{ "best_metric": 3.3106162548065186, "best_model_checkpoint": "./output/checkpoint-4800", "epoch": 28.91566265060241, "eval_steps": 150, "global_step": 4800, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.060240963855421686, "grad_norm": 52.598548889160156, "learning_rate": 1.25e-05, "loss": 1.4587, "step": 10 }, { "epoch": 0.12048192771084337, "grad_norm": 1.3145289421081543, "learning_rate": 2.5e-05, "loss": 0.1245, "step": 20 }, { "epoch": 0.18072289156626506, "grad_norm": 47.875640869140625, "learning_rate": 3.75e-05, "loss": 0.0925, "step": 30 }, { "epoch": 0.24096385542168675, "grad_norm": 25.39703369140625, "learning_rate": 5e-05, "loss": 0.1134, "step": 40 }, { "epoch": 0.30120481927710846, "grad_norm": 22.82965660095215, "learning_rate": 6.25e-05, "loss": 0.1348, "step": 50 }, { "epoch": 0.3614457831325301, "grad_norm": 0.446526437997818, "learning_rate": 7.5e-05, "loss": 0.5273, "step": 60 }, { "epoch": 0.42168674698795183, "grad_norm": 50.92262649536133, "learning_rate": 8.75e-05, "loss": 0.2283, "step": 70 }, { "epoch": 0.4819277108433735, "grad_norm": 65.35910034179688, "learning_rate": 0.0001, "loss": 0.3627, "step": 80 }, { "epoch": 0.5421686746987951, "grad_norm": 8.19798755645752, "learning_rate": 0.00011250000000000001, "loss": 0.2596, "step": 90 }, { "epoch": 0.6024096385542169, "grad_norm": 141.95716857910156, "learning_rate": 0.000125, "loss": 0.6644, "step": 100 }, { "epoch": 0.6626506024096386, "grad_norm": 736.8707885742188, "learning_rate": 0.00012499871543489787, "loss": 0.9693, "step": 110 }, { "epoch": 0.7228915662650602, "grad_norm": 249.3212432861328, "learning_rate": 0.00012499486179239495, "loss": 1.4457, "step": 120 }, { "epoch": 0.7831325301204819, "grad_norm": 30.746965408325195, "learning_rate": 0.00012498843923089938, "loss": 1.4675, "step": 130 }, { "epoch": 0.8433734939759037, "grad_norm": 41.87055587768555, "learning_rate": 0.0001249794480144175, "loss": 3.2459, "step": 140 }, { "epoch": 0.9036144578313253, "grad_norm": 64.25176239013672, "learning_rate": 0.000124967888512543, "loss": 5.6311, "step": 150 }, { "epoch": 0.9036144578313253, "eval_loss": 6.333486080169678, "eval_runtime": 3.9268, "eval_samples_per_second": 9.422, "eval_steps_per_second": 9.422, "step": 150 }, { "epoch": 0.963855421686747, "grad_norm": 20.342771530151367, "learning_rate": 0.00012495376120044173, "loss": 6.6599, "step": 160 }, { "epoch": 1.0240963855421688, "grad_norm": 14.641746520996094, "learning_rate": 0.00012493706665883217, "loss": 5.9994, "step": 170 }, { "epoch": 1.0843373493975903, "grad_norm": 10.879859924316406, "learning_rate": 0.00012491780557396154, "loss": 5.408, "step": 180 }, { "epoch": 1.144578313253012, "grad_norm": 11.879966735839844, "learning_rate": 0.00012489597873757756, "loss": 5.4315, "step": 190 }, { "epoch": 1.2048192771084336, "grad_norm": 14.315322875976562, "learning_rate": 0.00012487158704689602, "loss": 5.7639, "step": 200 }, { "epoch": 1.2650602409638554, "grad_norm": 12.6843843460083, "learning_rate": 0.0001248446315045638, "loss": 5.4332, "step": 210 }, { "epoch": 1.3253012048192772, "grad_norm": 