_norm": 0.838082492351532, + "learning_rate": 0.00010544598834603317, + "loss": 0.442, + "step": 32056 + }, + { + "epoch": 1.7910718775316368, + "grad_norm": 1.0987638235092163, + "learning_rate": 0.00010541797400268937, + "loss": 0.3747, + "step": 32057 + }, + { + "epoch": 1.7911277481353185, + "grad_norm": 0.5994178056716919, + "learning_rate": 0.00010538995965934559, + "loss": 0.486, + "step": 32058 + }, + { + "epoch": 1.7911836187390004, + "grad_norm": 0.5544427037239075, + "learning_rate": 0.0001053619453160018, + "loss": 0.4633, + "step": 32059 + }, + { + "epoch": 1.7912394893426824, + "grad_norm": 0.515326976776123, + "learning_rate": 0.000105333930972658, + "loss": 0.4362, + "step": 32060 + }, + { + "epoch": 1.7912953599463641, + "grad_norm": 0.4256957471370697, + "learning_rate": 0.00010530591662931421, + "loss": 0.3522, + "step": 32061 + }, + { + "epoch": 1.791351230550046, + "grad_norm": 0.32711899280548096, + "learning_rate": 0.00010527790228597043, + "loss": 0.3351, + "step": 32062 + }, + { + "epoch": 1.791407101153728, + "grad_norm": 0.6276299357414246, + "learning_rate": 0.00010524988794262663, + "loss": 0.3665, + "step": 32063 + }, + { + "epoch": 1.7914629717574098, + "grad_norm": 0.340925395488739, + "learning_rate": 0.00010522187359928284, + "loss": 0.3415, + "step": 32064 + }, + { + "epoch": 1.7915188423610917, + "grad_norm": 0.4122478663921356, + "learning_rate": 0.00010519385925593903, + "loss": 0.4506, + "step": 32065 + }, + { + "epoch": 1.7915747129647737, + "grad_norm": 0.3525846600532532, + "learning_rate": 0.00010516584491259525, + "loss": 0.4137, + "step": 32066 + }, + { + "epoch": 1.7916305835684554, + "grad_norm": 0.5300910472869873, + "learning_rate": 0.00010513783056925146, + "loss": 0.463, + "step": 32067 + }, + { + "epoch": 1.7916864541721373, + "grad_norm": 0.4531669616699219, + "learning_rate": 0.00010510981622590766, + "loss": 0.4382, + "step": 32068 + }, + { + "epoch": 1.7917423247758193, + "grad_norm": 0.4024210572242737, + "learning_rate": 0.00010508180188256387, + "loss": 0.4016, + "step": 32069 + }, + { + "epoch": 1.791798195379501, + "grad_norm": 2.55892276763916, + "learning_rate": 0.00010505378753922009, + "loss": 0.3457, + "step": 32070 + }, + { + "epoch": 1.791854065983183, + "grad_norm": 0.43230995535850525, + "learning_rate": 0.0001050257731958763, + "loss": 0.4471, + "step": 32071 + }, + { + "epoch": 1.791909936586865, + "grad_norm": 3.2905330657958984, + "learning_rate": 0.0001049977588525325, + "loss": 0.4476, + "step": 32072 + }, + { + "epoch": 1.7919658071905467, + "grad_norm": 0.6770856976509094, + "learning_rate": 0.0001049697445091887, + "loss": 0.3907, + "step": 32073 + }, + { + "epoch": 1.7920216777942286, + "grad_norm": 0.37827855348587036, + "learning_rate": 0.00010494173016584492, + "loss": 0.416, + "step": 32074 + }, + { + "epoch": 1.7920775483979106, + "grad_norm": 0.290313184261322, + "learning_rate": 0.00010491371582250113, + "loss": 0.3645, + "step": 32075 + }, + { + "epoch": 1.7921334190015923, + "grad_norm": 0.43834561109542847, + "learning_rate": 0.00010488570147915732, + "loss": 0.3645, + "step": 32076 + }, + { + "epoch": 1.7921892896052742, + "grad_norm": 2.1874499320983887, + "learning_rate": 0.00010485768713581353, + "loss": 0.4009, + "step": 32077 + }, + { + "epoch": 1.7922451602089562, + "grad_norm": 0.701718807220459, + "learning_rate": 0.00010482967279246975, + "loss": 0.3722, + "step": 32078 + }, + { + "epoch": 1.792301030812638, + "grad_norm": 0.4298023581504822, + "learning_rate": 0.00010480165844912595, + "loss": 0.4099, + "step": 32079 + }, + { + "epoch": 1.7923569014163196, + "grad_norm": 0.33678677678108215, + "learning_rate": 0.00010477364410578216, + "loss": 0.3798, + "step": 32080 + }, + { + "epoch": 1.7924127720200018, + "grad_norm": 0.4661508798599243, + "learning_rate": 0.00010474562976243837, + "loss": 0.3245, + "step": 32081 + }, + { + "epoch": 1.7924686426236836, + "grad_norm": 0.4329415261745453, + "learning_rate": 0.00010471761541909458, + "loss": 0.4779, + "step": 32082 + }, + { + "epoch": 1.7925245132273653, + "grad_norm": 0.4820594787597656, + "learning_rate": 0.00010468960107575079, + "loss": 0.4992, + "step": 32083 + }, + { + "epoch": 1.7925803838310475, + "grad_norm": 0.4126955568790436, + "learning_rate": 0.000104661586732407, + "loss": 0.4751, + "step": 32084 + }, + { + "epoch": 1.7926362544347292, + "grad_norm": 0.5851201415061951, + "learning_rate": 0.0001046335723890632, + "loss": 0.6209, + "step": 32085 + }, + { + "epoch": 1.792692125038411, + "grad_norm": 0.2948147654533386, + "learning_rate": 0.00010460555804571941, + "loss": 0.3569, + "step": 32086 + }, + { + "epoch": 1.7927479956420929, + "grad_norm": 0.34923774003982544, + "learning_rate": 0.00010457754370237561, + "loss": 0.3872, + "step": 32087 + }, + { + "epoch": 1.7928038662457748, + "grad_norm": 0.358916312456131, + "learning_rate": 0.00010454952935903182, + "loss": 0.4046, + "step": 32088 + }, + { + "epoch": 1.7928597368494565, + "grad_norm": 0.32111889123916626, + "learning_rate": 0.00010452151501568803, + "loss": 0.4207, + "step": 32089 + }, + { + "epoch": 1.7929156074531385, + "grad_norm": 0.5587243437767029, + "learning_rate": 0.00010449350067234425, + "loss": 0.5685, + "step": 32090 + }, + { + "epoch": 1.7929714780568204, + "grad_norm": 0.35585471987724304, + "learning_rate": 0.00010446548632900045, + "loss": 0.3321, + "step": 32091 + }, + { + "epoch": 1.7930273486605022, + "grad_norm": 0.39240172505378723, + "learning_rate": 0.00010443747198565666, + "loss": 0.4729, + "step": 32092 + }, + { + "epoch": 1.7930832192641841, + "grad_norm": 0.5814953446388245, + "learning_rate": 0.00010440945764231286, + "loss": 0.5745, + "step": 32093 + }, + { + "epoch": 1.793139089867866, + "grad_norm": 0.3261435925960541, + "learning_rate": 0.00010438144329896908, + "loss": 0.3602, + "step": 32094 + }, + { + "epoch": 1.7931949604715478, + "grad_norm": 0.6487747430801392, + "learning_rate": 0.00010435342895562529, + "loss": 0.4272, + "step": 32095 + }, + { + "epoch": 1.7932508310752298, + "grad_norm": 0.40280503034591675, + "learning_rate": 0.0001043254146122815, + "loss": 0.5657, + "step": 32096 + }, + { + "epoch": 1.7933067016789117, + "grad_norm": 0.5109310746192932, + "learning_rate": 0.00010429740026893769, + "loss": 0.4583, + "step": 32097 + }, + { + "epoch": 1.7933625722825934, + "grad_norm": 0.47950708866119385, + "learning_rate": 0.0001042693859255939, + "loss": 0.3839, + "step": 32098 + }, + { + "epoch": 1.7934184428862754, + "grad_norm": 3.1981558799743652, + "learning_rate": 0.00010424137158225011, + "loss": 0.3465, + "step": 32099 + }, + { + "epoch": 1.7934743134899573, + "grad_norm": 0.6619800925254822, + "learning_rate": 0.00010421335723890632, + "loss": 0.4775, + "step": 32100 + }, + { + "epoch": 1.793530184093639, + "grad_norm": 0.327280193567276, + "learning_rate": 0.00010418534289556252, + "loss": 0.3803, + "step": 32101 + }, + { + "epoch": 1.793586054697321, + "grad_norm": 0.5450477600097656, + "learning_rate": 0.00010415732855221874, + "loss": 0.4647, + "step": 32102 + }, + { + "epoch": 1.793641925301003, + "grad_norm": 0.39091143012046814, + "learning_rate": 0.00010412931420887495, + "loss": 0.455, + "step": 32103 + }, + { + "epoch": 1.7936977959046847, + "grad_norm": 0.40152713656425476, + "learning_rate": 0.00010410129986553115, + "loss": 0.4605, + "step": 32104 + }, + { + "epoch": 1.7937536665083667, + "grad_norm": 0.31872835755348206, + "learning_rate": 0.00010407328552218737, + "loss": 0.3484, + "step": 32105 + }, + { + "epoch": 1.7938095371120486, + "grad_norm": 1.1024017333984375, + "learning_rate": 0.00010404527117884358, + "loss": 0.367, + "step": 32106 + }, + { + "epoch": 1.7938654077157303, + "grad_norm": 0.5422174334526062, + "learning_rate": 0.00010401725683549977, + "loss": 0.6428, + "step": 32107 + }, + { + "epoch": 1.7939212783194123, + "grad_norm": 0.3486064076423645, + "learning_rate": 0.00010398924249215598, + "loss": 0.4005, + "step": 32108 + }, + { + "epoch": 1.7939771489230942, + "grad_norm": 0.471532940864563, + "learning_rate": 0.0001039612281488122, + "loss": 0.4478, + "step": 32109 + }, + { + "epoch": 1.794033019526776, + "grad_norm": 0.37275609374046326, + "learning_rate": 0.0001039332138054684, + "loss": 0.4567, + "step": 32110 + }, + { + "epoch": 1.794088890130458, + "grad_norm": 0.4718460738658905, + "learning_rate": 0.00010390519946212461, + "loss": 0.3219, + "step": 32111 + }, + { + "epoch": 1.7941447607341399, + "grad_norm": 0.7108396887779236, + "learning_rate": 0.00010387718511878081, + "loss": 0.3553, + "step": 32112 + }, + { + "epoch": 1.7942006313378216, + "grad_norm": 1.7412127256393433, + "learning_rate": 0.00010384917077543703, + "loss": 0.618, + "step": 32113 + }, + { + "epoch": 1.7942565019415033, + "grad_norm": 0.3074713349342346, + "learning_rate": 0.00010382115643209324, + "loss": 0.3926, + "step": 32114 + }, + { + "epoch": 1.7943123725451855, + "grad_norm": 0.37503185868263245, + "learning_rate": 0.00010379314208874944, + "loss": 0.4062, + "step": 32115 + }, + { + "epoch": 1.7943682431488672, + "grad_norm": 5.385427474975586, + "learning_rate": 0.00010376512774540565, + "loss": 0.3845, + "step": 32116 + }, + { + "epoch": 1.794424113752549, + "grad_norm": 0.7665802836418152, + "learning_rate": 0.00010373711340206187, + "loss": 0.4026, + "step": 32117 + }, + { + "epoch": 1.7944799843562311, + "grad_norm": 0.32550448179244995, + "learning_rate": 0.00010370909905871806, + "loss": 0.3367, + "step": 32118 + }, + { + "epoch": 1.7945358549599129, + "grad_norm": 0.41240569949150085, + "learning_rate": 0.00010368108471537427, + "loss": 0.5238, + "step": 32119 + }, + { + "epoch": 1.7945917255635946, + "grad_norm": 0.32674679160118103, + "learning_rate": 0.00010365307037203047, + "loss": 0.3486, + "step": 32120 + }, + { + "epoch": 1.7946475961672765, + "grad_norm": 0.5406381487846375, + "learning_rate": 0.00010362505602868669, + "loss": 0.4127, + "step": 32121 + }, + { + "epoch": 1.7947034667709585, + "grad_norm": 0.391112744808197, + "learning_rate": 0.0001035970416853429, + "loss": 0.4436, + "step": 32122 + }, + { + "epoch": 1.7947593373746402, + "grad_norm": 0.29874053597450256, + "learning_rate": 0.0001035690273419991, + "loss": 0.3394, + "step": 32123 + }, + { + "epoch": 1.7948152079783222, + "grad_norm": 0.6105486750602722, + "learning_rate": 0.00010354101299865531, + "loss": 0.6914, + "step": 32124 + }, + { + "epoch": 1.7948710785820041, + "grad_norm": 0.5440850853919983, + "learning_rate": 0.00010351299865531153, + "loss": 0.4441, + "step": 32125 + }, + { + "epoch": 1.7949269491856858, + "grad_norm": 0.4036461114883423, + "learning_rate": 0.00010348498431196774, + "loss": 0.4859, + "step": 32126 + }, + { + "epoch": 1.7949828197893678, + "grad_norm": 0.8165017366409302, + "learning_rate": 0.00010345696996862394, + "loss": 0.3936, + "step": 32127 + }, + { + "epoch": 1.7950386903930498, + "grad_norm": 0.486155241727829, + "learning_rate": 0.00010342895562528015, + "loss": 0.3759, + "step": 32128 + }, + { + "epoch": 1.7950945609967315, + "grad_norm": 0.6855242252349854, + "learning_rate": 0.00010340094128193635, + "loss": 0.4163, + "step": 32129 + }, + { + "epoch": 1.7951504316004134, + "grad_norm": 0.6624358892440796, + "learning_rate": 0.00010337292693859256, + "loss": 0.4422, + "step": 32130 + }, + { + "epoch": 1.7952063022040954, + "grad_norm": 0.47609853744506836, + "learning_rate": 0.00010334491259524876, + "loss": 0.4478, + "step": 32131 + }, + { + "epoch": 1.7952621728077771, + "grad_norm": 0.6029824614524841, + "learning_rate": 0.00010331689825190497, + "loss": 0.4774, + "step": 32132 + }, + { + "epoch": 1.795318043411459, + "grad_norm": 0.285302072763443, + "learning_rate": 0.00010328888390856119, + "loss": 0.3636, + "step": 32133 + }, + { + "epoch": 1.795373914015141, + "grad_norm": 0.3316420018672943, + "learning_rate": 0.0001032608695652174, + "loss": 0.3193, + "step": 32134 + }, + { + "epoch": 1.7954297846188227, + "grad_norm": 0.6690402626991272, + "learning_rate": 0.0001032328552218736, + "loss": 0.4722, + "step": 32135 + }, + { + "epoch": 1.7954856552225047, + "grad_norm": 0.5926231741905212, + "learning_rate": 0.00010320484087852981, + "loss": 0.331, + "step": 32136 + }, + { + "epoch": 1.7955415258261866, + "grad_norm": 0.33872488141059875, + "learning_rate": 0.00010317682653518603, + "loss": 0.3735, + "step": 32137 + }, + { + "epoch": 1.7955973964298684, + "grad_norm": 0.35484451055526733, + "learning_rate": 0.00010314881219184223, + "loss": 0.3575, + "step": 32138 + }, + { + "epoch": 1.7956532670335503, + "grad_norm": 0.47739171981811523, + "learning_rate": 0.00010312079784849842, + "loss": 0.4213, + "step": 32139 + }, + { + "epoch": 1.7957091376372323, + "grad_norm": 5.078860282897949, + "learning_rate": 0.00010309278350515463, + "loss": 0.5051, + "step": 32140 + }, + { + "epoch": 1.795765008240914, + "grad_norm": 0.51535964012146, + "learning_rate": 0.00010306476916181085, + "loss": 0.4508, + "step": 32141 + }, + { + "epoch": 1.795820878844596, + "grad_norm": 0.31010594964027405, + "learning_rate": 0.00010303675481846706, + "loss": 0.3312, + "step": 32142 + }, + { + "epoch": 1.795876749448278, + "grad_norm": 0.3624110817909241, + "learning_rate": 0.00010300874047512326, + "loss": 0.4241, + "step": 32143 + }, + { + "epoch": 1.7959326200519596, + "grad_norm": 0.3542773127555847, + "learning_rate": 0.00010298072613177947, + "loss": 0.4177, + "step": 32144 + }, + { + "epoch": 1.7959884906556416, + "grad_norm": 0.4761391878128052, + "learning_rate": 0.00010295271178843569, + "loss": 0.4373, + "step": 32145 + }, + { + "epoch": 1.7960443612593235, + "grad_norm": 0.576920211315155, + "learning_rate": 0.00010292469744509189, + "loss": 0.4445, + "step": 32146 + }, + { + "epoch": 1.7961002318630053, + "grad_norm": 1.173795223236084, + "learning_rate": 0.0001028966831017481, + "loss": 0.4295, + "step": 32147 + }, + { + "epoch": 1.796156102466687, + "grad_norm": 0.5111125707626343, + "learning_rate": 0.0001028686687584043, + "loss": 0.4223, + "step": 32148 + }, + { + "epoch": 1.7962119730703692, + "grad_norm": 0.2933606803417206, + "learning_rate": 0.00010284065441506052, + "loss": 0.2912, + "step": 32149 + }, + { + "epoch": 1.796267843674051, + "grad_norm": 0.4988614320755005, + "learning_rate": 0.00010281264007171672, + "loss": 0.3766, + "step": 32150 + }, + { + "epoch": 1.7963237142777326, + "grad_norm": 0.4470154941082001, + "learning_rate": 0.00010278462572837292, + "loss": 0.5199, + "step": 32151 + }, + { + "epoch": 1.7963795848814148, + "grad_norm": 0.43521350622177124, + "learning_rate": 0.00010275661138502913, + "loss": 0.5281, + "step": 32152 + }, + { + "epoch": 1.7964354554850965, + "grad_norm": 0.7526671290397644, + "learning_rate": 0.00010272859704168535, + "loss": 0.3675, + "step": 32153 + }, + { + "epoch": 1.7964913260887783, + "grad_norm": 0.4906253218650818, + "learning_rate": 0.00010270058269834155, + "loss": 0.4717, + "step": 32154 + }, + { + "epoch": 1.7965471966924602, + "grad_norm": 0.6133006811141968, + "learning_rate": 0.00010267256835499776, + "loss": 0.3924, + "step": 32155 + }, + { + "epoch": 1.7966030672961422, + "grad_norm": 1.677354335784912, + "learning_rate": 0.00010264455401165396, + "loss": 0.4256, + "step": 32156 + }, + { + "epoch": 1.796658937899824, + "grad_norm": 0.3330688774585724, + "learning_rate": 0.00010261653966831018, + "loss": 0.4214, + "step": 32157 + }, + { + "epoch": 1.7967148085035058, + "grad_norm": 0.6427372097969055, + "learning_rate": 0.00010258852532496639, + "loss": 0.3379, + "step": 32158 + }, + { + "epoch": 1.7967706791071878, + "grad_norm": 0.5160718560218811, + "learning_rate": 0.0001025605109816226, + "loss": 0.5489, + "step": 32159 + }, + { + "epoch": 1.7968265497108695, + "grad_norm": 1.3512331247329712, + "learning_rate": 0.0001025324966382788, + "loss": 0.4286, + "step": 32160 + }, + { + "epoch": 1.7968824203145515, + "grad_norm": 0.5147550702095032, + "learning_rate": 0.00010250448229493501, + "loss": 0.3849, + "step": 32161 + }, + { + "epoch": 1.7969382909182334, + "grad_norm": 0.6195976734161377, + "learning_rate": 0.00010247646795159121, + "loss": 0.4531, + "step": 32162 + }, + { + "epoch": 1.7969941615219152, + "grad_norm": 0.5793015956878662, + "learning_rate": 0.00010244845360824742, + "loss": 0.5197, + "step": 32163 + }, + { + "epoch": 1.797050032125597, + "grad_norm": 0.3246740698814392, + "learning_rate": 0.00010242043926490362, + "loss": 0.3363, + "step": 32164 + }, + { + "epoch": 1.797105902729279, + "grad_norm": 2.185816764831543, + "learning_rate": 0.00010239242492155984, + "loss": 0.4386, + "step": 32165 + }, + { + "epoch": 1.7971617733329608, + "grad_norm": 0.9246243238449097, + "learning_rate": 0.00010236441057821605, + "loss": 0.3065, + "step": 32166 + }, + { + "epoch": 1.7972176439366427, + "grad_norm": 0.6271084547042847, + "learning_rate": 0.00010233639623487226, + "loss": 0.3701, + "step": 32167 + }, + { + "epoch": 1.7972735145403247, + "grad_norm": 0.41680586338043213, + "learning_rate": 0.00010230838189152847, + "loss": 0.3089, + "step": 32168 + }, + { + "epoch": 1.7973293851440064, + "grad_norm": 0.4414445161819458, + "learning_rate": 0.00010228036754818468, + "loss": 0.3525, + "step": 32169 + }, + { + "epoch": 1.7973852557476884, + "grad_norm": 0.3563736081123352, + "learning_rate": 0.00010225235320484089, + "loss": 0.382, + "step": 32170 + }, + { + "epoch": 1.7974411263513703, + "grad_norm": 0.4441733956336975, + "learning_rate": 0.00010222433886149708, + "loss": 0.4303, + "step": 32171 + }, + { + "epoch": 1.797496996955052, + "grad_norm": 0.33959341049194336, + "learning_rate": 0.0001021963245181533, + "loss": 0.4517, + "step": 32172 + }, + { + "epoch": 1.797552867558734, + "grad_norm": 0.8667090535163879, + "learning_rate": 0.0001021683101748095, + "loss": 0.4458, + "step": 32173 + }, + { + "epoch": 1.797608738162416, + "grad_norm": 0.36744022369384766, + "learning_rate": 0.00010214029583146571, + "loss": 0.4498, + "step": 32174 + }, + { + "epoch": 1.7976646087660977, + "grad_norm": 0.330890953540802, + "learning_rate": 0.00010211228148812192, + "loss": 0.4349, + "step": 32175 + }, + { + "epoch": 1.7977204793697796, + "grad_norm": 0.33265554904937744, + "learning_rate": 0.00010208426714477813, + "loss": 0.3014, + "step": 32176 + }, + { + "epoch": 1.7977763499734616, + "grad_norm": 0.44974756240844727, + "learning_rate": 0.00010205625280143434, + "loss": 0.4746, + "step": 32177 + }, + { + "epoch": 1.7978322205771433, + "grad_norm": 0.4527336657047272, + "learning_rate": 0.00010202823845809055, + "loss": 0.447, + "step": 32178 + }, + { + "epoch": 1.797888091180825, + "grad_norm": 0.4121036231517792, + "learning_rate": 0.00010200022411474675, + "loss": 0.4432, + "step": 32179 + }, + { + "epoch": 1.7979439617845072, + "grad_norm": 0.4685678780078888, + "learning_rate": 0.00010197220977140297, + "loss": 0.373, + "step": 32180 + }, + { + "epoch": 1.797999832388189, + "grad_norm": 0.5137205123901367, + "learning_rate": 0.00010194419542805918, + "loss": 0.3671, + "step": 32181 + }, + { + "epoch": 1.7980557029918707, + "grad_norm": 0.5385423898696899, + "learning_rate": 0.00010191618108471537, + "loss": 0.3803, + "step": 32182 + }, + { + "epoch": 1.7981115735955528, + "grad_norm": 0.5154129862785339, + "learning_rate": 0.00010188816674137158, + "loss": 0.4285, + "step": 32183 + }, + { + "epoch": 1.7981674441992346, + "grad_norm": 0.4108971655368805, + "learning_rate": 0.0001018601523980278, + "loss": 0.4128, + "step": 32184 + }, + { + "epoch": 1.7982233148029163, + "grad_norm": 0.32855090498924255, + "learning_rate": 0.000101832138054684, + "loss": 0.3992, + "step": 32185 + }, + { + "epoch": 1.7982791854065985, + "grad_norm": 1.4293018579483032, + "learning_rate": 0.0001018041237113402, + "loss": 0.4964, + "step": 32186 + }, + { + "epoch": 1.7983350560102802, + "grad_norm": 0.565606415271759, + "learning_rate": 0.00010177610936799641, + "loss": 0.4622, + "step": 32187 + }, + { + "epoch": 1.798390926613962, + "grad_norm": 0.369072288274765, + "learning_rate": 0.00010174809502465263, + "loss": 0.4082, + "step": 32188 + }, + { + "epoch": 1.7984467972176439, + "grad_norm": 0.29724857211112976, + "learning_rate": 0.00010172008068130884, + "loss": 0.3247, + "step": 32189 + }, + { + "epoch": 1.7985026678213258, + "grad_norm": 0.6763718128204346, + "learning_rate": 0.00010169206633796504, + "loss": 0.4052, + "step": 32190 + }, + { + "epoch": 1.7985585384250076, + "grad_norm": 1.3506466150283813, + "learning_rate": 0.00010166405199462125, + "loss": 0.3917, + "step": 32191 + }, + { + "epoch": 1.7986144090286895, + "grad_norm": 0.44316238164901733, + "learning_rate": 0.00010163603765127746, + "loss": 0.4147, + "step": 32192 + }, + { + "epoch": 1.7986702796323715, + "grad_norm": 0.4197893440723419, + "learning_rate": 0.00010160802330793366, + "loss": 0.4045, + "step": 32193 + }, + { + "epoch": 1.7987261502360532, + "grad_norm": 0.772171676158905, + "learning_rate": 0.00010158000896458987, + "loss": 0.4903, + "step": 32194 + }, + { + "epoch": 1.7987820208397352, + "grad_norm": 0.4409593343734741, + "learning_rate": 0.00010155199462124607, + "loss": 0.4094, + "step": 32195 + }, + { + "epoch": 1.798837891443417, + "grad_norm": 0.31532952189445496, + "learning_rate": 0.00010152398027790229, + "loss": 0.4605, + "step": 32196 + }, + { + "epoch": 1.7988937620470988, + "grad_norm": 0.35377323627471924, + "learning_rate": 0.0001014959659345585, + "loss": 0.3197, + "step": 32197 + }, + { + "epoch": 1.7989496326507808, + "grad_norm": 0.3667739927768707, + "learning_rate": 0.0001014679515912147, + "loss": 0.3045, + "step": 32198 + }, + { + "epoch": 1.7990055032544627, + "grad_norm": 0.4204750955104828, + "learning_rate": 0.00010143993724787091, + "loss": 0.4354, + "step": 32199 + }, + { + "epoch": 1.7990613738581445, + "grad_norm": 0.38836807012557983, + "learning_rate": 0.00010141192290452713, + "loss": 0.4251, + "step": 32200 + }, + { + "epoch": 1.7991172444618264, + "grad_norm": 1.3796592950820923, + "learning_rate": 0.00010138390856118333, + "loss": 0.4871, + "step": 32201 + }, + { + "epoch": 1.7991731150655084, + "grad_norm": 0.8318590521812439, + "learning_rate": 0.00010135589421783954, + "loss": 0.3947, + "step": 32202 + }, + { + "epoch": 1.79922898566919, + "grad_norm": 0.3810541033744812, + "learning_rate": 0.00010132787987449573, + "loss": 0.4492, + "step": 32203 + }, + { + "epoch": 1.799284856272872, + "grad_norm": 0.4472671449184418, + "learning_rate": 0.00010129986553115195, + "loss": 0.3612, + "step": 32204 + }, + { + "epoch": 1.799340726876554, + "grad_norm": 0.5299384593963623, + "learning_rate": 0.00010127185118780816, + "loss": 0.4388, + "step": 32205 + }, + { + "epoch": 1.7993965974802357, + "grad_norm": 1.7054448127746582, + "learning_rate": 0.00010124383684446436, + "loss": 0.3342, + "step": 32206 + }, + { + "epoch": 1.7994524680839177, + "grad_norm": 0.29793161153793335, + "learning_rate": 0.00010121582250112057, + "loss": 0.3406, + "step": 32207 + }, + { + "epoch": 1.7995083386875996, + "grad_norm": 0.360792338848114, + "learning_rate": 0.00010118780815777679, + "loss": 0.394, + "step": 32208 + }, + { + "epoch": 1.7995642092912814, + "grad_norm": 0.4078877568244934, + "learning_rate": 0.000101159793814433, + "loss": 0.4518, + "step": 32209 + }, + { + "epoch": 1.7996200798949633, + "grad_norm": 5.983916759490967, + "learning_rate": 0.0001011317794710892, + "loss": 0.3299, + "step": 32210 + }, + { + "epoch": 1.7996759504986453, + "grad_norm": 0.4031793475151062, + "learning_rate": 0.0001011037651277454, + "loss": 0.406, + "step": 32211 + }, + { + "epoch": 1.799731821102327, + "grad_norm": 0.49351754784584045, + "learning_rate": 0.00010107575078440163, + "loss": 0.4023, + "step": 32212 + }, + { + "epoch": 1.7997876917060087, + "grad_norm": 0.327363520860672, + "learning_rate": 0.00010104773644105783, + "loss": 0.3797, + "step": 32213 + }, + { + "epoch": 1.799843562309691, + "grad_norm": 0.49227964878082275, + "learning_rate": 0.00010101972209771402, + "loss": 0.4603, + "step": 32214 + }, + { + "epoch": 1.7998994329133726, + "grad_norm": 0.3846569061279297, + "learning_rate": 0.00010099170775437023, + "loss": 0.3722, + "step": 32215 + }, + { + "epoch": 1.7999553035170544, + "grad_norm": 1.0728274583816528, + "learning_rate": 0.00010096369341102645, + "loss": 0.3771, + "step": 32216 + }, + { + "epoch": 1.8000111741207365, + "grad_norm": 0.3323317766189575, + "learning_rate": 0.00010093567906768265, + "loss": 0.4565, + "step": 32217 + }, + { + "epoch": 1.8000670447244183, + "grad_norm": 0.4822692573070526, + "learning_rate": 0.00010090766472433886, + "loss": 0.3614, + "step": 32218 + }, + { + "epoch": 1.8001229153281, + "grad_norm": 0.39569398760795593, + "learning_rate": 0.00010087965038099507, + "loss": 0.4311, + "step": 32219 + }, + { + "epoch": 1.8001787859317822, + "grad_norm": 0.7776076793670654, + "learning_rate": 0.00010085163603765129, + "loss": 0.4728, + "step": 32220 + }, + { + "epoch": 1.8002346565354639, + "grad_norm": 0.3745270371437073, + "learning_rate": 0.00010082362169430749, + "loss": 0.3253, + "step": 32221 + }, + { + "epoch": 1.8002905271391456, + "grad_norm": 0.813983678817749, + "learning_rate": 0.0001007956073509637, + "loss": 0.3563, + "step": 32222 + }, + { + "epoch": 1.8003463977428276, + "grad_norm": 0.5240582227706909, + "learning_rate": 0.0001007675930076199, + "loss": 0.4094, + "step": 32223 + }, + { + "epoch": 1.8004022683465095, + "grad_norm": 0.3394773006439209, + "learning_rate": 0.00010073957866427611, + "loss": 0.383, + "step": 32224 + }, + { + "epoch": 1.8004581389501912, + "grad_norm": 0.370312362909317, + "learning_rate": 0.00010071156432093231, + "loss": 0.3807, + "step": 32225 + }, + { + "epoch": 1.8005140095538732, + "grad_norm": 1.704925298690796, + "learning_rate": 0.00010068354997758852, + "loss": 0.3625, + "step": 32226 + }, + { + "epoch": 1.8005698801575551, + "grad_norm": 0.3544802665710449, + "learning_rate": 0.00010065553563424473, + "loss": 0.3344, + "step": 32227 + }, + { + "epoch": 1.8006257507612369, + "grad_norm": 0.40992093086242676, + "learning_rate": 0.00010062752129090095, + "loss": 0.4569, + "step": 32228 + }, + { + "epoch": 1.8006816213649188, + "grad_norm": 0.6871823072433472, + "learning_rate": 0.00010059950694755715, + "loss": 0.4174, + "step": 32229 + }, + { + "epoch": 1.8007374919686008, + "grad_norm": 0.4873432517051697, + "learning_rate": 0.00010057149260421336, + "loss": 0.3859, + "step": 32230 + }, + { + "epoch": 1.8007933625722825, + "grad_norm": 0.5891643166542053, + "learning_rate": 0.00010054347826086956, + "loss": 0.5246, + "step": 32231 + }, + { + "epoch": 1.8008492331759645, + "grad_norm": 1.204826831817627, + "learning_rate": 0.00010051546391752578, + "loss": 0.4974, + "step": 32232 + }, + { + "epoch": 1.8009051037796464, + "grad_norm": 0.3520817160606384, + "learning_rate": 0.00010048744957418199, + "loss": 0.5491, + "step": 32233 + }, + { + "epoch": 1.8009609743833281, + "grad_norm": 0.3944229781627655, + "learning_rate": 0.0001004594352308382, + "loss": 0.3998, + "step": 32234 + }, + { + "epoch": 1.80101684498701, + "grad_norm": 0.5144585371017456, + "learning_rate": 0.00010043142088749439, + "loss": 0.465, + "step": 32235 + }, + { + "epoch": 1.801072715590692, + "grad_norm": 0.7703235745429993, + "learning_rate": 0.0001004034065441506, + "loss": 0.3683, + "step": 32236 + }, + { + "epoch": 1.8011285861943738, + "grad_norm": 0.35460084676742554, + "learning_rate": 0.00010037539220080681, + "loss": 0.3833, + "step": 32237 + }, + { + "epoch": 1.8011844567980557, + "grad_norm": 0.472268670797348, + "learning_rate": 0.00010034737785746302, + "loss": 0.3344, + "step": 32238 + }, + { + "epoch": 1.8012403274017377, + "grad_norm": 0.533959686756134, + "learning_rate": 0.00010031936351411924, + "loss": 0.435, + "step": 32239 + }, + { + "epoch": 1.8012961980054194, + "grad_norm": 0.4317600131034851, + "learning_rate": 0.00010029134917077544, + "loss": 0.3716, + "step": 32240 + }, + { + "epoch": 1.8013520686091014, + "grad_norm": 0.46588313579559326, + "learning_rate": 0.00010026333482743165, + "loss": 0.4539, + "step": 32241 + }, + { + "epoch": 1.8014079392127833, + "grad_norm": 0.5147309899330139, + "learning_rate": 0.00010023532048408785, + "loss": 0.5266, + "step": 32242 + }, + { + "epoch": 1.801463809816465, + "grad_norm": 0.6395717263221741, + "learning_rate": 0.00010020730614074407, + "loss": 0.3765, + "step": 32243 + }, + { + "epoch": 1.801519680420147, + "grad_norm": 0.5324587225914001, + "learning_rate": 0.00010017929179740028, + "loss": 0.3641, + "step": 32244 + }, + { + "epoch": 1.801575551023829, + "grad_norm": 1.784595251083374, + "learning_rate": 0.00010015127745405649, + "loss": 0.4021, + "step": 32245 + }, + { + "epoch": 1.8016314216275107, + "grad_norm": 0.3502439260482788, + "learning_rate": 0.00010012326311071268, + "loss": 0.3581, + "step": 32246 + }, + { + "epoch": 1.8016872922311924, + "grad_norm": 3.2448878288269043, + "learning_rate": 0.0001000952487673689, + "loss": 0.3629, + "step": 32247 + }, + { + "epoch": 1.8017431628348746, + "grad_norm": 1.054725170135498, + "learning_rate": 0.0001000672344240251, + "loss": 0.4106, + "step": 32248 + }, + { + "epoch": 1.8017990334385563, + "grad_norm": 0.3552184998989105, + "learning_rate": 0.00010003922008068131, + "loss": 0.3547, + "step": 32249 + }, + { + "epoch": 1.801854904042238, + "grad_norm": 3.0251615047454834, + "learning_rate": 0.00010001120573733751, + "loss": 0.5341, + "step": 32250 + }, + { + "epoch": 1.8019107746459202, + "grad_norm": 2.424760580062866, + "learning_rate": 9.998319139399373e-05, + "loss": 0.5057, + "step": 32251 + }, + { + "epoch": 1.801966645249602, + "grad_norm": 0.36008310317993164, + "learning_rate": 9.995517705064994e-05, + "loss": 0.3507, + "step": 32252 + }, + { + "epoch": 1.8020225158532837, + "grad_norm": 0.41153866052627563, + "learning_rate": 9.992716270730615e-05, + "loss": 0.3514, + "step": 32253 + }, + { + "epoch": 1.8020783864569658, + "grad_norm": 0.465220183134079, + "learning_rate": 9.989914836396235e-05, + "loss": 0.3943, + "step": 32254 + }, + { + "epoch": 1.8021342570606476, + "grad_norm": 0.3958297073841095, + "learning_rate": 9.987113402061857e-05, + "loss": 0.3789, + "step": 32255 + }, + { + "epoch": 1.8021901276643293, + "grad_norm": 0.4727311432361603, + "learning_rate": 9.984311967727476e-05, + "loss": 0.4637, + "step": 32256 + }, + { + "epoch": 1.8022459982680112, + "grad_norm": 1.0321475267410278, + "learning_rate": 9.981510533393097e-05, + "loss": 0.3239, + "step": 32257 + }, + { + "epoch": 1.8023018688716932, + "grad_norm": 0.4576835334300995, + "learning_rate": 9.978709099058717e-05, + "loss": 0.3884, + "step": 32258 + }, + { + "epoch": 1.802357739475375, + "grad_norm": 0.3066233992576599, + "learning_rate": 9.97590766472434e-05, + "loss": 0.349, + "step": 32259 + }, + { + "epoch": 1.8024136100790569, + "grad_norm": 0.3274562358856201, + "learning_rate": 9.97310623038996e-05, + "loss": 0.288, + "step": 32260 + }, + { + "epoch": 1.8024694806827388, + "grad_norm": 0.4159696400165558, + "learning_rate": 9.97030479605558e-05, + "loss": 0.4175, + "step": 32261 + }, + { + "epoch": 1.8025253512864206, + "grad_norm": 0.5548173785209656, + "learning_rate": 9.967503361721201e-05, + "loss": 0.4809, + "step": 32262 + }, + { + "epoch": 1.8025812218901025, + "grad_norm": 0.4937436878681183, + "learning_rate": 9.964701927386823e-05, + "loss": 0.4392, + "step": 32263 + }, + { + "epoch": 1.8026370924937845, + "grad_norm": 0.7380626201629639, + "learning_rate": 9.961900493052444e-05, + "loss": 0.521, + "step": 32264 + }, + { + "epoch": 1.8026929630974662, + "grad_norm": 0.31832385063171387, + "learning_rate": 9.959099058718064e-05, + "loss": 0.3435, + "step": 32265 + }, + { + "epoch": 1.8027488337011481, + "grad_norm": 0.429177850484848, + "learning_rate": 9.956297624383685e-05, + "loss": 0.4025, + "step": 32266 + }, + { + "epoch": 1.80280470430483, + "grad_norm": 0.7510767579078674, + "learning_rate": 9.953496190049305e-05, + "loss": 0.4559, + "step": 32267 + }, + { + "epoch": 1.8028605749085118, + "grad_norm": 0.7886911630630493, + "learning_rate": 9.950694755714926e-05, + "loss": 0.3598, + "step": 32268 + }, + { + "epoch": 1.8029164455121938, + "grad_norm": 0.3964192569255829, + "learning_rate": 9.947893321380547e-05, + "loss": 0.4182, + "step": 32269 + }, + { + "epoch": 1.8029723161158757, + "grad_norm": 0.5220678448677063, + "learning_rate": 9.945091887046167e-05, + "loss": 0.506, + "step": 32270 + }, + { + "epoch": 1.8030281867195574, + "grad_norm": 0.3598325550556183, + "learning_rate": 9.942290452711789e-05, + "loss": 0.4856, + "step": 32271 + }, + { + "epoch": 1.8030840573232394, + "grad_norm": 0.35104647278785706, + "learning_rate": 9.93948901837741e-05, + "loss": 0.4266, + "step": 32272 + }, + { + "epoch": 1.8031399279269213, + "grad_norm": 1.7198710441589355, + "learning_rate": 9.93668758404303e-05, + "loss": 0.3145, + "step": 32273 + }, + { + "epoch": 1.803195798530603, + "grad_norm": 0.59626305103302, + "learning_rate": 9.933886149708651e-05, + "loss": 0.372, + "step": 32274 + }, + { + "epoch": 1.803251669134285, + "grad_norm": 0.40367433428764343, + "learning_rate": 9.931084715374273e-05, + "loss": 0.39, + "step": 32275 + }, + { + "epoch": 1.803307539737967, + "grad_norm": 1.0778961181640625, + "learning_rate": 9.928283281039893e-05, + "loss": 0.3644, + "step": 32276 + }, + { + "epoch": 1.8033634103416487, + "grad_norm": 0.3544921576976776, + "learning_rate": 9.925481846705513e-05, + "loss": 0.3388, + "step": 32277 + }, + { + "epoch": 1.8034192809453307, + "grad_norm": 0.5058964490890503, + "learning_rate": 9.922680412371133e-05, + "loss": 0.4914, + "step": 32278 + }, + { + "epoch": 1.8034751515490126, + "grad_norm": 0.33501556515693665, + "learning_rate": 9.919878978036755e-05, + "loss": 0.3053, + "step": 32279 + }, + { + "epoch": 1.8035310221526943, + "grad_norm": 0.8155794739723206, + "learning_rate": 9.917077543702376e-05, + "loss": 0.4423, + "step": 32280 + }, + { + "epoch": 1.803586892756376, + "grad_norm": 0.4715481102466583, + "learning_rate": 9.914276109367996e-05, + "loss": 0.4366, + "step": 32281 + }, + { + "epoch": 1.8036427633600582, + "grad_norm": 0.4038259983062744, + "learning_rate": 9.911474675033617e-05, + "loss": 0.3377, + "step": 32282 + }, + { + "epoch": 1.80369863396374, + "grad_norm": 0.4356226921081543, + "learning_rate": 9.908673240699239e-05, + "loss": 0.3326, + "step": 32283 + }, + { + "epoch": 1.8037545045674217, + "grad_norm": 5.346273899078369, + "learning_rate": 9.90587180636486e-05, + "loss": 0.3698, + "step": 32284 + }, + { + "epoch": 1.8038103751711039, + "grad_norm": 0.46056148409843445, + "learning_rate": 9.90307037203048e-05, + "loss": 0.5015, + "step": 32285 + }, + { + "epoch": 1.8038662457747856, + "grad_norm": 0.5268808007240295, + "learning_rate": 9.9002689376961e-05, + "loss": 0.381, + "step": 32286 + }, + { + "epoch": 1.8039221163784673, + "grad_norm": 0.4492740333080292, + "learning_rate": 9.897467503361722e-05, + "loss": 0.4175, + "step": 32287 + }, + { + "epoch": 1.8039779869821495, + "grad_norm": 0.6001808047294617, + "learning_rate": 9.894666069027342e-05, + "loss": 0.5724, + "step": 32288 + }, + { + "epoch": 1.8040338575858312, + "grad_norm": 0.47358494997024536, + "learning_rate": 9.891864634692962e-05, + "loss": 0.3509, + "step": 32289 + }, + { + "epoch": 1.804089728189513, + "grad_norm": 0.466736376285553, + "learning_rate": 9.889063200358583e-05, + "loss": 0.3714, + "step": 32290 + }, + { + "epoch": 1.804145598793195, + "grad_norm": 0.6926239728927612, + "learning_rate": 9.886261766024205e-05, + "loss": 0.4739, + "step": 32291 + }, + { + "epoch": 1.8042014693968769, + "grad_norm": 0.46614205837249756, + "learning_rate": 9.883460331689825e-05, + "loss": 0.4029, + "step": 32292 + }, + { + "epoch": 1.8042573400005586, + "grad_norm": 0.3578733205795288, + "learning_rate": 9.880658897355446e-05, + "loss": 0.361, + "step": 32293 + }, + { + "epoch": 1.8043132106042405, + "grad_norm": 0.6907479763031006, + "learning_rate": 9.877857463021067e-05, + "loss": 0.4697, + "step": 32294 + }, + { + "epoch": 1.8043690812079225, + "grad_norm": 0.3246879279613495, + "learning_rate": 9.875056028686688e-05, + "loss": 0.3094, + "step": 32295 + }, + { + "epoch": 1.8044249518116042, + "grad_norm": 0.3317734897136688, + "learning_rate": 9.872254594352309e-05, + "loss": 0.3353, + "step": 32296 + }, + { + "epoch": 1.8044808224152862, + "grad_norm": 4.45686674118042, + "learning_rate": 9.86945316001793e-05, + "loss": 0.3928, + "step": 32297 + }, + { + "epoch": 1.8045366930189681, + "grad_norm": 0.838014543056488, + "learning_rate": 9.86665172568355e-05, + "loss": 0.6037, + "step": 32298 + }, + { + "epoch": 1.8045925636226499, + "grad_norm": 1.342090368270874, + "learning_rate": 9.863850291349171e-05, + "loss": 0.3177, + "step": 32299 + }, + { + "epoch": 1.8046484342263318, + "grad_norm": 0.6915030479431152, + "learning_rate": 9.861048857014791e-05, + "loss": 0.3475, + "step": 32300 + }, + { + "epoch": 1.8047043048300138, + "grad_norm": 0.36533787846565247, + "learning_rate": 9.858247422680412e-05, + "loss": 0.3959, + "step": 32301 + }, + { + "epoch": 1.8047601754336955, + "grad_norm": 0.5275892615318298, + "learning_rate": 9.855445988346033e-05, + "loss": 0.5158, + "step": 32302 + }, + { + "epoch": 1.8048160460373774, + "grad_norm": 1.498059630393982, + "learning_rate": 9.852644554011654e-05, + "loss": 0.3841, + "step": 32303 + }, + { + "epoch": 1.8048719166410594, + "grad_norm": 0.7770636677742004, + "learning_rate": 9.849843119677275e-05, + "loss": 0.4367, + "step": 32304 + }, + { + "epoch": 1.8049277872447411, + "grad_norm": 0.45057666301727295, + "learning_rate": 9.847041685342896e-05, + "loss": 0.4643, + "step": 32305 + }, + { + "epoch": 1.804983657848423, + "grad_norm": 0.40695643424987793, + "learning_rate": 9.844240251008518e-05, + "loss": 0.419, + "step": 32306 + }, + { + "epoch": 1.805039528452105, + "grad_norm": 1.6682591438293457, + "learning_rate": 9.841438816674138e-05, + "loss": 0.4899, + "step": 32307 + }, + { + "epoch": 1.8050953990557868, + "grad_norm": 0.6680001020431519, + "learning_rate": 9.838637382339759e-05, + "loss": 0.5622, + "step": 32308 + }, + { + "epoch": 1.8051512696594687, + "grad_norm": 0.9332525730133057, + "learning_rate": 9.835835948005378e-05, + "loss": 0.4601, + "step": 32309 + }, + { + "epoch": 1.8052071402631507, + "grad_norm": 0.7501946091651917, + "learning_rate": 9.833034513671e-05, + "loss": 0.3704, + "step": 32310 + }, + { + "epoch": 1.8052630108668324, + "grad_norm": 0.4567849636077881, + "learning_rate": 9.83023307933662e-05, + "loss": 0.3351, + "step": 32311 + }, + { + "epoch": 1.8053188814705143, + "grad_norm": 0.549588680267334, + "learning_rate": 9.827431645002241e-05, + "loss": 0.3591, + "step": 32312 + }, + { + "epoch": 1.8053747520741963, + "grad_norm": 0.309909850358963, + "learning_rate": 9.824630210667862e-05, + "loss": 0.3766, + "step": 32313 + }, + { + "epoch": 1.805430622677878, + "grad_norm": 0.45730501413345337, + "learning_rate": 9.821828776333484e-05, + "loss": 0.5297, + "step": 32314 + }, + { + "epoch": 1.8054864932815597, + "grad_norm": 0.39213719964027405, + "learning_rate": 9.819027341999104e-05, + "loss": 0.4598, + "step": 32315 + }, + { + "epoch": 1.805542363885242, + "grad_norm": 0.5937513113021851, + "learning_rate": 9.816225907664725e-05, + "loss": 0.4047, + "step": 32316 + }, + { + "epoch": 1.8055982344889236, + "grad_norm": 0.5528897047042847, + "learning_rate": 9.813424473330345e-05, + "loss": 0.4876, + "step": 32317 + }, + { + "epoch": 1.8056541050926054, + "grad_norm": 0.34053757786750793, + "learning_rate": 9.810623038995967e-05, + "loss": 0.3638, + "step": 32318 + }, + { + "epoch": 1.8057099756962876, + "grad_norm": 0.43883299827575684, + "learning_rate": 9.807821604661588e-05, + "loss": 0.3754, + "step": 32319 + }, + { + "epoch": 1.8057658462999693, + "grad_norm": 0.41843605041503906, + "learning_rate": 9.805020170327207e-05, + "loss": 0.4144, + "step": 32320 + }, + { + "epoch": 1.805821716903651, + "grad_norm": 0.6696506142616272, + "learning_rate": 9.802218735992828e-05, + "loss": 0.4199, + "step": 32321 + }, + { + "epoch": 1.8058775875073332, + "grad_norm": 0.4944695234298706, + "learning_rate": 9.79941730165845e-05, + "loss": 0.4257, + "step": 32322 + }, + { + "epoch": 1.805933458111015, + "grad_norm": 0.35491445660591125, + "learning_rate": 9.79661586732407e-05, + "loss": 0.3551, + "step": 32323 + }, + { + "epoch": 1.8059893287146966, + "grad_norm": 0.6000068187713623, + "learning_rate": 9.793814432989691e-05, + "loss": 0.442, + "step": 32324 + }, + { + "epoch": 1.8060451993183786, + "grad_norm": 0.6292842030525208, + "learning_rate": 9.791012998655311e-05, + "loss": 0.5191, + "step": 32325 + }, + { + "epoch": 1.8061010699220605, + "grad_norm": 0.4662516415119171, + "learning_rate": 9.788211564320933e-05, + "loss": 0.3566, + "step": 32326 + }, + { + "epoch": 1.8061569405257423, + "grad_norm": 0.42551425099372864, + "learning_rate": 9.785410129986554e-05, + "loss": 0.3672, + "step": 32327 + }, + { + "epoch": 1.8062128111294242, + "grad_norm": 0.32744449377059937, + "learning_rate": 9.782608695652174e-05, + "loss": 0.3824, + "step": 32328 + }, + { + "epoch": 1.8062686817331062, + "grad_norm": 1.7681734561920166, + "learning_rate": 9.779807261317795e-05, + "loss": 0.3371, + "step": 32329 + }, + { + "epoch": 1.806324552336788, + "grad_norm": 2.287611246109009, + "learning_rate": 9.777005826983417e-05, + "loss": 0.372, + "step": 32330 + }, + { + "epoch": 1.8063804229404699, + "grad_norm": 0.3303265869617462, + "learning_rate": 9.774204392649036e-05, + "loss": 0.4123, + "step": 32331 + }, + { + "epoch": 1.8064362935441518, + "grad_norm": 0.5138015747070312, + "learning_rate": 9.771402958314657e-05, + "loss": 0.4163, + "step": 32332 + }, + { + "epoch": 1.8064921641478335, + "grad_norm": 0.33786430954933167, + "learning_rate": 9.768601523980277e-05, + "loss": 0.279, + "step": 32333 + }, + { + "epoch": 1.8065480347515155, + "grad_norm": 0.4266160726547241, + "learning_rate": 9.765800089645899e-05, + "loss": 0.3526, + "step": 32334 + }, + { + "epoch": 1.8066039053551974, + "grad_norm": 0.39610591530799866, + "learning_rate": 9.76299865531152e-05, + "loss": 0.4553, + "step": 32335 + }, + { + "epoch": 1.8066597759588792, + "grad_norm": 0.46305301785469055, + "learning_rate": 9.76019722097714e-05, + "loss": 0.3911, + "step": 32336 + }, + { + "epoch": 1.8067156465625611, + "grad_norm": 0.4321480691432953, + "learning_rate": 9.757395786642761e-05, + "loss": 0.5497, + "step": 32337 + }, + { + "epoch": 1.806771517166243, + "grad_norm": 1.0371084213256836, + "learning_rate": 9.754594352308383e-05, + "loss": 0.4211, + "step": 32338 + }, + { + "epoch": 1.8068273877699248, + "grad_norm": 0.5585001111030579, + "learning_rate": 9.751792917974004e-05, + "loss": 0.3752, + "step": 32339 + }, + { + "epoch": 1.8068832583736067, + "grad_norm": 0.38322365283966064, + "learning_rate": 9.748991483639624e-05, + "loss": 0.384, + "step": 32340 + }, + { + "epoch": 1.8069391289772887, + "grad_norm": 0.48952341079711914, + "learning_rate": 9.746190049305243e-05, + "loss": 0.505, + "step": 32341 + }, + { + "epoch": 1.8069949995809704, + "grad_norm": 0.5849735140800476, + "learning_rate": 9.743388614970865e-05, + "loss": 0.4656, + "step": 32342 + }, + { + "epoch": 1.8070508701846524, + "grad_norm": 0.4024132788181305, + "learning_rate": 9.740587180636486e-05, + "loss": 0.4366, + "step": 32343 + }, + { + "epoch": 1.8071067407883343, + "grad_norm": 0.3614051938056946, + "learning_rate": 9.737785746302106e-05, + "loss": 0.4119, + "step": 32344 + }, + { + "epoch": 1.807162611392016, + "grad_norm": 8.846004486083984, + "learning_rate": 9.734984311967727e-05, + "loss": 0.432, + "step": 32345 + }, + { + "epoch": 1.807218481995698, + "grad_norm": 0.5499487519264221, + "learning_rate": 9.732182877633349e-05, + "loss": 0.4147, + "step": 32346 + }, + { + "epoch": 1.80727435259938, + "grad_norm": 0.3616614043712616, + "learning_rate": 9.72938144329897e-05, + "loss": 0.4213, + "step": 32347 + }, + { + "epoch": 1.8073302232030617, + "grad_norm": 0.9421991109848022, + "learning_rate": 9.72658000896459e-05, + "loss": 0.5091, + "step": 32348 + }, + { + "epoch": 1.8073860938067434, + "grad_norm": 0.5435714721679688, + "learning_rate": 9.723778574630211e-05, + "loss": 0.4255, + "step": 32349 + }, + { + "epoch": 1.8074419644104256, + "grad_norm": 0.3667694330215454, + "learning_rate": 9.720977140295833e-05, + "loss": 0.45, + "step": 32350 + }, + { + "epoch": 1.8074978350141073, + "grad_norm": 0.3578639626502991, + "learning_rate": 9.718175705961453e-05, + "loss": 0.3619, + "step": 32351 + }, + { + "epoch": 1.807553705617789, + "grad_norm": 1.3268781900405884, + "learning_rate": 9.715374271627072e-05, + "loss": 0.5025, + "step": 32352 + }, + { + "epoch": 1.8076095762214712, + "grad_norm": 0.3617112934589386, + "learning_rate": 9.712572837292693e-05, + "loss": 0.4529, + "step": 32353 + }, + { + "epoch": 1.807665446825153, + "grad_norm": 0.5730541944503784, + "learning_rate": 9.709771402958315e-05, + "loss": 0.5536, + "step": 32354 + }, + { + "epoch": 1.8077213174288347, + "grad_norm": 0.5533154606819153, + "learning_rate": 9.706969968623936e-05, + "loss": 0.5334, + "step": 32355 + }, + { + "epoch": 1.8077771880325169, + "grad_norm": 1.5146812200546265, + "learning_rate": 9.704168534289556e-05, + "loss": 0.4081, + "step": 32356 + }, + { + "epoch": 1.8078330586361986, + "grad_norm": 0.5382184982299805, + "learning_rate": 9.701367099955177e-05, + "loss": 0.4494, + "step": 32357 + }, + { + "epoch": 1.8078889292398803, + "grad_norm": 0.8223549723625183, + "learning_rate": 9.698565665620799e-05, + "loss": 0.5234, + "step": 32358 + }, + { + "epoch": 1.8079447998435623, + "grad_norm": 0.37924644351005554, + "learning_rate": 9.695764231286419e-05, + "loss": 0.3506, + "step": 32359 + }, + { + "epoch": 1.8080006704472442, + "grad_norm": 0.3750298023223877, + "learning_rate": 9.69296279695204e-05, + "loss": 0.4176, + "step": 32360 + }, + { + "epoch": 1.808056541050926, + "grad_norm": 0.687561571598053, + "learning_rate": 9.69016136261766e-05, + "loss": 0.5467, + "step": 32361 + }, + { + "epoch": 1.808112411654608, + "grad_norm": 0.7965558171272278, + "learning_rate": 9.687359928283281e-05, + "loss": 0.3109, + "step": 32362 + }, + { + "epoch": 1.8081682822582898, + "grad_norm": 0.3546595573425293, + "learning_rate": 9.684558493948902e-05, + "loss": 0.371, + "step": 32363 + }, + { + "epoch": 1.8082241528619716, + "grad_norm": 0.4211459755897522, + "learning_rate": 9.681757059614522e-05, + "loss": 0.3582, + "step": 32364 + }, + { + "epoch": 1.8082800234656535, + "grad_norm": 0.6584033966064453, + "learning_rate": 9.678955625280143e-05, + "loss": 0.3869, + "step": 32365 + }, + { + "epoch": 1.8083358940693355, + "grad_norm": 0.4094834327697754, + "learning_rate": 9.676154190945765e-05, + "loss": 0.4144, + "step": 32366 + }, + { + "epoch": 1.8083917646730172, + "grad_norm": 12.780075073242188, + "learning_rate": 9.673352756611385e-05, + "loss": 0.4277, + "step": 32367 + }, + { + "epoch": 1.8084476352766992, + "grad_norm": 3.9254770278930664, + "learning_rate": 9.670551322277006e-05, + "loss": 0.5014, + "step": 32368 + }, + { + "epoch": 1.8085035058803811, + "grad_norm": 0.5234978795051575, + "learning_rate": 9.667749887942626e-05, + "loss": 0.5289, + "step": 32369 + }, + { + "epoch": 1.8085593764840628, + "grad_norm": 0.3439868986606598, + "learning_rate": 9.664948453608248e-05, + "loss": 0.3445, + "step": 32370 + }, + { + "epoch": 1.8086152470877448, + "grad_norm": 0.4670388698577881, + "learning_rate": 9.662147019273869e-05, + "loss": 0.5066, + "step": 32371 + }, + { + "epoch": 1.8086711176914267, + "grad_norm": 0.4898632764816284, + "learning_rate": 9.65934558493949e-05, + "loss": 0.3062, + "step": 32372 + }, + { + "epoch": 1.8087269882951085, + "grad_norm": 0.6397726535797119, + "learning_rate": 9.656544150605109e-05, + "loss": 0.53, + "step": 32373 + }, + { + "epoch": 1.8087828588987904, + "grad_norm": 0.3772043287754059, + "learning_rate": 9.65374271627073e-05, + "loss": 0.4655, + "step": 32374 + }, + { + "epoch": 1.8088387295024724, + "grad_norm": 0.3091883659362793, + "learning_rate": 9.650941281936351e-05, + "loss": 0.3179, + "step": 32375 + }, + { + "epoch": 1.808894600106154, + "grad_norm": 1.1258729696273804, + "learning_rate": 9.648139847601972e-05, + "loss": 0.4147, + "step": 32376 + }, + { + "epoch": 1.808950470709836, + "grad_norm": 0.39458462595939636, + "learning_rate": 9.645338413267594e-05, + "loss": 0.3583, + "step": 32377 + }, + { + "epoch": 1.809006341313518, + "grad_norm": 0.6166146993637085, + "learning_rate": 9.642536978933214e-05, + "loss": 0.5169, + "step": 32378 + }, + { + "epoch": 1.8090622119171997, + "grad_norm": 0.38769686222076416, + "learning_rate": 9.639735544598835e-05, + "loss": 0.4111, + "step": 32379 + }, + { + "epoch": 1.8091180825208817, + "grad_norm": 0.3778383433818817, + "learning_rate": 9.636934110264455e-05, + "loss": 0.3876, + "step": 32380 + }, + { + "epoch": 1.8091739531245636, + "grad_norm": 1.3931761980056763, + "learning_rate": 9.634132675930077e-05, + "loss": 0.4169, + "step": 32381 + }, + { + "epoch": 1.8092298237282454, + "grad_norm": 0.5872327089309692, + "learning_rate": 9.631331241595698e-05, + "loss": 0.3931, + "step": 32382 + }, + { + "epoch": 1.809285694331927, + "grad_norm": 2.121551275253296, + "learning_rate": 9.628529807261319e-05, + "loss": 0.3993, + "step": 32383 + }, + { + "epoch": 1.8093415649356093, + "grad_norm": 0.861067533493042, + "learning_rate": 9.625728372926938e-05, + "loss": 0.3345, + "step": 32384 + }, + { + "epoch": 1.809397435539291, + "grad_norm": 0.41388311982154846, + "learning_rate": 9.62292693859256e-05, + "loss": 0.3916, + "step": 32385 + }, + { + "epoch": 1.8094533061429727, + "grad_norm": 0.3483756482601166, + "learning_rate": 9.62012550425818e-05, + "loss": 0.3571, + "step": 32386 + }, + { + "epoch": 1.809509176746655, + "grad_norm": 3.2946152687072754, + "learning_rate": 9.617324069923801e-05, + "loss": 0.543, + "step": 32387 + }, + { + "epoch": 1.8095650473503366, + "grad_norm": 0.5983998775482178, + "learning_rate": 9.614522635589422e-05, + "loss": 0.3846, + "step": 32388 + }, + { + "epoch": 1.8096209179540184, + "grad_norm": 0.318753182888031, + "learning_rate": 9.611721201255043e-05, + "loss": 0.3289, + "step": 32389 + }, + { + "epoch": 1.8096767885577003, + "grad_norm": 0.8671567440032959, + "learning_rate": 9.608919766920664e-05, + "loss": 0.438, + "step": 32390 + }, + { + "epoch": 1.8097326591613823, + "grad_norm": 1.4143060445785522, + "learning_rate": 9.606118332586285e-05, + "loss": 0.5584, + "step": 32391 + }, + { + "epoch": 1.809788529765064, + "grad_norm": 0.5030646324157715, + "learning_rate": 9.603316898251905e-05, + "loss": 0.4418, + "step": 32392 + }, + { + "epoch": 1.809844400368746, + "grad_norm": 0.5610209703445435, + "learning_rate": 9.600515463917527e-05, + "loss": 0.4238, + "step": 32393 + }, + { + "epoch": 1.809900270972428, + "grad_norm": 0.397064745426178, + "learning_rate": 9.597714029583146e-05, + "loss": 0.4369, + "step": 32394 + }, + { + "epoch": 1.8099561415761096, + "grad_norm": 1.7195106744766235, + "learning_rate": 9.594912595248767e-05, + "loss": 0.387, + "step": 32395 + }, + { + "epoch": 1.8100120121797916, + "grad_norm": 0.3547340929508209, + "learning_rate": 9.592111160914388e-05, + "loss": 0.397, + "step": 32396 + }, + { + "epoch": 1.8100678827834735, + "grad_norm": 0.7112062573432922, + "learning_rate": 9.58930972658001e-05, + "loss": 0.4534, + "step": 32397 + }, + { + "epoch": 1.8101237533871553, + "grad_norm": 0.2996442914009094, + "learning_rate": 9.58650829224563e-05, + "loss": 0.3527, + "step": 32398 + }, + { + "epoch": 1.8101796239908372, + "grad_norm": 0.3367116451263428, + "learning_rate": 9.58370685791125e-05, + "loss": 0.4217, + "step": 32399 + }, + { + "epoch": 1.8102354945945192, + "grad_norm": 0.34167835116386414, + "learning_rate": 9.580905423576871e-05, + "loss": 0.3975, + "step": 32400 + }, + { + "epoch": 1.8102913651982009, + "grad_norm": 0.3740330636501312, + "learning_rate": 9.578103989242493e-05, + "loss": 0.3952, + "step": 32401 + }, + { + "epoch": 1.8103472358018828, + "grad_norm": 0.8206799030303955, + "learning_rate": 9.575302554908114e-05, + "loss": 0.4204, + "step": 32402 + }, + { + "epoch": 1.8104031064055648, + "grad_norm": 0.37736383080482483, + "learning_rate": 9.572501120573734e-05, + "loss": 0.386, + "step": 32403 + }, + { + "epoch": 1.8104589770092465, + "grad_norm": 0.6835668683052063, + "learning_rate": 9.569699686239355e-05, + "loss": 0.4296, + "step": 32404 + }, + { + "epoch": 1.8105148476129285, + "grad_norm": 4.272944450378418, + "learning_rate": 9.566898251904975e-05, + "loss": 0.4633, + "step": 32405 + }, + { + "epoch": 1.8105707182166104, + "grad_norm": 0.34161943197250366, + "learning_rate": 9.564096817570596e-05, + "loss": 0.4569, + "step": 32406 + }, + { + "epoch": 1.8106265888202921, + "grad_norm": 0.3690085709095001, + "learning_rate": 9.561295383236217e-05, + "loss": 0.3383, + "step": 32407 + }, + { + "epoch": 1.810682459423974, + "grad_norm": 0.5325530767440796, + "learning_rate": 9.558493948901837e-05, + "loss": 0.3779, + "step": 32408 + }, + { + "epoch": 1.810738330027656, + "grad_norm": 0.591647744178772, + "learning_rate": 9.555692514567459e-05, + "loss": 0.3291, + "step": 32409 + }, + { + "epoch": 1.8107942006313378, + "grad_norm": 0.9353770613670349, + "learning_rate": 9.55289108023308e-05, + "loss": 0.4534, + "step": 32410 + }, + { + "epoch": 1.8108500712350197, + "grad_norm": 4.076931476593018, + "learning_rate": 9.5500896458987e-05, + "loss": 0.5462, + "step": 32411 + }, + { + "epoch": 1.8109059418387017, + "grad_norm": 0.5061087012290955, + "learning_rate": 9.547288211564321e-05, + "loss": 0.4727, + "step": 32412 + }, + { + "epoch": 1.8109618124423834, + "grad_norm": 0.3583645522594452, + "learning_rate": 9.544486777229943e-05, + "loss": 0.4116, + "step": 32413 + }, + { + "epoch": 1.8110176830460654, + "grad_norm": 0.7196466326713562, + "learning_rate": 9.541685342895563e-05, + "loss": 0.3877, + "step": 32414 + }, + { + "epoch": 1.8110735536497473, + "grad_norm": 0.36492520570755005, + "learning_rate": 9.538883908561184e-05, + "loss": 0.4042, + "step": 32415 + }, + { + "epoch": 1.811129424253429, + "grad_norm": 1.2580244541168213, + "learning_rate": 9.536082474226803e-05, + "loss": 0.4591, + "step": 32416 + }, + { + "epoch": 1.8111852948571108, + "grad_norm": 0.46374940872192383, + "learning_rate": 9.533281039892425e-05, + "loss": 0.3944, + "step": 32417 + }, + { + "epoch": 1.811241165460793, + "grad_norm": 0.42924970388412476, + "learning_rate": 9.530479605558046e-05, + "loss": 0.4137, + "step": 32418 + }, + { + "epoch": 1.8112970360644747, + "grad_norm": 0.6497094035148621, + "learning_rate": 9.527678171223666e-05, + "loss": 0.3558, + "step": 32419 + }, + { + "epoch": 1.8113529066681564, + "grad_norm": 0.5104524493217468, + "learning_rate": 9.524876736889287e-05, + "loss": 0.6551, + "step": 32420 + }, + { + "epoch": 1.8114087772718386, + "grad_norm": 0.5374468564987183, + "learning_rate": 9.522075302554909e-05, + "loss": 0.4199, + "step": 32421 + }, + { + "epoch": 1.8114646478755203, + "grad_norm": 0.9418482184410095, + "learning_rate": 9.51927386822053e-05, + "loss": 0.3108, + "step": 32422 + }, + { + "epoch": 1.811520518479202, + "grad_norm": 0.38718074560165405, + "learning_rate": 9.51647243388615e-05, + "loss": 0.3726, + "step": 32423 + }, + { + "epoch": 1.811576389082884, + "grad_norm": 0.3372994363307953, + "learning_rate": 9.51367099955177e-05, + "loss": 0.3995, + "step": 32424 + }, + { + "epoch": 1.811632259686566, + "grad_norm": 0.38491377234458923, + "learning_rate": 9.510869565217393e-05, + "loss": 0.3626, + "step": 32425 + }, + { + "epoch": 1.8116881302902477, + "grad_norm": 0.6300144791603088, + "learning_rate": 9.508068130883012e-05, + "loss": 0.3496, + "step": 32426 + }, + { + "epoch": 1.8117440008939296, + "grad_norm": 0.6204205751419067, + "learning_rate": 9.505266696548632e-05, + "loss": 0.4528, + "step": 32427 + }, + { + "epoch": 1.8117998714976116, + "grad_norm": 0.31104806065559387, + "learning_rate": 9.502465262214253e-05, + "loss": 0.3455, + "step": 32428 + }, + { + "epoch": 1.8118557421012933, + "grad_norm": 0.8238840699195862, + "learning_rate": 9.499663827879875e-05, + "loss": 0.5221, + "step": 32429 + }, + { + "epoch": 1.8119116127049752, + "grad_norm": 0.790984570980072, + "learning_rate": 9.496862393545495e-05, + "loss": 0.7236, + "step": 32430 + }, + { + "epoch": 1.8119674833086572, + "grad_norm": 0.639743983745575, + "learning_rate": 9.494060959211116e-05, + "loss": 0.4808, + "step": 32431 + }, + { + "epoch": 1.812023353912339, + "grad_norm": 0.46149566769599915, + "learning_rate": 9.491259524876737e-05, + "loss": 0.5618, + "step": 32432 + }, + { + "epoch": 1.8120792245160209, + "grad_norm": 0.595741331577301, + "learning_rate": 9.488458090542359e-05, + "loss": 0.5561, + "step": 32433 + }, + { + "epoch": 1.8121350951197028, + "grad_norm": 0.36009928584098816, + "learning_rate": 9.485656656207979e-05, + "loss": 0.4848, + "step": 32434 + }, + { + "epoch": 1.8121909657233846, + "grad_norm": 0.41629040241241455, + "learning_rate": 9.4828552218736e-05, + "loss": 0.4469, + "step": 32435 + }, + { + "epoch": 1.8122468363270665, + "grad_norm": 0.34139513969421387, + "learning_rate": 9.48005378753922e-05, + "loss": 0.3512, + "step": 32436 + }, + { + "epoch": 1.8123027069307485, + "grad_norm": 0.34173914790153503, + "learning_rate": 9.477252353204841e-05, + "loss": 0.409, + "step": 32437 + }, + { + "epoch": 1.8123585775344302, + "grad_norm": 0.3868900239467621, + "learning_rate": 9.474450918870461e-05, + "loss": 0.3612, + "step": 32438 + }, + { + "epoch": 1.8124144481381121, + "grad_norm": 0.41386130452156067, + "learning_rate": 9.471649484536082e-05, + "loss": 0.6141, + "step": 32439 + }, + { + "epoch": 1.812470318741794, + "grad_norm": 0.6090137958526611, + "learning_rate": 9.468848050201703e-05, + "loss": 0.5564, + "step": 32440 + }, + { + "epoch": 1.8125261893454758, + "grad_norm": 0.4148716628551483, + "learning_rate": 9.466046615867325e-05, + "loss": 0.4227, + "step": 32441 + }, + { + "epoch": 1.8125820599491578, + "grad_norm": 0.6034636497497559, + "learning_rate": 9.463245181532945e-05, + "loss": 0.4169, + "step": 32442 + }, + { + "epoch": 1.8126379305528397, + "grad_norm": 0.5093014240264893, + "learning_rate": 9.460443747198566e-05, + "loss": 0.4484, + "step": 32443 + }, + { + "epoch": 1.8126938011565215, + "grad_norm": 0.7229387164115906, + "learning_rate": 9.457642312864188e-05, + "loss": 0.3431, + "step": 32444 + }, + { + "epoch": 1.8127496717602034, + "grad_norm": 0.5374104976654053, + "learning_rate": 9.454840878529808e-05, + "loss": 0.446, + "step": 32445 + }, + { + "epoch": 1.8128055423638854, + "grad_norm": 0.4254245460033417, + "learning_rate": 9.452039444195429e-05, + "loss": 0.3453, + "step": 32446 + }, + { + "epoch": 1.812861412967567, + "grad_norm": 1.0132571458816528, + "learning_rate": 9.449238009861048e-05, + "loss": 0.4196, + "step": 32447 + }, + { + "epoch": 1.812917283571249, + "grad_norm": 0.37004321813583374, + "learning_rate": 9.44643657552667e-05, + "loss": 0.393, + "step": 32448 + }, + { + "epoch": 1.812973154174931, + "grad_norm": 0.39998385310173035, + "learning_rate": 9.44363514119229e-05, + "loss": 0.4964, + "step": 32449 + }, + { + "epoch": 1.8130290247786127, + "grad_norm": 0.36705076694488525, + "learning_rate": 9.440833706857911e-05, + "loss": 0.4099, + "step": 32450 + }, + { + "epoch": 1.8130848953822944, + "grad_norm": 0.482724666595459, + "learning_rate": 9.438032272523532e-05, + "loss": 0.3347, + "step": 32451 + }, + { + "epoch": 1.8131407659859766, + "grad_norm": 6.117278575897217, + "learning_rate": 9.435230838189154e-05, + "loss": 0.4364, + "step": 32452 + }, + { + "epoch": 1.8131966365896584, + "grad_norm": 0.574894905090332, + "learning_rate": 9.432429403854774e-05, + "loss": 0.4498, + "step": 32453 + }, + { + "epoch": 1.81325250719334, + "grad_norm": 0.4854852557182312, + "learning_rate": 9.429627969520395e-05, + "loss": 0.6327, + "step": 32454 + }, + { + "epoch": 1.8133083777970223, + "grad_norm": 8.404415130615234, + "learning_rate": 9.426826535186015e-05, + "loss": 0.4427, + "step": 32455 + }, + { + "epoch": 1.813364248400704, + "grad_norm": 13.511711120605469, + "learning_rate": 9.424025100851637e-05, + "loss": 0.3736, + "step": 32456 + }, + { + "epoch": 1.8134201190043857, + "grad_norm": 0.4131469428539276, + "learning_rate": 9.421223666517258e-05, + "loss": 0.3772, + "step": 32457 + }, + { + "epoch": 1.8134759896080677, + "grad_norm": 0.4815536141395569, + "learning_rate": 9.418422232182877e-05, + "loss": 0.6008, + "step": 32458 + }, + { + "epoch": 1.8135318602117496, + "grad_norm": 0.7317570447921753, + "learning_rate": 9.415620797848498e-05, + "loss": 0.4502, + "step": 32459 + }, + { + "epoch": 1.8135877308154313, + "grad_norm": 0.32637640833854675, + "learning_rate": 9.41281936351412e-05, + "loss": 0.3343, + "step": 32460 + }, + { + "epoch": 1.8136436014191133, + "grad_norm": 0.5268176198005676, + "learning_rate": 9.41001792917974e-05, + "loss": 0.4156, + "step": 32461 + }, + { + "epoch": 1.8136994720227952, + "grad_norm": 8.957411766052246, + "learning_rate": 9.407216494845361e-05, + "loss": 0.3951, + "step": 32462 + }, + { + "epoch": 1.813755342626477, + "grad_norm": 0.9244331121444702, + "learning_rate": 9.404415060510981e-05, + "loss": 0.493, + "step": 32463 + }, + { + "epoch": 1.813811213230159, + "grad_norm": 0.3888901472091675, + "learning_rate": 9.401613626176603e-05, + "loss": 0.4068, + "step": 32464 + }, + { + "epoch": 1.8138670838338409, + "grad_norm": 0.3766835331916809, + "learning_rate": 9.398812191842224e-05, + "loss": 0.3852, + "step": 32465 + }, + { + "epoch": 1.8139229544375226, + "grad_norm": 0.9352009296417236, + "learning_rate": 9.396010757507844e-05, + "loss": 0.4479, + "step": 32466 + }, + { + "epoch": 1.8139788250412046, + "grad_norm": 0.527678370475769, + "learning_rate": 9.393209323173465e-05, + "loss": 0.4437, + "step": 32467 + }, + { + "epoch": 1.8140346956448865, + "grad_norm": 0.4973185956478119, + "learning_rate": 9.390407888839087e-05, + "loss": 0.5018, + "step": 32468 + }, + { + "epoch": 1.8140905662485682, + "grad_norm": 0.7872640490531921, + "learning_rate": 9.387606454504706e-05, + "loss": 0.3632, + "step": 32469 + }, + { + "epoch": 1.8141464368522502, + "grad_norm": 0.42138397693634033, + "learning_rate": 9.384805020170327e-05, + "loss": 0.3167, + "step": 32470 + }, + { + "epoch": 1.8142023074559321, + "grad_norm": 0.33858823776245117, + "learning_rate": 9.382003585835947e-05, + "loss": 0.441, + "step": 32471 + }, + { + "epoch": 1.8142581780596139, + "grad_norm": 0.5937570333480835, + "learning_rate": 9.37920215150157e-05, + "loss": 0.3725, + "step": 32472 + }, + { + "epoch": 1.8143140486632958, + "grad_norm": 0.33960288763046265, + "learning_rate": 9.37640071716719e-05, + "loss": 0.3741, + "step": 32473 + }, + { + "epoch": 1.8143699192669778, + "grad_norm": 0.3542039692401886, + "learning_rate": 9.37359928283281e-05, + "loss": 0.4344, + "step": 32474 + }, + { + "epoch": 1.8144257898706595, + "grad_norm": 0.4551481604576111, + "learning_rate": 9.370797848498431e-05, + "loss": 0.4598, + "step": 32475 + }, + { + "epoch": 1.8144816604743415, + "grad_norm": 1.519762635231018, + "learning_rate": 9.367996414164053e-05, + "loss": 0.5157, + "step": 32476 + }, + { + "epoch": 1.8145375310780234, + "grad_norm": 0.8059765696525574, + "learning_rate": 9.365194979829674e-05, + "loss": 0.4612, + "step": 32477 + }, + { + "epoch": 1.8145934016817051, + "grad_norm": 0.3982715308666229, + "learning_rate": 9.362393545495294e-05, + "loss": 0.4821, + "step": 32478 + }, + { + "epoch": 1.814649272285387, + "grad_norm": 1.9304505586624146, + "learning_rate": 9.359592111160913e-05, + "loss": 0.4705, + "step": 32479 + }, + { + "epoch": 1.814705142889069, + "grad_norm": 0.7005486488342285, + "learning_rate": 9.356790676826535e-05, + "loss": 0.3612, + "step": 32480 + }, + { + "epoch": 1.8147610134927508, + "grad_norm": 0.5514013171195984, + "learning_rate": 9.353989242492156e-05, + "loss": 0.5342, + "step": 32481 + }, + { + "epoch": 1.8148168840964327, + "grad_norm": 0.4358164668083191, + "learning_rate": 9.351187808157776e-05, + "loss": 0.3722, + "step": 32482 + }, + { + "epoch": 1.8148727547001147, + "grad_norm": 0.34742069244384766, + "learning_rate": 9.348386373823397e-05, + "loss": 0.397, + "step": 32483 + }, + { + "epoch": 1.8149286253037964, + "grad_norm": 0.38549110293388367, + "learning_rate": 9.345584939489019e-05, + "loss": 0.3371, + "step": 32484 + }, + { + "epoch": 1.8149844959074781, + "grad_norm": 0.49271920323371887, + "learning_rate": 9.34278350515464e-05, + "loss": 0.3435, + "step": 32485 + }, + { + "epoch": 1.8150403665111603, + "grad_norm": 0.31246986985206604, + "learning_rate": 9.33998207082026e-05, + "loss": 0.3785, + "step": 32486 + }, + { + "epoch": 1.815096237114842, + "grad_norm": 0.3757127821445465, + "learning_rate": 9.337180636485881e-05, + "loss": 0.4056, + "step": 32487 + }, + { + "epoch": 1.8151521077185238, + "grad_norm": 4.149107456207275, + "learning_rate": 9.334379202151503e-05, + "loss": 0.4735, + "step": 32488 + }, + { + "epoch": 1.815207978322206, + "grad_norm": 0.45842796564102173, + "learning_rate": 9.331577767817123e-05, + "loss": 0.3905, + "step": 32489 + }, + { + "epoch": 1.8152638489258877, + "grad_norm": 4.028582572937012, + "learning_rate": 9.328776333482743e-05, + "loss": 0.328, + "step": 32490 + }, + { + "epoch": 1.8153197195295694, + "grad_norm": 0.5515214800834656, + "learning_rate": 9.325974899148363e-05, + "loss": 0.3701, + "step": 32491 + }, + { + "epoch": 1.8153755901332513, + "grad_norm": 0.40889137983322144, + "learning_rate": 9.323173464813985e-05, + "loss": 0.5239, + "step": 32492 + }, + { + "epoch": 1.8154314607369333, + "grad_norm": 0.42583924531936646, + "learning_rate": 9.320372030479606e-05, + "loss": 0.4549, + "step": 32493 + }, + { + "epoch": 1.815487331340615, + "grad_norm": 0.32265082001686096, + "learning_rate": 9.317570596145226e-05, + "loss": 0.3766, + "step": 32494 + }, + { + "epoch": 1.815543201944297, + "grad_norm": 0.4477250576019287, + "learning_rate": 9.314769161810847e-05, + "loss": 0.4148, + "step": 32495 + }, + { + "epoch": 1.815599072547979, + "grad_norm": 0.36804038286209106, + "learning_rate": 9.311967727476469e-05, + "loss": 0.4194, + "step": 32496 + }, + { + "epoch": 1.8156549431516606, + "grad_norm": 0.40665021538734436, + "learning_rate": 9.309166293142089e-05, + "loss": 0.4148, + "step": 32497 + }, + { + "epoch": 1.8157108137553426, + "grad_norm": 0.6071017980575562, + "learning_rate": 9.30636485880771e-05, + "loss": 0.4607, + "step": 32498 + }, + { + "epoch": 1.8157666843590246, + "grad_norm": 0.7141764760017395, + "learning_rate": 9.30356342447333e-05, + "loss": 0.4069, + "step": 32499 + }, + { + "epoch": 1.8158225549627063, + "grad_norm": 0.5293461084365845, + "learning_rate": 9.300761990138952e-05, + "loss": 0.3154, + "step": 32500 + }, + { + "epoch": 1.8158225549627063, + "eval_cer": 0.08312875737837278, + "eval_loss": 0.3127972483634949, + "eval_runtime": 55.8824, + "eval_samples_per_second": 81.206, + "eval_steps_per_second": 5.082, + "eval_wer": 0.32962766403173094, + "step": 32500 + }, + { + "epoch": 1.8158784255663882, + "grad_norm": 0.6653088331222534, + "learning_rate": 9.297960555804572e-05, + "loss": 0.4206, + "step": 32501 + }, + { + "epoch": 1.8159342961700702, + "grad_norm": 0.46232250332832336, + "learning_rate": 9.295159121470192e-05, + "loss": 0.3288, + "step": 32502 + }, + { + "epoch": 1.815990166773752, + "grad_norm": 0.8732602000236511, + "learning_rate": 9.292357687135813e-05, + "loss": 0.4148, + "step": 32503 + }, + { + "epoch": 1.8160460373774339, + "grad_norm": 0.43408825993537903, + "learning_rate": 9.289556252801435e-05, + "loss": 0.5136, + "step": 32504 + }, + { + "epoch": 1.8161019079811158, + "grad_norm": 0.3185996413230896, + "learning_rate": 9.286754818467055e-05, + "loss": 0.3735, + "step": 32505 + }, + { + "epoch": 1.8161577785847975, + "grad_norm": 1.7085633277893066, + "learning_rate": 9.283953384132676e-05, + "loss": 0.4421, + "step": 32506 + }, + { + "epoch": 1.8162136491884795, + "grad_norm": 1.2258549928665161, + "learning_rate": 9.281151949798298e-05, + "loss": 0.4959, + "step": 32507 + }, + { + "epoch": 1.8162695197921614, + "grad_norm": 0.793949544429779, + "learning_rate": 9.278350515463918e-05, + "loss": 0.4115, + "step": 32508 + }, + { + "epoch": 1.8163253903958432, + "grad_norm": 1.221781611442566, + "learning_rate": 9.275549081129539e-05, + "loss": 0.4119, + "step": 32509 + }, + { + "epoch": 1.8163812609995251, + "grad_norm": 0.564397394657135, + "learning_rate": 9.27274764679516e-05, + "loss": 0.5183, + "step": 32510 + }, + { + "epoch": 1.816437131603207, + "grad_norm": 0.3755481541156769, + "learning_rate": 9.26994621246078e-05, + "loss": 0.4507, + "step": 32511 + }, + { + "epoch": 1.8164930022068888, + "grad_norm": 0.3223164975643158, + "learning_rate": 9.267144778126401e-05, + "loss": 0.3771, + "step": 32512 + }, + { + "epoch": 1.8165488728105708, + "grad_norm": 0.3327215611934662, + "learning_rate": 9.264343343792021e-05, + "loss": 0.4307, + "step": 32513 + }, + { + "epoch": 1.8166047434142527, + "grad_norm": 0.3800283968448639, + "learning_rate": 9.261541909457642e-05, + "loss": 0.4043, + "step": 32514 + }, + { + "epoch": 1.8166606140179344, + "grad_norm": 0.3115030825138092, + "learning_rate": 9.258740475123264e-05, + "loss": 0.3303, + "step": 32515 + }, + { + "epoch": 1.8167164846216164, + "grad_norm": 0.8653485774993896, + "learning_rate": 9.255939040788884e-05, + "loss": 0.3827, + "step": 32516 + }, + { + "epoch": 1.8167723552252983, + "grad_norm": 0.41242116689682007, + "learning_rate": 9.253137606454505e-05, + "loss": 0.4281, + "step": 32517 + }, + { + "epoch": 1.81682822582898, + "grad_norm": 0.3482825458049774, + "learning_rate": 9.250336172120126e-05, + "loss": 0.5055, + "step": 32518 + }, + { + "epoch": 1.8168840964326618, + "grad_norm": 0.5130195617675781, + "learning_rate": 9.247534737785747e-05, + "loss": 0.4456, + "step": 32519 + }, + { + "epoch": 1.816939967036344, + "grad_norm": 0.4884238541126251, + "learning_rate": 9.244733303451368e-05, + "loss": 0.4138, + "step": 32520 + }, + { + "epoch": 1.8169958376400257, + "grad_norm": 0.30373722314834595, + "learning_rate": 9.241931869116989e-05, + "loss": 0.3797, + "step": 32521 + }, + { + "epoch": 1.8170517082437074, + "grad_norm": 2.5771100521087646, + "learning_rate": 9.239130434782608e-05, + "loss": 0.3613, + "step": 32522 + }, + { + "epoch": 1.8171075788473896, + "grad_norm": 0.8482797145843506, + "learning_rate": 9.23632900044823e-05, + "loss": 