14.997090339660645, "learning_rate": 0.00012481511321861763, "loss": 5.7137, "step": 220 }, { "epoch": 1.3855421686746987, "grad_norm": 46041.421875, "learning_rate": 0.00012478303340243864, "loss": 7.4686, "step": 230 }, { "epoch": 1.4457831325301205, "grad_norm": 116.35897064208984, "learning_rate": 0.00012474839337470246, "loss": 5.9303, "step": 240 }, { "epoch": 1.5060240963855422, "grad_norm": 25.29050064086914, "learning_rate": 0.0001247111945593249, "loss": 5.6893, "step": 250 }, { "epoch": 1.5662650602409638, "grad_norm": 30.405521392822266, "learning_rate": 0.00012467143848540359, "loss": 5.4675, "step": 260 }, { "epoch": 1.6265060240963856, "grad_norm": 18.263336181640625, "learning_rate": 0.000124629126787155, "loss": 5.1431, "step": 270 }, { "epoch": 1.6867469879518073, "grad_norm": 30.267963409423828, "learning_rate": 0.00012458426120384738, "loss": 5.3887, "step": 280 }, { "epoch": 1.7469879518072289, "grad_norm": 16.499534606933594, "learning_rate": 0.00012453684357972906, "loss": 5.0634, "step": 290 }, { "epoch": 1.8072289156626506, "grad_norm": 13.954059600830078, "learning_rate": 0.00012448687586395289, "loss": 5.243, "step": 300 }, { "epoch": 1.8072289156626506, "eval_loss": 5.087627410888672, "eval_runtime": 3.8181, "eval_samples_per_second": 9.691, "eval_steps_per_second": 9.691, "step": 300 }, { "epoch": 1.8674698795180724, "grad_norm": 19.865482330322266, "learning_rate": 0.00012443436011049593, "loss": 4.7389, "step": 310 }, { "epoch": 1.927710843373494, "grad_norm": 46.38461685180664, "learning_rate": 0.0001243792984780751, "loss": 4.8623, "step": 320 }, { "epoch": 1.9879518072289155, "grad_norm": 21.573978424072266, "learning_rate": 0.00012432169323005853, "loss": 4.9039, "step": 330 }, { "epoch": 2.0481927710843375, "grad_norm": 21.620838165283203, "learning_rate": 0.00012426154673437223, "loss": 4.863, "step": 340 }, { "epoch": 2.108433734939759, "grad_norm": 64.13438415527344, "learning_rate": 0.00012419886146340314, "loss": 4.6726, "step": 350 }, { "epoch": 2.1686746987951806, "grad_norm": 32.13075256347656, "learning_rate": 0.0001241336399938972, "loss": 4.6064, "step": 360 }, { "epoch": 2.2289156626506026, "grad_norm": 62.58379364013672, "learning_rate": 0.00012406588500685355, "loss": 4.5968, "step": 370 }, { "epoch": 2.289156626506024, "grad_norm": 2010.449462890625, "learning_rate": 0.00012399559928741435, "loss": 5.0049, "step": 380 }, { "epoch": 2.3493975903614457, "grad_norm": 17.136829376220703, "learning_rate": 0.00012392278572475023, "loss": 5.0645, "step": 390 }, { "epoch": 2.4096385542168672, "grad_norm": 17.000410079956055, "learning_rate": 0.0001238474473119416, "loss": 5.2628, "step": 400 }, { "epoch": 2.4698795180722892, "grad_norm": 80.07215118408203, "learning_rate": 0.00012376958714585545, "loss": 4.8465, "step": 410 }, { "epoch": 2.5301204819277108, "grad_norm": 77.65228271484375, "learning_rate": 0.0001236892084270183, "loss": 4.5762, "step": 420 }, { "epoch": 2.5903614457831328, "grad_norm": 26.298023223876953, "learning_rate": 0.00012360631445948448, "loss": 4.6785, "step": 430 }, { "epoch": 2.6506024096385543, "grad_norm": 