0.4081, + "step": 32523 + }, + { + "epoch": 1.8171634494510713, + "grad_norm": 0.7645508646965027, + "learning_rate": 9.23352756611385e-05, + "loss": 0.5125, + "step": 32524 + }, + { + "epoch": 1.817219320054753, + "grad_norm": 0.48069506883621216, + "learning_rate": 9.230726131779471e-05, + "loss": 0.4547, + "step": 32525 + }, + { + "epoch": 1.817275190658435, + "grad_norm": 0.5193696022033691, + "learning_rate": 9.227924697445092e-05, + "loss": 0.5152, + "step": 32526 + }, + { + "epoch": 1.817331061262117, + "grad_norm": 3.2294743061065674, + "learning_rate": 9.225123263110714e-05, + "loss": 0.46, + "step": 32527 + }, + { + "epoch": 1.8173869318657987, + "grad_norm": 0.4781751036643982, + "learning_rate": 9.222321828776334e-05, + "loss": 0.3145, + "step": 32528 + }, + { + "epoch": 1.8174428024694806, + "grad_norm": 0.4096963703632355, + "learning_rate": 9.219520394441955e-05, + "loss": 0.5265, + "step": 32529 + }, + { + "epoch": 1.8174986730731626, + "grad_norm": 0.5728110074996948, + "learning_rate": 9.216718960107575e-05, + "loss": 0.3617, + "step": 32530 + }, + { + "epoch": 1.8175545436768443, + "grad_norm": 0.9301915168762207, + "learning_rate": 9.213917525773197e-05, + "loss": 0.3862, + "step": 32531 + }, + { + "epoch": 1.8176104142805263, + "grad_norm": 1.1300740242004395, + "learning_rate": 9.211116091438816e-05, + "loss": 0.3545, + "step": 32532 + }, + { + "epoch": 1.8176662848842082, + "grad_norm": 0.29928433895111084, + "learning_rate": 9.208314657104437e-05, + "loss": 0.3078, + "step": 32533 + }, + { + "epoch": 1.81772215548789, + "grad_norm": 0.34859973192214966, + "learning_rate": 9.205513222770058e-05, + "loss": 0.3542, + "step": 32534 + }, + { + "epoch": 1.817778026091572, + "grad_norm": 0.3184810280799866, + "learning_rate": 9.20271178843568e-05, + "loss": 0.4712, + "step": 32535 + }, + { + "epoch": 1.8178338966952539, + "grad_norm": 0.3844888508319855, + "learning_rate": 9.1999103541013e-05, + "loss": 0.3515, + "step": 32536 + }, + { + "epoch": 1.8178897672989356, + "grad_norm": 0.33808332681655884, + "learning_rate": 9.19710891976692e-05, + "loss": 0.3157, + "step": 32537 + }, + { + "epoch": 1.8179456379026175, + "grad_norm": 0.389936238527298, + "learning_rate": 9.194307485432541e-05, + "loss": 0.4346, + "step": 32538 + }, + { + "epoch": 1.8180015085062995, + "grad_norm": 0.456673264503479, + "learning_rate": 9.191506051098163e-05, + "loss": 0.5253, + "step": 32539 + }, + { + "epoch": 1.8180573791099812, + "grad_norm": 0.621004045009613, + "learning_rate": 9.188704616763784e-05, + "loss": 0.4898, + "step": 32540 + }, + { + "epoch": 1.8181132497136632, + "grad_norm": 0.34965384006500244, + "learning_rate": 9.185903182429404e-05, + "loss": 0.3383, + "step": 32541 + }, + { + "epoch": 1.8181691203173451, + "grad_norm": 0.5049427151679993, + "learning_rate": 9.183101748095025e-05, + "loss": 0.3878, + "step": 32542 + }, + { + "epoch": 1.8182249909210269, + "grad_norm": 0.3601321280002594, + "learning_rate": 9.180300313760646e-05, + "loss": 0.3938, + "step": 32543 + }, + { + "epoch": 1.8182808615247088, + "grad_norm": 0.3247752785682678, + "learning_rate": 9.177498879426266e-05, + "loss": 0.3332, + "step": 32544 + }, + { + "epoch": 1.8183367321283908, + "grad_norm": 1.2692288160324097, + "learning_rate": 9.174697445091887e-05, + "loss": 0.5147, + "step": 32545 + }, + { + "epoch": 1.8183926027320725, + "grad_norm": 0.446115642786026, + "learning_rate": 9.171896010757507e-05, + "loss": 0.3404, + "step": 32546 + }, + { + "epoch": 1.8184484733357544, + "grad_norm": 0.30747470259666443, + "learning_rate": 9.169094576423129e-05, + "loss": 0.3719, + "step": 32547 + }, + { + "epoch": 1.8185043439394364, + "grad_norm": 0.44200581312179565, + "learning_rate": 9.16629314208875e-05, + "loss": 0.5357, + "step": 32548 + }, + { + "epoch": 1.8185602145431181, + "grad_norm": 1.3370931148529053, + "learning_rate": 9.16349170775437e-05, + "loss": 0.4258, + "step": 32549 + }, + { + "epoch": 1.8186160851468, + "grad_norm": 0.8436364531517029, + "learning_rate": 9.160690273419991e-05, + "loss": 0.3464, + "step": 32550 + }, + { + "epoch": 1.818671955750482, + "grad_norm": 0.40700194239616394, + "learning_rate": 9.157888839085613e-05, + "loss": 0.6691, + "step": 32551 + }, + { + "epoch": 1.8187278263541637, + "grad_norm": 0.759835958480835, + "learning_rate": 9.155087404751233e-05, + "loss": 0.3941, + "step": 32552 + }, + { + "epoch": 1.8187836969578455, + "grad_norm": 0.4287680685520172, + "learning_rate": 9.152285970416854e-05, + "loss": 0.3481, + "step": 32553 + }, + { + "epoch": 1.8188395675615276, + "grad_norm": 0.49506762623786926, + "learning_rate": 9.149484536082473e-05, + "loss": 0.378, + "step": 32554 + }, + { + "epoch": 1.8188954381652094, + "grad_norm": 0.3504091799259186, + "learning_rate": 9.146683101748095e-05, + "loss": 0.3735, + "step": 32555 + }, + { + "epoch": 1.818951308768891, + "grad_norm": 0.4317473769187927, + "learning_rate": 9.143881667413716e-05, + "loss": 0.6264, + "step": 32556 + }, + { + "epoch": 1.8190071793725733, + "grad_norm": 0.37420693039894104, + "learning_rate": 9.141080233079336e-05, + "loss": 0.405, + "step": 32557 + }, + { + "epoch": 1.819063049976255, + "grad_norm": 0.44333672523498535, + "learning_rate": 9.138278798744957e-05, + "loss": 0.443, + "step": 32558 + }, + { + "epoch": 1.8191189205799367, + "grad_norm": 0.32927170395851135, + "learning_rate": 9.135477364410579e-05, + "loss": 0.3675, + "step": 32559 + }, + { + "epoch": 1.8191747911836187, + "grad_norm": 0.40413957834243774, + "learning_rate": 9.1326759300762e-05, + "loss": 0.4086, + "step": 32560 + }, + { + "epoch": 1.8192306617873006, + "grad_norm": 0.601325273513794, + "learning_rate": 9.12987449574182e-05, + "loss": 0.3406, + "step": 32561 + }, + { + "epoch": 1.8192865323909824, + "grad_norm": 0.42418405413627625, + "learning_rate": 9.12707306140744e-05, + "loss": 0.4133, + "step": 32562 + }, + { + "epoch": 1.8193424029946643, + "grad_norm": 0.6964157223701477, + "learning_rate": 9.124271627073063e-05, + "loss": 0.4076, + "step": 32563 + }, + { + "epoch": 1.8193982735983463, + "grad_norm": 0.6728548407554626, + "learning_rate": 9.121470192738682e-05, + "loss": 0.5458, + "step": 32564 + }, + { + "epoch": 1.819454144202028, + "grad_norm": 0.5280389785766602, + "learning_rate": 9.118668758404302e-05, + "loss": 0.3603, + "step": 32565 + }, + { + "epoch": 1.81951001480571, + "grad_norm": 0.4482826292514801, + "learning_rate": 9.115867324069923e-05, + "loss": 0.4723, + "step": 32566 + }, + { + "epoch": 1.819565885409392, + "grad_norm": 0.36412355303764343, + "learning_rate": 9.113065889735545e-05, + "loss": 0.4413, + "step": 32567 + }, + { + "epoch": 1.8196217560130736, + "grad_norm": 1.2507410049438477, + "learning_rate": 9.110264455401165e-05, + "loss": 0.5626, + "step": 32568 + }, + { + "epoch": 1.8196776266167556, + "grad_norm": 0.4253033399581909, + "learning_rate": 9.107463021066786e-05, + "loss": 0.3676, + "step": 32569 + }, + { + "epoch": 1.8197334972204375, + "grad_norm": 0.4385090470314026, + "learning_rate": 9.104661586732407e-05, + "loss": 0.4223, + "step": 32570 + }, + { + "epoch": 1.8197893678241193, + "grad_norm": 0.3647114336490631, + "learning_rate": 9.101860152398029e-05, + "loss": 0.4136, + "step": 32571 + }, + { + "epoch": 1.8198452384278012, + "grad_norm": 1.2754600048065186, + "learning_rate": 9.099058718063649e-05, + "loss": 0.4817, + "step": 32572 + }, + { + "epoch": 1.8199011090314832, + "grad_norm": 2.7498586177825928, + "learning_rate": 9.09625728372927e-05, + "loss": 0.3783, + "step": 32573 + }, + { + "epoch": 1.819956979635165, + "grad_norm": 1.258906602859497, + "learning_rate": 9.093455849394892e-05, + "loss": 0.3204, + "step": 32574 + }, + { + "epoch": 1.8200128502388468, + "grad_norm": 0.5071038603782654, + "learning_rate": 9.090654415060511e-05, + "loss": 0.4556, + "step": 32575 + }, + { + "epoch": 1.8200687208425288, + "grad_norm": 0.43021759390830994, + "learning_rate": 9.087852980726131e-05, + "loss": 0.4361, + "step": 32576 + }, + { + "epoch": 1.8201245914462105, + "grad_norm": 0.5080472230911255, + "learning_rate": 9.085051546391752e-05, + "loss": 0.4606, + "step": 32577 + }, + { + "epoch": 1.8201804620498925, + "grad_norm": 0.7317853569984436, + "learning_rate": 9.082250112057374e-05, + "loss": 0.4146, + "step": 32578 + }, + { + "epoch": 1.8202363326535744, + "grad_norm": 0.3626214861869812, + "learning_rate": 9.079448677722995e-05, + "loss": 0.393, + "step": 32579 + }, + { + "epoch": 1.8202922032572562, + "grad_norm": 0.31248733401298523, + "learning_rate": 9.076647243388615e-05, + "loss": 0.4533, + "step": 32580 + }, + { + "epoch": 1.820348073860938, + "grad_norm": 0.3188362419605255, + "learning_rate": 9.073845809054236e-05, + "loss": 0.2848, + "step": 32581 + }, + { + "epoch": 1.82040394446462, + "grad_norm": 1.2452759742736816, + "learning_rate": 9.071044374719858e-05, + "loss": 0.3963, + "step": 32582 + }, + { + "epoch": 1.8204598150683018, + "grad_norm": 0.4821965992450714, + "learning_rate": 9.068242940385478e-05, + "loss": 0.428, + "step": 32583 + }, + { + "epoch": 1.8205156856719837, + "grad_norm": 1.2750962972640991, + "learning_rate": 9.065441506051099e-05, + "loss": 0.4502, + "step": 32584 + }, + { + "epoch": 1.8205715562756657, + "grad_norm": 0.33922410011291504, + "learning_rate": 9.06264007171672e-05, + "loss": 0.478, + "step": 32585 + }, + { + "epoch": 1.8206274268793474, + "grad_norm": 0.4323895573616028, + "learning_rate": 9.05983863738234e-05, + "loss": 0.3906, + "step": 32586 + }, + { + "epoch": 1.8206832974830292, + "grad_norm": 0.49382561445236206, + "learning_rate": 9.05703720304796e-05, + "loss": 0.3625, + "step": 32587 + }, + { + "epoch": 1.8207391680867113, + "grad_norm": 0.4754297435283661, + "learning_rate": 9.054235768713581e-05, + "loss": 0.3641, + "step": 32588 + }, + { + "epoch": 1.820795038690393, + "grad_norm": 0.35114532709121704, + "learning_rate": 9.051434334379202e-05, + "loss": 0.3814, + "step": 32589 + }, + { + "epoch": 1.8208509092940748, + "grad_norm": 0.39383646845817566, + "learning_rate": 9.048632900044824e-05, + "loss": 0.3424, + "step": 32590 + }, + { + "epoch": 1.820906779897757, + "grad_norm": 0.4086364507675171, + "learning_rate": 9.045831465710444e-05, + "loss": 0.4064, + "step": 32591 + }, + { + "epoch": 1.8209626505014387, + "grad_norm": 0.43864449858665466, + "learning_rate": 9.043030031376065e-05, + "loss": 0.3177, + "step": 32592 + }, + { + "epoch": 1.8210185211051204, + "grad_norm": 0.33234789967536926, + "learning_rate": 9.040228597041685e-05, + "loss": 0.3726, + "step": 32593 + }, + { + "epoch": 1.8210743917088024, + "grad_norm": 0.43807879090309143, + "learning_rate": 9.037427162707307e-05, + "loss": 0.5211, + "step": 32594 + }, + { + "epoch": 1.8211302623124843, + "grad_norm": 0.5509362816810608, + "learning_rate": 9.034625728372928e-05, + "loss": 0.2657, + "step": 32595 + }, + { + "epoch": 1.821186132916166, + "grad_norm": 0.3725467026233673, + "learning_rate": 9.031824294038547e-05, + "loss": 0.3407, + "step": 32596 + }, + { + "epoch": 1.821242003519848, + "grad_norm": 0.31692931056022644, + "learning_rate": 9.029022859704168e-05, + "loss": 0.3013, + "step": 32597 + }, + { + "epoch": 1.82129787412353, + "grad_norm": 0.5240983963012695, + "learning_rate": 9.02622142536979e-05, + "loss": 0.3114, + "step": 32598 + }, + { + "epoch": 1.8213537447272117, + "grad_norm": 0.8342991471290588, + "learning_rate": 9.02341999103541e-05, + "loss": 0.5483, + "step": 32599 + }, + { + "epoch": 1.8214096153308936, + "grad_norm": 0.5111089944839478, + "learning_rate": 9.020618556701031e-05, + "loss": 0.4624, + "step": 32600 + }, + { + "epoch": 1.8214654859345756, + "grad_norm": 0.8262082934379578, + "learning_rate": 9.017817122366651e-05, + "loss": 0.5797, + "step": 32601 + }, + { + "epoch": 1.8215213565382573, + "grad_norm": 6.770280361175537, + "learning_rate": 9.015015688032273e-05, + "loss": 0.4481, + "step": 32602 + }, + { + "epoch": 1.8215772271419393, + "grad_norm": 0.3912312090396881, + "learning_rate": 9.012214253697894e-05, + "loss": 0.3882, + "step": 32603 + }, + { + "epoch": 1.8216330977456212, + "grad_norm": 0.3937259912490845, + "learning_rate": 9.009412819363515e-05, + "loss": 0.5171, + "step": 32604 + }, + { + "epoch": 1.821688968349303, + "grad_norm": 0.35557982325553894, + "learning_rate": 9.006611385029135e-05, + "loss": 0.4784, + "step": 32605 + }, + { + "epoch": 1.821744838952985, + "grad_norm": 0.41458240151405334, + "learning_rate": 9.003809950694757e-05, + "loss": 0.4667, + "step": 32606 + }, + { + "epoch": 1.8218007095566668, + "grad_norm": 0.32831329107284546, + "learning_rate": 9.001008516360376e-05, + "loss": 0.2732, + "step": 32607 + }, + { + "epoch": 1.8218565801603486, + "grad_norm": 0.5389516353607178, + "learning_rate": 8.998207082025997e-05, + "loss": 0.3582, + "step": 32608 + }, + { + "epoch": 1.8219124507640305, + "grad_norm": 0.5212104916572571, + "learning_rate": 8.995405647691617e-05, + "loss": 0.5114, + "step": 32609 + }, + { + "epoch": 1.8219683213677125, + "grad_norm": 0.9086574912071228, + "learning_rate": 8.99260421335724e-05, + "loss": 0.4366, + "step": 32610 + }, + { + "epoch": 1.8220241919713942, + "grad_norm": 0.36206182837486267, + "learning_rate": 8.98980277902286e-05, + "loss": 0.4658, + "step": 32611 + }, + { + "epoch": 1.8220800625750762, + "grad_norm": 0.3922532796859741, + "learning_rate": 8.98700134468848e-05, + "loss": 0.349, + "step": 32612 + }, + { + "epoch": 1.822135933178758, + "grad_norm": 0.5684999227523804, + "learning_rate": 8.984199910354101e-05, + "loss": 0.4557, + "step": 32613 + }, + { + "epoch": 1.8221918037824398, + "grad_norm": 0.31264135241508484, + "learning_rate": 8.981398476019723e-05, + "loss": 0.354, + "step": 32614 + }, + { + "epoch": 1.8222476743861218, + "grad_norm": 0.46321630477905273, + "learning_rate": 8.978597041685344e-05, + "loss": 0.4246, + "step": 32615 + }, + { + "epoch": 1.8223035449898037, + "grad_norm": 2.6482160091400146, + "learning_rate": 8.975795607350964e-05, + "loss": 0.3381, + "step": 32616 + }, + { + "epoch": 1.8223594155934855, + "grad_norm": 0.6674730777740479, + "learning_rate": 8.972994173016583e-05, + "loss": 0.5738, + "step": 32617 + }, + { + "epoch": 1.8224152861971672, + "grad_norm": 0.6854463815689087, + "learning_rate": 8.970192738682205e-05, + "loss": 0.5493, + "step": 32618 + }, + { + "epoch": 1.8224711568008494, + "grad_norm": 0.45408138632774353, + "learning_rate": 8.967391304347826e-05, + "loss": 0.5375, + "step": 32619 + }, + { + "epoch": 1.822527027404531, + "grad_norm": 0.32394859194755554, + "learning_rate": 8.964589870013447e-05, + "loss": 0.4298, + "step": 32620 + }, + { + "epoch": 1.8225828980082128, + "grad_norm": 1.0733610391616821, + "learning_rate": 8.961788435679067e-05, + "loss": 0.3695, + "step": 32621 + }, + { + "epoch": 1.822638768611895, + "grad_norm": 0.3442154824733734, + "learning_rate": 8.958987001344689e-05, + "loss": 0.411, + "step": 32622 + }, + { + "epoch": 1.8226946392155767, + "grad_norm": 1.4539341926574707, + "learning_rate": 8.95618556701031e-05, + "loss": 0.5156, + "step": 32623 + }, + { + "epoch": 1.8227505098192585, + "grad_norm": 0.5309671759605408, + "learning_rate": 8.95338413267593e-05, + "loss": 0.3863, + "step": 32624 + }, + { + "epoch": 1.8228063804229406, + "grad_norm": 0.3928808867931366, + "learning_rate": 8.950582698341551e-05, + "loss": 0.448, + "step": 32625 + }, + { + "epoch": 1.8228622510266224, + "grad_norm": 0.3270750939846039, + "learning_rate": 8.947781264007173e-05, + "loss": 0.4001, + "step": 32626 + }, + { + "epoch": 1.822918121630304, + "grad_norm": 0.5254973769187927, + "learning_rate": 8.944979829672793e-05, + "loss": 0.444, + "step": 32627 + }, + { + "epoch": 1.822973992233986, + "grad_norm": 4.891864776611328, + "learning_rate": 8.942178395338413e-05, + "loss": 0.4452, + "step": 32628 + }, + { + "epoch": 1.823029862837668, + "grad_norm": 0.5048308968544006, + "learning_rate": 8.939376961004033e-05, + "loss": 0.4162, + "step": 32629 + }, + { + "epoch": 1.8230857334413497, + "grad_norm": 0.31015744805336, + "learning_rate": 8.936575526669655e-05, + "loss": 0.4245, + "step": 32630 + }, + { + "epoch": 1.8231416040450317, + "grad_norm": 0.5078670978546143, + "learning_rate": 8.933774092335276e-05, + "loss": 0.5102, + "step": 32631 + }, + { + "epoch": 1.8231974746487136, + "grad_norm": 0.4111892282962799, + "learning_rate": 8.930972658000896e-05, + "loss": 0.3592, + "step": 32632 + }, + { + "epoch": 1.8232533452523954, + "grad_norm": 0.4632844924926758, + "learning_rate": 8.928171223666517e-05, + "loss": 0.4187, + "step": 32633 + }, + { + "epoch": 1.8233092158560773, + "grad_norm": 0.4957256317138672, + "learning_rate": 8.925369789332139e-05, + "loss": 0.491, + "step": 32634 + }, + { + "epoch": 1.8233650864597593, + "grad_norm": 0.458670049905777, + "learning_rate": 8.92256835499776e-05, + "loss": 0.3882, + "step": 32635 + }, + { + "epoch": 1.823420957063441, + "grad_norm": 1.3238328695297241, + "learning_rate": 8.91976692066338e-05, + "loss": 0.504, + "step": 32636 + }, + { + "epoch": 1.823476827667123, + "grad_norm": 0.28553181886672974, + "learning_rate": 8.916965486329e-05, + "loss": 0.3497, + "step": 32637 + }, + { + "epoch": 1.8235326982708049, + "grad_norm": 0.3311823904514313, + "learning_rate": 8.914164051994622e-05, + "loss": 0.3437, + "step": 32638 + }, + { + "epoch": 1.8235885688744866, + "grad_norm": 0.6872339844703674, + "learning_rate": 8.911362617660242e-05, + "loss": 0.4976, + "step": 32639 + }, + { + "epoch": 1.8236444394781686, + "grad_norm": 0.36720260977745056, + "learning_rate": 8.908561183325862e-05, + "loss": 0.3927, + "step": 32640 + }, + { + "epoch": 1.8237003100818505, + "grad_norm": 0.7211514115333557, + "learning_rate": 8.905759748991483e-05, + "loss": 0.6816, + "step": 32641 + }, + { + "epoch": 1.8237561806855322, + "grad_norm": 6.280039310455322, + "learning_rate": 8.902958314657105e-05, + "loss": 0.3381, + "step": 32642 + }, + { + "epoch": 1.8238120512892142, + "grad_norm": 0.38385528326034546, + "learning_rate": 8.900156880322725e-05, + "loss": 0.3476, + "step": 32643 + }, + { + "epoch": 1.8238679218928961, + "grad_norm": 0.6742785573005676, + "learning_rate": 8.897355445988346e-05, + "loss": 0.3695, + "step": 32644 + }, + { + "epoch": 1.8239237924965779, + "grad_norm": 0.40217357873916626, + "learning_rate": 8.894554011653968e-05, + "loss": 0.4872, + "step": 32645 + }, + { + "epoch": 1.8239796631002598, + "grad_norm": 0.5788240432739258, + "learning_rate": 8.891752577319588e-05, + "loss": 0.4121, + "step": 32646 + }, + { + "epoch": 1.8240355337039418, + "grad_norm": 0.4126443862915039, + "learning_rate": 8.888951142985209e-05, + "loss": 0.3949, + "step": 32647 + }, + { + "epoch": 1.8240914043076235, + "grad_norm": 0.5141621232032776, + "learning_rate": 8.88614970865083e-05, + "loss": 0.3886, + "step": 32648 + }, + { + "epoch": 1.8241472749113055, + "grad_norm": 0.39065274596214294, + "learning_rate": 8.88334827431645e-05, + "loss": 0.4978, + "step": 32649 + }, + { + "epoch": 1.8242031455149874, + "grad_norm": 0.3403256833553314, + "learning_rate": 8.880546839982071e-05, + "loss": 0.3792, + "step": 32650 + }, + { + "epoch": 1.8242590161186691, + "grad_norm": 11.867629051208496, + "learning_rate": 8.877745405647691e-05, + "loss": 0.3815, + "step": 32651 + }, + { + "epoch": 1.8243148867223509, + "grad_norm": 0.42894306778907776, + "learning_rate": 8.874943971313312e-05, + "loss": 0.3659, + "step": 32652 + }, + { + "epoch": 1.824370757326033, + "grad_norm": 0.37356799840927124, + "learning_rate": 8.872142536978934e-05, + "loss": 0.4368, + "step": 32653 + }, + { + "epoch": 1.8244266279297148, + "grad_norm": 0.36987531185150146, + "learning_rate": 8.869341102644554e-05, + "loss": 0.3722, + "step": 32654 + }, + { + "epoch": 1.8244824985333965, + "grad_norm": 0.3708203434944153, + "learning_rate": 8.866539668310175e-05, + "loss": 0.379, + "step": 32655 + }, + { + "epoch": 1.8245383691370787, + "grad_norm": 0.5055073499679565, + "learning_rate": 8.863738233975796e-05, + "loss": 0.4357, + "step": 32656 + }, + { + "epoch": 1.8245942397407604, + "grad_norm": 0.379313200712204, + "learning_rate": 8.860936799641418e-05, + "loss": 0.4825, + "step": 32657 + }, + { + "epoch": 1.8246501103444421, + "grad_norm": 0.3618811368942261, + "learning_rate": 8.858135365307038e-05, + "loss": 0.4745, + "step": 32658 + }, + { + "epoch": 1.8247059809481243, + "grad_norm": 0.4619145691394806, + "learning_rate": 8.855333930972659e-05, + "loss": 0.4759, + "step": 32659 + }, + { + "epoch": 1.824761851551806, + "grad_norm": 0.4952583909034729, + "learning_rate": 8.852532496638278e-05, + "loss": 0.4474, + "step": 32660 + }, + { + "epoch": 1.8248177221554878, + "grad_norm": 0.3791593909263611, + "learning_rate": 8.8497310623039e-05, + "loss": 0.4493, + "step": 32661 + }, + { + "epoch": 1.8248735927591697, + "grad_norm": 0.5558996200561523, + "learning_rate": 8.84692962796952e-05, + "loss": 0.4599, + "step": 32662 + }, + { + "epoch": 1.8249294633628517, + "grad_norm": 0.48642346262931824, + "learning_rate": 8.844128193635141e-05, + "loss": 0.4961, + "step": 32663 + }, + { + "epoch": 1.8249853339665334, + "grad_norm": 0.2961120009422302, + "learning_rate": 8.841326759300762e-05, + "loss": 0.3429, + "step": 32664 + }, + { + "epoch": 1.8250412045702153, + "grad_norm": 0.4587043821811676, + "learning_rate": 8.838525324966384e-05, + "loss": 0.588, + "step": 32665 + }, + { + "epoch": 1.8250970751738973, + "grad_norm": 0.7218194007873535, + "learning_rate": 8.835723890632004e-05, + "loss": 0.5451, + "step": 32666 + }, + { + "epoch": 1.825152945777579, + "grad_norm": 0.3695017695426941, + "learning_rate": 8.832922456297625e-05, + "loss": 0.2853, + "step": 32667 + }, + { + "epoch": 1.825208816381261, + "grad_norm": 0.5091305375099182, + "learning_rate": 8.830121021963245e-05, + "loss": 0.3428, + "step": 32668 + }, + { + "epoch": 1.825264686984943, + "grad_norm": 0.5220282673835754, + "learning_rate": 8.827319587628867e-05, + "loss": 0.4625, + "step": 32669 + }, + { + "epoch": 1.8253205575886247, + "grad_norm": 0.5233119130134583, + "learning_rate": 8.824518153294488e-05, + "loss": 0.4055, + "step": 32670 + }, + { + "epoch": 1.8253764281923066, + "grad_norm": 0.4037000238895416, + "learning_rate": 8.821716718960107e-05, + "loss": 0.4712, + "step": 32671 + }, + { + "epoch": 1.8254322987959886, + "grad_norm": 0.34906062483787537, + "learning_rate": 8.818915284625728e-05, + "loss": 0.4455, + "step": 32672 + }, + { + "epoch": 1.8254881693996703, + "grad_norm": 1.0178866386413574, + "learning_rate": 8.81611385029135e-05, + "loss": 0.4179, + "step": 32673 + }, + { + "epoch": 1.8255440400033522, + "grad_norm": 0.3596630394458771, + "learning_rate": 8.81331241595697e-05, + "loss": 0.431, + "step": 32674 + }, + { + "epoch": 1.8255999106070342, + "grad_norm": 0.37298551201820374, + "learning_rate": 8.810510981622591e-05, + "loss": 0.4186, + "step": 32675 + }, + { + "epoch": 1.825655781210716, + "grad_norm": 0.47867903113365173, + "learning_rate": 8.807709547288211e-05, + "loss": 0.4823, + "step": 32676 + }, + { + "epoch": 1.8257116518143979, + "grad_norm": 0.39461806416511536, + "learning_rate": 8.804908112953833e-05, + "loss": 0.3754, + "step": 32677 + }, + { + "epoch": 1.8257675224180798, + "grad_norm": 0.37617164850234985, + "learning_rate": 8.802106678619454e-05, + "loss": 0.5202, + "step": 32678 + }, + { + "epoch": 1.8258233930217616, + "grad_norm": 0.36601707339286804, + "learning_rate": 8.799305244285074e-05, + "loss": 0.361, + "step": 32679 + }, + { + "epoch": 1.8258792636254435, + "grad_norm": 1.9883246421813965, + "learning_rate": 8.796503809950695e-05, + "loss": 0.4635, + "step": 32680 + }, + { + "epoch": 1.8259351342291255, + "grad_norm": 0.3663940727710724, + "learning_rate": 8.793702375616316e-05, + "loss": 0.4888, + "step": 32681 + }, + { + "epoch": 1.8259910048328072, + "grad_norm": 0.8663790225982666, + "learning_rate": 8.790900941281936e-05, + "loss": 0.4915, + "step": 32682 + }, + { + "epoch": 1.8260468754364891, + "grad_norm": 0.5251271724700928, + "learning_rate": 8.788099506947557e-05, + "loss": 0.3643, + "step": 32683 + }, + { + "epoch": 1.826102746040171, + "grad_norm": 1.0799036026000977, + "learning_rate": 8.785298072613177e-05, + "loss": 0.4269, + "step": 32684 + }, + { + "epoch": 1.8261586166438528, + "grad_norm": 0.8637546896934509, + "learning_rate": 8.782496638278799e-05, + "loss": 0.4241, + "step": 32685 + }, + { + "epoch": 1.8262144872475345, + "grad_norm": 0.36941349506378174, + "learning_rate": 8.77969520394442e-05, + "loss": 0.4806, + "step": 32686 + }, + { + "epoch": 1.8262703578512167, + "grad_norm": 0.5779221653938293, + "learning_rate": 8.77689376961004e-05, + "loss": 0.4361, + "step": 32687 + }, + { + "epoch": 1.8263262284548984, + "grad_norm": 0.31867095828056335, + "learning_rate": 8.774092335275661e-05, + "loss": 0.3343, + "step": 32688 + }, + { + "epoch": 1.8263820990585802, + "grad_norm": 2.622779607772827, + "learning_rate": 8.771290900941283e-05, + "loss": 0.3924, + "step": 32689 + }, + { + "epoch": 1.8264379696622624, + "grad_norm": 0.2617053985595703, + "learning_rate": 8.768489466606904e-05, + "loss": 0.3635, + "step": 32690 + }, + { + "epoch": 1.826493840265944, + "grad_norm": 0.4027954936027527, + "learning_rate": 8.765688032272524e-05, + "loss": 0.5429, + "step": 32691 + }, + { + "epoch": 1.8265497108696258, + "grad_norm": 0.38758933544158936, + "learning_rate": 8.762886597938143e-05, + "loss": 0.422, + "step": 32692 + }, + { + "epoch": 1.826605581473308, + "grad_norm": 0.412352591753006, + "learning_rate": 8.760085163603765e-05, + "loss": 0.4186, + "step": 32693 + }, + { + "epoch": 1.8266614520769897, + "grad_norm": 0.3389891982078552, + "learning_rate": 8.757283729269386e-05, + "loss": 0.4381, + "step": 32694 + }, + { + "epoch": 1.8267173226806714, + "grad_norm": 0.4800606071949005, + "learning_rate": 8.754482294935006e-05, + "loss": 0.4121, + "step": 32695 + }, + { + "epoch": 1.8267731932843534, + "grad_norm": 1.1981359720230103, + "learning_rate": 8.751680860600627e-05, + "loss": 0.4211, + "step": 32696 + }, + { + "epoch": 1.8268290638880353, + "grad_norm": 0.39636045694351196, + "learning_rate": 8.748879426266249e-05, + "loss": 0.4453, + "step": 32697 + }, + { + "epoch": 1.826884934491717, + "grad_norm": 0.3641802966594696, + "learning_rate": 8.74607799193187e-05, + "loss": 0.3735, + "step": 32698 + }, + { + "epoch": 1.826940805095399, + "grad_norm": 0.43713924288749695, + "learning_rate": 8.74327655759749e-05, + "loss": 0.4931, + "step": 32699 + }, + { + "epoch": 1.826996675699081, + "grad_norm": 0.4104662239551544, + "learning_rate": 8.740475123263111e-05, + "loss": 0.5247, + "step": 32700 + }, + { + "epoch": 1.8270525463027627, + "grad_norm": 0.4149661064147949, + "learning_rate": 8.737673688928733e-05, + "loss": 0.3914, + "step": 32701 + }, + { + "epoch": 1.8271084169064447, + "grad_norm": 1.1765981912612915, + "learning_rate": 8.734872254594352e-05, + "loss": 0.6119, + "step": 32702 + }, + { + "epoch": 1.8271642875101266, + "grad_norm": 0.3791041374206543, + "learning_rate": 8.732070820259972e-05, + "loss": 0.3413, + "step": 32703 + }, + { + "epoch": 1.8272201581138083, + "grad_norm": 0.477271169424057, + "learning_rate": 8.729269385925593e-05, + "loss": 0.3881, + "step": 32704 + }, + { + "epoch": 1.8272760287174903, + "grad_norm": 0.40712982416152954, + "learning_rate": 8.726467951591215e-05, + "loss": 0.3599, + "step": 32705 + }, + { + "epoch": 1.8273318993211722, + "grad_norm": 0.36567389965057373, + "learning_rate": 8.723666517256836e-05, + "loss": 0.468, + "step": 32706 + }, + { + "epoch": 1.827387769924854, + "grad_norm": 0.5755578279495239, + "learning_rate": 8.720865082922456e-05, + "loss": 0.4476, + "step": 32707 + }, + { + "epoch": 1.827443640528536, + "grad_norm": 1.7578039169311523, + "learning_rate": 8.718063648588077e-05, + "loss": 0.3563, + "step": 32708 + }, + { + "epoch": 1.8274995111322179, + "grad_norm": 0.8825699090957642, + "learning_rate": 8.715262214253699e-05, + "loss": 0.6931, + "step": 32709 + }, + { + "epoch": 1.8275553817358996, + "grad_norm": 0.39829063415527344, + "learning_rate": 8.712460779919319e-05, + "loss": 0.335, + "step": 32710 + }, + { + "epoch": 1.8276112523395815, + "grad_norm": 0.7966462969779968, + "learning_rate": 8.70965934558494e-05, + "loss": 0.4719, + "step": 32711 + }, + { + "epoch": 1.8276671229432635, + "grad_norm": 0.44598644971847534, + "learning_rate": 8.706857911250562e-05, + "loss": 0.4273, + "step": 32712 + }, + { + "epoch": 1.8277229935469452, + "grad_norm": 0.30554571747779846, + "learning_rate": 8.704056476916181e-05, + "loss": 0.371, + "step": 32713 + }, + { + "epoch": 1.8277788641506272, + "grad_norm": 0.4376177191734314, + "learning_rate": 8.701255042581802e-05, + "loss": 0.4366, + "step": 32714 + }, + { + "epoch": 1.8278347347543091, + "grad_norm": 0.6572985649108887, + "learning_rate": 8.698453608247422e-05, + "loss": 0.425, + "step": 32715 + }, + { + "epoch": 1.8278906053579909, + "grad_norm": 2.694395065307617, + "learning_rate": 8.695652173913044e-05, + "loss": 0.3833, + "step": 32716 + }, + { + "epoch": 1.8279464759616728, + "grad_norm": 0.39475616812705994, + "learning_rate": 8.692850739578665e-05, + "loss": 0.4693, + "step": 32717 + }, + { + "epoch": 1.8280023465653548, + "grad_norm": 0.309169203042984, + "learning_rate": 8.690049305244285e-05, + "loss": 0.3794, + "step": 32718 + }, + { + "epoch": 1.8280582171690365, + "grad_norm": 0.32120662927627563, + "learning_rate": 8.687247870909906e-05, + "loss": 0.3826, + "step": 32719 + }, + { + "epoch": 1.8281140877727182, + "grad_norm": 0.3339031934738159, + "learning_rate": 8.684446436575528e-05, + "loss": 0.356, + "step": 32720 + }, + { + "epoch": 1.8281699583764004, + "grad_norm": 0.8471054434776306, + "learning_rate": 8.681645002241148e-05, + "loss": 0.3551, + "step": 32721 + }, + { + "epoch": 1.8282258289800821, + "grad_norm": 0.38110437989234924, + "learning_rate": 8.678843567906769e-05, + "loss": 0.4169, + "step": 32722 + }, + { + "epoch": 1.8282816995837639, + "grad_norm": 0.4576033651828766, + "learning_rate": 8.67604213357239e-05, + "loss": 0.4287, + "step": 32723 + }, + { + "epoch": 1.828337570187446, + "grad_norm": 0.5527513027191162, + "learning_rate": 8.67324069923801e-05, + "loss": 0.3324, + "step": 32724 + }, + { + "epoch": 1.8283934407911278, + "grad_norm": 0.35816875100135803, + "learning_rate": 8.67043926490363e-05, + "loss": 0.359, + "step": 32725 + }, + { + "epoch": 1.8284493113948095, + "grad_norm": 0.35263440012931824, + "learning_rate": 8.667637830569251e-05, + "loss": 0.4493, + "step": 32726 + }, + { + "epoch": 1.8285051819984917, + "grad_norm": 0.6201557517051697, + "learning_rate": 8.664836396234872e-05, + "loss": 0.3523, + "step": 32727 + }, + { + "epoch": 1.8285610526021734, + "grad_norm": 0.3486427962779999, + "learning_rate": 8.662034961900494e-05, + "loss": 0.3561, + "step": 32728 + }, + { + "epoch": 1.8286169232058551, + "grad_norm": 0.5453665256500244, + "learning_rate": 8.659233527566114e-05, + "loss": 0.3742, + "step": 32729 + }, + { + "epoch": 1.828672793809537, + "grad_norm": 0.42633435130119324, + "learning_rate": 8.656432093231735e-05, + "loss": 0.3611, + "step": 32730 + }, + { + "epoch": 1.828728664413219, + "grad_norm": 0.42084944248199463, + "learning_rate": 8.653630658897356e-05, + "loss": 0.455, + "step": 32731 + }, + { + "epoch": 1.8287845350169007, + "grad_norm": 0.3635386526584625, + "learning_rate": 8.650829224562977e-05, + "loss": 0.3981, + "step": 32732 + }, + { + "epoch": 1.8288404056205827, + "grad_norm": 0.5200009346008301, + "learning_rate": 8.648027790228598e-05, + "loss": 0.4974, + "step": 32733 + }, + { + "epoch": 1.8288962762242646, + "grad_norm": 0.7889485359191895, + "learning_rate": 8.645226355894217e-05, + "loss": 0.4438, + "step": 32734 + }, + { + "epoch": 1.8289521468279464, + "grad_norm": 0.8671538829803467, + "learning_rate": 8.642424921559838e-05, + "loss": 0.3609, + "step": 32735 + }, + { + "epoch": 1.8290080174316283, + "grad_norm": 0.4722672700881958, + "learning_rate": 8.63962348722546e-05, + "loss": 0.4655, + "step": 32736 + }, + { + "epoch": 1.8290638880353103, + "grad_norm": 0.5887338519096375, + "learning_rate": 8.63682205289108e-05, + "loss": 0.4341, + "step": 32737 + }, + { + "epoch": 1.829119758638992, + "grad_norm": 0.5655964612960815, + "learning_rate": 8.634020618556701e-05, + "loss": 0.4322, + "step": 32738 + }, + { + "epoch": 1.829175629242674, + "grad_norm": 0.6668345332145691, + "learning_rate": 8.631219184222322e-05, + "loss": 0.4407, + "step": 32739 + }, + { + "epoch": 1.829231499846356, + "grad_norm": 0.5087136626243591, + "learning_rate": 8.628417749887943e-05, + "loss": 0.4677, + "step": 32740 + }, + { + "epoch": 1.8292873704500376, + "grad_norm": 0.390455424785614, + "learning_rate": 8.625616315553564e-05, + "loss": 0.3662, + "step": 32741 + }, + { + "epoch": 1.8293432410537196, + "grad_norm": 0.2725785970687866, + "learning_rate": 8.622814881219185e-05, + "loss": 0.3771, + "step": 32742 + }, + { + "epoch": 1.8293991116574015, + "grad_norm": 0.33725032210350037, + "learning_rate": 8.620013446884805e-05, + "loss": 0.3699, + "step": 32743 + }, + { + "epoch": 1.8294549822610833, + "grad_norm": 1.8284152746200562, + "learning_rate": 8.617212012550427e-05, + "loss": 0.4742, + "step": 32744 + }, + { + "epoch": 1.8295108528647652, + "grad_norm": 1.319706916809082, + "learning_rate": 8.614410578216046e-05, + "loss": 0.3637, + "step": 32745 + }, + { + "epoch": 1.8295667234684472, + "grad_norm": 0.6174635887145996, + "learning_rate": 8.611609143881667e-05, + "loss": 0.4852, + "step": 32746 + }, + { + "epoch": 1.829622594072129, + "grad_norm": 0.481128066778183, + "learning_rate": 8.608807709547288e-05, + "loss": 0.4398, + "step": 32747 + }, + { + "epoch": 1.8296784646758109, + "grad_norm": 1.6469953060150146, + "learning_rate": 8.60600627521291e-05, + "loss": 0.5777, + "step": 32748 + }, + { + "epoch": 1.8297343352794928, + "grad_norm": 0.5434072613716125, + "learning_rate": 8.60320484087853e-05, + "loss": 0.5028, + "step": 32749 + }, + { + "epoch": 1.8297902058831745, + "grad_norm": 0.3864447772502899, + "learning_rate": 8.60040340654415e-05, + "loss": 0.4762, + "step": 32750 + }, + { + "epoch": 1.8298460764868565, + "grad_norm": 0.3676880896091461, + "learning_rate": 8.597601972209771e-05, + "loss": 0.3475, + "step": 32751 + }, + { + "epoch": 1.8299019470905384, + "grad_norm": 0.3302278220653534, + "learning_rate": 8.594800537875393e-05, + "loss": 0.3371, + "step": 32752 + }, + { + "epoch": 1.8299578176942202, + "grad_norm": 1.4054367542266846, + "learning_rate": 8.591999103541014e-05, + "loss": 0.3317, + "step": 32753 + }, + { + "epoch": 1.830013688297902, + "grad_norm": 0.6814680099487305, + "learning_rate": 8.589197669206634e-05, + "loss": 0.4964, + "step": 32754 + }, + { + "epoch": 1.830069558901584, + "grad_norm": 0.7393497824668884, + "learning_rate": 8.586396234872255e-05, + "loss": 0.4134, + "step": 32755 + }, + { + "epoch": 1.8301254295052658, + "grad_norm": 0.5243531465530396, + "learning_rate": 8.583594800537875e-05, + "loss": 0.5843, + "step": 32756 + }, + { + "epoch": 1.8301813001089475, + "grad_norm": 0.25018471479415894, + "learning_rate": 8.580793366203496e-05, + "loss": 0.2559, + "step": 32757 + }, + { + "epoch": 1.8302371707126297, + "grad_norm": 0.41834771633148193, + "learning_rate": 8.577991931869117e-05, + "loss": 0.3271, + "step": 32758 + }, + { + "epoch": 1.8302930413163114, + "grad_norm": 0.5412806868553162, + "learning_rate": 8.575190497534737e-05, + "loss": 0.5505, + "step": 32759 + }, + { + "epoch": 1.8303489119199932, + "grad_norm": 0.36346435546875, + "learning_rate": 8.572389063200359e-05, + "loss": 0.4526, + "step": 32760 + }, + { + "epoch": 1.8304047825236753, + "grad_norm": 0.5245670676231384, + "learning_rate": 8.56958762886598e-05, + "loss": 0.4034, + "step": 32761 + }, + { + "epoch": 1.830460653127357, + "grad_norm": 0.2890668213367462, + "learning_rate": 8.5667861945316e-05, + "loss": 0.3134, + "step": 32762 + }, + { + "epoch": 1.8305165237310388, + "grad_norm": 0.3523153066635132, + "learning_rate": 8.563984760197221e-05, + "loss": 0.3602, + "step": 32763 + }, + { + "epoch": 1.8305723943347207, + "grad_norm": 0.30864500999450684, + "learning_rate": 8.561183325862843e-05, + "loss": 0.3411, + "step": 32764 + }, + { + "epoch": 1.8306282649384027, + "grad_norm": 0.5289540886878967, + "learning_rate": 8.558381891528463e-05, + "loss": 0.4287, + "step": 32765 + }, + { + "epoch": 1.8306841355420844, + "grad_norm": 0.4329964816570282, + "learning_rate": 8.555580457194083e-05, + "loss": 0.3953, + "step": 32766 + }, + { + "epoch": 1.8307400061457664, + "grad_norm": 0.45230424404144287, + "learning_rate": 8.552779022859703e-05, + "loss": 0.4198, + "step": 32767 + }, + { + "epoch": 1.8307958767494483, + "grad_norm": 0.361547589302063, + "learning_rate": 8.549977588525325e-05, + "loss": 0.3492, + "step": 32768 + }, + { + "epoch": 1.83085174735313, + "grad_norm": 1.4699985980987549, + "learning_rate": 8.547176154190946e-05, + "loss": 0.391, + "step": 32769 + }, + { + "epoch": 1.830907617956812, + "grad_norm": 0.3823464512825012, + "learning_rate": 8.544374719856566e-05, + "loss": 0.4073, + "step": 32770 + }, + { + "epoch": 1.830963488560494, + "grad_norm": 0.3744846284389496, + "learning_rate": 8.541573285522187e-05, + "loss": 0.3767, + "step": 32771 + }, + { + "epoch": 1.8310193591641757, + "grad_norm": 0.47874176502227783, + "learning_rate": 8.538771851187809e-05, + "loss": 0.3697, + "step": 32772 + }, + { + "epoch": 1.8310752297678576, + "grad_norm": 0.3408127725124359, + "learning_rate": 8.53597041685343e-05, + "loss": 0.3952, + "step": 32773 + }, + { + "epoch": 1.8311311003715396, + "grad_norm": 0.42382287979125977, + "learning_rate": 8.53316898251905e-05, + "loss": 0.4216, + "step": 32774 + }, + { + "epoch": 1.8311869709752213, + "grad_norm": 0.3834471106529236, + "learning_rate": 8.53036754818467e-05, + "loss": 0.4042, + "step": 32775 + }, + { + "epoch": 1.8312428415789033, + "grad_norm": 0.39370232820510864, + "learning_rate": 8.527566113850293e-05, + "loss": 0.3241, + "step": 32776 + }, + { + "epoch": 1.8312987121825852, + "grad_norm": 0.5100993514060974, + "learning_rate": 8.524764679515912e-05, + "loss": 0.3861, + "step": 32777 + }, + { + "epoch": 1.831354582786267, + "grad_norm": 0.7251783609390259, + "learning_rate": 8.521963245181532e-05, + "loss": 0.3369, + "step": 32778 + }, + { + "epoch": 1.831410453389949, + "grad_norm": 0.4337495267391205, + "learning_rate": 8.519161810847153e-05, + "loss": 0.446, + "step": 32779 + }, + { + "epoch": 1.8314663239936309, + "grad_norm": 0.4384002387523651, + "learning_rate": 8.516360376512775e-05, + "loss": 0.4343, + "step": 32780 + }, + { + "epoch": 1.8315221945973126, + "grad_norm": 0.5903328657150269, + "learning_rate": 8.513558942178395e-05, + "loss": 0.861, + "step": 32781 + }, + { + "epoch": 1.8315780652009945, + "grad_norm": 3.40973162651062, + "learning_rate": 8.510757507844016e-05, + "loss": 0.3851, + "step": 32782 + }, + { + "epoch": 1.8316339358046765, + "grad_norm": 0.36614251136779785, + "learning_rate": 8.507956073509638e-05, + "loss": 0.442, + "step": 32783 + }, + { + "epoch": 1.8316898064083582, + "grad_norm": 0.3395480513572693, + "learning_rate": 8.505154639175259e-05, + "loss": 0.3471, + "step": 32784 + }, + { + "epoch": 1.8317456770120402, + "grad_norm": 0.3579520881175995, + "learning_rate": 8.502353204840879e-05, + "loss": 0.3392, + "step": 32785 + }, + { + "epoch": 1.8318015476157221, + "grad_norm": 0.8224445581436157, + "learning_rate": 8.4995517705065e-05, + "loss": 0.5659, + "step": 32786 + }, + { + "epoch": 1.8318574182194038, + "grad_norm": 0.783920168876648, + "learning_rate": 8.49675033617212e-05, + "loss": 0.3919, + "step": 32787 + }, + { + "epoch": 1.8319132888230856, + "grad_norm": 0.5099315047264099, + "learning_rate": 8.493948901837741e-05, + "loss": 0.3965, + "step": 32788 + }, + { + "epoch": 1.8319691594267677, + "grad_norm": 0.4985925853252411, + "learning_rate": 8.491147467503361e-05, + "loss": 0.4291, + "step": 32789 + }, + { + "epoch": 1.8320250300304495, + "grad_norm": 2.1264545917510986, + "learning_rate": 8.488346033168982e-05, + "loss": 0.3495, + "step": 32790 + }, + { + "epoch": 1.8320809006341312, + "grad_norm": 0.3220621347427368, + "learning_rate": 8.485544598834604e-05, + "loss": 0.2964, + "step": 32791 + }, + { + "epoch": 1.8321367712378134, + "grad_norm": 0.41862067580223083, + "learning_rate": 8.482743164500225e-05, + "loss": 0.4602, + "step": 32792 + }, + { + "epoch": 1.832192641841495, + "grad_norm": 0.8739029169082642, + "learning_rate": 8.479941730165845e-05, + "loss": 0.4587, + "step": 32793 + }, + { + "epoch": 1.8322485124451768, + "grad_norm": 0.5607500672340393, + "learning_rate": 8.477140295831466e-05, + "loss": 0.5012, + "step": 32794 + }, + { + "epoch": 1.832304383048859, + "grad_norm": 0.4431382417678833, + "learning_rate": 8.474338861497088e-05, + "loss": 0.441, + "step": 32795 + }, + { + "epoch": 1.8323602536525407, + "grad_norm": 0.4366603493690491, + "learning_rate": 8.471537427162708e-05, + "loss": 0.3171, + "step": 32796 + }, + { + "epoch": 1.8324161242562225, + "grad_norm": 0.43217548727989197, + "learning_rate": 8.468735992828329e-05, + "loss": 0.4397, + "step": 32797 + }, + { + "epoch": 1.8324719948599044, + "grad_norm": 0.8155645132064819, + "learning_rate": 8.465934558493948e-05, + "loss": 0.2923, + "step": 32798 + }, + { + "epoch": 1.8325278654635864, + "grad_norm": 0.39099061489105225, + "learning_rate": 8.46313312415957e-05, + "loss": 0.5009, + "step": 32799 + }, + { + "epoch": 1.832583736067268, + "grad_norm": 1.2317041158676147, + "learning_rate": 8.46033168982519e-05, + "loss": 0.4146, + "step": 32800 + }, + { + "epoch": 1.83263960667095, + "grad_norm": 0.7692990899085999, + "learning_rate": 8.457530255490811e-05, + "loss": 0.4457, + "step": 32801 + }, + { + "epoch": 1.832695477274632, + "grad_norm": 0.29813510179519653, + "learning_rate": 8.454728821156432e-05, + "loss": 0.3421, + "step": 32802 + }, + { + "epoch": 1.8327513478783137, + "grad_norm": 0.3592136800289154, + "learning_rate": 8.451927386822054e-05, + "loss": 0.4354, + "step": 32803 + }, + { + "epoch": 1.8328072184819957, + "grad_norm": 0.4014095366001129, + "learning_rate": 8.449125952487674e-05, + "loss": 0.4687, + "step": 32804 + }, + { + "epoch": 1.8328630890856776, + "grad_norm": 0.45243772864341736, + "learning_rate": 8.446324518153295e-05, + "loss": 0.4044, + "step": 32805 + }, + { + "epoch": 1.8329189596893594, + "grad_norm": 0.4634053111076355, + "learning_rate": 8.443523083818915e-05, + "loss": 0.3999, + "step": 32806 + }, + { + "epoch": 1.8329748302930413, + "grad_norm": 0.29330140352249146, + "learning_rate": 8.440721649484537e-05, + "loss": 0.3379, + "step": 32807 + }, + { + "epoch": 1.8330307008967233, + "grad_norm": 0.616818368434906, + "learning_rate": 8.437920215150158e-05, + "loss": 0.6101, + "step": 32808 + }, + { + "epoch": 1.833086571500405, + "grad_norm": 0.389492005109787, + "learning_rate": 8.435118780815777e-05, + "loss": 0.4034, + "step": 32809 + }, + { + "epoch": 1.833142442104087, + "grad_norm": 0.4052749276161194, + "learning_rate": 8.432317346481398e-05, + "loss": 0.4737, + "step": 32810 + }, + { + "epoch": 1.833198312707769, + "grad_norm": 1.462843656539917, + "learning_rate": 8.42951591214702e-05, + "loss": 0.413, + "step": 32811 + }, + { + "epoch": 1.8332541833114506, + "grad_norm": 0.37385934591293335, + "learning_rate": 8.42671447781264e-05, + "loss": 0.4261, + "step": 32812 + }, + { + "epoch": 1.8333100539151326, + "grad_norm": 0.31689080595970154, + "learning_rate": 8.423913043478261e-05, + "loss": 0.4143, + "step": 32813 + }, + { + "epoch": 1.8333659245188145, + "grad_norm": 0.34548068046569824, + "learning_rate": 8.421111609143881e-05, + "loss": 0.4606, + "step": 32814 + }, + { + "epoch": 1.8334217951224963, + "grad_norm": 0.6340919137001038, + "learning_rate": 8.418310174809503e-05, + "loss": 0.3598, + "step": 32815 + }, + { + "epoch": 1.8334776657261782, + "grad_norm": 0.498259037733078, + "learning_rate": 8.415508740475124e-05, + "loss": 0.4742, + "step": 32816 + }, + { + "epoch": 1.8335335363298602, + "grad_norm": 0.3974669873714447, + "learning_rate": 8.412707306140744e-05, + "loss": 0.3455, + "step": 32817 + }, + { + "epoch": 1.8335894069335419, + "grad_norm": 0.3461504578590393, + "learning_rate": 8.409905871806365e-05, + "loss": 0.3742, + "step": 32818 + }, + { + "epoch": 1.8336452775372238, + "grad_norm": 0.418295681476593, + "learning_rate": 8.407104437471986e-05, + "loss": 0.3614, + "step": 32819 + }, + { + "epoch": 1.8337011481409058, + "grad_norm": 3.940268039703369, + "learning_rate": 8.404303003137606e-05, + "loss": 0.3556, + "step": 32820 + }, + { + "epoch": 1.8337570187445875, + "grad_norm": 0.7775163054466248, + "learning_rate": 8.401501568803227e-05, + "loss": 0.4348, + "step": 32821 + }, + { + "epoch": 1.8338128893482692, + "grad_norm": 0.3466717302799225, + "learning_rate": 8.398700134468847e-05, + "loss": 0.3777, + "step": 32822 + }, + { + "epoch": 1.8338687599519514, + "grad_norm": 2.3898892402648926, + "learning_rate": 8.39589870013447e-05, + "loss": 0.4285, + "step": 32823 + }, + { + "epoch": 1.8339246305556332, + "grad_norm": 0.4038471579551697, + "learning_rate": 8.39309726580009e-05, + "loss": 0.433, + "step": 32824 + }, + { + "epoch": 1.8339805011593149, + "grad_norm": 0.429552286863327, + "learning_rate": 8.39029583146571e-05, + "loss": 0.4318, + "step": 32825 + }, + { + "epoch": 1.834036371762997, + "grad_norm": 0.7589590549468994, + "learning_rate": 8.387494397131331e-05, + "loss": 0.4485, + "step": 32826 + }, + { + "epoch": 1.8340922423666788, + "grad_norm": 0.4279058575630188, + "learning_rate": 8.384692962796953e-05, + "loss": 0.4857, + "step": 32827 + }, + { + "epoch": 1.8341481129703605, + "grad_norm": 0.524907112121582, + "learning_rate": 8.381891528462574e-05, + "loss": 0.4295, + "step": 32828 + }, + { + "epoch": 1.8342039835740425, + "grad_norm": 0.48243772983551025, + "learning_rate": 8.379090094128194e-05, + "loss": 0.3112, + "step": 32829 + }, + { + "epoch": 1.8342598541777244, + "grad_norm": 0.7055292725563049, + "learning_rate": 8.376288659793813e-05, + "loss": 0.4336, + "step": 32830 + }, + { + "epoch": 1.8343157247814061, + "grad_norm": 0.5159034132957458, + "learning_rate": 8.373487225459435e-05, + "loss": 0.4852, + "step": 32831 + }, + { + "epoch": 1.834371595385088, + "grad_norm": 0.7204219698905945, + "learning_rate": 8.370685791125056e-05, + "loss": 0.5139, + "step": 32832 + }, + { + "epoch": 1.83442746598877, + "grad_norm": 0.40596917271614075, + "learning_rate": 8.367884356790677e-05, + "loss": 0.4092, + "step": 32833 + }, + { + "epoch": 1.8344833365924518, + "grad_norm": 3.1423590183258057, + "learning_rate": 8.365082922456297e-05, + "loss": 0.5731, + "step": 32834 + }, + { + "epoch": 1.8345392071961337, + "grad_norm": 1.5014262199401855, + "learning_rate": 8.362281488121919e-05, + "loss": 0.42, + "step": 32835 + }, + { + "epoch": 1.8345950777998157, + "grad_norm": 0.3620970845222473, + "learning_rate": 8.35948005378754e-05, + "loss": 0.4509, + "step": 32836 + }, + { + "epoch": 1.8346509484034974, + "grad_norm": 0.329174280166626, + "learning_rate": 8.35667861945316e-05, + "loss": 0.3316, + "step": 32837 + }, + { + "epoch": 1.8347068190071794, + "grad_norm": 0.7065838575363159, + "learning_rate": 8.353877185118781e-05, + "loss": 0.4862, + "step": 32838 + }, + { + "epoch": 1.8347626896108613, + "grad_norm": 0.4205038547515869, + "learning_rate": 8.351075750784403e-05, + "loss": 0.4475, + "step": 32839 + }, + { + "epoch": 1.834818560214543, + "grad_norm": 0.30798256397247314, + "learning_rate": 8.348274316450023e-05, + "loss": 0.4092, + "step": 32840 + }, + { + "epoch": 1.834874430818225, + "grad_norm": 0.4675159752368927, + "learning_rate": 8.345472882115643e-05, + "loss": 0.4106, + "step": 32841 + }, + { + "epoch": 1.834930301421907, + "grad_norm": 0.5133211016654968, + "learning_rate": 8.342671447781263e-05, + "loss": 0.3982, + "step": 32842 + }, + { + "epoch": 1.8349861720255887, + "grad_norm": 2.366584300994873, + "learning_rate": 8.339870013446885e-05, + "loss": 0.349, + "step": 32843 + }, + { + "epoch": 1.8350420426292706, + "grad_norm": 0.44677338004112244, + "learning_rate": 8.337068579112506e-05, + "loss": 0.3791, + "step": 32844 + }, + { + "epoch": 1.8350979132329526, + "grad_norm": 0.3541729152202606, + "learning_rate": 8.334267144778126e-05, + "loss": 0.4056, + "step": 32845 + }, + { + "epoch": 1.8351537838366343, + "grad_norm": 0.6423866152763367, + "learning_rate": 8.331465710443748e-05, + "loss": 0.3843, + "step": 32846 + }, + { + "epoch": 1.8352096544403163, + "grad_norm": 0.934320330619812, + "learning_rate": 8.328664276109369e-05, + "loss": 0.408, + "step": 32847 + }, + { + "epoch": 1.8352655250439982, + "grad_norm": 0.41240328550338745, + "learning_rate": 8.325862841774989e-05, + "loss": 0.4212, + "step": 32848 + }, + { + "epoch": 1.83532139564768, + "grad_norm": 0.5745297074317932, + "learning_rate": 8.32306140744061e-05, + "loss": 0.4374, + "step": 32849 + }, + { + "epoch": 1.8353772662513619, + "grad_norm": 0.4433744251728058, + "learning_rate": 8.320259973106232e-05, + "loss": 0.4182, + "step": 32850 + }, + { + "epoch": 1.8354331368550438, + "grad_norm": 0.3076957166194916, + "learning_rate": 8.317458538771851e-05, + "loss": 0.3725, + "step": 32851 + }, + { + "epoch": 1.8354890074587256, + "grad_norm": 7.030282974243164, + "learning_rate": 8.314657104437472e-05, + "loss": 0.4812, + "step": 32852 + }, + { + "epoch": 1.8355448780624075, + "grad_norm": 0.38968372344970703, + "learning_rate": 8.311855670103092e-05, + "loss": 0.4228, + "step": 32853 + }, + { + "epoch": 1.8356007486660895, + "grad_norm": 2.060056209564209, + "learning_rate": 8.309054235768714e-05, + "loss": 0.3785, + "step": 32854 + }, + { + "epoch": 1.8356566192697712, + "grad_norm": 0.3281131088733673, + "learning_rate": 8.306252801434335e-05, + "loss": 0.3007, + "step": 32855 + }, + { + "epoch": 1.835712489873453, + "grad_norm": 0.2588299810886383, + "learning_rate": 8.303451367099955e-05, + "loss": 0.3122, + "step": 32856 + }, + { + "epoch": 1.835768360477135, + "grad_norm": 0.7498422861099243, + "learning_rate": 8.300649932765576e-05, + "loss": 0.377, + "step": 32857 + }, + { + "epoch": 1.8358242310808168, + "grad_norm": 0.42063140869140625, + "learning_rate": 8.297848498431198e-05, + "loss": 0.4639, + "step": 32858 + }, + { + "epoch": 1.8358801016844986, + "grad_norm": 0.3155960440635681, + "learning_rate": 8.295047064096818e-05, + "loss": 0.3744, + "step": 32859 + }, + { + "epoch": 1.8359359722881807, + "grad_norm": 0.38158515095710754, + "learning_rate": 8.292245629762439e-05, + "loss": 0.4516, + "step": 32860 + }, + { + "epoch": 1.8359918428918625, + "grad_norm": 0.4004345238208771, + "learning_rate": 8.28944419542806e-05, + "loss": 0.4388, + "step": 32861 + }, + { + "epoch": 1.8360477134955442, + "grad_norm": 0.7741023302078247, + "learning_rate": 8.28664276109368e-05, + "loss": 0.4203, + "step": 32862 + }, + { + "epoch": 1.8361035840992261, + "grad_norm": 0.4512966275215149, + "learning_rate": 8.283841326759301e-05, + "loss": 0.4725, + "step": 32863 + }, + { + "epoch": 1.836159454702908, + "grad_norm": 0.3649766743183136, + "learning_rate": 8.281039892424921e-05, + "loss": 0.371, + "step": 32864 + }, + { + "epoch": 1.8362153253065898, + "grad_norm": 1.0246310234069824, + "learning_rate": 8.278238458090542e-05, + "loss": 0.4236, + "step": 32865 + }, + { + "epoch": 1.8362711959102718, + "grad_norm": 0.5901214480400085, + "learning_rate": 8.275437023756164e-05, + "loss": 0.4909, + "step": 32866 + }, + { + "epoch": 1.8363270665139537, + "grad_norm": 0.4318733811378479, + "learning_rate": 8.272635589421784e-05, + "loss": 0.3688, + "step": 32867 + }, + { + "epoch": 1.8363829371176354, + "grad_norm": 0.4441783130168915, + "learning_rate": 8.269834155087405e-05, + "loss": 0.3788, + "step": 32868 + }, + { + "epoch": 1.8364388077213174, + "grad_norm": 0.6645383238792419, + "learning_rate": 8.267032720753026e-05, + "loss": 0.4537, + "step": 32869 + }, + { + "epoch": 1.8364946783249994, + "grad_norm": 0.3449426591396332, + "learning_rate": 8.264231286418648e-05, + "loss": 0.3657, + "step": 32870 + }, + { + "epoch": 1.836550548928681, + "grad_norm": 0.5484299063682556, + "learning_rate": 8.261429852084268e-05, + "loss": 0.4518, + "step": 32871 + }, + { + "epoch": 1.836606419532363, + "grad_norm": 0.5816342234611511, + "learning_rate": 8.258628417749887e-05, + "loss": 0.4371, + "step": 32872 + }, + { + "epoch": 1.836662290136045, + "grad_norm": 0.3770466446876526, + "learning_rate": 8.255826983415508e-05, + "loss": 0.4082, + "step": 32873 + }, + { + "epoch": 1.8367181607397267, + "grad_norm": 1.2297581434249878, + "learning_rate": 8.25302554908113e-05, + "loss": 0.4131, + "step": 32874 + }, + { + "epoch": 1.8367740313434087, + "grad_norm": 0.44993889331817627, + "learning_rate": 8.25022411474675e-05, + "loss": 0.4049, + "step": 32875 + }, + { + "epoch": 1.8368299019470906, + "grad_norm": 0.6340737342834473, + "learning_rate": 8.247422680412371e-05, + "loss": 0.427, + "step": 32876 + }, + { + "epoch": 1.8368857725507723, + "grad_norm": 0.34926480054855347, + "learning_rate": 8.244621246077992e-05, + "loss": 0.2897, + "step": 32877 + }, + { + "epoch": 1.8369416431544543, + "grad_norm": 0.41594332456588745, + "learning_rate": 8.241819811743614e-05, + "loss": 0.4187, + "step": 32878 + }, + { + "epoch": 1.8369975137581362, + "grad_norm": 0.4846256375312805, + "learning_rate": 8.239018377409234e-05, + "loss": 0.3855, + "step": 32879 + }, + { + "epoch": 1.837053384361818, + "grad_norm": 1.189725637435913, + "learning_rate": 8.236216943074855e-05, + "loss": 0.3761, + "step": 32880 + }, + { + "epoch": 1.8371092549655, + "grad_norm": 0.324385404586792, + "learning_rate": 8.233415508740475e-05, + "loss": 0.2968, + "step": 32881 + }, + { + "epoch": 1.8371651255691819, + "grad_norm": 0.5318078994750977, + "learning_rate": 8.230614074406097e-05, + "loss": 0.4512, + "step": 32882 + }, + { + "epoch": 1.8372209961728636, + "grad_norm": 0.49802911281585693, + "learning_rate": 8.227812640071716e-05, + "loss": 0.3807, + "step": 32883 + }, + { + "epoch": 1.8372768667765456, + "grad_norm": 1.6308503150939941, + "learning_rate": 8.225011205737337e-05, + "loss": 0.4645, + "step": 32884 + }, + { + "epoch": 1.8373327373802275, + "grad_norm": 0.3422342538833618, + "learning_rate": 8.222209771402958e-05, + "loss": 0.3311, + "step": 32885 + }, + { + "epoch": 1.8373886079839092, + "grad_norm": 0.582682728767395, + "learning_rate": 8.21940833706858e-05, + "loss": 0.4119, + "step": 32886 + }, + { + "epoch": 1.8374444785875912, + "grad_norm": 0.43060529232025146, + "learning_rate": 8.2166069027342e-05, + "loss": 0.5262, + "step": 32887 + }, + { + "epoch": 1.8375003491912731, + "grad_norm": 0.38373106718063354, + "learning_rate": 8.213805468399821e-05, + "loss": 0.4718, + "step": 32888 + }, + { + "epoch": 1.8375562197949549, + "grad_norm": 0.3564751446247101, + "learning_rate": 8.211004034065441e-05, + "loss": 0.4286, + "step": 32889 + }, + { + "epoch": 1.8376120903986366, + "grad_norm": 0.33382099866867065, + "learning_rate": 8.208202599731063e-05, + "loss": 0.4246, + "step": 32890 + }, + { + "epoch": 1.8376679610023188, + "grad_norm": 0.5429525375366211, + "learning_rate": 8.205401165396684e-05, + "loss": 0.4123, + "step": 32891 + }, + { + "epoch": 1.8377238316060005, + "grad_norm": 1.5095220804214478, + "learning_rate": 8.202599731062304e-05, + "loss": 0.4124, + "step": 32892 + }, + { + "epoch": 1.8377797022096822, + "grad_norm": 0.3845561146736145, + "learning_rate": 8.199798296727925e-05, + "loss": 0.4709, + "step": 32893 + }, + { + "epoch": 1.8378355728133644, + "grad_norm": 0.32055044174194336, + "learning_rate": 8.196996862393546e-05, + "loss": 0.4226, + "step": 32894 + }, + { + "epoch": 1.8378914434170461, + "grad_norm": 0.619599461555481, + "learning_rate": 8.194195428059166e-05, + "loss": 0.5181, + "step": 32895 + }, + { + "epoch": 1.8379473140207279, + "grad_norm": 1.0156373977661133, + "learning_rate": 8.191393993724787e-05, + "loss": 0.4005, + "step": 32896 + }, + { + "epoch": 1.8380031846244098, + "grad_norm": 0.4670063853263855, + "learning_rate": 8.188592559390407e-05, + "loss": 0.4043, + "step": 32897 + }, + { + "epoch": 1.8380590552280918, + "grad_norm": 0.7825227379798889, + "learning_rate": 8.185791125056029e-05, + "loss": 0.5216, + "step": 32898 + }, + { + "epoch": 1.8381149258317735, + "grad_norm": 0.35551103949546814, + "learning_rate": 8.18298969072165e-05, + "loss": 0.3822, + "step": 32899 + }, + { + "epoch": 1.8381707964354554, + "grad_norm": 0.37710651755332947, + "learning_rate": 8.18018825638727e-05, + "loss": 0.4169, + "step": 32900 + }, + { + "epoch": 1.8382266670391374, + "grad_norm": 0.49866360425949097, + "learning_rate": 8.177386822052891e-05, + "loss": 0.4128, + "step": 32901 + }, + { + "epoch": 1.8382825376428191, + "grad_norm": 0.3481580913066864, + "learning_rate": 8.174585387718513e-05, + "loss": 0.5255, + "step": 32902 + }, + { + "epoch": 1.838338408246501, + "grad_norm": 0.6607549786567688, + "learning_rate": 8.171783953384133e-05, + "loss": 0.5724, + "step": 32903 + }, + { + "epoch": 1.838394278850183, + "grad_norm": 0.3331376612186432, + "learning_rate": 8.168982519049753e-05, + "loss": 0.374, + "step": 32904 + }, + { + "epoch": 1.8384501494538648, + "grad_norm": 0.8921276330947876, + "learning_rate": 8.166181084715373e-05, + "loss": 0.3934, + "step": 32905 + }, + { + "epoch": 1.8385060200575467, + "grad_norm": 2.4783880710601807, + "learning_rate": 8.163379650380995e-05, + "loss": 0.5441, + "step": 32906 + }, + { + "epoch": 1.8385618906612287, + "grad_norm": 0.3889084756374359, + "learning_rate": 8.160578216046616e-05, + "loss": 0.3582, + "step": 32907 + }, + { + "epoch": 1.8386177612649104, + "grad_norm": 0.3184555470943451, + "learning_rate": 8.157776781712236e-05, + "loss": 0.385, + "step": 32908 + }, + { + "epoch": 1.8386736318685923, + "grad_norm": 0.4191698729991913, + "learning_rate": 8.154975347377857e-05, + "loss": 0.3824, + "step": 32909 + }, + { + "epoch": 1.8387295024722743, + "grad_norm": 0.36340823769569397, + "learning_rate": 8.152173913043479e-05, + "loss": 0.3928, + "step": 32910 + }, + { + "epoch": 1.838785373075956, + "grad_norm": 1.2904974222183228, + "learning_rate": 8.1493724787091e-05, + "loss": 0.5057, + "step": 32911 + }, + { + "epoch": 1.838841243679638, + "grad_norm": 0.6816427707672119, + "learning_rate": 8.14657104437472e-05, + "loss": 0.5039, + "step": 32912 + }, + { + "epoch": 1.83889711428332, + "grad_norm": 0.43448421359062195, + "learning_rate": 8.143769610040342e-05, + "loss": 0.3187, + "step": 32913 + }, + { + "epoch": 1.8389529848870017, + "grad_norm": 0.6967939734458923, + "learning_rate": 8.140968175705963e-05, + "loss": 0.385, + "step": 32914 + }, + { + "epoch": 1.8390088554906836, + "grad_norm": 0.3620549440383911, + "learning_rate": 8.138166741371582e-05, + "loss": 0.3551, + "step": 32915 + }, + { + "epoch": 1.8390647260943656, + "grad_norm": 0.3669598400592804, + "learning_rate": 8.135365307037202e-05, + "loss": 0.3803, + "step": 32916 + }, + { + "epoch": 1.8391205966980473, + "grad_norm": 0.3481784164905548, + "learning_rate": 8.132563872702824e-05, + "loss": 0.4137, + "step": 32917 + }, + { + "epoch": 1.8391764673017292, + "grad_norm": 0.3055291175842285, + "learning_rate": 8.129762438368445e-05, + "loss": 0.3963, + "step": 32918 + }, + { + "epoch": 1.8392323379054112, + "grad_norm": 0.4688158929347992, + "learning_rate": 8.126961004034065e-05, + "loss": 0.4405, + "step": 32919 + }, + { + "epoch": 1.839288208509093, + "grad_norm": 0.3705710470676422, + "learning_rate": 8.124159569699686e-05, + "loss": 0.4011, + "step": 32920 + }, + { + "epoch": 1.8393440791127749, + "grad_norm": 0.38556981086730957, + "learning_rate": 8.121358135365308e-05, + "loss": 0.343, + "step": 32921 + }, + { + "epoch": 1.8393999497164568, + "grad_norm": 2.521008253097534, + "learning_rate": 8.118556701030929e-05, + "loss": 0.3195, + "step": 32922 + }, + { + "epoch": 1.8394558203201385, + "grad_norm": 1.1496831178665161, + "learning_rate": 8.115755266696549e-05, + "loss": 0.4792, + "step": 32923 + }, + { + "epoch": 1.8395116909238203, + "grad_norm": 1.5373460054397583, + "learning_rate": 8.11295383236217e-05, + "loss": 0.3943, + "step": 32924 + }, + { + "epoch": 1.8395675615275024, + "grad_norm": 0.5067114233970642, + "learning_rate": 8.110152398027792e-05, + "loss": 0.5567, + "step": 32925 + }, + { + "epoch": 1.8396234321311842, + "grad_norm": 0.3137572407722473, + "learning_rate": 8.107350963693411e-05, + "loss": 0.3722, + "step": 32926 + }, + { + "epoch": 1.839679302734866, + "grad_norm": 0.5903536677360535, + "learning_rate": 8.104549529359031e-05, + "loss": 0.5472, + "step": 32927 + }, + { + "epoch": 1.839735173338548, + "grad_norm": 0.3700695037841797, + "learning_rate": 8.101748095024652e-05, + "loss": 0.3542, + "step": 32928 + }, + { + "epoch": 1.8397910439422298, + "grad_norm": 0.4500535726547241, + "learning_rate": 8.098946660690274e-05, + "loss": 0.3558, + "step": 32929 + }, + { + "epoch": 1.8398469145459115, + "grad_norm": 0.3946380615234375, + "learning_rate": 8.096145226355895e-05, + "loss": 0.4933, + "step": 32930 + }, + { + "epoch": 1.8399027851495935, + "grad_norm": 0.9403205513954163, + "learning_rate": 8.093343792021515e-05, + "loss": 0.4114, + "step": 32931 + }, + { + "epoch": 1.8399586557532754, + "grad_norm": 0.8017126321792603, + "learning_rate": 8.090542357687136e-05, + "loss": 0.472, + "step": 32932 + }, + { + "epoch": 1.8400145263569572, + "grad_norm": 0.3750063478946686, + "learning_rate": 8.087740923352758e-05, + "loss": 0.3801, + "step": 32933 + }, + { + "epoch": 1.8400703969606391, + "grad_norm": 0.3696344792842865, + "learning_rate": 8.084939489018378e-05, + "loss": 0.3671, + "step": 32934 + }, + { + "epoch": 1.840126267564321, + "grad_norm": 0.3777081072330475, + "learning_rate": 8.082138054683999e-05, + "loss": 0.3444, + "step": 32935 + }, + { + "epoch": 1.8401821381680028, + "grad_norm": 0.5440378189086914, + "learning_rate": 8.079336620349618e-05, + "loss": 0.5005, + "step": 32936 + }, + { + "epoch": 1.8402380087716848, + "grad_norm": 3.2170658111572266, + "learning_rate": 8.07653518601524e-05, + "loss": 0.4037, + "step": 32937 + }, + { + "epoch": 1.8402938793753667, + "grad_norm": 0.4196476638317108, + "learning_rate": 8.07373375168086e-05, + "loss": 0.4293, + "step": 32938 + }, + { + "epoch": 1.8403497499790484, + "grad_norm": 0.5212541818618774, + "learning_rate": 8.070932317346481e-05, + "loss": 0.4451, + "step": 32939 + }, + { + "epoch": 1.8404056205827304, + "grad_norm": 2.2976012229919434, + "learning_rate": 8.068130883012102e-05, + "loss": 0.3519, + "step": 32940 + }, + { + "epoch": 1.8404614911864123, + "grad_norm": 1.905513882637024, + "learning_rate": 8.065329448677724e-05, + "loss": 0.3957, + "step": 32941 + }, + { + "epoch": 1.840517361790094, + "grad_norm": 0.4896196722984314, + "learning_rate": 8.062528014343344e-05, + "loss": 0.4746, + "step": 32942 + }, + { + "epoch": 1.840573232393776, + "grad_norm": 0.41830241680145264, + "learning_rate": 8.059726580008965e-05, + "loss": 0.4138, + "step": 32943 + }, + { + "epoch": 1.840629102997458, + "grad_norm": 0.45899975299835205, + "learning_rate": 8.056925145674585e-05, + "loss": 0.37, + "step": 32944 + }, + { + "epoch": 1.8406849736011397, + "grad_norm": 1.0295288562774658, + "learning_rate": 8.054123711340207e-05, + "loss": 0.416, + "step": 32945 + }, + { + "epoch": 1.8407408442048216, + "grad_norm": 2.045736789703369, + "learning_rate": 8.051322277005828e-05, + "loss": 0.4082, + "step": 32946 + }, + { + "epoch": 1.8407967148085036, + "grad_norm": 0.4509471356868744, + "learning_rate": 8.048520842671447e-05, + "loss": 0.4113, + "step": 32947 + }, + { + "epoch": 1.8408525854121853, + "grad_norm": 0.4846443235874176, + "learning_rate": 8.045719408337068e-05, + "loss": 0.6335, + "step": 32948 + }, + { + "epoch": 1.8409084560158673, + "grad_norm": 0.6190320253372192, + "learning_rate": 8.04291797400269e-05, + "loss": 0.3979, + "step": 32949 + }, + { + "epoch": 1.8409643266195492, + "grad_norm": 0.42704012989997864, + "learning_rate": 8.04011653966831e-05, + "loss": 0.3799, + "step": 32950 + }, + { + "epoch": 1.841020197223231, + "grad_norm": 0.3707919418811798, + "learning_rate": 8.037315105333931e-05, + "loss": 0.3769, + "step": 32951 + }, + { + "epoch": 1.841076067826913, + "grad_norm": 0.8640062808990479, + "learning_rate": 8.034513670999551e-05, + "loss": 0.3937, + "step": 32952 + }, + { + "epoch": 1.8411319384305949, + "grad_norm": 0.5922791957855225, + "learning_rate": 8.031712236665173e-05, + "loss": 0.4041, + "step": 32953 + }, + { + "epoch": 1.8411878090342766, + "grad_norm": 0.604373574256897, + "learning_rate": 8.028910802330794e-05, + "loss": 0.45, + "step": 32954 + }, + { + "epoch": 1.8412436796379585, + "grad_norm": 0.5955703258514404, + "learning_rate": 8.026109367996415e-05, + "loss": 0.3576, + "step": 32955 + }, + { + "epoch": 1.8412995502416405, + "grad_norm": 0.31612685322761536, + "learning_rate": 8.023307933662035e-05, + "loss": 0.3523, + "step": 32956 + }, + { + "epoch": 1.8413554208453222, + "grad_norm": 0.40153008699417114, + "learning_rate": 8.020506499327656e-05, + "loss": 0.4477, + "step": 32957 + }, + { + "epoch": 1.841411291449004, + "grad_norm": 0.3716791570186615, + "learning_rate": 8.017705064993276e-05, + "loss": 0.3662, + "step": 32958 + }, + { + "epoch": 1.8414671620526861, + "grad_norm": 0.3884313106536865, + "learning_rate": 8.014903630658897e-05, + "loss": 0.4035, + "step": 32959 + }, + { + "epoch": 1.8415230326563679, + "grad_norm": 0.4007450044155121, + "learning_rate": 8.012102196324517e-05, + "loss": 0.3905, + "step": 32960 + }, + { + "epoch": 1.8415789032600496, + "grad_norm": 2.5198404788970947, + "learning_rate": 8.00930076199014e-05, + "loss": 0.5174, + "step": 32961 + }, + { + "epoch": 1.8416347738637318, + "grad_norm": 0.5380000472068787, + "learning_rate": 8.00649932765576e-05, + "loss": 0.4041, + "step": 32962 + }, + { + "epoch": 1.8416906444674135, + "grad_norm": 1.0139411687850952, + "learning_rate": 8.00369789332138e-05, + "loss": 0.335, + "step": 32963 + }, + { + "epoch": 1.8417465150710952, + "grad_norm": 0.427912175655365, + "learning_rate": 8.000896458987001e-05, + "loss": 0.4523, + "step": 32964 + }, + { + "epoch": 1.8418023856747772, + "grad_norm": 0.3507050573825836, + "learning_rate": 7.998095024652623e-05, + "loss": 0.292, + "step": 32965 + }, + { + "epoch": 1.8418582562784591, + "grad_norm": 0.564624547958374, + "learning_rate": 7.995293590318244e-05, + "loss": 0.78, + "step": 32966 + }, + { + "epoch": 1.8419141268821408, + "grad_norm": 0.367221474647522, + "learning_rate": 7.992492155983864e-05, + "loss": 0.3841, + "step": 32967 + }, + { + "epoch": 1.8419699974858228, + "grad_norm": 0.4495657682418823, + "learning_rate": 7.989690721649483e-05, + "loss": 0.3796, + "step": 32968 + }, + { + "epoch": 1.8420258680895047, + "grad_norm": 0.5189723372459412, + "learning_rate": 7.986889287315105e-05, + "loss": 0.4824, + "step": 32969 + }, + { + "epoch": 1.8420817386931865, + "grad_norm": 0.31678271293640137, + "learning_rate": 7.984087852980726e-05, + "loss": 0.3065, + "step": 32970 + }, + { + "epoch": 1.8421376092968684, + "grad_norm": 0.6245924830436707, + "learning_rate": 7.981286418646347e-05, + "loss": 0.4783, + "step": 32971 + }, + { + "epoch": 1.8421934799005504, + "grad_norm": 0.814864993095398, + "learning_rate": 7.978484984311967e-05, + "loss": 0.4125, + "step": 32972 + }, + { + "epoch": 1.842249350504232, + "grad_norm": 0.35234177112579346, + "learning_rate": 7.975683549977589e-05, + "loss": 0.3726, + "step": 32973 + }, + { + "epoch": 1.842305221107914, + "grad_norm": 0.7020549178123474, + "learning_rate": 7.97288211564321e-05, + "loss": 0.4933, + "step": 32974 + }, + { + "epoch": 1.842361091711596, + "grad_norm": 0.5858356952667236, + "learning_rate": 7.97008068130883e-05, + "loss": 0.3948, + "step": 32975 + }, + { + "epoch": 1.8424169623152777, + "grad_norm": 0.36798539757728577, + "learning_rate": 7.967279246974451e-05, + "loss": 0.4456, + "step": 32976 + }, + { + "epoch": 1.8424728329189597, + "grad_norm": 1.5562158823013306, + "learning_rate": 7.964477812640073e-05, + "loss": 0.3618, + "step": 32977 + }, + { + "epoch": 1.8425287035226416, + "grad_norm": 1.2253155708312988, + "learning_rate": 7.961676378305693e-05, + "loss": 0.3887, + "step": 32978 + }, + { + "epoch": 1.8425845741263234, + "grad_norm": 0.44535937905311584, + "learning_rate": 7.958874943971313e-05, + "loss": 0.4163, + "step": 32979 + }, + { + "epoch": 1.8426404447300053, + "grad_norm": 0.5056455135345459, + "learning_rate": 7.956073509636933e-05, + "loss": 0.4186, + "step": 32980 + }, + { + "epoch": 1.8426963153336873, + "grad_norm": 1.42159903049469, + "learning_rate": 7.953272075302555e-05, + "loss": 0.4343, + "step": 32981 + }, + { + "epoch": 1.842752185937369, + "grad_norm": 0.3510211706161499, + "learning_rate": 7.950470640968176e-05, + "loss": 0.4278, + "step": 32982 + }, + { + "epoch": 1.842808056541051, + "grad_norm": 2.3212778568267822, + "learning_rate": 7.947669206633796e-05, + "loss": 0.4243, + "step": 32983 + }, + { + "epoch": 1.842863927144733, + "grad_norm": 0.35033178329467773, + "learning_rate": 7.944867772299418e-05, + "loss": 0.4367, + "step": 32984 + }, + { + "epoch": 1.8429197977484146, + "grad_norm": 0.7234821319580078, + "learning_rate": 7.942066337965039e-05, + "loss": 0.378, + "step": 32985 + }, + { + "epoch": 1.8429756683520966, + "grad_norm": 0.5032725930213928, + "learning_rate": 7.93926490363066e-05, + "loss": 0.3818, + "step": 32986 + }, + { + "epoch": 1.8430315389557785, + "grad_norm": 0.3761288821697235, + "learning_rate": 7.93646346929628e-05, + "loss": 0.4823, + "step": 32987 + }, + { + "epoch": 1.8430874095594603, + "grad_norm": 1.1027830839157104, + "learning_rate": 7.933662034961902e-05, + "loss": 0.4428, + "step": 32988 + }, + { + "epoch": 1.8431432801631422, + "grad_norm": 0.5539515018463135, + "learning_rate": 7.930860600627521e-05, + "loss": 0.3912, + "step": 32989 + }, + { + "epoch": 1.8431991507668242, + "grad_norm": 0.39707162976264954, + "learning_rate": 7.928059166293142e-05, + "loss": 0.5219, + "step": 32990 + }, + { + "epoch": 1.843255021370506, + "grad_norm": 0.39202335476875305, + "learning_rate": 7.925257731958762e-05, + "loss": 0.4697, + "step": 32991 + }, + { + "epoch": 1.8433108919741876, + "grad_norm": 0.40007898211479187, + "learning_rate": 7.922456297624384e-05, + "loss": 0.4415, + "step": 32992 + }, + { + "epoch": 1.8433667625778698, + "grad_norm": 0.3678523898124695, + "learning_rate": 7.919654863290005e-05, + "loss": 0.3585, + "step": 32993 + }, + { + "epoch": 1.8434226331815515, + "grad_norm": 