44.22318649291992, "learning_rate": 0.00012352090865070026, "loss": 5.1179, "step": 440 }, { "epoch": 2.710843373493976, "grad_norm": 70.8712387084961, "learning_rate": 0.00012343299451136397, "loss": 4.2638, "step": 450 }, { "epoch": 2.710843373493976, "eval_loss": 4.652755260467529, "eval_runtime": 3.8238, "eval_samples_per_second": 9.676, "eval_steps_per_second": 9.676, "step": 450 }, { "epoch": 2.7710843373493974, "grad_norm": 155.15814208984375, "learning_rate": 0.00012334257565528155, "loss": 4.7236, "step": 460 }, { "epoch": 2.8313253012048194, "grad_norm": 18.455474853515625, "learning_rate": 0.000123249655799218, "loss": 5.0244, "step": 470 }, { "epoch": 2.891566265060241, "grad_norm": 26.79853057861328, "learning_rate": 0.00012315423876274468, "loss": 5.1311, "step": 480 }, { "epoch": 2.9518072289156625, "grad_norm": 151.131591796875, "learning_rate": 0.0001230563284680822, "loss": 5.1015, "step": 490 }, { "epoch": 3.0120481927710845, "grad_norm": 19.262746810913086, "learning_rate": 0.00012295592893993935, "loss": 4.6265, "step": 500 }, { "epoch": 3.072289156626506, "grad_norm": 21.435794830322266, "learning_rate": 0.00012285304430534745, "loss": 4.8465, "step": 510 }, { "epoch": 3.1325301204819276, "grad_norm": 122.01215362548828, "learning_rate": 0.00012274767879349083, "loss": 4.2186, "step": 520 }, { "epoch": 3.1927710843373496, "grad_norm": 42.04435729980469, "learning_rate": 0.00012263983673553306, "loss": 4.9252, "step": 530 }, { "epoch": 3.253012048192771, "grad_norm": 37.1818962097168, "learning_rate": 0.0001225295225644387, "loss": 4.5153, "step": 540 }, { "epoch": 3.3132530120481927, "grad_norm": 40.17600631713867, "learning_rate": 0.0001224167408147913, "loss": 4.7539, "step": 550 }, { "epoch": 3.3734939759036147, "grad_norm": 11.658145904541016, "learning_rate": 0.0001223014961226068, "loss": 4.4217, "step": 560 }, { "epoch": 3.433734939759036, "grad_norm": 53.181190490722656, "learning_rate": 0.00012218379322514317, "loss": 4.1959, "step": 570 }, { "epoch": 3.4939759036144578, "grad_norm": 27.87154197692871, "learning_rate": 0.00012206363696070545, "loss": 4.1385, "step": 580 }, { "epoch": 3.5542168674698793, "grad_norm": 131.2779541015625, "learning_rate": 0.0001219410322684471, "loss": 3.9121, "step": 590 }, { "epoch": 3.6144578313253013, "grad_norm": 88.18590545654297, "learning_rate": 0.0001218159841881668, "loss": 4.5732, "step": 600 }, { "epoch": 3.6144578313253013, "eval_loss": 4.325500965118408, "eval_runtime": 3.8207, "eval_samples_per_second": 9.684, "eval_steps_per_second": 9.684, "step": 600 }, { "epoch": 3.674698795180723, "grad_norm": 137.2965087890625, "learning_rate": 0.00012168849786010133, "loss": 4.3792, "step": 610 }, { "epoch": 3.734939759036145, "grad_norm": 247.1873779296875, "learning_rate": 0.00012155857852471433, "loss": 3.6815, "step": 620 }, { "epoch": 3.7951807228915664, "grad_norm": 243.79896545410156, "learning_rate": 0.0001214262315224808, "loss": 4.3236, "step": 630 }, { "epoch": 3.855421686746988, "grad_norm": 81.76631927490234, "learning_rate": 0.00012129146229366766, "loss": 4.2099, "step": 640 }, { "epoch": 3.9156626506024095, "grad_norm": 