0.5015842318534851, + "learning_rate": 7.916853428955625e-05, + "loss": 0.4327, + "step": 32994 + }, + { + "epoch": 1.8434785037852333, + "grad_norm": 0.4891580641269684, + "learning_rate": 7.914051994621246e-05, + "loss": 0.5152, + "step": 32995 + }, + { + "epoch": 1.8435343743889154, + "grad_norm": 0.39536258578300476, + "learning_rate": 7.911250560286868e-05, + "loss": 0.4236, + "step": 32996 + }, + { + "epoch": 1.8435902449925972, + "grad_norm": 0.5215957760810852, + "learning_rate": 7.908449125952488e-05, + "loss": 0.4158, + "step": 32997 + }, + { + "epoch": 1.843646115596279, + "grad_norm": 0.4704603850841522, + "learning_rate": 7.905647691618109e-05, + "loss": 0.4744, + "step": 32998 + }, + { + "epoch": 1.8437019861999608, + "grad_norm": 0.37727466225624084, + "learning_rate": 7.90284625728373e-05, + "loss": 0.453, + "step": 32999 + }, + { + "epoch": 1.8437578568036428, + "grad_norm": 0.3677610456943512, + "learning_rate": 7.90004482294935e-05, + "loss": 0.429, + "step": 33000 + }, + { + "epoch": 1.8437578568036428, + "eval_cer": 0.08405616837419397, + "eval_loss": 0.31344661116600037, + "eval_runtime": 56.2462, + "eval_samples_per_second": 80.681, + "eval_steps_per_second": 5.049, + "eval_wer": 0.33462752437082766, + "step": 33000 + }, + { + "epoch": 1.8438137274073245, + "grad_norm": 0.4065755307674408, + "learning_rate": 7.897243388614971e-05, + "loss": 0.371, + "step": 33001 + }, + { + "epoch": 1.8438695980110065, + "grad_norm": 0.38152846693992615, + "learning_rate": 7.894441954280591e-05, + "loss": 0.4009, + "step": 33002 + }, + { + "epoch": 1.8439254686146884, + "grad_norm": 0.3318037688732147, + "learning_rate": 7.891640519946212e-05, + "loss": 0.3485, + "step": 33003 + }, + { + "epoch": 1.8439813392183702, + "grad_norm": 1.5725977420806885, + "learning_rate": 7.888839085611834e-05, + "loss": 0.3828, + "step": 33004 + }, + { + "epoch": 1.844037209822052, + "grad_norm": 0.37654542922973633, + "learning_rate": 7.886037651277454e-05, + "loss": 0.3791, + "step": 33005 + }, + { + "epoch": 1.844093080425734, + "grad_norm": 0.4873170852661133, + "learning_rate": 7.883236216943075e-05, + "loss": 0.4826, + "step": 33006 + }, + { + "epoch": 1.8441489510294158, + "grad_norm": 0.329811155796051, + "learning_rate": 7.880434782608696e-05, + "loss": 0.4544, + "step": 33007 + }, + { + "epoch": 1.8442048216330977, + "grad_norm": 0.5878603458404541, + "learning_rate": 7.877633348274318e-05, + "loss": 0.3565, + "step": 33008 + }, + { + "epoch": 1.8442606922367797, + "grad_norm": 0.5262031555175781, + "learning_rate": 7.874831913939938e-05, + "loss": 0.3858, + "step": 33009 + }, + { + "epoch": 1.8443165628404614, + "grad_norm": 0.42082691192626953, + "learning_rate": 7.872030479605559e-05, + "loss": 0.4007, + "step": 33010 + }, + { + "epoch": 1.8443724334441434, + "grad_norm": 0.6167688965797424, + "learning_rate": 7.869229045271178e-05, + "loss": 0.4631, + "step": 33011 + }, + { + "epoch": 1.8444283040478253, + "grad_norm": 1.6909124851226807, + "learning_rate": 7.8664276109368e-05, + "loss": 0.4507, + "step": 33012 + }, + { + "epoch": 1.844484174651507, + "grad_norm": 0.5336065292358398, + "learning_rate": 7.86362617660242e-05, + "loss": 0.4816, + "step": 33013 + }, + { + "epoch": 1.844540045255189, + "grad_norm": 1.4819608926773071, + "learning_rate": 7.860824742268041e-05, + "loss": 0.3582, + "step": 33014 + }, + { + "epoch": 1.844595915858871, + "grad_norm": 0.29028502106666565, + "learning_rate": 7.858023307933662e-05, + "loss": 0.3527, + "step": 33015 + }, + { + "epoch": 1.8446517864625527, + "grad_norm": 0.3382931649684906, + "learning_rate": 7.855221873599284e-05, + "loss": 0.3716, + "step": 33016 + }, + { + "epoch": 1.8447076570662346, + "grad_norm": 0.49146100878715515, + "learning_rate": 7.852420439264904e-05, + "loss": 0.4095, + "step": 33017 + }, + { + "epoch": 1.8447635276699166, + "grad_norm": 0.34802350401878357, + "learning_rate": 7.849619004930525e-05, + "loss": 0.4, + "step": 33018 + }, + { + "epoch": 1.8448193982735983, + "grad_norm": 0.2974494993686676, + "learning_rate": 7.846817570596145e-05, + "loss": 0.4356, + "step": 33019 + }, + { + "epoch": 1.8448752688772803, + "grad_norm": 0.3632320761680603, + "learning_rate": 7.844016136261767e-05, + "loss": 0.4254, + "step": 33020 + }, + { + "epoch": 1.8449311394809622, + "grad_norm": 0.5908074975013733, + "learning_rate": 7.841214701927386e-05, + "loss": 0.3941, + "step": 33021 + }, + { + "epoch": 1.844987010084644, + "grad_norm": 0.40064752101898193, + "learning_rate": 7.838413267593007e-05, + "loss": 0.3502, + "step": 33022 + }, + { + "epoch": 1.845042880688326, + "grad_norm": 1.05625319480896, + "learning_rate": 7.835611833258628e-05, + "loss": 0.374, + "step": 33023 + }, + { + "epoch": 1.8450987512920078, + "grad_norm": 0.36056581139564514, + "learning_rate": 7.83281039892425e-05, + "loss": 0.3988, + "step": 33024 + }, + { + "epoch": 1.8451546218956896, + "grad_norm": 0.45577895641326904, + "learning_rate": 7.83000896458987e-05, + "loss": 0.3621, + "step": 33025 + }, + { + "epoch": 1.8452104924993713, + "grad_norm": 0.4212741553783417, + "learning_rate": 7.827207530255491e-05, + "loss": 0.3765, + "step": 33026 + }, + { + "epoch": 1.8452663631030535, + "grad_norm": 0.5260246992111206, + "learning_rate": 7.824406095921111e-05, + "loss": 0.3985, + "step": 33027 + }, + { + "epoch": 1.8453222337067352, + "grad_norm": 0.4020356833934784, + "learning_rate": 7.821604661586733e-05, + "loss": 0.4185, + "step": 33028 + }, + { + "epoch": 1.845378104310417, + "grad_norm": 0.3792063593864441, + "learning_rate": 7.818803227252354e-05, + "loss": 0.4542, + "step": 33029 + }, + { + "epoch": 1.845433974914099, + "grad_norm": 0.37317711114883423, + "learning_rate": 7.816001792917974e-05, + "loss": 0.3379, + "step": 33030 + }, + { + "epoch": 1.8454898455177808, + "grad_norm": 0.3257739245891571, + "learning_rate": 7.813200358583595e-05, + "loss": 0.3734, + "step": 33031 + }, + { + "epoch": 1.8455457161214626, + "grad_norm": 0.46461915969848633, + "learning_rate": 7.810398924249216e-05, + "loss": 0.4624, + "step": 33032 + }, + { + "epoch": 1.8456015867251445, + "grad_norm": 0.6147592067718506, + "learning_rate": 7.807597489914836e-05, + "loss": 0.6494, + "step": 33033 + }, + { + "epoch": 1.8456574573288265, + "grad_norm": 0.30340296030044556, + "learning_rate": 7.804796055580457e-05, + "loss": 0.3336, + "step": 33034 + }, + { + "epoch": 1.8457133279325082, + "grad_norm": 0.5162088871002197, + "learning_rate": 7.801994621246077e-05, + "loss": 0.3618, + "step": 33035 + }, + { + "epoch": 1.8457691985361901, + "grad_norm": 0.38183116912841797, + "learning_rate": 7.799193186911699e-05, + "loss": 0.4415, + "step": 33036 + }, + { + "epoch": 1.845825069139872, + "grad_norm": 0.3983537256717682, + "learning_rate": 7.79639175257732e-05, + "loss": 0.4422, + "step": 33037 + }, + { + "epoch": 1.8458809397435538, + "grad_norm": 0.4117516279220581, + "learning_rate": 7.79359031824294e-05, + "loss": 0.3829, + "step": 33038 + }, + { + "epoch": 1.8459368103472358, + "grad_norm": 0.3142840266227722, + "learning_rate": 7.790788883908561e-05, + "loss": 0.3294, + "step": 33039 + }, + { + "epoch": 1.8459926809509177, + "grad_norm": 0.33275341987609863, + "learning_rate": 7.787987449574183e-05, + "loss": 0.3934, + "step": 33040 + }, + { + "epoch": 1.8460485515545995, + "grad_norm": 0.44530028104782104, + "learning_rate": 7.785186015239804e-05, + "loss": 0.5235, + "step": 33041 + }, + { + "epoch": 1.8461044221582814, + "grad_norm": 0.44178876280784607, + "learning_rate": 7.782384580905423e-05, + "loss": 0.4074, + "step": 33042 + }, + { + "epoch": 1.8461602927619634, + "grad_norm": 0.35838258266448975, + "learning_rate": 7.779583146571043e-05, + "loss": 0.4073, + "step": 33043 + }, + { + "epoch": 1.846216163365645, + "grad_norm": 0.808847963809967, + "learning_rate": 7.776781712236665e-05, + "loss": 0.406, + "step": 33044 + }, + { + "epoch": 1.846272033969327, + "grad_norm": 0.35871031880378723, + "learning_rate": 7.773980277902286e-05, + "loss": 0.4093, + "step": 33045 + }, + { + "epoch": 1.846327904573009, + "grad_norm": 1.7368496656417847, + "learning_rate": 7.771178843567906e-05, + "loss": 0.4365, + "step": 33046 + }, + { + "epoch": 1.8463837751766907, + "grad_norm": 0.3537823557853699, + "learning_rate": 7.768377409233527e-05, + "loss": 0.3055, + "step": 33047 + }, + { + "epoch": 1.8464396457803727, + "grad_norm": 2.5747833251953125, + "learning_rate": 7.765575974899149e-05, + "loss": 0.4179, + "step": 33048 + }, + { + "epoch": 1.8464955163840546, + "grad_norm": 1.085663914680481, + "learning_rate": 7.76277454056477e-05, + "loss": 0.3806, + "step": 33049 + }, + { + "epoch": 1.8465513869877364, + "grad_norm": 0.34524887800216675, + "learning_rate": 7.75997310623039e-05, + "loss": 0.476, + "step": 33050 + }, + { + "epoch": 1.8466072575914183, + "grad_norm": 0.5157546997070312, + "learning_rate": 7.757171671896012e-05, + "loss": 0.4934, + "step": 33051 + }, + { + "epoch": 1.8466631281951003, + "grad_norm": 0.5574813485145569, + "learning_rate": 7.754370237561633e-05, + "loss": 0.5688, + "step": 33052 + }, + { + "epoch": 1.846718998798782, + "grad_norm": 0.3423078954219818, + "learning_rate": 7.751568803227252e-05, + "loss": 0.3981, + "step": 33053 + }, + { + "epoch": 1.846774869402464, + "grad_norm": 0.5142806768417358, + "learning_rate": 7.748767368892872e-05, + "loss": 0.5448, + "step": 33054 + }, + { + "epoch": 1.8468307400061459, + "grad_norm": 0.6696507930755615, + "learning_rate": 7.745965934558494e-05, + "loss": 0.5407, + "step": 33055 + }, + { + "epoch": 1.8468866106098276, + "grad_norm": 0.542449414730072, + "learning_rate": 7.743164500224115e-05, + "loss": 0.4275, + "step": 33056 + }, + { + "epoch": 1.8469424812135093, + "grad_norm": 0.4312054216861725, + "learning_rate": 7.740363065889736e-05, + "loss": 0.4785, + "step": 33057 + }, + { + "epoch": 1.8469983518171915, + "grad_norm": 0.4037191569805145, + "learning_rate": 7.737561631555356e-05, + "loss": 0.4865, + "step": 33058 + }, + { + "epoch": 1.8470542224208732, + "grad_norm": 0.37918156385421753, + "learning_rate": 7.734760197220978e-05, + "loss": 0.3457, + "step": 33059 + }, + { + "epoch": 1.847110093024555, + "grad_norm": 0.9030241370201111, + "learning_rate": 7.731958762886599e-05, + "loss": 0.506, + "step": 33060 + }, + { + "epoch": 1.8471659636282372, + "grad_norm": 0.38818177580833435, + "learning_rate": 7.729157328552219e-05, + "loss": 0.4332, + "step": 33061 + }, + { + "epoch": 1.8472218342319189, + "grad_norm": 0.4353932738304138, + "learning_rate": 7.72635589421784e-05, + "loss": 0.3803, + "step": 33062 + }, + { + "epoch": 1.8472777048356006, + "grad_norm": 0.542762279510498, + "learning_rate": 7.723554459883462e-05, + "loss": 0.4594, + "step": 33063 + }, + { + "epoch": 1.8473335754392828, + "grad_norm": 0.3864377737045288, + "learning_rate": 7.720753025549081e-05, + "loss": 0.3859, + "step": 33064 + }, + { + "epoch": 1.8473894460429645, + "grad_norm": 0.6891465783119202, + "learning_rate": 7.717951591214702e-05, + "loss": 0.4418, + "step": 33065 + }, + { + "epoch": 1.8474453166466462, + "grad_norm": 0.8048942685127258, + "learning_rate": 7.715150156880322e-05, + "loss": 0.3424, + "step": 33066 + }, + { + "epoch": 1.8475011872503282, + "grad_norm": 0.4887744188308716, + "learning_rate": 7.712348722545944e-05, + "loss": 0.3445, + "step": 33067 + }, + { + "epoch": 1.8475570578540101, + "grad_norm": 0.716437041759491, + "learning_rate": 7.709547288211565e-05, + "loss": 0.3138, + "step": 33068 + }, + { + "epoch": 1.8476129284576919, + "grad_norm": 0.6223667860031128, + "learning_rate": 7.706745853877185e-05, + "loss": 0.3272, + "step": 33069 + }, + { + "epoch": 1.8476687990613738, + "grad_norm": 0.5027972459793091, + "learning_rate": 7.703944419542806e-05, + "loss": 0.3996, + "step": 33070 + }, + { + "epoch": 1.8477246696650558, + "grad_norm": 1.3420658111572266, + "learning_rate": 7.701142985208428e-05, + "loss": 0.4703, + "step": 33071 + }, + { + "epoch": 1.8477805402687375, + "grad_norm": 0.5987066030502319, + "learning_rate": 7.698341550874048e-05, + "loss": 0.5151, + "step": 33072 + }, + { + "epoch": 1.8478364108724195, + "grad_norm": 4.5486578941345215, + "learning_rate": 7.695540116539669e-05, + "loss": 0.4394, + "step": 33073 + }, + { + "epoch": 1.8478922814761014, + "grad_norm": 0.4595004618167877, + "learning_rate": 7.692738682205288e-05, + "loss": 0.3965, + "step": 33074 + }, + { + "epoch": 1.8479481520797831, + "grad_norm": 0.36354097723960876, + "learning_rate": 7.68993724787091e-05, + "loss": 0.3834, + "step": 33075 + }, + { + "epoch": 1.848004022683465, + "grad_norm": 0.41438522934913635, + "learning_rate": 7.68713581353653e-05, + "loss": 0.5052, + "step": 33076 + }, + { + "epoch": 1.848059893287147, + "grad_norm": 0.42870527505874634, + "learning_rate": 7.684334379202151e-05, + "loss": 0.3864, + "step": 33077 + }, + { + "epoch": 1.8481157638908288, + "grad_norm": 0.45371803641319275, + "learning_rate": 7.681532944867772e-05, + "loss": 0.4357, + "step": 33078 + }, + { + "epoch": 1.8481716344945107, + "grad_norm": 0.3601784408092499, + "learning_rate": 7.678731510533394e-05, + "loss": 0.3264, + "step": 33079 + }, + { + "epoch": 1.8482275050981927, + "grad_norm": 0.43821731209754944, + "learning_rate": 7.675930076199014e-05, + "loss": 0.3938, + "step": 33080 + }, + { + "epoch": 1.8482833757018744, + "grad_norm": 0.6271788477897644, + "learning_rate": 7.673128641864635e-05, + "loss": 0.4453, + "step": 33081 + }, + { + "epoch": 1.8483392463055563, + "grad_norm": 0.5243375897407532, + "learning_rate": 7.670327207530256e-05, + "loss": 0.5796, + "step": 33082 + }, + { + "epoch": 1.8483951169092383, + "grad_norm": 0.3931160569190979, + "learning_rate": 7.667525773195877e-05, + "loss": 0.4496, + "step": 33083 + }, + { + "epoch": 1.84845098751292, + "grad_norm": 0.5009152293205261, + "learning_rate": 7.664724338861498e-05, + "loss": 0.3856, + "step": 33084 + }, + { + "epoch": 1.848506858116602, + "grad_norm": 0.5381882786750793, + "learning_rate": 7.661922904527117e-05, + "loss": 0.3346, + "step": 33085 + }, + { + "epoch": 1.848562728720284, + "grad_norm": 0.4090999662876129, + "learning_rate": 7.659121470192738e-05, + "loss": 0.5228, + "step": 33086 + }, + { + "epoch": 1.8486185993239657, + "grad_norm": 0.355263352394104, + "learning_rate": 7.65632003585836e-05, + "loss": 0.3995, + "step": 33087 + }, + { + "epoch": 1.8486744699276476, + "grad_norm": 0.4124086797237396, + "learning_rate": 7.65351860152398e-05, + "loss": 0.491, + "step": 33088 + }, + { + "epoch": 1.8487303405313296, + "grad_norm": 2.197101593017578, + "learning_rate": 7.650717167189601e-05, + "loss": 0.3635, + "step": 33089 + }, + { + "epoch": 1.8487862111350113, + "grad_norm": 0.352927565574646, + "learning_rate": 7.647915732855222e-05, + "loss": 0.4573, + "step": 33090 + }, + { + "epoch": 1.848842081738693, + "grad_norm": 0.3960360884666443, + "learning_rate": 7.645114298520843e-05, + "loss": 0.433, + "step": 33091 + }, + { + "epoch": 1.8488979523423752, + "grad_norm": 0.4054833650588989, + "learning_rate": 7.642312864186464e-05, + "loss": 0.447, + "step": 33092 + }, + { + "epoch": 1.848953822946057, + "grad_norm": 0.6666194796562195, + "learning_rate": 7.639511429852085e-05, + "loss": 0.4065, + "step": 33093 + }, + { + "epoch": 1.8490096935497387, + "grad_norm": 0.5511000156402588, + "learning_rate": 7.636709995517705e-05, + "loss": 0.4307, + "step": 33094 + }, + { + "epoch": 1.8490655641534208, + "grad_norm": 0.4548628330230713, + "learning_rate": 7.633908561183326e-05, + "loss": 0.479, + "step": 33095 + }, + { + "epoch": 1.8491214347571026, + "grad_norm": 1.0453823804855347, + "learning_rate": 7.631107126848946e-05, + "loss": 0.4696, + "step": 33096 + }, + { + "epoch": 1.8491773053607843, + "grad_norm": 0.40074828267097473, + "learning_rate": 7.628305692514567e-05, + "loss": 0.3708, + "step": 33097 + }, + { + "epoch": 1.8492331759644665, + "grad_norm": 0.5072028636932373, + "learning_rate": 7.625504258180188e-05, + "loss": 0.4728, + "step": 33098 + }, + { + "epoch": 1.8492890465681482, + "grad_norm": 0.3345775008201599, + "learning_rate": 7.62270282384581e-05, + "loss": 0.3821, + "step": 33099 + }, + { + "epoch": 1.84934491717183, + "grad_norm": 0.48072633147239685, + "learning_rate": 7.61990138951143e-05, + "loss": 0.3534, + "step": 33100 + }, + { + "epoch": 1.8494007877755119, + "grad_norm": 0.48358169198036194, + "learning_rate": 7.61709995517705e-05, + "loss": 0.4315, + "step": 33101 + }, + { + "epoch": 1.8494566583791938, + "grad_norm": 0.47605565190315247, + "learning_rate": 7.614298520842671e-05, + "loss": 0.3529, + "step": 33102 + }, + { + "epoch": 1.8495125289828755, + "grad_norm": 0.5026987791061401, + "learning_rate": 7.611497086508293e-05, + "loss": 0.5188, + "step": 33103 + }, + { + "epoch": 1.8495683995865575, + "grad_norm": 0.361778199672699, + "learning_rate": 7.608695652173914e-05, + "loss": 0.4204, + "step": 33104 + }, + { + "epoch": 1.8496242701902395, + "grad_norm": 0.3966955244541168, + "learning_rate": 7.605894217839534e-05, + "loss": 0.4842, + "step": 33105 + }, + { + "epoch": 1.8496801407939212, + "grad_norm": 0.4507737457752228, + "learning_rate": 7.603092783505154e-05, + "loss": 0.3998, + "step": 33106 + }, + { + "epoch": 1.8497360113976031, + "grad_norm": 2.661221981048584, + "learning_rate": 7.600291349170775e-05, + "loss": 0.3802, + "step": 33107 + }, + { + "epoch": 1.849791882001285, + "grad_norm": 0.5065662264823914, + "learning_rate": 7.597489914836396e-05, + "loss": 0.4242, + "step": 33108 + }, + { + "epoch": 1.8498477526049668, + "grad_norm": 2.9406116008758545, + "learning_rate": 7.594688480502017e-05, + "loss": 0.321, + "step": 33109 + }, + { + "epoch": 1.8499036232086488, + "grad_norm": 0.46097084879875183, + "learning_rate": 7.591887046167637e-05, + "loss": 0.3454, + "step": 33110 + }, + { + "epoch": 1.8499594938123307, + "grad_norm": 5.509661674499512, + "learning_rate": 7.589085611833259e-05, + "loss": 0.3361, + "step": 33111 + }, + { + "epoch": 1.8500153644160124, + "grad_norm": 0.611015796661377, + "learning_rate": 7.58628417749888e-05, + "loss": 0.3861, + "step": 33112 + }, + { + "epoch": 1.8500712350196944, + "grad_norm": 0.4860779047012329, + "learning_rate": 7.5834827431645e-05, + "loss": 0.3664, + "step": 33113 + }, + { + "epoch": 1.8501271056233763, + "grad_norm": 0.4153894782066345, + "learning_rate": 7.580681308830122e-05, + "loss": 0.3484, + "step": 33114 + }, + { + "epoch": 1.850182976227058, + "grad_norm": 2.267616033554077, + "learning_rate": 7.577879874495743e-05, + "loss": 0.4065, + "step": 33115 + }, + { + "epoch": 1.85023884683074, + "grad_norm": 0.41485804319381714, + "learning_rate": 7.575078440161363e-05, + "loss": 0.3929, + "step": 33116 + }, + { + "epoch": 1.850294717434422, + "grad_norm": 1.4100978374481201, + "learning_rate": 7.572277005826983e-05, + "loss": 0.3464, + "step": 33117 + }, + { + "epoch": 1.8503505880381037, + "grad_norm": 0.603288471698761, + "learning_rate": 7.569475571492605e-05, + "loss": 0.5672, + "step": 33118 + }, + { + "epoch": 1.8504064586417857, + "grad_norm": 0.6712586879730225, + "learning_rate": 7.566674137158225e-05, + "loss": 0.4699, + "step": 33119 + }, + { + "epoch": 1.8504623292454676, + "grad_norm": 0.835722029209137, + "learning_rate": 7.563872702823846e-05, + "loss": 0.4067, + "step": 33120 + }, + { + "epoch": 1.8505181998491493, + "grad_norm": 2.6469295024871826, + "learning_rate": 7.561071268489466e-05, + "loss": 0.3614, + "step": 33121 + }, + { + "epoch": 1.8505740704528313, + "grad_norm": 0.6195850372314453, + "learning_rate": 7.558269834155088e-05, + "loss": 0.369, + "step": 33122 + }, + { + "epoch": 1.8506299410565132, + "grad_norm": 0.8464590907096863, + "learning_rate": 7.555468399820709e-05, + "loss": 0.3542, + "step": 33123 + }, + { + "epoch": 1.850685811660195, + "grad_norm": 0.4500897228717804, + "learning_rate": 7.55266696548633e-05, + "loss": 0.382, + "step": 33124 + }, + { + "epoch": 1.8507416822638767, + "grad_norm": 0.5220641493797302, + "learning_rate": 7.54986553115195e-05, + "loss": 0.411, + "step": 33125 + }, + { + "epoch": 1.8507975528675589, + "grad_norm": 0.5567074418067932, + "learning_rate": 7.547064096817572e-05, + "loss": 0.4679, + "step": 33126 + }, + { + "epoch": 1.8508534234712406, + "grad_norm": 0.46157166361808777, + "learning_rate": 7.544262662483191e-05, + "loss": 0.437, + "step": 33127 + }, + { + "epoch": 1.8509092940749223, + "grad_norm": 0.3282259404659271, + "learning_rate": 7.541461228148812e-05, + "loss": 0.3703, + "step": 33128 + }, + { + "epoch": 1.8509651646786045, + "grad_norm": 0.3981834352016449, + "learning_rate": 7.538659793814432e-05, + "loss": 0.4651, + "step": 33129 + }, + { + "epoch": 1.8510210352822862, + "grad_norm": 0.5208078026771545, + "learning_rate": 7.535858359480054e-05, + "loss": 0.4096, + "step": 33130 + }, + { + "epoch": 1.851076905885968, + "grad_norm": 0.39842483401298523, + "learning_rate": 7.533056925145675e-05, + "loss": 0.339, + "step": 33131 + }, + { + "epoch": 1.8511327764896501, + "grad_norm": 0.4247796833515167, + "learning_rate": 7.530255490811295e-05, + "loss": 0.4174, + "step": 33132 + }, + { + "epoch": 1.8511886470933319, + "grad_norm": 0.487761914730072, + "learning_rate": 7.527454056476916e-05, + "loss": 0.3268, + "step": 33133 + }, + { + "epoch": 1.8512445176970136, + "grad_norm": 0.39884692430496216, + "learning_rate": 7.524652622142538e-05, + "loss": 0.432, + "step": 33134 + }, + { + "epoch": 1.8513003883006955, + "grad_norm": 0.4539135992527008, + "learning_rate": 7.521851187808159e-05, + "loss": 0.4242, + "step": 33135 + }, + { + "epoch": 1.8513562589043775, + "grad_norm": 1.286074161529541, + "learning_rate": 7.519049753473779e-05, + "loss": 0.3536, + "step": 33136 + }, + { + "epoch": 1.8514121295080592, + "grad_norm": 2.7077248096466064, + "learning_rate": 7.5162483191394e-05, + "loss": 0.4307, + "step": 33137 + }, + { + "epoch": 1.8514680001117412, + "grad_norm": 0.359323114156723, + "learning_rate": 7.51344688480502e-05, + "loss": 0.3798, + "step": 33138 + }, + { + "epoch": 1.8515238707154231, + "grad_norm": 0.5546415448188782, + "learning_rate": 7.510645450470641e-05, + "loss": 0.4266, + "step": 33139 + }, + { + "epoch": 1.8515797413191049, + "grad_norm": 0.6153076887130737, + "learning_rate": 7.507844016136261e-05, + "loss": 0.4011, + "step": 33140 + }, + { + "epoch": 1.8516356119227868, + "grad_norm": 0.4984629452228546, + "learning_rate": 7.505042581801882e-05, + "loss": 0.5642, + "step": 33141 + }, + { + "epoch": 1.8516914825264688, + "grad_norm": 0.5197604894638062, + "learning_rate": 7.502241147467504e-05, + "loss": 0.4091, + "step": 33142 + }, + { + "epoch": 1.8517473531301505, + "grad_norm": 0.277371883392334, + "learning_rate": 7.499439713133125e-05, + "loss": 0.293, + "step": 33143 + }, + { + "epoch": 1.8518032237338324, + "grad_norm": 0.6169501543045044, + "learning_rate": 7.496638278798745e-05, + "loss": 0.4268, + "step": 33144 + }, + { + "epoch": 1.8518590943375144, + "grad_norm": 0.7985299229621887, + "learning_rate": 7.493836844464366e-05, + "loss": 0.4155, + "step": 33145 + }, + { + "epoch": 1.8519149649411961, + "grad_norm": 0.41812366247177124, + "learning_rate": 7.491035410129988e-05, + "loss": 0.362, + "step": 33146 + }, + { + "epoch": 1.851970835544878, + "grad_norm": 0.4834993779659271, + "learning_rate": 7.488233975795608e-05, + "loss": 0.3966, + "step": 33147 + }, + { + "epoch": 1.85202670614856, + "grad_norm": 0.3666814863681793, + "learning_rate": 7.485432541461229e-05, + "loss": 0.3706, + "step": 33148 + }, + { + "epoch": 1.8520825767522417, + "grad_norm": 0.7272896766662598, + "learning_rate": 7.482631107126848e-05, + "loss": 0.3679, + "step": 33149 + }, + { + "epoch": 1.8521384473559237, + "grad_norm": 0.3634657859802246, + "learning_rate": 7.47982967279247e-05, + "loss": 0.3916, + "step": 33150 + }, + { + "epoch": 1.8521943179596057, + "grad_norm": 0.7998027205467224, + "learning_rate": 7.47702823845809e-05, + "loss": 0.6855, + "step": 33151 + }, + { + "epoch": 1.8522501885632874, + "grad_norm": 13.759552001953125, + "learning_rate": 7.474226804123711e-05, + "loss": 0.4093, + "step": 33152 + }, + { + "epoch": 1.8523060591669693, + "grad_norm": 1.4222593307495117, + "learning_rate": 7.471425369789332e-05, + "loss": 0.4696, + "step": 33153 + }, + { + "epoch": 1.8523619297706513, + "grad_norm": 1.171268343925476, + "learning_rate": 7.468623935454954e-05, + "loss": 0.3786, + "step": 33154 + }, + { + "epoch": 1.852417800374333, + "grad_norm": 0.5160524249076843, + "learning_rate": 7.465822501120574e-05, + "loss": 0.3728, + "step": 33155 + }, + { + "epoch": 1.852473670978015, + "grad_norm": 0.4990294873714447, + "learning_rate": 7.463021066786195e-05, + "loss": 0.4842, + "step": 33156 + }, + { + "epoch": 1.852529541581697, + "grad_norm": 0.36780819296836853, + "learning_rate": 7.460219632451815e-05, + "loss": 0.3685, + "step": 33157 + }, + { + "epoch": 1.8525854121853786, + "grad_norm": 0.51551353931427, + "learning_rate": 7.457418198117437e-05, + "loss": 0.3754, + "step": 33158 + }, + { + "epoch": 1.8526412827890604, + "grad_norm": 0.528724730014801, + "learning_rate": 7.454616763783057e-05, + "loss": 0.4265, + "step": 33159 + }, + { + "epoch": 1.8526971533927425, + "grad_norm": 2.5339083671569824, + "learning_rate": 7.451815329448677e-05, + "loss": 0.4649, + "step": 33160 + }, + { + "epoch": 1.8527530239964243, + "grad_norm": 0.43018436431884766, + "learning_rate": 7.449013895114298e-05, + "loss": 0.3131, + "step": 33161 + }, + { + "epoch": 1.852808894600106, + "grad_norm": 0.3766195476055145, + "learning_rate": 7.44621246077992e-05, + "loss": 0.4287, + "step": 33162 + }, + { + "epoch": 1.8528647652037882, + "grad_norm": 0.6450812816619873, + "learning_rate": 7.44341102644554e-05, + "loss": 0.4349, + "step": 33163 + }, + { + "epoch": 1.85292063580747, + "grad_norm": 0.3701653480529785, + "learning_rate": 7.440609592111161e-05, + "loss": 0.3495, + "step": 33164 + }, + { + "epoch": 1.8529765064111516, + "grad_norm": 0.35711994767189026, + "learning_rate": 7.437808157776781e-05, + "loss": 0.3203, + "step": 33165 + }, + { + "epoch": 1.8530323770148338, + "grad_norm": 0.40324294567108154, + "learning_rate": 7.435006723442403e-05, + "loss": 0.4163, + "step": 33166 + }, + { + "epoch": 1.8530882476185155, + "grad_norm": 1.5191315412521362, + "learning_rate": 7.432205289108024e-05, + "loss": 0.3849, + "step": 33167 + }, + { + "epoch": 1.8531441182221973, + "grad_norm": 0.4714263081550598, + "learning_rate": 7.429403854773645e-05, + "loss": 0.4568, + "step": 33168 + }, + { + "epoch": 1.8531999888258792, + "grad_norm": 0.9164767861366272, + "learning_rate": 7.426602420439265e-05, + "loss": 0.4725, + "step": 33169 + }, + { + "epoch": 1.8532558594295612, + "grad_norm": 0.41484397649765015, + "learning_rate": 7.423800986104886e-05, + "loss": 0.3521, + "step": 33170 + }, + { + "epoch": 1.853311730033243, + "grad_norm": 0.38305428624153137, + "learning_rate": 7.420999551770506e-05, + "loss": 0.416, + "step": 33171 + }, + { + "epoch": 1.8533676006369249, + "grad_norm": 0.3933025896549225, + "learning_rate": 7.418198117436127e-05, + "loss": 0.3735, + "step": 33172 + }, + { + "epoch": 1.8534234712406068, + "grad_norm": 0.4051043689250946, + "learning_rate": 7.415396683101747e-05, + "loss": 0.5033, + "step": 33173 + }, + { + "epoch": 1.8534793418442885, + "grad_norm": 2.298140525817871, + "learning_rate": 7.41259524876737e-05, + "loss": 0.4127, + "step": 33174 + }, + { + "epoch": 1.8535352124479705, + "grad_norm": 0.888774573802948, + "learning_rate": 7.40979381443299e-05, + "loss": 0.4209, + "step": 33175 + }, + { + "epoch": 1.8535910830516524, + "grad_norm": 1.2837977409362793, + "learning_rate": 7.40699238009861e-05, + "loss": 0.401, + "step": 33176 + }, + { + "epoch": 1.8536469536553342, + "grad_norm": 0.3295147716999054, + "learning_rate": 7.404190945764231e-05, + "loss": 0.3109, + "step": 33177 + }, + { + "epoch": 1.8537028242590161, + "grad_norm": 1.1153770685195923, + "learning_rate": 7.401389511429853e-05, + "loss": 0.4511, + "step": 33178 + }, + { + "epoch": 1.853758694862698, + "grad_norm": 0.8499010801315308, + "learning_rate": 7.398588077095474e-05, + "loss": 0.5005, + "step": 33179 + }, + { + "epoch": 1.8538145654663798, + "grad_norm": 0.658917248249054, + "learning_rate": 7.395786642761093e-05, + "loss": 0.4337, + "step": 33180 + }, + { + "epoch": 1.8538704360700617, + "grad_norm": 0.48940348625183105, + "learning_rate": 7.392985208426713e-05, + "loss": 0.4663, + "step": 33181 + }, + { + "epoch": 1.8539263066737437, + "grad_norm": 0.6275519728660583, + "learning_rate": 7.390183774092335e-05, + "loss": 0.3373, + "step": 33182 + }, + { + "epoch": 1.8539821772774254, + "grad_norm": 0.5635482668876648, + "learning_rate": 7.387382339757956e-05, + "loss": 0.4466, + "step": 33183 + }, + { + "epoch": 1.8540380478811074, + "grad_norm": 0.37735363841056824, + "learning_rate": 7.384580905423577e-05, + "loss": 0.4534, + "step": 33184 + }, + { + "epoch": 1.8540939184847893, + "grad_norm": 0.32762980461120605, + "learning_rate": 7.381779471089198e-05, + "loss": 0.3664, + "step": 33185 + }, + { + "epoch": 1.854149789088471, + "grad_norm": 1.1293600797653198, + "learning_rate": 7.378978036754819e-05, + "loss": 0.4248, + "step": 33186 + }, + { + "epoch": 1.854205659692153, + "grad_norm": 0.6230453848838806, + "learning_rate": 7.37617660242044e-05, + "loss": 0.4411, + "step": 33187 + }, + { + "epoch": 1.854261530295835, + "grad_norm": 0.4029426872730255, + "learning_rate": 7.37337516808606e-05, + "loss": 0.4705, + "step": 33188 + }, + { + "epoch": 1.8543174008995167, + "grad_norm": 1.0914218425750732, + "learning_rate": 7.370573733751682e-05, + "loss": 0.3756, + "step": 33189 + }, + { + "epoch": 1.8543732715031986, + "grad_norm": 0.38033080101013184, + "learning_rate": 7.367772299417303e-05, + "loss": 0.3389, + "step": 33190 + }, + { + "epoch": 1.8544291421068806, + "grad_norm": 0.37038877606391907, + "learning_rate": 7.364970865082922e-05, + "loss": 0.3778, + "step": 33191 + }, + { + "epoch": 1.8544850127105623, + "grad_norm": 0.7971298694610596, + "learning_rate": 7.362169430748543e-05, + "loss": 0.5067, + "step": 33192 + }, + { + "epoch": 1.854540883314244, + "grad_norm": 0.4393012821674347, + "learning_rate": 7.359367996414164e-05, + "loss": 0.4455, + "step": 33193 + }, + { + "epoch": 1.8545967539179262, + "grad_norm": 0.48506563901901245, + "learning_rate": 7.356566562079785e-05, + "loss": 0.4107, + "step": 33194 + }, + { + "epoch": 1.854652624521608, + "grad_norm": 0.4475899636745453, + "learning_rate": 7.353765127745406e-05, + "loss": 0.4402, + "step": 33195 + }, + { + "epoch": 1.8547084951252897, + "grad_norm": 0.6818153858184814, + "learning_rate": 7.350963693411026e-05, + "loss": 0.5727, + "step": 33196 + }, + { + "epoch": 1.8547643657289719, + "grad_norm": 0.3336114287376404, + "learning_rate": 7.348162259076648e-05, + "loss": 0.3807, + "step": 33197 + }, + { + "epoch": 1.8548202363326536, + "grad_norm": 0.4760880768299103, + "learning_rate": 7.345360824742269e-05, + "loss": 0.4363, + "step": 33198 + }, + { + "epoch": 1.8548761069363353, + "grad_norm": 0.42375293374061584, + "learning_rate": 7.342559390407889e-05, + "loss": 0.4241, + "step": 33199 + }, + { + "epoch": 1.8549319775400175, + "grad_norm": 0.6978812217712402, + "learning_rate": 7.33975795607351e-05, + "loss": 0.3893, + "step": 33200 + }, + { + "epoch": 1.8549878481436992, + "grad_norm": 0.35615819692611694, + "learning_rate": 7.336956521739132e-05, + "loss": 0.4541, + "step": 33201 + }, + { + "epoch": 1.855043718747381, + "grad_norm": 0.4437476396560669, + "learning_rate": 7.334155087404751e-05, + "loss": 0.3316, + "step": 33202 + }, + { + "epoch": 1.855099589351063, + "grad_norm": 0.4644714891910553, + "learning_rate": 7.331353653070372e-05, + "loss": 0.3936, + "step": 33203 + }, + { + "epoch": 1.8551554599547448, + "grad_norm": 0.31734585762023926, + "learning_rate": 7.328552218735992e-05, + "loss": 0.3547, + "step": 33204 + }, + { + "epoch": 1.8552113305584266, + "grad_norm": 0.6339156031608582, + "learning_rate": 7.325750784401614e-05, + "loss": 0.4901, + "step": 33205 + }, + { + "epoch": 1.8552672011621085, + "grad_norm": 0.3670881688594818, + "learning_rate": 7.322949350067235e-05, + "loss": 0.3852, + "step": 33206 + }, + { + "epoch": 1.8553230717657905, + "grad_norm": 1.0013114213943481, + "learning_rate": 7.320147915732855e-05, + "loss": 0.4851, + "step": 33207 + }, + { + "epoch": 1.8553789423694722, + "grad_norm": 0.3083060383796692, + "learning_rate": 7.317346481398476e-05, + "loss": 0.4431, + "step": 33208 + }, + { + "epoch": 1.8554348129731542, + "grad_norm": 0.656182587146759, + "learning_rate": 7.314545047064098e-05, + "loss": 0.4444, + "step": 33209 + }, + { + "epoch": 1.855490683576836, + "grad_norm": 0.6726865172386169, + "learning_rate": 7.311743612729718e-05, + "loss": 0.4342, + "step": 33210 + }, + { + "epoch": 1.8555465541805178, + "grad_norm": 0.3406350910663605, + "learning_rate": 7.308942178395339e-05, + "loss": 0.3144, + "step": 33211 + }, + { + "epoch": 1.8556024247841998, + "grad_norm": 0.35005033016204834, + "learning_rate": 7.306140744060958e-05, + "loss": 0.3673, + "step": 33212 + }, + { + "epoch": 1.8556582953878817, + "grad_norm": 0.5162157416343689, + "learning_rate": 7.30333930972658e-05, + "loss": 0.4257, + "step": 33213 + }, + { + "epoch": 1.8557141659915635, + "grad_norm": 0.3125956952571869, + "learning_rate": 7.300537875392201e-05, + "loss": 0.3307, + "step": 33214 + }, + { + "epoch": 1.8557700365952454, + "grad_norm": 0.4424608051776886, + "learning_rate": 7.297736441057821e-05, + "loss": 0.4193, + "step": 33215 + }, + { + "epoch": 1.8558259071989274, + "grad_norm": 2.152892827987671, + "learning_rate": 7.294935006723442e-05, + "loss": 0.3627, + "step": 33216 + }, + { + "epoch": 1.855881777802609, + "grad_norm": 0.3573732078075409, + "learning_rate": 7.292133572389064e-05, + "loss": 0.3886, + "step": 33217 + }, + { + "epoch": 1.855937648406291, + "grad_norm": 1.0437473058700562, + "learning_rate": 7.289332138054684e-05, + "loss": 0.3517, + "step": 33218 + }, + { + "epoch": 1.855993519009973, + "grad_norm": 0.3897626996040344, + "learning_rate": 7.286530703720305e-05, + "loss": 0.4338, + "step": 33219 + }, + { + "epoch": 1.8560493896136547, + "grad_norm": 0.3968331515789032, + "learning_rate": 7.283729269385926e-05, + "loss": 0.4591, + "step": 33220 + }, + { + "epoch": 1.8561052602173367, + "grad_norm": 0.3692198395729065, + "learning_rate": 7.280927835051548e-05, + "loss": 0.3746, + "step": 33221 + }, + { + "epoch": 1.8561611308210186, + "grad_norm": 0.37460678815841675, + "learning_rate": 7.278126400717168e-05, + "loss": 0.3511, + "step": 33222 + }, + { + "epoch": 1.8562170014247004, + "grad_norm": 0.3792080581188202, + "learning_rate": 7.275324966382787e-05, + "loss": 0.3934, + "step": 33223 + }, + { + "epoch": 1.8562728720283823, + "grad_norm": 0.37607160210609436, + "learning_rate": 7.272523532048408e-05, + "loss": 0.3346, + "step": 33224 + }, + { + "epoch": 1.8563287426320643, + "grad_norm": 0.46423813700675964, + "learning_rate": 7.26972209771403e-05, + "loss": 0.4022, + "step": 33225 + }, + { + "epoch": 1.856384613235746, + "grad_norm": 0.4183676838874817, + "learning_rate": 7.26692066337965e-05, + "loss": 0.3615, + "step": 33226 + }, + { + "epoch": 1.8564404838394277, + "grad_norm": 0.4441477060317993, + "learning_rate": 7.264119229045271e-05, + "loss": 0.4522, + "step": 33227 + }, + { + "epoch": 1.85649635444311, + "grad_norm": 0.8715426325798035, + "learning_rate": 7.261317794710892e-05, + "loss": 0.4568, + "step": 33228 + }, + { + "epoch": 1.8565522250467916, + "grad_norm": 0.5705441832542419, + "learning_rate": 7.258516360376514e-05, + "loss": 0.4396, + "step": 33229 + }, + { + "epoch": 1.8566080956504734, + "grad_norm": 0.41585153341293335, + "learning_rate": 7.255714926042134e-05, + "loss": 0.4712, + "step": 33230 + }, + { + "epoch": 1.8566639662541555, + "grad_norm": 1.108453392982483, + "learning_rate": 7.252913491707755e-05, + "loss": 0.4466, + "step": 33231 + }, + { + "epoch": 1.8567198368578373, + "grad_norm": 1.1356561183929443, + "learning_rate": 7.250112057373375e-05, + "loss": 0.4347, + "step": 33232 + }, + { + "epoch": 1.856775707461519, + "grad_norm": 0.37507274746894836, + "learning_rate": 7.247310623038997e-05, + "loss": 0.455, + "step": 33233 + }, + { + "epoch": 1.8568315780652012, + "grad_norm": 1.03776216506958, + "learning_rate": 7.244509188704616e-05, + "loss": 0.475, + "step": 33234 + }, + { + "epoch": 1.856887448668883, + "grad_norm": 0.37231752276420593, + "learning_rate": 7.241707754370237e-05, + "loss": 0.4659, + "step": 33235 + }, + { + "epoch": 1.8569433192725646, + "grad_norm": 0.8156972527503967, + "learning_rate": 7.238906320035858e-05, + "loss": 0.3304, + "step": 33236 + }, + { + "epoch": 1.8569991898762466, + "grad_norm": 0.3634747862815857, + "learning_rate": 7.23610488570148e-05, + "loss": 0.3749, + "step": 33237 + }, + { + "epoch": 1.8570550604799285, + "grad_norm": 0.44898536801338196, + "learning_rate": 7.2333034513671e-05, + "loss": 0.4343, + "step": 33238 + }, + { + "epoch": 1.8571109310836103, + "grad_norm": 2.9900269508361816, + "learning_rate": 7.230502017032721e-05, + "loss": 0.5525, + "step": 33239 + }, + { + "epoch": 1.8571668016872922, + "grad_norm": 0.4013199508190155, + "learning_rate": 7.227700582698341e-05, + "loss": 0.4673, + "step": 33240 + }, + { + "epoch": 1.8572226722909742, + "grad_norm": 0.38984760642051697, + "learning_rate": 7.224899148363963e-05, + "loss": 0.3754, + "step": 33241 + }, + { + "epoch": 1.8572785428946559, + "grad_norm": 1.721587061882019, + "learning_rate": 7.222097714029584e-05, + "loss": 0.4181, + "step": 33242 + }, + { + "epoch": 1.8573344134983378, + "grad_norm": 0.6136160492897034, + "learning_rate": 7.219296279695204e-05, + "loss": 0.4578, + "step": 33243 + }, + { + "epoch": 1.8573902841020198, + "grad_norm": 0.502068042755127, + "learning_rate": 7.216494845360824e-05, + "loss": 0.4411, + "step": 33244 + }, + { + "epoch": 1.8574461547057015, + "grad_norm": 0.46963435411453247, + "learning_rate": 7.213693411026446e-05, + "loss": 0.3552, + "step": 33245 + }, + { + "epoch": 1.8575020253093835, + "grad_norm": 0.3522191345691681, + "learning_rate": 7.210891976692066e-05, + "loss": 0.361, + "step": 33246 + }, + { + "epoch": 1.8575578959130654, + "grad_norm": 0.7439832091331482, + "learning_rate": 7.208090542357687e-05, + "loss": 0.5106, + "step": 33247 + }, + { + "epoch": 1.8576137665167471, + "grad_norm": 0.5065428018569946, + "learning_rate": 7.205289108023307e-05, + "loss": 0.4426, + "step": 33248 + }, + { + "epoch": 1.857669637120429, + "grad_norm": 0.3914973735809326, + "learning_rate": 7.202487673688929e-05, + "loss": 0.3627, + "step": 33249 + }, + { + "epoch": 1.857725507724111, + "grad_norm": 0.5193977952003479, + "learning_rate": 7.19968623935455e-05, + "loss": 0.4696, + "step": 33250 + }, + { + "epoch": 1.8577813783277928, + "grad_norm": 0.4074508547782898, + "learning_rate": 7.19688480502017e-05, + "loss": 0.4114, + "step": 33251 + }, + { + "epoch": 1.8578372489314747, + "grad_norm": 0.41177091002464294, + "learning_rate": 7.194083370685792e-05, + "loss": 0.3601, + "step": 33252 + }, + { + "epoch": 1.8578931195351567, + "grad_norm": 0.4115317165851593, + "learning_rate": 7.191281936351413e-05, + "loss": 0.3547, + "step": 33253 + }, + { + "epoch": 1.8579489901388384, + "grad_norm": 0.44875645637512207, + "learning_rate": 7.188480502017033e-05, + "loss": 0.4416, + "step": 33254 + }, + { + "epoch": 1.8580048607425204, + "grad_norm": 0.8959817290306091, + "learning_rate": 7.185679067682653e-05, + "loss": 0.4527, + "step": 33255 + }, + { + "epoch": 1.8580607313462023, + "grad_norm": 0.3561689555644989, + "learning_rate": 7.182877633348275e-05, + "loss": 0.5174, + "step": 33256 + }, + { + "epoch": 1.858116601949884, + "grad_norm": 0.28069645166397095, + "learning_rate": 7.180076199013895e-05, + "loss": 0.3508, + "step": 33257 + }, + { + "epoch": 1.858172472553566, + "grad_norm": 0.3504849672317505, + "learning_rate": 7.177274764679516e-05, + "loss": 0.3121, + "step": 33258 + }, + { + "epoch": 1.858228343157248, + "grad_norm": 0.3752122223377228, + "learning_rate": 7.174473330345136e-05, + "loss": 0.4501, + "step": 33259 + }, + { + "epoch": 1.8582842137609297, + "grad_norm": 0.4325581192970276, + "learning_rate": 7.171671896010758e-05, + "loss": 0.4536, + "step": 33260 + }, + { + "epoch": 1.8583400843646114, + "grad_norm": 0.54692143201828, + "learning_rate": 7.168870461676379e-05, + "loss": 0.5063, + "step": 33261 + }, + { + "epoch": 1.8583959549682936, + "grad_norm": 0.4844164252281189, + "learning_rate": 7.166069027342e-05, + "loss": 0.3849, + "step": 33262 + }, + { + "epoch": 1.8584518255719753, + "grad_norm": 0.4458923935890198, + "learning_rate": 7.16326759300762e-05, + "loss": 0.4833, + "step": 33263 + }, + { + "epoch": 1.858507696175657, + "grad_norm": 0.7042177319526672, + "learning_rate": 7.160466158673242e-05, + "loss": 0.5251, + "step": 33264 + }, + { + "epoch": 1.8585635667793392, + "grad_norm": 0.5289865732192993, + "learning_rate": 7.157664724338861e-05, + "loss": 0.48, + "step": 33265 + }, + { + "epoch": 1.858619437383021, + "grad_norm": 0.4116239845752716, + "learning_rate": 7.154863290004482e-05, + "loss": 0.3938, + "step": 33266 + }, + { + "epoch": 1.8586753079867027, + "grad_norm": 0.43603983521461487, + "learning_rate": 7.152061855670102e-05, + "loss": 0.4362, + "step": 33267 + }, + { + "epoch": 1.8587311785903846, + "grad_norm": 0.9376553297042847, + "learning_rate": 7.149260421335724e-05, + "loss": 0.4366, + "step": 33268 + }, + { + "epoch": 1.8587870491940666, + "grad_norm": 1.3627618551254272, + "learning_rate": 7.146458987001345e-05, + "loss": 0.3086, + "step": 33269 + }, + { + "epoch": 1.8588429197977483, + "grad_norm": 0.2769615352153778, + "learning_rate": 7.143657552666966e-05, + "loss": 0.3133, + "step": 33270 + }, + { + "epoch": 1.8588987904014302, + "grad_norm": 0.3622414171695709, + "learning_rate": 7.140856118332586e-05, + "loss": 0.41, + "step": 33271 + }, + { + "epoch": 1.8589546610051122, + "grad_norm": 0.4449443221092224, + "learning_rate": 7.138054683998208e-05, + "loss": 0.444, + "step": 33272 + }, + { + "epoch": 1.859010531608794, + "grad_norm": 0.33408117294311523, + "learning_rate": 7.135253249663829e-05, + "loss": 0.2984, + "step": 33273 + }, + { + "epoch": 1.8590664022124759, + "grad_norm": 0.46573713421821594, + "learning_rate": 7.132451815329449e-05, + "loss": 0.4974, + "step": 33274 + }, + { + "epoch": 1.8591222728161578, + "grad_norm": 0.7564700245857239, + "learning_rate": 7.12965038099507e-05, + "loss": 0.4079, + "step": 33275 + }, + { + "epoch": 1.8591781434198396, + "grad_norm": 0.3524746298789978, + "learning_rate": 7.12684894666069e-05, + "loss": 0.4784, + "step": 33276 + }, + { + "epoch": 1.8592340140235215, + "grad_norm": 0.6689967513084412, + "learning_rate": 7.124047512326311e-05, + "loss": 0.4368, + "step": 33277 + }, + { + "epoch": 1.8592898846272035, + "grad_norm": 0.3461688756942749, + "learning_rate": 7.121246077991932e-05, + "loss": 0.3192, + "step": 33278 + }, + { + "epoch": 1.8593457552308852, + "grad_norm": 0.6745437979698181, + "learning_rate": 7.118444643657552e-05, + "loss": 0.324, + "step": 33279 + }, + { + "epoch": 1.8594016258345671, + "grad_norm": 0.9729037284851074, + "learning_rate": 7.115643209323174e-05, + "loss": 0.3765, + "step": 33280 + }, + { + "epoch": 1.859457496438249, + "grad_norm": 0.32138171792030334, + "learning_rate": 7.112841774988795e-05, + "loss": 0.4387, + "step": 33281 + }, + { + "epoch": 1.8595133670419308, + "grad_norm": 0.4549672603607178, + "learning_rate": 7.110040340654415e-05, + "loss": 0.3737, + "step": 33282 + }, + { + "epoch": 1.8595692376456128, + "grad_norm": 0.3882914185523987, + "learning_rate": 7.107238906320036e-05, + "loss": 0.4123, + "step": 33283 + }, + { + "epoch": 1.8596251082492947, + "grad_norm": 2.0474460124969482, + "learning_rate": 7.104437471985658e-05, + "loss": 0.2965, + "step": 33284 + }, + { + "epoch": 1.8596809788529765, + "grad_norm": 0.4147179424762726, + "learning_rate": 7.101636037651278e-05, + "loss": 0.3953, + "step": 33285 + }, + { + "epoch": 1.8597368494566584, + "grad_norm": 0.401673287153244, + "learning_rate": 7.098834603316899e-05, + "loss": 0.3994, + "step": 33286 + }, + { + "epoch": 1.8597927200603404, + "grad_norm": 1.1313868761062622, + "learning_rate": 7.096033168982518e-05, + "loss": 0.4102, + "step": 33287 + }, + { + "epoch": 1.859848590664022, + "grad_norm": 1.8171223402023315, + "learning_rate": 7.09323173464814e-05, + "loss": 0.3753, + "step": 33288 + }, + { + "epoch": 1.859904461267704, + "grad_norm": 0.7040718197822571, + "learning_rate": 7.09043030031376e-05, + "loss": 0.4174, + "step": 33289 + }, + { + "epoch": 1.859960331871386, + "grad_norm": 0.9715284109115601, + "learning_rate": 7.087628865979381e-05, + "loss": 0.4578, + "step": 33290 + }, + { + "epoch": 1.8600162024750677, + "grad_norm": 0.7936530709266663, + "learning_rate": 7.084827431645002e-05, + "loss": 0.4641, + "step": 33291 + }, + { + "epoch": 1.8600720730787497, + "grad_norm": 0.36094751954078674, + "learning_rate": 7.082025997310624e-05, + "loss": 0.4667, + "step": 33292 + }, + { + "epoch": 1.8601279436824316, + "grad_norm": 0.7547724843025208, + "learning_rate": 7.079224562976244e-05, + "loss": 0.3045, + "step": 33293 + }, + { + "epoch": 1.8601838142861133, + "grad_norm": 0.49365219473838806, + "learning_rate": 7.076423128641865e-05, + "loss": 0.5442, + "step": 33294 + }, + { + "epoch": 1.860239684889795, + "grad_norm": 0.49702560901641846, + "learning_rate": 7.073621694307485e-05, + "loss": 0.3407, + "step": 33295 + }, + { + "epoch": 1.8602955554934772, + "grad_norm": 1.1603838205337524, + "learning_rate": 7.070820259973107e-05, + "loss": 0.3192, + "step": 33296 + }, + { + "epoch": 1.860351426097159, + "grad_norm": 0.31287136673927307, + "learning_rate": 7.068018825638727e-05, + "loss": 0.4076, + "step": 33297 + }, + { + "epoch": 1.8604072967008407, + "grad_norm": 0.3493267297744751, + "learning_rate": 7.065217391304347e-05, + "loss": 0.3286, + "step": 33298 + }, + { + "epoch": 1.8604631673045229, + "grad_norm": 0.6481277346611023, + "learning_rate": 7.062415956969968e-05, + "loss": 0.4493, + "step": 33299 + }, + { + "epoch": 1.8605190379082046, + "grad_norm": 0.44085344672203064, + "learning_rate": 7.05961452263559e-05, + "loss": 0.4611, + "step": 33300 + }, + { + "epoch": 1.8605749085118863, + "grad_norm": 0.4575200378894806, + "learning_rate": 7.05681308830121e-05, + "loss": 0.4043, + "step": 33301 + }, + { + "epoch": 1.8606307791155683, + "grad_norm": 0.6417617201805115, + "learning_rate": 7.054011653966831e-05, + "loss": 0.3534, + "step": 33302 + }, + { + "epoch": 1.8606866497192502, + "grad_norm": 0.32399410009384155, + "learning_rate": 7.051210219632451e-05, + "loss": 0.3959, + "step": 33303 + }, + { + "epoch": 1.860742520322932, + "grad_norm": 0.5660884380340576, + "learning_rate": 7.048408785298073e-05, + "loss": 0.4619, + "step": 33304 + }, + { + "epoch": 1.860798390926614, + "grad_norm": 0.34823843836784363, + "learning_rate": 7.045607350963694e-05, + "loss": 0.384, + "step": 33305 + }, + { + "epoch": 1.8608542615302959, + "grad_norm": 0.33023861050605774, + "learning_rate": 7.042805916629315e-05, + "loss": 0.3475, + "step": 33306 + }, + { + "epoch": 1.8609101321339776, + "grad_norm": 0.46336302161216736, + "learning_rate": 7.040004482294935e-05, + "loss": 0.3906, + "step": 33307 + }, + { + "epoch": 1.8609660027376596, + "grad_norm": 0.5508529543876648, + "learning_rate": 7.037203047960556e-05, + "loss": 0.3553, + "step": 33308 + }, + { + "epoch": 1.8610218733413415, + "grad_norm": 0.31391191482543945, + "learning_rate": 7.034401613626176e-05, + "loss": 0.3897, + "step": 33309 + }, + { + "epoch": 1.8610777439450232, + "grad_norm": 0.4885827302932739, + "learning_rate": 7.031600179291797e-05, + "loss": 0.4022, + "step": 33310 + }, + { + "epoch": 1.8611336145487052, + "grad_norm": 0.48673319816589355, + "learning_rate": 7.028798744957417e-05, + "loss": 0.4277, + "step": 33311 + }, + { + "epoch": 1.8611894851523871, + "grad_norm": 0.4259131848812103, + "learning_rate": 7.02599731062304e-05, + "loss": 0.3835, + "step": 33312 + }, + { + "epoch": 1.8612453557560689, + "grad_norm": 0.5545757412910461, + "learning_rate": 7.02319587628866e-05, + "loss": 0.4305, + "step": 33313 + }, + { + "epoch": 1.8613012263597508, + "grad_norm": 0.3723834455013275, + "learning_rate": 7.02039444195428e-05, + "loss": 0.379, + "step": 33314 + }, + { + "epoch": 1.8613570969634328, + "grad_norm": 0.33681607246398926, + "learning_rate": 7.017593007619901e-05, + "loss": 0.3369, + "step": 33315 + }, + { + "epoch": 1.8614129675671145, + "grad_norm": 0.4746062159538269, + "learning_rate": 7.014791573285523e-05, + "loss": 0.326, + "step": 33316 + }, + { + "epoch": 1.8614688381707964, + "grad_norm": 6.454277038574219, + "learning_rate": 7.011990138951144e-05, + "loss": 0.4167, + "step": 33317 + }, + { + "epoch": 1.8615247087744784, + "grad_norm": 0.5208922624588013, + "learning_rate": 7.009188704616764e-05, + "loss": 0.5263, + "step": 33318 + }, + { + "epoch": 1.8615805793781601, + "grad_norm": 0.3456449508666992, + "learning_rate": 7.006387270282383e-05, + "loss": 0.4104, + "step": 33319 + }, + { + "epoch": 1.861636449981842, + "grad_norm": 0.7628393769264221, + "learning_rate": 7.003585835948005e-05, + "loss": 0.3877, + "step": 33320 + }, + { + "epoch": 1.861692320585524, + "grad_norm": 0.3386150300502777, + "learning_rate": 7.000784401613626e-05, + "loss": 0.3999, + "step": 33321 + }, + { + "epoch": 1.8617481911892058, + "grad_norm": 1.4114211797714233, + "learning_rate": 6.997982967279247e-05, + "loss": 0.5102, + "step": 33322 + }, + { + "epoch": 1.8618040617928877, + "grad_norm": 0.2801196873188019, + "learning_rate": 6.995181532944869e-05, + "loss": 0.3091, + "step": 33323 + }, + { + "epoch": 1.8618599323965697, + "grad_norm": 0.6978117823600769, + "learning_rate": 6.992380098610489e-05, + "loss": 0.3958, + "step": 33324 + }, + { + "epoch": 1.8619158030002514, + "grad_norm": 0.5829229354858398, + "learning_rate": 6.98957866427611e-05, + "loss": 0.4493, + "step": 33325 + }, + { + "epoch": 1.8619716736039333, + "grad_norm": 1.427825927734375, + "learning_rate": 6.98677722994173e-05, + "loss": 0.4974, + "step": 33326 + }, + { + "epoch": 1.8620275442076153, + "grad_norm": 0.5932033658027649, + "learning_rate": 6.983975795607352e-05, + "loss": 0.4626, + "step": 33327 + }, + { + "epoch": 1.862083414811297, + "grad_norm": 0.4329274296760559, + "learning_rate": 6.981174361272973e-05, + "loss": 0.4319, + "step": 33328 + }, + { + "epoch": 1.8621392854149788, + "grad_norm": 0.3802347779273987, + "learning_rate": 6.978372926938592e-05, + "loss": 0.5117, + "step": 33329 + }, + { + "epoch": 1.862195156018661, + "grad_norm": 0.7424359917640686, + "learning_rate": 6.975571492604213e-05, + "loss": 0.556, + "step": 33330 + }, + { + "epoch": 1.8622510266223427, + "grad_norm": 0.45339497923851013, + "learning_rate": 6.972770058269835e-05, + "loss": 0.4495, + "step": 33331 + }, + { + "epoch": 1.8623068972260244, + "grad_norm": 0.4637817442417145, + "learning_rate": 6.969968623935455e-05, + "loss": 0.5642, + "step": 33332 + }, + { + "epoch": 1.8623627678297066, + "grad_norm": 0.33088985085487366, + "learning_rate": 6.967167189601076e-05, + "loss": 0.3657, + "step": 33333 + }, + { + "epoch": 1.8624186384333883, + "grad_norm": 1.6933037042617798, + "learning_rate": 6.964365755266696e-05, + "loss": 0.4543, + "step": 33334 + }, + { + "epoch": 1.86247450903707, + "grad_norm": 0.37309888005256653, + "learning_rate": 6.961564320932318e-05, + "loss": 0.4472, + "step": 33335 + }, + { + "epoch": 1.862530379640752, + "grad_norm": 0.3528648018836975, + "learning_rate": 6.958762886597939e-05, + "loss": 0.4337, + "step": 33336 + }, + { + "epoch": 1.862586250244434, + "grad_norm": 0.6363543272018433, + "learning_rate": 6.95596145226356e-05, + "loss": 0.4035, + "step": 33337 + }, + { + "epoch": 1.8626421208481156, + "grad_norm": 0.5067612528800964, + "learning_rate": 6.95316001792918e-05, + "loss": 0.4831, + "step": 33338 + }, + { + "epoch": 1.8626979914517976, + "grad_norm": 0.43966543674468994, + "learning_rate": 6.950358583594802e-05, + "loss": 0.4415, + "step": 33339 + }, + { + "epoch": 1.8627538620554795, + "grad_norm": 0.4040336012840271, + "learning_rate": 6.947557149260421e-05, + "loss": 0.3722, + "step": 33340 + }, + { + "epoch": 1.8628097326591613, + "grad_norm": 1.8082979917526245, + "learning_rate": 6.944755714926042e-05, + "loss": 0.5194, + "step": 33341 + }, + { + "epoch": 1.8628656032628432, + "grad_norm": 0.5780319571495056, + "learning_rate": 6.941954280591662e-05, + "loss": 0.4318, + "step": 33342 + }, + { + "epoch": 1.8629214738665252, + "grad_norm": 0.4342341423034668, + "learning_rate": 6.939152846257284e-05, + "loss": 0.418, + "step": 33343 + }, + { + "epoch": 1.862977344470207, + "grad_norm": 0.3892075717449188, + "learning_rate": 6.936351411922905e-05, + "loss": 0.3095, + "step": 33344 + }, + { + "epoch": 1.8630332150738889, + "grad_norm": 0.36813706159591675, + "learning_rate": 6.933549977588525e-05, + "loss": 0.3517, + "step": 33345 + }, + { + "epoch": 1.8630890856775708, + "grad_norm": 0.35504215955734253, + "learning_rate": 6.930748543254146e-05, + "loss": 0.448, + "step": 33346 + }, + { + "epoch": 1.8631449562812525, + "grad_norm": 0.4771301746368408, + "learning_rate": 6.927947108919768e-05, + "loss": 0.2951, + "step": 33347 + }, + { + "epoch": 1.8632008268849345, + "grad_norm": 0.3863832950592041, + "learning_rate": 6.925145674585388e-05, + "loss": 0.4477, + "step": 33348 + }, + { + "epoch": 1.8632566974886164, + "grad_norm": 0.5558767318725586, + "learning_rate": 6.922344240251009e-05, + "loss": 0.4376, + "step": 33349 + }, + { + "epoch": 1.8633125680922982, + "grad_norm": 0.3419276177883148, + "learning_rate": 6.919542805916628e-05, + "loss": 0.3049, + "step": 33350 + }, + { + "epoch": 1.8633684386959801, + "grad_norm": 0.5083890557289124, + "learning_rate": 6.91674137158225e-05, + "loss": 0.3707, + "step": 33351 + }, + { + "epoch": 1.863424309299662, + "grad_norm": 0.5906623601913452, + "learning_rate": 6.913939937247871e-05, + "loss": 0.384, + "step": 33352 + }, + { + "epoch": 1.8634801799033438, + "grad_norm": 0.3641810715198517, + "learning_rate": 6.911138502913491e-05, + "loss": 0.3479, + "step": 33353 + }, + { + "epoch": 1.8635360505070258, + "grad_norm": 0.4080997407436371, + "learning_rate": 6.908337068579112e-05, + "loss": 0.4002, + "step": 33354 + }, + { + "epoch": 1.8635919211107077, + "grad_norm": 0.38570940494537354, + "learning_rate": 6.905535634244734e-05, + "loss": 0.3808, + "step": 33355 + }, + { + "epoch": 1.8636477917143894, + "grad_norm": 0.36116474866867065, + "learning_rate": 6.902734199910354e-05, + "loss": 0.4108, + "step": 33356 + }, + { + "epoch": 1.8637036623180714, + "grad_norm": 0.3645663261413574, + "learning_rate": 6.899932765575975e-05, + "loss": 0.4108, + "step": 33357 + }, + { + "epoch": 1.8637595329217533, + "grad_norm": 0.46780797839164734, + "learning_rate": 6.897131331241596e-05, + "loss": 0.4111, + "step": 33358 + }, + { + "epoch": 1.863815403525435, + "grad_norm": 0.6849843263626099, + "learning_rate": 6.894329896907218e-05, + "loss": 0.3206, + "step": 33359 + }, + { + "epoch": 1.863871274129117, + "grad_norm": 0.4968203008174896, + "learning_rate": 6.891528462572838e-05, + "loss": 0.3177, + "step": 33360 + }, + { + "epoch": 1.863927144732799, + "grad_norm": 0.330303430557251, + "learning_rate": 6.888727028238457e-05, + "loss": 0.4398, + "step": 33361 + }, + { + "epoch": 1.8639830153364807, + "grad_norm": 1.731706976890564, + "learning_rate": 6.885925593904078e-05, + "loss": 0.4048, + "step": 33362 + }, + { + "epoch": 1.8640388859401624, + "grad_norm": 0.5595253705978394, + "learning_rate": 6.8831241595697e-05, + "loss": 0.3977, + "step": 33363 + }, + { + "epoch": 1.8640947565438446, + "grad_norm": 1.2139759063720703, + "learning_rate": 6.88032272523532e-05, + "loss": 0.3441, + "step": 33364 + }, + { + "epoch": 1.8641506271475263, + "grad_norm": 0.41545194387435913, + "learning_rate": 6.877521290900941e-05, + "loss": 0.3832, + "step": 33365 + }, + { + "epoch": 1.864206497751208, + "grad_norm": 0.4444138705730438, + "learning_rate": 6.874719856566562e-05, + "loss": 0.3505, + "step": 33366 + }, + { + "epoch": 1.8642623683548902, + "grad_norm": 0.3759993612766266, + "learning_rate": 6.871918422232184e-05, + "loss": 0.3907, + "step": 33367 + }, + { + "epoch": 1.864318238958572, + "grad_norm": 0.7055286169052124, + "learning_rate": 6.869116987897804e-05, + "loss": 0.3329, + "step": 33368 + }, + { + "epoch": 1.8643741095622537, + "grad_norm": 0.380613774061203, + "learning_rate": 6.866315553563425e-05, + "loss": 0.4523, + "step": 33369 + }, + { + "epoch": 1.8644299801659356, + "grad_norm": 2.0179619789123535, + "learning_rate": 6.863514119229045e-05, + "loss": 0.3273, + "step": 33370 + }, + { + "epoch": 1.8644858507696176, + "grad_norm": 0.7475593090057373, + "learning_rate": 6.860712684894667e-05, + "loss": 0.3768, + "step": 33371 + }, + { + "epoch": 1.8645417213732993, + "grad_norm": 1.5207558870315552, + "learning_rate": 6.857911250560287e-05, + "loss": 0.3896, + "step": 33372 + }, + { + "epoch": 1.8645975919769813, + "grad_norm": 1.1231706142425537, + "learning_rate": 6.855109816225907e-05, + "loss": 0.3911, + "step": 33373 + }, + { + "epoch": 1.8646534625806632, + "grad_norm": 0.6419461369514465, + "learning_rate": 6.852308381891528e-05, + "loss": 0.378, + "step": 33374 + }, + { + "epoch": 1.864709333184345, + "grad_norm": 0.44170504808425903, + "learning_rate": 6.84950694755715e-05, + "loss": 0.4769, + "step": 33375 + }, + { + "epoch": 1.864765203788027, + "grad_norm": 0.37091967463493347, + "learning_rate": 6.84670551322277e-05, + "loss": 0.4254, + "step": 33376 + }, + { + "epoch": 1.8648210743917089, + "grad_norm": 1.3322581052780151, + "learning_rate": 6.843904078888391e-05, + "loss": 0.4192, + "step": 33377 + }, + { + "epoch": 1.8648769449953906, + "grad_norm": 0.4513862431049347, + "learning_rate": 6.841102644554011e-05, + "loss": 0.4291, + "step": 33378 + }, + { + "epoch": 1.8649328155990725, + "grad_norm": 0.3716732859611511, + "learning_rate": 6.838301210219633e-05, + "loss": 0.502, + "step": 33379 + }, + { + "epoch": 1.8649886862027545, + "grad_norm": 0.31618961691856384, + "learning_rate": 6.835499775885254e-05, + "loss": 0.2904, + "step": 33380 + }, + { + "epoch": 1.8650445568064362, + "grad_norm": 0.45500198006629944, + "learning_rate": 6.832698341550874e-05, + "loss": 0.3544, + "step": 33381 + }, + { + "epoch": 1.8651004274101182, + "grad_norm": 0.4217720925807953, + "learning_rate": 6.829896907216494e-05, + "loss": 0.358, + "step": 33382 + }, + { + "epoch": 1.8651562980138001, + "grad_norm": 0.4886710047721863, + "learning_rate": 6.827095472882116e-05, + "loss": 0.4309, + "step": 33383 + }, + { + "epoch": 1.8652121686174818, + "grad_norm": 0.32353803515434265, + "learning_rate": 6.824294038547736e-05, + "loss": 0.3724, + "step": 33384 + }, + { + "epoch": 1.8652680392211638, + "grad_norm": 0.4162328541278839, + "learning_rate": 6.821492604213357e-05, + "loss": 0.3749, + "step": 33385 + }, + { + "epoch": 1.8653239098248457, + "grad_norm": 0.4406258463859558, + "learning_rate": 6.818691169878977e-05, + "loss": 0.4985, + "step": 33386 + }, + { + "epoch": 1.8653797804285275, + "grad_norm": 0.5632652640342712, + "learning_rate": 6.815889735544599e-05, + "loss": 0.3868, + "step": 33387 + }, + { + "epoch": 1.8654356510322094, + "grad_norm": 0.6779837608337402, + "learning_rate": 6.81308830121022e-05, + "loss": 0.3929, + "step": 33388 + }, + { + "epoch": 1.8654915216358914, + "grad_norm": 0.5458515882492065, + "learning_rate": 6.81028686687584e-05, + "loss": 0.4946, + "step": 33389 + }, + { + "epoch": 1.865547392239573, + "grad_norm": 0.4438040852546692, + "learning_rate": 6.807485432541462e-05, + "loss": 0.5245, + "step": 33390 + }, + { + "epoch": 1.865603262843255, + "grad_norm": 0.39184674620628357, + "learning_rate": 6.804683998207083e-05, + "loss": 0.3399, + "step": 33391 + }, + { + "epoch": 1.865659133446937, + "grad_norm": 0.3382805585861206, + "learning_rate": 6.801882563872704e-05, + "loss": 0.3713, + "step": 33392 + }, + { + "epoch": 1.8657150040506187, + "grad_norm": 0.6471397876739502, + "learning_rate": 6.799081129538323e-05, + "loss": 0.4389, + "step": 33393 + }, + { + "epoch": 1.8657708746543007, + "grad_norm": 0.6445754170417786, + "learning_rate": 6.796279695203945e-05, + "loss": 0.4526, + "step": 33394 + }, + { + "epoch": 1.8658267452579826, + "grad_norm": 1.1158334016799927, + "learning_rate": 6.793478260869565e-05, + "loss": 0.3621, + "step": 33395 + }, + { + "epoch": 1.8658826158616644, + "grad_norm": 0.29335537552833557, + "learning_rate": 6.790676826535186e-05, + "loss": 0.3742, + "step": 33396 + }, + { + "epoch": 1.865938486465346, + "grad_norm": 0.560205340385437, + "learning_rate": 6.787875392200806e-05, + "loss": 0.4202, + "step": 33397 + }, + { + "epoch": 1.8659943570690283, + "grad_norm": 0.34781011939048767, + "learning_rate": 6.785073957866428e-05, + "loss": 0.4451, + "step": 33398 + }, + { + "epoch": 1.86605022767271, + "grad_norm": 0.32374951243400574, + "learning_rate": 6.782272523532049e-05, + "loss": 0.3524, + "step": 33399 + }, + { + "epoch": 1.8661060982763917, + "grad_norm": 0.5462645888328552, + "learning_rate": 6.77947108919767e-05, + "loss": 0.5202, + "step": 33400 + }, + { + "epoch": 1.866161968880074, + "grad_norm": 0.4110119342803955, + "learning_rate": 6.77666965486329e-05, + "loss": 0.3098, + "step": 33401 + }, + { + "epoch": 1.8662178394837556, + "grad_norm": 0.598021924495697, + "learning_rate": 6.773868220528912e-05, + "loss": 0.3408, + "step": 33402 + }, + { + "epoch": 1.8662737100874374, + "grad_norm": 0.2637447416782379, + "learning_rate": 6.771066786194533e-05, + "loss": 0.3562, + "step": 33403 + }, + { + "epoch": 1.8663295806911193, + "grad_norm": 0.436822772026062, + "learning_rate": 6.768265351860152e-05, + "loss": 0.4549, + "step": 33404 + }, + { + "epoch": 1.8663854512948013, + "grad_norm": 0.5774064064025879, + "learning_rate": 6.765463917525772e-05, + "loss": 0.5516, + "step": 33405 + }, + { + "epoch": 1.866441321898483, + "grad_norm": 0.2598268687725067, + "learning_rate": 6.762662483191394e-05, + "loss": 0.3134, + "step": 33406 + }, + { + "epoch": 1.866497192502165, + "grad_norm": 0.389077365398407, + "learning_rate": 6.759861048857015e-05, + "loss": 0.3793, + "step": 33407 + }, + { + "epoch": 1.866553063105847, + "grad_norm": 0.5207592248916626, + "learning_rate": 6.757059614522636e-05, + "loss": 0.3517, + "step": 33408 + }, + { + "epoch": 1.8666089337095286, + "grad_norm": 0.5015153884887695, + "learning_rate": 6.754258180188256e-05, + "loss": 0.4192, + "step": 33409 + }, + { + "epoch": 1.8666648043132106, + "grad_norm": 4.26040506362915, + "learning_rate": 6.751456745853878e-05, + "loss": 0.4003, + "step": 33410 + }, + { + "epoch": 1.8667206749168925, + "grad_norm": 0.3175593912601471, + "learning_rate": 6.748655311519499e-05, + "loss": 0.3515, + "step": 33411 + }, + { + "epoch": 1.8667765455205743, + "grad_norm": 0.45714065432548523, + "learning_rate": 6.745853877185119e-05, + "loss": 0.362, + "step": 33412 + }, + { + "epoch": 1.8668324161242562, + "grad_norm": 0.5353235602378845, + "learning_rate": 6.74305244285074e-05, + "loss": 0.4292, + "step": 33413 + }, + { + "epoch": 1.8668882867279382, + "grad_norm": 0.5297164916992188, + "learning_rate": 6.74025100851636e-05, + "loss": 0.4385, + "step": 33414 + }, + { + "epoch": 1.86694415733162, + "grad_norm": 0.3833010196685791, + "learning_rate": 6.737449574181981e-05, + "loss": 0.4748, + "step": 33415 + }, + { + "epoch": 1.8670000279353018, + "grad_norm": 0.3608248233795166, + "learning_rate": 6.734648139847602e-05, + "loss": 0.4027, + "step": 33416 + }, + { + "epoch": 1.8670558985389838, + "grad_norm": 2.031564474105835, + "learning_rate": 6.731846705513222e-05, + "loss": 0.4202, + "step": 33417 + }, + { + "epoch": 1.8671117691426655, + "grad_norm": 0.5654221773147583, + "learning_rate": 6.729045271178844e-05, + "loss": 0.4211, + "step": 33418 + }, + { + "epoch": 1.8671676397463475, + "grad_norm": 0.5311909317970276, + "learning_rate": 6.726243836844465e-05, + "loss": 0.3073, + "step": 33419 + }, + { + "epoch": 1.8672235103500294, + "grad_norm": 0.3729856610298157, + "learning_rate": 6.723442402510085e-05, + "loss": 0.4248, + "step": 33420 + }, + { + "epoch": 1.8672793809537112, + "grad_norm": 0.38758307695388794, + "learning_rate": 6.720640968175706e-05, + "loss": 0.4181, + "step": 33421 + }, + { + "epoch": 1.867335251557393, + "grad_norm": 0.3668435513973236, + "learning_rate": 6.717839533841328e-05, + "loss": 0.5072, + "step": 33422 + }, + { + "epoch": 1.867391122161075, + "grad_norm": 0.865195095539093, + "learning_rate": 6.715038099506948e-05, + "loss": 0.4634, + "step": 33423 + }, + { + "epoch": 1.8674469927647568, + "grad_norm": 0.40398892760276794, + "learning_rate": 6.712236665172569e-05, + "loss": 0.3425, + "step": 33424 + }, + { + "epoch": 1.8675028633684387, + "grad_norm": 0.6100161075592041, + "learning_rate": 6.709435230838188e-05, + "loss": 0.4303, + "step": 33425 + }, + { + "epoch": 1.8675587339721207, + "grad_norm": 0.39422231912612915, + "learning_rate": 6.70663379650381e-05, + "loss": 0.4569, + "step": 33426 + }, + { + "epoch": 1.8676146045758024, + "grad_norm": 0.5140197277069092, + "learning_rate": 6.703832362169431e-05, + "loss": 0.4003, + "step": 33427 + }, + { + "epoch": 1.8676704751794844, + "grad_norm": 0.30161231756210327, + "learning_rate": 6.701030927835051e-05, + "loss": 0.3675, + "step": 33428 + }, + { + "epoch": 1.8677263457831663, + "grad_norm": 0.6099092364311218, + "learning_rate": 6.698229493500672e-05, + "loss": 0.3847, + "step": 33429 + }, + { + "epoch": 1.867782216386848, + "grad_norm": 0.6928983926773071, + "learning_rate": 6.695428059166294e-05, + "loss": 0.5656, + "step": 33430 + }, + { + "epoch": 1.8678380869905298, + "grad_norm": 0.40816381573677063, + "learning_rate": 6.692626624831914e-05, + "loss": 0.4447, + "step": 33431 + }, + { + "epoch": 1.867893957594212, + "grad_norm": 3.3517680168151855, + "learning_rate": 6.689825190497535e-05, + "loss": 0.4406, + "step": 33432 + }, + { + "epoch": 1.8679498281978937, + "grad_norm": 0.35272759199142456, + "learning_rate": 6.687023756163156e-05, + "loss": 0.4421, + "step": 33433 + }, + { + "epoch": 1.8680056988015754, + "grad_norm": 1.0532200336456299, + "learning_rate": 6.684222321828777e-05, + "loss": 0.4279, + "step": 33434 + }, + { + "epoch": 1.8680615694052576, + "grad_norm": 0.50798499584198, + "learning_rate": 6.681420887494397e-05, + "loss": 0.3807, + "step": 33435 + }, + { + "epoch": 1.8681174400089393, + "grad_norm": 0.5452051162719727, + "learning_rate": 6.678619453160017e-05, + "loss": 0.3542, + "step": 33436 + }, + { + "epoch": 1.868173310612621, + "grad_norm": 0.35215428471565247, + "learning_rate": 6.675818018825638e-05, + "loss": 0.5221, + "step": 33437 + }, + { + "epoch": 1.868229181216303, + "grad_norm": 0.33557718992233276, + "learning_rate": 6.67301658449126e-05, + "loss": 0.2989, + "step": 33438 + }, + { + "epoch": 1.868285051819985, + "grad_norm": 0.37871554493904114, + "learning_rate": 6.67021515015688e-05, + "loss": 0.4176, + "step": 33439 + }, + { + "epoch": 1.8683409224236667, + "grad_norm": 0.3187194764614105, + "learning_rate": 6.667413715822501e-05, + "loss": 0.3768, + "step": 33440 + }, + { + "epoch": 1.8683967930273486, + "grad_norm": 0.4163011312484741, + "learning_rate": 6.664612281488122e-05, + "loss": 0.6202, + "step": 33441 + }, + { + "epoch": 1.8684526636310306, + "grad_norm": 1.7900282144546509, + "learning_rate": 6.661810847153743e-05, + "loss": 0.4345, + "step": 33442 + }, + { + "epoch": 1.8685085342347123, + "grad_norm": 0.5866592526435852, + "learning_rate": 6.659009412819364e-05, + "loss": 0.4268, + "step": 33443 + }, + { + "epoch": 1.8685644048383943, + "grad_norm": 0.3644791841506958, + "learning_rate": 6.656207978484985e-05, + "loss": 0.3431, + "step": 33444 + }, + { + "epoch": 1.8686202754420762, + "grad_norm": 2.3929545879364014, + "learning_rate": 6.653406544150605e-05, + "loss": 0.4297, + "step": 33445 + }, + { + "epoch": 1.868676146045758, + "grad_norm": 0.3594454526901245, + "learning_rate": 6.650605109816226e-05, + "loss": 0.3481, + "step": 33446 + }, + { + "epoch": 1.8687320166494399, + "grad_norm": 0.4995846152305603, + "learning_rate": 6.647803675481846e-05, + "loss": 0.4133, + "step": 33447 + }, + { + "epoch": 1.8687878872531218, + "grad_norm": 0.41786766052246094, + "learning_rate": 6.645002241147467e-05, + "loss": 0.3977, + "step": 33448 + }, + { + "epoch": 1.8688437578568036, + "grad_norm": 0.3294534385204315, + "learning_rate": 6.642200806813088e-05, + "loss": 0.3921, + "step": 33449 + }, + { + "epoch": 1.8688996284604855, + "grad_norm": 0.46017929911613464, + "learning_rate": 6.63939937247871e-05, + "loss": 0.3486, + "step": 33450 + }, + { + "epoch": 1.8689554990641675, + "grad_norm": 0.35101014375686646, + "learning_rate": 6.63659793814433e-05, + "loss": 0.3585, + "step": 33451 + }, + { + "epoch": 1.8690113696678492, + "grad_norm": 0.30699482560157776, + "learning_rate": 6.63379650380995e-05, + "loss": 0.4138, + "step": 33452 + }, + { + "epoch": 1.8690672402715311, + "grad_norm": 0.3823170065879822, + "learning_rate": 6.630995069475573e-05, + "loss": 0.3355, + "step": 33453 + }, + { + "epoch": 1.869123110875213, + "grad_norm": 0.4800308048725128, + "learning_rate": 6.628193635141193e-05, + "loss": 0.3944, + "step": 33454 + }, + { + "epoch": 1.8691789814788948, + "grad_norm": 0.5142921805381775, + "learning_rate": 6.625392200806814e-05, + "loss": 0.4466, + "step": 33455 + }, + { + "epoch": 1.8692348520825768, + "grad_norm": 0.4079895317554474, + "learning_rate": 6.622590766472434e-05, + "loss": 0.2992, + "step": 33456 + }, + { + "epoch": 1.8692907226862587, + "grad_norm": 0.40119314193725586, + "learning_rate": 6.619789332138055e-05, + "loss": 0.4247, + "step": 33457 + }, + { + "epoch": 1.8693465932899405, + "grad_norm": 1.2186819314956665, + "learning_rate": 6.616987897803675e-05, + "loss": 0.3629, + "step": 33458 + }, + { + "epoch": 1.8694024638936224, + "grad_norm": 0.4172484874725342, + "learning_rate": 6.614186463469296e-05, + "loss": 0.3528, + "step": 33459 + }, + { + "epoch": 1.8694583344973044, + "grad_norm": 0.7139580845832825, + "learning_rate": 6.611385029134917e-05, + "loss": 0.504, + "step": 33460 + }, + { + "epoch": 1.869514205100986, + "grad_norm": 0.5400996208190918, + "learning_rate": 6.608583594800539e-05, + "loss": 0.4974, + "step": 33461 + }, + { + "epoch": 1.869570075704668, + "grad_norm": 0.5375036001205444, + "learning_rate": 6.605782160466159e-05, + "loss": 0.3533, + "step": 33462 + }, + { + "epoch": 1.86962594630835, + "grad_norm": 1.0131808519363403, + "learning_rate": 6.60298072613178e-05, + "loss": 0.3294, + "step": 33463 + }, + { + "epoch": 1.8696818169120317, + "grad_norm": 0.35594937205314636, + "learning_rate": 6.6001792917974e-05, + "loss": 0.4454, + "step": 33464 + }, + { + "epoch": 1.8697376875157135, + "grad_norm": 0.4124080538749695, + "learning_rate": 6.597377857463022e-05, + "loss": 0.414, + "step": 33465 + }, + { + "epoch": 1.8697935581193956, + "grad_norm": 0.4650404751300812, + "learning_rate": 6.594576423128643e-05, + "loss": 0.3726, + "step": 33466 + }, + { + "epoch": 1.8698494287230774, + "grad_norm": 0.5092276334762573, + "learning_rate": 6.591774988794262e-05, + "loss": 0.3942, + "step": 33467 + }, + { + "epoch": 1.869905299326759, + "grad_norm": 0.38606858253479004, + "learning_rate": 6.588973554459883e-05, + "loss": 0.3868, + "step": 33468 + }, + { + "epoch": 1.8699611699304413, + "grad_norm": 0.39061352610588074, + "learning_rate": 6.586172120125505e-05, + "loss": 0.3101, + "step": 33469 + }, + { + "epoch": 1.870017040534123, + "grad_norm": 0.5376286506652832, + "learning_rate": 6.583370685791125e-05, + "loss": 0.3982, + "step": 33470 + }, + { + "epoch": 1.8700729111378047, + "grad_norm": 0.303803414106369, + "learning_rate": 6.580569251456746e-05, + "loss": 0.3332, + "step": 33471 + }, + { + "epoch": 1.8701287817414867, + "grad_norm": 1.847053050994873, + "learning_rate": 6.577767817122366e-05, + "loss": 0.4963, + "step": 33472 + }, + { + "epoch": 1.8701846523451686, + "grad_norm": 0.5145503878593445, + "learning_rate": 6.574966382787988e-05, + "loss": 0.3892, + "step": 33473 + }, + { + "epoch": 1.8702405229488503, + "grad_norm": 2.8284332752227783, + "learning_rate": 6.572164948453609e-05, + "loss": 0.5002, + "step": 33474 + }, + { + "epoch": 1.8702963935525323, + "grad_norm": 0.8383220434188843, + "learning_rate": 6.56936351411923e-05, + "loss": 0.4407, + "step": 33475 + }, + { + "epoch": 1.8703522641562143, + "grad_norm": 0.35525044798851013, + "learning_rate": 6.56656207978485e-05, + "loss": 0.429, + "step": 33476 + }, + { + "epoch": 1.870408134759896, + "grad_norm": 0.3224989175796509, + "learning_rate": 6.563760645450472e-05, + "loss": 0.3665, + "step": 33477 + }, + { + "epoch": 1.870464005363578, + "grad_norm": 0.4106627404689789, + "learning_rate": 6.560959211116091e-05, + "loss": 0.3478, + "step": 33478 + }, + { + "epoch": 1.8705198759672599, + "grad_norm": 0.42355313897132874, + "learning_rate": 6.558157776781712e-05, + "loss": 0.4436, + "step": 33479 + }, + { + "epoch": 1.8705757465709416, + "grad_norm": 0.3803893029689789, + "learning_rate": 6.555356342447332e-05, + "loss": 0.3829, + "step": 33480 + }, + { + "epoch": 1.8706316171746236, + "grad_norm": 0.4280180335044861, + "learning_rate": 6.552554908112954e-05, + "loss": 0.4049, + "step": 33481 + }, + { + "epoch": 1.8706874877783055, + "grad_norm": 0.4148230254650116, + "learning_rate": 6.549753473778575e-05, + "loss": 0.3787, + "step": 33482 + }, + { + "epoch": 1.8707433583819872, + "grad_norm": 1.325531005859375, + "learning_rate": 6.546952039444195e-05, + "loss": 0.4014, + "step": 33483 + }, + { + "epoch": 1.8707992289856692, + "grad_norm": 0.7382338643074036, + "learning_rate": 6.544150605109816e-05, + "loss": 0.396, + "step": 33484 + }, + { + "epoch": 1.8708550995893511, + "grad_norm": 0.6310577392578125, + "learning_rate": 6.541349170775438e-05, + "loss": 0.428, + "step": 33485 + }, + { + "epoch": 1.8709109701930329, + "grad_norm": 0.45361870527267456, + "learning_rate": 6.538547736441059e-05, + "loss": 0.4211, + "step": 33486 + }, + { + "epoch": 1.8709668407967148, + "grad_norm": 1.0186258554458618, + "learning_rate": 6.535746302106679e-05, + "loss": 0.7012, + "step": 33487 + }, + { + "epoch": 1.8710227114003968, + "grad_norm": 0.4794420301914215, + "learning_rate": 6.5329448677723e-05, + "loss": 0.3786, + "step": 33488 + }, + { + "epoch": 1.8710785820040785, + "grad_norm": 0.4684213697910309, + "learning_rate": 6.53014343343792e-05, + "loss": 0.3788, + "step": 33489 + }, + { + "epoch": 1.8711344526077605, + "grad_norm": 0.7868956327438354, + "learning_rate": 6.527341999103541e-05, + "loss": 0.4835, + "step": 33490 + }, + { + "epoch": 1.8711903232114424, + "grad_norm": 0.34326890110969543, + "learning_rate": 6.524540564769161e-05, + "loss": 0.4108, + "step": 33491 + }, + { + "epoch": 1.8712461938151241, + "grad_norm": 1.8000527620315552, + "learning_rate": 6.521739130434782e-05, + "loss": 0.4528, + "step": 33492 + }, + { + "epoch": 1.871302064418806, + "grad_norm": 0.3849748373031616, + "learning_rate": 6.518937696100404e-05, + "loss": 0.3355, + "step": 33493 + }, + { + "epoch": 1.871357935022488, + "grad_norm": 0.4219738245010376, + "learning_rate": 6.516136261766025e-05, + "loss": 0.5087, + "step": 33494 + }, + { + "epoch": 1.8714138056261698, + "grad_norm": 0.34853824973106384, + "learning_rate": 6.513334827431645e-05, + "loss": 0.3211, + "step": 33495 + }, + { + "epoch": 1.8714696762298515, + "grad_norm": 0.41340330243110657, + "learning_rate": 6.510533393097266e-05, + "loss": 0.467, + "step": 33496 + }, + { + "epoch": 1.8715255468335337, + "grad_norm": 0.33704861998558044, + "learning_rate": 6.507731958762888e-05, + "loss": 0.391, + "step": 33497 + }, + { + "epoch": 1.8715814174372154, + "grad_norm": 0.7888216972351074, + "learning_rate": 6.504930524428508e-05, + "loss": 0.4374, + "step": 33498 + }, + { + "epoch": 1.8716372880408971, + "grad_norm": 3.074949026107788, + "learning_rate": 6.502129090094127e-05, + "loss": 0.3394, + "step": 33499 + }, + { + "epoch": 1.8716931586445793, + "grad_norm": 0.5371533036231995, + "learning_rate": 6.499327655759748e-05, + "loss": 0.3474, + "step": 33500 + }, + { + "epoch": 1.8716931586445793, + "eval_cer": 0.08359791823508232, + "eval_loss": 0.3136485815048218, + "eval_runtime": 56.0532, + "eval_samples_per_second": 80.959, + "eval_steps_per_second": 5.067, + "eval_wer": 0.33105220524566353, + "step": 33500 + }, + { + "epoch": 1.871749029248261, + "grad_norm": 0.548852264881134, + "learning_rate": 6.49652622142537e-05, + "loss": 0.4587, + "step": 33501 + }, + { + "epoch": 1.8718048998519428, + "grad_norm": 0.38016366958618164, + "learning_rate": 6.49372478709099e-05, + "loss": 0.3388, + "step": 33502 + }, + { + "epoch": 1.871860770455625, + "grad_norm": 0.6367019414901733, + "learning_rate": 6.490923352756611e-05, + "loss": 0.5604, + "step": 33503 + }, + { + "epoch": 1.8719166410593067, + "grad_norm": 0.44747570157051086, + "learning_rate": 6.488121918422232e-05, + "loss": 0.2954, + "step": 33504 + }, + { + "epoch": 1.8719725116629884, + "grad_norm": 0.329344242811203, + "learning_rate": 6.485320484087854e-05, + "loss": 0.3163, + "step": 33505 + }, + { + "epoch": 1.8720283822666703, + "grad_norm": 0.3121519386768341, + "learning_rate": 6.482519049753474e-05, + "loss": 0.3762, + "step": 33506 + }, + { + "epoch": 1.8720842528703523, + "grad_norm": 0.4660194516181946, + "learning_rate": 6.479717615419095e-05, + "loss": 0.4002, + "step": 33507 + }, + { + "epoch": 1.872140123474034, + "grad_norm": 0.3382743000984192, + "learning_rate": 6.476916181084715e-05, + "loss": 0.3271, + "step": 33508 + }, + { + "epoch": 1.872195994077716, + "grad_norm": 0.4025447368621826, + "learning_rate": 6.474114746750337e-05, + "loss": 0.4801, + "step": 33509 + }, + { + "epoch": 1.872251864681398, + "grad_norm": 0.3833082318305969, + "learning_rate": 6.471313312415957e-05, + "loss": 0.4165, + "step": 33510 + }, + { + "epoch": 1.8723077352850797, + "grad_norm": 0.45783761143684387, + "learning_rate": 6.468511878081577e-05, + "loss": 0.3249, + "step": 33511 + }, + { + "epoch": 1.8723636058887616, + "grad_norm": 0.9589014649391174, + "learning_rate": 6.465710443747198e-05, + "loss": 0.4923, + "step": 33512 + }, + { + "epoch": 1.8724194764924436, + "grad_norm": 0.3349100351333618, + "learning_rate": 6.46290900941282e-05, + "loss": 0.3771, + "step": 33513 + }, + { + "epoch": 1.8724753470961253, + "grad_norm": 0.6499332785606384, + "learning_rate": 6.46010757507844e-05, + "loss": 0.4073, + "step": 33514 + }, + { + "epoch": 1.8725312176998072, + "grad_norm": 0.6287620663642883, + "learning_rate": 6.457306140744061e-05, + "loss": 0.4093, + "step": 33515 + }, + { + "epoch": 1.8725870883034892, + "grad_norm": 0.6636718511581421, + "learning_rate": 6.454504706409681e-05, + "loss": 0.3842, + "step": 33516 + }, + { + "epoch": 1.872642958907171, + "grad_norm": 0.3163396418094635, + "learning_rate": 6.451703272075303e-05, + "loss": 0.3569, + "step": 33517 + }, + { + "epoch": 1.8726988295108529, + "grad_norm": 0.3709493577480316, + "learning_rate": 6.448901837740924e-05, + "loss": 0.3717, + "step": 33518 + }, + { + "epoch": 1.8727547001145348, + "grad_norm": 8.756085395812988, + "learning_rate": 6.446100403406545e-05, + "loss": 0.4616, + "step": 33519 + }, + { + "epoch": 1.8728105707182165, + "grad_norm": 0.3927733600139618, + "learning_rate": 6.443298969072164e-05, + "loss": 0.3595, + "step": 33520 + }, + { + "epoch": 1.8728664413218985, + "grad_norm": 0.34712111949920654, + "learning_rate": 6.440497534737786e-05, + "loss": 0.3701, + "step": 33521 + }, + { + "epoch": 1.8729223119255805, + "grad_norm": 0.5568222999572754, + "learning_rate": 6.437696100403406e-05, + "loss": 0.5409, + "step": 33522 + }, + { + "epoch": 1.8729781825292622, + "grad_norm": 1.1448090076446533, + "learning_rate": 6.434894666069027e-05, + "loss": 0.4065, + "step": 33523 + }, + { + "epoch": 1.8730340531329441, + "grad_norm": 0.3176319897174835, + "learning_rate": 6.432093231734649e-05, + "loss": 0.3578, + "step": 33524 + }, + { + "epoch": 1.873089923736626, + "grad_norm": 0.9525318145751953, + "learning_rate": 6.42929179740027e-05, + "loss": 0.4134, + "step": 33525 + }, + { + "epoch": 1.8731457943403078, + "grad_norm": 0.3032667934894562, + "learning_rate": 6.42649036306589e-05, + "loss": 0.3046, + "step": 33526 + }, + { + "epoch": 1.8732016649439898, + "grad_norm": 0.3857688903808594, + "learning_rate": 6.42368892873151e-05, + "loss": 0.3997, + "step": 33527 + }, + { + "epoch": 1.8732575355476717, + "grad_norm": 2.2771222591400146, + "learning_rate": 6.420887494397132e-05, + "loss": 0.5485, + "step": 33528 + }, + { + "epoch": 1.8733134061513534, + "grad_norm": 0.45497292280197144, + "learning_rate": 6.418086060062753e-05, + "loss": 0.3314, + "step": 33529 + }, + { + "epoch": 1.8733692767550352, + "grad_norm": 0.43181970715522766, + "learning_rate": 6.415284625728374e-05, + "loss": 0.4086, + "step": 33530 + }, + { + "epoch": 1.8734251473587173, + "grad_norm": 0.4528804421424866, + "learning_rate": 6.412483191393993e-05, + "loss": 0.3519, + "step": 33531 + }, + { + "epoch": 1.873481017962399, + "grad_norm": 0.2917894721031189, + "learning_rate": 6.409681757059615e-05, + "loss": 0.3362, + "step": 33532 + }, + { + "epoch": 1.8735368885660808, + "grad_norm": 0.9387564063072205, + "learning_rate": 6.406880322725235e-05, + "loss": 0.4272, + "step": 33533 + }, + { + "epoch": 1.873592759169763, + "grad_norm": 0.38161781430244446, + "learning_rate": 6.404078888390856e-05, + "loss": 0.4525, + "step": 33534 + }, + { + "epoch": 1.8736486297734447, + "grad_norm": 0.3893602788448334, + "learning_rate": 6.401277454056477e-05, + "loss": 0.4541, + "step": 33535 + }, + { + "epoch": 1.8737045003771264, + "grad_norm": 0.7509521842002869, + "learning_rate": 6.398476019722098e-05, + "loss": 0.3775, + "step": 33536 + }, + { + "epoch": 1.8737603709808086, + "grad_norm": 0.4313136041164398, + "learning_rate": 6.395674585387719e-05, + "loss": 0.4779, + "step": 33537 + }, + { + "epoch": 1.8738162415844903, + "grad_norm": 0.3785873055458069, + "learning_rate": 6.39287315105334e-05, + "loss": 0.3806, + "step": 33538 + }, + { + "epoch": 1.873872112188172, + "grad_norm": 0.5326160192489624, + "learning_rate": 6.39007171671896e-05, + "loss": 0.5816, + "step": 33539 + }, + { + "epoch": 1.873927982791854, + "grad_norm": 0.6020594239234924, + "learning_rate": 6.387270282384582e-05, + "loss": 0.5263, + "step": 33540 + }, + { + "epoch": 1.873983853395536, + "grad_norm": 0.4279043972492218, + "learning_rate": 6.384468848050203e-05, + "loss": 0.4714, + "step": 33541 + }, + { + "epoch": 1.8740397239992177, + "grad_norm": 0.3196878135204315, + "learning_rate": 6.381667413715822e-05, + "loss": 0.3458, + "step": 33542 + }, + { + "epoch": 1.8740955946028997, + "grad_norm": 0.5974107384681702, + "learning_rate": 6.378865979381443e-05, + "loss": 0.3289, + "step": 33543 + }, + { + "epoch": 1.8741514652065816, + "grad_norm": 0.35247287154197693, + "learning_rate": 6.376064545047064e-05, + "loss": 0.408, + "step": 33544 + }, + { + "epoch": 1.8742073358102633, + "grad_norm": 0.6769580841064453, + "learning_rate": 6.373263110712685e-05, + "loss": 0.5, + "step": 33545 + }, + { + "epoch": 1.8742632064139453, + "grad_norm": 0.83806973695755, + "learning_rate": 6.370461676378306e-05, + "loss": 0.3592, + "step": 33546 + }, + { + "epoch": 1.8743190770176272, + "grad_norm": 0.36293479800224304, + "learning_rate": 6.367660242043926e-05, + "loss": 0.3746, + "step": 33547 + }, + { + "epoch": 1.874374947621309, + "grad_norm": 0.5821139216423035, + "learning_rate": 6.364858807709548e-05, + "loss": 0.5368, + "step": 33548 + }, + { + "epoch": 1.874430818224991, + "grad_norm": 0.4178779423236847, + "learning_rate": 6.362057373375169e-05, + "loss": 0.3913, + "step": 33549 + }, + { + "epoch": 1.8744866888286729, + "grad_norm": 0.3990817368030548, + "learning_rate": 6.359255939040789e-05, + "loss": 0.366, + "step": 33550 + }, + { + "epoch": 1.8745425594323546, + "grad_norm": 0.28411149978637695, + "learning_rate": 6.35645450470641e-05, + "loss": 0.4023, + "step": 33551 + }, + { + "epoch": 1.8745984300360365, + "grad_norm": 0.7083909511566162, + "learning_rate": 6.35365307037203e-05, + "loss": 0.3096, + "step": 33552 + }, + { + "epoch": 1.8746543006397185, + "grad_norm": 0.3577840328216553, + "learning_rate": 6.350851636037651e-05, + "loss": 0.3421, + "step": 33553 + }, + { + "epoch": 1.8747101712434002, + "grad_norm": 0.6491180658340454, + "learning_rate": 6.348050201703272e-05, + "loss": 0.4697, + "step": 33554 + }, + { + "epoch": 1.8747660418470822, + "grad_norm": 0.4017687737941742, + "learning_rate": 6.345248767368892e-05, + "loss": 0.4015, + "step": 33555 + }, + { + "epoch": 1.8748219124507641, + "grad_norm": 0.644782304763794, + "learning_rate": 6.342447333034514e-05, + "loss": 0.4153, + "step": 33556 + }, + { + "epoch": 1.8748777830544459, + "grad_norm": 0.39242473244667053, + "learning_rate": 6.339645898700135e-05, + "loss": 0.3057, + "step": 33557 + }, + { + "epoch": 1.8749336536581278, + "grad_norm": 0.3987269401550293, + "learning_rate": 6.336844464365755e-05, + "loss": 0.4928, + "step": 33558 + }, + { + "epoch": 1.8749895242618098, + "grad_norm": 0.7329679727554321, + "learning_rate": 6.334043030031376e-05, + "loss": 0.4835, + "step": 33559 + }, + { + "epoch": 1.8750453948654915, + "grad_norm": 1.113215684890747, + "learning_rate": 6.331241595696998e-05, + "loss": 0.4486, + "step": 33560 + }, + { + "epoch": 1.8751012654691734, + "grad_norm": 0.2926773726940155, + "learning_rate": 6.328440161362618e-05, + "loss": 0.3682, + "step": 33561 + }, + { + "epoch": 1.8751571360728554, + "grad_norm": 0.5033329129219055, + "learning_rate": 6.325638727028239e-05, + "loss": 0.3847, + "step": 33562 + }, + { + "epoch": 1.8752130066765371, + "grad_norm": 0.8065803050994873, + "learning_rate": 6.322837292693858e-05, + "loss": 0.4142, + "step": 33563 + }, + { + "epoch": 1.8752688772802188, + "grad_norm": 0.4758709669113159, + "learning_rate": 6.32003585835948e-05, + "loss": 0.4338, + "step": 33564 + }, + { + "epoch": 1.875324747883901, + "grad_norm": 0.6176608204841614, + "learning_rate": 6.317234424025101e-05, + "loss": 0.3699, + "step": 33565 + }, + { + "epoch": 1.8753806184875828, + "grad_norm": 0.4479329288005829, + "learning_rate": 6.314432989690721e-05, + "loss": 0.3858, + "step": 33566 + }, + { + "epoch": 1.8754364890912645, + "grad_norm": 2.1033895015716553, + "learning_rate": 6.311631555356342e-05, + "loss": 0.4351, + "step": 33567 + }, + { + "epoch": 1.8754923596949467, + "grad_norm": 0.3383899927139282, + "learning_rate": 6.308830121021964e-05, + "loss": 0.4264, + "step": 33568 + }, + { + "epoch": 1.8755482302986284, + "grad_norm": 0.3503607511520386, + "learning_rate": 6.306028686687584e-05, + "loss": 0.4427, + "step": 33569 + }, + { + "epoch": 1.8756041009023101, + "grad_norm": 1.1657464504241943, + "learning_rate": 6.303227252353205e-05, + "loss": 0.3649, + "step": 33570 + }, + { + "epoch": 1.8756599715059923, + "grad_norm": 0.5140806436538696, + "learning_rate": 6.300425818018826e-05, + "loss": 0.3593, + "step": 33571 + }, + { + "epoch": 1.875715842109674, + "grad_norm": 0.6383193135261536, + "learning_rate": 6.297624383684448e-05, + "loss": 0.5431, + "step": 33572 + }, + { + "epoch": 1.8757717127133557, + "grad_norm": 0.38334882259368896, + "learning_rate": 6.294822949350068e-05, + "loss": 0.3268, + "step": 33573 + }, + { + "epoch": 1.8758275833170377, + "grad_norm": 0.4675576686859131, + "learning_rate": 6.292021515015687e-05, + "loss": 0.4678, + "step": 33574 + }, + { + "epoch": 1.8758834539207196, + "grad_norm": 0.5027660727500916, + "learning_rate": 6.289220080681308e-05, + "loss": 0.3324, + "step": 33575 + }, + { + "epoch": 1.8759393245244014, + "grad_norm": 1.4418091773986816, + "learning_rate": 6.28641864634693e-05, + "loss": 0.3063, + "step": 33576 + }, + { + "epoch": 1.8759951951280833, + "grad_norm": 0.544094979763031, + "learning_rate": 6.28361721201255e-05, + "loss": 0.4668, + "step": 33577 + }, + { + "epoch": 1.8760510657317653, + "grad_norm": 0.44140616059303284, + "learning_rate": 6.280815777678171e-05, + "loss": 0.3925, + "step": 33578 + }, + { + "epoch": 1.876106936335447, + "grad_norm": 0.3128601014614105, + "learning_rate": 6.278014343343792e-05, + "loss": 0.3156, + "step": 33579 + }, + { + "epoch": 1.876162806939129, + "grad_norm": 2.00406813621521, + "learning_rate": 6.275212909009414e-05, + "loss": 0.427, + "step": 33580 + }, + { + "epoch": 1.876218677542811, + "grad_norm": 3.101161241531372, + "learning_rate": 6.272411474675034e-05, + "loss": 0.4243, + "step": 33581 + }, + { + "epoch": 1.8762745481464926, + "grad_norm": 0.34877535700798035, + "learning_rate": 6.269610040340655e-05, + "loss": 0.3692, + "step": 33582 + }, + { + "epoch": 1.8763304187501746, + "grad_norm": 0.8754026889801025, + "learning_rate": 6.266808606006275e-05, + "loss": 0.4461, + "step": 33583 + }, + { + "epoch": 1.8763862893538565, + "grad_norm": 0.37111517786979675, + "learning_rate": 6.264007171671896e-05, + "loss": 0.3918, + "step": 33584 + }, + { + "epoch": 1.8764421599575383, + "grad_norm": 0.5411069393157959, + "learning_rate": 6.261205737337516e-05, + "loss": 0.4708, + "step": 33585 + }, + { + "epoch": 1.8764980305612202, + "grad_norm": 0.49663209915161133, + "learning_rate": 6.258404303003137e-05, + "loss": 0.3834, + "step": 33586 + }, + { + "epoch": 1.8765539011649022, + "grad_norm": 0.48616382479667664, + "learning_rate": 6.255602868668758e-05, + "loss": 0.4724, + "step": 33587 + }, + { + "epoch": 1.876609771768584, + "grad_norm": 0.3550819456577301, + "learning_rate": 6.25280143433438e-05, + "loss": 0.4079, + "step": 33588 + }, + { + "epoch": 1.8766656423722659, + "grad_norm": 0.32342368364334106, + "learning_rate": 6.25e-05, + "loss": 0.4178, + "step": 33589 + }, + { + "epoch": 1.8767215129759478, + "grad_norm": 2.2929162979125977, + "learning_rate": 6.247198565665621e-05, + "loss": 0.3606, + "step": 33590 + }, + { + "epoch": 1.8767773835796295, + "grad_norm": 0.8160014748573303, + "learning_rate": 6.244397131331243e-05, + "loss": 0.5592, + "step": 33591 + }, + { + "epoch": 1.8768332541833115, + "grad_norm": 0.5783356428146362, + "learning_rate": 6.241595696996862e-05, + "loss": 0.396, + "step": 33592 + }, + { + "epoch": 1.8768891247869934, + "grad_norm": 0.37676867842674255, + "learning_rate": 6.238794262662484e-05, + "loss": 0.3966, + "step": 33593 + }, + { + "epoch": 1.8769449953906752, + "grad_norm": 0.36656659841537476, + "learning_rate": 6.235992828328104e-05, + "loss": 0.4067, + "step": 33594 + }, + { + "epoch": 1.8770008659943571, + "grad_norm": 1.684362769126892, + "learning_rate": 6.233191393993725e-05, + "loss": 0.4657, + "step": 33595 + }, + { + "epoch": 1.877056736598039, + "grad_norm": 0.7500331401824951, + "learning_rate": 6.230389959659346e-05, + "loss": 0.571, + "step": 33596 + }, + { + "epoch": 1.8771126072017208, + "grad_norm": 0.6341351270675659, + "learning_rate": 6.227588525324966e-05, + "loss": 0.4101, + "step": 33597 + }, + { + "epoch": 1.8771684778054025, + "grad_norm": 0.45272210240364075, + "learning_rate": 6.224787090990587e-05, + "loss": 0.4867, + "step": 33598 + }, + { + "epoch": 1.8772243484090847, + "grad_norm": 0.38514021039009094, + "learning_rate": 6.221985656656209e-05, + "loss": 0.3945, + "step": 33599 + }, + { + "epoch": 1.8772802190127664, + "grad_norm": 0.4279957115650177, + "learning_rate": 6.219184222321829e-05, + "loss": 0.4112, + "step": 33600 + }, + { + "epoch": 1.8773360896164482, + "grad_norm": 1.4102996587753296, + "learning_rate": 6.21638278798745e-05, + "loss": 0.3409, + "step": 33601 + }, + { + "epoch": 1.8773919602201303, + "grad_norm": 7.961245536804199, + "learning_rate": 6.21358135365307e-05, + "loss": 0.3434, + "step": 33602 + }, + { + "epoch": 1.877447830823812, + "grad_norm": 0.7655268311500549, + "learning_rate": 6.210779919318691e-05, + "loss": 0.3422, + "step": 33603 + }, + { + "epoch": 1.8775037014274938, + "grad_norm": 0.7575247287750244, + "learning_rate": 6.207978484984312e-05, + "loss": 0.5147, + "step": 33604 + }, + { + "epoch": 1.877559572031176, + "grad_norm": 0.5847148895263672, + "learning_rate": 6.205177050649934e-05, + "loss": 0.4063, + "step": 33605 + }, + { + "epoch": 1.8776154426348577, + "grad_norm": 0.3239635229110718, + "learning_rate": 6.202375616315554e-05, + "loss": 0.3256, + "step": 33606 + }, + { + "epoch": 1.8776713132385394, + "grad_norm": 0.29545503854751587, + "learning_rate": 6.199574181981175e-05, + "loss": 0.389, + "step": 33607 + }, + { + "epoch": 1.8777271838422214, + "grad_norm": 2.97273588180542, + "learning_rate": 6.196772747646795e-05, + "loss": 0.4493, + "step": 33608 + }, + { + "epoch": 1.8777830544459033, + "grad_norm": 0.4641498923301697, + "learning_rate": 6.193971313312416e-05, + "loss": 0.4444, + "step": 33609 + }, + { + "epoch": 1.877838925049585, + "grad_norm": 0.5178989171981812, + "learning_rate": 6.191169878978036e-05, + "loss": 0.4578, + "step": 33610 + }, + { + "epoch": 1.877894795653267, + "grad_norm": 0.4870426654815674, + "learning_rate": 6.188368444643658e-05, + "loss": 0.437, + "step": 33611 + }, + { + "epoch": 1.877950666256949, + "grad_norm": 0.3719465434551239, + "learning_rate": 6.185567010309279e-05, + "loss": 0.4164, + "step": 33612 + }, + { + "epoch": 1.8780065368606307, + "grad_norm": 0.4625658690929413, + "learning_rate": 6.1827655759749e-05, + "loss": 0.321, + "step": 33613 + }, + { + "epoch": 1.8780624074643126, + "grad_norm": 0.3052750825881958, + "learning_rate": 6.17996414164052e-05, + "loss": 0.3952, + "step": 33614 + }, + { + "epoch": 1.8781182780679946, + "grad_norm": 0.9113104939460754, + "learning_rate": 6.17716270730614e-05, + "loss": 0.4808, + "step": 33615 + }, + { + "epoch": 1.8781741486716763, + "grad_norm": 0.5879862904548645, + "learning_rate": 6.174361272971761e-05, + "loss": 0.4112, + "step": 33616 + }, + { + "epoch": 1.8782300192753583, + "grad_norm": 0.3747504651546478, + "learning_rate": 6.171559838637383e-05, + "loss": 0.3572, + "step": 33617 + }, + { + "epoch": 1.8782858898790402, + "grad_norm": 0.49284231662750244, + "learning_rate": 6.168758404303004e-05, + "loss": 0.4449, + "step": 33618 + }, + { + "epoch": 1.878341760482722, + "grad_norm": 0.3454899191856384, + "learning_rate": 6.165956969968624e-05, + "loss": 0.4141, + "step": 33619 + }, + { + "epoch": 1.878397631086404, + "grad_norm": 0.3933853507041931, + "learning_rate": 6.163155535634245e-05, + "loss": 0.3236, + "step": 33620 + }, + { + "epoch": 1.8784535016900858, + "grad_norm": 0.8555752038955688, + "learning_rate": 6.160354101299866e-05, + "loss": 0.5227, + "step": 33621 + }, + { + "epoch": 1.8785093722937676, + "grad_norm": 0.3955298662185669, + "learning_rate": 6.157552666965486e-05, + "loss": 0.3819, + "step": 33622 + }, + { + "epoch": 1.8785652428974495, + "grad_norm": 0.5482027530670166, + "learning_rate": 6.154751232631108e-05, + "loss": 0.3763, + "step": 33623 + }, + { + "epoch": 1.8786211135011315, + "grad_norm": 2.4385414123535156, + "learning_rate": 6.151949798296727e-05, + "loss": 0.4186, + "step": 33624 + }, + { + "epoch": 1.8786769841048132, + "grad_norm": 0.7996050119400024, + "learning_rate": 6.149148363962349e-05, + "loss": 0.4646, + "step": 33625 + }, + { + "epoch": 1.8787328547084952, + "grad_norm": 0.7323092222213745, + "learning_rate": 6.14634692962797e-05, + "loss": 0.4238, + "step": 33626 + }, + { + "epoch": 1.878788725312177, + "grad_norm": 0.35817208886146545, + "learning_rate": 6.14354549529359e-05, + "loss": 0.3006, + "step": 33627 + }, + { + "epoch": 1.8788445959158588, + "grad_norm": 0.33474525809288025, + "learning_rate": 6.140744060959211e-05, + "loss": 0.368, + "step": 33628 + }, + { + "epoch": 1.8789004665195408, + "grad_norm": 0.5311456918716431, + "learning_rate": 6.137942626624832e-05, + "loss": 0.3461, + "step": 33629 + }, + { + "epoch": 1.8789563371232227, + "grad_norm": 0.4906572103500366, + "learning_rate": 6.135141192290452e-05, + "loss": 0.471, + "step": 33630 + }, + { + "epoch": 1.8790122077269045, + "grad_norm": 5.080285549163818, + "learning_rate": 6.132339757956074e-05, + "loss": 0.4157, + "step": 33631 + }, + { + "epoch": 1.8790680783305862, + "grad_norm": 0.48111552000045776, + "learning_rate": 6.129538323621695e-05, + "loss": 0.5005, + "step": 33632 + }, + { + "epoch": 1.8791239489342684, + "grad_norm": 0.551455020904541, + "learning_rate": 6.126736889287315e-05, + "loss": 0.5277, + "step": 33633 + }, + { + "epoch": 1.87917981953795, + "grad_norm": 0.529376208782196, + "learning_rate": 6.123935454952936e-05, + "loss": 0.4141, + "step": 33634 + }, + { + "epoch": 1.8792356901416318, + "grad_norm": 0.46296426653862, + "learning_rate": 6.121134020618556e-05, + "loss": 0.4291, + "step": 33635 + }, + { + "epoch": 1.879291560745314, + "grad_norm": 0.48719361424446106, + "learning_rate": 6.118332586284177e-05, + "loss": 0.4003, + "step": 33636 + }, + { + "epoch": 1.8793474313489957, + "grad_norm": 0.37383246421813965, + "learning_rate": 6.115531151949799e-05, + "loss": 0.3709, + "step": 33637 + }, + { + "epoch": 1.8794033019526775, + "grad_norm": 0.3766135573387146, + "learning_rate": 6.11272971761542e-05, + "loss": 0.4281, + "step": 33638 + }, + { + "epoch": 1.8794591725563596, + "grad_norm": 0.47992846369743347, + "learning_rate": 6.10992828328104e-05, + "loss": 0.5224, + "step": 33639 + }, + { + "epoch": 1.8795150431600414, + "grad_norm": 1.319037675857544, + "learning_rate": 6.10712684894666e-05, + "loss": 0.4011, + "step": 33640 + }, + { + "epoch": 1.879570913763723, + "grad_norm": 0.4574945569038391, + "learning_rate": 6.104325414612281e-05, + "loss": 0.3851, + "step": 33641 + }, + { + "epoch": 1.879626784367405, + "grad_norm": 0.43861737847328186, + "learning_rate": 6.1015239802779025e-05, + "loss": 0.4129, + "step": 33642 + }, + { + "epoch": 1.879682654971087, + "grad_norm": 0.4670538306236267, + "learning_rate": 6.098722545943523e-05, + "loss": 0.4866, + "step": 33643 + }, + { + "epoch": 1.8797385255747687, + "grad_norm": 0.3316364586353302, + "learning_rate": 6.095921111609144e-05, + "loss": 0.4345, + "step": 33644 + }, + { + "epoch": 1.8797943961784507, + "grad_norm": 0.3774320185184479, + "learning_rate": 6.093119677274764e-05, + "loss": 0.4084, + "step": 33645 + }, + { + "epoch": 1.8798502667821326, + "grad_norm": 0.44247886538505554, + "learning_rate": 6.0903182429403855e-05, + "loss": 0.2845, + "step": 33646 + }, + { + "epoch": 1.8799061373858144, + "grad_norm": 1.0796630382537842, + "learning_rate": 6.087516808606006e-05, + "loss": 0.2981, + "step": 33647 + }, + { + "epoch": 1.8799620079894963, + "grad_norm": 0.5913124084472656, + "learning_rate": 6.084715374271627e-05, + "loss": 0.3762, + "step": 33648 + }, + { + "epoch": 1.8800178785931783, + "grad_norm": 0.4016968309879303, + "learning_rate": 6.081913939937248e-05, + "loss": 0.4394, + "step": 33649 + }, + { + "epoch": 1.88007374919686, + "grad_norm": 0.3643381595611572, + "learning_rate": 6.079112505602869e-05, + "loss": 0.4545, + "step": 33650 + }, + { + "epoch": 1.880129619800542, + "grad_norm": 0.3884713649749756, + "learning_rate": 6.076311071268489e-05, + "loss": 0.2908, + "step": 33651 + }, + { + "epoch": 1.880185490404224, + "grad_norm": 0.48470813035964966, + "learning_rate": 6.07350963693411e-05, + "loss": 0.4836, + "step": 33652 + }, + { + "epoch": 1.8802413610079056, + "grad_norm": 0.3237745761871338, + "learning_rate": 6.070708202599731e-05, + "loss": 0.3119, + "step": 33653 + }, + { + "epoch": 1.8802972316115876, + "grad_norm": 0.33742234110832214, + "learning_rate": 6.067906768265352e-05, + "loss": 0.3921, + "step": 33654 + }, + { + "epoch": 1.8803531022152695, + "grad_norm": 0.3883976936340332, + "learning_rate": 6.065105333930973e-05, + "loss": 0.4947, + "step": 33655 + }, + { + "epoch": 1.8804089728189513, + "grad_norm": 0.5753815770149231, + "learning_rate": 6.062303899596593e-05, + "loss": 0.3972, + "step": 33656 + }, + { + "epoch": 1.8804648434226332, + "grad_norm": 0.3176683485507965, + "learning_rate": 6.059502465262214e-05, + "loss": 0.4344, + "step": 33657 + }, + { + "epoch": 1.8805207140263152, + "grad_norm": 0.9830725193023682, + "learning_rate": 6.056701030927835e-05, + "loss": 0.4735, + "step": 33658 + }, + { + "epoch": 1.8805765846299969, + "grad_norm": 0.4248071014881134, + "learning_rate": 6.0538995965934564e-05, + "loss": 0.3871, + "step": 33659 + }, + { + "epoch": 1.8806324552336788, + "grad_norm": 0.41707172989845276, + "learning_rate": 6.051098162259077e-05, + "loss": 0.3621, + "step": 33660 + }, + { + "epoch": 1.8806883258373608, + "grad_norm": 0.6513271331787109, + "learning_rate": 6.0482967279246976e-05, + "loss": 0.361, + "step": 33661 + }, + { + "epoch": 1.8807441964410425, + "grad_norm": 0.34370043873786926, + "learning_rate": 6.045495293590318e-05, + "loss": 0.3615, + "step": 33662 + }, + { + "epoch": 1.8808000670447245, + "grad_norm": 0.6562367081642151, + "learning_rate": 6.0426938592559394e-05, + "loss": 0.508, + "step": 33663 + }, + { + "epoch": 1.8808559376484064, + "grad_norm": 0.3373478651046753, + "learning_rate": 6.03989242492156e-05, + "loss": 0.3631, + "step": 33664 + }, + { + "epoch": 1.8809118082520881, + "grad_norm": 0.47856348752975464, + "learning_rate": 6.037090990587181e-05, + "loss": 0.5303, + "step": 33665 + }, + { + "epoch": 1.8809676788557699, + "grad_norm": 0.34003835916519165, + "learning_rate": 6.034289556252802e-05, + "loss": 0.4026, + "step": 33666 + }, + { + "epoch": 1.881023549459452, + "grad_norm": 0.5376291871070862, + "learning_rate": 6.0314881219184224e-05, + "loss": 0.5715, + "step": 33667 + }, + { + "epoch": 1.8810794200631338, + "grad_norm": 0.39697882533073425, + "learning_rate": 6.028686687584043e-05, + "loss": 0.3511, + "step": 33668 + }, + { + "epoch": 1.8811352906668155, + "grad_norm": 0.37549275159835815, + "learning_rate": 6.025885253249664e-05, + "loss": 0.4842, + "step": 33669 + }, + { + "epoch": 1.8811911612704977, + "grad_norm": 2.0245444774627686, + "learning_rate": 6.023083818915285e-05, + "loss": 0.4092, + "step": 33670 + }, + { + "epoch": 1.8812470318741794, + "grad_norm": 0.3196111023426056, + "learning_rate": 6.020282384580906e-05, + "loss": 0.3729, + "step": 33671 + }, + { + "epoch": 1.8813029024778611, + "grad_norm": 1.8841211795806885, + "learning_rate": 6.017480950246526e-05, + "loss": 0.4304, + "step": 33672 + }, + { + "epoch": 1.8813587730815433, + "grad_norm": 0.3621622622013092, + "learning_rate": 6.014679515912147e-05, + "loss": 0.4188, + "step": 33673 + }, + { + "epoch": 1.881414643685225, + "grad_norm": 0.40744587779045105, + "learning_rate": 6.011878081577768e-05, + "loss": 0.3945, + "step": 33674 + }, + { + "epoch": 1.8814705142889068, + "grad_norm": 0.6618550419807434, + "learning_rate": 6.009076647243389e-05, + "loss": 0.6711, + "step": 33675 + }, + { + "epoch": 1.8815263848925887, + "grad_norm": 0.4131869077682495, + "learning_rate": 6.00627521290901e-05, + "loss": 0.639, + "step": 33676 + }, + { + "epoch": 1.8815822554962707, + "grad_norm": 0.4998542368412018, + "learning_rate": 6.00347377857463e-05, + "loss": 0.4236, + "step": 33677 + }, + { + "epoch": 1.8816381260999524, + "grad_norm": 0.3717058300971985, + "learning_rate": 6.000672344240251e-05, + "loss": 0.4108, + "step": 33678 + }, + { + "epoch": 1.8816939967036344, + "grad_norm": 0.8415923118591309, + "learning_rate": 5.997870909905872e-05, + "loss": 0.4229, + "step": 33679 + }, + { + "epoch": 1.8817498673073163, + "grad_norm": 0.5597980618476868, + "learning_rate": 5.995069475571493e-05, + "loss": 0.3652, + "step": 33680 + }, + { + "epoch": 1.881805737910998, + "grad_norm": 0.8050516843795776, + "learning_rate": 5.992268041237114e-05, + "loss": 0.3763, + "step": 33681 + }, + { + "epoch": 1.88186160851468, + "grad_norm": 0.6861029863357544, + "learning_rate": 5.989466606902734e-05, + "loss": 0.4066, + "step": 33682 + }, + { + "epoch": 1.881917479118362, + "grad_norm": 6.249019622802734, + "learning_rate": 5.986665172568355e-05, + "loss": 0.3324, + "step": 33683 + }, + { + "epoch": 1.8819733497220437, + "grad_norm": 0.8213231563568115, + "learning_rate": 5.983863738233976e-05, + "loss": 0.2878, + "step": 33684 + }, + { + "epoch": 1.8820292203257256, + "grad_norm": 0.5643568634986877, + "learning_rate": 5.981062303899597e-05, + "loss": 0.6253, + "step": 33685 + }, + { + "epoch": 1.8820850909294076, + "grad_norm": 0.6186957955360413, + "learning_rate": 5.9782608695652175e-05, + "loss": 0.3897, + "step": 33686 + }, + { + "epoch": 1.8821409615330893, + "grad_norm": 0.3512934446334839, + "learning_rate": 5.975459435230839e-05, + "loss": 0.4844, + "step": 33687 + }, + { + "epoch": 1.8821968321367712, + "grad_norm": 0.4658972918987274, + "learning_rate": 5.972658000896459e-05, + "loss": 0.4567, + "step": 33688 + }, + { + "epoch": 1.8822527027404532, + "grad_norm": 0.3627346456050873, + "learning_rate": 5.96985656656208e-05, + "loss": 0.3314, + "step": 33689 + }, + { + "epoch": 1.882308573344135, + "grad_norm": 0.31169599294662476, + "learning_rate": 5.9670551322277005e-05, + "loss": 0.3123, + "step": 33690 + }, + { + "epoch": 1.8823644439478169, + "grad_norm": 0.35051557421684265, + "learning_rate": 5.964253697893322e-05, + "loss": 0.3443, + "step": 33691 + }, + { + "epoch": 1.8824203145514988, + "grad_norm": 0.8885484933853149, + "learning_rate": 5.9614522635589424e-05, + "loss": 0.4415, + "step": 33692 + }, + { + "epoch": 1.8824761851551806, + "grad_norm": 0.6732179522514343, + "learning_rate": 5.958650829224563e-05, + "loss": 0.333, + "step": 33693 + }, + { + "epoch": 1.8825320557588625, + "grad_norm": 1.5600768327713013, + "learning_rate": 5.9558493948901836e-05, + "loss": 0.4315, + "step": 33694 + }, + { + "epoch": 1.8825879263625445, + "grad_norm": 3.9667277336120605, + "learning_rate": 5.953047960555805e-05, + "loss": 0.4987, + "step": 33695 + }, + { + "epoch": 1.8826437969662262, + "grad_norm": 0.5637587904930115, + "learning_rate": 5.9502465262214254e-05, + "loss": 0.3266, + "step": 33696 + }, + { + "epoch": 1.8826996675699081, + "grad_norm": 0.40798771381378174, + "learning_rate": 5.9474450918870467e-05, + "loss": 0.3446, + "step": 33697 + }, + { + "epoch": 1.88275553817359, + "grad_norm": 0.391683429479599, + "learning_rate": 5.9446436575526666e-05, + "loss": 0.3394, + "step": 33698 + }, + { + "epoch": 1.8828114087772718, + "grad_norm": 0.39397379755973816, + "learning_rate": 5.941842223218288e-05, + "loss": 0.377, + "step": 33699 + }, + { + "epoch": 1.8828672793809536, + "grad_norm": 0.28892120718955994, + "learning_rate": 5.9390407888839084e-05, + "loss": 0.3126, + "step": 33700 + }, + { + "epoch": 1.8829231499846357, + "grad_norm": 0.3385789394378662, + "learning_rate": 5.9362393545495297e-05, + "loss": 0.3773, + "step": 33701 + }, + { + "epoch": 1.8829790205883175, + "grad_norm": 0.40511220693588257, + "learning_rate": 5.93343792021515e-05, + "loss": 0.3574, + "step": 33702 + }, + { + "epoch": 1.8830348911919992, + "grad_norm": 0.3705117702484131, + "learning_rate": 5.9306364858807715e-05, + "loss": 0.4133, + "step": 33703 + }, + { + "epoch": 1.8830907617956814, + "grad_norm": 0.4163115620613098, + "learning_rate": 5.9278350515463914e-05, + "loss": 0.424, + "step": 33704 + }, + { + "epoch": 1.883146632399363, + "grad_norm": 0.37685996294021606, + "learning_rate": 5.9250336172120127e-05, + "loss": 0.4251, + "step": 33705 + }, + { + "epoch": 1.8832025030030448, + "grad_norm": 1.5424500703811646, + "learning_rate": 5.922232182877633e-05, + "loss": 0.4569, + "step": 33706 + }, + { + "epoch": 1.8832583736067268, + "grad_norm": 0.4311223328113556, + "learning_rate": 5.9194307485432545e-05, + "loss": 0.4075, + "step": 33707 + }, + { + "epoch": 1.8833142442104087, + "grad_norm": 0.5496557950973511, + "learning_rate": 5.916629314208875e-05, + "loss": 0.4288, + "step": 33708 + }, + { + "epoch": 1.8833701148140904, + "grad_norm": 0.5474255084991455, + "learning_rate": 5.913827879874496e-05, + "loss": 0.4561, + "step": 33709 + }, + { + "epoch": 1.8834259854177724, + "grad_norm": 0.41941559314727783, + "learning_rate": 5.911026445540116e-05, + "loss": 0.6448, + "step": 33710 + }, + { + "epoch": 1.8834818560214543, + "grad_norm": 0.3979928493499756, + "learning_rate": 5.9082250112057375e-05, + "loss": 0.4903, + "step": 33711 + }, + { + "epoch": 1.883537726625136, + "grad_norm": 2.198906421661377, + "learning_rate": 5.905423576871358e-05, + "loss": 0.3714, + "step": 33712 + }, + { + "epoch": 1.883593597228818, + "grad_norm": 2.477229356765747, + "learning_rate": 5.9026221425369793e-05, + "loss": 0.4818, + "step": 33713 + }, + { + "epoch": 1.8836494678325, + "grad_norm": 0.5073561668395996, + "learning_rate": 5.899820708202599e-05, + "loss": 0.4802, + "step": 33714 + }, + { + "epoch": 1.8837053384361817, + "grad_norm": 0.8925992250442505, + "learning_rate": 5.8970192738682205e-05, + "loss": 0.3983, + "step": 33715 + }, + { + "epoch": 1.8837612090398637, + "grad_norm": 1.1073909997940063, + "learning_rate": 5.894217839533841e-05, + "loss": 0.4977, + "step": 33716 + }, + { + "epoch": 1.8838170796435456, + "grad_norm": 1.83687424659729, + "learning_rate": 5.8914164051994623e-05, + "loss": 0.5465, + "step": 33717 + }, + { + "epoch": 1.8838729502472273, + "grad_norm": 0.5136669874191284, + "learning_rate": 5.888614970865083e-05, + "loss": 0.4711, + "step": 33718 + }, + { + "epoch": 1.8839288208509093, + "grad_norm": 0.343605637550354, + "learning_rate": 5.885813536530704e-05, + "loss": 0.4125, + "step": 33719 + }, + { + "epoch": 1.8839846914545912, + "grad_norm": 0.3227352797985077, + "learning_rate": 5.883012102196324e-05, + "loss": 0.3057, + "step": 33720 + }, + { + "epoch": 1.884040562058273, + "grad_norm": 0.3207407295703888, + "learning_rate": 5.8802106678619454e-05, + "loss": 0.4067, + "step": 33721 + }, + { + "epoch": 1.884096432661955, + "grad_norm": 0.33921435475349426, + "learning_rate": 5.877409233527566e-05, + "loss": 0.3697, + "step": 33722 + }, + { + "epoch": 1.8841523032656369, + "grad_norm": 0.4029504358768463, + "learning_rate": 5.874607799193187e-05, + "loss": 0.4252, + "step": 33723 + }, + { + "epoch": 1.8842081738693186, + "grad_norm": 0.5555506944656372, + "learning_rate": 5.8718063648588085e-05, + "loss": 0.41, + "step": 33724 + }, + { + "epoch": 1.8842640444730006, + "grad_norm": 0.5627740621566772, + "learning_rate": 5.8690049305244284e-05, + "loss": 0.397, + "step": 33725 + }, + { + "epoch": 1.8843199150766825, + "grad_norm": 3.375183582305908, + "learning_rate": 5.8662034961900496e-05, + "loss": 0.3485, + "step": 33726 + }, + { + "epoch": 1.8843757856803642, + "grad_norm": 0.5108027458190918, + "learning_rate": 5.86340206185567e-05, + "loss": 0.4694, + "step": 33727 + }, + { + "epoch": 1.8844316562840462, + "grad_norm": 0.5317733287811279, + "learning_rate": 5.8606006275212915e-05, + "loss": 0.4341, + "step": 33728 + }, + { + "epoch": 1.8844875268877281, + "grad_norm": 0.3432862460613251, + "learning_rate": 5.857799193186912e-05, + "loss": 0.3587, + "step": 33729 + }, + { + "epoch": 1.8845433974914099, + "grad_norm": 0.5401731133460999, + "learning_rate": 5.8549977588525326e-05, + "loss": 0.5878, + "step": 33730 + }, + { + "epoch": 1.8845992680950918, + "grad_norm": 0.3872847855091095, + "learning_rate": 5.852196324518153e-05, + "loss": 0.3989, + "step": 33731 + }, + { + "epoch": 1.8846551386987738, + "grad_norm": 1.5058729648590088, + "learning_rate": 5.8493948901837745e-05, + "loss": 0.4739, + "step": 33732 + }, + { + "epoch": 1.8847110093024555, + "grad_norm": 4.166914939880371, + "learning_rate": 5.846593455849395e-05, + "loss": 0.4096, + "step": 33733 + }, + { + "epoch": 1.8847668799061372, + "grad_norm": 0.5735809803009033, + "learning_rate": 5.843792021515016e-05, + "loss": 0.5506, + "step": 33734 + }, + { + "epoch": 1.8848227505098194, + "grad_norm": 0.445576936006546, + "learning_rate": 5.840990587180637e-05, + "loss": 0.5281, + "step": 33735 + }, + { + "epoch": 1.8848786211135011, + "grad_norm": 0.43428078293800354, + "learning_rate": 5.8381891528462575e-05, + "loss": 0.5762, + "step": 33736 + }, + { + "epoch": 1.8849344917171829, + "grad_norm": 0.3942629098892212, + "learning_rate": 5.835387718511878e-05, + "loss": 0.3991, + "step": 33737 + }, + { + "epoch": 1.884990362320865, + "grad_norm": 0.43536216020584106, + "learning_rate": 5.832586284177499e-05, + "loss": 0.4926, + "step": 33738 + }, + { + "epoch": 1.8850462329245468, + "grad_norm": 0.8541391491889954, + "learning_rate": 5.82978484984312e-05, + "loss": 0.3152, + "step": 33739 + }, + { + "epoch": 1.8851021035282285, + "grad_norm": 0.4226725697517395, + "learning_rate": 5.826983415508741e-05, + "loss": 0.4012, + "step": 33740 + }, + { + "epoch": 1.8851579741319104, + "grad_norm": 0.3795933425426483, + "learning_rate": 5.824181981174361e-05, + "loss": 0.3417, + "step": 33741 + }, + { + "epoch": 1.8852138447355924, + "grad_norm": 0.4915877878665924, + "learning_rate": 5.821380546839982e-05, + "loss": 0.3567, + "step": 33742 + }, + { + "epoch": 1.8852697153392741, + "grad_norm": 0.49547335505485535, + "learning_rate": 5.818579112505603e-05, + "loss": 0.2859, + "step": 33743 + }, + { + "epoch": 1.885325585942956, + "grad_norm": 0.4600854516029358, + "learning_rate": 5.815777678171224e-05, + "loss": 0.3763, + "step": 33744 + }, + { + "epoch": 1.885381456546638, + "grad_norm": 0.4558922350406647, + "learning_rate": 5.812976243836845e-05, + "loss": 0.3683, + "step": 33745 + }, + { + "epoch": 1.8854373271503198, + "grad_norm": 1.1565515995025635, + "learning_rate": 5.810174809502465e-05, + "loss": 0.4019, + "step": 33746 + }, + { + "epoch": 1.8854931977540017, + "grad_norm": 0.33650892972946167, + "learning_rate": 5.807373375168086e-05, + "loss": 0.3817, + "step": 33747 + }, + { + "epoch": 1.8855490683576837, + "grad_norm": 0.6196249127388, + "learning_rate": 5.804571940833707e-05, + "loss": 0.3376, + "step": 33748 + }, + { + "epoch": 1.8856049389613654, + "grad_norm": 0.6577578186988831, + "learning_rate": 5.801770506499328e-05, + "loss": 0.4431, + "step": 33749 + }, + { + "epoch": 1.8856608095650473, + "grad_norm": 0.47853440046310425, + "learning_rate": 5.798969072164949e-05, + "loss": 0.3529, + "step": 33750 + }, + { + "epoch": 1.8857166801687293, + "grad_norm": 0.40958669781684875, + "learning_rate": 5.7961676378305696e-05, + "loss": 0.3359, + "step": 33751 + }, + { + "epoch": 1.885772550772411, + "grad_norm": 0.4030410051345825, + "learning_rate": 5.79336620349619e-05, + "loss": 0.3962, + "step": 33752 + }, + { + "epoch": 1.885828421376093, + "grad_norm": 5.871595859527588, + "learning_rate": 5.790564769161811e-05, + "loss": 0.5357, + "step": 33753 + }, + { + "epoch": 1.885884291979775, + "grad_norm": 0.3145682215690613, + "learning_rate": 5.787763334827432e-05, + "loss": 0.4439, + "step": 33754 + }, + { + "epoch": 1.8859401625834566, + "grad_norm": 0.2921803295612335, + "learning_rate": 5.7849619004930526e-05, + "loss": 0.3127, + "step": 33755 + }, + { + "epoch": 1.8859960331871386, + "grad_norm": 0.3905917704105377, + "learning_rate": 5.782160466158674e-05, + "loss": 0.4255, + "step": 33756 + }, + { + "epoch": 1.8860519037908205, + "grad_norm": 0.40316227078437805, + "learning_rate": 5.779359031824294e-05, + "loss": 0.4092, + "step": 33757 + }, + { + "epoch": 1.8861077743945023, + "grad_norm": 0.44969454407691956, + "learning_rate": 5.776557597489915e-05, + "loss": 0.4207, + "step": 33758 + }, + { + "epoch": 1.8861636449981842, + "grad_norm": 0.29543039202690125, + "learning_rate": 5.7737561631555356e-05, + "loss": 0.3192, + "step": 33759 + }, + { + "epoch": 1.8862195156018662, + "grad_norm": 0.5796817541122437, + "learning_rate": 5.770954728821157e-05, + "loss": 0.4862, + "step": 33760 + }, + { + "epoch": 1.886275386205548, + "grad_norm": 0.9143925309181213, + "learning_rate": 5.7681532944867774e-05, + "loss": 0.3814, + "step": 33761 + }, + { + "epoch": 1.8863312568092299, + "grad_norm": 0.48023906350135803, + "learning_rate": 5.765351860152398e-05, + "loss": 0.3717, + "step": 33762 + }, + { + "epoch": 1.8863871274129118, + "grad_norm": 0.5372528433799744, + "learning_rate": 5.7625504258180186e-05, + "loss": 0.3249, + "step": 33763 + }, + { + "epoch": 1.8864429980165935, + "grad_norm": 0.48777535557746887, + "learning_rate": 5.75974899148364e-05, + "loss": 0.643, + "step": 33764 + }, + { + "epoch": 1.8864988686202755, + "grad_norm": 0.3577386140823364, + "learning_rate": 5.7569475571492604e-05, + "loss": 0.4241, + "step": 33765 + }, + { + "epoch": 1.8865547392239574, + "grad_norm": 0.38987573981285095, + "learning_rate": 5.754146122814882e-05, + "loss": 0.3457, + "step": 33766 + }, + { + "epoch": 1.8866106098276392, + "grad_norm": 0.4526149034500122, + "learning_rate": 5.7513446884805016e-05, + "loss": 0.4255, + "step": 33767 + }, + { + "epoch": 1.886666480431321, + "grad_norm": 0.42551538348197937, + "learning_rate": 5.748543254146123e-05, + "loss": 0.4523, + "step": 33768 + }, + { + "epoch": 1.886722351035003, + "grad_norm": 0.37254971265792847, + "learning_rate": 5.7457418198117434e-05, + "loss": 0.3826, + "step": 33769 + }, + { + "epoch": 1.8867782216386848, + "grad_norm": 0.402055025100708, + "learning_rate": 5.742940385477365e-05, + "loss": 0.3868, + "step": 33770 + }, + { + "epoch": 1.8868340922423665, + "grad_norm": 0.9263764023780823, + "learning_rate": 5.740138951142985e-05, + "loss": 0.3251, + "step": 33771 + }, + { + "epoch": 1.8868899628460487, + "grad_norm": 0.5523051023483276, + "learning_rate": 5.7373375168086065e-05, + "loss": 0.549, + "step": 33772 + }, + { + "epoch": 1.8869458334497304, + "grad_norm": 0.7885268926620483, + "learning_rate": 5.7345360824742264e-05, + "loss": 0.438, + "step": 33773 + }, + { + "epoch": 1.8870017040534122, + "grad_norm": 1.0546255111694336, + "learning_rate": 5.731734648139848e-05, + "loss": 0.4136, + "step": 33774 + }, + { + "epoch": 1.8870575746570941, + "grad_norm": 0.5091260671615601, + "learning_rate": 5.728933213805468e-05, + "loss": 0.4107, + "step": 33775 + }, + { + "epoch": 1.887113445260776, + "grad_norm": 0.7330193519592285, + "learning_rate": 5.7261317794710895e-05, + "loss": 0.4998, + "step": 33776 + }, + { + "epoch": 1.8871693158644578, + "grad_norm": 0.7786251902580261, + "learning_rate": 5.72333034513671e-05, + "loss": 0.4926, + "step": 33777 + }, + { + "epoch": 1.8872251864681397, + "grad_norm": 0.6807215213775635, + "learning_rate": 5.720528910802331e-05, + "loss": 0.4188, + "step": 33778 + }, + { + "epoch": 1.8872810570718217, + "grad_norm": 0.3894605338573456, + "learning_rate": 5.717727476467951e-05, + "loss": 0.3458, + "step": 33779 + }, + { + "epoch": 1.8873369276755034, + "grad_norm": 0.7511091232299805, + "learning_rate": 5.7149260421335725e-05, + "loss": 0.4415, + "step": 33780 + }, + { + "epoch": 1.8873927982791854, + "grad_norm": 0.9305733442306519, + "learning_rate": 5.712124607799193e-05, + "loss": 0.5381, + "step": 33781 + }, + { + "epoch": 1.8874486688828673, + "grad_norm": 0.7400648593902588, + "learning_rate": 5.7093231734648144e-05, + "loss": 0.4399, + "step": 33782 + }, + { + "epoch": 1.887504539486549, + "grad_norm": 0.5024551153182983, + "learning_rate": 5.706521739130434e-05, + "loss": 0.448, + "step": 33783 + }, + { + "epoch": 1.887560410090231, + "grad_norm": 0.42965802550315857, + "learning_rate": 5.7037203047960555e-05, + "loss": 0.3459, + "step": 33784 + }, + { + "epoch": 1.887616280693913, + "grad_norm": 0.43468308448791504, + "learning_rate": 5.700918870461676e-05, + "loss": 0.4495, + "step": 33785 + }, + { + "epoch": 1.8876721512975947, + "grad_norm": 2.411900758743286, + "learning_rate": 5.6981174361272974e-05, + "loss": 0.4255, + "step": 33786 + }, + { + "epoch": 1.8877280219012766, + "grad_norm": 0.5723736882209778, + "learning_rate": 5.695316001792918e-05, + "loss": 0.3782, + "step": 33787 + }, + { + "epoch": 1.8877838925049586, + "grad_norm": 0.427412211894989, + "learning_rate": 5.692514567458539e-05, + "loss": 0.3809, + "step": 33788 + }, + { + "epoch": 1.8878397631086403, + "grad_norm": 0.5827173590660095, + "learning_rate": 5.689713133124159e-05, + "loss": 0.3919, + "step": 33789 + }, + { + "epoch": 1.8878956337123223, + "grad_norm": 0.39833158254623413, + "learning_rate": 5.6869116987897804e-05, + "loss": 0.455, + "step": 33790 + }, + { + "epoch": 1.8879515043160042, + "grad_norm": 0.4799654483795166, + "learning_rate": 5.684110264455401e-05, + "loss": 0.4216, + "step": 33791 + }, + { + "epoch": 1.888007374919686, + "grad_norm": 0.4143189489841461, + "learning_rate": 5.681308830121022e-05, + "loss": 0.4453, + "step": 33792 + }, + { + "epoch": 1.888063245523368, + "grad_norm": 0.5489906668663025, + "learning_rate": 5.6785073957866435e-05, + "loss": 0.4676, + "step": 33793 + }, + { + "epoch": 1.8881191161270499, + "grad_norm": 0.438204824924469, + "learning_rate": 5.6757059614522634e-05, + "loss": 0.3885, + "step": 33794 + }, + { + "epoch": 1.8881749867307316, + "grad_norm": 0.47333666682243347, + "learning_rate": 5.6729045271178846e-05, + "loss": 0.3991, + "step": 33795 + }, + { + "epoch": 1.8882308573344135, + "grad_norm": 0.5929920077323914, + "learning_rate": 5.670103092783505e-05, + "loss": 0.3773, + "step": 33796 + }, + { + "epoch": 1.8882867279380955, + "grad_norm": 0.35315388441085815, + "learning_rate": 5.6673016584491265e-05, + "loss": 0.4635, + "step": 33797 + }, + { + "epoch": 1.8883425985417772, + "grad_norm": 0.7371035814285278, + "learning_rate": 5.664500224114747e-05, + "loss": 0.4231, + "step": 33798 + }, + { + "epoch": 1.8883984691454592, + "grad_norm": 0.5185027122497559, + "learning_rate": 5.6616987897803677e-05, + "loss": 0.4185, + "step": 33799 + }, + { + "epoch": 1.8884543397491411, + "grad_norm": 0.4498947262763977, + "learning_rate": 5.658897355445988e-05, + "loss": 0.367, + "step": 33800 + }, + { + "epoch": 1.8885102103528228, + "grad_norm": 0.9471501111984253, + "learning_rate": 5.6560959211116095e-05, + "loss": 0.4916, + "step": 33801 + }, + { + "epoch": 1.8885660809565046, + "grad_norm": 0.43813377618789673, + "learning_rate": 5.65329448677723e-05, + "loss": 0.4437, + "step": 33802 + }, + { + "epoch": 1.8886219515601868, + "grad_norm": 2.3076982498168945, + "learning_rate": 5.650493052442851e-05, + "loss": 0.485, + "step": 33803 + }, + { + "epoch": 1.8886778221638685, + "grad_norm": 0.8722963333129883, + "learning_rate": 5.647691618108472e-05, + "loss": 0.3436, + "step": 33804 + }, + { + "epoch": 1.8887336927675502, + "grad_norm": 1.2634778022766113, + "learning_rate": 5.6448901837740925e-05, + "loss": 0.4238, + "step": 33805 + }, + { + "epoch": 1.8887895633712324, + "grad_norm": 23.049283981323242, + "learning_rate": 5.642088749439713e-05, + "loss": 0.4117, + "step": 33806 + }, + { + "epoch": 1.8888454339749141, + "grad_norm": 0.6996992826461792, + "learning_rate": 5.639287315105334e-05, + "loss": 0.589, + "step": 33807 + }, + { + "epoch": 1.8889013045785958, + "grad_norm": 1.3217602968215942, + "learning_rate": 5.636485880770955e-05, + "loss": 0.3714, + "step": 33808 + }, + { + "epoch": 1.8889571751822778, + "grad_norm": 0.762109100818634, + "learning_rate": 5.633684446436576e-05, + "loss": 0.4951, + "step": 33809 + }, + { + "epoch": 1.8890130457859597, + "grad_norm": 0.401608407497406, + "learning_rate": 5.630883012102196e-05, + "loss": 0.5235, + "step": 33810 + }, + { + "epoch": 1.8890689163896415, + "grad_norm": 0.6864024996757507, + "learning_rate": 5.6280815777678173e-05, + "loss": 0.4237, + "step": 33811 + }, + { + "epoch": 1.8891247869933234, + "grad_norm": 0.3382447063922882, + "learning_rate": 5.625280143433438e-05, + "loss": 0.3407, + "step": 33812 + }, + { + "epoch": 1.8891806575970054, + "grad_norm": 0.6158654689788818, + "learning_rate": 5.622478709099059e-05, + "loss": 0.4864, + "step": 33813 + }, + { + "epoch": 1.889236528200687, + "grad_norm": 0.42978063225746155, + "learning_rate": 5.61967727476468e-05, + "loss": 0.434, + "step": 33814 + }, + { + "epoch": 1.889292398804369, + "grad_norm": 0.443751722574234, + "learning_rate": 5.6168758404303003e-05, + "loss": 0.3235, + "step": 33815 + }, + { + "epoch": 1.889348269408051, + "grad_norm": 0.7813637256622314, + "learning_rate": 5.614074406095921e-05, + "loss": 0.4606, + "step": 33816 + }, + { + "epoch": 1.8894041400117327, + "grad_norm": 0.9085599184036255, + "learning_rate": 5.611272971761542e-05, + "loss": 0.4067, + "step": 33817 + }, + { + "epoch": 1.8894600106154147, + "grad_norm": 1.2867217063903809, + "learning_rate": 5.608471537427163e-05, + "loss": 0.3604, + "step": 33818 + }, + { + "epoch": 1.8895158812190966, + "grad_norm": 0.42817074060440063, + "learning_rate": 5.605670103092784e-05, + "loss": 0.3924, + "step": 33819 + }, + { + "epoch": 1.8895717518227784, + "grad_norm": 0.4733828604221344, + "learning_rate": 5.6028686687584046e-05, + "loss": 0.5552, + "step": 33820 + }, + { + "epoch": 1.8896276224264603, + "grad_norm": 0.42827004194259644, + "learning_rate": 5.600067234424025e-05, + "loss": 0.4383, + "step": 33821 + }, + { + "epoch": 1.8896834930301423, + "grad_norm": 0.5520772337913513, + "learning_rate": 5.597265800089646e-05, + "loss": 0.4632, + "step": 33822 + }, + { + "epoch": 1.889739363633824, + "grad_norm": 1.7946430444717407, + "learning_rate": 5.594464365755267e-05, + "loss": 0.4183, + "step": 33823 + }, + { + "epoch": 1.889795234237506, + "grad_norm": 2.1439547538757324, + "learning_rate": 5.5916629314208876e-05, + "loss": 0.4112, + "step": 33824 + }, + { + "epoch": 1.889851104841188, + "grad_norm": 0.586154580116272, + "learning_rate": 5.588861497086509e-05, + "loss": 0.317, + "step": 33825 + }, + { + "epoch": 1.8899069754448696, + "grad_norm": 0.3722447454929352, + "learning_rate": 5.586060062752129e-05, + "loss": 0.3272, + "step": 33826 + }, + { + "epoch": 1.8899628460485516, + "grad_norm": 0.3980291485786438, + "learning_rate": 5.58325862841775e-05, + "loss": 0.3836, + "step": 33827 + }, + { + "epoch": 1.8900187166522335, + "grad_norm": 0.45132559537887573, + "learning_rate": 5.5804571940833706e-05, + "loss": 0.3281, + "step": 33828 + }, + { + "epoch": 1.8900745872559153, + "grad_norm": 0.4224640429019928, + "learning_rate": 5.577655759748992e-05, + "loss": 0.4219, + "step": 33829 + }, + { + "epoch": 1.8901304578595972, + "grad_norm": 0.3454042077064514, + "learning_rate": 5.5748543254146125e-05, + "loss": 0.3195, + "step": 33830 + }, + { + "epoch": 1.8901863284632792, + "grad_norm": 1.09646475315094, + "learning_rate": 5.572052891080233e-05, + "loss": 0.4252, + "step": 33831 + }, + { + "epoch": 1.890242199066961, + "grad_norm": 0.7772781848907471, + "learning_rate": 5.5692514567458536e-05, + "loss": 0.4011, + "step": 33832 + }, + { + "epoch": 1.8902980696706428, + "grad_norm": 0.4106890857219696, + "learning_rate": 5.566450022411475e-05, + "loss": 0.3894, + "step": 33833 + }, + { + "epoch": 1.8903539402743248, + "grad_norm": 0.3629416525363922, + "learning_rate": 5.5636485880770955e-05, + "loss": 0.4193, + "step": 33834 + }, + { + "epoch": 1.8904098108780065, + "grad_norm": 1.0851796865463257, + "learning_rate": 5.560847153742717e-05, + "loss": 0.4316, + "step": 33835 + }, + { + "epoch": 1.8904656814816883, + "grad_norm": 0.7155457139015198, + "learning_rate": 5.558045719408337e-05, + "loss": 0.476, + "step": 33836 + }, + { + "epoch": 1.8905215520853704, + "grad_norm": 0.4346226155757904, + "learning_rate": 5.555244285073958e-05, + "loss": 0.3874, + "step": 33837 + }, + { + "epoch": 1.8905774226890522, + "grad_norm": 0.815687894821167, + "learning_rate": 5.5524428507395785e-05, + "loss": 0.518, + "step": 33838 + }, + { + "epoch": 1.8906332932927339, + "grad_norm": 0.6869335174560547, + "learning_rate": 5.5496414164052e-05, + "loss": 0.3572, + "step": 33839 + }, + { + "epoch": 1.890689163896416, + "grad_norm": 1.1183921098709106, + "learning_rate": 5.54683998207082e-05, + "loss": 0.3537, + "step": 33840 + }, + { + "epoch": 1.8907450345000978, + "grad_norm": 1.6485602855682373, + "learning_rate": 5.5440385477364416e-05, + "loss": 0.4675, + "step": 33841 + }, + { + "epoch": 1.8908009051037795, + "grad_norm": 0.40356943011283875, + "learning_rate": 5.5412371134020615e-05, + "loss": 0.3685, + "step": 33842 + }, + { + "epoch": 1.8908567757074615, + "grad_norm": 1.6808357238769531, + "learning_rate": 5.538435679067683e-05, + "loss": 0.4436, + "step": 33843 + }, + { + "epoch": 1.8909126463111434, + "grad_norm": 0.566108763217926, + "learning_rate": 5.535634244733303e-05, + "loss": 0.4824, + "step": 33844 + }, + { + "epoch": 1.8909685169148251, + "grad_norm": 0.5419701933860779, + "learning_rate": 5.5328328103989246e-05, + "loss": 0.3452, + "step": 33845 + }, + { + "epoch": 1.891024387518507, + "grad_norm": 1.9168987274169922, + "learning_rate": 5.530031376064545e-05, + "loss": 0.4832, + "step": 33846 + }, + { + "epoch": 1.891080258122189, + "grad_norm": 0.42280134558677673, + "learning_rate": 5.527229941730166e-05, + "loss": 0.2956, + "step": 33847 + }, + { + "epoch": 1.8911361287258708, + "grad_norm": 0.6202862858772278, + "learning_rate": 5.524428507395786e-05, + "loss": 0.6451, + "step": 33848 + }, + { + "epoch": 1.8911919993295527, + "grad_norm": 0.4256094992160797, + "learning_rate": 5.5216270730614076e-05, + "loss": 0.4595, + "step": 33849 + }, + { + "epoch": 1.8912478699332347, + "grad_norm": 0.8671340346336365, + "learning_rate": 5.518825638727028e-05, + "loss": 0.4073, + "step": 33850 + }, + { + "epoch": 1.8913037405369164, + "grad_norm": 0.68622225522995, + "learning_rate": 5.5160242043926494e-05, + "loss": 0.3747, + "step": 33851 + }, + { + "epoch": 1.8913596111405984, + "grad_norm": 0.49908360838890076, + "learning_rate": 5.513222770058269e-05, + "loss": 0.4084, + "step": 33852 + }, + { + "epoch": 1.8914154817442803, + "grad_norm": 0.5174360871315002, + "learning_rate": 5.5104213357238906e-05, + "loss": 0.5305, + "step": 33853 + }, + { + "epoch": 1.891471352347962, + "grad_norm": 0.5664327144622803, + "learning_rate": 5.507619901389511e-05, + "loss": 0.3864, + "step": 33854 + }, + { + "epoch": 1.891527222951644, + "grad_norm": 0.6892515420913696, + "learning_rate": 5.5048184670551324e-05, + "loss": 0.3945, + "step": 33855 + }, + { + "epoch": 1.891583093555326, + "grad_norm": 0.418325275182724, + "learning_rate": 5.502017032720753e-05, + "loss": 0.524, + "step": 33856 + }, + { + "epoch": 1.8916389641590077, + "grad_norm": 0.9345685839653015, + "learning_rate": 5.499215598386374e-05, + "loss": 0.3547, + "step": 33857 + }, + { + "epoch": 1.8916948347626896, + "grad_norm": 0.5010624527931213, + "learning_rate": 5.496414164051994e-05, + "loss": 0.502, + "step": 33858 + }, + { + "epoch": 1.8917507053663716, + "grad_norm": 0.3753693997859955, + "learning_rate": 5.4936127297176154e-05, + "loss": 0.4391, + "step": 33859 + }, + { + "epoch": 1.8918065759700533, + "grad_norm": 0.603979229927063, + "learning_rate": 5.490811295383237e-05, + "loss": 0.6281, + "step": 33860 + }, + { + "epoch": 1.8918624465737353, + "grad_norm": 0.33995139598846436, + "learning_rate": 5.488009861048857e-05, + "loss": 0.4679, + "step": 33861 + }, + { + "epoch": 1.8919183171774172, + "grad_norm": 0.3436858057975769, + "learning_rate": 5.4852084267144785e-05, + "loss": 0.527, + "step": 33862 + }, + { + "epoch": 1.891974187781099, + "grad_norm": 0.38881629705429077, + "learning_rate": 5.4824069923800984e-05, + "loss": 0.3507, + "step": 33863 + }, + { + "epoch": 1.8920300583847809, + "grad_norm": 0.3711162209510803, + "learning_rate": 5.47960555804572e-05, + "loss": 0.3352, + "step": 33864 + }, + { + "epoch": 1.8920859289884628, + "grad_norm": 0.8599634766578674, + "learning_rate": 5.47680412371134e-05, + "loss": 0.3766, + "step": 33865 + }, + { + "epoch": 1.8921417995921446, + "grad_norm": 0.6148390173912048, + "learning_rate": 5.4740026893769615e-05, + "loss": 0.6086, + "step": 33866 + }, + { + "epoch": 1.8921976701958265, + "grad_norm": 0.36063918471336365, + "learning_rate": 5.471201255042582e-05, + "loss": 0.3606, + "step": 33867 + }, + { + "epoch": 1.8922535407995085, + "grad_norm": 5.897696495056152, + "learning_rate": 5.468399820708203e-05, + "loss": 0.4115, + "step": 33868 + }, + { + "epoch": 1.8923094114031902, + "grad_norm": 0.4347590506076813, + "learning_rate": 5.465598386373823e-05, + "loss": 0.42, + "step": 33869 + }, + { + "epoch": 1.892365282006872, + "grad_norm": 0.4963781237602234, + "learning_rate": 5.4627969520394445e-05, + "loss": 0.3306, + "step": 33870 + }, + { + "epoch": 1.892421152610554, + "grad_norm": 1.4552351236343384, + "learning_rate": 5.459995517705065e-05, + "loss": 0.2937, + "step": 33871 + }, + { + "epoch": 1.8924770232142358, + "grad_norm": 0.5625531673431396, + "learning_rate": 5.4571940833706864e-05, + "loss": 0.414, + "step": 33872 + }, + { + "epoch": 1.8925328938179176, + "grad_norm": 0.5369486212730408, + "learning_rate": 5.454392649036307e-05, + "loss": 0.4465, + "step": 33873 + }, + { + "epoch": 1.8925887644215997, + "grad_norm": 1.2658953666687012, + "learning_rate": 5.4515912147019275e-05, + "loss": 0.3839, + "step": 33874 + }, + { + "epoch": 1.8926446350252815, + "grad_norm": 0.5293002724647522, + "learning_rate": 5.448789780367548e-05, + "loss": 0.4381, + "step": 33875 + }, + { + "epoch": 1.8927005056289632, + "grad_norm": 0.5388786792755127, + "learning_rate": 5.4459883460331694e-05, + "loss": 0.4324, + "step": 33876 + }, + { + "epoch": 1.8927563762326451, + "grad_norm": 0.623543918132782, + "learning_rate": 5.44318691169879e-05, + "loss": 0.3756, + "step": 33877 + }, + { + "epoch": 1.892812246836327, + "grad_norm": 0.36770910024642944, + "learning_rate": 5.440385477364411e-05, + "loss": 0.4404, + "step": 33878 + }, + { + "epoch": 1.8928681174400088, + "grad_norm": 0.37572765350341797, + "learning_rate": 5.437584043030031e-05, + "loss": 0.3615, + "step": 33879 + }, + { + "epoch": 1.8929239880436908, + "grad_norm": 0.5905071496963501, + "learning_rate": 5.4347826086956524e-05, + "loss": 0.3998, + "step": 33880 + }, + { + "epoch": 1.8929798586473727, + "grad_norm": 0.40489789843559265, + "learning_rate": 5.431981174361273e-05, + "loss": 0.4275, + "step": 33881 + }, + { + "epoch": 1.8930357292510545, + "grad_norm": 0.43382519483566284, + "learning_rate": 5.429179740026894e-05, + "loss": 0.4882, + "step": 33882 + }, + { + "epoch": 1.8930915998547364, + "grad_norm": 1.1042331457138062, + "learning_rate": 5.426378305692515e-05, + "loss": 0.361, + "step": 33883 + }, + { + "epoch": 1.8931474704584184, + "grad_norm": 0.34555816650390625, + "learning_rate": 5.4235768713581354e-05, + "loss": 0.4813, + "step": 33884 + }, + { + "epoch": 1.8932033410621, + "grad_norm": 0.4400337338447571, + "learning_rate": 5.420775437023756e-05, + "loss": 0.3792, + "step": 33885 + }, + { + "epoch": 1.893259211665782, + "grad_norm": 0.4619087874889374, + "learning_rate": 5.417974002689377e-05, + "loss": 0.4491, + "step": 33886 + }, + { + "epoch": 1.893315082269464, + "grad_norm": 0.3227425813674927, + "learning_rate": 5.415172568354998e-05, + "loss": 0.3827, + "step": 33887 + }, + { + "epoch": 1.8933709528731457, + "grad_norm": 0.6534332633018494, + "learning_rate": 5.412371134020619e-05, + "loss": 0.4068, + "step": 33888 + }, + { + "epoch": 1.8934268234768277, + "grad_norm": 1.2490122318267822, + "learning_rate": 5.4095696996862396e-05, + "loss": 0.4299, + "step": 33889 + }, + { + "epoch": 1.8934826940805096, + "grad_norm": 0.48924222588539124, + "learning_rate": 5.40676826535186e-05, + "loss": 0.4313, + "step": 33890 + }, + { + "epoch": 1.8935385646841913, + "grad_norm": 0.5871970057487488, + "learning_rate": 5.403966831017481e-05, + "loss": 0.3251, + "step": 33891 + }, + { + "epoch": 1.8935944352878733, + "grad_norm": 0.38770028948783875, + "learning_rate": 5.401165396683102e-05, + "loss": 0.473, + "step": 33892 + }, + { + "epoch": 1.8936503058915553, + "grad_norm": 0.42346128821372986, + "learning_rate": 5.3983639623487226e-05, + "loss": 0.3738, + "step": 33893 + }, + { + "epoch": 1.893706176495237, + "grad_norm": 0.40875643491744995, + "learning_rate": 5.395562528014344e-05, + "loss": 0.4684, + "step": 33894 + }, + { + "epoch": 1.893762047098919, + "grad_norm": 0.43315473198890686, + "learning_rate": 5.392761093679964e-05, + "loss": 0.4161, + "step": 33895 + }, + { + "epoch": 1.8938179177026009, + "grad_norm": 2.14300274848938, + "learning_rate": 5.389959659345585e-05, + "loss": 0.4537, + "step": 33896 + }, + { + "epoch": 1.8938737883062826, + "grad_norm": 0.9617534279823303, + "learning_rate": 5.3871582250112056e-05, + "loss": 0.475, + "step": 33897 + }, + { + "epoch": 1.8939296589099646, + "grad_norm": 0.43197187781333923, + "learning_rate": 5.384356790676827e-05, + "loss": 0.3437, + "step": 33898 + }, + { + "epoch": 1.8939855295136465, + "grad_norm": 0.4029042720794678, + "learning_rate": 5.3815553563424475e-05, + "loss": 0.4358, + "step": 33899 + }, + { + "epoch": 1.8940414001173282, + "grad_norm": 0.374306321144104, + "learning_rate": 5.378753922008068e-05, + "loss": 0.442, + "step": 33900 + }, + { + "epoch": 1.8940972707210102, + "grad_norm": 0.34734171628952026, + "learning_rate": 5.3759524876736887e-05, + "loss": 0.331, + "step": 33901 + }, + { + "epoch": 1.8941531413246921, + "grad_norm": 0.40672171115875244, + "learning_rate": 5.37315105333931e-05, + "loss": 0.3809, + "step": 33902 + }, + { + "epoch": 1.8942090119283739, + "grad_norm": 0.4266183078289032, + "learning_rate": 5.3703496190049305e-05, + "loss": 0.499, + "step": 33903 + }, + { + "epoch": 1.8942648825320556, + "grad_norm": 0.3839362561702728, + "learning_rate": 5.367548184670552e-05, + "loss": 0.3728, + "step": 33904 + }, + { + "epoch": 1.8943207531357378, + "grad_norm": 0.9264284372329712, + "learning_rate": 5.364746750336172e-05, + "loss": 0.471, + "step": 33905 + }, + { + "epoch": 1.8943766237394195, + "grad_norm": 0.4661270081996918, + "learning_rate": 5.361945316001793e-05, + "loss": 0.4409, + "step": 33906 + }, + { + "epoch": 1.8944324943431012, + "grad_norm": 1.2585169076919556, + "learning_rate": 5.3591438816674135e-05, + "loss": 0.4049, + "step": 33907 + }, + { + "epoch": 1.8944883649467834, + "grad_norm": 1.0743927955627441, + "learning_rate": 5.356342447333035e-05, + "loss": 0.4571, + "step": 33908 + }, + { + "epoch": 1.8945442355504651, + "grad_norm": 0.5163407325744629, + "learning_rate": 5.353541012998655e-05, + "loss": 0.458, + "step": 33909 + }, + { + "epoch": 1.8946001061541469, + "grad_norm": 0.44419369101524353, + "learning_rate": 5.3507395786642766e-05, + "loss": 0.4163, + "step": 33910 + }, + { + "epoch": 1.8946559767578288, + "grad_norm": 0.429655522108078, + "learning_rate": 5.3479381443298965e-05, + "loss": 0.4059, + "step": 33911 + }, + { + "epoch": 1.8947118473615108, + "grad_norm": 0.4206882119178772, + "learning_rate": 5.345136709995518e-05, + "loss": 0.6854, + "step": 33912 + }, + { + "epoch": 1.8947677179651925, + "grad_norm": 0.5061456561088562, + "learning_rate": 5.3423352756611383e-05, + "loss": 0.3827, + "step": 33913 + }, + { + "epoch": 1.8948235885688745, + "grad_norm": 0.39502954483032227, + "learning_rate": 5.3395338413267596e-05, + "loss": 0.5245, + "step": 33914 + }, + { + "epoch": 1.8948794591725564, + "grad_norm": 0.480203777551651, + "learning_rate": 5.33673240699238e-05, + "loss": 0.5351, + "step": 33915 + }, + { + "epoch": 1.8949353297762381, + "grad_norm": 0.4162672758102417, + "learning_rate": 5.333930972658001e-05, + "loss": 0.35, + "step": 33916 + }, + { + "epoch": 1.89499120037992, + "grad_norm": 0.4918067753314972, + "learning_rate": 5.3311295383236213e-05, + "loss": 0.4519, + "step": 33917 + }, + { + "epoch": 1.895047070983602, + "grad_norm": 0.5703198313713074, + "learning_rate": 5.3283281039892426e-05, + "loss": 0.541, + "step": 33918 + }, + { + "epoch": 1.8951029415872838, + "grad_norm": 0.3752855658531189, + "learning_rate": 5.325526669654863e-05, + "loss": 0.3422, + "step": 33919 + }, + { + "epoch": 1.8951588121909657, + "grad_norm": 0.4065428674221039, + "learning_rate": 5.3227252353204844e-05, + "loss": 0.5398, + "step": 33920 + }, + { + "epoch": 1.8952146827946477, + "grad_norm": 0.4851628541946411, + "learning_rate": 5.3199238009861043e-05, + "loss": 0.4029, + "step": 33921 + }, + { + "epoch": 1.8952705533983294, + "grad_norm": 0.4001314342021942, + "learning_rate": 5.3171223666517256e-05, + "loss": 0.3566, + "step": 33922 + }, + { + "epoch": 1.8953264240020113, + "grad_norm": 0.7668794393539429, + "learning_rate": 5.314320932317346e-05, + "loss": 0.4006, + "step": 33923 + }, + { + "epoch": 1.8953822946056933, + "grad_norm": 0.43619388341903687, + "learning_rate": 5.3115194979829674e-05, + "loss": 0.4349, + "step": 33924 + }, + { + "epoch": 1.895438165209375, + "grad_norm": 0.4295825958251953, + "learning_rate": 5.308718063648588e-05, + "loss": 0.4847, + "step": 33925 + }, + { + "epoch": 1.895494035813057, + "grad_norm": 0.48599934577941895, + "learning_rate": 5.305916629314209e-05, + "loss": 0.4055, + "step": 33926 + }, + { + "epoch": 1.895549906416739, + "grad_norm": 0.43958723545074463, + "learning_rate": 5.303115194979829e-05, + "loss": 0.3867, + "step": 33927 + }, + { + "epoch": 1.8956057770204207, + "grad_norm": 0.46240290999412537, + "learning_rate": 5.3003137606454505e-05, + "loss": 0.4803, + "step": 33928 + }, + { + "epoch": 1.8956616476241026, + "grad_norm": 1.5011141300201416, + "learning_rate": 5.297512326311072e-05, + "loss": 0.5775, + "step": 33929 + }, + { + "epoch": 1.8957175182277846, + "grad_norm": 0.4046623408794403, + "learning_rate": 5.294710891976692e-05, + "loss": 0.4187, + "step": 