83.27767944335938, "learning_rate": 0.00012115427637811003, "loss": 3.6707, "step": 650 }, { "epoch": 3.9759036144578315, "grad_norm": 13.061168670654297, "learning_rate": 0.00012101467941498357, "loss": 3.9052, "step": 660 }, { "epoch": 4.036144578313253, "grad_norm": 13.092235565185547, "learning_rate": 0.0001208726771425727, "loss": 4.3889, "step": 670 }, { "epoch": 4.096385542168675, "grad_norm": 254.2332305908203, "learning_rate": 0.00012072827539803463, "loss": 4.1295, "step": 680 }, { "epoch": 4.156626506024097, "grad_norm": 57.88543701171875, "learning_rate": 0.00012058148011715949, "loss": 3.8834, "step": 690 }, { "epoch": 4.216867469879518, "grad_norm": 12.289281845092773, "learning_rate": 0.00012043229733412636, "loss": 4.0815, "step": 700 }, { "epoch": 4.27710843373494, "grad_norm": 16.690227508544922, "learning_rate": 0.0001202807331812551, "loss": 4.2312, "step": 710 }, { "epoch": 4.337349397590361, "grad_norm": 88.23700714111328, "learning_rate": 0.00012012679388875441, "loss": 4.0464, "step": 720 }, { "epoch": 4.397590361445783, "grad_norm": 150.62051391601562, "learning_rate": 0.00011997048578446568, "loss": 4.5441, "step": 730 }, { "epoch": 4.457831325301205, "grad_norm": 50.538490295410156, "learning_rate": 0.00011981181529360282, "loss": 3.5866, "step": 740 }, { "epoch": 4.518072289156627, "grad_norm": 619.7337646484375, "learning_rate": 0.00011965078893848828, "loss": 4.3225, "step": 750 }, { "epoch": 4.518072289156627, "eval_loss": 4.134085178375244, "eval_runtime": 3.8365, "eval_samples_per_second": 9.644, "eval_steps_per_second": 9.644, "step": 750 }, { "epoch": 4.578313253012048, "grad_norm": 13.040363311767578, "learning_rate": 0.00011948741333828481, "loss": 4.1669, "step": 760 }, { "epoch": 4.63855421686747, "grad_norm": 26.328935623168945, "learning_rate": 0.00011932169520872344, "loss": 3.5026, "step": 770 }, { "epoch": 4.698795180722891, "grad_norm": 12.123993873596191, "learning_rate": 0.00011915364136182738, "loss": 3.7511, "step": 780 }, { "epoch": 4.759036144578313, "grad_norm": 34.736244201660156, "learning_rate": 0.0001189832587056321, "loss": 3.8989, "step": 790 }, { "epoch": 4.8192771084337345, "grad_norm": 50.64377212524414, "learning_rate": 0.00011881055424390119, "loss": 3.9904, "step": 800 }, { "epoch": 4.879518072289157, "grad_norm": 31.657766342163086, "learning_rate": 0.00011863553507583869, "loss": 3.8077, "step": 810 }, { "epoch": 4.9397590361445785, "grad_norm": 16.779949188232422, "learning_rate": 0.00011845820839579708, "loss": 3.9674, "step": 820 }, { "epoch": 5.0, "grad_norm": 14.840353965759277, "learning_rate": 0.00011827858149298162, "loss": 4.0888, "step": 830 }, { "epoch": 5.0602409638554215, "grad_norm": 61.29264450073242, "learning_rate": 0.00011809666175115075, "loss": 3.8837, "step": 840 }, { "epoch": 5.120481927710843, "grad_norm": 683.336181640625, "learning_rate": 0.00011791245664831251, "loss": 4.0266, "step": 850 }, { "epoch": 5.180722891566265, "grad_norm": 144.3917694091797, "learning_rate": 0.0001177259737564172, "loss": 4.3238, "step": 860 }, { "epoch": 5.240963855421687, "grad_norm": 16.853086471557617, "learning_rate": 0.00011753722074104613, "loss": 3.8228, "step": 870 }, { "epoch": 5.301204819277109, "grad_norm": 21.043733596801758, "learning_rate": 0.00011734620536109644, "loss": 4.0891, "step": 880 }, { "epoch": 5.36144578313253, "grad_norm": 27.56604766845703, "learning_rate": 0.00011715293546846223, "loss": 4.0359, "step": 890 }, { "epoch": 5.421686746987952, "grad_norm": 162.16494750976562, "learning_rate": 0.00011695741900771184, "loss": 4.3176, "step": 900 }, { "epoch": 5.421686746987952, "eval_loss": 3.8886878490448, "eval_runtime": 3.8151, "eval_samples_per_second": 9.698, "eval_steps_per_second": 9.698, "step": 900 }, { "epoch": 5.481927710843373, "grad_norm": 65.58177185058594, "learning_rate": 0.00011675966401576116, "loss": 3.6477, "step": 910 }, { "epoch": 5.542168674698795, "grad_norm": 80.09622955322266, "learning_rate": 0.00011655967862154335, "loss": 3.9318, "step": 920 }, { "epoch": 5.602409638554217, "grad_norm": 21.606958389282227, "learning_rate": 0.0001163574710456747, "loss": 4.1184, "step": 930 }, { "epoch": 5.662650602409639, "grad_norm": 139.5963134765625, "learning_rate": 0.00011615304960011663, "loss": 3.7183, "step": 940 }, { "epoch": 5.72289156626506, "grad_norm": 36.027828216552734, "learning_rate": 0.00011594642268783415, "loss": 3.6794, "step": 950 }, { "epoch": 5.783132530120482, "grad_norm": 8.31830883026123, "learning_rate": 0.00011573759880245027, "loss": 3.7807, "step": 960 }, { "epoch": 5.843373493975903, "grad_norm": 14.214261054992676, "learning_rate": 0.00011552658652789703, "loss": 4.1366, "step": 970 }, { "epoch": 5.903614457831325, "grad_norm": 14.497884750366211, "learning_rate": 0.00011531339453806258, "loss": 3.6411, "step": 980 }, { "epoch": 5.9638554216867465, "grad_norm": 16.32625389099121, "learning_rate": 0.00011509803159643458, "loss": 4.0479, "step": 990 }, { "epoch": 6.024096385542169, "grad_norm": 238.30307006835938, "learning_rate": 0.00011488050655574003, "loss": 3.9169, "step": 1000 }, { "epoch": 6.0843373493975905, "grad_norm": 12.584395408630371, "learning_rate": 0.00011466082835758141, "loss": 3.6927, "step": 1010 }, { "epoch": 6.144578313253012, "grad_norm": 34.47623825073242, "learning_rate": 0.000114439006032069, "loss": 3.524, "step": 1020 }, { "epoch": 6.204819277108434, "grad_norm": 14.770285606384277, "learning_rate": 0.00011421504869744978, "loss": 3.6258, "step": 1030 }, { "epoch": 6.265060240963855, "grad_norm": 24.838224411010742, "learning_rate": 0.0001139889655597326, "loss": 3.7966, "step": 1040 }, { "epoch": 6.325301204819277, "grad_norm": 18.04807472229004, "learning_rate": 0.00011376076591230974, "loss": 4.263, "step": 1050 }, { "epoch": 6.325301204819277, "eval_loss": 4.747550010681152, "eval_runtime": 3.8135, "eval_samples_per_second": 9.702, "eval_steps_per_second": 9.702, "step": 1050 }, { "epoch": 6.385542168674699, "grad_norm": 472.16766357421875, "learning_rate": 0.00011353045913557492, "loss": 4.1538, "step": 1060 }, { "epoch": 6.445783132530121, "grad_norm": 19.558197021484375, "learning_rate": 0.00011329805469653768, "loss": 4.3733, "step": 1070 }, { "epoch": 