33930 + }, + { + "epoch": 1.8957733888314663, + "grad_norm": 0.36999204754829407, + "learning_rate": 5.2919094576423136e-05, + "loss": 0.3601, + "step": 33931 + }, + { + "epoch": 1.8958292594351482, + "grad_norm": 0.4428631365299225, + "learning_rate": 5.2891080233079335e-05, + "loss": 0.4397, + "step": 33932 + }, + { + "epoch": 1.8958851300388302, + "grad_norm": 0.412064790725708, + "learning_rate": 5.286306588973555e-05, + "loss": 0.4647, + "step": 33933 + }, + { + "epoch": 1.895941000642512, + "grad_norm": 0.4849878251552582, + "learning_rate": 5.283505154639175e-05, + "loss": 0.3853, + "step": 33934 + }, + { + "epoch": 1.8959968712461936, + "grad_norm": 0.4581514000892639, + "learning_rate": 5.2807037203047966e-05, + "loss": 0.4844, + "step": 33935 + }, + { + "epoch": 1.8960527418498758, + "grad_norm": 0.3575570583343506, + "learning_rate": 5.277902285970417e-05, + "loss": 0.3619, + "step": 33936 + }, + { + "epoch": 1.8961086124535576, + "grad_norm": 0.43986615538597107, + "learning_rate": 5.275100851636038e-05, + "loss": 0.3875, + "step": 33937 + }, + { + "epoch": 1.8961644830572393, + "grad_norm": 0.44353893399238586, + "learning_rate": 5.272299417301658e-05, + "loss": 0.4517, + "step": 33938 + }, + { + "epoch": 1.8962203536609215, + "grad_norm": 0.5010743141174316, + "learning_rate": 5.2694979829672796e-05, + "loss": 0.3527, + "step": 33939 + }, + { + "epoch": 1.8962762242646032, + "grad_norm": 0.45234963297843933, + "learning_rate": 5.2666965486329e-05, + "loss": 0.3803, + "step": 33940 + }, + { + "epoch": 1.896332094868285, + "grad_norm": 0.41169029474258423, + "learning_rate": 5.2638951142985214e-05, + "loss": 0.3527, + "step": 33941 + }, + { + "epoch": 1.896387965471967, + "grad_norm": 0.519895613193512, + "learning_rate": 5.261093679964142e-05, + "loss": 0.4763, + "step": 33942 + }, + { + "epoch": 1.8964438360756488, + "grad_norm": 0.4630723297595978, + "learning_rate": 5.2582922456297626e-05, + "loss": 0.4277, + "step": 33943 + }, + { + "epoch": 1.8964997066793305, + "grad_norm": 1.0739799737930298, + "learning_rate": 5.255490811295383e-05, + "loss": 0.419, + "step": 33944 + }, + { + "epoch": 1.8965555772830125, + "grad_norm": 0.3538402020931244, + "learning_rate": 5.2526893769610044e-05, + "loss": 0.3682, + "step": 33945 + }, + { + "epoch": 1.8966114478866944, + "grad_norm": 2.612133026123047, + "learning_rate": 5.249887942626625e-05, + "loss": 0.4206, + "step": 33946 + }, + { + "epoch": 1.8966673184903762, + "grad_norm": 0.4383772909641266, + "learning_rate": 5.247086508292246e-05, + "loss": 0.6418, + "step": 33947 + }, + { + "epoch": 1.8967231890940581, + "grad_norm": 0.39638015627861023, + "learning_rate": 5.244285073957866e-05, + "loss": 0.3918, + "step": 33948 + }, + { + "epoch": 1.89677905969774, + "grad_norm": 0.546771228313446, + "learning_rate": 5.2414836396234874e-05, + "loss": 0.3598, + "step": 33949 + }, + { + "epoch": 1.8968349303014218, + "grad_norm": 0.3634411096572876, + "learning_rate": 5.238682205289108e-05, + "loss": 0.323, + "step": 33950 + }, + { + "epoch": 1.8968908009051038, + "grad_norm": 0.4510881006717682, + "learning_rate": 5.235880770954729e-05, + "loss": 0.4253, + "step": 33951 + }, + { + "epoch": 1.8969466715087857, + "grad_norm": 0.6558862328529358, + "learning_rate": 5.23307933662035e-05, + "loss": 0.3242, + "step": 33952 + }, + { + "epoch": 1.8970025421124674, + "grad_norm": 1.0728424787521362, + "learning_rate": 5.2302779022859704e-05, + "loss": 0.4407, + "step": 33953 + }, + { + "epoch": 1.8970584127161494, + "grad_norm": 1.55135178565979, + "learning_rate": 5.227476467951591e-05, + "loss": 0.3527, + "step": 33954 + }, + { + "epoch": 1.8971142833198313, + "grad_norm": 0.3927832245826721, + "learning_rate": 5.224675033617212e-05, + "loss": 0.3255, + "step": 33955 + }, + { + "epoch": 1.897170153923513, + "grad_norm": 0.664405345916748, + "learning_rate": 5.221873599282833e-05, + "loss": 0.3761, + "step": 33956 + }, + { + "epoch": 1.897226024527195, + "grad_norm": 0.36571159958839417, + "learning_rate": 5.219072164948454e-05, + "loss": 0.313, + "step": 33957 + }, + { + "epoch": 1.897281895130877, + "grad_norm": 0.3506898880004883, + "learning_rate": 5.216270730614075e-05, + "loss": 0.333, + "step": 33958 + }, + { + "epoch": 1.8973377657345587, + "grad_norm": 0.4369656443595886, + "learning_rate": 5.213469296279695e-05, + "loss": 0.4978, + "step": 33959 + }, + { + "epoch": 1.8973936363382407, + "grad_norm": 0.42377880215644836, + "learning_rate": 5.210667861945316e-05, + "loss": 0.5009, + "step": 33960 + }, + { + "epoch": 1.8974495069419226, + "grad_norm": 0.34021684527397156, + "learning_rate": 5.207866427610937e-05, + "loss": 0.3497, + "step": 33961 + }, + { + "epoch": 1.8975053775456043, + "grad_norm": 0.4345000386238098, + "learning_rate": 5.205064993276558e-05, + "loss": 0.553, + "step": 33962 + }, + { + "epoch": 1.8975612481492863, + "grad_norm": 0.9427254796028137, + "learning_rate": 5.202263558942179e-05, + "loss": 0.3824, + "step": 33963 + }, + { + "epoch": 1.8976171187529682, + "grad_norm": 0.4008789360523224, + "learning_rate": 5.199462124607799e-05, + "loss": 0.4183, + "step": 33964 + }, + { + "epoch": 1.89767298935665, + "grad_norm": 0.8906373381614685, + "learning_rate": 5.19666069027342e-05, + "loss": 0.4302, + "step": 33965 + }, + { + "epoch": 1.897728859960332, + "grad_norm": 0.3974537253379822, + "learning_rate": 5.193859255939041e-05, + "loss": 0.402, + "step": 33966 + }, + { + "epoch": 1.8977847305640139, + "grad_norm": 3.19034743309021, + "learning_rate": 5.191057821604662e-05, + "loss": 0.3072, + "step": 33967 + }, + { + "epoch": 1.8978406011676956, + "grad_norm": 0.32099857926368713, + "learning_rate": 5.1882563872702825e-05, + "loss": 0.4316, + "step": 33968 + }, + { + "epoch": 1.8978964717713773, + "grad_norm": 0.44090554118156433, + "learning_rate": 5.185454952935903e-05, + "loss": 0.3927, + "step": 33969 + }, + { + "epoch": 1.8979523423750595, + "grad_norm": 3.652703285217285, + "learning_rate": 5.182653518601524e-05, + "loss": 0.3692, + "step": 33970 + }, + { + "epoch": 1.8980082129787412, + "grad_norm": 0.3565467894077301, + "learning_rate": 5.179852084267145e-05, + "loss": 0.3899, + "step": 33971 + }, + { + "epoch": 1.898064083582423, + "grad_norm": 0.4030906856060028, + "learning_rate": 5.1770506499327655e-05, + "loss": 0.3206, + "step": 33972 + }, + { + "epoch": 1.8981199541861051, + "grad_norm": 1.3432997465133667, + "learning_rate": 5.174249215598387e-05, + "loss": 0.4457, + "step": 33973 + }, + { + "epoch": 1.8981758247897869, + "grad_norm": 1.9551444053649902, + "learning_rate": 5.1714477812640074e-05, + "loss": 0.5808, + "step": 33974 + }, + { + "epoch": 1.8982316953934686, + "grad_norm": 0.5350936651229858, + "learning_rate": 5.168646346929628e-05, + "loss": 0.4131, + "step": 33975 + }, + { + "epoch": 1.8982875659971508, + "grad_norm": 1.333747148513794, + "learning_rate": 5.1658449125952485e-05, + "loss": 0.4414, + "step": 33976 + }, + { + "epoch": 1.8983434366008325, + "grad_norm": 0.488978773355484, + "learning_rate": 5.16304347826087e-05, + "loss": 0.4243, + "step": 33977 + }, + { + "epoch": 1.8983993072045142, + "grad_norm": 0.41322648525238037, + "learning_rate": 5.1602420439264904e-05, + "loss": 0.4432, + "step": 33978 + }, + { + "epoch": 1.8984551778081962, + "grad_norm": 0.4517848789691925, + "learning_rate": 5.1574406095921116e-05, + "loss": 0.3556, + "step": 33979 + }, + { + "epoch": 1.8985110484118781, + "grad_norm": 0.376099556684494, + "learning_rate": 5.1546391752577315e-05, + "loss": 0.3692, + "step": 33980 + }, + { + "epoch": 1.8985669190155599, + "grad_norm": 0.48529860377311707, + "learning_rate": 5.151837740923353e-05, + "loss": 0.5597, + "step": 33981 + }, + { + "epoch": 1.8986227896192418, + "grad_norm": 0.461670845746994, + "learning_rate": 5.1490363065889734e-05, + "loss": 0.3175, + "step": 33982 + }, + { + "epoch": 1.8986786602229238, + "grad_norm": 0.373965322971344, + "learning_rate": 5.1462348722545946e-05, + "loss": 0.3831, + "step": 33983 + }, + { + "epoch": 1.8987345308266055, + "grad_norm": 0.38097772002220154, + "learning_rate": 5.143433437920215e-05, + "loss": 0.3961, + "step": 33984 + }, + { + "epoch": 1.8987904014302874, + "grad_norm": 0.41078799962997437, + "learning_rate": 5.140632003585836e-05, + "loss": 0.422, + "step": 33985 + }, + { + "epoch": 1.8988462720339694, + "grad_norm": 0.33785074949264526, + "learning_rate": 5.1378305692514564e-05, + "loss": 0.2979, + "step": 33986 + }, + { + "epoch": 1.8989021426376511, + "grad_norm": 0.30382588505744934, + "learning_rate": 5.1350291349170776e-05, + "loss": 0.2485, + "step": 33987 + }, + { + "epoch": 1.898958013241333, + "grad_norm": 0.35372069478034973, + "learning_rate": 5.132227700582698e-05, + "loss": 0.3555, + "step": 33988 + }, + { + "epoch": 1.899013883845015, + "grad_norm": 0.44411295652389526, + "learning_rate": 5.1294262662483195e-05, + "loss": 0.5392, + "step": 33989 + }, + { + "epoch": 1.8990697544486967, + "grad_norm": 0.33698222041130066, + "learning_rate": 5.12662483191394e-05, + "loss": 0.3135, + "step": 33990 + }, + { + "epoch": 1.8991256250523787, + "grad_norm": 0.35466861724853516, + "learning_rate": 5.1238233975795606e-05, + "loss": 0.3155, + "step": 33991 + }, + { + "epoch": 1.8991814956560606, + "grad_norm": 0.4246235489845276, + "learning_rate": 5.121021963245181e-05, + "loss": 0.5092, + "step": 33992 + }, + { + "epoch": 1.8992373662597424, + "grad_norm": 1.7543342113494873, + "learning_rate": 5.1182205289108025e-05, + "loss": 0.3945, + "step": 33993 + }, + { + "epoch": 1.8992932368634243, + "grad_norm": 1.1550387144088745, + "learning_rate": 5.115419094576424e-05, + "loss": 0.4137, + "step": 33994 + }, + { + "epoch": 1.8993491074671063, + "grad_norm": 0.36913084983825684, + "learning_rate": 5.112617660242044e-05, + "loss": 0.3655, + "step": 33995 + }, + { + "epoch": 1.899404978070788, + "grad_norm": 6.485686779022217, + "learning_rate": 5.109816225907665e-05, + "loss": 0.4091, + "step": 33996 + }, + { + "epoch": 1.89946084867447, + "grad_norm": 1.2026903629302979, + "learning_rate": 5.1070147915732855e-05, + "loss": 0.4251, + "step": 33997 + }, + { + "epoch": 1.899516719278152, + "grad_norm": 0.37188300490379333, + "learning_rate": 5.104213357238907e-05, + "loss": 0.3532, + "step": 33998 + }, + { + "epoch": 1.8995725898818336, + "grad_norm": 0.509874165058136, + "learning_rate": 5.101411922904527e-05, + "loss": 0.3117, + "step": 33999 + }, + { + "epoch": 1.8996284604855156, + "grad_norm": 0.5207553505897522, + "learning_rate": 5.0986104885701486e-05, + "loss": 0.4074, + "step": 34000 + }, + { + "epoch": 1.8996284604855156, + "eval_cer": 0.083210587760357, + "eval_loss": 0.3129920959472656, + "eval_runtime": 56.1245, + "eval_samples_per_second": 80.856, + "eval_steps_per_second": 5.06, + "eval_wer": 0.33177844194296247, + "step": 34000 + }, + { + "epoch": 1.8996843310891975, + "grad_norm": 2.7292873859405518, + "learning_rate": 5.0958090542357685e-05, + "loss": 0.4206, + "step": 34001 + }, + { + "epoch": 1.8997402016928793, + "grad_norm": 0.49933162331581116, + "learning_rate": 5.09300761990139e-05, + "loss": 0.4704, + "step": 34002 + }, + { + "epoch": 1.899796072296561, + "grad_norm": 0.5525579452514648, + "learning_rate": 5.09020618556701e-05, + "loss": 0.5463, + "step": 34003 + }, + { + "epoch": 1.8998519429002432, + "grad_norm": 0.40591976046562195, + "learning_rate": 5.0874047512326316e-05, + "loss": 0.4803, + "step": 34004 + }, + { + "epoch": 1.899907813503925, + "grad_norm": 0.4725314974784851, + "learning_rate": 5.084603316898252e-05, + "loss": 0.5837, + "step": 34005 + }, + { + "epoch": 1.8999636841076066, + "grad_norm": 0.6700323224067688, + "learning_rate": 5.081801882563873e-05, + "loss": 0.4073, + "step": 34006 + }, + { + "epoch": 1.9000195547112888, + "grad_norm": 2.233515977859497, + "learning_rate": 5.079000448229493e-05, + "loss": 0.4622, + "step": 34007 + }, + { + "epoch": 1.9000754253149705, + "grad_norm": 1.1157536506652832, + "learning_rate": 5.0761990138951146e-05, + "loss": 0.4271, + "step": 34008 + }, + { + "epoch": 1.9001312959186523, + "grad_norm": 0.830605149269104, + "learning_rate": 5.073397579560735e-05, + "loss": 0.3518, + "step": 34009 + }, + { + "epoch": 1.9001871665223344, + "grad_norm": 0.36976784467697144, + "learning_rate": 5.0705961452263564e-05, + "loss": 0.4463, + "step": 34010 + }, + { + "epoch": 1.9002430371260162, + "grad_norm": 0.5936668515205383, + "learning_rate": 5.067794710891977e-05, + "loss": 0.3865, + "step": 34011 + }, + { + "epoch": 1.900298907729698, + "grad_norm": 0.5157992839813232, + "learning_rate": 5.0649932765575976e-05, + "loss": 0.435, + "step": 34012 + }, + { + "epoch": 1.9003547783333798, + "grad_norm": 0.36869940161705017, + "learning_rate": 5.062191842223218e-05, + "loss": 0.4481, + "step": 34013 + }, + { + "epoch": 1.9004106489370618, + "grad_norm": 0.43391117453575134, + "learning_rate": 5.0593904078888394e-05, + "loss": 0.3706, + "step": 34014 + }, + { + "epoch": 1.9004665195407435, + "grad_norm": 0.36533740162849426, + "learning_rate": 5.05658897355446e-05, + "loss": 0.3422, + "step": 34015 + }, + { + "epoch": 1.9005223901444255, + "grad_norm": 4.253903388977051, + "learning_rate": 5.053787539220081e-05, + "loss": 0.398, + "step": 34016 + }, + { + "epoch": 1.9005782607481074, + "grad_norm": 0.3452115058898926, + "learning_rate": 5.050986104885701e-05, + "loss": 0.3195, + "step": 34017 + }, + { + "epoch": 1.9006341313517892, + "grad_norm": 0.37108203768730164, + "learning_rate": 5.0481846705513224e-05, + "loss": 0.3689, + "step": 34018 + }, + { + "epoch": 1.900690001955471, + "grad_norm": 0.40776297450065613, + "learning_rate": 5.045383236216943e-05, + "loss": 0.4123, + "step": 34019 + }, + { + "epoch": 1.900745872559153, + "grad_norm": 0.8415020704269409, + "learning_rate": 5.042581801882564e-05, + "loss": 0.3813, + "step": 34020 + }, + { + "epoch": 1.9008017431628348, + "grad_norm": 0.2957282066345215, + "learning_rate": 5.039780367548185e-05, + "loss": 0.4148, + "step": 34021 + }, + { + "epoch": 1.9008576137665167, + "grad_norm": 2.5193517208099365, + "learning_rate": 5.0369789332138054e-05, + "loss": 0.3417, + "step": 34022 + }, + { + "epoch": 1.9009134843701987, + "grad_norm": 0.42753326892852783, + "learning_rate": 5.034177498879426e-05, + "loss": 0.4736, + "step": 34023 + }, + { + "epoch": 1.9009693549738804, + "grad_norm": 0.9638602137565613, + "learning_rate": 5.031376064545047e-05, + "loss": 0.4628, + "step": 34024 + }, + { + "epoch": 1.9010252255775624, + "grad_norm": 0.41126373410224915, + "learning_rate": 5.028574630210668e-05, + "loss": 0.4997, + "step": 34025 + }, + { + "epoch": 1.9010810961812443, + "grad_norm": 0.3055334687232971, + "learning_rate": 5.025773195876289e-05, + "loss": 0.4208, + "step": 34026 + }, + { + "epoch": 1.901136966784926, + "grad_norm": 0.6064284443855286, + "learning_rate": 5.02297176154191e-05, + "loss": 0.3947, + "step": 34027 + }, + { + "epoch": 1.901192837388608, + "grad_norm": 1.694380283355713, + "learning_rate": 5.02017032720753e-05, + "loss": 0.4689, + "step": 34028 + }, + { + "epoch": 1.90124870799229, + "grad_norm": 0.6599203944206238, + "learning_rate": 5.017368892873151e-05, + "loss": 0.3989, + "step": 34029 + }, + { + "epoch": 1.9013045785959717, + "grad_norm": 0.49025705456733704, + "learning_rate": 5.014567458538772e-05, + "loss": 0.5199, + "step": 34030 + }, + { + "epoch": 1.9013604491996536, + "grad_norm": 0.33674222230911255, + "learning_rate": 5.011766024204393e-05, + "loss": 0.4101, + "step": 34031 + }, + { + "epoch": 1.9014163198033356, + "grad_norm": 0.29282987117767334, + "learning_rate": 5.008964589870014e-05, + "loss": 0.3569, + "step": 34032 + }, + { + "epoch": 1.9014721904070173, + "grad_norm": 0.3578849136829376, + "learning_rate": 5.006163155535634e-05, + "loss": 0.3318, + "step": 34033 + }, + { + "epoch": 1.9015280610106993, + "grad_norm": 0.39547276496887207, + "learning_rate": 5.003361721201255e-05, + "loss": 0.3814, + "step": 34034 + }, + { + "epoch": 1.9015839316143812, + "grad_norm": 0.3101220428943634, + "learning_rate": 5.000560286866876e-05, + "loss": 0.4372, + "step": 34035 + }, + { + "epoch": 1.901639802218063, + "grad_norm": 0.5064871907234192, + "learning_rate": 4.997758852532497e-05, + "loss": 0.4242, + "step": 34036 + }, + { + "epoch": 1.9016956728217447, + "grad_norm": 0.6567855477333069, + "learning_rate": 4.9949574181981176e-05, + "loss": 0.3975, + "step": 34037 + }, + { + "epoch": 1.9017515434254268, + "grad_norm": 0.5114567279815674, + "learning_rate": 4.992155983863738e-05, + "loss": 0.3746, + "step": 34038 + }, + { + "epoch": 1.9018074140291086, + "grad_norm": 0.42445629835128784, + "learning_rate": 4.989354549529359e-05, + "loss": 0.3824, + "step": 34039 + }, + { + "epoch": 1.9018632846327903, + "grad_norm": 0.3565137982368469, + "learning_rate": 4.98655311519498e-05, + "loss": 0.449, + "step": 34040 + }, + { + "epoch": 1.9019191552364725, + "grad_norm": 0.5589709281921387, + "learning_rate": 4.9837516808606006e-05, + "loss": 0.4296, + "step": 34041 + }, + { + "epoch": 1.9019750258401542, + "grad_norm": 0.44063135981559753, + "learning_rate": 4.980950246526222e-05, + "loss": 0.3999, + "step": 34042 + }, + { + "epoch": 1.902030896443836, + "grad_norm": 0.5760193467140198, + "learning_rate": 4.9781488121918424e-05, + "loss": 0.6562, + "step": 34043 + }, + { + "epoch": 1.9020867670475181, + "grad_norm": 0.6273636221885681, + "learning_rate": 4.975347377857463e-05, + "loss": 0.4611, + "step": 34044 + }, + { + "epoch": 1.9021426376511998, + "grad_norm": 0.5334069728851318, + "learning_rate": 4.9725459435230836e-05, + "loss": 0.4904, + "step": 34045 + }, + { + "epoch": 1.9021985082548816, + "grad_norm": 0.3617416024208069, + "learning_rate": 4.969744509188705e-05, + "loss": 0.438, + "step": 34046 + }, + { + "epoch": 1.9022543788585635, + "grad_norm": 4.971535682678223, + "learning_rate": 4.9669430748543254e-05, + "loss": 0.3284, + "step": 34047 + }, + { + "epoch": 1.9023102494622455, + "grad_norm": 0.39464622735977173, + "learning_rate": 4.964141640519947e-05, + "loss": 0.5081, + "step": 34048 + }, + { + "epoch": 1.9023661200659272, + "grad_norm": 0.31598958373069763, + "learning_rate": 4.9613402061855666e-05, + "loss": 0.3656, + "step": 34049 + }, + { + "epoch": 1.9024219906696092, + "grad_norm": 0.6940006613731384, + "learning_rate": 4.958538771851188e-05, + "loss": 0.3929, + "step": 34050 + }, + { + "epoch": 1.902477861273291, + "grad_norm": 0.4437127113342285, + "learning_rate": 4.9557373375168084e-05, + "loss": 0.3524, + "step": 34051 + }, + { + "epoch": 1.9025337318769728, + "grad_norm": 0.425151526927948, + "learning_rate": 4.95293590318243e-05, + "loss": 0.4108, + "step": 34052 + }, + { + "epoch": 1.9025896024806548, + "grad_norm": 0.46305277943611145, + "learning_rate": 4.95013446884805e-05, + "loss": 0.4713, + "step": 34053 + }, + { + "epoch": 1.9026454730843367, + "grad_norm": 0.5790467262268066, + "learning_rate": 4.947333034513671e-05, + "loss": 0.3246, + "step": 34054 + }, + { + "epoch": 1.9027013436880185, + "grad_norm": 0.44017648696899414, + "learning_rate": 4.9445316001792914e-05, + "loss": 0.4704, + "step": 34055 + }, + { + "epoch": 1.9027572142917004, + "grad_norm": 0.5783828496932983, + "learning_rate": 4.941730165844913e-05, + "loss": 0.4498, + "step": 34056 + }, + { + "epoch": 1.9028130848953824, + "grad_norm": 0.4623919725418091, + "learning_rate": 4.938928731510533e-05, + "loss": 0.4579, + "step": 34057 + }, + { + "epoch": 1.902868955499064, + "grad_norm": 0.3924160301685333, + "learning_rate": 4.9361272971761545e-05, + "loss": 0.4241, + "step": 34058 + }, + { + "epoch": 1.902924826102746, + "grad_norm": 0.5952802300453186, + "learning_rate": 4.933325862841775e-05, + "loss": 0.4172, + "step": 34059 + }, + { + "epoch": 1.902980696706428, + "grad_norm": 0.4374030530452728, + "learning_rate": 4.930524428507396e-05, + "loss": 0.3714, + "step": 34060 + }, + { + "epoch": 1.9030365673101097, + "grad_norm": 0.35606035590171814, + "learning_rate": 4.927722994173016e-05, + "loss": 0.3991, + "step": 34061 + }, + { + "epoch": 1.9030924379137917, + "grad_norm": 0.5100721716880798, + "learning_rate": 4.9249215598386375e-05, + "loss": 0.3163, + "step": 34062 + }, + { + "epoch": 1.9031483085174736, + "grad_norm": 0.5956750512123108, + "learning_rate": 4.922120125504259e-05, + "loss": 0.4957, + "step": 34063 + }, + { + "epoch": 1.9032041791211554, + "grad_norm": 0.5695796012878418, + "learning_rate": 4.9193186911698794e-05, + "loss": 0.4167, + "step": 34064 + }, + { + "epoch": 1.9032600497248373, + "grad_norm": 0.6158276200294495, + "learning_rate": 4.9165172568355e-05, + "loss": 0.4046, + "step": 34065 + }, + { + "epoch": 1.9033159203285193, + "grad_norm": 0.4149850308895111, + "learning_rate": 4.9137158225011205e-05, + "loss": 0.483, + "step": 34066 + }, + { + "epoch": 1.903371790932201, + "grad_norm": 2.2227606773376465, + "learning_rate": 4.910914388166742e-05, + "loss": 0.3885, + "step": 34067 + }, + { + "epoch": 1.903427661535883, + "grad_norm": 1.015714168548584, + "learning_rate": 4.9081129538323624e-05, + "loss": 0.3438, + "step": 34068 + }, + { + "epoch": 1.903483532139565, + "grad_norm": 0.38126400113105774, + "learning_rate": 4.9053115194979836e-05, + "loss": 0.3362, + "step": 34069 + }, + { + "epoch": 1.9035394027432466, + "grad_norm": 0.6171360611915588, + "learning_rate": 4.9025100851636035e-05, + "loss": 0.5052, + "step": 34070 + }, + { + "epoch": 1.9035952733469284, + "grad_norm": 0.4608530104160309, + "learning_rate": 4.899708650829225e-05, + "loss": 0.428, + "step": 34071 + }, + { + "epoch": 1.9036511439506105, + "grad_norm": 0.4042859375476837, + "learning_rate": 4.8969072164948454e-05, + "loss": 0.4827, + "step": 34072 + }, + { + "epoch": 1.9037070145542923, + "grad_norm": 34.404109954833984, + "learning_rate": 4.8941057821604666e-05, + "loss": 0.4122, + "step": 34073 + }, + { + "epoch": 1.903762885157974, + "grad_norm": 0.3818959891796112, + "learning_rate": 4.891304347826087e-05, + "loss": 0.4383, + "step": 34074 + }, + { + "epoch": 1.9038187557616562, + "grad_norm": 2.0385022163391113, + "learning_rate": 4.8885029134917085e-05, + "loss": 0.4023, + "step": 34075 + }, + { + "epoch": 1.9038746263653379, + "grad_norm": 0.37040042877197266, + "learning_rate": 4.8857014791573284e-05, + "loss": 0.4386, + "step": 34076 + }, + { + "epoch": 1.9039304969690196, + "grad_norm": 0.45191672444343567, + "learning_rate": 4.8829000448229496e-05, + "loss": 0.566, + "step": 34077 + }, + { + "epoch": 1.9039863675727018, + "grad_norm": 0.37761086225509644, + "learning_rate": 4.88009861048857e-05, + "loss": 0.4695, + "step": 34078 + }, + { + "epoch": 1.9040422381763835, + "grad_norm": 2.309100866317749, + "learning_rate": 4.8772971761541915e-05, + "loss": 0.4309, + "step": 34079 + }, + { + "epoch": 1.9040981087800652, + "grad_norm": 0.5817417502403259, + "learning_rate": 4.874495741819812e-05, + "loss": 0.4001, + "step": 34080 + }, + { + "epoch": 1.9041539793837472, + "grad_norm": 0.42541366815567017, + "learning_rate": 4.8716943074854326e-05, + "loss": 0.495, + "step": 34081 + }, + { + "epoch": 1.9042098499874291, + "grad_norm": 0.3893144130706787, + "learning_rate": 4.868892873151053e-05, + "loss": 0.4256, + "step": 34082 + }, + { + "epoch": 1.9042657205911109, + "grad_norm": 0.3504246771335602, + "learning_rate": 4.8660914388166745e-05, + "loss": 0.4145, + "step": 34083 + }, + { + "epoch": 1.9043215911947928, + "grad_norm": 0.5716679096221924, + "learning_rate": 4.863290004482295e-05, + "loss": 0.3677, + "step": 34084 + }, + { + "epoch": 1.9043774617984748, + "grad_norm": 0.5060387849807739, + "learning_rate": 4.860488570147916e-05, + "loss": 0.5053, + "step": 34085 + }, + { + "epoch": 1.9044333324021565, + "grad_norm": 0.5497783422470093, + "learning_rate": 4.857687135813536e-05, + "loss": 0.5088, + "step": 34086 + }, + { + "epoch": 1.9044892030058385, + "grad_norm": 0.37139567732810974, + "learning_rate": 4.8548857014791575e-05, + "loss": 0.4925, + "step": 34087 + }, + { + "epoch": 1.9045450736095204, + "grad_norm": 0.4549939036369324, + "learning_rate": 4.852084267144778e-05, + "loss": 0.3614, + "step": 34088 + }, + { + "epoch": 1.9046009442132021, + "grad_norm": 0.5887765884399414, + "learning_rate": 4.849282832810399e-05, + "loss": 0.5092, + "step": 34089 + }, + { + "epoch": 1.904656814816884, + "grad_norm": 0.9912926554679871, + "learning_rate": 4.84648139847602e-05, + "loss": 0.3977, + "step": 34090 + }, + { + "epoch": 1.904712685420566, + "grad_norm": 2.591461181640625, + "learning_rate": 4.8436799641416405e-05, + "loss": 0.4241, + "step": 34091 + }, + { + "epoch": 1.9047685560242478, + "grad_norm": 0.6895063519477844, + "learning_rate": 4.840878529807261e-05, + "loss": 0.3693, + "step": 34092 + }, + { + "epoch": 1.9048244266279297, + "grad_norm": 0.3954276740550995, + "learning_rate": 4.838077095472882e-05, + "loss": 0.4112, + "step": 34093 + }, + { + "epoch": 1.9048802972316117, + "grad_norm": 0.6591778993606567, + "learning_rate": 4.835275661138503e-05, + "loss": 0.3831, + "step": 34094 + }, + { + "epoch": 1.9049361678352934, + "grad_norm": 0.4190099835395813, + "learning_rate": 4.832474226804124e-05, + "loss": 0.498, + "step": 34095 + }, + { + "epoch": 1.9049920384389754, + "grad_norm": 1.0280975103378296, + "learning_rate": 4.829672792469745e-05, + "loss": 0.4603, + "step": 34096 + }, + { + "epoch": 1.9050479090426573, + "grad_norm": 0.4987579882144928, + "learning_rate": 4.826871358135365e-05, + "loss": 0.3627, + "step": 34097 + }, + { + "epoch": 1.905103779646339, + "grad_norm": 0.46484261751174927, + "learning_rate": 4.824069923800986e-05, + "loss": 0.3764, + "step": 34098 + }, + { + "epoch": 1.905159650250021, + "grad_norm": 0.3454582393169403, + "learning_rate": 4.821268489466607e-05, + "loss": 0.3315, + "step": 34099 + }, + { + "epoch": 1.905215520853703, + "grad_norm": 0.4878591001033783, + "learning_rate": 4.818467055132228e-05, + "loss": 0.497, + "step": 34100 + }, + { + "epoch": 1.9052713914573847, + "grad_norm": 0.6225417852401733, + "learning_rate": 4.815665620797849e-05, + "loss": 0.4382, + "step": 34101 + }, + { + "epoch": 1.9053272620610666, + "grad_norm": 0.3205970525741577, + "learning_rate": 4.812864186463469e-05, + "loss": 0.404, + "step": 34102 + }, + { + "epoch": 1.9053831326647486, + "grad_norm": 0.5315133929252625, + "learning_rate": 4.81006275212909e-05, + "loss": 0.5319, + "step": 34103 + }, + { + "epoch": 1.9054390032684303, + "grad_norm": 0.3877382278442383, + "learning_rate": 4.807261317794711e-05, + "loss": 0.3003, + "step": 34104 + }, + { + "epoch": 1.905494873872112, + "grad_norm": 0.2979893386363983, + "learning_rate": 4.804459883460332e-05, + "loss": 0.383, + "step": 34105 + }, + { + "epoch": 1.9055507444757942, + "grad_norm": 0.32344850897789, + "learning_rate": 4.8016584491259526e-05, + "loss": 0.336, + "step": 34106 + }, + { + "epoch": 1.905606615079476, + "grad_norm": 0.5974989533424377, + "learning_rate": 4.798857014791573e-05, + "loss": 0.4674, + "step": 34107 + }, + { + "epoch": 1.9056624856831577, + "grad_norm": 0.48708823323249817, + "learning_rate": 4.796055580457194e-05, + "loss": 0.6252, + "step": 34108 + }, + { + "epoch": 1.9057183562868398, + "grad_norm": 0.34793776273727417, + "learning_rate": 4.793254146122815e-05, + "loss": 0.3385, + "step": 34109 + }, + { + "epoch": 1.9057742268905216, + "grad_norm": 0.4282264709472656, + "learning_rate": 4.7904527117884356e-05, + "loss": 0.3476, + "step": 34110 + }, + { + "epoch": 1.9058300974942033, + "grad_norm": 0.45278871059417725, + "learning_rate": 4.787651277454057e-05, + "loss": 0.3178, + "step": 34111 + }, + { + "epoch": 1.9058859680978855, + "grad_norm": 0.33419862389564514, + "learning_rate": 4.7848498431196774e-05, + "loss": 0.3795, + "step": 34112 + }, + { + "epoch": 1.9059418387015672, + "grad_norm": 0.4670722782611847, + "learning_rate": 4.782048408785298e-05, + "loss": 0.4358, + "step": 34113 + }, + { + "epoch": 1.905997709305249, + "grad_norm": 0.33954674005508423, + "learning_rate": 4.7792469744509186e-05, + "loss": 0.4281, + "step": 34114 + }, + { + "epoch": 1.9060535799089309, + "grad_norm": 0.3642888069152832, + "learning_rate": 4.77644554011654e-05, + "loss": 0.3971, + "step": 34115 + }, + { + "epoch": 1.9061094505126128, + "grad_norm": 0.32295727729797363, + "learning_rate": 4.7736441057821604e-05, + "loss": 0.4236, + "step": 34116 + }, + { + "epoch": 1.9061653211162946, + "grad_norm": 0.44074344635009766, + "learning_rate": 4.770842671447782e-05, + "loss": 0.4347, + "step": 34117 + }, + { + "epoch": 1.9062211917199765, + "grad_norm": 0.754938006401062, + "learning_rate": 4.7680412371134016e-05, + "loss": 0.4192, + "step": 34118 + }, + { + "epoch": 1.9062770623236585, + "grad_norm": 0.45145440101623535, + "learning_rate": 4.765239802779023e-05, + "loss": 0.3804, + "step": 34119 + }, + { + "epoch": 1.9063329329273402, + "grad_norm": 0.8313418030738831, + "learning_rate": 4.7624383684446434e-05, + "loss": 0.3334, + "step": 34120 + }, + { + "epoch": 1.9063888035310221, + "grad_norm": 0.33287373185157776, + "learning_rate": 4.759636934110265e-05, + "loss": 0.3458, + "step": 34121 + }, + { + "epoch": 1.906444674134704, + "grad_norm": 0.4577706456184387, + "learning_rate": 4.756835499775885e-05, + "loss": 0.5388, + "step": 34122 + }, + { + "epoch": 1.9065005447383858, + "grad_norm": 0.435867041349411, + "learning_rate": 4.754034065441506e-05, + "loss": 0.437, + "step": 34123 + }, + { + "epoch": 1.9065564153420678, + "grad_norm": 0.510905921459198, + "learning_rate": 4.7512326311071264e-05, + "loss": 0.3877, + "step": 34124 + }, + { + "epoch": 1.9066122859457497, + "grad_norm": 0.3301418423652649, + "learning_rate": 4.748431196772748e-05, + "loss": 0.3819, + "step": 34125 + }, + { + "epoch": 1.9066681565494314, + "grad_norm": 0.36155426502227783, + "learning_rate": 4.745629762438368e-05, + "loss": 0.4036, + "step": 34126 + }, + { + "epoch": 1.9067240271531134, + "grad_norm": 0.45283177495002747, + "learning_rate": 4.7428283281039895e-05, + "loss": 0.4008, + "step": 34127 + }, + { + "epoch": 1.9067798977567954, + "grad_norm": 1.1117850542068481, + "learning_rate": 4.74002689376961e-05, + "loss": 0.3091, + "step": 34128 + }, + { + "epoch": 1.906835768360477, + "grad_norm": 1.7444415092468262, + "learning_rate": 4.737225459435231e-05, + "loss": 0.4141, + "step": 34129 + }, + { + "epoch": 1.906891638964159, + "grad_norm": 0.4117007553577423, + "learning_rate": 4.734424025100851e-05, + "loss": 0.3271, + "step": 34130 + }, + { + "epoch": 1.906947509567841, + "grad_norm": 0.30958592891693115, + "learning_rate": 4.7316225907664726e-05, + "loss": 0.3595, + "step": 34131 + }, + { + "epoch": 1.9070033801715227, + "grad_norm": 0.3626241683959961, + "learning_rate": 4.728821156432094e-05, + "loss": 0.4251, + "step": 34132 + }, + { + "epoch": 1.9070592507752047, + "grad_norm": 0.3504946529865265, + "learning_rate": 4.7260197220977144e-05, + "loss": 0.3338, + "step": 34133 + }, + { + "epoch": 1.9071151213788866, + "grad_norm": 0.6255998611450195, + "learning_rate": 4.723218287763335e-05, + "loss": 0.3764, + "step": 34134 + }, + { + "epoch": 1.9071709919825683, + "grad_norm": 1.5956764221191406, + "learning_rate": 4.7204168534289556e-05, + "loss": 0.3567, + "step": 34135 + }, + { + "epoch": 1.9072268625862503, + "grad_norm": 0.43054595589637756, + "learning_rate": 4.717615419094577e-05, + "loss": 0.5611, + "step": 34136 + }, + { + "epoch": 1.9072827331899322, + "grad_norm": 0.4861234426498413, + "learning_rate": 4.7148139847601974e-05, + "loss": 0.4678, + "step": 34137 + }, + { + "epoch": 1.907338603793614, + "grad_norm": 0.5173541903495789, + "learning_rate": 4.7120125504258187e-05, + "loss": 0.3969, + "step": 34138 + }, + { + "epoch": 1.9073944743972957, + "grad_norm": 0.3904843330383301, + "learning_rate": 4.7092111160914386e-05, + "loss": 0.5213, + "step": 34139 + }, + { + "epoch": 1.9074503450009779, + "grad_norm": 0.5169283151626587, + "learning_rate": 4.70640968175706e-05, + "loss": 0.3681, + "step": 34140 + }, + { + "epoch": 1.9075062156046596, + "grad_norm": 0.38323739171028137, + "learning_rate": 4.7036082474226804e-05, + "loss": 0.4322, + "step": 34141 + }, + { + "epoch": 1.9075620862083413, + "grad_norm": 1.0091832876205444, + "learning_rate": 4.7008068130883017e-05, + "loss": 0.4264, + "step": 34142 + }, + { + "epoch": 1.9076179568120235, + "grad_norm": 0.4086916446685791, + "learning_rate": 4.698005378753922e-05, + "loss": 0.4851, + "step": 34143 + }, + { + "epoch": 1.9076738274157052, + "grad_norm": 0.34409570693969727, + "learning_rate": 4.6952039444195435e-05, + "loss": 0.3778, + "step": 34144 + }, + { + "epoch": 1.907729698019387, + "grad_norm": 0.4133690595626831, + "learning_rate": 4.6924025100851634e-05, + "loss": 0.2689, + "step": 34145 + }, + { + "epoch": 1.907785568623069, + "grad_norm": 0.52272629737854, + "learning_rate": 4.689601075750785e-05, + "loss": 0.4062, + "step": 34146 + }, + { + "epoch": 1.9078414392267509, + "grad_norm": 0.4555348753929138, + "learning_rate": 4.686799641416405e-05, + "loss": 0.4674, + "step": 34147 + }, + { + "epoch": 1.9078973098304326, + "grad_norm": 0.83127361536026, + "learning_rate": 4.6839982070820265e-05, + "loss": 0.3605, + "step": 34148 + }, + { + "epoch": 1.9079531804341145, + "grad_norm": 0.37815114855766296, + "learning_rate": 4.681196772747647e-05, + "loss": 0.3971, + "step": 34149 + }, + { + "epoch": 1.9080090510377965, + "grad_norm": 0.580857515335083, + "learning_rate": 4.678395338413268e-05, + "loss": 0.3789, + "step": 34150 + }, + { + "epoch": 1.9080649216414782, + "grad_norm": 0.45660144090652466, + "learning_rate": 4.675593904078888e-05, + "loss": 0.5042, + "step": 34151 + }, + { + "epoch": 1.9081207922451602, + "grad_norm": 0.8300501108169556, + "learning_rate": 4.6727924697445095e-05, + "loss": 0.4057, + "step": 34152 + }, + { + "epoch": 1.9081766628488421, + "grad_norm": 3.6818599700927734, + "learning_rate": 4.66999103541013e-05, + "loss": 0.4067, + "step": 34153 + }, + { + "epoch": 1.9082325334525239, + "grad_norm": 0.7902851700782776, + "learning_rate": 4.6671896010757513e-05, + "loss": 0.4575, + "step": 34154 + }, + { + "epoch": 1.9082884040562058, + "grad_norm": 0.5159029364585876, + "learning_rate": 4.664388166741371e-05, + "loss": 0.5867, + "step": 34155 + }, + { + "epoch": 1.9083442746598878, + "grad_norm": 0.391287237405777, + "learning_rate": 4.6615867324069925e-05, + "loss": 0.3881, + "step": 34156 + }, + { + "epoch": 1.9084001452635695, + "grad_norm": 2.649580240249634, + "learning_rate": 4.658785298072613e-05, + "loss": 0.4401, + "step": 34157 + }, + { + "epoch": 1.9084560158672514, + "grad_norm": 0.4726184606552124, + "learning_rate": 4.6559838637382344e-05, + "loss": 0.2928, + "step": 34158 + }, + { + "epoch": 1.9085118864709334, + "grad_norm": 0.8754472136497498, + "learning_rate": 4.653182429403855e-05, + "loss": 0.4567, + "step": 34159 + }, + { + "epoch": 1.9085677570746151, + "grad_norm": 0.43266844749450684, + "learning_rate": 4.650380995069476e-05, + "loss": 0.316, + "step": 34160 + }, + { + "epoch": 1.908623627678297, + "grad_norm": 0.45950427651405334, + "learning_rate": 4.647579560735096e-05, + "loss": 0.3729, + "step": 34161 + }, + { + "epoch": 1.908679498281979, + "grad_norm": 0.3796997666358948, + "learning_rate": 4.6447781264007174e-05, + "loss": 0.4499, + "step": 34162 + }, + { + "epoch": 1.9087353688856608, + "grad_norm": 0.4120330512523651, + "learning_rate": 4.641976692066338e-05, + "loss": 0.4242, + "step": 34163 + }, + { + "epoch": 1.9087912394893427, + "grad_norm": 0.38940051198005676, + "learning_rate": 4.639175257731959e-05, + "loss": 0.5157, + "step": 34164 + }, + { + "epoch": 1.9088471100930247, + "grad_norm": 1.4334467649459839, + "learning_rate": 4.63637382339758e-05, + "loss": 0.4678, + "step": 34165 + }, + { + "epoch": 1.9089029806967064, + "grad_norm": 0.34707701206207275, + "learning_rate": 4.6335723890632004e-05, + "loss": 0.3586, + "step": 34166 + }, + { + "epoch": 1.9089588513003883, + "grad_norm": 0.3333244025707245, + "learning_rate": 4.630770954728821e-05, + "loss": 0.3416, + "step": 34167 + }, + { + "epoch": 1.9090147219040703, + "grad_norm": 0.4891131520271301, + "learning_rate": 4.627969520394442e-05, + "loss": 0.4201, + "step": 34168 + }, + { + "epoch": 1.909070592507752, + "grad_norm": 0.560443639755249, + "learning_rate": 4.625168086060063e-05, + "loss": 0.4923, + "step": 34169 + }, + { + "epoch": 1.909126463111434, + "grad_norm": 0.4710027575492859, + "learning_rate": 4.622366651725684e-05, + "loss": 0.3573, + "step": 34170 + }, + { + "epoch": 1.909182333715116, + "grad_norm": 0.5493643879890442, + "learning_rate": 4.619565217391304e-05, + "loss": 0.4501, + "step": 34171 + }, + { + "epoch": 1.9092382043187976, + "grad_norm": 0.3460245430469513, + "learning_rate": 4.616763783056925e-05, + "loss": 0.4074, + "step": 34172 + }, + { + "epoch": 1.9092940749224794, + "grad_norm": 0.323703795671463, + "learning_rate": 4.613962348722546e-05, + "loss": 0.355, + "step": 34173 + }, + { + "epoch": 1.9093499455261616, + "grad_norm": 1.084197759628296, + "learning_rate": 4.611160914388167e-05, + "loss": 0.4407, + "step": 34174 + }, + { + "epoch": 1.9094058161298433, + "grad_norm": 0.32964572310447693, + "learning_rate": 4.6083594800537876e-05, + "loss": 0.3179, + "step": 34175 + }, + { + "epoch": 1.909461686733525, + "grad_norm": 0.42137330770492554, + "learning_rate": 4.605558045719408e-05, + "loss": 0.448, + "step": 34176 + }, + { + "epoch": 1.9095175573372072, + "grad_norm": 0.8118842244148254, + "learning_rate": 4.602756611385029e-05, + "loss": 0.4607, + "step": 34177 + }, + { + "epoch": 1.909573427940889, + "grad_norm": 1.9654608964920044, + "learning_rate": 4.59995517705065e-05, + "loss": 0.4247, + "step": 34178 + }, + { + "epoch": 1.9096292985445706, + "grad_norm": 0.5289742350578308, + "learning_rate": 4.5971537427162706e-05, + "loss": 0.314, + "step": 34179 + }, + { + "epoch": 1.9096851691482526, + "grad_norm": 0.5790948271751404, + "learning_rate": 4.594352308381892e-05, + "loss": 0.4258, + "step": 34180 + }, + { + "epoch": 1.9097410397519345, + "grad_norm": 0.6918966770172119, + "learning_rate": 4.5915508740475125e-05, + "loss": 0.3225, + "step": 34181 + }, + { + "epoch": 1.9097969103556163, + "grad_norm": 0.4916957914829254, + "learning_rate": 4.588749439713133e-05, + "loss": 0.3526, + "step": 34182 + }, + { + "epoch": 1.9098527809592982, + "grad_norm": 0.36339378356933594, + "learning_rate": 4.5859480053787536e-05, + "loss": 0.3435, + "step": 34183 + }, + { + "epoch": 1.9099086515629802, + "grad_norm": 0.3836834728717804, + "learning_rate": 4.583146571044375e-05, + "loss": 0.404, + "step": 34184 + }, + { + "epoch": 1.909964522166662, + "grad_norm": 0.579321563243866, + "learning_rate": 4.5803451367099955e-05, + "loss": 0.4808, + "step": 34185 + }, + { + "epoch": 1.9100203927703439, + "grad_norm": 4.702267646789551, + "learning_rate": 4.577543702375617e-05, + "loss": 0.4531, + "step": 34186 + }, + { + "epoch": 1.9100762633740258, + "grad_norm": 0.41328364610671997, + "learning_rate": 4.5747422680412366e-05, + "loss": 0.3642, + "step": 34187 + }, + { + "epoch": 1.9101321339777075, + "grad_norm": 2.155182361602783, + "learning_rate": 4.571940833706858e-05, + "loss": 0.3632, + "step": 34188 + }, + { + "epoch": 1.9101880045813895, + "grad_norm": 0.3851478397846222, + "learning_rate": 4.5691393993724785e-05, + "loss": 0.3689, + "step": 34189 + }, + { + "epoch": 1.9102438751850714, + "grad_norm": 0.41954973340034485, + "learning_rate": 4.5663379650381e-05, + "loss": 0.3755, + "step": 34190 + }, + { + "epoch": 1.9102997457887532, + "grad_norm": 0.55866539478302, + "learning_rate": 4.56353653070372e-05, + "loss": 0.4759, + "step": 34191 + }, + { + "epoch": 1.9103556163924351, + "grad_norm": 0.4929575026035309, + "learning_rate": 4.560735096369341e-05, + "loss": 0.4101, + "step": 34192 + }, + { + "epoch": 1.910411486996117, + "grad_norm": 0.4727829098701477, + "learning_rate": 4.5579336620349615e-05, + "loss": 0.4238, + "step": 34193 + }, + { + "epoch": 1.9104673575997988, + "grad_norm": 0.3692944049835205, + "learning_rate": 4.555132227700583e-05, + "loss": 0.3408, + "step": 34194 + }, + { + "epoch": 1.9105232282034808, + "grad_norm": 0.46246951818466187, + "learning_rate": 4.552330793366203e-05, + "loss": 0.4118, + "step": 34195 + }, + { + "epoch": 1.9105790988071627, + "grad_norm": 0.6007760167121887, + "learning_rate": 4.5495293590318246e-05, + "loss": 0.3258, + "step": 34196 + }, + { + "epoch": 1.9106349694108444, + "grad_norm": 0.3778511881828308, + "learning_rate": 4.546727924697446e-05, + "loss": 0.3729, + "step": 34197 + }, + { + "epoch": 1.9106908400145264, + "grad_norm": 0.4103360176086426, + "learning_rate": 4.543926490363066e-05, + "loss": 0.5797, + "step": 34198 + }, + { + "epoch": 1.9107467106182083, + "grad_norm": 0.34780654311180115, + "learning_rate": 4.541125056028687e-05, + "loss": 0.4029, + "step": 34199 + }, + { + "epoch": 1.91080258122189, + "grad_norm": 1.3025646209716797, + "learning_rate": 4.5383236216943076e-05, + "loss": 0.4111, + "step": 34200 + }, + { + "epoch": 1.910858451825572, + "grad_norm": 0.4106946885585785, + "learning_rate": 4.535522187359929e-05, + "loss": 0.2977, + "step": 34201 + }, + { + "epoch": 1.910914322429254, + "grad_norm": 1.2017337083816528, + "learning_rate": 4.5327207530255494e-05, + "loss": 0.5423, + "step": 34202 + }, + { + "epoch": 1.9109701930329357, + "grad_norm": 0.6356443762779236, + "learning_rate": 4.52991931869117e-05, + "loss": 0.5044, + "step": 34203 + }, + { + "epoch": 1.9110260636366176, + "grad_norm": 0.4082108438014984, + "learning_rate": 4.5271178843567906e-05, + "loss": 0.3747, + "step": 34204 + }, + { + "epoch": 1.9110819342402996, + "grad_norm": 1.3693883419036865, + "learning_rate": 4.524316450022412e-05, + "loss": 0.3677, + "step": 34205 + }, + { + "epoch": 1.9111378048439813, + "grad_norm": 0.9863170981407166, + "learning_rate": 4.5215150156880324e-05, + "loss": 0.4307, + "step": 34206 + }, + { + "epoch": 1.911193675447663, + "grad_norm": 1.4017888307571411, + "learning_rate": 4.518713581353654e-05, + "loss": 0.5134, + "step": 34207 + }, + { + "epoch": 1.9112495460513452, + "grad_norm": 0.35614311695098877, + "learning_rate": 4.5159121470192736e-05, + "loss": 0.3864, + "step": 34208 + }, + { + "epoch": 1.911305416655027, + "grad_norm": 0.47634613513946533, + "learning_rate": 4.513110712684895e-05, + "loss": 0.5463, + "step": 34209 + }, + { + "epoch": 1.9113612872587087, + "grad_norm": 0.6449847221374512, + "learning_rate": 4.5103092783505154e-05, + "loss": 0.4301, + "step": 34210 + }, + { + "epoch": 1.9114171578623909, + "grad_norm": 1.1317805051803589, + "learning_rate": 4.507507844016137e-05, + "loss": 0.4311, + "step": 34211 + }, + { + "epoch": 1.9114730284660726, + "grad_norm": 1.404512643814087, + "learning_rate": 4.504706409681757e-05, + "loss": 0.3568, + "step": 34212 + }, + { + "epoch": 1.9115288990697543, + "grad_norm": 0.3884095549583435, + "learning_rate": 4.5019049753473785e-05, + "loss": 0.3626, + "step": 34213 + }, + { + "epoch": 1.9115847696734363, + "grad_norm": 0.3802717626094818, + "learning_rate": 4.4991035410129984e-05, + "loss": 0.3907, + "step": 34214 + }, + { + "epoch": 1.9116406402771182, + "grad_norm": 0.497441828250885, + "learning_rate": 4.49630210667862e-05, + "loss": 0.4265, + "step": 34215 + }, + { + "epoch": 1.9116965108808, + "grad_norm": 0.6393193006515503, + "learning_rate": 4.49350067234424e-05, + "loss": 0.5676, + "step": 34216 + }, + { + "epoch": 1.911752381484482, + "grad_norm": 0.34649473428726196, + "learning_rate": 4.4906992380098615e-05, + "loss": 0.4576, + "step": 34217 + }, + { + "epoch": 1.9118082520881639, + "grad_norm": 0.48334813117980957, + "learning_rate": 4.487897803675482e-05, + "loss": 0.4458, + "step": 34218 + }, + { + "epoch": 1.9118641226918456, + "grad_norm": 0.8660749793052673, + "learning_rate": 4.485096369341103e-05, + "loss": 0.5483, + "step": 34219 + }, + { + "epoch": 1.9119199932955275, + "grad_norm": 0.3475062847137451, + "learning_rate": 4.482294935006723e-05, + "loss": 0.2957, + "step": 34220 + }, + { + "epoch": 1.9119758638992095, + "grad_norm": 0.2971285283565521, + "learning_rate": 4.4794935006723445e-05, + "loss": 0.3365, + "step": 34221 + }, + { + "epoch": 1.9120317345028912, + "grad_norm": 0.42710819840431213, + "learning_rate": 4.476692066337965e-05, + "loss": 0.4029, + "step": 34222 + }, + { + "epoch": 1.9120876051065732, + "grad_norm": 0.45588359236717224, + "learning_rate": 4.4738906320035864e-05, + "loss": 0.4257, + "step": 34223 + }, + { + "epoch": 1.9121434757102551, + "grad_norm": 0.8549709320068359, + "learning_rate": 4.471089197669206e-05, + "loss": 0.3288, + "step": 34224 + }, + { + "epoch": 1.9121993463139368, + "grad_norm": 0.36473792791366577, + "learning_rate": 4.4682877633348275e-05, + "loss": 0.4024, + "step": 34225 + }, + { + "epoch": 1.9122552169176188, + "grad_norm": 0.5666016340255737, + "learning_rate": 4.465486329000448e-05, + "loss": 0.3698, + "step": 34226 + }, + { + "epoch": 1.9123110875213007, + "grad_norm": 0.3124704360961914, + "learning_rate": 4.4626848946660694e-05, + "loss": 0.3804, + "step": 34227 + }, + { + "epoch": 1.9123669581249825, + "grad_norm": 0.3888112008571625, + "learning_rate": 4.45988346033169e-05, + "loss": 0.3888, + "step": 34228 + }, + { + "epoch": 1.9124228287286644, + "grad_norm": 0.6857258081436157, + "learning_rate": 4.457082025997311e-05, + "loss": 0.3316, + "step": 34229 + }, + { + "epoch": 1.9124786993323464, + "grad_norm": 0.37537816166877747, + "learning_rate": 4.454280591662931e-05, + "loss": 0.3698, + "step": 34230 + }, + { + "epoch": 1.912534569936028, + "grad_norm": 0.493400514125824, + "learning_rate": 4.4514791573285524e-05, + "loss": 0.3868, + "step": 34231 + }, + { + "epoch": 1.91259044053971, + "grad_norm": 1.728888750076294, + "learning_rate": 4.448677722994173e-05, + "loss": 0.3897, + "step": 34232 + }, + { + "epoch": 1.912646311143392, + "grad_norm": 0.43773993849754333, + "learning_rate": 4.445876288659794e-05, + "loss": 0.479, + "step": 34233 + }, + { + "epoch": 1.9127021817470737, + "grad_norm": 0.4888748824596405, + "learning_rate": 4.443074854325415e-05, + "loss": 0.3649, + "step": 34234 + }, + { + "epoch": 1.9127580523507557, + "grad_norm": 0.6154592037200928, + "learning_rate": 4.4402734199910354e-05, + "loss": 0.4338, + "step": 34235 + }, + { + "epoch": 1.9128139229544376, + "grad_norm": 0.968364417552948, + "learning_rate": 4.437471985656656e-05, + "loss": 0.4369, + "step": 34236 + }, + { + "epoch": 1.9128697935581194, + "grad_norm": 0.4669293761253357, + "learning_rate": 4.434670551322277e-05, + "loss": 0.4745, + "step": 34237 + }, + { + "epoch": 1.9129256641618013, + "grad_norm": 1.0751398801803589, + "learning_rate": 4.431869116987898e-05, + "loss": 0.4715, + "step": 34238 + }, + { + "epoch": 1.9129815347654833, + "grad_norm": 0.4402364492416382, + "learning_rate": 4.429067682653519e-05, + "loss": 0.437, + "step": 34239 + }, + { + "epoch": 1.913037405369165, + "grad_norm": 0.2822793424129486, + "learning_rate": 4.426266248319139e-05, + "loss": 0.307, + "step": 34240 + }, + { + "epoch": 1.9130932759728467, + "grad_norm": 0.6765398979187012, + "learning_rate": 4.42346481398476e-05, + "loss": 0.2775, + "step": 34241 + }, + { + "epoch": 1.913149146576529, + "grad_norm": 0.5679739713668823, + "learning_rate": 4.420663379650381e-05, + "loss": 0.3363, + "step": 34242 + }, + { + "epoch": 1.9132050171802106, + "grad_norm": 0.42868176102638245, + "learning_rate": 4.417861945316002e-05, + "loss": 0.4431, + "step": 34243 + }, + { + "epoch": 1.9132608877838924, + "grad_norm": 3.0035955905914307, + "learning_rate": 4.4150605109816227e-05, + "loss": 0.4535, + "step": 34244 + }, + { + "epoch": 1.9133167583875745, + "grad_norm": 0.8478909134864807, + "learning_rate": 4.412259076647244e-05, + "loss": 0.3559, + "step": 34245 + }, + { + "epoch": 1.9133726289912563, + "grad_norm": 0.44962218403816223, + "learning_rate": 4.409457642312864e-05, + "loss": 0.395, + "step": 34246 + }, + { + "epoch": 1.913428499594938, + "grad_norm": 2.0915424823760986, + "learning_rate": 4.406656207978485e-05, + "loss": 0.4332, + "step": 34247 + }, + { + "epoch": 1.91348437019862, + "grad_norm": 0.43334779143333435, + "learning_rate": 4.403854773644106e-05, + "loss": 0.4522, + "step": 34248 + }, + { + "epoch": 1.913540240802302, + "grad_norm": 0.382476806640625, + "learning_rate": 4.401053339309727e-05, + "loss": 0.4319, + "step": 34249 + }, + { + "epoch": 1.9135961114059836, + "grad_norm": 2.289149284362793, + "learning_rate": 4.3982519049753475e-05, + "loss": 0.3513, + "step": 34250 + }, + { + "epoch": 1.9136519820096656, + "grad_norm": 0.46357378363609314, + "learning_rate": 4.395450470640968e-05, + "loss": 0.435, + "step": 34251 + }, + { + "epoch": 1.9137078526133475, + "grad_norm": 0.7242608666419983, + "learning_rate": 4.392649036306589e-05, + "loss": 0.2943, + "step": 34252 + }, + { + "epoch": 1.9137637232170293, + "grad_norm": 0.6586125493049622, + "learning_rate": 4.38984760197221e-05, + "loss": 0.4084, + "step": 34253 + }, + { + "epoch": 1.9138195938207112, + "grad_norm": 0.9369167685508728, + "learning_rate": 4.3870461676378305e-05, + "loss": 0.5086, + "step": 34254 + }, + { + "epoch": 1.9138754644243932, + "grad_norm": 2.533966302871704, + "learning_rate": 4.384244733303452e-05, + "loss": 0.2581, + "step": 34255 + }, + { + "epoch": 1.9139313350280749, + "grad_norm": 0.39661240577697754, + "learning_rate": 4.381443298969072e-05, + "loss": 0.3611, + "step": 34256 + }, + { + "epoch": 1.9139872056317568, + "grad_norm": 0.429257869720459, + "learning_rate": 4.378641864634693e-05, + "loss": 0.4984, + "step": 34257 + }, + { + "epoch": 1.9140430762354388, + "grad_norm": 0.40350064635276794, + "learning_rate": 4.3758404303003135e-05, + "loss": 0.4956, + "step": 34258 + }, + { + "epoch": 1.9140989468391205, + "grad_norm": 0.31435227394104004, + "learning_rate": 4.373038995965935e-05, + "loss": 0.3102, + "step": 34259 + }, + { + "epoch": 1.9141548174428025, + "grad_norm": 0.4137342572212219, + "learning_rate": 4.3702375616315554e-05, + "loss": 0.569, + "step": 34260 + }, + { + "epoch": 1.9142106880464844, + "grad_norm": 0.48849862813949585, + "learning_rate": 4.367436127297176e-05, + "loss": 0.4927, + "step": 34261 + }, + { + "epoch": 1.9142665586501662, + "grad_norm": 0.7119266986846924, + "learning_rate": 4.3646346929627965e-05, + "loss": 0.4515, + "step": 34262 + }, + { + "epoch": 1.914322429253848, + "grad_norm": 0.9770709276199341, + "learning_rate": 4.361833258628418e-05, + "loss": 0.3714, + "step": 34263 + }, + { + "epoch": 1.91437829985753, + "grad_norm": 0.4106422960758209, + "learning_rate": 4.3590318242940384e-05, + "loss": 0.363, + "step": 34264 + }, + { + "epoch": 1.9144341704612118, + "grad_norm": 0.6092843413352966, + "learning_rate": 4.3562303899596596e-05, + "loss": 0.5558, + "step": 34265 + }, + { + "epoch": 1.9144900410648937, + "grad_norm": 1.3964316844940186, + "learning_rate": 4.353428955625281e-05, + "loss": 0.3788, + "step": 34266 + }, + { + "epoch": 1.9145459116685757, + "grad_norm": 0.4352029263973236, + "learning_rate": 4.350627521290901e-05, + "loss": 0.3867, + "step": 34267 + }, + { + "epoch": 1.9146017822722574, + "grad_norm": 0.5091961622238159, + "learning_rate": 4.347826086956522e-05, + "loss": 0.411, + "step": 34268 + }, + { + "epoch": 1.9146576528759394, + "grad_norm": 0.5088726282119751, + "learning_rate": 4.3450246526221426e-05, + "loss": 0.3896, + "step": 34269 + }, + { + "epoch": 1.9147135234796213, + "grad_norm": 0.46792590618133545, + "learning_rate": 4.342223218287764e-05, + "loss": 0.3889, + "step": 34270 + }, + { + "epoch": 1.914769394083303, + "grad_norm": 0.4207637310028076, + "learning_rate": 4.3394217839533845e-05, + "loss": 0.3123, + "step": 34271 + }, + { + "epoch": 1.914825264686985, + "grad_norm": 0.36659103631973267, + "learning_rate": 4.336620349619005e-05, + "loss": 0.389, + "step": 34272 + }, + { + "epoch": 1.914881135290667, + "grad_norm": 1.8973172903060913, + "learning_rate": 4.3338189152846256e-05, + "loss": 0.3436, + "step": 34273 + }, + { + "epoch": 1.9149370058943487, + "grad_norm": 0.30367985367774963, + "learning_rate": 4.331017480950247e-05, + "loss": 0.4117, + "step": 34274 + }, + { + "epoch": 1.9149928764980304, + "grad_norm": 0.4243701994419098, + "learning_rate": 4.3282160466158675e-05, + "loss": 0.3977, + "step": 34275 + }, + { + "epoch": 1.9150487471017126, + "grad_norm": 0.43293315172195435, + "learning_rate": 4.325414612281489e-05, + "loss": 0.4455, + "step": 34276 + }, + { + "epoch": 1.9151046177053943, + "grad_norm": 0.4271524250507355, + "learning_rate": 4.3226131779471086e-05, + "loss": 0.4309, + "step": 34277 + }, + { + "epoch": 1.915160488309076, + "grad_norm": 0.4816223084926605, + "learning_rate": 4.31981174361273e-05, + "loss": 0.43, + "step": 34278 + }, + { + "epoch": 1.9152163589127582, + "grad_norm": 0.34293991327285767, + "learning_rate": 4.3170103092783505e-05, + "loss": 0.3146, + "step": 34279 + }, + { + "epoch": 1.91527222951644, + "grad_norm": 0.6228797435760498, + "learning_rate": 4.314208874943972e-05, + "loss": 0.4457, + "step": 34280 + }, + { + "epoch": 1.9153281001201217, + "grad_norm": 1.4898720979690552, + "learning_rate": 4.311407440609592e-05, + "loss": 0.4482, + "step": 34281 + }, + { + "epoch": 1.9153839707238036, + "grad_norm": 0.3441813290119171, + "learning_rate": 4.3086060062752136e-05, + "loss": 0.4904, + "step": 34282 + }, + { + "epoch": 1.9154398413274856, + "grad_norm": 0.8991810083389282, + "learning_rate": 4.3058045719408335e-05, + "loss": 0.35, + "step": 34283 + }, + { + "epoch": 1.9154957119311673, + "grad_norm": 0.3679778575897217, + "learning_rate": 4.303003137606455e-05, + "loss": 0.4862, + "step": 34284 + }, + { + "epoch": 1.9155515825348493, + "grad_norm": 0.41680648922920227, + "learning_rate": 4.300201703272075e-05, + "loss": 0.4846, + "step": 34285 + }, + { + "epoch": 1.9156074531385312, + "grad_norm": 0.5573477745056152, + "learning_rate": 4.2974002689376966e-05, + "loss": 0.4495, + "step": 34286 + }, + { + "epoch": 1.915663323742213, + "grad_norm": 0.28624212741851807, + "learning_rate": 4.294598834603317e-05, + "loss": 0.3691, + "step": 34287 + }, + { + "epoch": 1.9157191943458949, + "grad_norm": 0.4706348478794098, + "learning_rate": 4.291797400268938e-05, + "loss": 0.4968, + "step": 34288 + }, + { + "epoch": 1.9157750649495768, + "grad_norm": 0.8037953972816467, + "learning_rate": 4.288995965934558e-05, + "loss": 0.5268, + "step": 34289 + }, + { + "epoch": 1.9158309355532586, + "grad_norm": 0.4223778545856476, + "learning_rate": 4.2861945316001796e-05, + "loss": 0.4076, + "step": 34290 + }, + { + "epoch": 1.9158868061569405, + "grad_norm": 6.864068031311035, + "learning_rate": 4.2833930972658e-05, + "loss": 0.4691, + "step": 34291 + }, + { + "epoch": 1.9159426767606225, + "grad_norm": 0.4562666118144989, + "learning_rate": 4.2805916629314214e-05, + "loss": 0.3974, + "step": 34292 + }, + { + "epoch": 1.9159985473643042, + "grad_norm": 0.4775877296924591, + "learning_rate": 4.277790228597041e-05, + "loss": 0.4683, + "step": 34293 + }, + { + "epoch": 1.9160544179679861, + "grad_norm": 0.4554876685142517, + "learning_rate": 4.2749887942626626e-05, + "loss": 0.4717, + "step": 34294 + }, + { + "epoch": 1.916110288571668, + "grad_norm": 0.7127218246459961, + "learning_rate": 4.272187359928283e-05, + "loss": 0.4943, + "step": 34295 + }, + { + "epoch": 1.9161661591753498, + "grad_norm": 0.5725153088569641, + "learning_rate": 4.2693859255939044e-05, + "loss": 0.4709, + "step": 34296 + }, + { + "epoch": 1.9162220297790318, + "grad_norm": 0.321032851934433, + "learning_rate": 4.266584491259525e-05, + "loss": 0.369, + "step": 34297 + }, + { + "epoch": 1.9162779003827137, + "grad_norm": 0.3633958399295807, + "learning_rate": 4.263783056925146e-05, + "loss": 0.3137, + "step": 34298 + }, + { + "epoch": 1.9163337709863955, + "grad_norm": 0.5405903458595276, + "learning_rate": 4.260981622590766e-05, + "loss": 0.498, + "step": 34299 + }, + { + "epoch": 1.9163896415900774, + "grad_norm": 0.37858614325523376, + "learning_rate": 4.2581801882563874e-05, + "loss": 0.4154, + "step": 34300 + }, + { + "epoch": 1.9164455121937594, + "grad_norm": 0.478780061006546, + "learning_rate": 4.255378753922008e-05, + "loss": 0.502, + "step": 34301 + }, + { + "epoch": 1.916501382797441, + "grad_norm": 4.087695598602295, + "learning_rate": 4.252577319587629e-05, + "loss": 0.4066, + "step": 34302 + }, + { + "epoch": 1.916557253401123, + "grad_norm": 1.0947344303131104, + "learning_rate": 4.24977588525325e-05, + "loss": 0.449, + "step": 34303 + }, + { + "epoch": 1.916613124004805, + "grad_norm": 0.5400902628898621, + "learning_rate": 4.2469744509188704e-05, + "loss": 0.3671, + "step": 34304 + }, + { + "epoch": 1.9166689946084867, + "grad_norm": 0.40932977199554443, + "learning_rate": 4.244173016584491e-05, + "loss": 0.4082, + "step": 34305 + }, + { + "epoch": 1.9167248652121687, + "grad_norm": 0.39133143424987793, + "learning_rate": 4.241371582250112e-05, + "loss": 0.4193, + "step": 34306 + }, + { + "epoch": 1.9167807358158506, + "grad_norm": 0.5038919448852539, + "learning_rate": 4.238570147915733e-05, + "loss": 0.2829, + "step": 34307 + }, + { + "epoch": 1.9168366064195324, + "grad_norm": 0.49355506896972656, + "learning_rate": 4.235768713581354e-05, + "loss": 0.3566, + "step": 34308 + }, + { + "epoch": 1.916892477023214, + "grad_norm": 0.4365871250629425, + "learning_rate": 4.232967279246974e-05, + "loss": 0.3939, + "step": 34309 + }, + { + "epoch": 1.9169483476268963, + "grad_norm": 0.6090376973152161, + "learning_rate": 4.230165844912595e-05, + "loss": 0.3085, + "step": 34310 + }, + { + "epoch": 1.917004218230578, + "grad_norm": 1.4619123935699463, + "learning_rate": 4.227364410578216e-05, + "loss": 0.4331, + "step": 34311 + }, + { + "epoch": 1.9170600888342597, + "grad_norm": 0.443119615316391, + "learning_rate": 4.224562976243837e-05, + "loss": 0.3711, + "step": 34312 + }, + { + "epoch": 1.9171159594379419, + "grad_norm": 0.32106080651283264, + "learning_rate": 4.221761541909458e-05, + "loss": 0.4036, + "step": 34313 + }, + { + "epoch": 1.9171718300416236, + "grad_norm": 0.35212138295173645, + "learning_rate": 4.218960107575079e-05, + "loss": 0.4312, + "step": 34314 + }, + { + "epoch": 1.9172277006453053, + "grad_norm": 0.43672260642051697, + "learning_rate": 4.216158673240699e-05, + "loss": 0.4509, + "step": 34315 + }, + { + "epoch": 1.9172835712489873, + "grad_norm": 0.6347072720527649, + "learning_rate": 4.21335723890632e-05, + "loss": 0.4255, + "step": 34316 + }, + { + "epoch": 1.9173394418526692, + "grad_norm": 0.3297475576400757, + "learning_rate": 4.210555804571941e-05, + "loss": 0.3245, + "step": 34317 + }, + { + "epoch": 1.917395312456351, + "grad_norm": 0.40941980481147766, + "learning_rate": 4.207754370237562e-05, + "loss": 0.3925, + "step": 34318 + }, + { + "epoch": 1.917451183060033, + "grad_norm": 0.34163805842399597, + "learning_rate": 4.2049529359031825e-05, + "loss": 0.3809, + "step": 34319 + }, + { + "epoch": 1.9175070536637149, + "grad_norm": 1.198486328125, + "learning_rate": 4.202151501568803e-05, + "loss": 0.628, + "step": 34320 + }, + { + "epoch": 1.9175629242673966, + "grad_norm": 0.3798491656780243, + "learning_rate": 4.199350067234424e-05, + "loss": 0.3934, + "step": 34321 + }, + { + "epoch": 1.9176187948710786, + "grad_norm": 1.7939612865447998, + "learning_rate": 4.196548632900045e-05, + "loss": 0.3712, + "step": 34322 + }, + { + "epoch": 1.9176746654747605, + "grad_norm": 0.37783968448638916, + "learning_rate": 4.1937471985656655e-05, + "loss": 0.348, + "step": 34323 + }, + { + "epoch": 1.9177305360784422, + "grad_norm": 0.4197753965854645, + "learning_rate": 4.190945764231287e-05, + "loss": 0.4728, + "step": 34324 + }, + { + "epoch": 1.9177864066821242, + "grad_norm": 0.4825836420059204, + "learning_rate": 4.188144329896907e-05, + "loss": 0.4438, + "step": 34325 + }, + { + "epoch": 1.9178422772858061, + "grad_norm": 1.367897629737854, + "learning_rate": 4.185342895562528e-05, + "loss": 0.517, + "step": 34326 + }, + { + "epoch": 1.9178981478894879, + "grad_norm": 1.111684799194336, + "learning_rate": 4.1825414612281485e-05, + "loss": 0.4035, + "step": 34327 + }, + { + "epoch": 1.9179540184931698, + "grad_norm": 2.4506494998931885, + "learning_rate": 4.17974002689377e-05, + "loss": 0.5143, + "step": 34328 + }, + { + "epoch": 1.9180098890968518, + "grad_norm": 0.414903849363327, + "learning_rate": 4.1769385925593904e-05, + "loss": 0.4339, + "step": 34329 + }, + { + "epoch": 1.9180657597005335, + "grad_norm": 0.32991036772727966, + "learning_rate": 4.1741371582250116e-05, + "loss": 0.4447, + "step": 34330 + }, + { + "epoch": 1.9181216303042155, + "grad_norm": 0.38050350546836853, + "learning_rate": 4.1713357238906315e-05, + "loss": 0.4352, + "step": 34331 + }, + { + "epoch": 1.9181775009078974, + "grad_norm": 0.40988799929618835, + "learning_rate": 4.168534289556253e-05, + "loss": 0.3246, + "step": 34332 + }, + { + "epoch": 1.9182333715115791, + "grad_norm": 0.369305819272995, + "learning_rate": 4.165732855221874e-05, + "loss": 0.4093, + "step": 34333 + }, + { + "epoch": 1.918289242115261, + "grad_norm": 0.4962928593158722, + "learning_rate": 4.1629314208874946e-05, + "loss": 0.3268, + "step": 34334 + }, + { + "epoch": 1.918345112718943, + "grad_norm": 0.4778684675693512, + "learning_rate": 4.160129986553116e-05, + "loss": 0.456, + "step": 34335 + }, + { + "epoch": 1.9184009833226248, + "grad_norm": 0.9261903166770935, + "learning_rate": 4.157328552218736e-05, + "loss": 0.4361, + "step": 34336 + }, + { + "epoch": 1.9184568539263067, + "grad_norm": 0.43509945273399353, + "learning_rate": 4.154527117884357e-05, + "loss": 0.2834, + "step": 34337 + }, + { + "epoch": 1.9185127245299887, + "grad_norm": 0.28730955719947815, + "learning_rate": 4.1517256835499777e-05, + "loss": 0.3201, + "step": 34338 + }, + { + "epoch": 1.9185685951336704, + "grad_norm": 1.3418009281158447, + "learning_rate": 4.148924249215599e-05, + "loss": 0.3665, + "step": 34339 + }, + { + "epoch": 1.9186244657373523, + "grad_norm": 0.3285287916660309, + "learning_rate": 4.1461228148812195e-05, + "loss": 0.3232, + "step": 34340 + }, + { + "epoch": 1.9186803363410343, + "grad_norm": 0.567524790763855, + "learning_rate": 4.14332138054684e-05, + "loss": 0.4421, + "step": 34341 + }, + { + "epoch": 1.918736206944716, + "grad_norm": 0.36297377943992615, + "learning_rate": 4.1405199462124607e-05, + "loss": 0.3659, + "step": 34342 + }, + { + "epoch": 1.9187920775483978, + "grad_norm": 2.056252956390381, + "learning_rate": 4.137718511878082e-05, + "loss": 0.37, + "step": 34343 + }, + { + "epoch": 1.91884794815208, + "grad_norm": 1.4104450941085815, + "learning_rate": 4.1349170775437025e-05, + "loss": 0.4075, + "step": 34344 + }, + { + "epoch": 1.9189038187557617, + "grad_norm": 1.630632996559143, + "learning_rate": 4.132115643209324e-05, + "loss": 0.4144, + "step": 34345 + }, + { + "epoch": 1.9189596893594434, + "grad_norm": 0.41014495491981506, + "learning_rate": 4.1293142088749437e-05, + "loss": 0.3841, + "step": 34346 + }, + { + "epoch": 1.9190155599631256, + "grad_norm": 0.416121244430542, + "learning_rate": 4.126512774540565e-05, + "loss": 0.3493, + "step": 34347 + }, + { + "epoch": 1.9190714305668073, + "grad_norm": 0.8764697909355164, + "learning_rate": 4.1237113402061855e-05, + "loss": 0.5919, + "step": 34348 + }, + { + "epoch": 1.919127301170489, + "grad_norm": 0.4808672368526459, + "learning_rate": 4.120909905871807e-05, + "loss": 0.5171, + "step": 34349 + }, + { + "epoch": 1.919183171774171, + "grad_norm": 0.4413376748561859, + "learning_rate": 4.1181084715374273e-05, + "loss": 0.3482, + "step": 34350 + }, + { + "epoch": 1.919239042377853, + "grad_norm": 0.3182777464389801, + "learning_rate": 4.1153070372030486e-05, + "loss": 0.3119, + "step": 34351 + }, + { + "epoch": 1.9192949129815347, + "grad_norm": 0.32238835096359253, + "learning_rate": 4.1125056028686685e-05, + "loss": 0.3732, + "step": 34352 + }, + { + "epoch": 1.9193507835852166, + "grad_norm": 0.5525793433189392, + "learning_rate": 4.10970416853429e-05, + "loss": 0.6421, + "step": 34353 + }, + { + "epoch": 1.9194066541888986, + "grad_norm": 0.3692116141319275, + "learning_rate": 4.1069027341999103e-05, + "loss": 0.4206, + "step": 34354 + }, + { + "epoch": 1.9194625247925803, + "grad_norm": 0.5363834500312805, + "learning_rate": 4.1041012998655316e-05, + "loss": 0.4135, + "step": 34355 + }, + { + "epoch": 1.9195183953962622, + "grad_norm": 0.3896879255771637, + "learning_rate": 4.101299865531152e-05, + "loss": 0.3537, + "step": 34356 + }, + { + "epoch": 1.9195742659999442, + "grad_norm": 0.7565752267837524, + "learning_rate": 4.098498431196773e-05, + "loss": 0.4636, + "step": 34357 + }, + { + "epoch": 1.919630136603626, + "grad_norm": 0.435890257358551, + "learning_rate": 4.0956969968623933e-05, + "loss": 0.379, + "step": 34358 + }, + { + "epoch": 1.9196860072073079, + "grad_norm": 0.3476204574108124, + "learning_rate": 4.0928955625280146e-05, + "loss": 0.5272, + "step": 34359 + }, + { + "epoch": 1.9197418778109898, + "grad_norm": 0.406837522983551, + "learning_rate": 4.090094128193635e-05, + "loss": 0.4514, + "step": 34360 + }, + { + "epoch": 1.9197977484146715, + "grad_norm": 0.957166850566864, + "learning_rate": 4.0872926938592564e-05, + "loss": 0.3582, + "step": 34361 + }, + { + "epoch": 1.9198536190183535, + "grad_norm": 0.31613242626190186, + "learning_rate": 4.0844912595248764e-05, + "loss": 0.3005, + "step": 34362 + }, + { + "epoch": 1.9199094896220354, + "grad_norm": 0.40637972950935364, + "learning_rate": 4.0816898251904976e-05, + "loss": 0.4648, + "step": 34363 + }, + { + "epoch": 1.9199653602257172, + "grad_norm": 0.3344730734825134, + "learning_rate": 4.078888390856118e-05, + "loss": 0.3178, + "step": 34364 + }, + { + "epoch": 1.9200212308293991, + "grad_norm": 0.40923750400543213, + "learning_rate": 4.0760869565217395e-05, + "loss": 0.4206, + "step": 34365 + }, + { + "epoch": 1.920077101433081, + "grad_norm": 0.45278751850128174, + "learning_rate": 4.07328552218736e-05, + "loss": 0.3435, + "step": 34366 + }, + { + "epoch": 1.9201329720367628, + "grad_norm": 0.3391481935977936, + "learning_rate": 4.070484087852981e-05, + "loss": 0.3292, + "step": 34367 + }, + { + "epoch": 1.9201888426404448, + "grad_norm": 0.4437628984451294, + "learning_rate": 4.067682653518601e-05, + "loss": 0.4241, + "step": 34368 + }, + { + "epoch": 1.9202447132441267, + "grad_norm": 0.44838112592697144, + "learning_rate": 4.0648812191842225e-05, + "loss": 0.3403, + "step": 34369 + }, + { + "epoch": 1.9203005838478084, + "grad_norm": 0.424727201461792, + "learning_rate": 4.062079784849843e-05, + "loss": 0.4419, + "step": 34370 + }, + { + "epoch": 1.9203564544514904, + "grad_norm": 0.4063621163368225, + "learning_rate": 4.059278350515464e-05, + "loss": 0.3789, + "step": 34371 + }, + { + "epoch": 1.9204123250551723, + "grad_norm": 0.3992321491241455, + "learning_rate": 4.056476916181085e-05, + "loss": 0.4609, + "step": 34372 + }, + { + "epoch": 1.920468195658854, + "grad_norm": 0.3840164542198181, + "learning_rate": 4.0536754818467055e-05, + "loss": 0.3411, + "step": 34373 + }, + { + "epoch": 1.9205240662625358, + "grad_norm": 0.5726984143257141, + "learning_rate": 4.050874047512326e-05, + "loss": 0.4423, + "step": 34374 + }, + { + "epoch": 1.920579936866218, + "grad_norm": 0.4032820165157318, + "learning_rate": 4.048072613177947e-05, + "loss": 0.38, + "step": 34375 + }, + { + "epoch": 1.9206358074698997, + "grad_norm": 0.6653097867965698, + "learning_rate": 4.045271178843568e-05, + "loss": 0.4144, + "step": 34376 + }, + { + "epoch": 1.9206916780735814, + "grad_norm": 0.4279140532016754, + "learning_rate": 4.042469744509189e-05, + "loss": 0.4577, + "step": 34377 + }, + { + "epoch": 1.9207475486772636, + "grad_norm": 0.6527025699615479, + "learning_rate": 4.039668310174809e-05, + "loss": 0.3765, + "step": 34378 + }, + { + "epoch": 1.9208034192809453, + "grad_norm": 0.45855712890625, + "learning_rate": 4.03686687584043e-05, + "loss": 0.469, + "step": 34379 + }, + { + "epoch": 1.920859289884627, + "grad_norm": 0.4895079433917999, + "learning_rate": 4.034065441506051e-05, + "loss": 0.4244, + "step": 34380 + }, + { + "epoch": 1.9209151604883092, + "grad_norm": 0.4190917909145355, + "learning_rate": 4.031264007171672e-05, + "loss": 0.4052, + "step": 34381 + }, + { + "epoch": 1.920971031091991, + "grad_norm": 0.7345749139785767, + "learning_rate": 4.028462572837293e-05, + "loss": 0.4265, + "step": 34382 + }, + { + "epoch": 1.9210269016956727, + "grad_norm": 0.37472793459892273, + "learning_rate": 4.025661138502914e-05, + "loss": 0.425, + "step": 34383 + }, + { + "epoch": 1.9210827722993546, + "grad_norm": 1.9790451526641846, + "learning_rate": 4.022859704168534e-05, + "loss": 0.4876, + "step": 34384 + }, + { + "epoch": 1.9211386429030366, + "grad_norm": 0.5686604976654053, + "learning_rate": 4.020058269834155e-05, + "loss": 0.3547, + "step": 34385 + }, + { + "epoch": 1.9211945135067183, + "grad_norm": 0.28960850834846497, + "learning_rate": 4.017256835499776e-05, + "loss": 0.3012, + "step": 34386 + }, + { + "epoch": 1.9212503841104003, + "grad_norm": 0.9621022939682007, + "learning_rate": 4.014455401165397e-05, + "loss": 0.384, + "step": 34387 + }, + { + "epoch": 1.9213062547140822, + "grad_norm": 0.360082745552063, + "learning_rate": 4.0116539668310176e-05, + "loss": 0.3574, + "step": 34388 + }, + { + "epoch": 1.921362125317764, + "grad_norm": 0.46664637327194214, + "learning_rate": 4.008852532496638e-05, + "loss": 0.3885, + "step": 34389 + }, + { + "epoch": 1.921417995921446, + "grad_norm": 0.3236210346221924, + "learning_rate": 4.006051098162259e-05, + "loss": 0.4293, + "step": 34390 + }, + { + "epoch": 1.9214738665251279, + "grad_norm": 0.30381304025650024, + "learning_rate": 4.00324966382788e-05, + "loss": 0.3535, + "step": 34391 + }, + { + "epoch": 1.9215297371288096, + "grad_norm": 0.41973772644996643, + "learning_rate": 4.0004482294935006e-05, + "loss": 0.3231, + "step": 34392 + }, + { + "epoch": 1.9215856077324915, + "grad_norm": 0.42850401997566223, + "learning_rate": 3.997646795159122e-05, + "loss": 0.4133, + "step": 34393 + }, + { + "epoch": 1.9216414783361735, + "grad_norm": 0.3758770823478699, + "learning_rate": 3.994845360824742e-05, + "loss": 0.3915, + "step": 34394 + }, + { + "epoch": 1.9216973489398552, + "grad_norm": 0.45978522300720215, + "learning_rate": 3.992043926490363e-05, + "loss": 0.3964, + "step": 34395 + }, + { + "epoch": 1.9217532195435372, + "grad_norm": 0.305429607629776, + "learning_rate": 3.9892424921559836e-05, + "loss": 0.3929, + "step": 34396 + }, + { + "epoch": 1.9218090901472191, + "grad_norm": 0.529990553855896, + "learning_rate": 3.986441057821605e-05, + "loss": 0.3845, + "step": 34397 + }, + { + "epoch": 1.9218649607509009, + "grad_norm": 