6.506024096385542, "grad_norm": 87.86190795898438, "learning_rate": 0.00011306356214843422, "loss": 3.6097, "step": 1080 }, { "epoch": 6.566265060240964, "grad_norm": 12.482193946838379, "learning_rate": 0.00011282699113033477, "loss": 4.1259, "step": 1090 }, { "epoch": 6.626506024096385, "grad_norm": 24.125751495361328, "learning_rate": 0.00011258835136674729, "loss": 3.9089, "step": 1100 }, { "epoch": 6.686746987951807, "grad_norm": 46.368099212646484, "learning_rate": 0.00011234765266721778, "loss": 3.6971, "step": 1110 }, { "epoch": 6.746987951807229, "grad_norm": 51.86185836791992, "learning_rate": 0.00011210490492592703, "loss": 3.8875, "step": 1120 }, { "epoch": 6.807228915662651, "grad_norm": 27.235157012939453, "learning_rate": 0.0001118601181212839, "loss": 4.3133, "step": 1130 }, { "epoch": 6.867469879518072, "grad_norm": 296.5423278808594, "learning_rate": 0.00011161330231551515, "loss": 4.1481, "step": 1140 }, { "epoch": 6.927710843373494, "grad_norm": 11.731728553771973, "learning_rate": 0.00011136446765425187, "loss": 4.192, "step": 1150 }, { "epoch": 6.9879518072289155, "grad_norm": 15.431469917297363, "learning_rate": 0.00011111362436611234, "loss": 3.8919, "step": 1160 }, { "epoch": 7.048192771084337, "grad_norm": 98.80924987792969, "learning_rate": 0.00011086078276228167, "loss": 3.7787, "step": 1170 }, { "epoch": 7.108433734939759, "grad_norm": 101.24623107910156, "learning_rate": 0.00011060595323608789, "loss": 3.8452, "step": 1180 }, { "epoch": 7.168674698795181, "grad_norm": 71.53185272216797, "learning_rate": 0.00011034914626257467, "loss": 3.874, "step": 1190 }, { "epoch": 7.228915662650603, "grad_norm": 66.60515594482422, "learning_rate": 0.0001100903723980709, "loss": 3.8769, "step": 1200 }, { "epoch": 7.228915662650603, "eval_loss": 3.916496992111206, "eval_runtime": 3.8065, "eval_samples_per_second": 9.72, "eval_steps_per_second": 9.72, "step": 1200 }, { "epoch": 7.289156626506024, "grad_norm": 22.101329803466797, "learning_rate": 0.00010982964227975658, "loss": 3.9201, "step": 1210 }, { "epoch": 7.349397590361446, "grad_norm": 14.925955772399902, "learning_rate": 0.00010956696662522569, "loss": 3.602, "step": 1220 }, { "epoch": 7.409638554216867, "grad_norm": 713.195556640625, "learning_rate": 0.00010930235623204551, "loss": 3.8217, "step": 1230 }, { "epoch": 7.469879518072289, "grad_norm": 51.607173919677734, "learning_rate": 0.00010903582197731294, "loss": 4.1866, "step": 1240 }, { "epoch": 7.530120481927711, "grad_norm": 10.283899307250977, "learning_rate": 0.00010876737481720722, "loss": 3.7481, "step": 1250 }, { "epoch": 7.590361445783133, "grad_norm": 257.76556396484375, "learning_rate": 0.0001084970257865397, "loss": 3.8203, "step": 1260 }, { "epoch": 7.650602409638554, "grad_norm": 14.792612075805664, "learning_rate": 0.00010822478599830008, "loss": 3.6593, "step": 1270 }, { "epoch": 7.710843373493976, "grad_norm": 303.11767578125, "learning_rate": 0.00010795066664319983, "loss": 3.753, "step": 1280 }, { "epoch": 7.771084337349397, "grad_norm": 32.58333206176758, "learning_rate": 0.00010767467898921197, "loss": 3.6544, "step": 1290 }, { "epoch": 