0.4399206042289734, + "learning_rate": 3.9836396234872254e-05, + "loss": 0.3575, + "step": 34398 + }, + { + "epoch": 1.9219208313545828, + "grad_norm": 0.41393619775772095, + "learning_rate": 3.980838189152847e-05, + "loss": 0.4393, + "step": 34399 + }, + { + "epoch": 1.9219767019582648, + "grad_norm": 0.3957655131816864, + "learning_rate": 3.9780367548184666e-05, + "loss": 0.397, + "step": 34400 + }, + { + "epoch": 1.9220325725619465, + "grad_norm": 9.238204002380371, + "learning_rate": 3.975235320484088e-05, + "loss": 0.3836, + "step": 34401 + }, + { + "epoch": 1.9220884431656284, + "grad_norm": 0.4644838273525238, + "learning_rate": 3.972433886149709e-05, + "loss": 0.5119, + "step": 34402 + }, + { + "epoch": 1.9221443137693104, + "grad_norm": 0.44145816564559937, + "learning_rate": 3.96963245181533e-05, + "loss": 0.3995, + "step": 34403 + }, + { + "epoch": 1.9222001843729921, + "grad_norm": 0.5054018497467041, + "learning_rate": 3.966831017480951e-05, + "loss": 0.396, + "step": 34404 + }, + { + "epoch": 1.922256054976674, + "grad_norm": 9.454402923583984, + "learning_rate": 3.964029583146571e-05, + "loss": 0.3799, + "step": 34405 + }, + { + "epoch": 1.922311925580356, + "grad_norm": 0.3360656499862671, + "learning_rate": 3.961228148812192e-05, + "loss": 0.3711, + "step": 34406 + }, + { + "epoch": 1.9223677961840377, + "grad_norm": 0.4442393481731415, + "learning_rate": 3.958426714477813e-05, + "loss": 0.3991, + "step": 34407 + }, + { + "epoch": 1.9224236667877195, + "grad_norm": 0.5351799130439758, + "learning_rate": 3.955625280143434e-05, + "loss": 0.3898, + "step": 34408 + }, + { + "epoch": 1.9224795373914016, + "grad_norm": 0.31341543793678284, + "learning_rate": 3.9528238458090545e-05, + "loss": 0.3938, + "step": 34409 + }, + { + "epoch": 1.9225354079950834, + "grad_norm": 0.3622884750366211, + "learning_rate": 3.950022411474675e-05, + "loss": 0.4041, + "step": 34410 + }, + { + "epoch": 1.922591278598765, + "grad_norm": 0.4285779595375061, + "learning_rate": 3.947220977140296e-05, + "loss": 0.46, + "step": 34411 + }, + { + "epoch": 1.9226471492024473, + "grad_norm": 0.3727155029773712, + "learning_rate": 3.944419542805917e-05, + "loss": 0.3803, + "step": 34412 + }, + { + "epoch": 1.922703019806129, + "grad_norm": 0.4688851237297058, + "learning_rate": 3.9416181084715375e-05, + "loss": 0.3312, + "step": 34413 + }, + { + "epoch": 1.9227588904098107, + "grad_norm": 0.418640673160553, + "learning_rate": 3.938816674137159e-05, + "loss": 0.4323, + "step": 34414 + }, + { + "epoch": 1.922814761013493, + "grad_norm": 0.4609624445438385, + "learning_rate": 3.9360152398027794e-05, + "loss": 0.5079, + "step": 34415 + }, + { + "epoch": 1.9228706316171746, + "grad_norm": 0.35610467195510864, + "learning_rate": 3.9332138054684e-05, + "loss": 0.3211, + "step": 34416 + }, + { + "epoch": 1.9229265022208564, + "grad_norm": 0.2769652307033539, + "learning_rate": 3.9304123711340205e-05, + "loss": 0.316, + "step": 34417 + }, + { + "epoch": 1.9229823728245383, + "grad_norm": 0.5993046164512634, + "learning_rate": 3.927610936799642e-05, + "loss": 0.4902, + "step": 34418 + }, + { + "epoch": 1.9230382434282203, + "grad_norm": 0.5783616304397583, + "learning_rate": 3.9248095024652624e-05, + "loss": 0.385, + "step": 34419 + }, + { + "epoch": 1.923094114031902, + "grad_norm": 0.3645850121974945, + "learning_rate": 3.9220080681308836e-05, + "loss": 0.4391, + "step": 34420 + }, + { + "epoch": 1.923149984635584, + "grad_norm": 0.3187403380870819, + "learning_rate": 3.9192066337965035e-05, + "loss": 0.2742, + "step": 34421 + }, + { + "epoch": 1.923205855239266, + "grad_norm": 1.0897845029830933, + "learning_rate": 3.916405199462125e-05, + "loss": 0.4266, + "step": 34422 + }, + { + "epoch": 1.9232617258429476, + "grad_norm": 0.3021298944950104, + "learning_rate": 3.9136037651277454e-05, + "loss": 0.3508, + "step": 34423 + }, + { + "epoch": 1.9233175964466296, + "grad_norm": 0.465518057346344, + "learning_rate": 3.9108023307933666e-05, + "loss": 0.378, + "step": 34424 + }, + { + "epoch": 1.9233734670503115, + "grad_norm": 0.35533660650253296, + "learning_rate": 3.908000896458987e-05, + "loss": 0.4276, + "step": 34425 + }, + { + "epoch": 1.9234293376539933, + "grad_norm": 0.9103195667266846, + "learning_rate": 3.905199462124608e-05, + "loss": 0.399, + "step": 34426 + }, + { + "epoch": 1.9234852082576752, + "grad_norm": 0.710067868232727, + "learning_rate": 3.9023980277902284e-05, + "loss": 0.7337, + "step": 34427 + }, + { + "epoch": 1.9235410788613572, + "grad_norm": 0.41249290108680725, + "learning_rate": 3.8995965934558496e-05, + "loss": 0.3784, + "step": 34428 + }, + { + "epoch": 1.923596949465039, + "grad_norm": 0.39959797263145447, + "learning_rate": 3.89679515912147e-05, + "loss": 0.4488, + "step": 34429 + }, + { + "epoch": 1.9236528200687208, + "grad_norm": 0.458814799785614, + "learning_rate": 3.8939937247870915e-05, + "loss": 0.4989, + "step": 34430 + }, + { + "epoch": 1.9237086906724028, + "grad_norm": 0.4139024317264557, + "learning_rate": 3.8911922904527114e-05, + "loss": 0.3944, + "step": 34431 + }, + { + "epoch": 1.9237645612760845, + "grad_norm": 0.36748799681663513, + "learning_rate": 3.8883908561183326e-05, + "loss": 0.3869, + "step": 34432 + }, + { + "epoch": 1.9238204318797665, + "grad_norm": 1.424985408782959, + "learning_rate": 3.885589421783953e-05, + "loss": 0.3622, + "step": 34433 + }, + { + "epoch": 1.9238763024834484, + "grad_norm": 4.544186592102051, + "learning_rate": 3.8827879874495745e-05, + "loss": 0.4027, + "step": 34434 + }, + { + "epoch": 1.9239321730871302, + "grad_norm": 0.44417664408683777, + "learning_rate": 3.879986553115195e-05, + "loss": 0.3925, + "step": 34435 + }, + { + "epoch": 1.923988043690812, + "grad_norm": 0.3176079988479614, + "learning_rate": 3.877185118780816e-05, + "loss": 0.3372, + "step": 34436 + }, + { + "epoch": 1.924043914294494, + "grad_norm": 0.3365500271320343, + "learning_rate": 3.874383684446436e-05, + "loss": 0.3519, + "step": 34437 + }, + { + "epoch": 1.9240997848981758, + "grad_norm": 0.561474084854126, + "learning_rate": 3.8715822501120575e-05, + "loss": 0.4229, + "step": 34438 + }, + { + "epoch": 1.9241556555018577, + "grad_norm": 0.480220228433609, + "learning_rate": 3.868780815777678e-05, + "loss": 0.4543, + "step": 34439 + }, + { + "epoch": 1.9242115261055397, + "grad_norm": 0.35136786103248596, + "learning_rate": 3.865979381443299e-05, + "loss": 0.4101, + "step": 34440 + }, + { + "epoch": 1.9242673967092214, + "grad_norm": 0.4422420859336853, + "learning_rate": 3.86317794710892e-05, + "loss": 0.3509, + "step": 34441 + }, + { + "epoch": 1.9243232673129032, + "grad_norm": 0.3071976900100708, + "learning_rate": 3.8603765127745405e-05, + "loss": 0.4475, + "step": 34442 + }, + { + "epoch": 1.9243791379165853, + "grad_norm": 1.1019513607025146, + "learning_rate": 3.857575078440161e-05, + "loss": 0.3683, + "step": 34443 + }, + { + "epoch": 1.924435008520267, + "grad_norm": 0.5266516208648682, + "learning_rate": 3.854773644105782e-05, + "loss": 0.3743, + "step": 34444 + }, + { + "epoch": 1.9244908791239488, + "grad_norm": 1.1812118291854858, + "learning_rate": 3.851972209771403e-05, + "loss": 0.4785, + "step": 34445 + }, + { + "epoch": 1.924546749727631, + "grad_norm": 0.41608959436416626, + "learning_rate": 3.849170775437024e-05, + "loss": 0.3805, + "step": 34446 + }, + { + "epoch": 1.9246026203313127, + "grad_norm": 0.381840318441391, + "learning_rate": 3.846369341102644e-05, + "loss": 0.4576, + "step": 34447 + }, + { + "epoch": 1.9246584909349944, + "grad_norm": 0.36750003695487976, + "learning_rate": 3.843567906768265e-05, + "loss": 0.2393, + "step": 34448 + }, + { + "epoch": 1.9247143615386766, + "grad_norm": 0.5425214171409607, + "learning_rate": 3.840766472433886e-05, + "loss": 0.4496, + "step": 34449 + }, + { + "epoch": 1.9247702321423583, + "grad_norm": 1.1650989055633545, + "learning_rate": 3.837965038099507e-05, + "loss": 0.3824, + "step": 34450 + }, + { + "epoch": 1.92482610274604, + "grad_norm": 0.45163071155548096, + "learning_rate": 3.835163603765128e-05, + "loss": 0.4118, + "step": 34451 + }, + { + "epoch": 1.924881973349722, + "grad_norm": 0.5177558064460754, + "learning_rate": 3.832362169430749e-05, + "loss": 0.4208, + "step": 34452 + }, + { + "epoch": 1.924937843953404, + "grad_norm": 0.9397730231285095, + "learning_rate": 3.829560735096369e-05, + "loss": 0.3666, + "step": 34453 + }, + { + "epoch": 1.9249937145570857, + "grad_norm": 0.5194541811943054, + "learning_rate": 3.82675930076199e-05, + "loss": 0.4596, + "step": 34454 + }, + { + "epoch": 1.9250495851607676, + "grad_norm": 0.2970203757286072, + "learning_rate": 3.823957866427611e-05, + "loss": 0.3054, + "step": 34455 + }, + { + "epoch": 1.9251054557644496, + "grad_norm": 0.3727014660835266, + "learning_rate": 3.821156432093232e-05, + "loss": 0.3951, + "step": 34456 + }, + { + "epoch": 1.9251613263681313, + "grad_norm": 0.5183871984481812, + "learning_rate": 3.8183549977588526e-05, + "loss": 0.4067, + "step": 34457 + }, + { + "epoch": 1.9252171969718133, + "grad_norm": 0.43049609661102295, + "learning_rate": 3.815553563424473e-05, + "loss": 0.4604, + "step": 34458 + }, + { + "epoch": 1.9252730675754952, + "grad_norm": 0.4322688579559326, + "learning_rate": 3.812752129090094e-05, + "loss": 0.4741, + "step": 34459 + }, + { + "epoch": 1.925328938179177, + "grad_norm": 0.4001346528530121, + "learning_rate": 3.809950694755715e-05, + "loss": 0.5018, + "step": 34460 + }, + { + "epoch": 1.925384808782859, + "grad_norm": 0.39848771691322327, + "learning_rate": 3.8071492604213356e-05, + "loss": 0.4235, + "step": 34461 + }, + { + "epoch": 1.9254406793865408, + "grad_norm": 1.1031758785247803, + "learning_rate": 3.804347826086957e-05, + "loss": 0.4182, + "step": 34462 + }, + { + "epoch": 1.9254965499902226, + "grad_norm": 1.3031487464904785, + "learning_rate": 3.801546391752577e-05, + "loss": 0.3699, + "step": 34463 + }, + { + "epoch": 1.9255524205939045, + "grad_norm": 0.4250287413597107, + "learning_rate": 3.798744957418198e-05, + "loss": 0.3649, + "step": 34464 + }, + { + "epoch": 1.9256082911975865, + "grad_norm": 1.4280849695205688, + "learning_rate": 3.7959435230838186e-05, + "loss": 0.3606, + "step": 34465 + }, + { + "epoch": 1.9256641618012682, + "grad_norm": 0.28692299127578735, + "learning_rate": 3.79314208874944e-05, + "loss": 0.3206, + "step": 34466 + }, + { + "epoch": 1.9257200324049502, + "grad_norm": 0.7679880857467651, + "learning_rate": 3.790340654415061e-05, + "loss": 0.3856, + "step": 34467 + }, + { + "epoch": 1.925775903008632, + "grad_norm": 0.4992069602012634, + "learning_rate": 3.787539220080682e-05, + "loss": 0.4559, + "step": 34468 + }, + { + "epoch": 1.9258317736123138, + "grad_norm": 0.5304147005081177, + "learning_rate": 3.784737785746302e-05, + "loss": 0.4322, + "step": 34469 + }, + { + "epoch": 1.9258876442159958, + "grad_norm": 0.3416355550289154, + "learning_rate": 3.781936351411923e-05, + "loss": 0.464, + "step": 34470 + }, + { + "epoch": 1.9259435148196777, + "grad_norm": 0.46232548356056213, + "learning_rate": 3.779134917077544e-05, + "loss": 0.4506, + "step": 34471 + }, + { + "epoch": 1.9259993854233595, + "grad_norm": 1.4116183519363403, + "learning_rate": 3.776333482743165e-05, + "loss": 0.4172, + "step": 34472 + }, + { + "epoch": 1.9260552560270414, + "grad_norm": 0.34797602891921997, + "learning_rate": 3.773532048408786e-05, + "loss": 0.4361, + "step": 34473 + }, + { + "epoch": 1.9261111266307234, + "grad_norm": 0.3821897804737091, + "learning_rate": 3.770730614074406e-05, + "loss": 0.343, + "step": 34474 + }, + { + "epoch": 1.926166997234405, + "grad_norm": 1.4994858503341675, + "learning_rate": 3.767929179740027e-05, + "loss": 0.5318, + "step": 34475 + }, + { + "epoch": 1.9262228678380868, + "grad_norm": 0.43359434604644775, + "learning_rate": 3.765127745405648e-05, + "loss": 0.3657, + "step": 34476 + }, + { + "epoch": 1.926278738441769, + "grad_norm": 0.8007287979125977, + "learning_rate": 3.762326311071269e-05, + "loss": 0.3624, + "step": 34477 + }, + { + "epoch": 1.9263346090454507, + "grad_norm": 0.6289954781532288, + "learning_rate": 3.7595248767368896e-05, + "loss": 0.4614, + "step": 34478 + }, + { + "epoch": 1.9263904796491325, + "grad_norm": 0.4194599986076355, + "learning_rate": 3.75672344240251e-05, + "loss": 0.4301, + "step": 34479 + }, + { + "epoch": 1.9264463502528146, + "grad_norm": 0.7934345602989197, + "learning_rate": 3.753922008068131e-05, + "loss": 0.4294, + "step": 34480 + }, + { + "epoch": 1.9265022208564964, + "grad_norm": 0.7096133828163147, + "learning_rate": 3.751120573733752e-05, + "loss": 0.4789, + "step": 34481 + }, + { + "epoch": 1.926558091460178, + "grad_norm": 0.3488916754722595, + "learning_rate": 3.7483191393993726e-05, + "loss": 0.3823, + "step": 34482 + }, + { + "epoch": 1.9266139620638603, + "grad_norm": 0.5258325338363647, + "learning_rate": 3.745517705064994e-05, + "loss": 0.4829, + "step": 34483 + }, + { + "epoch": 1.926669832667542, + "grad_norm": 0.5074584484100342, + "learning_rate": 3.7427162707306144e-05, + "loss": 0.3058, + "step": 34484 + }, + { + "epoch": 1.9267257032712237, + "grad_norm": 0.3149060010910034, + "learning_rate": 3.739914836396235e-05, + "loss": 0.3513, + "step": 34485 + }, + { + "epoch": 1.9267815738749057, + "grad_norm": 0.3674938380718231, + "learning_rate": 3.7371134020618556e-05, + "loss": 0.3867, + "step": 34486 + }, + { + "epoch": 1.9268374444785876, + "grad_norm": 0.34746280312538147, + "learning_rate": 3.734311967727477e-05, + "loss": 0.3317, + "step": 34487 + }, + { + "epoch": 1.9268933150822694, + "grad_norm": 0.3442893922328949, + "learning_rate": 3.7315105333930974e-05, + "loss": 0.3478, + "step": 34488 + }, + { + "epoch": 1.9269491856859513, + "grad_norm": 0.5434176921844482, + "learning_rate": 3.728709099058719e-05, + "loss": 0.4789, + "step": 34489 + }, + { + "epoch": 1.9270050562896333, + "grad_norm": 0.7284636497497559, + "learning_rate": 3.7259076647243386e-05, + "loss": 0.3574, + "step": 34490 + }, + { + "epoch": 1.927060926893315, + "grad_norm": 1.42918860912323, + "learning_rate": 3.72310623038996e-05, + "loss": 0.4221, + "step": 34491 + }, + { + "epoch": 1.927116797496997, + "grad_norm": 0.7368375062942505, + "learning_rate": 3.7203047960555804e-05, + "loss": 0.381, + "step": 34492 + }, + { + "epoch": 1.9271726681006789, + "grad_norm": 0.7260094881057739, + "learning_rate": 3.717503361721202e-05, + "loss": 0.3513, + "step": 34493 + }, + { + "epoch": 1.9272285387043606, + "grad_norm": 0.7004160284996033, + "learning_rate": 3.714701927386822e-05, + "loss": 0.4099, + "step": 34494 + }, + { + "epoch": 1.9272844093080426, + "grad_norm": 0.40637660026550293, + "learning_rate": 3.711900493052443e-05, + "loss": 0.4729, + "step": 34495 + }, + { + "epoch": 1.9273402799117245, + "grad_norm": 0.3416901230812073, + "learning_rate": 3.7090990587180634e-05, + "loss": 0.4468, + "step": 34496 + }, + { + "epoch": 1.9273961505154062, + "grad_norm": 0.4110446274280548, + "learning_rate": 3.706297624383685e-05, + "loss": 0.3702, + "step": 34497 + }, + { + "epoch": 1.9274520211190882, + "grad_norm": 0.34785929322242737, + "learning_rate": 3.703496190049305e-05, + "loss": 0.3494, + "step": 34498 + }, + { + "epoch": 1.9275078917227702, + "grad_norm": 0.34119343757629395, + "learning_rate": 3.7006947557149265e-05, + "loss": 0.4062, + "step": 34499 + }, + { + "epoch": 1.9275637623264519, + "grad_norm": 0.4147798418998718, + "learning_rate": 3.6978933213805464e-05, + "loss": 0.3559, + "step": 34500 + }, + { + "epoch": 1.9275637623264519, + "eval_cer": 0.08354336464709283, + "eval_loss": 0.3123348355293274, + "eval_runtime": 56.2496, + "eval_samples_per_second": 80.676, + "eval_steps_per_second": 5.049, + "eval_wer": 0.3333705762408871, + "step": 34500 + }, + { + "epoch": 1.9276196329301338, + "grad_norm": 0.5493800640106201, + "learning_rate": 3.695091887046168e-05, + "loss": 0.5491, + "step": 34501 + }, + { + "epoch": 1.9276755035338158, + "grad_norm": 0.30006372928619385, + "learning_rate": 3.692290452711788e-05, + "loss": 0.3375, + "step": 34502 + }, + { + "epoch": 1.9277313741374975, + "grad_norm": 0.3407485783100128, + "learning_rate": 3.6894890183774095e-05, + "loss": 0.3312, + "step": 34503 + }, + { + "epoch": 1.9277872447411795, + "grad_norm": 0.30694156885147095, + "learning_rate": 3.68668758404303e-05, + "loss": 0.3445, + "step": 34504 + }, + { + "epoch": 1.9278431153448614, + "grad_norm": 0.5319073796272278, + "learning_rate": 3.6838861497086514e-05, + "loss": 0.4113, + "step": 34505 + }, + { + "epoch": 1.9278989859485431, + "grad_norm": 0.5415977835655212, + "learning_rate": 3.681084715374271e-05, + "loss": 0.4538, + "step": 34506 + }, + { + "epoch": 1.927954856552225, + "grad_norm": 0.6221215724945068, + "learning_rate": 3.6782832810398925e-05, + "loss": 0.4079, + "step": 34507 + }, + { + "epoch": 1.928010727155907, + "grad_norm": 0.5531096458435059, + "learning_rate": 3.675481846705513e-05, + "loss": 0.4346, + "step": 34508 + }, + { + "epoch": 1.9280665977595888, + "grad_norm": 0.7627242207527161, + "learning_rate": 3.6726804123711344e-05, + "loss": 0.3747, + "step": 34509 + }, + { + "epoch": 1.9281224683632705, + "grad_norm": 0.6291773915290833, + "learning_rate": 3.669878978036755e-05, + "loss": 0.4694, + "step": 34510 + }, + { + "epoch": 1.9281783389669527, + "grad_norm": 0.34860196709632874, + "learning_rate": 3.6670775437023755e-05, + "loss": 0.3401, + "step": 34511 + }, + { + "epoch": 1.9282342095706344, + "grad_norm": 0.3759145140647888, + "learning_rate": 3.664276109367996e-05, + "loss": 0.3304, + "step": 34512 + }, + { + "epoch": 1.9282900801743161, + "grad_norm": 3.802002429962158, + "learning_rate": 3.6614746750336174e-05, + "loss": 0.3795, + "step": 34513 + }, + { + "epoch": 1.9283459507779983, + "grad_norm": 0.3899758458137512, + "learning_rate": 3.658673240699238e-05, + "loss": 0.3013, + "step": 34514 + }, + { + "epoch": 1.92840182138168, + "grad_norm": 0.9814776182174683, + "learning_rate": 3.655871806364859e-05, + "loss": 0.5583, + "step": 34515 + }, + { + "epoch": 1.9284576919853618, + "grad_norm": 0.3232879638671875, + "learning_rate": 3.653070372030479e-05, + "loss": 0.3653, + "step": 34516 + }, + { + "epoch": 1.928513562589044, + "grad_norm": 0.44970306754112244, + "learning_rate": 3.6502689376961004e-05, + "loss": 0.4243, + "step": 34517 + }, + { + "epoch": 1.9285694331927257, + "grad_norm": 0.3439804017543793, + "learning_rate": 3.647467503361721e-05, + "loss": 0.2947, + "step": 34518 + }, + { + "epoch": 1.9286253037964074, + "grad_norm": 0.3914286494255066, + "learning_rate": 3.644666069027342e-05, + "loss": 0.3376, + "step": 34519 + }, + { + "epoch": 1.9286811744000893, + "grad_norm": 1.000595211982727, + "learning_rate": 3.641864634692963e-05, + "loss": 0.3978, + "step": 34520 + }, + { + "epoch": 1.9287370450037713, + "grad_norm": 0.7306061387062073, + "learning_rate": 3.639063200358584e-05, + "loss": 0.4389, + "step": 34521 + }, + { + "epoch": 1.928792915607453, + "grad_norm": 0.6307012438774109, + "learning_rate": 3.636261766024204e-05, + "loss": 0.481, + "step": 34522 + }, + { + "epoch": 1.928848786211135, + "grad_norm": 1.8940296173095703, + "learning_rate": 3.633460331689825e-05, + "loss": 0.3562, + "step": 34523 + }, + { + "epoch": 1.928904656814817, + "grad_norm": 4.262044429779053, + "learning_rate": 3.630658897355446e-05, + "loss": 0.5667, + "step": 34524 + }, + { + "epoch": 1.9289605274184987, + "grad_norm": 0.3450900614261627, + "learning_rate": 3.627857463021067e-05, + "loss": 0.4252, + "step": 34525 + }, + { + "epoch": 1.9290163980221806, + "grad_norm": 0.33593589067459106, + "learning_rate": 3.6250560286866876e-05, + "loss": 0.4175, + "step": 34526 + }, + { + "epoch": 1.9290722686258626, + "grad_norm": 0.37478119134902954, + "learning_rate": 3.622254594352308e-05, + "loss": 0.3367, + "step": 34527 + }, + { + "epoch": 1.9291281392295443, + "grad_norm": 0.42749732732772827, + "learning_rate": 3.619453160017929e-05, + "loss": 0.5274, + "step": 34528 + }, + { + "epoch": 1.9291840098332262, + "grad_norm": 0.36133766174316406, + "learning_rate": 3.61665172568355e-05, + "loss": 0.5232, + "step": 34529 + }, + { + "epoch": 1.9292398804369082, + "grad_norm": 0.446828156709671, + "learning_rate": 3.6138502913491706e-05, + "loss": 0.4806, + "step": 34530 + }, + { + "epoch": 1.92929575104059, + "grad_norm": 0.6057213544845581, + "learning_rate": 3.611048857014792e-05, + "loss": 0.4796, + "step": 34531 + }, + { + "epoch": 1.9293516216442719, + "grad_norm": 0.36505335569381714, + "learning_rate": 3.608247422680412e-05, + "loss": 0.3676, + "step": 34532 + }, + { + "epoch": 1.9294074922479538, + "grad_norm": 0.49200087785720825, + "learning_rate": 3.605445988346033e-05, + "loss": 0.4063, + "step": 34533 + }, + { + "epoch": 1.9294633628516356, + "grad_norm": 0.3077874779701233, + "learning_rate": 3.6026445540116536e-05, + "loss": 0.336, + "step": 34534 + }, + { + "epoch": 1.9295192334553175, + "grad_norm": 0.3997020721435547, + "learning_rate": 3.599843119677275e-05, + "loss": 0.3336, + "step": 34535 + }, + { + "epoch": 1.9295751040589995, + "grad_norm": 1.657181739807129, + "learning_rate": 3.597041685342896e-05, + "loss": 0.3209, + "step": 34536 + }, + { + "epoch": 1.9296309746626812, + "grad_norm": 1.9332919120788574, + "learning_rate": 3.594240251008517e-05, + "loss": 0.4205, + "step": 34537 + }, + { + "epoch": 1.9296868452663631, + "grad_norm": 0.39734238386154175, + "learning_rate": 3.591438816674137e-05, + "loss": 0.4416, + "step": 34538 + }, + { + "epoch": 1.929742715870045, + "grad_norm": 0.40115827322006226, + "learning_rate": 3.588637382339758e-05, + "loss": 0.4701, + "step": 34539 + }, + { + "epoch": 1.9297985864737268, + "grad_norm": 0.455321729183197, + "learning_rate": 3.585835948005379e-05, + "loss": 0.6038, + "step": 34540 + }, + { + "epoch": 1.9298544570774088, + "grad_norm": 0.3794196546077728, + "learning_rate": 3.583034513671e-05, + "loss": 0.4538, + "step": 34541 + }, + { + "epoch": 1.9299103276810907, + "grad_norm": 0.44637224078178406, + "learning_rate": 3.580233079336621e-05, + "loss": 0.466, + "step": 34542 + }, + { + "epoch": 1.9299661982847724, + "grad_norm": 0.8015100359916687, + "learning_rate": 3.577431645002241e-05, + "loss": 0.5069, + "step": 34543 + }, + { + "epoch": 1.9300220688884542, + "grad_norm": 0.8752270936965942, + "learning_rate": 3.574630210667862e-05, + "loss": 0.5353, + "step": 34544 + }, + { + "epoch": 1.9300779394921364, + "grad_norm": 0.867617666721344, + "learning_rate": 3.571828776333483e-05, + "loss": 0.4494, + "step": 34545 + }, + { + "epoch": 1.930133810095818, + "grad_norm": 0.3374629020690918, + "learning_rate": 3.569027341999104e-05, + "loss": 0.3418, + "step": 34546 + }, + { + "epoch": 1.9301896806994998, + "grad_norm": 2.3250324726104736, + "learning_rate": 3.5662259076647246e-05, + "loss": 0.3697, + "step": 34547 + }, + { + "epoch": 1.930245551303182, + "grad_norm": 0.4004785120487213, + "learning_rate": 3.563424473330345e-05, + "loss": 0.3803, + "step": 34548 + }, + { + "epoch": 1.9303014219068637, + "grad_norm": 0.41887232661247253, + "learning_rate": 3.560623038995966e-05, + "loss": 0.3948, + "step": 34549 + }, + { + "epoch": 1.9303572925105454, + "grad_norm": 0.42769455909729004, + "learning_rate": 3.557821604661587e-05, + "loss": 0.3429, + "step": 34550 + }, + { + "epoch": 1.9304131631142276, + "grad_norm": 0.3530546724796295, + "learning_rate": 3.5550201703272076e-05, + "loss": 0.3953, + "step": 34551 + }, + { + "epoch": 1.9304690337179093, + "grad_norm": 0.44131162762641907, + "learning_rate": 3.552218735992829e-05, + "loss": 0.3138, + "step": 34552 + }, + { + "epoch": 1.930524904321591, + "grad_norm": 0.35916805267333984, + "learning_rate": 3.5494173016584494e-05, + "loss": 0.3949, + "step": 34553 + }, + { + "epoch": 1.930580774925273, + "grad_norm": 0.44831621646881104, + "learning_rate": 3.54661586732407e-05, + "loss": 0.4515, + "step": 34554 + }, + { + "epoch": 1.930636645528955, + "grad_norm": 0.5133212804794312, + "learning_rate": 3.5438144329896906e-05, + "loss": 0.582, + "step": 34555 + }, + { + "epoch": 1.9306925161326367, + "grad_norm": 0.3506346046924591, + "learning_rate": 3.541012998655312e-05, + "loss": 0.4231, + "step": 34556 + }, + { + "epoch": 1.9307483867363187, + "grad_norm": 0.44303497672080994, + "learning_rate": 3.5382115643209324e-05, + "loss": 0.4902, + "step": 34557 + }, + { + "epoch": 1.9308042573400006, + "grad_norm": 0.41234129667282104, + "learning_rate": 3.535410129986554e-05, + "loss": 0.4712, + "step": 34558 + }, + { + "epoch": 1.9308601279436823, + "grad_norm": 0.2961030900478363, + "learning_rate": 3.5326086956521736e-05, + "loss": 0.3808, + "step": 34559 + }, + { + "epoch": 1.9309159985473643, + "grad_norm": 0.820661723613739, + "learning_rate": 3.529807261317795e-05, + "loss": 0.4268, + "step": 34560 + }, + { + "epoch": 1.9309718691510462, + "grad_norm": 0.39451244473457336, + "learning_rate": 3.5270058269834154e-05, + "loss": 0.3962, + "step": 34561 + }, + { + "epoch": 1.931027739754728, + "grad_norm": 1.186583399772644, + "learning_rate": 3.524204392649037e-05, + "loss": 0.3703, + "step": 34562 + }, + { + "epoch": 1.93108361035841, + "grad_norm": 0.63531893491745, + "learning_rate": 3.521402958314657e-05, + "loss": 0.4754, + "step": 34563 + }, + { + "epoch": 1.9311394809620919, + "grad_norm": 0.3858046531677246, + "learning_rate": 3.518601523980278e-05, + "loss": 0.4318, + "step": 34564 + }, + { + "epoch": 1.9311953515657736, + "grad_norm": 0.3496503233909607, + "learning_rate": 3.5158000896458984e-05, + "loss": 0.3623, + "step": 34565 + }, + { + "epoch": 1.9312512221694556, + "grad_norm": 0.479705274105072, + "learning_rate": 3.51299865531152e-05, + "loss": 0.3978, + "step": 34566 + }, + { + "epoch": 1.9313070927731375, + "grad_norm": 0.42565521597862244, + "learning_rate": 3.51019722097714e-05, + "loss": 0.3661, + "step": 34567 + }, + { + "epoch": 1.9313629633768192, + "grad_norm": 0.3329249918460846, + "learning_rate": 3.5073957866427615e-05, + "loss": 0.3463, + "step": 34568 + }, + { + "epoch": 1.9314188339805012, + "grad_norm": 0.429065078496933, + "learning_rate": 3.504594352308382e-05, + "loss": 0.4451, + "step": 34569 + }, + { + "epoch": 1.9314747045841831, + "grad_norm": 0.37978002429008484, + "learning_rate": 3.501792917974003e-05, + "loss": 0.3193, + "step": 34570 + }, + { + "epoch": 1.9315305751878649, + "grad_norm": 1.637731909751892, + "learning_rate": 3.498991483639623e-05, + "loss": 0.3154, + "step": 34571 + }, + { + "epoch": 1.9315864457915468, + "grad_norm": 0.3908073902130127, + "learning_rate": 3.4961900493052446e-05, + "loss": 0.4011, + "step": 34572 + }, + { + "epoch": 1.9316423163952288, + "grad_norm": 0.37708181142807007, + "learning_rate": 3.493388614970865e-05, + "loss": 0.3381, + "step": 34573 + }, + { + "epoch": 1.9316981869989105, + "grad_norm": 0.515893280506134, + "learning_rate": 3.4905871806364864e-05, + "loss": 0.4496, + "step": 34574 + }, + { + "epoch": 1.9317540576025924, + "grad_norm": 0.7853911519050598, + "learning_rate": 3.487785746302106e-05, + "loss": 0.4106, + "step": 34575 + }, + { + "epoch": 1.9318099282062744, + "grad_norm": 0.49481770396232605, + "learning_rate": 3.4849843119677276e-05, + "loss": 0.6012, + "step": 34576 + }, + { + "epoch": 1.9318657988099561, + "grad_norm": 0.4792594015598297, + "learning_rate": 3.482182877633348e-05, + "loss": 0.3843, + "step": 34577 + }, + { + "epoch": 1.9319216694136379, + "grad_norm": 0.34969544410705566, + "learning_rate": 3.4793814432989694e-05, + "loss": 0.3476, + "step": 34578 + }, + { + "epoch": 1.93197754001732, + "grad_norm": 0.38755837082862854, + "learning_rate": 3.47658000896459e-05, + "loss": 0.3293, + "step": 34579 + }, + { + "epoch": 1.9320334106210018, + "grad_norm": 0.37803390622138977, + "learning_rate": 3.4737785746302106e-05, + "loss": 0.3423, + "step": 34580 + }, + { + "epoch": 1.9320892812246835, + "grad_norm": 0.26366057991981506, + "learning_rate": 3.470977140295831e-05, + "loss": 0.2746, + "step": 34581 + }, + { + "epoch": 1.9321451518283657, + "grad_norm": 0.3953426480293274, + "learning_rate": 3.4681757059614524e-05, + "loss": 0.3792, + "step": 34582 + }, + { + "epoch": 1.9322010224320474, + "grad_norm": 0.5923205614089966, + "learning_rate": 3.465374271627073e-05, + "loss": 0.3996, + "step": 34583 + }, + { + "epoch": 1.9322568930357291, + "grad_norm": 0.38816431164741516, + "learning_rate": 3.462572837292694e-05, + "loss": 0.501, + "step": 34584 + }, + { + "epoch": 1.932312763639411, + "grad_norm": 1.019743800163269, + "learning_rate": 3.459771402958314e-05, + "loss": 0.3609, + "step": 34585 + }, + { + "epoch": 1.932368634243093, + "grad_norm": 0.417155385017395, + "learning_rate": 3.4569699686239354e-05, + "loss": 0.3545, + "step": 34586 + }, + { + "epoch": 1.9324245048467747, + "grad_norm": 0.48291099071502686, + "learning_rate": 3.454168534289556e-05, + "loss": 0.4846, + "step": 34587 + }, + { + "epoch": 1.9324803754504567, + "grad_norm": 1.8596795797348022, + "learning_rate": 3.451367099955177e-05, + "loss": 0.321, + "step": 34588 + }, + { + "epoch": 1.9325362460541387, + "grad_norm": 0.38713133335113525, + "learning_rate": 3.448565665620798e-05, + "loss": 0.3897, + "step": 34589 + }, + { + "epoch": 1.9325921166578204, + "grad_norm": 0.3954804539680481, + "learning_rate": 3.445764231286419e-05, + "loss": 0.5012, + "step": 34590 + }, + { + "epoch": 1.9326479872615023, + "grad_norm": 0.38404256105422974, + "learning_rate": 3.442962796952039e-05, + "loss": 0.514, + "step": 34591 + }, + { + "epoch": 1.9327038578651843, + "grad_norm": 0.5269303917884827, + "learning_rate": 3.44016136261766e-05, + "loss": 0.3932, + "step": 34592 + }, + { + "epoch": 1.932759728468866, + "grad_norm": 5.51351261138916, + "learning_rate": 3.437359928283281e-05, + "loss": 0.4203, + "step": 34593 + }, + { + "epoch": 1.932815599072548, + "grad_norm": 0.4895313084125519, + "learning_rate": 3.434558493948902e-05, + "loss": 0.3521, + "step": 34594 + }, + { + "epoch": 1.93287146967623, + "grad_norm": 0.53565514087677, + "learning_rate": 3.431757059614523e-05, + "loss": 0.4385, + "step": 34595 + }, + { + "epoch": 1.9329273402799116, + "grad_norm": 0.39437246322631836, + "learning_rate": 3.428955625280143e-05, + "loss": 0.3919, + "step": 34596 + }, + { + "epoch": 1.9329832108835936, + "grad_norm": 0.8276225328445435, + "learning_rate": 3.426154190945764e-05, + "loss": 0.44, + "step": 34597 + }, + { + "epoch": 1.9330390814872755, + "grad_norm": 0.33253487944602966, + "learning_rate": 3.423352756611385e-05, + "loss": 0.3603, + "step": 34598 + }, + { + "epoch": 1.9330949520909573, + "grad_norm": 0.32541990280151367, + "learning_rate": 3.420551322277006e-05, + "loss": 0.362, + "step": 34599 + }, + { + "epoch": 1.9331508226946392, + "grad_norm": 1.312601089477539, + "learning_rate": 3.417749887942627e-05, + "loss": 0.3618, + "step": 34600 + }, + { + "epoch": 1.9332066932983212, + "grad_norm": 0.29593947529792786, + "learning_rate": 3.414948453608247e-05, + "loss": 0.426, + "step": 34601 + }, + { + "epoch": 1.933262563902003, + "grad_norm": 0.48420363664627075, + "learning_rate": 3.412147019273868e-05, + "loss": 0.4413, + "step": 34602 + }, + { + "epoch": 1.9333184345056849, + "grad_norm": 0.514251708984375, + "learning_rate": 3.409345584939489e-05, + "loss": 0.4488, + "step": 34603 + }, + { + "epoch": 1.9333743051093668, + "grad_norm": 0.4051116108894348, + "learning_rate": 3.40654415060511e-05, + "loss": 0.5299, + "step": 34604 + }, + { + "epoch": 1.9334301757130485, + "grad_norm": 0.3659229874610901, + "learning_rate": 3.403742716270731e-05, + "loss": 0.4316, + "step": 34605 + }, + { + "epoch": 1.9334860463167305, + "grad_norm": 1.670169472694397, + "learning_rate": 3.400941281936352e-05, + "loss": 0.4122, + "step": 34606 + }, + { + "epoch": 1.9335419169204124, + "grad_norm": 0.48653218150138855, + "learning_rate": 3.3981398476019724e-05, + "loss": 0.371, + "step": 34607 + }, + { + "epoch": 1.9335977875240942, + "grad_norm": 0.36534929275512695, + "learning_rate": 3.395338413267593e-05, + "loss": 0.382, + "step": 34608 + }, + { + "epoch": 1.9336536581277761, + "grad_norm": 0.3900347352027893, + "learning_rate": 3.392536978933214e-05, + "loss": 0.4998, + "step": 34609 + }, + { + "epoch": 1.933709528731458, + "grad_norm": 0.4209895133972168, + "learning_rate": 3.389735544598835e-05, + "loss": 0.5853, + "step": 34610 + }, + { + "epoch": 1.9337653993351398, + "grad_norm": 0.5438444018363953, + "learning_rate": 3.386934110264456e-05, + "loss": 0.3009, + "step": 34611 + }, + { + "epoch": 1.9338212699388215, + "grad_norm": 0.6872645616531372, + "learning_rate": 3.384132675930076e-05, + "loss": 0.4113, + "step": 34612 + }, + { + "epoch": 1.9338771405425037, + "grad_norm": 0.3606652021408081, + "learning_rate": 3.381331241595697e-05, + "loss": 0.4585, + "step": 34613 + }, + { + "epoch": 1.9339330111461854, + "grad_norm": 0.41253727674484253, + "learning_rate": 3.378529807261318e-05, + "loss": 0.4522, + "step": 34614 + }, + { + "epoch": 1.9339888817498672, + "grad_norm": 0.40709933638572693, + "learning_rate": 3.375728372926939e-05, + "loss": 0.4153, + "step": 34615 + }, + { + "epoch": 1.9340447523535493, + "grad_norm": 0.4353870153427124, + "learning_rate": 3.3729269385925596e-05, + "loss": 0.5672, + "step": 34616 + }, + { + "epoch": 1.934100622957231, + "grad_norm": 0.3438272476196289, + "learning_rate": 3.37012550425818e-05, + "loss": 0.3242, + "step": 34617 + }, + { + "epoch": 1.9341564935609128, + "grad_norm": 0.48111045360565186, + "learning_rate": 3.367324069923801e-05, + "loss": 0.5411, + "step": 34618 + }, + { + "epoch": 1.9342123641645947, + "grad_norm": 0.2861270606517792, + "learning_rate": 3.364522635589422e-05, + "loss": 0.3439, + "step": 34619 + }, + { + "epoch": 1.9342682347682767, + "grad_norm": 0.4326430559158325, + "learning_rate": 3.3617212012550426e-05, + "loss": 0.443, + "step": 34620 + }, + { + "epoch": 1.9343241053719584, + "grad_norm": 0.3403887152671814, + "learning_rate": 3.358919766920664e-05, + "loss": 0.464, + "step": 34621 + }, + { + "epoch": 1.9343799759756404, + "grad_norm": 0.5921106338500977, + "learning_rate": 3.3561183325862845e-05, + "loss": 0.4069, + "step": 34622 + }, + { + "epoch": 1.9344358465793223, + "grad_norm": 0.7261403203010559, + "learning_rate": 3.353316898251905e-05, + "loss": 0.35, + "step": 34623 + }, + { + "epoch": 1.934491717183004, + "grad_norm": 0.357506662607193, + "learning_rate": 3.3505154639175256e-05, + "loss": 0.4572, + "step": 34624 + }, + { + "epoch": 1.934547587786686, + "grad_norm": 0.37763771414756775, + "learning_rate": 3.347714029583147e-05, + "loss": 0.4049, + "step": 34625 + }, + { + "epoch": 1.934603458390368, + "grad_norm": 0.42851611971855164, + "learning_rate": 3.3449125952487675e-05, + "loss": 0.4341, + "step": 34626 + }, + { + "epoch": 1.9346593289940497, + "grad_norm": 0.41918691992759705, + "learning_rate": 3.342111160914389e-05, + "loss": 0.323, + "step": 34627 + }, + { + "epoch": 1.9347151995977316, + "grad_norm": 0.3371991813182831, + "learning_rate": 3.3393097265800086e-05, + "loss": 0.4261, + "step": 34628 + }, + { + "epoch": 1.9347710702014136, + "grad_norm": 0.3080897033214569, + "learning_rate": 3.33650829224563e-05, + "loss": 0.3824, + "step": 34629 + }, + { + "epoch": 1.9348269408050953, + "grad_norm": 0.41618049144744873, + "learning_rate": 3.3337068579112505e-05, + "loss": 0.4524, + "step": 34630 + }, + { + "epoch": 1.9348828114087773, + "grad_norm": 0.4975675046443939, + "learning_rate": 3.330905423576872e-05, + "loss": 0.5129, + "step": 34631 + }, + { + "epoch": 1.9349386820124592, + "grad_norm": 0.8335966467857361, + "learning_rate": 3.328103989242492e-05, + "loss": 0.455, + "step": 34632 + }, + { + "epoch": 1.934994552616141, + "grad_norm": 0.6304157376289368, + "learning_rate": 3.325302554908113e-05, + "loss": 0.6969, + "step": 34633 + }, + { + "epoch": 1.935050423219823, + "grad_norm": 0.4577392041683197, + "learning_rate": 3.3225011205737335e-05, + "loss": 0.5904, + "step": 34634 + }, + { + "epoch": 1.9351062938235049, + "grad_norm": 0.44391998648643494, + "learning_rate": 3.319699686239355e-05, + "loss": 0.3898, + "step": 34635 + }, + { + "epoch": 1.9351621644271866, + "grad_norm": 0.4142776131629944, + "learning_rate": 3.316898251904975e-05, + "loss": 0.4145, + "step": 34636 + }, + { + "epoch": 1.9352180350308685, + "grad_norm": 0.57902592420578, + "learning_rate": 3.3140968175705966e-05, + "loss": 0.4254, + "step": 34637 + }, + { + "epoch": 1.9352739056345505, + "grad_norm": 1.5755788087844849, + "learning_rate": 3.311295383236217e-05, + "loss": 0.4343, + "step": 34638 + }, + { + "epoch": 1.9353297762382322, + "grad_norm": 0.5186643600463867, + "learning_rate": 3.308493948901838e-05, + "loss": 0.3163, + "step": 34639 + }, + { + "epoch": 1.9353856468419142, + "grad_norm": 1.837416648864746, + "learning_rate": 3.305692514567458e-05, + "loss": 0.4215, + "step": 34640 + }, + { + "epoch": 1.9354415174455961, + "grad_norm": 0.5156635046005249, + "learning_rate": 3.3028910802330796e-05, + "loss": 0.4564, + "step": 34641 + }, + { + "epoch": 1.9354973880492778, + "grad_norm": 0.4906598627567291, + "learning_rate": 3.3000896458987e-05, + "loss": 0.3926, + "step": 34642 + }, + { + "epoch": 1.9355532586529598, + "grad_norm": 0.4306842088699341, + "learning_rate": 3.2972882115643214e-05, + "loss": 0.5016, + "step": 34643 + }, + { + "epoch": 1.9356091292566417, + "grad_norm": 0.38472363352775574, + "learning_rate": 3.294486777229941e-05, + "loss": 0.3046, + "step": 34644 + }, + { + "epoch": 1.9356649998603235, + "grad_norm": 0.4915589988231659, + "learning_rate": 3.2916853428955626e-05, + "loss": 0.5206, + "step": 34645 + }, + { + "epoch": 1.9357208704640052, + "grad_norm": 0.3618522882461548, + "learning_rate": 3.288883908561183e-05, + "loss": 0.3963, + "step": 34646 + }, + { + "epoch": 1.9357767410676874, + "grad_norm": 0.3333253860473633, + "learning_rate": 3.2860824742268044e-05, + "loss": 0.3909, + "step": 34647 + }, + { + "epoch": 1.935832611671369, + "grad_norm": 0.3065474331378937, + "learning_rate": 3.283281039892425e-05, + "loss": 0.2955, + "step": 34648 + }, + { + "epoch": 1.9358884822750508, + "grad_norm": 0.7281541228294373, + "learning_rate": 3.2804796055580456e-05, + "loss": 0.4356, + "step": 34649 + }, + { + "epoch": 1.935944352878733, + "grad_norm": 0.333962082862854, + "learning_rate": 3.277678171223666e-05, + "loss": 0.426, + "step": 34650 + }, + { + "epoch": 1.9360002234824147, + "grad_norm": 0.3872211277484894, + "learning_rate": 3.2748767368892874e-05, + "loss": 0.4993, + "step": 34651 + }, + { + "epoch": 1.9360560940860965, + "grad_norm": 0.629458487033844, + "learning_rate": 3.272075302554908e-05, + "loss": 0.4, + "step": 34652 + }, + { + "epoch": 1.9361119646897784, + "grad_norm": 0.38237905502319336, + "learning_rate": 3.269273868220529e-05, + "loss": 0.3781, + "step": 34653 + }, + { + "epoch": 1.9361678352934604, + "grad_norm": 0.36789780855178833, + "learning_rate": 3.26647243388615e-05, + "loss": 0.3548, + "step": 34654 + }, + { + "epoch": 1.936223705897142, + "grad_norm": 0.3487459719181061, + "learning_rate": 3.2636709995517704e-05, + "loss": 0.4214, + "step": 34655 + }, + { + "epoch": 1.936279576500824, + "grad_norm": 0.4319250285625458, + "learning_rate": 3.260869565217391e-05, + "loss": 0.3739, + "step": 34656 + }, + { + "epoch": 1.936335447104506, + "grad_norm": 1.876110553741455, + "learning_rate": 3.258068130883012e-05, + "loss": 0.3302, + "step": 34657 + }, + { + "epoch": 1.9363913177081877, + "grad_norm": 0.37849903106689453, + "learning_rate": 3.255266696548633e-05, + "loss": 0.4315, + "step": 34658 + }, + { + "epoch": 1.9364471883118697, + "grad_norm": 0.3880673050880432, + "learning_rate": 3.252465262214254e-05, + "loss": 0.3729, + "step": 34659 + }, + { + "epoch": 1.9365030589155516, + "grad_norm": 0.4346785545349121, + "learning_rate": 3.249663827879874e-05, + "loss": 0.4602, + "step": 34660 + }, + { + "epoch": 1.9365589295192334, + "grad_norm": 0.38537999987602234, + "learning_rate": 3.246862393545495e-05, + "loss": 0.3064, + "step": 34661 + }, + { + "epoch": 1.9366148001229153, + "grad_norm": 0.6629894375801086, + "learning_rate": 3.244060959211116e-05, + "loss": 0.5635, + "step": 34662 + }, + { + "epoch": 1.9366706707265973, + "grad_norm": 0.32438865303993225, + "learning_rate": 3.241259524876737e-05, + "loss": 0.4377, + "step": 34663 + }, + { + "epoch": 1.936726541330279, + "grad_norm": 0.37488552927970886, + "learning_rate": 3.238458090542358e-05, + "loss": 0.3738, + "step": 34664 + }, + { + "epoch": 1.936782411933961, + "grad_norm": 0.47687768936157227, + "learning_rate": 3.235656656207978e-05, + "loss": 0.3492, + "step": 34665 + }, + { + "epoch": 1.936838282537643, + "grad_norm": 0.3367181420326233, + "learning_rate": 3.232855221873599e-05, + "loss": 0.4319, + "step": 34666 + }, + { + "epoch": 1.9368941531413246, + "grad_norm": 0.3329147398471832, + "learning_rate": 3.23005378753922e-05, + "loss": 0.3703, + "step": 34667 + }, + { + "epoch": 1.9369500237450066, + "grad_norm": 0.5420816540718079, + "learning_rate": 3.227252353204841e-05, + "loss": 0.5496, + "step": 34668 + }, + { + "epoch": 1.9370058943486885, + "grad_norm": 0.38361990451812744, + "learning_rate": 3.224450918870462e-05, + "loss": 0.4771, + "step": 34669 + }, + { + "epoch": 1.9370617649523703, + "grad_norm": 1.3445786237716675, + "learning_rate": 3.221649484536082e-05, + "loss": 0.4326, + "step": 34670 + }, + { + "epoch": 1.9371176355560522, + "grad_norm": 0.3557622730731964, + "learning_rate": 3.218848050201703e-05, + "loss": 0.425, + "step": 34671 + }, + { + "epoch": 1.9371735061597342, + "grad_norm": 0.45620331168174744, + "learning_rate": 3.2160466158673244e-05, + "loss": 0.5534, + "step": 34672 + }, + { + "epoch": 1.9372293767634159, + "grad_norm": 0.40172481536865234, + "learning_rate": 3.213245181532945e-05, + "loss": 0.4283, + "step": 34673 + }, + { + "epoch": 1.9372852473670978, + "grad_norm": 0.619547426700592, + "learning_rate": 3.210443747198566e-05, + "loss": 0.4262, + "step": 34674 + }, + { + "epoch": 1.9373411179707798, + "grad_norm": 0.428457647562027, + "learning_rate": 3.207642312864187e-05, + "loss": 0.4997, + "step": 34675 + }, + { + "epoch": 1.9373969885744615, + "grad_norm": 0.574296236038208, + "learning_rate": 3.2048408785298074e-05, + "loss": 0.3818, + "step": 34676 + }, + { + "epoch": 1.9374528591781435, + "grad_norm": 0.38033929467201233, + "learning_rate": 3.202039444195428e-05, + "loss": 0.4434, + "step": 34677 + }, + { + "epoch": 1.9375087297818254, + "grad_norm": 0.39014142751693726, + "learning_rate": 3.199238009861049e-05, + "loss": 0.429, + "step": 34678 + }, + { + "epoch": 1.9375646003855072, + "grad_norm": 0.7192628979682922, + "learning_rate": 3.19643657552667e-05, + "loss": 0.6426, + "step": 34679 + }, + { + "epoch": 1.9376204709891889, + "grad_norm": 0.5256057977676392, + "learning_rate": 3.193635141192291e-05, + "loss": 0.307, + "step": 34680 + }, + { + "epoch": 1.937676341592871, + "grad_norm": 5.287993907928467, + "learning_rate": 3.190833706857911e-05, + "loss": 0.3909, + "step": 34681 + }, + { + "epoch": 1.9377322121965528, + "grad_norm": 0.5438193678855896, + "learning_rate": 3.188032272523532e-05, + "loss": 0.5314, + "step": 34682 + }, + { + "epoch": 1.9377880828002345, + "grad_norm": 0.46506422758102417, + "learning_rate": 3.185230838189153e-05, + "loss": 0.3777, + "step": 34683 + }, + { + "epoch": 1.9378439534039167, + "grad_norm": 0.4137091040611267, + "learning_rate": 3.182429403854774e-05, + "loss": 0.4875, + "step": 34684 + }, + { + "epoch": 1.9378998240075984, + "grad_norm": 1.013984203338623, + "learning_rate": 3.1796279695203947e-05, + "loss": 0.428, + "step": 34685 + }, + { + "epoch": 1.9379556946112801, + "grad_norm": 0.41852739453315735, + "learning_rate": 3.176826535186015e-05, + "loss": 0.3995, + "step": 34686 + }, + { + "epoch": 1.938011565214962, + "grad_norm": 6.523377895355225, + "learning_rate": 3.174025100851636e-05, + "loss": 0.303, + "step": 34687 + }, + { + "epoch": 1.938067435818644, + "grad_norm": 0.37497973442077637, + "learning_rate": 3.171223666517257e-05, + "loss": 0.358, + "step": 34688 + }, + { + "epoch": 1.9381233064223258, + "grad_norm": 0.5906387567520142, + "learning_rate": 3.168422232182878e-05, + "loss": 0.3763, + "step": 34689 + }, + { + "epoch": 1.9381791770260077, + "grad_norm": 0.39972934126853943, + "learning_rate": 3.165620797848499e-05, + "loss": 0.3452, + "step": 34690 + }, + { + "epoch": 1.9382350476296897, + "grad_norm": 0.40227776765823364, + "learning_rate": 3.1628193635141195e-05, + "loss": 0.3378, + "step": 34691 + }, + { + "epoch": 1.9382909182333714, + "grad_norm": 0.3728443682193756, + "learning_rate": 3.16001792917974e-05, + "loss": 0.4255, + "step": 34692 + }, + { + "epoch": 1.9383467888370534, + "grad_norm": 0.35628482699394226, + "learning_rate": 3.157216494845361e-05, + "loss": 0.432, + "step": 34693 + }, + { + "epoch": 1.9384026594407353, + "grad_norm": 0.41729605197906494, + "learning_rate": 3.154415060510982e-05, + "loss": 0.4216, + "step": 34694 + }, + { + "epoch": 1.938458530044417, + "grad_norm": 0.45856761932373047, + "learning_rate": 3.1516136261766025e-05, + "loss": 0.3717, + "step": 34695 + }, + { + "epoch": 1.938514400648099, + "grad_norm": 0.3552633225917816, + "learning_rate": 3.148812191842224e-05, + "loss": 0.4333, + "step": 34696 + }, + { + "epoch": 1.938570271251781, + "grad_norm": 2.8172848224639893, + "learning_rate": 3.146010757507844e-05, + "loss": 0.4563, + "step": 34697 + }, + { + "epoch": 1.9386261418554627, + "grad_norm": 0.8128816485404968, + "learning_rate": 3.143209323173465e-05, + "loss": 0.504, + "step": 34698 + }, + { + "epoch": 1.9386820124591446, + "grad_norm": 0.3828098773956299, + "learning_rate": 3.1404078888390855e-05, + "loss": 0.4211, + "step": 34699 + }, + { + "epoch": 1.9387378830628266, + "grad_norm": 0.5098896026611328, + "learning_rate": 3.137606454504707e-05, + "loss": 0.4586, + "step": 34700 + }, + { + "epoch": 1.9387937536665083, + "grad_norm": 0.5432751774787903, + "learning_rate": 3.1348050201703274e-05, + "loss": 0.4716, + "step": 34701 + }, + { + "epoch": 1.9388496242701903, + "grad_norm": 0.4062510132789612, + "learning_rate": 3.132003585835948e-05, + "loss": 0.3021, + "step": 34702 + }, + { + "epoch": 1.9389054948738722, + "grad_norm": 0.6473374962806702, + "learning_rate": 3.1292021515015685e-05, + "loss": 0.3528, + "step": 34703 + }, + { + "epoch": 1.938961365477554, + "grad_norm": 0.6162545084953308, + "learning_rate": 3.12640071716719e-05, + "loss": 0.6643, + "step": 34704 + }, + { + "epoch": 1.9390172360812359, + "grad_norm": 0.34897133708000183, + "learning_rate": 3.1235992828328104e-05, + "loss": 0.3458, + "step": 34705 + }, + { + "epoch": 1.9390731066849178, + "grad_norm": 0.29247811436653137, + "learning_rate": 3.120797848498431e-05, + "loss": 0.2872, + "step": 34706 + }, + { + "epoch": 1.9391289772885996, + "grad_norm": 4.197548866271973, + "learning_rate": 3.117996414164052e-05, + "loss": 0.2902, + "step": 34707 + }, + { + "epoch": 1.9391848478922815, + "grad_norm": 1.1531494855880737, + "learning_rate": 3.115194979829673e-05, + "loss": 0.3464, + "step": 34708 + }, + { + "epoch": 1.9392407184959635, + "grad_norm": 1.5173863172531128, + "learning_rate": 3.1123935454952934e-05, + "loss": 0.3707, + "step": 34709 + }, + { + "epoch": 1.9392965890996452, + "grad_norm": 0.4516448378562927, + "learning_rate": 3.1095921111609146e-05, + "loss": 0.408, + "step": 34710 + }, + { + "epoch": 1.9393524597033271, + "grad_norm": 0.5782080292701721, + "learning_rate": 3.106790676826535e-05, + "loss": 0.4122, + "step": 34711 + }, + { + "epoch": 1.939408330307009, + "grad_norm": 0.9206594228744507, + "learning_rate": 3.103989242492156e-05, + "loss": 0.4309, + "step": 34712 + }, + { + "epoch": 1.9394642009106908, + "grad_norm": 0.7553554177284241, + "learning_rate": 3.101187808157777e-05, + "loss": 0.3876, + "step": 34713 + }, + { + "epoch": 1.9395200715143726, + "grad_norm": 1.0047125816345215, + "learning_rate": 3.0983863738233976e-05, + "loss": 0.5118, + "step": 34714 + }, + { + "epoch": 1.9395759421180547, + "grad_norm": 0.4559648931026459, + "learning_rate": 3.095584939489018e-05, + "loss": 0.5298, + "step": 34715 + }, + { + "epoch": 1.9396318127217365, + "grad_norm": 0.4209601581096649, + "learning_rate": 3.0927835051546395e-05, + "loss": 0.391, + "step": 34716 + }, + { + "epoch": 1.9396876833254182, + "grad_norm": 0.33357444405555725, + "learning_rate": 3.08998207082026e-05, + "loss": 0.3094, + "step": 34717 + }, + { + "epoch": 1.9397435539291004, + "grad_norm": 0.404274582862854, + "learning_rate": 3.0871806364858806e-05, + "loss": 0.3368, + "step": 34718 + }, + { + "epoch": 1.939799424532782, + "grad_norm": 0.3770274817943573, + "learning_rate": 3.084379202151502e-05, + "loss": 0.3368, + "step": 34719 + }, + { + "epoch": 1.9398552951364638, + "grad_norm": 0.33760493993759155, + "learning_rate": 3.0815777678171225e-05, + "loss": 0.3847, + "step": 34720 + }, + { + "epoch": 1.9399111657401458, + "grad_norm": 0.420826256275177, + "learning_rate": 3.078776333482743e-05, + "loss": 0.4267, + "step": 34721 + }, + { + "epoch": 1.9399670363438277, + "grad_norm": 11.889344215393066, + "learning_rate": 3.0759748991483636e-05, + "loss": 0.493, + "step": 34722 + }, + { + "epoch": 1.9400229069475095, + "grad_norm": 0.789463996887207, + "learning_rate": 3.073173464813985e-05, + "loss": 0.4842, + "step": 34723 + }, + { + "epoch": 1.9400787775511914, + "grad_norm": 0.44276943802833557, + "learning_rate": 3.0703720304796055e-05, + "loss": 0.3369, + "step": 34724 + }, + { + "epoch": 1.9401346481548734, + "grad_norm": 0.5104917287826538, + "learning_rate": 3.067570596145226e-05, + "loss": 0.4694, + "step": 34725 + }, + { + "epoch": 1.940190518758555, + "grad_norm": 1.3395295143127441, + "learning_rate": 3.064769161810847e-05, + "loss": 0.4448, + "step": 34726 + }, + { + "epoch": 1.940246389362237, + "grad_norm": 0.3846125900745392, + "learning_rate": 3.061967727476468e-05, + "loss": 0.3844, + "step": 34727 + }, + { + "epoch": 1.940302259965919, + "grad_norm": 0.45875880122184753, + "learning_rate": 3.0591662931420885e-05, + "loss": 0.7053, + "step": 34728 + }, + { + "epoch": 1.9403581305696007, + "grad_norm": 0.5228897929191589, + "learning_rate": 3.05636485880771e-05, + "loss": 0.3599, + "step": 34729 + }, + { + "epoch": 1.9404140011732827, + "grad_norm": 0.5168948173522949, + "learning_rate": 3.05356342447333e-05, + "loss": 0.4158, + "step": 34730 + }, + { + "epoch": 1.9404698717769646, + "grad_norm": 0.5964422821998596, + "learning_rate": 3.0507619901389512e-05, + "loss": 0.4236, + "step": 34731 + }, + { + "epoch": 1.9405257423806463, + "grad_norm": 1.1337026357650757, + "learning_rate": 3.047960555804572e-05, + "loss": 0.4855, + "step": 34732 + }, + { + "epoch": 1.9405816129843283, + "grad_norm": 0.364128440618515, + "learning_rate": 3.0451591214701927e-05, + "loss": 0.3863, + "step": 34733 + }, + { + "epoch": 1.9406374835880102, + "grad_norm": 0.5518068075180054, + "learning_rate": 3.0423576871358137e-05, + "loss": 0.3856, + "step": 34734 + }, + { + "epoch": 1.940693354191692, + "grad_norm": 0.4229276478290558, + "learning_rate": 3.0395562528014346e-05, + "loss": 0.3675, + "step": 34735 + }, + { + "epoch": 1.940749224795374, + "grad_norm": 0.3927251100540161, + "learning_rate": 3.036754818467055e-05, + "loss": 0.4289, + "step": 34736 + }, + { + "epoch": 1.9408050953990559, + "grad_norm": 0.5801718831062317, + "learning_rate": 3.033953384132676e-05, + "loss": 0.4424, + "step": 34737 + }, + { + "epoch": 1.9408609660027376, + "grad_norm": 1.0321851968765259, + "learning_rate": 3.0311519497982967e-05, + "loss": 0.5893, + "step": 34738 + }, + { + "epoch": 1.9409168366064196, + "grad_norm": 0.2449156641960144, + "learning_rate": 3.0283505154639176e-05, + "loss": 0.2769, + "step": 34739 + }, + { + "epoch": 1.9409727072101015, + "grad_norm": 0.5273467898368835, + "learning_rate": 3.0255490811295385e-05, + "loss": 0.3587, + "step": 34740 + }, + { + "epoch": 1.9410285778137832, + "grad_norm": 0.36812007427215576, + "learning_rate": 3.022747646795159e-05, + "loss": 0.3983, + "step": 34741 + }, + { + "epoch": 1.9410844484174652, + "grad_norm": 6.148923397064209, + "learning_rate": 3.01994621246078e-05, + "loss": 0.3221, + "step": 34742 + }, + { + "epoch": 1.9411403190211471, + "grad_norm": 0.5106387138366699, + "learning_rate": 3.017144778126401e-05, + "loss": 0.4728, + "step": 34743 + }, + { + "epoch": 1.9411961896248289, + "grad_norm": 1.192569375038147, + "learning_rate": 3.0143433437920215e-05, + "loss": 0.2948, + "step": 34744 + }, + { + "epoch": 1.9412520602285108, + "grad_norm": 0.3179323971271515, + "learning_rate": 3.0115419094576424e-05, + "loss": 0.3667, + "step": 34745 + }, + { + "epoch": 1.9413079308321928, + "grad_norm": 0.651533305644989, + "learning_rate": 3.008740475123263e-05, + "loss": 0.4353, + "step": 34746 + }, + { + "epoch": 1.9413638014358745, + "grad_norm": 0.8146346211433411, + "learning_rate": 3.005939040788884e-05, + "loss": 0.3457, + "step": 34747 + }, + { + "epoch": 1.9414196720395562, + "grad_norm": 10.19890308380127, + "learning_rate": 3.003137606454505e-05, + "loss": 0.4828, + "step": 34748 + }, + { + "epoch": 1.9414755426432384, + "grad_norm": 0.5724304914474487, + "learning_rate": 3.0003361721201254e-05, + "loss": 0.4286, + "step": 34749 + }, + { + "epoch": 1.9415314132469201, + "grad_norm": 0.3501376509666443, + "learning_rate": 2.9975347377857464e-05, + "loss": 0.3484, + "step": 34750 + }, + { + "epoch": 1.9415872838506019, + "grad_norm": 1.523781657218933, + "learning_rate": 2.994733303451367e-05, + "loss": 0.4216, + "step": 34751 + }, + { + "epoch": 1.941643154454284, + "grad_norm": 0.4999578893184662, + "learning_rate": 2.991931869116988e-05, + "loss": 0.4107, + "step": 34752 + }, + { + "epoch": 1.9416990250579658, + "grad_norm": 0.382930189371109, + "learning_rate": 2.9891304347826088e-05, + "loss": 0.3294, + "step": 34753 + }, + { + "epoch": 1.9417548956616475, + "grad_norm": 0.358915776014328, + "learning_rate": 2.9863290004482294e-05, + "loss": 0.336, + "step": 34754 + }, + { + "epoch": 1.9418107662653294, + "grad_norm": 0.5067926645278931, + "learning_rate": 2.9835275661138503e-05, + "loss": 0.4979, + "step": 34755 + }, + { + "epoch": 1.9418666368690114, + "grad_norm": 0.4337778687477112, + "learning_rate": 2.9807261317794712e-05, + "loss": 0.4591, + "step": 34756 + }, + { + "epoch": 1.9419225074726931, + "grad_norm": 0.6212049126625061, + "learning_rate": 2.9779246974450918e-05, + "loss": 0.4477, + "step": 34757 + }, + { + "epoch": 1.941978378076375, + "grad_norm": 2.154020071029663, + "learning_rate": 2.9751232631107127e-05, + "loss": 0.381, + "step": 34758 + }, + { + "epoch": 1.942034248680057, + "grad_norm": 0.37892788648605347, + "learning_rate": 2.9723218287763333e-05, + "loss": 0.317, + "step": 34759 + }, + { + "epoch": 1.9420901192837388, + "grad_norm": 0.38254061341285706, + "learning_rate": 2.9695203944419542e-05, + "loss": 0.3571, + "step": 34760 + }, + { + "epoch": 1.9421459898874207, + "grad_norm": 0.4328794479370117, + "learning_rate": 2.966718960107575e-05, + "loss": 0.4809, + "step": 34761 + }, + { + "epoch": 1.9422018604911027, + "grad_norm": 0.629059374332428, + "learning_rate": 2.9639175257731957e-05, + "loss": 0.4101, + "step": 34762 + }, + { + "epoch": 1.9422577310947844, + "grad_norm": 0.8671553134918213, + "learning_rate": 2.9611160914388166e-05, + "loss": 0.4044, + "step": 34763 + }, + { + "epoch": 1.9423136016984663, + "grad_norm": 2.245042324066162, + "learning_rate": 2.9583146571044375e-05, + "loss": 0.3728, + "step": 34764 + }, + { + "epoch": 1.9423694723021483, + "grad_norm": 0.48647430539131165, + "learning_rate": 2.955513222770058e-05, + "loss": 0.3961, + "step": 34765 + }, + { + "epoch": 1.94242534290583, + "grad_norm": 2.6163907051086426, + "learning_rate": 2.952711788435679e-05, + "loss": 0.3915, + "step": 34766 + }, + { + "epoch": 1.942481213509512, + "grad_norm": 0.666840136051178, + "learning_rate": 2.9499103541012996e-05, + "loss": 0.4596, + "step": 34767 + }, + { + "epoch": 1.942537084113194, + "grad_norm": 0.49172258377075195, + "learning_rate": 2.9471089197669205e-05, + "loss": 0.3857, + "step": 34768 + }, + { + "epoch": 1.9425929547168757, + "grad_norm": 0.39691397547721863, + "learning_rate": 2.9443074854325415e-05, + "loss": 0.4442, + "step": 34769 + }, + { + "epoch": 1.9426488253205576, + "grad_norm": 0.6125882863998413, + "learning_rate": 2.941506051098162e-05, + "loss": 0.4853, + "step": 34770 + }, + { + "epoch": 1.9427046959242396, + "grad_norm": 0.7291705012321472, + "learning_rate": 2.938704616763783e-05, + "loss": 0.4898, + "step": 34771 + }, + { + "epoch": 1.9427605665279213, + "grad_norm": 0.3058040142059326, + "learning_rate": 2.9359031824294042e-05, + "loss": 0.3932, + "step": 34772 + }, + { + "epoch": 1.9428164371316032, + "grad_norm": 0.4465928375720978, + "learning_rate": 2.9331017480950248e-05, + "loss": 0.3774, + "step": 34773 + }, + { + "epoch": 1.9428723077352852, + "grad_norm": 0.9166715145111084, + "learning_rate": 2.9303003137606457e-05, + "loss": 0.334, + "step": 34774 + }, + { + "epoch": 1.942928178338967, + "grad_norm": 0.3617401123046875, + "learning_rate": 2.9274988794262663e-05, + "loss": 0.4189, + "step": 34775 + }, + { + "epoch": 1.9429840489426489, + "grad_norm": 3.956185817718506, + "learning_rate": 2.9246974450918872e-05, + "loss": 0.335, + "step": 34776 + }, + { + "epoch": 1.9430399195463308, + "grad_norm": 0.30954208970069885, + "learning_rate": 2.921896010757508e-05, + "loss": 0.3831, + "step": 34777 + }, + { + "epoch": 1.9430957901500125, + "grad_norm": 0.47611552476882935, + "learning_rate": 2.9190945764231287e-05, + "loss": 0.5332, + "step": 34778 + }, + { + "epoch": 1.9431516607536945, + "grad_norm": 0.4020768105983734, + "learning_rate": 2.9162931420887497e-05, + "loss": 0.3679, + "step": 34779 + }, + { + "epoch": 1.9432075313573764, + "grad_norm": 0.5139366388320923, + "learning_rate": 2.9134917077543706e-05, + "loss": 0.4067, + "step": 34780 + }, + { + "epoch": 1.9432634019610582, + "grad_norm": 1.624606728553772, + "learning_rate": 2.910690273419991e-05, + "loss": 0.3684, + "step": 34781 + }, + { + "epoch": 1.94331927256474, + "grad_norm": 0.3897649943828583, + "learning_rate": 2.907888839085612e-05, + "loss": 0.4359, + "step": 34782 + }, + { + "epoch": 1.943375143168422, + "grad_norm": 0.48156997561454773, + "learning_rate": 2.9050874047512327e-05, + "loss": 0.4798, + "step": 34783 + }, + { + "epoch": 1.9434310137721038, + "grad_norm": 0.48266011476516724, + "learning_rate": 2.9022859704168536e-05, + "loss": 0.4082, + "step": 34784 + }, + { + "epoch": 1.9434868843757855, + "grad_norm": 0.31548622250556946, + "learning_rate": 2.8994845360824745e-05, + "loss": 0.3507, + "step": 34785 + }, + { + "epoch": 1.9435427549794677, + "grad_norm": 0.4254055321216583, + "learning_rate": 2.896683101748095e-05, + "loss": 0.4587, + "step": 34786 + }, + { + "epoch": 1.9435986255831494, + "grad_norm": 0.40257906913757324, + "learning_rate": 2.893881667413716e-05, + "loss": 0.3912, + "step": 34787 + }, + { + "epoch": 1.9436544961868312, + "grad_norm": 0.3649793565273285, + "learning_rate": 2.891080233079337e-05, + "loss": 0.422, + "step": 34788 + }, + { + "epoch": 1.9437103667905131, + "grad_norm": 0.8793378472328186, + "learning_rate": 2.8882787987449575e-05, + "loss": 0.3496, + "step": 34789 + }, + { + "epoch": 1.943766237394195, + "grad_norm": 0.4674241840839386, + "learning_rate": 2.8854773644105784e-05, + "loss": 0.4503, + "step": 34790 + }, + { + "epoch": 1.9438221079978768, + "grad_norm": 0.6863811016082764, + "learning_rate": 2.882675930076199e-05, + "loss": 0.3208, + "step": 34791 + }, + { + "epoch": 1.9438779786015588, + "grad_norm": 0.46482518315315247, + "learning_rate": 2.87987449574182e-05, + "loss": 0.5181, + "step": 34792 + }, + { + "epoch": 1.9439338492052407, + "grad_norm": 0.38722047209739685, + "learning_rate": 2.877073061407441e-05, + "loss": 0.5007, + "step": 34793 + }, + { + "epoch": 1.9439897198089224, + "grad_norm": 0.6776148080825806, + "learning_rate": 2.8742716270730614e-05, + "loss": 0.413, + "step": 34794 + }, + { + "epoch": 1.9440455904126044, + "grad_norm": 0.43135422468185425, + "learning_rate": 2.8714701927386823e-05, + "loss": 0.4616, + "step": 34795 + }, + { + "epoch": 1.9441014610162863, + "grad_norm": 0.4941811263561249, + "learning_rate": 2.8686687584043033e-05, + "loss": 0.5005, + "step": 34796 + }, + { + "epoch": 1.944157331619968, + "grad_norm": 0.48716306686401367, + "learning_rate": 2.865867324069924e-05, + "loss": 0.4754, + "step": 34797 + }, + { + "epoch": 1.94421320222365, + "grad_norm": 0.33403754234313965, + "learning_rate": 2.8630658897355448e-05, + "loss": 0.3787, + "step": 34798 + }, + { + "epoch": 1.944269072827332, + "grad_norm": 0.7172821760177612, + "learning_rate": 2.8602644554011653e-05, + "loss": 0.4694, + "step": 34799 + }, + { + "epoch": 1.9443249434310137, + "grad_norm": 0.48335596919059753, + "learning_rate": 2.8574630210667863e-05, + "loss": 0.3793, + "step": 34800 + }, + { + "epoch": 1.9443808140346956, + "grad_norm": 0.5470888018608093, + "learning_rate": 2.8546615867324072e-05, + "loss": 0.4299, + "step": 34801 + }, + { + "epoch": 1.9444366846383776, + "grad_norm": 0.5494381189346313, + "learning_rate": 2.8518601523980278e-05, + "loss": 0.4534, + "step": 34802 + }, + { + "epoch": 1.9444925552420593, + "grad_norm": 0.3690171539783478, + "learning_rate": 2.8490587180636487e-05, + "loss": 0.4633, + "step": 34803 + }, + { + "epoch": 1.9445484258457413, + "grad_norm": 0.46607622504234314, + "learning_rate": 2.8462572837292696e-05, + "loss": 0.3735, + "step": 34804 + }, + { + "epoch": 1.9446042964494232, + "grad_norm": 0.38053977489471436, + "learning_rate": 2.8434558493948902e-05, + "loss": 0.3987, + "step": 34805 + }, + { + "epoch": 1.944660167053105, + "grad_norm": 0.41611146926879883, + "learning_rate": 2.840654415060511e-05, + "loss": 0.3747, + "step": 34806 + }, + { + "epoch": 1.944716037656787, + "grad_norm": 0.4140450954437256, + "learning_rate": 2.8378529807261317e-05, + "loss": 0.3999, + "step": 34807 + }, + { + "epoch": 1.9447719082604689, + "grad_norm": 0.6078356504440308, + "learning_rate": 2.8350515463917526e-05, + "loss": 0.5084, + "step": 34808 + }, + { + "epoch": 1.9448277788641506, + "grad_norm": 0.4055252969264984, + "learning_rate": 2.8322501120573735e-05, + "loss": 0.442, + "step": 34809 + }, + { + "epoch": 1.9448836494678325, + "grad_norm": 0.39598679542541504, + "learning_rate": 2.829448677722994e-05, + "loss": 0.3558, + "step": 34810 + }, + { + "epoch": 1.9449395200715145, + "grad_norm": 1.4600489139556885, + "learning_rate": 2.826647243388615e-05, + "loss": 0.6102, + "step": 34811 + }, + { + "epoch": 1.9449953906751962, + "grad_norm": 1.073744297027588, + "learning_rate": 2.823845809054236e-05, + "loss": 0.4468, + "step": 34812 + }, + { + "epoch": 1.945051261278878, + "grad_norm": 0.31486666202545166, + "learning_rate": 2.8210443747198565e-05, + "loss": 0.3341, + "step": 34813 + }, + { + "epoch": 1.9451071318825601, + "grad_norm": 0.4650283455848694, + "learning_rate": 2.8182429403854775e-05, + "loss": 0.3912, + "step": 34814 + }, + { + "epoch": 1.9451630024862419, + "grad_norm": 0.8865023255348206, + "learning_rate": 2.815441506051098e-05, + "loss": 0.3559, + "step": 34815 + }, + { + "epoch": 1.9452188730899236, + "grad_norm": 0.33342787623405457, + "learning_rate": 2.812640071716719e-05, + "loss": 0.4489, + "step": 34816 + }, + { + "epoch": 1.9452747436936058, + "grad_norm": 1.8867014646530151, + "learning_rate": 2.80983863738234e-05, + "loss": 0.3332, + "step": 34817 + }, + { + "epoch": 1.9453306142972875, + "grad_norm": 0.4713660478591919, + "learning_rate": 2.8070372030479605e-05, + "loss": 0.3602, + "step": 34818 + }, + { + "epoch": 1.9453864849009692, + "grad_norm": 0.5616967678070068, + "learning_rate": 2.8042357687135814e-05, + "loss": 0.4488, + "step": 34819 + }, + { + "epoch": 1.9454423555046514, + "grad_norm": 0.4189523458480835, + "learning_rate": 2.8014343343792023e-05, + "loss": 0.3985, + "step": 34820 + }, + { + "epoch": 1.9454982261083331, + "grad_norm": 0.4267442524433136, + "learning_rate": 2.798632900044823e-05, + "loss": 0.4465, + "step": 34821 + }, + { + "epoch": 1.9455540967120148, + "grad_norm": 0.3675283193588257, + "learning_rate": 2.7958314657104438e-05, + "loss": 0.3668, + "step": 34822 + }, + { + "epoch": 1.9456099673156968, + "grad_norm": 0.4180164337158203, + "learning_rate": 2.7930300313760644e-05, + "loss": 0.4392, + "step": 34823 + }, + { + "epoch": 1.9456658379193787, + "grad_norm": 0.39195767045021057, + "learning_rate": 2.7902285970416853e-05, + "loss": 0.3727, + "step": 34824 + }, + { + "epoch": 1.9457217085230605, + "grad_norm": 0.4654119908809662, + "learning_rate": 2.7874271627073062e-05, + "loss": 0.3002, + "step": 34825 + }, + { + "epoch": 1.9457775791267424, + "grad_norm": 0.3843443691730499, + "learning_rate": 2.7846257283729268e-05, + "loss": 0.4777, + "step": 34826 + }, + { + "epoch": 1.9458334497304244, + "grad_norm": 0.3745386600494385, + "learning_rate": 2.7818242940385477e-05, + "loss": 0.4633, + "step": 34827 + }, + { + "epoch": 1.945889320334106, + "grad_norm": 0.4629959762096405, + "learning_rate": 2.7790228597041687e-05, + "loss": 0.5399, + "step": 34828 + }, + { + "epoch": 1.945945190937788, + "grad_norm": 0.38437825441360474, + "learning_rate": 2.7762214253697892e-05, + "loss": 0.3536, + "step": 34829 + }, + { + "epoch": 1.94600106154147, + "grad_norm": 1.059781789779663, + "learning_rate": 2.77341999103541e-05, + "loss": 0.4887, + "step": 34830 + }, + { + "epoch": 1.9460569321451517, + "grad_norm": 0.38688480854034424, + "learning_rate": 2.7706185567010307e-05, + "loss": 0.4247, + "step": 34831 + }, + { + "epoch": 1.9461128027488337, + "grad_norm": 0.4567547142505646, + "learning_rate": 2.7678171223666517e-05, + "loss": 0.3815, + "step": 34832 + }, + { + "epoch": 1.9461686733525156, + "grad_norm": 0.32930490374565125, + "learning_rate": 2.7650156880322726e-05, + "loss": 0.3487, + "step": 34833 + }, + { + "epoch": 1.9462245439561974, + "grad_norm": 0.33316662907600403, + "learning_rate": 2.762214253697893e-05, + "loss": 0.4142, + "step": 34834 + }, + { + "epoch": 1.9462804145598793, + "grad_norm": 0.36287155747413635, + "learning_rate": 2.759412819363514e-05, + "loss": 0.446, + "step": 34835 + }, + { + "epoch": 1.9463362851635613, + "grad_norm": 0.5131442546844482, + "learning_rate": 2.7566113850291347e-05, + "loss": 0.4536, + "step": 34836 + }, + { + "epoch": 1.946392155767243, + "grad_norm": 0.3176761865615845, + "learning_rate": 2.7538099506947556e-05, + "loss": 0.4019, + "step": 34837 + }, + { + "epoch": 1.946448026370925, + "grad_norm": 0.7335029244422913, + "learning_rate": 2.7510085163603765e-05, + "loss": 0.3811, + "step": 34838 + }, + { + "epoch": 1.946503896974607, + "grad_norm": 0.8696825504302979, + "learning_rate": 2.748207082025997e-05, + "loss": 0.315, + "step": 34839 + }, + { + "epoch": 1.9465597675782886, + "grad_norm": 0.8684741854667664, + "learning_rate": 2.7454056476916183e-05, + "loss": 0.5955, + "step": 34840 + }, + { + "epoch": 1.9466156381819706, + "grad_norm": 0.43719249963760376, + "learning_rate": 2.7426042133572393e-05, + "loss": 0.4199, + "step": 34841 + }, + { + "epoch": 1.9466715087856525, + "grad_norm": 0.37638670206069946, + "learning_rate": 2.73980277902286e-05, + "loss": 0.3348, + "step": 34842 + }, + { + "epoch": 1.9467273793893343, + "grad_norm": 0.4539429545402527, + "learning_rate": 2.7370013446884808e-05, + "loss": 0.464, + "step": 34843 + }, + { + "epoch": 1.9467832499930162, + "grad_norm": 0.4451451897621155, + "learning_rate": 2.7341999103541013e-05, + "loss": 0.3478, + "step": 34844 + }, + { + "epoch": 1.9468391205966982, + "grad_norm": 2.101501941680908, + "learning_rate": 2.7313984760197223e-05, + "loss": 0.371, + "step": 34845 + }, + { + "epoch": 1.94689499120038, + "grad_norm": 0.37078526616096497, + "learning_rate": 2.7285970416853432e-05, + "loss": 0.3967, + "step": 34846 + }, + { + "epoch": 1.9469508618040616, + "grad_norm": 0.3792608380317688, + "learning_rate": 2.7257956073509638e-05, + "loss": 0.5618, + "step": 34847 + }, + { + "epoch": 1.9470067324077438, + "grad_norm": 0.3979594111442566, + "learning_rate": 2.7229941730165847e-05, + "loss": 0.3818, + "step": 34848 + }, + { + "epoch": 1.9470626030114255, + "grad_norm": 0.6186492443084717, + "learning_rate": 2.7201927386822056e-05, + "loss": 0.4596, + "step": 34849 + }, + { + "epoch": 1.9471184736151073, + "grad_norm": 0.34125587344169617, + "learning_rate": 2.7173913043478262e-05, + "loss": 0.4353, + "step": 34850 + }, + { + "epoch": 1.9471743442187894, + "grad_norm": 0.3888828456401825, + "learning_rate": 2.714589870013447e-05, + "loss": 0.3419, + "step": 34851 + }, + { + "epoch": 1.9472302148224712, + "grad_norm": 0.49176084995269775, + "learning_rate": 2.7117884356790677e-05, + "loss": 0.4524, + "step": 34852 + }, + { + "epoch": 1.947286085426153, + "grad_norm": 0.389321893453598, + "learning_rate": 2.7089870013446886e-05, + "loss": 0.4093, + "step": 34853 + }, + { + "epoch": 1.947341956029835, + "grad_norm": 0.5707189440727234, + "learning_rate": 2.7061855670103095e-05, + "loss": 0.4459, + "step": 34854 + }, + { + "epoch": 1.9473978266335168, + "grad_norm": 0.3958985209465027, + "learning_rate": 2.70338413267593e-05, + "loss": 0.4104, + "step": 34855 + }, + { + "epoch": 1.9474536972371985, + "grad_norm": 0.4778175354003906, + "learning_rate": 2.700582698341551e-05, + "loss": 0.5436, + "step": 34856 + }, + { + "epoch": 1.9475095678408805, + "grad_norm": 26.285930633544922, + "learning_rate": 2.697781264007172e-05, + "loss": 0.4289, + "step": 34857 + }, + { + "epoch": 1.9475654384445624, + "grad_norm": 1.0443861484527588, + "learning_rate": 2.6949798296727925e-05, + "loss": 0.5225, + "step": 34858 + }, + { + "epoch": 1.9476213090482442, + "grad_norm": 0.4000419080257416, + "learning_rate": 2.6921783953384135e-05, + "loss": 0.3385, + "step": 34859 + }, + { + "epoch": 1.947677179651926, + "grad_norm": 0.8857975602149963, + "learning_rate": 2.689376961004034e-05, + "loss": 0.3246, + "step": 34860 + }, + { + "epoch": 1.947733050255608, + "grad_norm": 0.473580002784729, + "learning_rate": 2.686575526669655e-05, + "loss": 0.4977, + "step": 34861 + }, + { + "epoch": 1.9477889208592898, + "grad_norm": 3.1626105308532715, + "learning_rate": 2.683774092335276e-05, + "loss": 0.5332, + "step": 34862 + }, + { + "epoch": 1.9478447914629717, + "grad_norm": 0.3771395981311798, + "learning_rate": 2.6809726580008965e-05, + "loss": 0.4057, + "step": 34863 + }, + { + "epoch": 1.9479006620666537, + "grad_norm": 0.4667014479637146, + "learning_rate": 2.6781712236665174e-05, + "loss": 0.3254, + "step": 34864 + }, + { + "epoch": 1.9479565326703354, + "grad_norm": 0.5655879974365234, + "learning_rate": 