7.831325301204819, "grad_norm": 17.639245986938477, "learning_rate": 0.00010739683438110797, "loss": 3.5928, "step": 1300 }, { "epoch": 7.891566265060241, "grad_norm": 10.270660400390625, "learning_rate": 0.00010711714423999145, "loss": 3.7059, "step": 1310 }, { "epoch": 7.951807228915663, "grad_norm": 16.441017150878906, "learning_rate": 0.00010683562006282861, "loss": 3.957, "step": 1320 }, { "epoch": 8.012048192771084, "grad_norm": 24.458555221557617, "learning_rate": 0.00010655227342197574, "loss": 4.0428, "step": 1330 }, { "epoch": 8.072289156626505, "grad_norm": 9.778626441955566, "learning_rate": 0.00010626711596470343, "loss": 4.0835, "step": 1340 }, { "epoch": 8.132530120481928, "grad_norm": 12.588565826416016, "learning_rate": 0.0001059801594127179, "loss": 3.523, "step": 1350 }, { "epoch": 8.132530120481928, "eval_loss": 3.8830435276031494, "eval_runtime": 3.8155, "eval_samples_per_second": 9.697, "eval_steps_per_second": 9.697, "step": 1350 }, { "epoch": 8.19277108433735, "grad_norm": 16.958955764770508, "learning_rate": 0.00010569141556167905, "loss": 3.8287, "step": 1360 }, { "epoch": 8.25301204819277, "grad_norm": 21.932573318481445, "learning_rate": 0.00010540089628071566, "loss": 3.7502, "step": 1370 }, { "epoch": 8.313253012048193, "grad_norm": 21.345739364624023, "learning_rate": 0.00010510861351193747, "loss": 4.3312, "step": 1380 }, { "epoch": 8.373493975903614, "grad_norm": 11.423644065856934, "learning_rate": 0.00010481457926994435, "loss": 4.3564, "step": 1390 }, { "epoch": 8.433734939759036, "grad_norm": 13.70189094543457, "learning_rate": 0.0001045188056413323, "loss": 3.2986, "step": 1400 }, { "epoch": 8.493975903614459, "grad_norm": 17.401643753051758, "learning_rate": 0.00010422130478419676, "loss": 3.9109, "step": 1410 }, { "epoch": 8.55421686746988, "grad_norm": 484.1435546875, "learning_rate": 0.00010392208892763269, "loss": 3.717, "step": 1420 }, { "epoch": 8.614457831325302, "grad_norm": 23.173097610473633, "learning_rate": 0.00010362117037123204, "loss": 3.4596, "step": 1430 }, { "epoch": 8.674698795180722, "grad_norm": 417.0069274902344, "learning_rate": 0.00010331856148457803, "loss": 3.722, "step": 1440 }, { "epoch": 8.734939759036145, "grad_norm": 55.5108528137207, "learning_rate": 0.00010301427470673678, "loss": 3.665, "step": 1450 }, { "epoch": 8.795180722891565, "grad_norm": 18.918964385986328, "learning_rate": 0.00010270832254574588, "loss": 3.6538, "step": 1460 }, { "epoch": 8.855421686746988, "grad_norm": 12.27521800994873, "learning_rate": 0.00010240071757810036, "loss": 3.8557, "step": 1470 }, { "epoch": 8.91566265060241, "grad_norm": 17.036134719848633, "learning_rate": 0.00010209147244823564, "loss": 3.448, "step": 1480 }, { "epoch": 8.975903614457831, "grad_norm": 9.00616455078125, "learning_rate": 0.00010178059986800773, "loss": 3.754, "step": 1490 }, { "epoch": 9.036144578313253, "grad_norm": 8.838912010192871, "learning_rate": 0.00010146811261617085, "loss": 3.9134, "step": 1500 }, { "epoch": 9.036144578313253, "eval_loss": 3.791945695877075, "eval_runtime": 3.8181, "eval_samples_per_second": 9.691, "eval_steps_per_second": 9.691, "step": 1500 }, { "epoch": 9.096385542168674, "grad_norm": 14.983063697814941, "learning_rate": 0.00010115402353785197, "loss": 3.8232, "step": 1510 }, { "epoch": 9.156626506024097, "grad_norm": 26.87819480895996, "learning_rate": 0.00010083834554402292, "loss": 3.7914, "step": 1520 }, { "epoch": 9.216867469879517, "grad_norm": 18.9931583404541, "learning_rate": 0.00010052109161096958, "loss": 3.7873, "step": 1530 }, { "epoch": 9.27710843373494, "grad_norm": 27333.392578125, "learning_rate": 0.00010020227477975852, "loss": 3.6831, "step": 1540 }, { "epoch": 9.337349397590362, "grad_norm": 13.019943237304688, "learning_rate": 9.9881908155701e-05, "loss": 4.0784, "step": 1550 }, { "epoch": 9.397590361445783, "grad_norm": 11.35181999206543, "learning_rate": 9.956000490781411e-05, "loss": 3.7456, "step": 1560 }, { "epoch": 9.457831325301205, "grad_norm": 28.795074462890625, "learning_rate": 9.923657826827957e-05, "loss": 3.6005, "step": 1570 }, { "epoch": 9.518072289156626, "grad_norm": 18.929288864135742, "learning_rate": 9.891164153189976e-05, "loss": 3.8908, "step": 1580 }, { "epoch": 9.578313253012048, "grad_norm": 236.947509765625, "learning_rate": 9.858520805555123e-05, "loss": 3.7814, "step": 1590 }, { "epoch": 9.638554216867469, "grad_norm": 30.308324813842773, "learning_rate": 9.825729125763561e-05, "loss": 3.8457, "step": 1600 }, { "epoch": 9.698795180722891, "grad_norm": 41.033992767333984, "learning_rate": 9.792790461752813e-05, "loss": 3.4012, "step": 1610 }, { "epoch": 9.759036144578314, "grad_norm": 25.953258514404297, "learning_rate": 9.759706167502343e-05, "loss": 3.5834, "step": 1620 }, { "epoch": 9.819277108433734, "grad_norm": 15.013717651367188, "learning_rate": 9.726477602977905e-05, "loss": 3.6479, "step": 1630 }, { "epoch": 9.879518072289157, "grad_norm": 39.97005081176758, "learning_rate": 9.69310613407564e-05, "loss": 3.3718, "step": 1640 }, { "epoch": 9.939759036144578, "grad_norm": 13.676387786865234, "learning_rate": 9.659593132565929e-05, "loss": 3.6831, "step": 1650 }, { "epoch": 9.939759036144578, "eval_loss": 3.70894193649292, "eval_runtime": 3.8103, "eval_samples_per_second": 9.711, "eval_steps_per_second": 9.711, "step": 1650 }, { "epoch": 10.0, "grad_norm": 30.147296905517578, "learning_rate": 9.625939976037002e-05, "loss": 3.4134, "step": 1660 }, { "epoch": 10.060240963855422, "grad_norm": 130.5504150390625, "learning_rate": 9.59214804783831e-05, "loss": 3.4339, "step": 1670 }, { "epoch": 10.120481927710843, "grad_norm": 8.657685279846191, "learning_rate": 9.558218737023671e-05, "loss": 4.0876, "step": 1680 }, { "epoch": 10.180722891566266, "grad_norm": 48.8166618347168, "learning_rate": 9.524153438294159e-05, "loss": 3.4347, "step": 1690 }, { "epoch": 10.240963855421686, "grad_norm": 15.538662910461426, "learning_rate": 9.489953551940783e-05, "loss": 3.7002, "step": 1700 }, { "epoch": 10.301204819277109, "grad_norm": 30.31039047241211, "learning_rate": 9.455620483786914e-05, "loss": 3.3832, "step": 1710 }, { "epoch": 10.36144578313253, "grad_norm": 118.9944839477539, "learning_rate": 9.421155645130514e-05, "loss": 3.3828, "