2.6753697893321383e-05, + "loss": 0.3997, + "step": 34865 + }, + { + "epoch": 1.9480124032740174, + "grad_norm": 0.40999114513397217, + "learning_rate": 2.672568354997759e-05, + "loss": 0.4085, + "step": 34866 + }, + { + "epoch": 1.9480682738776993, + "grad_norm": 2.2313339710235596, + "learning_rate": 2.6697669206633798e-05, + "loss": 0.3063, + "step": 34867 + }, + { + "epoch": 1.948124144481381, + "grad_norm": 7.844938278198242, + "learning_rate": 2.6669654863290004e-05, + "loss": 0.4342, + "step": 34868 + }, + { + "epoch": 1.948180015085063, + "grad_norm": 0.3185787498950958, + "learning_rate": 2.6641640519946213e-05, + "loss": 0.4081, + "step": 34869 + }, + { + "epoch": 1.948235885688745, + "grad_norm": 2.0198910236358643, + "learning_rate": 2.6613626176602422e-05, + "loss": 0.3976, + "step": 34870 + }, + { + "epoch": 1.9482917562924267, + "grad_norm": 1.6879161596298218, + "learning_rate": 2.6585611833258628e-05, + "loss": 0.4163, + "step": 34871 + }, + { + "epoch": 1.9483476268961086, + "grad_norm": 1.058830976486206, + "learning_rate": 2.6557597489914837e-05, + "loss": 0.5902, + "step": 34872 + }, + { + "epoch": 1.9484034974997906, + "grad_norm": 0.5358595848083496, + "learning_rate": 2.6529583146571046e-05, + "loss": 0.4216, + "step": 34873 + }, + { + "epoch": 1.9484593681034723, + "grad_norm": 0.37240177392959595, + "learning_rate": 2.6501568803227252e-05, + "loss": 0.5004, + "step": 34874 + }, + { + "epoch": 1.9485152387071543, + "grad_norm": 0.37679100036621094, + "learning_rate": 2.647355445988346e-05, + "loss": 0.3386, + "step": 34875 + }, + { + "epoch": 1.9485711093108362, + "grad_norm": 0.403704971075058, + "learning_rate": 2.6445540116539667e-05, + "loss": 0.4567, + "step": 34876 + }, + { + "epoch": 1.948626979914518, + "grad_norm": 0.5514911413192749, + "learning_rate": 2.6417525773195876e-05, + "loss": 0.5256, + "step": 34877 + }, + { + "epoch": 1.9486828505182, + "grad_norm": 0.4773176908493042, + "learning_rate": 2.6389511429852086e-05, + "loss": 0.3896, + "step": 34878 + }, + { + "epoch": 1.9487387211218818, + "grad_norm": 0.47868093848228455, + "learning_rate": 2.636149708650829e-05, + "loss": 0.5362, + "step": 34879 + }, + { + "epoch": 1.9487945917255636, + "grad_norm": 0.6090668439865112, + "learning_rate": 2.63334827431645e-05, + "loss": 0.6446, + "step": 34880 + }, + { + "epoch": 1.9488504623292453, + "grad_norm": 0.558916449546814, + "learning_rate": 2.630546839982071e-05, + "loss": 0.3079, + "step": 34881 + }, + { + "epoch": 1.9489063329329275, + "grad_norm": 0.32970449328422546, + "learning_rate": 2.6277454056476916e-05, + "loss": 0.3651, + "step": 34882 + }, + { + "epoch": 1.9489622035366092, + "grad_norm": 0.394188791513443, + "learning_rate": 2.6249439713133125e-05, + "loss": 0.3587, + "step": 34883 + }, + { + "epoch": 1.949018074140291, + "grad_norm": 0.47630414366722107, + "learning_rate": 2.622142536978933e-05, + "loss": 0.3976, + "step": 34884 + }, + { + "epoch": 1.949073944743973, + "grad_norm": 0.37334030866622925, + "learning_rate": 2.619341102644554e-05, + "loss": 0.4121, + "step": 34885 + }, + { + "epoch": 1.9491298153476548, + "grad_norm": 0.42327409982681274, + "learning_rate": 2.616539668310175e-05, + "loss": 0.3936, + "step": 34886 + }, + { + "epoch": 1.9491856859513366, + "grad_norm": 0.36684492230415344, + "learning_rate": 2.6137382339757955e-05, + "loss": 0.3326, + "step": 34887 + }, + { + "epoch": 1.9492415565550187, + "grad_norm": 0.34538230299949646, + "learning_rate": 2.6109367996414164e-05, + "loss": 0.3741, + "step": 34888 + }, + { + "epoch": 1.9492974271587005, + "grad_norm": 7.195596694946289, + "learning_rate": 2.6081353653070373e-05, + "loss": 0.4491, + "step": 34889 + }, + { + "epoch": 1.9493532977623822, + "grad_norm": 3.3644537925720215, + "learning_rate": 2.605333930972658e-05, + "loss": 0.4332, + "step": 34890 + }, + { + "epoch": 1.9494091683660641, + "grad_norm": 0.40119922161102295, + "learning_rate": 2.602532496638279e-05, + "loss": 0.5401, + "step": 34891 + }, + { + "epoch": 1.949465038969746, + "grad_norm": 0.5305694341659546, + "learning_rate": 2.5997310623038994e-05, + "loss": 0.4614, + "step": 34892 + }, + { + "epoch": 1.9495209095734278, + "grad_norm": 0.3079487383365631, + "learning_rate": 2.5969296279695203e-05, + "loss": 0.3443, + "step": 34893 + }, + { + "epoch": 1.9495767801771098, + "grad_norm": 0.4830012023448944, + "learning_rate": 2.5941281936351413e-05, + "loss": 0.4197, + "step": 34894 + }, + { + "epoch": 1.9496326507807917, + "grad_norm": 0.47671133279800415, + "learning_rate": 2.591326759300762e-05, + "loss": 0.3755, + "step": 34895 + }, + { + "epoch": 1.9496885213844735, + "grad_norm": 0.5137268304824829, + "learning_rate": 2.5885253249663828e-05, + "loss": 0.3868, + "step": 34896 + }, + { + "epoch": 1.9497443919881554, + "grad_norm": 0.3081226050853729, + "learning_rate": 2.5857238906320037e-05, + "loss": 0.343, + "step": 34897 + }, + { + "epoch": 1.9498002625918374, + "grad_norm": 0.5563020706176758, + "learning_rate": 2.5829224562976243e-05, + "loss": 0.4928, + "step": 34898 + }, + { + "epoch": 1.949856133195519, + "grad_norm": 0.3821716010570526, + "learning_rate": 2.5801210219632452e-05, + "loss": 0.3911, + "step": 34899 + }, + { + "epoch": 1.949912003799201, + "grad_norm": 0.35020169615745544, + "learning_rate": 2.5773195876288658e-05, + "loss": 0.3646, + "step": 34900 + }, + { + "epoch": 1.949967874402883, + "grad_norm": 32.81454086303711, + "learning_rate": 2.5745181532944867e-05, + "loss": 0.3717, + "step": 34901 + }, + { + "epoch": 1.9500237450065647, + "grad_norm": 0.3239165246486664, + "learning_rate": 2.5717167189601076e-05, + "loss": 0.4236, + "step": 34902 + }, + { + "epoch": 1.9500796156102467, + "grad_norm": 0.4158688187599182, + "learning_rate": 2.5689152846257282e-05, + "loss": 0.3968, + "step": 34903 + }, + { + "epoch": 1.9501354862139286, + "grad_norm": 0.4513828158378601, + "learning_rate": 2.566113850291349e-05, + "loss": 0.4527, + "step": 34904 + }, + { + "epoch": 1.9501913568176104, + "grad_norm": 0.4403612017631531, + "learning_rate": 2.56331241595697e-05, + "loss": 0.3842, + "step": 34905 + }, + { + "epoch": 1.9502472274212923, + "grad_norm": 0.4887315034866333, + "learning_rate": 2.5605109816225906e-05, + "loss": 0.4497, + "step": 34906 + }, + { + "epoch": 1.9503030980249743, + "grad_norm": 1.3847177028656006, + "learning_rate": 2.557709547288212e-05, + "loss": 0.3965, + "step": 34907 + }, + { + "epoch": 1.950358968628656, + "grad_norm": 0.4429575800895691, + "learning_rate": 2.5549081129538325e-05, + "loss": 0.4685, + "step": 34908 + }, + { + "epoch": 1.950414839232338, + "grad_norm": 0.43088021874427795, + "learning_rate": 2.5521066786194534e-05, + "loss": 0.2891, + "step": 34909 + }, + { + "epoch": 1.95047070983602, + "grad_norm": 0.5149264335632324, + "learning_rate": 2.5493052442850743e-05, + "loss": 0.3887, + "step": 34910 + }, + { + "epoch": 1.9505265804397016, + "grad_norm": 0.32675665616989136, + "learning_rate": 2.546503809950695e-05, + "loss": 0.3505, + "step": 34911 + }, + { + "epoch": 1.9505824510433836, + "grad_norm": 0.40591108798980713, + "learning_rate": 2.5437023756163158e-05, + "loss": 0.4335, + "step": 34912 + }, + { + "epoch": 1.9506383216470655, + "grad_norm": 0.4386378526687622, + "learning_rate": 2.5409009412819364e-05, + "loss": 0.5124, + "step": 34913 + }, + { + "epoch": 1.9506941922507472, + "grad_norm": 1.0110071897506714, + "learning_rate": 2.5380995069475573e-05, + "loss": 0.4687, + "step": 34914 + }, + { + "epoch": 1.950750062854429, + "grad_norm": 0.4328303337097168, + "learning_rate": 2.5352980726131782e-05, + "loss": 0.4877, + "step": 34915 + }, + { + "epoch": 1.9508059334581112, + "grad_norm": 0.4122275114059448, + "learning_rate": 2.5324966382787988e-05, + "loss": 0.5942, + "step": 34916 + }, + { + "epoch": 1.9508618040617929, + "grad_norm": 0.8111280202865601, + "learning_rate": 2.5296952039444197e-05, + "loss": 0.649, + "step": 34917 + }, + { + "epoch": 1.9509176746654746, + "grad_norm": 0.3586811125278473, + "learning_rate": 2.5268937696100406e-05, + "loss": 0.4532, + "step": 34918 + }, + { + "epoch": 1.9509735452691568, + "grad_norm": 0.5348096489906311, + "learning_rate": 2.5240923352756612e-05, + "loss": 0.3383, + "step": 34919 + }, + { + "epoch": 1.9510294158728385, + "grad_norm": 0.3886922299861908, + "learning_rate": 2.521290900941282e-05, + "loss": 0.439, + "step": 34920 + }, + { + "epoch": 1.9510852864765202, + "grad_norm": 0.33688804507255554, + "learning_rate": 2.5184894666069027e-05, + "loss": 0.3479, + "step": 34921 + }, + { + "epoch": 1.9511411570802024, + "grad_norm": 0.35586196184158325, + "learning_rate": 2.5156880322725236e-05, + "loss": 0.4152, + "step": 34922 + }, + { + "epoch": 1.9511970276838841, + "grad_norm": 3.173177480697632, + "learning_rate": 2.5128865979381446e-05, + "loss": 0.3661, + "step": 34923 + }, + { + "epoch": 1.9512528982875659, + "grad_norm": 0.4203706979751587, + "learning_rate": 2.510085163603765e-05, + "loss": 0.3858, + "step": 34924 + }, + { + "epoch": 1.9513087688912478, + "grad_norm": 0.5637423396110535, + "learning_rate": 2.507283729269386e-05, + "loss": 0.5325, + "step": 34925 + }, + { + "epoch": 1.9513646394949298, + "grad_norm": 0.30118605494499207, + "learning_rate": 2.504482294935007e-05, + "loss": 0.3917, + "step": 34926 + }, + { + "epoch": 1.9514205100986115, + "grad_norm": 0.4587344229221344, + "learning_rate": 2.5016808606006276e-05, + "loss": 0.505, + "step": 34927 + }, + { + "epoch": 1.9514763807022935, + "grad_norm": 0.333673357963562, + "learning_rate": 2.4988794262662485e-05, + "loss": 0.3174, + "step": 34928 + }, + { + "epoch": 1.9515322513059754, + "grad_norm": 0.37995973229408264, + "learning_rate": 2.496077991931869e-05, + "loss": 0.3143, + "step": 34929 + }, + { + "epoch": 1.9515881219096571, + "grad_norm": 0.3554840683937073, + "learning_rate": 2.49327655759749e-05, + "loss": 0.4594, + "step": 34930 + }, + { + "epoch": 1.951643992513339, + "grad_norm": 0.5965527296066284, + "learning_rate": 2.490475123263111e-05, + "loss": 0.4212, + "step": 34931 + }, + { + "epoch": 1.951699863117021, + "grad_norm": 0.3562812805175781, + "learning_rate": 2.4876736889287315e-05, + "loss": 0.3662, + "step": 34932 + }, + { + "epoch": 1.9517557337207028, + "grad_norm": 0.3639959990978241, + "learning_rate": 2.4848722545943524e-05, + "loss": 0.3329, + "step": 34933 + }, + { + "epoch": 1.9518116043243847, + "grad_norm": 0.4197746217250824, + "learning_rate": 2.4820708202599733e-05, + "loss": 0.438, + "step": 34934 + }, + { + "epoch": 1.9518674749280667, + "grad_norm": 0.7016283869743347, + "learning_rate": 2.479269385925594e-05, + "loss": 0.5214, + "step": 34935 + }, + { + "epoch": 1.9519233455317484, + "grad_norm": 0.34985533356666565, + "learning_rate": 2.476467951591215e-05, + "loss": 0.3136, + "step": 34936 + }, + { + "epoch": 1.9519792161354304, + "grad_norm": 0.5642797350883484, + "learning_rate": 2.4736665172568354e-05, + "loss": 0.4826, + "step": 34937 + }, + { + "epoch": 1.9520350867391123, + "grad_norm": 1.1460039615631104, + "learning_rate": 2.4708650829224563e-05, + "loss": 0.4334, + "step": 34938 + }, + { + "epoch": 1.952090957342794, + "grad_norm": 0.36661744117736816, + "learning_rate": 2.4680636485880773e-05, + "loss": 0.4319, + "step": 34939 + }, + { + "epoch": 1.952146827946476, + "grad_norm": 0.3551940619945526, + "learning_rate": 2.465262214253698e-05, + "loss": 0.4005, + "step": 34940 + }, + { + "epoch": 1.952202698550158, + "grad_norm": 1.0496081113815308, + "learning_rate": 2.4624607799193188e-05, + "loss": 0.4535, + "step": 34941 + }, + { + "epoch": 1.9522585691538397, + "grad_norm": 0.3239435851573944, + "learning_rate": 2.4596593455849397e-05, + "loss": 0.3535, + "step": 34942 + }, + { + "epoch": 1.9523144397575216, + "grad_norm": 0.26693853735923767, + "learning_rate": 2.4568579112505603e-05, + "loss": 0.3431, + "step": 34943 + }, + { + "epoch": 1.9523703103612036, + "grad_norm": 0.44413328170776367, + "learning_rate": 2.4540564769161812e-05, + "loss": 0.4871, + "step": 34944 + }, + { + "epoch": 1.9524261809648853, + "grad_norm": 0.4794631600379944, + "learning_rate": 2.4512550425818018e-05, + "loss": 0.3117, + "step": 34945 + }, + { + "epoch": 1.9524820515685672, + "grad_norm": 0.5050163269042969, + "learning_rate": 2.4484536082474227e-05, + "loss": 0.3452, + "step": 34946 + }, + { + "epoch": 1.9525379221722492, + "grad_norm": 0.3751164376735687, + "learning_rate": 2.4456521739130436e-05, + "loss": 0.4659, + "step": 34947 + }, + { + "epoch": 1.952593792775931, + "grad_norm": 0.452149361371994, + "learning_rate": 2.4428507395786642e-05, + "loss": 0.5245, + "step": 34948 + }, + { + "epoch": 1.9526496633796127, + "grad_norm": 0.9857299327850342, + "learning_rate": 2.440049305244285e-05, + "loss": 0.556, + "step": 34949 + }, + { + "epoch": 1.9527055339832948, + "grad_norm": 0.3957718014717102, + "learning_rate": 2.437247870909906e-05, + "loss": 0.3757, + "step": 34950 + }, + { + "epoch": 1.9527614045869766, + "grad_norm": 0.45053553581237793, + "learning_rate": 2.4344464365755266e-05, + "loss": 0.3875, + "step": 34951 + }, + { + "epoch": 1.9528172751906583, + "grad_norm": 0.5093410015106201, + "learning_rate": 2.4316450022411475e-05, + "loss": 0.6208, + "step": 34952 + }, + { + "epoch": 1.9528731457943405, + "grad_norm": 4.762423515319824, + "learning_rate": 2.428843567906768e-05, + "loss": 0.4244, + "step": 34953 + }, + { + "epoch": 1.9529290163980222, + "grad_norm": 0.956049382686615, + "learning_rate": 2.426042133572389e-05, + "loss": 0.3829, + "step": 34954 + }, + { + "epoch": 1.952984887001704, + "grad_norm": 0.40768206119537354, + "learning_rate": 2.42324069923801e-05, + "loss": 0.3451, + "step": 34955 + }, + { + "epoch": 1.953040757605386, + "grad_norm": 0.9147008657455444, + "learning_rate": 2.4204392649036305e-05, + "loss": 0.4371, + "step": 34956 + }, + { + "epoch": 1.9530966282090678, + "grad_norm": 0.39451998472213745, + "learning_rate": 2.4176378305692515e-05, + "loss": 0.5663, + "step": 34957 + }, + { + "epoch": 1.9531524988127495, + "grad_norm": 1.36188805103302, + "learning_rate": 2.4148363962348724e-05, + "loss": 0.367, + "step": 34958 + }, + { + "epoch": 1.9532083694164315, + "grad_norm": 0.5149770975112915, + "learning_rate": 2.412034961900493e-05, + "loss": 0.5208, + "step": 34959 + }, + { + "epoch": 1.9532642400201135, + "grad_norm": 0.37218016386032104, + "learning_rate": 2.409233527566114e-05, + "loss": 0.4248, + "step": 34960 + }, + { + "epoch": 1.9533201106237952, + "grad_norm": 0.6750142574310303, + "learning_rate": 2.4064320932317345e-05, + "loss": 0.3799, + "step": 34961 + }, + { + "epoch": 1.9533759812274771, + "grad_norm": 0.4582483172416687, + "learning_rate": 2.4036306588973554e-05, + "loss": 0.4055, + "step": 34962 + }, + { + "epoch": 1.953431851831159, + "grad_norm": 0.6735495328903198, + "learning_rate": 2.4008292245629763e-05, + "loss": 0.4726, + "step": 34963 + }, + { + "epoch": 1.9534877224348408, + "grad_norm": 0.6086174845695496, + "learning_rate": 2.398027790228597e-05, + "loss": 0.432, + "step": 34964 + }, + { + "epoch": 1.9535435930385228, + "grad_norm": 0.3633480966091156, + "learning_rate": 2.3952263558942178e-05, + "loss": 0.4097, + "step": 34965 + }, + { + "epoch": 1.9535994636422047, + "grad_norm": 1.196553111076355, + "learning_rate": 2.3924249215598387e-05, + "loss": 0.4021, + "step": 34966 + }, + { + "epoch": 1.9536553342458864, + "grad_norm": 0.4171019494533539, + "learning_rate": 2.3896234872254593e-05, + "loss": 0.393, + "step": 34967 + }, + { + "epoch": 1.9537112048495684, + "grad_norm": 0.32389479875564575, + "learning_rate": 2.3868220528910802e-05, + "loss": 0.3597, + "step": 34968 + }, + { + "epoch": 1.9537670754532503, + "grad_norm": 1.6136513948440552, + "learning_rate": 2.3840206185567008e-05, + "loss": 0.4139, + "step": 34969 + }, + { + "epoch": 1.953822946056932, + "grad_norm": 0.5314156413078308, + "learning_rate": 2.3812191842223217e-05, + "loss": 0.2722, + "step": 34970 + }, + { + "epoch": 1.953878816660614, + "grad_norm": 0.41440919041633606, + "learning_rate": 2.3784177498879426e-05, + "loss": 0.5017, + "step": 34971 + }, + { + "epoch": 1.953934687264296, + "grad_norm": 0.3552356958389282, + "learning_rate": 2.3756163155535632e-05, + "loss": 0.3934, + "step": 34972 + }, + { + "epoch": 1.9539905578679777, + "grad_norm": 0.4760270118713379, + "learning_rate": 2.372814881219184e-05, + "loss": 0.4416, + "step": 34973 + }, + { + "epoch": 1.9540464284716597, + "grad_norm": 0.4790276885032654, + "learning_rate": 2.370013446884805e-05, + "loss": 0.4988, + "step": 34974 + }, + { + "epoch": 1.9541022990753416, + "grad_norm": 0.3589388132095337, + "learning_rate": 2.3672120125504256e-05, + "loss": 0.4585, + "step": 34975 + }, + { + "epoch": 1.9541581696790233, + "grad_norm": 1.310416340827942, + "learning_rate": 2.364410578216047e-05, + "loss": 0.5036, + "step": 34976 + }, + { + "epoch": 1.9542140402827053, + "grad_norm": 0.5084202885627747, + "learning_rate": 2.3616091438816675e-05, + "loss": 0.4274, + "step": 34977 + }, + { + "epoch": 1.9542699108863872, + "grad_norm": 0.46654683351516724, + "learning_rate": 2.3588077095472884e-05, + "loss": 0.4965, + "step": 34978 + }, + { + "epoch": 1.954325781490069, + "grad_norm": 0.5494911074638367, + "learning_rate": 2.3560062752129093e-05, + "loss": 0.4698, + "step": 34979 + }, + { + "epoch": 1.954381652093751, + "grad_norm": 0.8865321278572083, + "learning_rate": 2.35320484087853e-05, + "loss": 0.3985, + "step": 34980 + }, + { + "epoch": 1.9544375226974329, + "grad_norm": 0.31925293803215027, + "learning_rate": 2.3504034065441508e-05, + "loss": 0.348, + "step": 34981 + }, + { + "epoch": 1.9544933933011146, + "grad_norm": 0.5578247904777527, + "learning_rate": 2.3476019722097717e-05, + "loss": 0.3909, + "step": 34982 + }, + { + "epoch": 1.9545492639047963, + "grad_norm": 0.3190237283706665, + "learning_rate": 2.3448005378753923e-05, + "loss": 0.3118, + "step": 34983 + }, + { + "epoch": 1.9546051345084785, + "grad_norm": 0.43295907974243164, + "learning_rate": 2.3419991035410133e-05, + "loss": 0.3583, + "step": 34984 + }, + { + "epoch": 1.9546610051121602, + "grad_norm": 0.39791184663772583, + "learning_rate": 2.339197669206634e-05, + "loss": 0.3722, + "step": 34985 + }, + { + "epoch": 1.954716875715842, + "grad_norm": 0.6071617603302002, + "learning_rate": 2.3363962348722548e-05, + "loss": 0.3213, + "step": 34986 + }, + { + "epoch": 1.9547727463195241, + "grad_norm": 0.37147054076194763, + "learning_rate": 2.3335948005378757e-05, + "loss": 0.3685, + "step": 34987 + }, + { + "epoch": 1.9548286169232059, + "grad_norm": 0.41310957074165344, + "learning_rate": 2.3307933662034963e-05, + "loss": 0.467, + "step": 34988 + }, + { + "epoch": 1.9548844875268876, + "grad_norm": 0.4707541763782501, + "learning_rate": 2.3279919318691172e-05, + "loss": 0.3891, + "step": 34989 + }, + { + "epoch": 1.9549403581305698, + "grad_norm": 0.429647296667099, + "learning_rate": 2.325190497534738e-05, + "loss": 0.3105, + "step": 34990 + }, + { + "epoch": 1.9549962287342515, + "grad_norm": 0.7474184632301331, + "learning_rate": 2.3223890632003587e-05, + "loss": 0.4932, + "step": 34991 + }, + { + "epoch": 1.9550520993379332, + "grad_norm": 0.38221192359924316, + "learning_rate": 2.3195876288659796e-05, + "loss": 0.3928, + "step": 34992 + }, + { + "epoch": 1.9551079699416152, + "grad_norm": 0.720056414604187, + "learning_rate": 2.3167861945316002e-05, + "loss": 0.3327, + "step": 34993 + }, + { + "epoch": 1.9551638405452971, + "grad_norm": 0.354531854391098, + "learning_rate": 2.313984760197221e-05, + "loss": 0.3801, + "step": 34994 + }, + { + "epoch": 1.9552197111489789, + "grad_norm": 0.3993018865585327, + "learning_rate": 2.311183325862842e-05, + "loss": 0.3678, + "step": 34995 + }, + { + "epoch": 1.9552755817526608, + "grad_norm": 0.4871785044670105, + "learning_rate": 2.3083818915284626e-05, + "loss": 0.3411, + "step": 34996 + }, + { + "epoch": 1.9553314523563428, + "grad_norm": 0.4915942847728729, + "learning_rate": 2.3055804571940835e-05, + "loss": 0.3948, + "step": 34997 + }, + { + "epoch": 1.9553873229600245, + "grad_norm": 1.3353888988494873, + "learning_rate": 2.302779022859704e-05, + "loss": 0.3883, + "step": 34998 + }, + { + "epoch": 1.9554431935637064, + "grad_norm": 0.6307648420333862, + "learning_rate": 2.299977588525325e-05, + "loss": 0.4351, + "step": 34999 + }, + { + "epoch": 1.9554990641673884, + "grad_norm": 0.8194971680641174, + "learning_rate": 2.297176154190946e-05, + "loss": 0.3606, + "step": 35000 + }, + { + "epoch": 1.9554990641673884, + "eval_cer": 0.08340698067711913, + "eval_loss": 0.31193068623542786, + "eval_runtime": 56.2867, + "eval_samples_per_second": 80.623, + "eval_steps_per_second": 5.046, + "eval_wer": 0.33183430630429317, + "step": 35000 + }, + { + "epoch": 1.9555549347710701, + "grad_norm": 0.3775785565376282, + "learning_rate": 2.2943747198565665e-05, + "loss": 0.3667, + "step": 35001 + }, + { + "epoch": 1.955610805374752, + "grad_norm": 0.3289203941822052, + "learning_rate": 2.2915732855221874e-05, + "loss": 0.3833, + "step": 35002 + }, + { + "epoch": 1.955666675978434, + "grad_norm": 0.3449989855289459, + "learning_rate": 2.2887718511878084e-05, + "loss": 0.4464, + "step": 35003 + }, + { + "epoch": 1.9557225465821158, + "grad_norm": 9.516732215881348, + "learning_rate": 2.285970416853429e-05, + "loss": 0.4537, + "step": 35004 + }, + { + "epoch": 1.9557784171857977, + "grad_norm": 0.41750702261924744, + "learning_rate": 2.28316898251905e-05, + "loss": 0.3533, + "step": 35005 + }, + { + "epoch": 1.9558342877894797, + "grad_norm": 0.5086398124694824, + "learning_rate": 2.2803675481846704e-05, + "loss": 0.4353, + "step": 35006 + }, + { + "epoch": 1.9558901583931614, + "grad_norm": 1.5038363933563232, + "learning_rate": 2.2775661138502914e-05, + "loss": 0.3685, + "step": 35007 + }, + { + "epoch": 1.9559460289968433, + "grad_norm": 0.33740127086639404, + "learning_rate": 2.2747646795159123e-05, + "loss": 0.3202, + "step": 35008 + }, + { + "epoch": 1.9560018996005253, + "grad_norm": 0.6700924038887024, + "learning_rate": 2.271963245181533e-05, + "loss": 0.6072, + "step": 35009 + }, + { + "epoch": 1.956057770204207, + "grad_norm": 1.259704351425171, + "learning_rate": 2.2691618108471538e-05, + "loss": 0.5257, + "step": 35010 + }, + { + "epoch": 1.956113640807889, + "grad_norm": 0.3127596378326416, + "learning_rate": 2.2663603765127747e-05, + "loss": 0.3936, + "step": 35011 + }, + { + "epoch": 1.956169511411571, + "grad_norm": 1.2233704328536987, + "learning_rate": 2.2635589421783953e-05, + "loss": 0.4902, + "step": 35012 + }, + { + "epoch": 1.9562253820152526, + "grad_norm": 0.38112300634384155, + "learning_rate": 2.2607575078440162e-05, + "loss": 0.3954, + "step": 35013 + }, + { + "epoch": 1.9562812526189346, + "grad_norm": 2.8314099311828613, + "learning_rate": 2.2579560735096368e-05, + "loss": 0.3859, + "step": 35014 + }, + { + "epoch": 1.9563371232226165, + "grad_norm": 0.5044242143630981, + "learning_rate": 2.2551546391752577e-05, + "loss": 0.5139, + "step": 35015 + }, + { + "epoch": 1.9563929938262983, + "grad_norm": 0.3469448387622833, + "learning_rate": 2.2523532048408786e-05, + "loss": 0.4391, + "step": 35016 + }, + { + "epoch": 1.95644886442998, + "grad_norm": 0.38662752509117126, + "learning_rate": 2.2495517705064992e-05, + "loss": 0.4736, + "step": 35017 + }, + { + "epoch": 1.9565047350336622, + "grad_norm": 0.5046682953834534, + "learning_rate": 2.24675033617212e-05, + "loss": 0.498, + "step": 35018 + }, + { + "epoch": 1.956560605637344, + "grad_norm": 0.3607742488384247, + "learning_rate": 2.243948901837741e-05, + "loss": 0.4167, + "step": 35019 + }, + { + "epoch": 1.9566164762410256, + "grad_norm": 0.7041375637054443, + "learning_rate": 2.2411474675033616e-05, + "loss": 0.4086, + "step": 35020 + }, + { + "epoch": 1.9566723468447078, + "grad_norm": 0.3174569010734558, + "learning_rate": 2.2383460331689826e-05, + "loss": 0.3806, + "step": 35021 + }, + { + "epoch": 1.9567282174483895, + "grad_norm": 0.3639940321445465, + "learning_rate": 2.235544598834603e-05, + "loss": 0.3149, + "step": 35022 + }, + { + "epoch": 1.9567840880520713, + "grad_norm": 0.40199390053749084, + "learning_rate": 2.232743164500224e-05, + "loss": 0.407, + "step": 35023 + }, + { + "epoch": 1.9568399586557532, + "grad_norm": 0.4595865309238434, + "learning_rate": 2.229941730165845e-05, + "loss": 0.4256, + "step": 35024 + }, + { + "epoch": 1.9568958292594352, + "grad_norm": 0.2907261550426483, + "learning_rate": 2.2271402958314656e-05, + "loss": 0.3516, + "step": 35025 + }, + { + "epoch": 1.956951699863117, + "grad_norm": 0.6580731868743896, + "learning_rate": 2.2243388614970865e-05, + "loss": 0.5045, + "step": 35026 + }, + { + "epoch": 1.9570075704667989, + "grad_norm": 0.43950381875038147, + "learning_rate": 2.2215374271627074e-05, + "loss": 0.391, + "step": 35027 + }, + { + "epoch": 1.9570634410704808, + "grad_norm": 0.5050404667854309, + "learning_rate": 2.218735992828328e-05, + "loss": 0.2826, + "step": 35028 + }, + { + "epoch": 1.9571193116741625, + "grad_norm": 4.582131385803223, + "learning_rate": 2.215934558493949e-05, + "loss": 0.4375, + "step": 35029 + }, + { + "epoch": 1.9571751822778445, + "grad_norm": 0.4098758399486542, + "learning_rate": 2.2131331241595695e-05, + "loss": 0.334, + "step": 35030 + }, + { + "epoch": 1.9572310528815264, + "grad_norm": 0.3906839191913605, + "learning_rate": 2.2103316898251904e-05, + "loss": 0.3705, + "step": 35031 + }, + { + "epoch": 1.9572869234852082, + "grad_norm": 0.6041945219039917, + "learning_rate": 2.2075302554908113e-05, + "loss": 0.3641, + "step": 35032 + }, + { + "epoch": 1.9573427940888901, + "grad_norm": 0.5342503190040588, + "learning_rate": 2.204728821156432e-05, + "loss": 0.4389, + "step": 35033 + }, + { + "epoch": 1.957398664692572, + "grad_norm": 0.4767759144306183, + "learning_rate": 2.201927386822053e-05, + "loss": 0.5709, + "step": 35034 + }, + { + "epoch": 1.9574545352962538, + "grad_norm": 3.7190194129943848, + "learning_rate": 2.1991259524876738e-05, + "loss": 0.2858, + "step": 35035 + }, + { + "epoch": 1.9575104058999357, + "grad_norm": 0.9699122309684753, + "learning_rate": 2.1963245181532943e-05, + "loss": 0.3624, + "step": 35036 + }, + { + "epoch": 1.9575662765036177, + "grad_norm": 0.41158175468444824, + "learning_rate": 2.1935230838189153e-05, + "loss": 0.4935, + "step": 35037 + }, + { + "epoch": 1.9576221471072994, + "grad_norm": 0.5035500526428223, + "learning_rate": 2.190721649484536e-05, + "loss": 0.4622, + "step": 35038 + }, + { + "epoch": 1.9576780177109814, + "grad_norm": 0.6054421067237854, + "learning_rate": 2.1879202151501568e-05, + "loss": 0.5059, + "step": 35039 + }, + { + "epoch": 1.9577338883146633, + "grad_norm": 0.3142821788787842, + "learning_rate": 2.1851187808157777e-05, + "loss": 0.3703, + "step": 35040 + }, + { + "epoch": 1.957789758918345, + "grad_norm": 0.4178810119628906, + "learning_rate": 2.1823173464813983e-05, + "loss": 0.426, + "step": 35041 + }, + { + "epoch": 1.957845629522027, + "grad_norm": 0.3914826810359955, + "learning_rate": 2.1795159121470192e-05, + "loss": 0.3669, + "step": 35042 + }, + { + "epoch": 1.957901500125709, + "grad_norm": 5.555047988891602, + "learning_rate": 2.1767144778126404e-05, + "loss": 0.3511, + "step": 35043 + }, + { + "epoch": 1.9579573707293907, + "grad_norm": 0.43227049708366394, + "learning_rate": 2.173913043478261e-05, + "loss": 0.4098, + "step": 35044 + }, + { + "epoch": 1.9580132413330726, + "grad_norm": 0.43777430057525635, + "learning_rate": 2.171111609143882e-05, + "loss": 0.4484, + "step": 35045 + }, + { + "epoch": 1.9580691119367546, + "grad_norm": 0.3623625636100769, + "learning_rate": 2.1683101748095025e-05, + "loss": 0.4676, + "step": 35046 + }, + { + "epoch": 1.9581249825404363, + "grad_norm": 0.7942073345184326, + "learning_rate": 2.1655087404751234e-05, + "loss": 0.4832, + "step": 35047 + }, + { + "epoch": 1.9581808531441183, + "grad_norm": 0.44222116470336914, + "learning_rate": 2.1627073061407444e-05, + "loss": 0.4224, + "step": 35048 + }, + { + "epoch": 1.9582367237478002, + "grad_norm": 0.892195999622345, + "learning_rate": 2.159905871806365e-05, + "loss": 0.3904, + "step": 35049 + }, + { + "epoch": 1.958292594351482, + "grad_norm": 0.31217461824417114, + "learning_rate": 2.157104437471986e-05, + "loss": 0.3381, + "step": 35050 + }, + { + "epoch": 1.9583484649551637, + "grad_norm": 0.4661027789115906, + "learning_rate": 2.1543030031376068e-05, + "loss": 0.4312, + "step": 35051 + }, + { + "epoch": 1.9584043355588459, + "grad_norm": 0.37676262855529785, + "learning_rate": 2.1515015688032274e-05, + "loss": 0.3533, + "step": 35052 + }, + { + "epoch": 1.9584602061625276, + "grad_norm": 1.3164879083633423, + "learning_rate": 2.1487001344688483e-05, + "loss": 0.448, + "step": 35053 + }, + { + "epoch": 1.9585160767662093, + "grad_norm": 0.4110380709171295, + "learning_rate": 2.145898700134469e-05, + "loss": 0.5033, + "step": 35054 + }, + { + "epoch": 1.9585719473698915, + "grad_norm": 0.3721495270729065, + "learning_rate": 2.1430972658000898e-05, + "loss": 0.397, + "step": 35055 + }, + { + "epoch": 1.9586278179735732, + "grad_norm": 1.0114713907241821, + "learning_rate": 2.1402958314657107e-05, + "loss": 0.3881, + "step": 35056 + }, + { + "epoch": 1.958683688577255, + "grad_norm": 0.5304999351501465, + "learning_rate": 2.1374943971313313e-05, + "loss": 0.4173, + "step": 35057 + }, + { + "epoch": 1.958739559180937, + "grad_norm": 0.47871577739715576, + "learning_rate": 2.1346929627969522e-05, + "loss": 0.3096, + "step": 35058 + }, + { + "epoch": 1.9587954297846188, + "grad_norm": 0.49732112884521484, + "learning_rate": 2.131891528462573e-05, + "loss": 0.4367, + "step": 35059 + }, + { + "epoch": 1.9588513003883006, + "grad_norm": 0.37721410393714905, + "learning_rate": 2.1290900941281937e-05, + "loss": 0.3433, + "step": 35060 + }, + { + "epoch": 1.9589071709919825, + "grad_norm": 0.8509459495544434, + "learning_rate": 2.1262886597938146e-05, + "loss": 0.426, + "step": 35061 + }, + { + "epoch": 1.9589630415956645, + "grad_norm": 0.39213088154792786, + "learning_rate": 2.1234872254594352e-05, + "loss": 0.3376, + "step": 35062 + }, + { + "epoch": 1.9590189121993462, + "grad_norm": 0.338619589805603, + "learning_rate": 2.120685791125056e-05, + "loss": 0.4536, + "step": 35063 + }, + { + "epoch": 1.9590747828030282, + "grad_norm": 0.9875851273536682, + "learning_rate": 2.117884356790677e-05, + "loss": 0.4891, + "step": 35064 + }, + { + "epoch": 1.95913065340671, + "grad_norm": 0.4755150079727173, + "learning_rate": 2.1150829224562976e-05, + "loss": 0.4028, + "step": 35065 + }, + { + "epoch": 1.9591865240103918, + "grad_norm": 0.4832228422164917, + "learning_rate": 2.1122814881219186e-05, + "loss": 0.4203, + "step": 35066 + }, + { + "epoch": 1.9592423946140738, + "grad_norm": 0.620010256767273, + "learning_rate": 2.1094800537875395e-05, + "loss": 0.4693, + "step": 35067 + }, + { + "epoch": 1.9592982652177557, + "grad_norm": 0.4208616018295288, + "learning_rate": 2.10667861945316e-05, + "loss": 0.5742, + "step": 35068 + }, + { + "epoch": 1.9593541358214375, + "grad_norm": 0.4218832552433014, + "learning_rate": 2.103877185118781e-05, + "loss": 0.4798, + "step": 35069 + }, + { + "epoch": 1.9594100064251194, + "grad_norm": 1.094462513923645, + "learning_rate": 2.1010757507844016e-05, + "loss": 0.3846, + "step": 35070 + }, + { + "epoch": 1.9594658770288014, + "grad_norm": 0.5039265751838684, + "learning_rate": 2.0982743164500225e-05, + "loss": 0.3689, + "step": 35071 + }, + { + "epoch": 1.959521747632483, + "grad_norm": 3.9647178649902344, + "learning_rate": 2.0954728821156434e-05, + "loss": 0.4959, + "step": 35072 + }, + { + "epoch": 1.959577618236165, + "grad_norm": 0.594390869140625, + "learning_rate": 2.092671447781264e-05, + "loss": 0.4175, + "step": 35073 + }, + { + "epoch": 1.959633488839847, + "grad_norm": 0.8682819604873657, + "learning_rate": 2.089870013446885e-05, + "loss": 0.3903, + "step": 35074 + }, + { + "epoch": 1.9596893594435287, + "grad_norm": 0.4387446343898773, + "learning_rate": 2.0870685791125058e-05, + "loss": 0.4805, + "step": 35075 + }, + { + "epoch": 1.9597452300472107, + "grad_norm": 0.40756654739379883, + "learning_rate": 2.0842671447781264e-05, + "loss": 0.4236, + "step": 35076 + }, + { + "epoch": 1.9598011006508926, + "grad_norm": 0.45161959528923035, + "learning_rate": 2.0814657104437473e-05, + "loss": 0.3219, + "step": 35077 + }, + { + "epoch": 1.9598569712545744, + "grad_norm": 0.5558258891105652, + "learning_rate": 2.078664276109368e-05, + "loss": 0.3526, + "step": 35078 + }, + { + "epoch": 1.9599128418582563, + "grad_norm": 0.39413022994995117, + "learning_rate": 2.0758628417749888e-05, + "loss": 0.4562, + "step": 35079 + }, + { + "epoch": 1.9599687124619383, + "grad_norm": 0.5307700634002686, + "learning_rate": 2.0730614074406097e-05, + "loss": 0.463, + "step": 35080 + }, + { + "epoch": 1.96002458306562, + "grad_norm": 0.9359146952629089, + "learning_rate": 2.0702599731062303e-05, + "loss": 0.4712, + "step": 35081 + }, + { + "epoch": 1.960080453669302, + "grad_norm": 0.31991592049598694, + "learning_rate": 2.0674585387718512e-05, + "loss": 0.3325, + "step": 35082 + }, + { + "epoch": 1.960136324272984, + "grad_norm": 1.06733238697052, + "learning_rate": 2.0646571044374718e-05, + "loss": 0.3859, + "step": 35083 + }, + { + "epoch": 1.9601921948766656, + "grad_norm": 0.47502920031547546, + "learning_rate": 2.0618556701030927e-05, + "loss": 0.3963, + "step": 35084 + }, + { + "epoch": 1.9602480654803474, + "grad_norm": 3.0690722465515137, + "learning_rate": 2.0590542357687137e-05, + "loss": 0.3412, + "step": 35085 + }, + { + "epoch": 1.9603039360840295, + "grad_norm": 1.093799352645874, + "learning_rate": 2.0562528014343343e-05, + "loss": 0.4022, + "step": 35086 + }, + { + "epoch": 1.9603598066877113, + "grad_norm": 0.5045552849769592, + "learning_rate": 2.0534513670999552e-05, + "loss": 0.5263, + "step": 35087 + }, + { + "epoch": 1.960415677291393, + "grad_norm": 0.35828492045402527, + "learning_rate": 2.050649932765576e-05, + "loss": 0.3275, + "step": 35088 + }, + { + "epoch": 1.9604715478950752, + "grad_norm": 0.4994213283061981, + "learning_rate": 2.0478484984311967e-05, + "loss": 0.3225, + "step": 35089 + }, + { + "epoch": 1.960527418498757, + "grad_norm": 0.37657687067985535, + "learning_rate": 2.0450470640968176e-05, + "loss": 0.3253, + "step": 35090 + }, + { + "epoch": 1.9605832891024386, + "grad_norm": 0.5540342926979065, + "learning_rate": 2.0422456297624382e-05, + "loss": 0.5671, + "step": 35091 + }, + { + "epoch": 1.9606391597061206, + "grad_norm": 0.5734858512878418, + "learning_rate": 2.039444195428059e-05, + "loss": 0.3701, + "step": 35092 + }, + { + "epoch": 1.9606950303098025, + "grad_norm": 0.4539072811603546, + "learning_rate": 2.03664276109368e-05, + "loss": 0.3504, + "step": 35093 + }, + { + "epoch": 1.9607509009134843, + "grad_norm": 0.41693365573883057, + "learning_rate": 2.0338413267593006e-05, + "loss": 0.4938, + "step": 35094 + }, + { + "epoch": 1.9608067715171662, + "grad_norm": 0.3832407295703888, + "learning_rate": 2.0310398924249215e-05, + "loss": 0.4242, + "step": 35095 + }, + { + "epoch": 1.9608626421208482, + "grad_norm": 16.382169723510742, + "learning_rate": 2.0282384580905424e-05, + "loss": 0.3781, + "step": 35096 + }, + { + "epoch": 1.9609185127245299, + "grad_norm": 0.43154430389404297, + "learning_rate": 2.025437023756163e-05, + "loss": 0.4347, + "step": 35097 + }, + { + "epoch": 1.9609743833282118, + "grad_norm": 0.5850375890731812, + "learning_rate": 2.022635589421784e-05, + "loss": 0.295, + "step": 35098 + }, + { + "epoch": 1.9610302539318938, + "grad_norm": 0.3899644911289215, + "learning_rate": 2.0198341550874045e-05, + "loss": 0.454, + "step": 35099 + }, + { + "epoch": 1.9610861245355755, + "grad_norm": 0.38368353247642517, + "learning_rate": 2.0170327207530254e-05, + "loss": 0.3763, + "step": 35100 + }, + { + "epoch": 1.9611419951392575, + "grad_norm": 0.3062717318534851, + "learning_rate": 2.0142312864186464e-05, + "loss": 0.3457, + "step": 35101 + }, + { + "epoch": 1.9611978657429394, + "grad_norm": 0.42859914898872375, + "learning_rate": 2.011429852084267e-05, + "loss": 0.5428, + "step": 35102 + }, + { + "epoch": 1.9612537363466211, + "grad_norm": 0.40725550055503845, + "learning_rate": 2.008628417749888e-05, + "loss": 0.4242, + "step": 35103 + }, + { + "epoch": 1.961309606950303, + "grad_norm": 0.5616305470466614, + "learning_rate": 2.0058269834155088e-05, + "loss": 0.4528, + "step": 35104 + }, + { + "epoch": 1.961365477553985, + "grad_norm": 0.5797846913337708, + "learning_rate": 2.0030255490811294e-05, + "loss": 0.4712, + "step": 35105 + }, + { + "epoch": 1.9614213481576668, + "grad_norm": 0.48942920565605164, + "learning_rate": 2.0002241147467503e-05, + "loss": 0.3992, + "step": 35106 + }, + { + "epoch": 1.9614772187613487, + "grad_norm": 0.6421518325805664, + "learning_rate": 1.997422680412371e-05, + "loss": 0.3641, + "step": 35107 + }, + { + "epoch": 1.9615330893650307, + "grad_norm": 0.3752550482749939, + "learning_rate": 1.9946212460779918e-05, + "loss": 0.4157, + "step": 35108 + }, + { + "epoch": 1.9615889599687124, + "grad_norm": 1.1835190057754517, + "learning_rate": 1.9918198117436127e-05, + "loss": 0.3935, + "step": 35109 + }, + { + "epoch": 1.9616448305723944, + "grad_norm": 1.0512290000915527, + "learning_rate": 1.9890183774092333e-05, + "loss": 0.3136, + "step": 35110 + }, + { + "epoch": 1.9617007011760763, + "grad_norm": 0.27145132422447205, + "learning_rate": 1.9862169430748545e-05, + "loss": 0.3024, + "step": 35111 + }, + { + "epoch": 1.961756571779758, + "grad_norm": 0.5200066566467285, + "learning_rate": 1.9834155087404755e-05, + "loss": 0.4848, + "step": 35112 + }, + { + "epoch": 1.96181244238344, + "grad_norm": 0.6473117470741272, + "learning_rate": 1.980614074406096e-05, + "loss": 0.4191, + "step": 35113 + }, + { + "epoch": 1.961868312987122, + "grad_norm": 0.5494121313095093, + "learning_rate": 1.977812640071717e-05, + "loss": 0.5302, + "step": 35114 + }, + { + "epoch": 1.9619241835908037, + "grad_norm": 0.5595334768295288, + "learning_rate": 1.9750112057373376e-05, + "loss": 0.4024, + "step": 35115 + }, + { + "epoch": 1.9619800541944856, + "grad_norm": 0.3913322687149048, + "learning_rate": 1.9722097714029585e-05, + "loss": 0.3008, + "step": 35116 + }, + { + "epoch": 1.9620359247981676, + "grad_norm": 0.9521633386611938, + "learning_rate": 1.9694083370685794e-05, + "loss": 0.5031, + "step": 35117 + }, + { + "epoch": 1.9620917954018493, + "grad_norm": 0.39047864079475403, + "learning_rate": 1.9666069027342e-05, + "loss": 0.3211, + "step": 35118 + }, + { + "epoch": 1.962147666005531, + "grad_norm": 1.6735872030258179, + "learning_rate": 1.963805468399821e-05, + "loss": 0.4305, + "step": 35119 + }, + { + "epoch": 1.9622035366092132, + "grad_norm": 0.36002489924430847, + "learning_rate": 1.9610040340654418e-05, + "loss": 0.4713, + "step": 35120 + }, + { + "epoch": 1.962259407212895, + "grad_norm": 0.41529974341392517, + "learning_rate": 1.9582025997310624e-05, + "loss": 0.3696, + "step": 35121 + }, + { + "epoch": 1.9623152778165767, + "grad_norm": 0.28723135590553284, + "learning_rate": 1.9554011653966833e-05, + "loss": 0.3512, + "step": 35122 + }, + { + "epoch": 1.9623711484202588, + "grad_norm": 0.7715911865234375, + "learning_rate": 1.952599731062304e-05, + "loss": 0.3882, + "step": 35123 + }, + { + "epoch": 1.9624270190239406, + "grad_norm": 0.460662305355072, + "learning_rate": 1.9497982967279248e-05, + "loss": 0.3819, + "step": 35124 + }, + { + "epoch": 1.9624828896276223, + "grad_norm": 0.43061667680740356, + "learning_rate": 1.9469968623935457e-05, + "loss": 0.502, + "step": 35125 + }, + { + "epoch": 1.9625387602313042, + "grad_norm": 0.3600395619869232, + "learning_rate": 1.9441954280591663e-05, + "loss": 0.3956, + "step": 35126 + }, + { + "epoch": 1.9625946308349862, + "grad_norm": 0.46403631567955017, + "learning_rate": 1.9413939937247872e-05, + "loss": 0.4709, + "step": 35127 + }, + { + "epoch": 1.962650501438668, + "grad_norm": 0.4317598044872284, + "learning_rate": 1.938592559390408e-05, + "loss": 0.5264, + "step": 35128 + }, + { + "epoch": 1.9627063720423499, + "grad_norm": 1.6972817182540894, + "learning_rate": 1.9357911250560287e-05, + "loss": 0.3747, + "step": 35129 + }, + { + "epoch": 1.9627622426460318, + "grad_norm": 0.234371155500412, + "learning_rate": 1.9329896907216497e-05, + "loss": 0.2418, + "step": 35130 + }, + { + "epoch": 1.9628181132497136, + "grad_norm": 0.3389555811882019, + "learning_rate": 1.9301882563872702e-05, + "loss": 0.3558, + "step": 35131 + }, + { + "epoch": 1.9628739838533955, + "grad_norm": 0.5646786689758301, + "learning_rate": 1.927386822052891e-05, + "loss": 0.4825, + "step": 35132 + }, + { + "epoch": 1.9629298544570775, + "grad_norm": 0.3673233985900879, + "learning_rate": 1.924585387718512e-05, + "loss": 0.3759, + "step": 35133 + }, + { + "epoch": 1.9629857250607592, + "grad_norm": 1.0776273012161255, + "learning_rate": 1.9217839533841327e-05, + "loss": 0.3154, + "step": 35134 + }, + { + "epoch": 1.9630415956644411, + "grad_norm": 0.5600674748420715, + "learning_rate": 1.9189825190497536e-05, + "loss": 0.5734, + "step": 35135 + }, + { + "epoch": 1.963097466268123, + "grad_norm": 0.31662386655807495, + "learning_rate": 1.9161810847153745e-05, + "loss": 0.3436, + "step": 35136 + }, + { + "epoch": 1.9631533368718048, + "grad_norm": 0.5415621399879456, + "learning_rate": 1.913379650380995e-05, + "loss": 0.3499, + "step": 35137 + }, + { + "epoch": 1.9632092074754868, + "grad_norm": 0.4466671049594879, + "learning_rate": 1.910578216046616e-05, + "loss": 0.3958, + "step": 35138 + }, + { + "epoch": 1.9632650780791687, + "grad_norm": 0.37103182077407837, + "learning_rate": 1.9077767817122366e-05, + "loss": 0.3854, + "step": 35139 + }, + { + "epoch": 1.9633209486828505, + "grad_norm": 0.3715423345565796, + "learning_rate": 1.9049753473778575e-05, + "loss": 0.347, + "step": 35140 + }, + { + "epoch": 1.9633768192865324, + "grad_norm": 0.2944261133670807, + "learning_rate": 1.9021739130434784e-05, + "loss": 0.3923, + "step": 35141 + }, + { + "epoch": 1.9634326898902144, + "grad_norm": 0.5496199727058411, + "learning_rate": 1.899372478709099e-05, + "loss": 0.4537, + "step": 35142 + }, + { + "epoch": 1.963488560493896, + "grad_norm": 0.4475196599960327, + "learning_rate": 1.89657104437472e-05, + "loss": 0.4226, + "step": 35143 + }, + { + "epoch": 1.963544431097578, + "grad_norm": 0.3139975070953369, + "learning_rate": 1.893769610040341e-05, + "loss": 0.3673, + "step": 35144 + }, + { + "epoch": 1.96360030170126, + "grad_norm": 2.008916139602661, + "learning_rate": 1.8909681757059614e-05, + "loss": 0.3558, + "step": 35145 + }, + { + "epoch": 1.9636561723049417, + "grad_norm": 3.509674072265625, + "learning_rate": 1.8881667413715824e-05, + "loss": 0.4495, + "step": 35146 + }, + { + "epoch": 1.9637120429086237, + "grad_norm": 0.39259472489356995, + "learning_rate": 1.885365307037203e-05, + "loss": 0.4254, + "step": 35147 + }, + { + "epoch": 1.9637679135123056, + "grad_norm": 0.32883507013320923, + "learning_rate": 1.882563872702824e-05, + "loss": 0.4259, + "step": 35148 + }, + { + "epoch": 1.9638237841159873, + "grad_norm": 1.3035303354263306, + "learning_rate": 1.8797624383684448e-05, + "loss": 0.4107, + "step": 35149 + }, + { + "epoch": 1.9638796547196693, + "grad_norm": 0.331350713968277, + "learning_rate": 1.8769610040340654e-05, + "loss": 0.4208, + "step": 35150 + }, + { + "epoch": 1.9639355253233512, + "grad_norm": 0.48227187991142273, + "learning_rate": 1.8741595696996863e-05, + "loss": 0.4749, + "step": 35151 + }, + { + "epoch": 1.963991395927033, + "grad_norm": 0.3820458948612213, + "learning_rate": 1.8713581353653072e-05, + "loss": 0.5327, + "step": 35152 + }, + { + "epoch": 1.9640472665307147, + "grad_norm": 1.5840853452682495, + "learning_rate": 1.8685567010309278e-05, + "loss": 0.4077, + "step": 35153 + }, + { + "epoch": 1.9641031371343969, + "grad_norm": 0.34333422780036926, + "learning_rate": 1.8657552666965487e-05, + "loss": 0.4016, + "step": 35154 + }, + { + "epoch": 1.9641590077380786, + "grad_norm": 0.2861166000366211, + "learning_rate": 1.8629538323621693e-05, + "loss": 0.3673, + "step": 35155 + }, + { + "epoch": 1.9642148783417603, + "grad_norm": 0.5153173804283142, + "learning_rate": 1.8601523980277902e-05, + "loss": 0.4019, + "step": 35156 + }, + { + "epoch": 1.9642707489454425, + "grad_norm": 0.3562248945236206, + "learning_rate": 1.857350963693411e-05, + "loss": 0.3176, + "step": 35157 + }, + { + "epoch": 1.9643266195491242, + "grad_norm": 1.9684510231018066, + "learning_rate": 1.8545495293590317e-05, + "loss": 0.297, + "step": 35158 + }, + { + "epoch": 1.964382490152806, + "grad_norm": 0.36647793650627136, + "learning_rate": 1.8517480950246526e-05, + "loss": 0.3478, + "step": 35159 + }, + { + "epoch": 1.964438360756488, + "grad_norm": 0.9096338748931885, + "learning_rate": 1.8489466606902732e-05, + "loss": 0.3833, + "step": 35160 + }, + { + "epoch": 1.9644942313601699, + "grad_norm": 0.6509581804275513, + "learning_rate": 1.846145226355894e-05, + "loss": 0.5178, + "step": 35161 + }, + { + "epoch": 1.9645501019638516, + "grad_norm": 0.6919894814491272, + "learning_rate": 1.843343792021515e-05, + "loss": 0.546, + "step": 35162 + }, + { + "epoch": 1.9646059725675336, + "grad_norm": 0.503948986530304, + "learning_rate": 1.8405423576871356e-05, + "loss": 0.3979, + "step": 35163 + }, + { + "epoch": 1.9646618431712155, + "grad_norm": 0.3718375265598297, + "learning_rate": 1.8377409233527566e-05, + "loss": 0.4217, + "step": 35164 + }, + { + "epoch": 1.9647177137748972, + "grad_norm": 0.5655865669250488, + "learning_rate": 1.8349394890183775e-05, + "loss": 0.4107, + "step": 35165 + }, + { + "epoch": 1.9647735843785792, + "grad_norm": 0.32770660519599915, + "learning_rate": 1.832138054683998e-05, + "loss": 0.397, + "step": 35166 + }, + { + "epoch": 1.9648294549822611, + "grad_norm": 0.4253676235675812, + "learning_rate": 1.829336620349619e-05, + "loss": 0.2985, + "step": 35167 + }, + { + "epoch": 1.9648853255859429, + "grad_norm": 0.558607816696167, + "learning_rate": 1.8265351860152396e-05, + "loss": 0.4355, + "step": 35168 + }, + { + "epoch": 1.9649411961896248, + "grad_norm": 0.5531201958656311, + "learning_rate": 1.8237337516808605e-05, + "loss": 0.4554, + "step": 35169 + }, + { + "epoch": 1.9649970667933068, + "grad_norm": 0.3821670711040497, + "learning_rate": 1.8209323173464814e-05, + "loss": 0.3977, + "step": 35170 + }, + { + "epoch": 1.9650529373969885, + "grad_norm": 1.9951872825622559, + "learning_rate": 1.818130883012102e-05, + "loss": 0.4296, + "step": 35171 + }, + { + "epoch": 1.9651088080006704, + "grad_norm": 1.804835557937622, + "learning_rate": 1.815329448677723e-05, + "loss": 0.4815, + "step": 35172 + }, + { + "epoch": 1.9651646786043524, + "grad_norm": 5.842530250549316, + "learning_rate": 1.8125280143433438e-05, + "loss": 0.3789, + "step": 35173 + }, + { + "epoch": 1.9652205492080341, + "grad_norm": 1.1214741468429565, + "learning_rate": 1.8097265800089644e-05, + "loss": 0.3461, + "step": 35174 + }, + { + "epoch": 1.965276419811716, + "grad_norm": 0.38702329993247986, + "learning_rate": 1.8069251456745853e-05, + "loss": 0.3724, + "step": 35175 + }, + { + "epoch": 1.965332290415398, + "grad_norm": 0.3432091474533081, + "learning_rate": 1.804123711340206e-05, + "loss": 0.3725, + "step": 35176 + }, + { + "epoch": 1.9653881610190798, + "grad_norm": 0.5281915068626404, + "learning_rate": 1.8013222770058268e-05, + "loss": 0.3387, + "step": 35177 + }, + { + "epoch": 1.9654440316227617, + "grad_norm": 4.2905497550964355, + "learning_rate": 1.798520842671448e-05, + "loss": 0.328, + "step": 35178 + }, + { + "epoch": 1.9654999022264437, + "grad_norm": 2.072777032852173, + "learning_rate": 1.7957194083370687e-05, + "loss": 0.4099, + "step": 35179 + }, + { + "epoch": 1.9655557728301254, + "grad_norm": 0.3317776918411255, + "learning_rate": 1.7929179740026896e-05, + "loss": 0.3623, + "step": 35180 + }, + { + "epoch": 1.9656116434338073, + "grad_norm": 0.4044872522354126, + "learning_rate": 1.7901165396683105e-05, + "loss": 0.3732, + "step": 35181 + }, + { + "epoch": 1.9656675140374893, + "grad_norm": 1.0407843589782715, + "learning_rate": 1.787315105333931e-05, + "loss": 0.3319, + "step": 35182 + }, + { + "epoch": 1.965723384641171, + "grad_norm": 0.36356040835380554, + "learning_rate": 1.784513670999552e-05, + "loss": 0.3931, + "step": 35183 + }, + { + "epoch": 1.965779255244853, + "grad_norm": 0.3268255889415741, + "learning_rate": 1.7817122366651726e-05, + "loss": 0.3897, + "step": 35184 + }, + { + "epoch": 1.965835125848535, + "grad_norm": 0.7652891278266907, + "learning_rate": 1.7789108023307935e-05, + "loss": 0.522, + "step": 35185 + }, + { + "epoch": 1.9658909964522167, + "grad_norm": 0.6671881675720215, + "learning_rate": 1.7761093679964144e-05, + "loss": 0.436, + "step": 35186 + }, + { + "epoch": 1.9659468670558984, + "grad_norm": 0.41406816244125366, + "learning_rate": 1.773307933662035e-05, + "loss": 0.463, + "step": 35187 + }, + { + "epoch": 1.9660027376595806, + "grad_norm": 0.5086569786071777, + "learning_rate": 1.770506499327656e-05, + "loss": 0.5686, + "step": 35188 + }, + { + "epoch": 1.9660586082632623, + "grad_norm": 1.6365602016448975, + "learning_rate": 1.767705064993277e-05, + "loss": 0.4375, + "step": 35189 + }, + { + "epoch": 1.966114478866944, + "grad_norm": 0.5463235974311829, + "learning_rate": 1.7649036306588974e-05, + "loss": 0.4126, + "step": 35190 + }, + { + "epoch": 1.9661703494706262, + "grad_norm": 0.2622164487838745, + "learning_rate": 1.7621021963245184e-05, + "loss": 0.3112, + "step": 35191 + }, + { + "epoch": 1.966226220074308, + "grad_norm": 0.3908533453941345, + "learning_rate": 1.759300761990139e-05, + "loss": 0.3592, + "step": 35192 + }, + { + "epoch": 1.9662820906779896, + "grad_norm": 0.38852450251579285, + "learning_rate": 1.75649932765576e-05, + "loss": 0.3849, + "step": 35193 + }, + { + "epoch": 1.9663379612816716, + "grad_norm": 0.3332635462284088, + "learning_rate": 1.7536978933213808e-05, + "loss": 0.3162, + "step": 35194 + }, + { + "epoch": 1.9663938318853535, + "grad_norm": 0.4447583854198456, + "learning_rate": 1.7508964589870014e-05, + "loss": 0.3581, + "step": 35195 + }, + { + "epoch": 1.9664497024890353, + "grad_norm": 0.3590472340583801, + "learning_rate": 1.7480950246526223e-05, + "loss": 0.4054, + "step": 35196 + }, + { + "epoch": 1.9665055730927172, + "grad_norm": 0.8402016758918762, + "learning_rate": 1.7452935903182432e-05, + "loss": 0.4377, + "step": 35197 + }, + { + "epoch": 1.9665614436963992, + "grad_norm": 1.2762351036071777, + "learning_rate": 1.7424921559838638e-05, + "loss": 0.4463, + "step": 35198 + }, + { + "epoch": 1.966617314300081, + "grad_norm": 1.8691726922988892, + "learning_rate": 1.7396907216494847e-05, + "loss": 0.3626, + "step": 35199 + }, + { + "epoch": 1.9666731849037629, + "grad_norm": 0.3120005130767822, + "learning_rate": 1.7368892873151053e-05, + "loss": 0.4026, + "step": 35200 + }, + { + "epoch": 1.9667290555074448, + "grad_norm": 0.37045910954475403, + "learning_rate": 1.7340878529807262e-05, + "loss": 0.4663, + "step": 35201 + }, + { + "epoch": 1.9667849261111265, + "grad_norm": 0.339084655046463, + "learning_rate": 1.731286418646347e-05, + "loss": 0.4157, + "step": 35202 + }, + { + "epoch": 1.9668407967148085, + "grad_norm": 1.1518828868865967, + "learning_rate": 1.7284849843119677e-05, + "loss": 0.3898, + "step": 35203 + }, + { + "epoch": 1.9668966673184904, + "grad_norm": 0.39826419949531555, + "learning_rate": 1.7256835499775886e-05, + "loss": 0.3713, + "step": 35204 + }, + { + "epoch": 1.9669525379221722, + "grad_norm": 0.4281449019908905, + "learning_rate": 1.7228821156432095e-05, + "loss": 0.4153, + "step": 35205 + }, + { + "epoch": 1.9670084085258541, + "grad_norm": 0.3349481523036957, + "learning_rate": 1.72008068130883e-05, + "loss": 0.4479, + "step": 35206 + }, + { + "epoch": 1.967064279129536, + "grad_norm": 0.3813696801662445, + "learning_rate": 1.717279246974451e-05, + "loss": 0.4564, + "step": 35207 + }, + { + "epoch": 1.9671201497332178, + "grad_norm": 0.8667809963226318, + "learning_rate": 1.7144778126400716e-05, + "loss": 0.4701, + "step": 35208 + }, + { + "epoch": 1.9671760203368998, + "grad_norm": 0.3691703677177429, + "learning_rate": 1.7116763783056925e-05, + "loss": 0.4524, + "step": 35209 + }, + { + "epoch": 1.9672318909405817, + "grad_norm": 0.3843746483325958, + "learning_rate": 1.7088749439713135e-05, + "loss": 0.4737, + "step": 35210 + }, + { + "epoch": 1.9672877615442634, + "grad_norm": 0.5006298422813416, + "learning_rate": 1.706073509636934e-05, + "loss": 0.3676, + "step": 35211 + }, + { + "epoch": 1.9673436321479454, + "grad_norm": 0.6923978328704834, + "learning_rate": 1.703272075302555e-05, + "loss": 0.4156, + "step": 35212 + }, + { + "epoch": 1.9673995027516273, + "grad_norm": 0.8398563265800476, + "learning_rate": 1.700470640968176e-05, + "loss": 0.342, + "step": 35213 + }, + { + "epoch": 1.967455373355309, + "grad_norm": 0.3644918203353882, + "learning_rate": 1.6976692066337965e-05, + "loss": 0.4531, + "step": 35214 + }, + { + "epoch": 1.967511243958991, + "grad_norm": 0.34679529070854187, + "learning_rate": 1.6948677722994174e-05, + "loss": 0.3803, + "step": 35215 + }, + { + "epoch": 1.967567114562673, + "grad_norm": 0.4583982527256012, + "learning_rate": 1.692066337965038e-05, + "loss": 0.4636, + "step": 35216 + }, + { + "epoch": 1.9676229851663547, + "grad_norm": 1.1519744396209717, + "learning_rate": 1.689264903630659e-05, + "loss": 0.4735, + "step": 35217 + }, + { + "epoch": 1.9676788557700366, + "grad_norm": 0.39482200145721436, + "learning_rate": 1.6864634692962798e-05, + "loss": 0.4516, + "step": 35218 + }, + { + "epoch": 1.9677347263737186, + "grad_norm": 0.47675183415412903, + "learning_rate": 1.6836620349619004e-05, + "loss": 0.4612, + "step": 35219 + }, + { + "epoch": 1.9677905969774003, + "grad_norm": 0.3867076635360718, + "learning_rate": 1.6808606006275213e-05, + "loss": 0.3985, + "step": 35220 + }, + { + "epoch": 1.967846467581082, + "grad_norm": 0.4631907343864441, + "learning_rate": 1.6780591662931422e-05, + "loss": 0.3982, + "step": 35221 + }, + { + "epoch": 1.9679023381847642, + "grad_norm": 0.34107285737991333, + "learning_rate": 1.6752577319587628e-05, + "loss": 0.344, + "step": 35222 + }, + { + "epoch": 1.967958208788446, + "grad_norm": 0.5682148933410645, + "learning_rate": 1.6724562976243837e-05, + "loss": 0.4134, + "step": 35223 + }, + { + "epoch": 1.9680140793921277, + "grad_norm": 0.7029966711997986, + "learning_rate": 1.6696548632900043e-05, + "loss": 0.4961, + "step": 35224 + }, + { + "epoch": 1.9680699499958099, + "grad_norm": 0.3284512460231781, + "learning_rate": 1.6668534289556252e-05, + "loss": 0.4048, + "step": 35225 + }, + { + "epoch": 1.9681258205994916, + "grad_norm": 0.3305002450942993, + "learning_rate": 1.664051994621246e-05, + "loss": 0.4235, + "step": 35226 + }, + { + "epoch": 1.9681816912031733, + "grad_norm": 0.4203590154647827, + "learning_rate": 1.6612505602868667e-05, + "loss": 0.4109, + "step": 35227 + }, + { + "epoch": 1.9682375618068553, + "grad_norm": 0.46679866313934326, + "learning_rate": 1.6584491259524877e-05, + "loss": 0.4729, + "step": 35228 + }, + { + "epoch": 1.9682934324105372, + "grad_norm": 0.47902828454971313, + "learning_rate": 1.6556476916181086e-05, + "loss": 0.4681, + "step": 35229 + }, + { + "epoch": 1.968349303014219, + "grad_norm": 0.4284192621707916, + "learning_rate": 1.652846257283729e-05, + "loss": 0.4995, + "step": 35230 + }, + { + "epoch": 1.968405173617901, + "grad_norm": 0.45257487893104553, + "learning_rate": 1.65004482294935e-05, + "loss": 0.5605, + "step": 35231 + }, + { + "epoch": 1.9684610442215829, + "grad_norm": 1.6922435760498047, + "learning_rate": 1.6472433886149707e-05, + "loss": 0.3885, + "step": 35232 + }, + { + "epoch": 1.9685169148252646, + "grad_norm": 0.36862826347351074, + "learning_rate": 1.6444419542805916e-05, + "loss": 0.3969, + "step": 35233 + }, + { + "epoch": 1.9685727854289465, + "grad_norm": 0.6461756229400635, + "learning_rate": 1.6416405199462125e-05, + "loss": 0.5469, + "step": 35234 + }, + { + "epoch": 1.9686286560326285, + "grad_norm": 0.40880075097084045, + "learning_rate": 1.638839085611833e-05, + "loss": 0.3882, + "step": 35235 + }, + { + "epoch": 1.9686845266363102, + "grad_norm": 0.4211095869541168, + "learning_rate": 1.636037651277454e-05, + "loss": 0.3708, + "step": 35236 + }, + { + "epoch": 1.9687403972399922, + "grad_norm": 0.4694100320339203, + "learning_rate": 1.633236216943075e-05, + "loss": 0.5077, + "step": 35237 + }, + { + "epoch": 1.9687962678436741, + "grad_norm": 0.3899674713611603, + "learning_rate": 1.6304347826086955e-05, + "loss": 0.4209, + "step": 35238 + }, + { + "epoch": 1.9688521384473558, + "grad_norm": 0.8081586360931396, + "learning_rate": 1.6276333482743164e-05, + "loss": 0.3994, + "step": 35239 + }, + { + "epoch": 1.9689080090510378, + "grad_norm": 0.4116978645324707, + "learning_rate": 1.624831913939937e-05, + "loss": 0.429, + "step": 35240 + }, + { + "epoch": 1.9689638796547198, + "grad_norm": 0.3790639638900757, + "learning_rate": 1.622030479605558e-05, + "loss": 0.3574, + "step": 35241 + }, + { + "epoch": 1.9690197502584015, + "grad_norm": 2.998286485671997, + "learning_rate": 1.619229045271179e-05, + "loss": 0.624, + "step": 35242 + }, + { + "epoch": 1.9690756208620834, + "grad_norm": 0.4271169602870941, + "learning_rate": 1.6164276109367994e-05, + "loss": 0.3996, + "step": 35243 + }, + { + "epoch": 1.9691314914657654, + "grad_norm": 0.4649761915206909, + "learning_rate": 1.6136261766024204e-05, + "loss": 0.4593, + "step": 35244 + }, + { + "epoch": 1.9691873620694471, + "grad_norm": 0.5670130848884583, + "learning_rate": 1.610824742268041e-05, + "loss": 0.4818, + "step": 35245 + }, + { + "epoch": 1.969243232673129, + "grad_norm": 0.3861340284347534, + "learning_rate": 1.6080233079336622e-05, + "loss": 0.3182, + "step": 35246 + }, + { + "epoch": 1.969299103276811, + "grad_norm": 0.45759761333465576, + "learning_rate": 1.605221873599283e-05, + "loss": 0.4309, + "step": 35247 + }, + { + "epoch": 1.9693549738804927, + "grad_norm": 0.5834199786186218, + "learning_rate": 1.6024204392649037e-05, + "loss": 0.4418, + "step": 35248 + }, + { + "epoch": 1.9694108444841747, + "grad_norm": 0.4056153893470764, + "learning_rate": 1.5996190049305246e-05, + "loss": 0.3617, + "step": 35249 + }, + { + "epoch": 1.9694667150878566, + "grad_norm": 1.1991982460021973, + "learning_rate": 1.5968175705961455e-05, + "loss": 0.3746, + "step": 35250 + }, + { + "epoch": 1.9695225856915384, + "grad_norm": 0.5794174075126648, + "learning_rate": 1.594016136261766e-05, + "loss": 0.3696, + "step": 35251 + }, + { + "epoch": 1.96957845629522, + "grad_norm": 0.7341796159744263, + "learning_rate": 1.591214701927387e-05, + "loss": 0.4462, + "step": 35252 + }, + { + "epoch": 1.9696343268989023, + "grad_norm": 0.48963794112205505, + "learning_rate": 1.5884132675930076e-05, + "loss": 0.6735, + "step": 35253 + }, + { + "epoch": 1.969690197502584, + "grad_norm": 0.35346826910972595, + "learning_rate": 1.5856118332586285e-05, + "loss": 0.4439, + "step": 35254 + }, + { + "epoch": 1.9697460681062657, + "grad_norm": 0.33615219593048096, + "learning_rate": 1.5828103989242495e-05, + "loss": 0.3697, + "step": 35255 + }, + { + "epoch": 1.969801938709948, + "grad_norm": 0.5258422493934631, + "learning_rate": 1.58000896458987e-05, + "loss": 0.4414, + "step": 35256 + }, + { + "epoch": 1.9698578093136296, + "grad_norm": 0.5710558891296387, + "learning_rate": 1.577207530255491e-05, + "loss": 0.3843, + "step": 35257 + }, + { + "epoch": 1.9699136799173114, + "grad_norm": 0.5514925718307495, + "learning_rate": 1.574406095921112e-05, + "loss": 0.3739, + "step": 35258 + }, + { + "epoch": 1.9699695505209935, + "grad_norm": 0.9132239818572998, + "learning_rate": 1.5716046615867325e-05, + "loss": 0.535, + "step": 35259 + }, + { + "epoch": 1.9700254211246753, + "grad_norm": 0.3889336884021759, + "learning_rate": 1.5688032272523534e-05, + "loss": 0.49, + "step": 35260 + }, + { + "epoch": 1.970081291728357, + "grad_norm": 0.8641798496246338, + "learning_rate": 1.566001792917974e-05, + "loss": 0.3808, + "step": 35261 + }, + { + "epoch": 1.970137162332039, + "grad_norm": 7.148865699768066, + "learning_rate": 1.563200358583595e-05, + "loss": 0.4274, + "step": 35262 + }, + { + "epoch": 1.970193032935721, + "grad_norm": 0.41186943650245667, + "learning_rate": 1.5603989242492155e-05, + "loss": 0.564, + "step": 35263 + }, + { + "epoch": 1.9702489035394026, + "grad_norm": 0.5236369371414185, + "learning_rate": 1.5575974899148364e-05, + "loss": 0.356, + "step": 35264 + }, + { + "epoch": 1.9703047741430846, + "grad_norm": 0.46556851267814636, + "learning_rate": 1.5547960555804573e-05, + "loss": 0.3726, + "step": 35265 + }, + { + "epoch": 1.9703606447467665, + "grad_norm": 0.8942652344703674, + "learning_rate": 1.551994621246078e-05, + "loss": 0.4562, + "step": 35266 + }, + { + "epoch": 1.9704165153504483, + "grad_norm": 0.7025408148765564, + "learning_rate": 1.5491931869116988e-05, + "loss": 0.4903, + "step": 35267 + }, + { + "epoch": 1.9704723859541302, + "grad_norm": 0.34000638127326965, + "learning_rate": 1.5463917525773197e-05, + "loss": 0.3742, + "step": 35268 + }, + { + "epoch": 1.9705282565578122, + "grad_norm": 0.6306625008583069, + "learning_rate": 1.5435903182429403e-05, + "loss": 0.3878, + "step": 35269 + }, + { + "epoch": 1.970584127161494, + "grad_norm": 0.5040526986122131, + "learning_rate": 1.5407888839085612e-05, + "loss": 0.3586, + "step": 35270 + }, + { + "epoch": 1.9706399977651758, + "grad_norm": 0.5481066703796387, + "learning_rate": 1.5379874495741818e-05, + "loss": 0.6021, + "step": 35271 + }, + { + "epoch": 1.9706958683688578, + "grad_norm": 0.4591140151023865, + "learning_rate": 1.5351860152398027e-05, + "loss": 0.4349, + "step": 35272 + }, + { + "epoch": 1.9707517389725395, + "grad_norm": 0.3916412591934204, + "learning_rate": 1.5323845809054237e-05, + "loss": 0.3007, + "step": 35273 + }, + { + "epoch": 1.9708076095762215, + "grad_norm": 1.560625433921814, + "learning_rate": 1.5295831465710442e-05, + "loss": 0.5531, + "step": 35274 + }, + { + "epoch": 1.9708634801799034, + "grad_norm": 0.3510339558124542, + "learning_rate": 1.526781712236665e-05, + "loss": 0.3477, + "step": 35275 + }, + { + "epoch": 1.9709193507835852, + "grad_norm": 0.37701183557510376, + "learning_rate": 1.523980277902286e-05, + "loss": 0.4257, + "step": 35276 + }, + { + "epoch": 1.970975221387267, + "grad_norm": 1.0126398801803589, + "learning_rate": 1.5211788435679068e-05, + "loss": 0.3328, + "step": 35277 + }, + { + "epoch": 1.971031091990949, + "grad_norm": 0.37289944291114807, + "learning_rate": 1.5183774092335276e-05, + "loss": 0.4157, + "step": 35278 + }, + { + "epoch": 1.9710869625946308, + "grad_norm": 0.5329359173774719, + "learning_rate": 1.5155759748991483e-05, + "loss": 0.4698, + "step": 35279 + }, + { + "epoch": 1.9711428331983127, + "grad_norm": 0.4355868399143219, + "learning_rate": 1.5127745405647693e-05, + "loss": 0.4148, + "step": 35280 + }, + { + "epoch": 1.9711987038019947, + "grad_norm": 0.7701157331466675, + "learning_rate": 1.50997310623039e-05, + "loss": 0.3817, + "step": 35281 + }, + { + "epoch": 1.9712545744056764, + "grad_norm": 2.2527880668640137, + "learning_rate": 1.5071716718960108e-05, + "loss": 0.5179, + "step": 35282 + }, + { + "epoch": 1.9713104450093584, + "grad_norm": 0.41233736276626587, + "learning_rate": 1.5043702375616315e-05, + "loss": 0.3713, + "step": 35283 + }, + { + "epoch": 1.9713663156130403, + "grad_norm": 0.31079623103141785, + "learning_rate": 1.5015688032272524e-05, + "loss": 0.4456, + "step": 35284 + }, + { + "epoch": 1.971422186216722, + "grad_norm": 0.42161667346954346, + "learning_rate": 1.4987673688928732e-05, + "loss": 0.4296, + "step": 35285 + }, + { + "epoch": 1.9714780568204038, + "grad_norm": 0.31922900676727295, + "learning_rate": 1.495965934558494e-05, + "loss": 0.3876, + "step": 35286 + }, + { + "epoch": 1.971533927424086, + "grad_norm": 0.5481165051460266, + "learning_rate": 1.4931645002241147e-05, + "loss": 0.5232, + "step": 35287 + }, + { + "epoch": 1.9715897980277677, + "grad_norm": 0.4144391417503357, + "learning_rate": 1.4903630658897356e-05, + "loss": 0.4279, + "step": 35288 + }, + { + "epoch": 1.9716456686314494, + "grad_norm": 0.44656452536582947, + "learning_rate": 1.4875616315553563e-05, + "loss": 0.3625, + "step": 35289 + }, + { + "epoch": 1.9717015392351316, + "grad_norm": 0.4513012170791626, + "learning_rate": 1.4847601972209771e-05, + "loss": 0.4774, + "step": 35290 + }, + { + "epoch": 1.9717574098388133, + "grad_norm": 0.41149890422821045, + "learning_rate": 1.4819587628865979e-05, + "loss": 0.3534, + "step": 35291 + }, + { + "epoch": 1.971813280442495, + "grad_norm": 0.4076889157295227, + "learning_rate": 1.4791573285522188e-05, + "loss": 0.3595, + "step": 35292 + }, + { + "epoch": 1.9718691510461772, + "grad_norm": 0.34817078709602356, + "learning_rate": 1.4763558942178395e-05, + "loss": 0.3777, + "step": 35293 + }, + { + "epoch": 1.971925021649859, + "grad_norm": 0.2774334251880646, + "learning_rate": 1.4735544598834603e-05, + "loss": 0.3215, + "step": 35294 + }, + { + "epoch": 1.9719808922535407, + "grad_norm": 0.3214719295501709, + "learning_rate": 1.470753025549081e-05, + "loss": 0.3451, + "step": 35295 + }, + { + "epoch": 1.9720367628572226, + "grad_norm": 0.39714518189430237, + "learning_rate": 1.4679515912147021e-05, + "loss": 0.3281, + "step": 35296 + }, + { + "epoch": 1.9720926334609046, + "grad_norm": 0.3374534845352173, + "learning_rate": 1.4651501568803229e-05, + "loss": 0.3864, + "step": 35297 + }, + { + "epoch": 1.9721485040645863, + "grad_norm": 0.3287605047225952, + "learning_rate": 1.4623487225459436e-05, + "loss": 0.3213, + "step": 35298 + }, + { + "epoch": 1.9722043746682683, + "grad_norm": 0.8247247934341431, + "learning_rate": 1.4595472882115644e-05, + "loss": 0.4442, + "step": 35299 + }, + { + "epoch": 1.9722602452719502, + "grad_norm": 1.0226812362670898, + "learning_rate": 1.4567458538771853e-05, + "loss": 0.3708, + "step": 35300 + }, + { + "epoch": 1.972316115875632, + "grad_norm": 0.3859289586544037, + "learning_rate": 1.453944419542806e-05, + "loss": 0.3639, + "step": 35301 + }, + { + "epoch": 1.9723719864793139, + "grad_norm": 0.3122573494911194, + "learning_rate": 1.4511429852084268e-05, + "loss": 0.3742, + "step": 35302 + }, + { + "epoch": 1.9724278570829958, + "grad_norm": 0.30165138840675354, + "learning_rate": 1.4483415508740475e-05, + "loss": 0.3487, + "step": 35303 + }, + { + "epoch": 1.9724837276866776, + "grad_norm": 0.37233006954193115, + "learning_rate": 1.4455401165396685e-05, + "loss": 0.4154, + "step": 35304 + }, + { + "epoch": 1.9725395982903595, + "grad_norm": 0.3415948450565338, + "learning_rate": 1.4427386822052892e-05, + "loss": 0.3487, + "step": 35305 + }, + { + "epoch": 1.9725954688940415, + "grad_norm": 0.331565260887146, + "learning_rate": 1.43993724787091e-05, + "loss": 0.3911, + "step": 35306 + }, + { + "epoch": 1.9726513394977232, + "grad_norm": 1.0243853330612183, + "learning_rate": 1.4371358135365307e-05, + "loss": 0.4062, + "step": 35307 + }, + { + "epoch": 1.9727072101014052, + "grad_norm": 1.012930989265442, + "learning_rate": 1.4343343792021516e-05, + "loss": 0.3888, + "step": 35308 + }, + { + "epoch": 1.972763080705087, + "grad_norm": 0.4306751787662506, + "learning_rate": 1.4315329448677724e-05, + "loss": 0.3635, + "step": 35309 + }, + { + "epoch": 1.9728189513087688, + "grad_norm": 0.44946420192718506, + "learning_rate": 1.4287315105333931e-05, + "loss": 0.509, + "step": 35310 + }, + { + "epoch": 1.9728748219124508, + "grad_norm": 0.35199210047721863, + "learning_rate": 1.4259300761990139e-05, + "loss": 0.3696, + "step": 35311 + }, + { + "epoch": 1.9729306925161327, + "grad_norm": 0.449562668800354, + "learning_rate": 1.4231286418646348e-05, + "loss": 0.4507, + "step": 35312 + }, + { + "epoch": 1.9729865631198145, + "grad_norm": 0.40688714385032654, + "learning_rate": 1.4203272075302556e-05, + "loss": 0.4225, + "step": 35313 + }, + { + "epoch": 1.9730424337234964, + "grad_norm": 0.3414492607116699, + "learning_rate": 1.4175257731958763e-05, + "loss": 0.3721, + "step": 35314 + }, + { + "epoch": 1.9730983043271784, + "grad_norm": 1.4660786390304565, + "learning_rate": 1.414724338861497e-05, + "loss": 0.4938, + "step": 35315 + }, + { + "epoch": 1.97315417493086, + "grad_norm": 0.4567111134529114, + "learning_rate": 1.411922904527118e-05, + "loss": 0.4931, + "step": 35316 + }, + { + "epoch": 1.973210045534542, + "grad_norm": 0.42869025468826294, + "learning_rate": 1.4091214701927387e-05, + "loss": 0.4061, + "step": 35317 + }, + { + "epoch": 1.973265916138224, + "grad_norm": 0.5000063180923462, + "learning_rate": 1.4063200358583595e-05, + "loss": 0.4193, + "step": 35318 + }, + { + "epoch": 1.9733217867419057, + "grad_norm": 0.35139450430870056, + "learning_rate": 1.4035186015239802e-05, + "loss": 0.4176, + "step": 35319 + }, + { + "epoch": 1.9733776573455875, + "grad_norm": 0.41805997490882874, + "learning_rate": 1.4007171671896012e-05, + "loss": 0.4102, + "step": 35320 + }, + { + "epoch": 1.9734335279492696, + "grad_norm": 0.33719107508659363, + "learning_rate": 1.3979157328552219e-05, + "loss": 0.4263, + "step": 35321 + }, + { + "epoch": 1.9734893985529514, + "grad_norm": 0.41814616322517395, + "learning_rate": 1.3951142985208427e-05, + "loss": 0.3575, + "step": 35322 + }, + { + "epoch": 1.973545269156633, + "grad_norm": 0.425140917301178, + "learning_rate": 1.3923128641864634e-05, + "loss": 0.3281, + "step": 35323 + }, + { + "epoch": 1.9736011397603153, + "grad_norm": 0.5218266844749451, + "learning_rate": 1.3895114298520843e-05, + "loss": 0.2948, + "step": 35324 + }, + { + "epoch": 1.973657010363997, + "grad_norm": 0.6118056774139404, + "learning_rate": 1.386709995517705e-05, + "loss": 0.491, + "step": 35325 + }, + { + "epoch": 1.9737128809676787, + "grad_norm": 0.6867071986198425, + "learning_rate": 1.3839085611833258e-05, + "loss": 0.4552, + "step": 35326 + }, + { + "epoch": 1.973768751571361, + "grad_norm": 0.566696286201477, + "learning_rate": 1.3811071268489466e-05, + "loss": 0.3807, + "step": 35327 + }, + { + "epoch": 1.9738246221750426, + "grad_norm": 0.385211706161499, + "learning_rate": 1.3783056925145673e-05, + "loss": 0.4841, + "step": 35328 + }, + { + "epoch": 1.9738804927787243, + "grad_norm": 0.5399830937385559, + "learning_rate": 1.3755042581801882e-05, + "loss": 0.3678, + "step": 35329 + }, + { + "epoch": 1.9739363633824063, + "grad_norm": 0.3064306676387787, + "learning_rate": 1.3727028238458092e-05, + "loss": 0.3673, + "step": 35330 + }, + { + "epoch": 1.9739922339860883, + "grad_norm": 0.7752071022987366, + "learning_rate": 1.36990138951143e-05, + "loss": 0.3565, + "step": 35331 + }, + { + "epoch": 1.97404810458977, + "grad_norm": 0.3764433264732361, + "learning_rate": 1.3670999551770507e-05, + "loss": 0.3385, + "step": 35332 + }, + { + "epoch": 1.974103975193452, + "grad_norm": 0.7953189015388489, + "learning_rate": 1.3642985208426716e-05, + "loss": 0.3522, + "step": 35333 + }, + { + "epoch": 1.9741598457971339, + "grad_norm": 0.3528911769390106, + "learning_rate": 1.3614970865082923e-05, + "loss": 0.3318, + "step": 35334 + }, + { + "epoch": 1.9742157164008156, + "grad_norm": 0.3512069284915924, + "learning_rate": 1.3586956521739131e-05, + "loss": 0.4393, + "step": 35335 + }, + { + "epoch": 1.9742715870044976, + "grad_norm": 0.2486758977174759, + "learning_rate": 1.3558942178395338e-05, + "loss": 0.3857, + "step": 35336 + }, + { + "epoch": 1.9743274576081795, + "grad_norm": 0.4450356066226959, + "learning_rate": 1.3530927835051548e-05, + "loss": 0.5153, + "step": 35337 + }, + { + "epoch": 1.9743833282118612, + "grad_norm": 0.4743556082248688, + "learning_rate": 1.3502913491707755e-05, + "loss": 0.4237, + "step": 35338 + }, + { + "epoch": 1.9744391988155432, + "grad_norm": 0.5748738646507263, + "learning_rate": 1.3474899148363963e-05, + "loss": 0.5516, + "step": 35339 + }, + { + "epoch": 1.9744950694192251, + "grad_norm": 0.35722091794013977, + "learning_rate": 1.344688480502017e-05, + "loss": 0.3207, + "step": 35340 + }, + { + "epoch": 1.9745509400229069, + "grad_norm": 0.41249996423721313, + "learning_rate": 1.341887046167638e-05, + "loss": 0.5002, + "step": 35341 + }, + { + "epoch": 1.9746068106265888, + "grad_norm": 0.48129087686538696, + "learning_rate": 1.3390856118332587e-05, + "loss": 0.367, + "step": 35342 + }, + { + "epoch": 1.9746626812302708, + "grad_norm": 0.33433809876441956, + "learning_rate": 1.3362841774988794e-05, + "loss": 0.4004, + "step": 35343 + }, + { + "epoch": 1.9747185518339525, + "grad_norm": 0.4104599952697754, + "learning_rate": 1.3334827431645002e-05, + "loss": 0.4424, + "step": 35344 + }, + { + "epoch": 1.9747744224376345, + "grad_norm": 0.6066135764122009, + "learning_rate": 1.3306813088301211e-05, + "loss": 0.4, + "step": 35345 + }, + { + "epoch": 1.9748302930413164, + "grad_norm": 0.42426109313964844, + "learning_rate": 1.3278798744957419e-05, + "loss": 0.3335, + "step": 35346 + }, + { + "epoch": 1.9748861636449981, + "grad_norm": 0.5885132551193237, + "learning_rate": 1.3250784401613626e-05, + "loss": 0.5459, + "step": 35347 + }, + { + "epoch": 1.97494203424868, + "grad_norm": 1.4446831941604614, + "learning_rate": 1.3222770058269834e-05, + "loss": 0.3559, + "step": 35348 + }, + { + "epoch": 1.974997904852362, + "grad_norm": 19.544252395629883, + "learning_rate": 1.3194755714926043e-05, + "loss": 0.6949, + "step": 35349 + }, + { + "epoch": 1.9750537754560438, + "grad_norm": 0.3774154782295227, + "learning_rate": 1.316674137158225e-05, + "loss": 0.3814, + "step": 35350 + }, + { + "epoch": 1.9751096460597257, + "grad_norm": 0.6812028288841248, + "learning_rate": 1.3138727028238458e-05, + "loss": 0.3895, + "step": 35351 + }, + { + "epoch": 1.9751655166634077, + "grad_norm": 2.523672103881836, + "learning_rate": 1.3110712684894665e-05, + "loss": 0.4171, + "step": 35352 + }, + { + "epoch": 1.9752213872670894, + "grad_norm": 0.35425665974617004, + "learning_rate": 1.3082698341550875e-05, + "loss": 0.4202, + "step": 35353 + }, + { + "epoch": 1.9752772578707711, + "grad_norm": 0.6819718480110168, + "learning_rate": 1.3054683998207082e-05, + "loss": 0.4516, + "step": 35354 + }, + { + "epoch": 1.9753331284744533, + "grad_norm": 0.41050103306770325, + "learning_rate": 1.302666965486329e-05, + "loss": 0.3578, + "step": 35355 + }, + { + "epoch": 1.975388999078135, + "grad_norm": 0.3947691321372986, + "learning_rate": 1.2998655311519497e-05, + "loss": 0.6227, + "step": 35356 + }, + { + "epoch": 1.9754448696818168, + "grad_norm": 0.40063437819480896, + "learning_rate": 1.2970640968175706e-05, + "loss": 0.4629, + "step": 35357 + }, + { + "epoch": 1.975500740285499, + "grad_norm": 0.4988287091255188, + "learning_rate": 1.2942626624831914e-05, + "loss": 0.4926, + "step": 35358 + }, + { + "epoch": 1.9755566108891807, + "grad_norm": 0.272115021944046, + "learning_rate": 1.2914612281488121e-05, + "loss": 0.3485, + "step": 35359 + }, + { + "epoch": 1.9756124814928624, + "grad_norm": 0.4375048577785492, + "learning_rate": 1.2886597938144329e-05, + "loss": 0.3466, + "step": 35360 + }, + { + "epoch": 1.9756683520965446, + "grad_norm": 0.3443738520145416, + "learning_rate": 1.2858583594800538e-05, + "loss": 0.4048, + "step": 35361 + }, + { + "epoch": 1.9757242227002263, + "grad_norm": 0.35768893361091614, + "learning_rate": 1.2830569251456746e-05, + "loss": 0.3427, + "step": 35362 + }, + { + "epoch": 1.975780093303908, + "grad_norm": 0.6116833686828613, + "learning_rate": 1.2802554908112953e-05, + "loss": 0.446, + "step": 35363 + }, + { + "epoch": 1.97583596390759, + "grad_norm": 1.4107757806777954, + "learning_rate": 1.2774540564769162e-05, + "loss": 0.447, + "step": 35364 + }, + { + "epoch": 1.975891834511272, + "grad_norm": 0.35649362206459045, + "learning_rate": 1.2746526221425371e-05, + "loss": 0.3725, + "step": 35365 + }, + { + "epoch": 1.9759477051149537, + "grad_norm": 0.3843716084957123, + "learning_rate": 1.2718511878081579e-05, + "loss": 0.3588, + "step": 35366 + }, + { + "epoch": 1.9760035757186356, + "grad_norm": 0.30584704875946045, + "learning_rate": 1.2690497534737786e-05, + "loss": 0.3947, + "step": 35367 + }, + { + "epoch": 1.9760594463223176, + "grad_norm": 0.3409220576286316, + "learning_rate": 1.2662483191393994e-05, + "loss": 0.3866, + "step": 35368 + }, + { + "epoch": 1.9761153169259993, + "grad_norm": 0.3389098644256592, + "learning_rate": 1.2634468848050203e-05, + "loss": 0.3942, + "step": 35369 + }, + { + "epoch": 1.9761711875296812, + "grad_norm": 0.3676108121871948, + "learning_rate": 1.260645450470641e-05, + "loss": 0.4862, + "step": 35370 + }, + { + "epoch": 1.9762270581333632, + "grad_norm": 0.4112115502357483, + "learning_rate": 1.2578440161362618e-05, + "loss": 0.3435, + "step": 35371 + }, + { + "epoch": 1.976282928737045, + "grad_norm": 0.7224171161651611, + "learning_rate": 1.2550425818018826e-05, + "loss": 0.4276, + "step": 35372 + }, + { + "epoch": 1.9763387993407269, + "grad_norm": 0.3431493639945984, + "learning_rate": 1.2522411474675035e-05, + "loss": 0.3857, + "step": 35373 + }, + { + "epoch": 1.9763946699444088, + "grad_norm": 0.4269741475582123, + "learning_rate": 1.2494397131331242e-05, + "loss": 0.3726, + "step": 35374 + }, + { + "epoch": 1.9764505405480906, + "grad_norm": 0.4641093313694, + "learning_rate": 1.246638278798745e-05, + "loss": 0.3357, + "step": 35375 + }, + { + "epoch": 1.9765064111517725, + "grad_norm": 0.36588814854621887, + "learning_rate": 1.2438368444643657e-05, + "loss": 0.494, + "step": 35376 + }, + { + "epoch": 1.9765622817554545, + "grad_norm": 0.6898702383041382, + "learning_rate": 1.2410354101299867e-05, + "loss": 0.3811, + "step": 35377 + }, + { + "epoch": 1.9766181523591362, + "grad_norm": 0.4298263490200043, + "learning_rate": 1.2382339757956074e-05, + "loss": 0.3881, + "step": 35378 + }, + { + "epoch": 1.9766740229628181, + "grad_norm": 0.36528128385543823, + "learning_rate": 1.2354325414612282e-05, + "loss": 0.3351, + "step": 35379 + }, + { + "epoch": 1.9767298935665, + "grad_norm": 0.49937954545021057, + "learning_rate": 1.232631107126849e-05, + "loss": 0.3451, + "step": 35380 + }, + { + "epoch": 1.9767857641701818, + "grad_norm": 0.4069293141365051, + "learning_rate": 1.2298296727924698e-05, + "loss": 0.3627, + "step": 35381 + }, + { + "epoch": 1.9768416347738638, + "grad_norm": 0.3120575249195099, + "learning_rate": 1.2270282384580906e-05, + "loss": 0.3523, + "step": 35382 + }, + { + "epoch": 1.9768975053775457, + "grad_norm": 0.31371694803237915, + "learning_rate": 1.2242268041237113e-05, + "loss": 0.3251, + "step": 35383 + }, + { + "epoch": 1.9769533759812274, + "grad_norm": 0.8830664157867432, + "learning_rate": 1.2214253697893321e-05, + "loss": 0.3907, + "step": 35384 + }, + { + "epoch": 1.9770092465849094, + "grad_norm": 0.3745341897010803, + "learning_rate": 1.218623935454953e-05, + "loss": 0.5001, + "step": 35385 + }, + { + "epoch": 1.9770651171885913, + "grad_norm": 0.4260234236717224, + "learning_rate": 1.2158225011205738e-05, + "loss": 0.4015, + "step": 35386 + }, + { + "epoch": 1.977120987792273, + "grad_norm": 0.6010230779647827, + "learning_rate": 1.2130210667861945e-05, + "loss": 0.3294, + "step": 35387 + }, + { + "epoch": 1.9771768583959548, + "grad_norm": 0.4387933015823364, + "learning_rate": 1.2102196324518153e-05, + "loss": 0.4213, + "step": 35388 + }, + { + "epoch": 1.977232728999637, + "grad_norm": 0.47194865345954895, + "learning_rate": 1.2074181981174362e-05, + "loss": 0.4308, + "step": 35389 + }, + { + "epoch": 1.9772885996033187, + "grad_norm": 0.44220203161239624, + "learning_rate": 1.204616763783057e-05, + "loss": 0.466, + "step": 35390 + }, + { + "epoch": 1.9773444702070004, + "grad_norm": 3.148615837097168, + "learning_rate": 1.2018153294486777e-05, + "loss": 0.3193, + "step": 35391 + }, + { + "epoch": 1.9774003408106826, + "grad_norm": 0.6938915848731995, + "learning_rate": 1.1990138951142984e-05, + "loss": 0.4817, + "step": 35392 + }, + { + "epoch": 1.9774562114143643, + "grad_norm": 0.35449910163879395, + "learning_rate": 1.1962124607799194e-05, + "loss": 0.3875, + "step": 35393 + }, + { + "epoch": 1.977512082018046, + "grad_norm": 0.31625989079475403, + "learning_rate": 1.1934110264455401e-05, + "loss": 0.3819, + "step": 35394 + }, + { + "epoch": 1.9775679526217282, + "grad_norm": 0.3193192183971405, + "learning_rate": 1.1906095921111609e-05, + "loss": 0.4218, + "step": 35395 + }, + { + "epoch": 1.97762382322541, + "grad_norm": 0.450026273727417, + "learning_rate": 1.1878081577767816e-05, + "loss": 0.4561, + "step": 35396 + }, + { + "epoch": 1.9776796938290917, + "grad_norm": 0.4834858179092407, + "learning_rate": 1.1850067234424025e-05, + "loss": 0.409, + "step": 35397 + }, + { + "epoch": 1.9777355644327737, + "grad_norm": 0.3597108721733093, + "learning_rate": 1.1822052891080235e-05, + "loss": 0.4624, + "step": 35398 + }, + { + "epoch": 1.9777914350364556, + "grad_norm": 0.350680947303772, + "learning_rate": 1.1794038547736442e-05, + "loss": 0.402, + "step": 35399 + }, + { + "epoch": 1.9778473056401373, + "grad_norm": 5.227167129516602, + "learning_rate": 1.176602420439265e-05, + "loss": 0.2985, + "step": 35400 + }, + { + "epoch": 1.9779031762438193, + "grad_norm": 0.3840988874435425, + "learning_rate": 1.1738009861048859e-05, + "loss": 0.3331, + "step": 35401 + }, + { + "epoch": 1.9779590468475012, + "grad_norm": 0.43913838267326355, + "learning_rate": 1.1709995517705066e-05, + "loss": 0.3876, + "step": 35402 + }, + { + "epoch": 1.978014917451183, + "grad_norm": 0.5022783875465393, + "learning_rate": 1.1681981174361274e-05, + "loss": 0.3208, + "step": 35403 + }, + { + "epoch": 1.978070788054865, + "grad_norm": 0.2944175899028778, + "learning_rate": 1.1653966831017481e-05, + "loss": 0.3156, + "step": 35404 + }, + { + "epoch": 1.9781266586585469, + "grad_norm": 0.45665857195854187, + "learning_rate": 1.162595248767369e-05, + "loss": 0.5157, + "step": 35405 + }, + { + "epoch": 1.9781825292622286, + "grad_norm": 0.31047606468200684, + "learning_rate": 1.1597938144329898e-05, + "loss": 0.4378, + "step": 35406 + }, + { + "epoch": 1.9782383998659105, + "grad_norm": 0.3672226071357727, + "learning_rate": 1.1569923800986105e-05, + "loss": 0.4717, + "step": 35407 + }, + { + "epoch": 1.9782942704695925, + "grad_norm": 0.38622599840164185, + "learning_rate": 1.1541909457642313e-05, + "loss": 0.3331, + "step": 35408 + }, + { + "epoch": 1.9783501410732742, + "grad_norm": 0.6430656313896179, + "learning_rate": 1.151389511429852e-05, + "loss": 0.5489, + "step": 35409 + }, + { + "epoch": 1.9784060116769562, + "grad_norm": 0.39556631445884705, + "learning_rate": 1.148588077095473e-05, + "loss": 0.3706, + "step": 35410 + }, + { + "epoch": 1.9784618822806381, + "grad_norm": 0.5180029273033142, + "learning_rate": 1.1457866427610937e-05, + "loss": 0.4843, + "step": 35411 + }, + { + "epoch": 1.9785177528843199, + "grad_norm": 0.46748024225234985, + "learning_rate": 1.1429852084267145e-05, + "loss": 0.383, + "step": 35412 + }, + { + "epoch": 1.9785736234880018, + "grad_norm": 0.40457406640052795, + "learning_rate": 1.1401837740923352e-05, + "loss": 0.3301, + "step": 35413 + }, + { + "epoch": 1.9786294940916838, + "grad_norm": 0.348421186208725, + "learning_rate": 1.1373823397579561e-05, + "loss": 0.3415, + "step": 35414 + }, + { + "epoch": 1.9786853646953655, + "grad_norm": 0.2776881456375122, + "learning_rate": 1.1345809054235769e-05, + "loss": 0.3162, + "step": 35415 + }, + { + "epoch": 1.9787412352990474, + "grad_norm": 0.41264453530311584, + "learning_rate": 1.1317794710891976e-05, + "loss": 0.435, + "step": 35416 + }, + { + "epoch": 1.9787971059027294, + "grad_norm": 0.32829952239990234, + "learning_rate": 1.1289780367548184e-05, + "loss": 0.3918, + "step": 35417 + }, + { + "epoch": 1.9788529765064111, + "grad_norm": 0.5285181403160095, + "learning_rate": 1.1261766024204393e-05, + "loss": 0.4506, + "step": 35418 + }, + { + "epoch": 1.978908847110093, + "grad_norm": 0.5209811329841614, + "learning_rate": 1.12337516808606e-05, + "loss": 0.3717, + "step": 35419 + }, + { + "epoch": 1.978964717713775, + "grad_norm": 0.6160881519317627, + "learning_rate": 1.1205737337516808e-05, + "loss": 0.3929, + "step": 35420 + }, + { + "epoch": 1.9790205883174568, + "grad_norm": 0.395696759223938, + "learning_rate": 1.1177722994173016e-05, + "loss": 0.3334, + "step": 35421 + }, + { + "epoch": 1.9790764589211385, + "grad_norm": 0.4388158917427063, + "learning_rate": 1.1149708650829225e-05, + "loss": 0.4027, + "step": 35422 + }, + { + "epoch": 1.9791323295248207, + "grad_norm": 0.8262894153594971, + "learning_rate": 1.1121694307485432e-05, + "loss": 0.4215, + "step": 35423 + }, + { + "epoch": 1.9791882001285024, + "grad_norm": 1.5190175771713257, + "learning_rate": 1.109367996414164e-05, + "loss": 0.5155, + "step": 35424 + }, + { + "epoch": 1.9792440707321841, + "grad_norm": 0.583678662776947, + "learning_rate": 1.1065665620797847e-05, + "loss": 0.4063, + "step": 35425 + }, + { + "epoch": 1.9792999413358663, + "grad_norm": 0.5916831493377686, + "learning_rate": 1.1037651277454057e-05, + "loss": 0.389, + "step": 35426 + }, + { + "epoch": 1.979355811939548, + "grad_norm": 0.40686699748039246, + "learning_rate": 1.1009636934110264e-05, + "loss": 0.3291, + "step": 35427 + }, + { + "epoch": 1.9794116825432297, + "grad_norm": 0.4145815372467041, + "learning_rate": 1.0981622590766472e-05, + "loss": 0.4141, + "step": 35428 + }, + { + "epoch": 1.979467553146912, + "grad_norm": 0.4327212870121002, + "learning_rate": 1.095360824742268e-05, + "loss": 0.3366, + "step": 35429 + }, + { + "epoch": 1.9795234237505936, + "grad_norm": 0.34650126099586487, + "learning_rate": 1.0925593904078888e-05, + "loss": 0.4629, + "step": 35430 + }, + { + "epoch": 1.9795792943542754, + "grad_norm": 0.37681716680526733, + "learning_rate": 1.0897579560735096e-05, + "loss": 0.5307, + "step": 35431 + }, + { + "epoch": 1.9796351649579573, + "grad_norm": 0.8809183835983276, + "learning_rate": 1.0869565217391305e-05, + "loss": 0.7693, + "step": 35432 + }, + { + "epoch": 1.9796910355616393, + "grad_norm": 0.47985610365867615, + "learning_rate": 1.0841550874047513e-05, + "loss": 0.3482, + "step": 35433 + }, + { + "epoch": 1.979746906165321, + "grad_norm": 0.28807491064071655, + "learning_rate": 1.0813536530703722e-05, + "loss": 0.317, + "step": 35434 + }, + { + "epoch": 1.979802776769003, + "grad_norm": 0.4803768992424011, + "learning_rate": 1.078552218735993e-05, + "loss": 0.4321, + "step": 35435 + }, + { + "epoch": 1.979858647372685, + "grad_norm": 0.34992507100105286, + "learning_rate": 1.0757507844016137e-05, + "loss": 0.3989, + "step": 35436 + }, + { + "epoch": 1.9799145179763666, + "grad_norm": 0.4196276366710663, + "learning_rate": 1.0729493500672344e-05, + "loss": 0.3747, + "step": 35437 + }, + { + "epoch": 1.9799703885800486, + "grad_norm": 0.510502815246582, + "learning_rate": 1.0701479157328554e-05, + "loss": 0.3791, + "step": 35438 + }, + { + "epoch": 1.9800262591837305, + "grad_norm": 0.4567020833492279, + "learning_rate": 1.0673464813984761e-05, + "loss": 0.4355, + "step": 35439 + }, + { + "epoch": 1.9800821297874123, + "grad_norm": 0.3724379241466522, + "learning_rate": 1.0645450470640969e-05, + "loss": 0.3704, + "step": 35440 + }, + { + "epoch": 1.9801380003910942, + "grad_norm": 4.14894962310791, + "learning_rate": 1.0617436127297176e-05, + "loss": 0.3729, + "step": 35441 + }, + { + "epoch": 1.9801938709947762, + "grad_norm": 0.27241992950439453, + "learning_rate": 1.0589421783953385e-05, + "loss": 0.3962, + "step": 35442 + }, + { + "epoch": 1.980249741598458, + "grad_norm": 0.578941285610199, + "learning_rate": 1.0561407440609593e-05, + "loss": 0.4418, + "step": 35443 + }, + { + "epoch": 1.9803056122021399, + "grad_norm": 0.3262757360935211, + "learning_rate": 1.05333930972658e-05, + "loss": 0.3751, + "step": 35444 + }, + { + "epoch": 1.9803614828058218, + "grad_norm": 0.3220713436603546, + "learning_rate": 1.0505378753922008e-05, + "loss": 0.3515, + "step": 35445 + }, + { + "epoch": 1.9804173534095035, + "grad_norm": 0.4874691069126129, + "learning_rate": 1.0477364410578217e-05, + "loss": 0.3675, + "step": 35446 + }, + { + "epoch": 1.9804732240131855, + "grad_norm": 0.353204607963562, + "learning_rate": 1.0449350067234425e-05, + "loss": 0.4066, + "step": 35447 + }, + { + "epoch": 1.9805290946168674, + "grad_norm": 0.4824671745300293, + "learning_rate": 1.0421335723890632e-05, + "loss": 0.3786, + "step": 35448 + }, + { + "epoch": 1.9805849652205492, + "grad_norm": 0.42167434096336365, + "learning_rate": 1.039332138054684e-05, + "loss": 0.4748, + "step": 35449 + }, + { + "epoch": 1.9806408358242311, + "grad_norm": 0.3735155165195465, + "learning_rate": 1.0365307037203049e-05, + "loss": 0.3619, + "step": 35450 + }, + { + "epoch": 1.980696706427913, + "grad_norm": 0.5023233294487, + "learning_rate": 1.0337292693859256e-05, + "loss": 0.4292, + "step": 35451 + }, + { + "epoch": 1.9807525770315948, + "grad_norm": 0.48237353563308716, + "learning_rate": 1.0309278350515464e-05, + "loss": 0.5513, + "step": 35452 + }, + { + "epoch": 1.9808084476352767, + "grad_norm": 0.4192497432231903, + "learning_rate": 1.0281264007171671e-05, + "loss": 0.4316, + "step": 35453 + }, + { + "epoch": 1.9808643182389587, + "grad_norm": 0.428616464138031, + "learning_rate": 1.025324966382788e-05, + "loss": 0.4565, + "step": 35454 + }, + { + "epoch": 1.9809201888426404, + "grad_norm": 0.6998700499534607, + "learning_rate": 1.0225235320484088e-05, + "loss": 0.4118, + "step": 35455 + }, + { + "epoch": 1.9809760594463222, + "grad_norm": 0.3009698688983917, + "learning_rate": 1.0197220977140295e-05, + "loss": 0.3928, + "step": 35456 + }, + { + "epoch": 1.9810319300500043, + "grad_norm": 0.3434673845767975, + "learning_rate": 1.0169206633796503e-05, + "loss": 0.4639, + "step": 35457 + }, + { + "epoch": 1.981087800653686, + "grad_norm": 0.38892677426338196, + "learning_rate": 1.0141192290452712e-05, + "loss": 0.4286, + "step": 35458 + }, + { + "epoch": 1.9811436712573678, + "grad_norm": 0.3358103632926941, + "learning_rate": 1.011317794710892e-05, + "loss": 0.2714, + "step": 35459 + }, + { + "epoch": 1.98119954186105, + "grad_norm": 0.3765749931335449, + "learning_rate": 1.0085163603765127e-05, + "loss": 0.4464, + "step": 35460 + }, + { + "epoch": 1.9812554124647317, + "grad_norm": 0.9778949022293091, + "learning_rate": 1.0057149260421335e-05, + "loss": 0.6097, + "step": 35461 + }, + { + "epoch": 1.9813112830684134, + "grad_norm": 0.4465412497520447, + "learning_rate": 1.0029134917077544e-05, + "loss": 0.3198, + "step": 35462 + }, + { + "epoch": 1.9813671536720956, + "grad_norm": 0.3509390652179718, + "learning_rate": 1.0001120573733751e-05, + "loss": 0.3813, + "step": 35463 + }, + { + "epoch": 1.9814230242757773, + "grad_norm": 0.32738298177719116, + "learning_rate": 9.973106230389959e-06, + "loss": 0.4735, + "step": 35464 + }, + { + "epoch": 1.981478894879459, + "grad_norm": 0.36386892199516296, + "learning_rate": 9.945091887046166e-06, + "loss": 0.3746, + "step": 35465 + }, + { + "epoch": 1.981534765483141, + "grad_norm": 0.3590438663959503, + "learning_rate": 9.917077543702377e-06, + "loss": 0.3868, + "step": 35466 + }, + { + "epoch": 1.981590636086823, + "grad_norm": 0.591023862361908, + "learning_rate": 9.889063200358585e-06, + "loss": 0.3664, + "step": 35467 + }, + { + "epoch": 1.9816465066905047, + "grad_norm": 1.2088146209716797, + "learning_rate": 9.861048857014792e-06, + "loss": 0.3944, + "step": 35468 + }, + { + "epoch": 1.9817023772941866, + "grad_norm": 0.3029077649116516, + "learning_rate": 9.833034513671e-06, + "loss": 0.4604, + "step": 35469 + }, + { + "epoch": 1.9817582478978686, + "grad_norm": 0.3527134656906128, + "learning_rate": 9.805020170327209e-06, + "loss": 0.4073, + "step": 35470 + }, + { + "epoch": 1.9818141185015503, + "grad_norm": 0.43339017033576965, + "learning_rate": 9.777005826983417e-06, + "loss": 0.4574, + "step": 35471 + }, + { + "epoch": 1.9818699891052323, + "grad_norm": 0.4482857286930084, + "learning_rate": 9.748991483639624e-06, + "loss": 0.4495, + "step": 35472 + }, + { + "epoch": 1.9819258597089142, + "grad_norm": 0.5071635246276855, + "learning_rate": 9.720977140295832e-06, + "loss": 0.4287, + "step": 35473 + }, + { + "epoch": 1.981981730312596, + "grad_norm": 0.6989907026290894, + "learning_rate": 9.69296279695204e-06, + "loss": 0.3816, + "step": 35474 + }, + { + "epoch": 1.982037600916278, + "grad_norm": 0.40269726514816284, + "learning_rate": 9.664948453608248e-06, + "loss": 0.4111, + "step": 35475 + }, + { + "epoch": 1.9820934715199598, + "grad_norm": 0.477560818195343, + "learning_rate": 9.636934110264456e-06, + "loss": 0.5542, + "step": 35476 + }, + { + "epoch": 1.9821493421236416, + "grad_norm": 0.3623398542404175, + "learning_rate": 9.608919766920663e-06, + "loss": 0.3269, + "step": 35477 + }, + { + "epoch": 1.9822052127273235, + "grad_norm": 0.6108466982841492, + "learning_rate": 9.580905423576873e-06, + "loss": 0.6467, + "step": 35478 + }, + { + "epoch": 1.9822610833310055, + "grad_norm": 0.3136386275291443, + "learning_rate": 9.55289108023308e-06, + "loss": 0.3225, + "step": 35479 + }, + { + "epoch": 1.9823169539346872, + "grad_norm": 0.5038298964500427, + "learning_rate": 9.524876736889288e-06, + "loss": 0.5162, + "step": 35480 + }, + { + "epoch": 1.9823728245383692, + "grad_norm": 0.42994096875190735, + "learning_rate": 9.496862393545495e-06, + "loss": 0.3614, + "step": 35481 + }, + { + "epoch": 1.9824286951420511, + "grad_norm": 1.2804393768310547, + "learning_rate": 9.468848050201704e-06, + "loss": 0.4144, + "step": 35482 + }, + { + "epoch": 1.9824845657457328, + "grad_norm": 1.4602397680282593, + "learning_rate": 9.440833706857912e-06, + "loss": 0.3813, + "step": 35483 + }, + { + "epoch": 1.9825404363494148, + "grad_norm": 1.1154212951660156, + "learning_rate": 9.41281936351412e-06, + "loss": 0.4055, + "step": 35484 + }, + { + "epoch": 1.9825963069530967, + "grad_norm": 0.41717204451560974, + "learning_rate": 9.384805020170327e-06, + "loss": 0.4318, + "step": 35485 + }, + { + "epoch": 1.9826521775567785, + "grad_norm": 0.2770557999610901, + "learning_rate": 9.356790676826536e-06, + "loss": 0.3335, + "step": 35486 + }, + { + "epoch": 1.9827080481604604, + "grad_norm": 0.46297991275787354, + "learning_rate": 9.328776333482744e-06, + "loss": 0.3827, + "step": 35487 + }, + { + "epoch": 1.9827639187641424, + "grad_norm": 0.4676477313041687, + "learning_rate": 9.300761990138951e-06, + "loss": 0.4628, + "step": 35488 + }, + { + "epoch": 1.982819789367824, + "grad_norm": 0.3904455602169037, + "learning_rate": 9.272747646795159e-06, + "loss": 0.3877, + "step": 35489 + }, + { + "epoch": 1.9828756599715058, + "grad_norm": 0.4714444577693939, + "learning_rate": 9.244733303451366e-06, + "loss": 0.3205, + "step": 35490 + }, + { + "epoch": 1.982931530575188, + "grad_norm": 0.5209878087043762, + "learning_rate": 9.216718960107575e-06, + "loss": 0.4531, + "step": 35491 + }, + { + "epoch": 1.9829874011788697, + "grad_norm": 0.44477784633636475, + "learning_rate": 9.188704616763783e-06, + "loss": 0.4364, + "step": 35492 + }, + { + "epoch": 1.9830432717825515, + "grad_norm": 0.4104710817337036, + "learning_rate": 9.16069027341999e-06, + "loss": 0.3715, + "step": 35493 + }, + { + "epoch": 1.9830991423862336, + "grad_norm": 0.8476564288139343, + "learning_rate": 9.132675930076198e-06, + "loss": 0.5627, + "step": 35494 + }, + { + "epoch": 1.9831550129899154, + "grad_norm": 0.8691827058792114, + "learning_rate": 9.104661586732407e-06, + "loss": 0.4253, + "step": 35495 + }, + { + "epoch": 1.983210883593597, + "grad_norm": 0.3529101312160492, + "learning_rate": 9.076647243388614e-06, + "loss": 0.4906, + "step": 35496 + }, + { + "epoch": 1.983266754197279, + "grad_norm": 0.49434030055999756, + "learning_rate": 9.048632900044822e-06, + "loss": 0.3273, + "step": 35497 + }, + { + "epoch": 1.983322624800961, + "grad_norm": 0.4417986571788788, + "learning_rate": 9.02061855670103e-06, + "loss": 0.3893, + "step": 35498 + }, + { + "epoch": 1.9833784954046427, + "grad_norm": 0.6164150834083557, + "learning_rate": 8.99260421335724e-06, + "loss": 0.4153, + "step": 35499 + }, + { + "epoch": 1.9834343660083247, + "grad_norm": 0.40743938088417053, + "learning_rate": 8.964589870013448e-06, + "loss": 0.4125, + "step": 35500 + }, + { + "epoch": 1.9834343660083247, + "eval_cer": 0.0830414716375896, + "eval_loss": 0.31228604912757874, + "eval_runtime": 56.2382, + "eval_samples_per_second": 80.693, + "eval_steps_per_second": 5.05, + "eval_wer": 0.33007457892237646, + "step": 35500 + }, + { + "epoch": 1.9834902366120066, + "grad_norm": 0.35909488797187805, + "learning_rate": 8.936575526669655e-06, + "loss": 0.3719, + "step": 35501 + }, + { + "epoch": 1.9835461072156884, + "grad_norm": 0.3382745385169983, + "learning_rate": 8.908561183325863e-06, + "loss": 0.3708, + "step": 35502 + }, + { + "epoch": 1.9836019778193703, + "grad_norm": 0.8339303731918335, + "learning_rate": 8.880546839982072e-06, + "loss": 0.498, + "step": 35503 + }, + { + "epoch": 1.9836578484230523, + "grad_norm": 0.4797423779964447, + "learning_rate": 8.85253249663828e-06, + "loss": 0.4083, + "step": 35504 + }, + { + "epoch": 1.983713719026734, + "grad_norm": 0.3672046363353729, + "learning_rate": 8.824518153294487e-06, + "loss": 0.3951, + "step": 35505 + }, + { + "epoch": 1.983769589630416, + "grad_norm": 0.5558099746704102, + "learning_rate": 8.796503809950695e-06, + "loss": 0.4776, + "step": 35506 + }, + { + "epoch": 1.983825460234098, + "grad_norm": 0.39278724789619446, + "learning_rate": 8.768489466606904e-06, + "loss": 0.4, + "step": 35507 + }, + { + "epoch": 1.9838813308377796, + "grad_norm": 0.43823352456092834, + "learning_rate": 8.740475123263111e-06, + "loss": 0.3349, + "step": 35508 + }, + { + "epoch": 1.9839372014414616, + "grad_norm": 0.4868876039981842, + "learning_rate": 8.712460779919319e-06, + "loss": 0.5445, + "step": 35509 + }, + { + "epoch": 1.9839930720451435, + "grad_norm": 6.821183204650879, + "learning_rate": 8.684446436575526e-06, + "loss": 0.3436, + "step": 35510 + }, + { + "epoch": 1.9840489426488253, + "grad_norm": 0.732600748538971, + "learning_rate": 8.656432093231736e-06, + "loss": 0.4389, + "step": 35511 + }, + { + "epoch": 1.9841048132525072, + "grad_norm": 0.36254218220710754, + "learning_rate": 8.628417749887943e-06, + "loss": 0.4585, + "step": 35512 + }, + { + "epoch": 1.9841606838561892, + "grad_norm": 4.028898239135742, + "learning_rate": 8.60040340654415e-06, + "loss": 0.4307, + "step": 35513 + }, + { + "epoch": 1.9842165544598709, + "grad_norm": 0.3494948744773865, + "learning_rate": 8.572389063200358e-06, + "loss": 0.435, + "step": 35514 + }, + { + "epoch": 1.9842724250635528, + "grad_norm": 0.46224936842918396, + "learning_rate": 8.544374719856567e-06, + "loss": 0.3859, + "step": 35515 + }, + { + "epoch": 1.9843282956672348, + "grad_norm": 0.4225270450115204, + "learning_rate": 8.516360376512775e-06, + "loss": 0.439, + "step": 35516 + }, + { + "epoch": 1.9843841662709165, + "grad_norm": 0.3549323081970215, + "learning_rate": 8.488346033168982e-06, + "loss": 0.3355, + "step": 35517 + }, + { + "epoch": 1.9844400368745985, + "grad_norm": 0.3221336305141449, + "learning_rate": 8.46033168982519e-06, + "loss": 0.3476, + "step": 35518 + }, + { + "epoch": 1.9844959074782804, + "grad_norm": 0.3674732744693756, + "learning_rate": 8.432317346481399e-06, + "loss": 0.3942, + "step": 35519 + }, + { + "epoch": 1.9845517780819621, + "grad_norm": 1.9651713371276855, + "learning_rate": 8.404303003137607e-06, + "loss": 0.3578, + "step": 35520 + }, + { + "epoch": 1.984607648685644, + "grad_norm": 0.8961592316627502, + "learning_rate": 8.376288659793814e-06, + "loss": 0.4066, + "step": 35521 + }, + { + "epoch": 1.984663519289326, + "grad_norm": 5.089651107788086, + "learning_rate": 8.348274316450022e-06, + "loss": 0.6254, + "step": 35522 + }, + { + "epoch": 1.9847193898930078, + "grad_norm": 0.426183819770813, + "learning_rate": 8.32025997310623e-06, + "loss": 0.3776, + "step": 35523 + }, + { + "epoch": 1.9847752604966895, + "grad_norm": 8.405844688415527, + "learning_rate": 8.292245629762438e-06, + "loss": 0.4637, + "step": 35524 + }, + { + "epoch": 1.9848311311003717, + "grad_norm": 1.7383610010147095, + "learning_rate": 8.264231286418646e-06, + "loss": 0.3886, + "step": 35525 + }, + { + "epoch": 1.9848870017040534, + "grad_norm": 0.7031764984130859, + "learning_rate": 8.236216943074853e-06, + "loss": 0.4605, + "step": 35526 + }, + { + "epoch": 1.9849428723077351, + "grad_norm": 1.245998740196228, + "learning_rate": 8.208202599731063e-06, + "loss": 0.2979, + "step": 35527 + }, + { + "epoch": 1.9849987429114173, + "grad_norm": 0.34439054131507874, + "learning_rate": 8.18018825638727e-06, + "loss": 0.4229, + "step": 35528 + }, + { + "epoch": 1.985054613515099, + "grad_norm": 0.3131870925426483, + "learning_rate": 8.152173913043478e-06, + "loss": 0.3173, + "step": 35529 + }, + { + "epoch": 1.9851104841187808, + "grad_norm": 0.9430181384086609, + "learning_rate": 8.124159569699685e-06, + "loss": 0.4843, + "step": 35530 + }, + { + "epoch": 1.9851663547224627, + "grad_norm": 0.30864274501800537, + "learning_rate": 8.096145226355894e-06, + "loss": 0.4327, + "step": 35531 + }, + { + "epoch": 1.9852222253261447, + "grad_norm": 1.1974354982376099, + "learning_rate": 8.068130883012102e-06, + "loss": 0.3159, + "step": 35532 + }, + { + "epoch": 1.9852780959298264, + "grad_norm": 1.0238425731658936, + "learning_rate": 8.040116539668311e-06, + "loss": 0.392, + "step": 35533 + }, + { + "epoch": 1.9853339665335084, + "grad_norm": 0.37550023198127747, + "learning_rate": 8.012102196324518e-06, + "loss": 0.4225, + "step": 35534 + }, + { + "epoch": 1.9853898371371903, + "grad_norm": 0.47601020336151123, + "learning_rate": 7.984087852980728e-06, + "loss": 0.4271, + "step": 35535 + }, + { + "epoch": 1.985445707740872, + "grad_norm": 0.5666808485984802, + "learning_rate": 7.956073509636935e-06, + "loss": 0.4317, + "step": 35536 + }, + { + "epoch": 1.985501578344554, + "grad_norm": 0.6952075958251953, + "learning_rate": 7.928059166293143e-06, + "loss": 0.3961, + "step": 35537 + }, + { + "epoch": 1.985557448948236, + "grad_norm": 0.7124210596084595, + "learning_rate": 7.90004482294935e-06, + "loss": 0.5169, + "step": 35538 + }, + { + "epoch": 1.9856133195519177, + "grad_norm": 0.8222705721855164, + "learning_rate": 7.87203047960556e-06, + "loss": 0.437, + "step": 35539 + }, + { + "epoch": 1.9856691901555996, + "grad_norm": 0.39075806736946106, + "learning_rate": 7.844016136261767e-06, + "loss": 0.4136, + "step": 35540 + }, + { + "epoch": 1.9857250607592816, + "grad_norm": 0.3291119635105133, + "learning_rate": 7.816001792917974e-06, + "loss": 0.354, + "step": 35541 + }, + { + "epoch": 1.9857809313629633, + "grad_norm": 0.504656970500946, + "learning_rate": 7.787987449574182e-06, + "loss": 0.4536, + "step": 35542 + }, + { + "epoch": 1.9858368019666452, + "grad_norm": 0.3573819398880005, + "learning_rate": 7.75997310623039e-06, + "loss": 0.3156, + "step": 35543 + }, + { + "epoch": 1.9858926725703272, + "grad_norm": 0.46166133880615234, + "learning_rate": 7.731958762886599e-06, + "loss": 0.4145, + "step": 35544 + }, + { + "epoch": 1.985948543174009, + "grad_norm": 0.6468780040740967, + "learning_rate": 7.703944419542806e-06, + "loss": 0.3131, + "step": 35545 + }, + { + "epoch": 1.9860044137776909, + "grad_norm": 0.4011163115501404, + "learning_rate": 7.675930076199014e-06, + "loss": 0.3849, + "step": 35546 + }, + { + "epoch": 1.9860602843813728, + "grad_norm": 0.4362679123878479, + "learning_rate": 7.647915732855221e-06, + "loss": 0.4621, + "step": 35547 + }, + { + "epoch": 1.9861161549850546, + "grad_norm": 0.37139442563056946, + "learning_rate": 7.61990138951143e-06, + "loss": 0.4438, + "step": 35548 + }, + { + "epoch": 1.9861720255887365, + "grad_norm": 0.4842933714389801, + "learning_rate": 7.591887046167638e-06, + "loss": 0.386, + "step": 35549 + }, + { + "epoch": 1.9862278961924185, + "grad_norm": 0.40010741353034973, + "learning_rate": 7.563872702823846e-06, + "loss": 0.3769, + "step": 35550 + }, + { + "epoch": 1.9862837667961002, + "grad_norm": 1.3058829307556152, + "learning_rate": 7.535858359480054e-06, + "loss": 0.4095, + "step": 35551 + }, + { + "epoch": 1.9863396373997821, + "grad_norm": 0.6068435311317444, + "learning_rate": 7.507844016136262e-06, + "loss": 0.3781, + "step": 35552 + }, + { + "epoch": 1.986395508003464, + "grad_norm": 0.2988492250442505, + "learning_rate": 7.47982967279247e-06, + "loss": 0.3714, + "step": 35553 + }, + { + "epoch": 1.9864513786071458, + "grad_norm": 0.4857252538204193, + "learning_rate": 7.451815329448678e-06, + "loss": 0.3266, + "step": 35554 + }, + { + "epoch": 1.9865072492108278, + "grad_norm": 0.4408091902732849, + "learning_rate": 7.4238009861048855e-06, + "loss": 0.6667, + "step": 35555 + }, + { + "epoch": 1.9865631198145097, + "grad_norm": 0.4366621673107147, + "learning_rate": 7.395786642761094e-06, + "loss": 0.5909, + "step": 35556 + }, + { + "epoch": 1.9866189904181915, + "grad_norm": 0.3932151794433594, + "learning_rate": 7.367772299417301e-06, + "loss": 0.4325, + "step": 35557 + }, + { + "epoch": 1.9866748610218732, + "grad_norm": 0.38174957036972046, + "learning_rate": 7.3397579560735106e-06, + "loss": 0.3223, + "step": 35558 + }, + { + "epoch": 1.9867307316255554, + "grad_norm": 0.9424348473548889, + "learning_rate": 7.311743612729718e-06, + "loss": 0.3732, + "step": 35559 + }, + { + "epoch": 1.986786602229237, + "grad_norm": 0.5926947593688965, + "learning_rate": 7.2837292693859264e-06, + "loss": 0.3328, + "step": 35560 + }, + { + "epoch": 1.9868424728329188, + "grad_norm": 0.7140779495239258, + "learning_rate": 7.255714926042134e-06, + "loss": 0.317, + "step": 35561 + }, + { + "epoch": 1.986898343436601, + "grad_norm": 0.91305011510849, + "learning_rate": 7.227700582698342e-06, + "loss": 0.3402, + "step": 35562 + }, + { + "epoch": 1.9869542140402827, + "grad_norm": 0.6470022201538086, + "learning_rate": 7.19968623935455e-06, + "loss": 0.3905, + "step": 35563 + }, + { + "epoch": 1.9870100846439644, + "grad_norm": 0.5955722332000732, + "learning_rate": 7.171671896010758e-06, + "loss": 0.3381, + "step": 35564 + }, + { + "epoch": 1.9870659552476464, + "grad_norm": 0.4546647369861603, + "learning_rate": 7.143657552666966e-06, + "loss": 0.3811, + "step": 35565 + }, + { + "epoch": 1.9871218258513283, + "grad_norm": 0.5097374320030212, + "learning_rate": 7.115643209323174e-06, + "loss": 0.486, + "step": 35566 + }, + { + "epoch": 1.98717769645501, + "grad_norm": 0.405180960893631, + "learning_rate": 7.0876288659793815e-06, + "loss": 0.2808, + "step": 35567 + }, + { + "epoch": 1.987233567058692, + "grad_norm": 0.31157517433166504, + "learning_rate": 7.05961452263559e-06, + "loss": 0.3709, + "step": 35568 + }, + { + "epoch": 1.987289437662374, + "grad_norm": 0.6494206190109253, + "learning_rate": 7.031600179291797e-06, + "loss": 0.3874, + "step": 35569 + }, + { + "epoch": 1.9873453082660557, + "grad_norm": 0.40976187586784363, + "learning_rate": 7.003585835948006e-06, + "loss": 0.4717, + "step": 35570 + }, + { + "epoch": 1.9874011788697377, + "grad_norm": 0.40248897671699524, + "learning_rate": 6.975571492604213e-06, + "loss": 0.4955, + "step": 35571 + }, + { + "epoch": 1.9874570494734196, + "grad_norm": 0.2785916328430176, + "learning_rate": 6.947557149260422e-06, + "loss": 0.311, + "step": 35572 + }, + { + "epoch": 1.9875129200771013, + "grad_norm": 0.32967856526374817, + "learning_rate": 6.919542805916629e-06, + "loss": 0.3775, + "step": 35573 + }, + { + "epoch": 1.9875687906807833, + "grad_norm": 0.4427777826786041, + "learning_rate": 6.891528462572837e-06, + "loss": 0.4187, + "step": 35574 + }, + { + "epoch": 1.9876246612844652, + "grad_norm": 0.4474479854106903, + "learning_rate": 6.863514119229046e-06, + "loss": 0.3862, + "step": 35575 + }, + { + "epoch": 1.987680531888147, + "grad_norm": 0.47385528683662415, + "learning_rate": 6.835499775885253e-06, + "loss": 0.4133, + "step": 35576 + }, + { + "epoch": 1.987736402491829, + "grad_norm": 1.0852779150009155, + "learning_rate": 6.807485432541462e-06, + "loss": 0.4392, + "step": 35577 + }, + { + "epoch": 1.9877922730955109, + "grad_norm": 0.3990510106086731, + "learning_rate": 6.779471089197669e-06, + "loss": 0.4381, + "step": 35578 + }, + { + "epoch": 1.9878481436991926, + "grad_norm": 2.7998342514038086, + "learning_rate": 6.751456745853878e-06, + "loss": 0.3924, + "step": 35579 + }, + { + "epoch": 1.9879040143028746, + "grad_norm": 0.5957509279251099, + "learning_rate": 6.723442402510085e-06, + "loss": 0.431, + "step": 35580 + }, + { + "epoch": 1.9879598849065565, + "grad_norm": 0.6058845520019531, + "learning_rate": 6.6954280591662934e-06, + "loss": 0.3978, + "step": 35581 + }, + { + "epoch": 1.9880157555102382, + "grad_norm": 1.357000470161438, + "learning_rate": 6.667413715822501e-06, + "loss": 0.5157, + "step": 35582 + }, + { + "epoch": 1.9880716261139202, + "grad_norm": 1.0181946754455566, + "learning_rate": 6.639399372478709e-06, + "loss": 0.5089, + "step": 35583 + }, + { + "epoch": 1.9881274967176021, + "grad_norm": 0.9358936548233032, + "learning_rate": 6.611385029134917e-06, + "loss": 0.5175, + "step": 35584 + }, + { + "epoch": 1.9881833673212839, + "grad_norm": 0.3722173571586609, + "learning_rate": 6.583370685791125e-06, + "loss": 0.2973, + "step": 35585 + }, + { + "epoch": 1.9882392379249658, + "grad_norm": 0.38294851779937744, + "learning_rate": 6.555356342447333e-06, + "loss": 0.4133, + "step": 35586 + }, + { + "epoch": 1.9882951085286478, + "grad_norm": 0.5658478736877441, + "learning_rate": 6.527341999103541e-06, + "loss": 0.364, + "step": 35587 + }, + { + "epoch": 1.9883509791323295, + "grad_norm": 0.4914742112159729, + "learning_rate": 6.4993276557597486e-06, + "loss": 0.4937, + "step": 35588 + }, + { + "epoch": 1.9884068497360115, + "grad_norm": 0.46896040439605713, + "learning_rate": 6.471313312415957e-06, + "loss": 0.4018, + "step": 35589 + }, + { + "epoch": 1.9884627203396934, + "grad_norm": 0.6467366218566895, + "learning_rate": 6.443298969072164e-06, + "loss": 0.4097, + "step": 35590 + }, + { + "epoch": 1.9885185909433751, + "grad_norm": 0.42334118485450745, + "learning_rate": 6.415284625728373e-06, + "loss": 0.3419, + "step": 35591 + }, + { + "epoch": 1.9885744615470569, + "grad_norm": 0.36591318249702454, + "learning_rate": 6.387270282384581e-06, + "loss": 0.321, + "step": 35592 + }, + { + "epoch": 1.988630332150739, + "grad_norm": 0.5230004787445068, + "learning_rate": 6.3592559390407895e-06, + "loss": 0.3848, + "step": 35593 + }, + { + "epoch": 1.9886862027544208, + "grad_norm": 0.8307427763938904, + "learning_rate": 6.331241595696997e-06, + "loss": 0.4345, + "step": 35594 + }, + { + "epoch": 1.9887420733581025, + "grad_norm": 0.4458899199962616, + "learning_rate": 6.303227252353205e-06, + "loss": 0.3683, + "step": 35595 + }, + { + "epoch": 1.9887979439617847, + "grad_norm": 0.3931237757205963, + "learning_rate": 6.275212909009413e-06, + "loss": 0.3495, + "step": 35596 + }, + { + "epoch": 1.9888538145654664, + "grad_norm": 0.3088665008544922, + "learning_rate": 6.247198565665621e-06, + "loss": 0.3581, + "step": 35597 + }, + { + "epoch": 1.9889096851691481, + "grad_norm": 0.594006359577179, + "learning_rate": 6.219184222321829e-06, + "loss": 0.4945, + "step": 35598 + }, + { + "epoch": 1.98896555577283, + "grad_norm": 0.6453005075454712, + "learning_rate": 6.191169878978037e-06, + "loss": 0.3901, + "step": 35599 + }, + { + "epoch": 1.989021426376512, + "grad_norm": 2.255666494369507, + "learning_rate": 6.163155535634245e-06, + "loss": 0.4174, + "step": 35600 + }, + { + "epoch": 1.9890772969801938, + "grad_norm": 0.331491082906723, + "learning_rate": 6.135141192290453e-06, + "loss": 0.3426, + "step": 35601 + }, + { + "epoch": 1.9891331675838757, + "grad_norm": 0.7598323225975037, + "learning_rate": 6.1071268489466605e-06, + "loss": 0.353, + "step": 35602 + }, + { + "epoch": 1.9891890381875577, + "grad_norm": 0.3529798984527588, + "learning_rate": 6.079112505602869e-06, + "loss": 0.3526, + "step": 35603 + }, + { + "epoch": 1.9892449087912394, + "grad_norm": 0.33721959590911865, + "learning_rate": 6.051098162259076e-06, + "loss": 0.308, + "step": 35604 + }, + { + "epoch": 1.9893007793949213, + "grad_norm": 0.7156705260276794, + "learning_rate": 6.023083818915285e-06, + "loss": 0.5021, + "step": 35605 + }, + { + "epoch": 1.9893566499986033, + "grad_norm": 0.422794371843338, + "learning_rate": 5.995069475571492e-06, + "loss": 0.4408, + "step": 35606 + }, + { + "epoch": 1.989412520602285, + "grad_norm": 0.3381930887699127, + "learning_rate": 5.9670551322277005e-06, + "loss": 0.383, + "step": 35607 + }, + { + "epoch": 1.989468391205967, + "grad_norm": 0.8939307332038879, + "learning_rate": 5.939040788883908e-06, + "loss": 0.5566, + "step": 35608 + }, + { + "epoch": 1.989524261809649, + "grad_norm": 1.0366275310516357, + "learning_rate": 5.911026445540117e-06, + "loss": 0.3701, + "step": 35609 + }, + { + "epoch": 1.9895801324133306, + "grad_norm": 0.49525532126426697, + "learning_rate": 5.883012102196325e-06, + "loss": 0.3493, + "step": 35610 + }, + { + "epoch": 1.9896360030170126, + "grad_norm": 0.4776369631290436, + "learning_rate": 5.854997758852533e-06, + "loss": 0.4398, + "step": 35611 + }, + { + "epoch": 1.9896918736206946, + "grad_norm": 0.4125860333442688, + "learning_rate": 5.826983415508741e-06, + "loss": 0.399, + "step": 35612 + }, + { + "epoch": 1.9897477442243763, + "grad_norm": 0.5332386493682861, + "learning_rate": 5.798969072164949e-06, + "loss": 0.537, + "step": 35613 + }, + { + "epoch": 1.9898036148280582, + "grad_norm": 0.49249401688575745, + "learning_rate": 5.7709547288211565e-06, + "loss": 0.4397, + "step": 35614 + }, + { + "epoch": 1.9898594854317402, + "grad_norm": 1.8167508840560913, + "learning_rate": 5.742940385477365e-06, + "loss": 0.415, + "step": 35615 + }, + { + "epoch": 1.989915356035422, + "grad_norm": 0.5566772818565369, + "learning_rate": 5.714926042133572e-06, + "loss": 0.467, + "step": 35616 + }, + { + "epoch": 1.9899712266391039, + "grad_norm": 0.771764874458313, + "learning_rate": 5.686911698789781e-06, + "loss": 0.397, + "step": 35617 + }, + { + "epoch": 1.9900270972427858, + "grad_norm": 0.3983426094055176, + "learning_rate": 5.658897355445988e-06, + "loss": 0.4231, + "step": 35618 + }, + { + "epoch": 1.9900829678464675, + "grad_norm": 0.5181986093521118, + "learning_rate": 5.630883012102197e-06, + "loss": 0.422, + "step": 35619 + }, + { + "epoch": 1.9901388384501495, + "grad_norm": 0.2970748543739319, + "learning_rate": 5.602868668758404e-06, + "loss": 0.3695, + "step": 35620 + }, + { + "epoch": 1.9901947090538314, + "grad_norm": 0.36532360315322876, + "learning_rate": 5.5748543254146125e-06, + "loss": 0.4305, + "step": 35621 + }, + { + "epoch": 1.9902505796575132, + "grad_norm": 1.2562991380691528, + "learning_rate": 5.54683998207082e-06, + "loss": 0.4233, + "step": 35622 + }, + { + "epoch": 1.9903064502611951, + "grad_norm": 2.1763346195220947, + "learning_rate": 5.518825638727028e-06, + "loss": 0.3236, + "step": 35623 + }, + { + "epoch": 1.990362320864877, + "grad_norm": 2.0094268321990967, + "learning_rate": 5.490811295383236e-06, + "loss": 0.454, + "step": 35624 + }, + { + "epoch": 1.9904181914685588, + "grad_norm": 0.3941686451435089, + "learning_rate": 5.462796952039444e-06, + "loss": 0.3921, + "step": 35625 + }, + { + "epoch": 1.9904740620722405, + "grad_norm": 0.5518307685852051, + "learning_rate": 5.4347826086956525e-06, + "loss": 0.463, + "step": 35626 + }, + { + "epoch": 1.9905299326759227, + "grad_norm": 0.6246505975723267, + "learning_rate": 5.406768265351861e-06, + "loss": 0.3704, + "step": 35627 + }, + { + "epoch": 1.9905858032796044, + "grad_norm": 0.31642913818359375, + "learning_rate": 5.378753922008068e-06, + "loss": 0.3497, + "step": 35628 + }, + { + "epoch": 1.9906416738832862, + "grad_norm": 0.34609153866767883, + "learning_rate": 5.350739578664277e-06, + "loss": 0.3261, + "step": 35629 + }, + { + "epoch": 1.9906975444869683, + "grad_norm": 0.48956969380378723, + "learning_rate": 5.322725235320484e-06, + "loss": 0.3987, + "step": 35630 + }, + { + "epoch": 1.99075341509065, + "grad_norm": 0.5008858442306519, + "learning_rate": 5.294710891976693e-06, + "loss": 0.4418, + "step": 35631 + }, + { + "epoch": 1.9908092856943318, + "grad_norm": 0.5567123889923096, + "learning_rate": 5.2666965486329e-06, + "loss": 0.5443, + "step": 35632 + }, + { + "epoch": 1.9908651562980137, + "grad_norm": 0.8035559058189392, + "learning_rate": 5.2386822052891085e-06, + "loss": 0.3826, + "step": 35633 + }, + { + "epoch": 1.9909210269016957, + "grad_norm": 1.7202852964401245, + "learning_rate": 5.210667861945316e-06, + "loss": 0.3738, + "step": 35634 + }, + { + "epoch": 1.9909768975053774, + "grad_norm": 1.1167091131210327, + "learning_rate": 5.182653518601524e-06, + "loss": 0.32, + "step": 35635 + }, + { + "epoch": 1.9910327681090594, + "grad_norm": 0.5962890982627869, + "learning_rate": 5.154639175257732e-06, + "loss": 0.4447, + "step": 35636 + }, + { + "epoch": 1.9910886387127413, + "grad_norm": 0.3742331564426422, + "learning_rate": 5.12662483191394e-06, + "loss": 0.3814, + "step": 35637 + }, + { + "epoch": 1.991144509316423, + "grad_norm": 0.3489070236682892, + "learning_rate": 5.098610488570148e-06, + "loss": 0.4326, + "step": 35638 + }, + { + "epoch": 1.991200379920105, + "grad_norm": 0.6165893077850342, + "learning_rate": 5.070596145226356e-06, + "loss": 0.4206, + "step": 35639 + }, + { + "epoch": 1.991256250523787, + "grad_norm": 0.37275221943855286, + "learning_rate": 5.042581801882564e-06, + "loss": 0.3771, + "step": 35640 + }, + { + "epoch": 1.9913121211274687, + "grad_norm": 1.564069390296936, + "learning_rate": 5.014567458538772e-06, + "loss": 0.448, + "step": 35641 + }, + { + "epoch": 1.9913679917311506, + "grad_norm": 0.3773597180843353, + "learning_rate": 4.9865531151949795e-06, + "loss": 0.5225, + "step": 35642 + }, + { + "epoch": 1.9914238623348326, + "grad_norm": 1.4172966480255127, + "learning_rate": 4.958538771851189e-06, + "loss": 0.4195, + "step": 35643 + }, + { + "epoch": 1.9914797329385143, + "grad_norm": 0.7818450331687927, + "learning_rate": 4.930524428507396e-06, + "loss": 0.3952, + "step": 35644 + }, + { + "epoch": 1.9915356035421963, + "grad_norm": 0.9646872282028198, + "learning_rate": 4.9025100851636045e-06, + "loss": 0.4364, + "step": 35645 + }, + { + "epoch": 1.9915914741458782, + "grad_norm": 0.49168506264686584, + "learning_rate": 4.874495741819812e-06, + "loss": 0.47, + "step": 35646 + }, + { + "epoch": 1.99164734474956, + "grad_norm": 0.34980347752571106, + "learning_rate": 4.84648139847602e-06, + "loss": 0.3407, + "step": 35647 + }, + { + "epoch": 1.991703215353242, + "grad_norm": 0.41109946370124817, + "learning_rate": 4.818467055132228e-06, + "loss": 0.3703, + "step": 35648 + }, + { + "epoch": 1.9917590859569239, + "grad_norm": 0.5629669427871704, + "learning_rate": 4.790452711788436e-06, + "loss": 0.5423, + "step": 35649 + }, + { + "epoch": 1.9918149565606056, + "grad_norm": 0.6993848085403442, + "learning_rate": 4.762438368444644e-06, + "loss": 0.4655, + "step": 35650 + }, + { + "epoch": 1.9918708271642875, + "grad_norm": 0.29591235518455505, + "learning_rate": 4.734424025100852e-06, + "loss": 0.4035, + "step": 35651 + }, + { + "epoch": 1.9919266977679695, + "grad_norm": 0.4148690700531006, + "learning_rate": 4.70640968175706e-06, + "loss": 0.4823, + "step": 35652 + }, + { + "epoch": 1.9919825683716512, + "grad_norm": 0.41839852929115295, + "learning_rate": 4.678395338413268e-06, + "loss": 0.3149, + "step": 35653 + }, + { + "epoch": 1.9920384389753332, + "grad_norm": 0.6558884978294373, + "learning_rate": 4.6503809950694755e-06, + "loss": 0.4419, + "step": 35654 + }, + { + "epoch": 1.9920943095790151, + "grad_norm": 2.13547945022583, + "learning_rate": 4.622366651725683e-06, + "loss": 0.3628, + "step": 35655 + }, + { + "epoch": 1.9921501801826969, + "grad_norm": 0.4250819981098175, + "learning_rate": 4.594352308381891e-06, + "loss": 0.4777, + "step": 35656 + }, + { + "epoch": 1.9922060507863788, + "grad_norm": 0.4190877079963684, + "learning_rate": 4.566337965038099e-06, + "loss": 0.4579, + "step": 35657 + }, + { + "epoch": 1.9922619213900608, + "grad_norm": 0.43166565895080566, + "learning_rate": 4.538323621694307e-06, + "loss": 0.4086, + "step": 35658 + }, + { + "epoch": 1.9923177919937425, + "grad_norm": 0.3695400357246399, + "learning_rate": 4.510309278350515e-06, + "loss": 0.3967, + "step": 35659 + }, + { + "epoch": 1.9923736625974242, + "grad_norm": 0.4528349041938782, + "learning_rate": 4.482294935006724e-06, + "loss": 0.4366, + "step": 35660 + }, + { + "epoch": 1.9924295332011064, + "grad_norm": 2.1584620475769043, + "learning_rate": 4.4542805916629315e-06, + "loss": 0.3667, + "step": 35661 + }, + { + "epoch": 1.9924854038047881, + "grad_norm": 0.755148708820343, + "learning_rate": 4.42626624831914e-06, + "loss": 0.5729, + "step": 35662 + }, + { + "epoch": 1.9925412744084698, + "grad_norm": 0.33959534764289856, + "learning_rate": 4.398251904975347e-06, + "loss": 0.3679, + "step": 35663 + }, + { + "epoch": 1.992597145012152, + "grad_norm": 0.3616888225078583, + "learning_rate": 4.370237561631556e-06, + "loss": 0.4089, + "step": 35664 + }, + { + "epoch": 1.9926530156158337, + "grad_norm": 0.3760923147201538, + "learning_rate": 4.342223218287763e-06, + "loss": 0.4127, + "step": 35665 + }, + { + "epoch": 1.9927088862195155, + "grad_norm": 0.3937622308731079, + "learning_rate": 4.3142088749439716e-06, + "loss": 0.4097, + "step": 35666 + }, + { + "epoch": 1.9927647568231974, + "grad_norm": 3.27081298828125, + "learning_rate": 4.286194531600179e-06, + "loss": 0.4338, + "step": 35667 + }, + { + "epoch": 1.9928206274268794, + "grad_norm": 0.5191303491592407, + "learning_rate": 4.258180188256387e-06, + "loss": 0.3219, + "step": 35668 + }, + { + "epoch": 1.992876498030561, + "grad_norm": 0.5500243902206421, + "learning_rate": 4.230165844912595e-06, + "loss": 0.4592, + "step": 35669 + }, + { + "epoch": 1.992932368634243, + "grad_norm": 0.30893978476524353, + "learning_rate": 4.202151501568803e-06, + "loss": 0.3874, + "step": 35670 + }, + { + "epoch": 1.992988239237925, + "grad_norm": 0.39090651273727417, + "learning_rate": 4.174137158225011e-06, + "loss": 0.3158, + "step": 35671 + }, + { + "epoch": 1.9930441098416067, + "grad_norm": 0.8593867421150208, + "learning_rate": 4.146122814881219e-06, + "loss": 0.4178, + "step": 35672 + }, + { + "epoch": 1.9930999804452887, + "grad_norm": 0.31048858165740967, + "learning_rate": 4.118108471537427e-06, + "loss": 0.4268, + "step": 35673 + }, + { + "epoch": 1.9931558510489706, + "grad_norm": 0.47987034916877747, + "learning_rate": 4.090094128193635e-06, + "loss": 0.3391, + "step": 35674 + }, + { + "epoch": 1.9932117216526524, + "grad_norm": 0.5533677339553833, + "learning_rate": 4.0620797848498425e-06, + "loss": 0.4306, + "step": 35675 + }, + { + "epoch": 1.9932675922563343, + "grad_norm": 0.45375755429267883, + "learning_rate": 4.034065441506051e-06, + "loss": 0.4273, + "step": 35676 + }, + { + "epoch": 1.9933234628600163, + "grad_norm": 0.5116195678710938, + "learning_rate": 4.006051098162259e-06, + "loss": 0.5615, + "step": 35677 + }, + { + "epoch": 1.993379333463698, + "grad_norm": 0.6796606779098511, + "learning_rate": 3.978036754818468e-06, + "loss": 0.588, + "step": 35678 + }, + { + "epoch": 1.99343520406738, + "grad_norm": 0.3177330791950226, + "learning_rate": 3.950022411474675e-06, + "loss": 0.3409, + "step": 35679 + }, + { + "epoch": 1.993491074671062, + "grad_norm": 0.8640015125274658, + "learning_rate": 3.9220080681308835e-06, + "loss": 0.4571, + "step": 35680 + }, + { + "epoch": 1.9935469452747436, + "grad_norm": 1.2481305599212646, + "learning_rate": 3.893993724787091e-06, + "loss": 0.4236, + "step": 35681 + }, + { + "epoch": 1.9936028158784256, + "grad_norm": 0.9822991490364075, + "learning_rate": 3.865979381443299e-06, + "loss": 0.3857, + "step": 35682 + }, + { + "epoch": 1.9936586864821075, + "grad_norm": 0.4168742001056671, + "learning_rate": 3.837965038099507e-06, + "loss": 0.4213, + "step": 35683 + }, + { + "epoch": 1.9937145570857893, + "grad_norm": 0.39503344893455505, + "learning_rate": 3.809950694755715e-06, + "loss": 0.4252, + "step": 35684 + }, + { + "epoch": 1.9937704276894712, + "grad_norm": 0.7102245688438416, + "learning_rate": 3.781936351411923e-06, + "loss": 0.4296, + "step": 35685 + }, + { + "epoch": 1.9938262982931532, + "grad_norm": 0.3205803632736206, + "learning_rate": 3.753922008068131e-06, + "loss": 0.4413, + "step": 35686 + }, + { + "epoch": 1.993882168896835, + "grad_norm": 0.7924851179122925, + "learning_rate": 3.725907664724339e-06, + "loss": 0.567, + "step": 35687 + }, + { + "epoch": 1.9939380395005168, + "grad_norm": 0.39628762006759644, + "learning_rate": 3.697893321380547e-06, + "loss": 0.4533, + "step": 35688 + }, + { + "epoch": 1.9939939101041988, + "grad_norm": 0.5665674209594727, + "learning_rate": 3.6698789780367553e-06, + "loss": 0.6353, + "step": 35689 + }, + { + "epoch": 1.9940497807078805, + "grad_norm": 0.3430337905883789, + "learning_rate": 3.6418646346929632e-06, + "loss": 0.3897, + "step": 35690 + }, + { + "epoch": 1.9941056513115623, + "grad_norm": 0.3386296033859253, + "learning_rate": 3.613850291349171e-06, + "loss": 0.3761, + "step": 35691 + }, + { + "epoch": 1.9941615219152444, + "grad_norm": 3.9185922145843506, + "learning_rate": 3.585835948005379e-06, + "loss": 0.4653, + "step": 35692 + }, + { + "epoch": 1.9942173925189262, + "grad_norm": 0.4390910267829895, + "learning_rate": 3.557821604661587e-06, + "loss": 0.5364, + "step": 35693 + }, + { + "epoch": 1.9942732631226079, + "grad_norm": 0.5199807286262512, + "learning_rate": 3.529807261317795e-06, + "loss": 0.3548, + "step": 35694 + }, + { + "epoch": 1.99432913372629, + "grad_norm": 0.44168663024902344, + "learning_rate": 3.501792917974003e-06, + "loss": 0.2867, + "step": 35695 + }, + { + "epoch": 1.9943850043299718, + "grad_norm": 0.34231945872306824, + "learning_rate": 3.473778574630211e-06, + "loss": 0.4586, + "step": 35696 + }, + { + "epoch": 1.9944408749336535, + "grad_norm": 0.3256148397922516, + "learning_rate": 3.4457642312864183e-06, + "loss": 0.4616, + "step": 35697 + }, + { + "epoch": 1.9944967455373357, + "grad_norm": 0.3383201062679291, + "learning_rate": 3.4177498879426267e-06, + "loss": 0.4552, + "step": 35698 + }, + { + "epoch": 1.9945526161410174, + "grad_norm": 0.3679378926753998, + "learning_rate": 3.3897355445988346e-06, + "loss": 0.4197, + "step": 35699 + }, + { + "epoch": 1.9946084867446991, + "grad_norm": 0.5196458697319031, + "learning_rate": 3.3617212012550425e-06, + "loss": 0.4495, + "step": 35700 + }, + { + "epoch": 1.994664357348381, + "grad_norm": 0.7566943764686584, + "learning_rate": 3.3337068579112505e-06, + "loss": 0.468, + "step": 35701 + }, + { + "epoch": 1.994720227952063, + "grad_norm": 0.3826761841773987, + "learning_rate": 3.3056925145674584e-06, + "loss": 0.4336, + "step": 35702 + }, + { + "epoch": 1.9947760985557448, + "grad_norm": 0.34002578258514404, + "learning_rate": 3.2776781712236663e-06, + "loss": 0.3124, + "step": 35703 + }, + { + "epoch": 1.9948319691594267, + "grad_norm": 0.4090016484260559, + "learning_rate": 3.2496638278798743e-06, + "loss": 0.3633, + "step": 35704 + }, + { + "epoch": 1.9948878397631087, + "grad_norm": 0.3951379060745239, + "learning_rate": 3.221649484536082e-06, + "loss": 0.3178, + "step": 35705 + }, + { + "epoch": 1.9949437103667904, + "grad_norm": 0.43041297793388367, + "learning_rate": 3.1936351411922906e-06, + "loss": 0.5499, + "step": 35706 + }, + { + "epoch": 1.9949995809704724, + "grad_norm": 1.334377646446228, + "learning_rate": 3.1656207978484985e-06, + "loss": 0.5113, + "step": 35707 + }, + { + "epoch": 1.9950554515741543, + "grad_norm": 0.4992438554763794, + "learning_rate": 3.1376064545047064e-06, + "loss": 0.4911, + "step": 35708 + }, + { + "epoch": 1.995111322177836, + "grad_norm": 0.5223579406738281, + "learning_rate": 3.1095921111609144e-06, + "loss": 0.4595, + "step": 35709 + }, + { + "epoch": 1.995167192781518, + "grad_norm": 0.7210167050361633, + "learning_rate": 3.0815777678171223e-06, + "loss": 0.3302, + "step": 35710 + }, + { + "epoch": 1.9952230633852, + "grad_norm": 0.46947163343429565, + "learning_rate": 3.0535634244733302e-06, + "loss": 0.4987, + "step": 35711 + }, + { + "epoch": 1.9952789339888817, + "grad_norm": 0.4218412935733795, + "learning_rate": 3.025549081129538e-06, + "loss": 0.4471, + "step": 35712 + }, + { + "epoch": 1.9953348045925636, + "grad_norm": 0.38259798288345337, + "learning_rate": 2.997534737785746e-06, + "loss": 0.3334, + "step": 35713 + }, + { + "epoch": 1.9953906751962456, + "grad_norm": 0.3752696216106415, + "learning_rate": 2.969520394441954e-06, + "loss": 0.5284, + "step": 35714 + }, + { + "epoch": 1.9954465457999273, + "grad_norm": 0.701297402381897, + "learning_rate": 2.9415060510981624e-06, + "loss": 0.4992, + "step": 35715 + }, + { + "epoch": 1.9955024164036093, + "grad_norm": 0.3770781457424164, + "learning_rate": 2.9134917077543703e-06, + "loss": 0.3387, + "step": 35716 + }, + { + "epoch": 1.9955582870072912, + "grad_norm": 0.4973782002925873, + "learning_rate": 2.8854773644105783e-06, + "loss": 0.3328, + "step": 35717 + }, + { + "epoch": 1.995614157610973, + "grad_norm": 0.38852933049201965, + "learning_rate": 2.857463021066786e-06, + "loss": 0.4157, + "step": 35718 + }, + { + "epoch": 1.995670028214655, + "grad_norm": 0.7471222877502441, + "learning_rate": 2.829448677722994e-06, + "loss": 0.3935, + "step": 35719 + }, + { + "epoch": 1.9957258988183368, + "grad_norm": 0.4504157602787018, + "learning_rate": 2.801434334379202e-06, + "loss": 0.3502, + "step": 35720 + }, + { + "epoch": 1.9957817694220186, + "grad_norm": 0.39333853125572205, + "learning_rate": 2.77341999103541e-06, + "loss": 0.3403, + "step": 35721 + }, + { + "epoch": 1.9958376400257005, + "grad_norm": 0.8432984948158264, + "learning_rate": 2.745405647691618e-06, + "loss": 0.5151, + "step": 35722 + }, + { + "epoch": 1.9958935106293825, + "grad_norm": 0.6508733630180359, + "learning_rate": 2.7173913043478263e-06, + "loss": 0.3335, + "step": 35723 + }, + { + "epoch": 1.9959493812330642, + "grad_norm": 0.44302603602409363, + "learning_rate": 2.689376961004034e-06, + "loss": 0.4978, + "step": 35724 + }, + { + "epoch": 1.996005251836746, + "grad_norm": 0.3624066710472107, + "learning_rate": 2.661362617660242e-06, + "loss": 0.3993, + "step": 35725 + }, + { + "epoch": 1.996061122440428, + "grad_norm": 0.6377848982810974, + "learning_rate": 2.63334827431645e-06, + "loss": 0.4133, + "step": 35726 + }, + { + "epoch": 1.9961169930441098, + "grad_norm": 0.28055113554000854, + "learning_rate": 2.605333930972658e-06, + "loss": 0.2633, + "step": 35727 + }, + { + "epoch": 1.9961728636477916, + "grad_norm": 1.1968340873718262, + "learning_rate": 2.577319587628866e-06, + "loss": 0.438, + "step": 35728 + }, + { + "epoch": 1.9962287342514737, + "grad_norm": 0.46781423687934875, + "learning_rate": 2.549305244285074e-06, + "loss": 0.3365, + "step": 35729 + }, + { + "epoch": 1.9962846048551555, + "grad_norm": 0.4113905727863312, + "learning_rate": 2.521290900941282e-06, + "loss": 0.4246, + "step": 35730 + }, + { + "epoch": 1.9963404754588372, + "grad_norm": 0.3434446156024933, + "learning_rate": 2.4932765575974897e-06, + "loss": 0.3459, + "step": 35731 + }, + { + "epoch": 1.9963963460625194, + "grad_norm": 0.3993925452232361, + "learning_rate": 2.465262214253698e-06, + "loss": 0.405, + "step": 35732 + }, + { + "epoch": 1.996452216666201, + "grad_norm": 0.44391128420829773, + "learning_rate": 2.437247870909906e-06, + "loss": 0.4618, + "step": 35733 + }, + { + "epoch": 1.9965080872698828, + "grad_norm": 0.33705461025238037, + "learning_rate": 2.409233527566114e-06, + "loss": 0.3303, + "step": 35734 + }, + { + "epoch": 1.9965639578735648, + "grad_norm": 0.7711524963378906, + "learning_rate": 2.381219184222322e-06, + "loss": 0.458, + "step": 35735 + }, + { + "epoch": 1.9966198284772467, + "grad_norm": 0.4266740381717682, + "learning_rate": 2.35320484087853e-06, + "loss": 0.3696, + "step": 35736 + }, + { + "epoch": 1.9966756990809285, + "grad_norm": 2.350389003753662, + "learning_rate": 2.3251904975347378e-06, + "loss": 0.8004, + "step": 35737 + }, + { + "epoch": 1.9967315696846104, + "grad_norm": 3.0154495239257812, + "learning_rate": 2.2971761541909457e-06, + "loss": 0.2821, + "step": 35738 + }, + { + "epoch": 1.9967874402882924, + "grad_norm": 0.35069483518600464, + "learning_rate": 2.2691618108471536e-06, + "loss": 0.4083, + "step": 35739 + }, + { + "epoch": 1.996843310891974, + "grad_norm": 0.7912352085113525, + "learning_rate": 2.241147467503362e-06, + "loss": 0.6244, + "step": 35740 + }, + { + "epoch": 1.996899181495656, + "grad_norm": 0.3499909043312073, + "learning_rate": 2.21313312415957e-06, + "loss": 0.3829, + "step": 35741 + }, + { + "epoch": 1.996955052099338, + "grad_norm": 0.6558854579925537, + "learning_rate": 2.185118780815778e-06, + "loss": 0.4792, + "step": 35742 + }, + { + "epoch": 1.9970109227030197, + "grad_norm": 0.34658315777778625, + "learning_rate": 2.1571044374719858e-06, + "loss": 0.3661, + "step": 35743 + }, + { + "epoch": 1.9970667933067017, + "grad_norm": 0.7110089659690857, + "learning_rate": 2.1290900941281937e-06, + "loss": 0.4445, + "step": 35744 + }, + { + "epoch": 1.9971226639103836, + "grad_norm": 0.5083531141281128, + "learning_rate": 2.1010757507844016e-06, + "loss": 0.3672, + "step": 35745 + }, + { + "epoch": 1.9971785345140654, + "grad_norm": 0.5138866901397705, + "learning_rate": 2.0730614074406096e-06, + "loss": 0.3997, + "step": 35746 + }, + { + "epoch": 1.9972344051177473, + "grad_norm": 0.4278574287891388, + "learning_rate": 2.0450470640968175e-06, + "loss": 0.4416, + "step": 35747 + }, + { + "epoch": 1.9972902757214293, + "grad_norm": 0.5405464768409729, + "learning_rate": 2.0170327207530254e-06, + "loss": 0.42, + "step": 35748 + }, + { + "epoch": 1.997346146325111, + "grad_norm": 0.36886394023895264, + "learning_rate": 1.989018377409234e-06, + "loss": 0.4112, + "step": 35749 + }, + { + "epoch": 1.997402016928793, + "grad_norm": 0.7545531392097473, + "learning_rate": 1.9610040340654417e-06, + "loss": 0.5417, + "step": 35750 + }, + { + "epoch": 1.9974578875324749, + "grad_norm": 0.358664333820343, + "learning_rate": 1.9329896907216497e-06, + "loss": 0.3856, + "step": 35751 + }, + { + "epoch": 1.9975137581361566, + "grad_norm": 0.5060238242149353, + "learning_rate": 1.9049753473778576e-06, + "loss": 0.4683, + "step": 35752 + }, + { + "epoch": 1.9975696287398386, + "grad_norm": 0.9404985904693604, + "learning_rate": 1.8769610040340655e-06, + "loss": 0.4266, + "step": 35753 + }, + { + "epoch": 1.9976254993435205, + "grad_norm": 2.033193588256836, + "learning_rate": 1.8489466606902735e-06, + "loss": 0.3735, + "step": 35754 + }, + { + "epoch": 1.9976813699472022, + "grad_norm": 0.6215830445289612, + "learning_rate": 1.8209323173464816e-06, + "loss": 0.4256, + "step": 35755 + }, + { + "epoch": 1.9977372405508842, + "grad_norm": 0.5656476616859436, + "learning_rate": 1.7929179740026895e-06, + "loss": 0.438, + "step": 35756 + }, + { + "epoch": 1.9977931111545661, + "grad_norm": 0.46710363030433655, + "learning_rate": 1.7649036306588975e-06, + "loss": 0.5694, + "step": 35757 + }, + { + "epoch": 1.9978489817582479, + "grad_norm": 0.4282217025756836, + "learning_rate": 1.7368892873151054e-06, + "loss": 0.5269, + "step": 35758 + }, + { + "epoch": 1.9979048523619296, + "grad_norm": 0.4144870340824127, + "learning_rate": 1.7088749439713133e-06, + "loss": 0.5584, + "step": 35759 + }, + { + "epoch": 1.9979607229656118, + "grad_norm": 0.4742681086063385, + "learning_rate": 1.6808606006275213e-06, + "loss": 0.436, + "step": 35760 + }, + { + "epoch": 1.9980165935692935, + "grad_norm": 0.35541436076164246, + "learning_rate": 1.6528462572837292e-06, + "loss": 0.4082, + "step": 35761 + }, + { + "epoch": 1.9980724641729752, + "grad_norm": 1.5273208618164062, + "learning_rate": 1.6248319139399371e-06, + "loss": 0.4301, + "step": 35762 + }, + { + "epoch": 1.9981283347766574, + "grad_norm": 0.8457242250442505, + "learning_rate": 1.5968175705961453e-06, + "loss": 0.336, + "step": 35763 + }, + { + "epoch": 1.9981842053803391, + "grad_norm": 0.4340015649795532, + "learning_rate": 1.5688032272523532e-06, + "loss": 0.4649, + "step": 35764 + }, + { + "epoch": 1.9982400759840209, + "grad_norm": 12.19112777709961, + "learning_rate": 1.5407888839085611e-06, + "loss": 0.3788, + "step": 35765 + }, + { + "epoch": 1.998295946587703, + "grad_norm": 0.39418140053749084, + "learning_rate": 1.512774540564769e-06, + "loss": 0.3861, + "step": 35766 + }, + { + "epoch": 1.9983518171913848, + "grad_norm": 0.4960680902004242, + "learning_rate": 1.484760197220977e-06, + "loss": 0.6901, + "step": 35767 + }, + { + "epoch": 1.9984076877950665, + "grad_norm": 0.3991592228412628, + "learning_rate": 1.4567458538771852e-06, + "loss": 0.4925, + "step": 35768 + }, + { + "epoch": 1.9984635583987485, + "grad_norm": 3.753572463989258, + "learning_rate": 1.428731510533393e-06, + "loss": 0.9797, + "step": 35769 + }, + { + "epoch": 1.9985194290024304, + "grad_norm": 0.5106097459793091, + "learning_rate": 1.400717167189601e-06, + "loss": 0.4008, + "step": 35770 + }, + { + "epoch": 1.9985752996061121, + "grad_norm": 0.7105627059936523, + "learning_rate": 1.372702823845809e-06, + "loss": 0.4728, + "step": 35771 + }, + { + "epoch": 1.998631170209794, + "grad_norm": 0.940258264541626, + "learning_rate": 1.344688480502017e-06, + "loss": 0.5268, + "step": 35772 + }, + { + "epoch": 1.998687040813476, + "grad_norm": 0.5970829129219055, + "learning_rate": 1.316674137158225e-06, + "loss": 0.5207, + "step": 35773 + }, + { + "epoch": 1.9987429114171578, + "grad_norm": 0.44508057832717896, + "learning_rate": 1.288659793814433e-06, + "loss": 0.432, + "step": 35774 + }, + { + "epoch": 1.9987987820208397, + "grad_norm": 0.34798336029052734, + "learning_rate": 1.260645450470641e-06, + "loss": 0.4107, + "step": 35775 + }, + { + "epoch": 1.9988546526245217, + "grad_norm": 0.3538365662097931, + "learning_rate": 1.232631107126849e-06, + "loss": 0.4546, + "step": 35776 + }, + { + "epoch": 1.9989105232282034, + "grad_norm": 0.493001252412796, + "learning_rate": 1.204616763783057e-06, + "loss": 0.3631, + "step": 35777 + }, + { + "epoch": 1.9989663938318853, + "grad_norm": 1.4015655517578125, + "learning_rate": 1.176602420439265e-06, + "loss": 0.347, + "step": 35778 + }, + { + "epoch": 1.9990222644355673, + "grad_norm": 0.8624745607376099, + "learning_rate": 1.1485880770954728e-06, + "loss": 0.4739, + "step": 35779 + }, + { + "epoch": 1.999078135039249, + "grad_norm": 0.31685954332351685, + "learning_rate": 1.120573733751681e-06, + "loss": 0.3431, + "step": 35780 + }, + { + "epoch": 1.999134005642931, + "grad_norm": 0.3837191164493561, + "learning_rate": 1.092559390407889e-06, + "loss": 0.3493, + "step": 35781 + }, + { + "epoch": 1.999189876246613, + "grad_norm": 0.46204572916030884, + "learning_rate": 1.0645450470640969e-06, + "loss": 0.394, + "step": 35782 + }, + { + "epoch": 1.9992457468502947, + "grad_norm": 0.583669126033783, + "learning_rate": 1.0365307037203048e-06, + "loss": 0.3222, + "step": 35783 + }, + { + "epoch": 1.9993016174539766, + "grad_norm": 0.48441991209983826, + "learning_rate": 1.0085163603765127e-06, + "loss": 0.3126, + "step": 35784 + }, + { + "epoch": 1.9993574880576586, + "grad_norm": 0.5357797741889954, + "learning_rate": 9.805020170327209e-07, + "loss": 0.336, + "step": 35785 + }, + { + "epoch": 1.9994133586613403, + "grad_norm": 0.43379923701286316, + "learning_rate": 9.524876736889288e-07, + "loss": 0.4133, + "step": 35786 + }, + { + "epoch": 1.9994692292650222, + "grad_norm": 0.4764619767665863, + "learning_rate": 9.244733303451367e-07, + "loss": 0.404, + "step": 35787 + }, + { + "epoch": 1.9995250998687042, + "grad_norm": 1.5264904499053955, + "learning_rate": 8.964589870013448e-07, + "loss": 0.4206, + "step": 35788 + }, + { + "epoch": 1.999580970472386, + "grad_norm": 0.3946182429790497, + "learning_rate": 8.684446436575527e-07, + "loss": 0.4515, + "step": 35789 + }, + { + "epoch": 1.9996368410760679, + "grad_norm": 0.5990998148918152, + "learning_rate": 8.404303003137606e-07, + "loss": 0.4207, + "step": 35790 + }, + { + "epoch": 1.9996927116797498, + "grad_norm": 0.30601683259010315, + "learning_rate": 8.124159569699686e-07, + "loss": 0.4167, + "step": 35791 + }, + { + "epoch": 1.9997485822834316, + "grad_norm": 0.5732196569442749, + "learning_rate": 7.844016136261766e-07, + "loss": 0.6548, + "step": 35792 + }, + { + "epoch": 1.9998044528871133, + "grad_norm": 0.37823235988616943, + "learning_rate": 7.563872702823845e-07, + "loss": 0.4705, + "step": 35793 + }, + { + "epoch": 1.9998603234907955, + "grad_norm": 1.2745391130447388, + "learning_rate": 7.283729269385926e-07, + "loss": 0.4359, + "step": 35794 + }, + { + "epoch": 1.9999161940944772, + "grad_norm": 1.0162562131881714, + "learning_rate": 7.003585835948005e-07, + "loss": 0.4505, + "step": 35795 + }, + { + "epoch": 1.999972064698159, + "grad_norm": 0.5632713437080383, + "learning_rate": 6.723442402510086e-07, + "loss": 0.3466, + "step": 35796 + } + ], + "logging_steps": 1.0, + "max_steps": 35796, + "num_input_tokens_seen": 0, + "num_train_epochs": 2, + "save_steps": 500, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 5.8041710762050716e+20, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +}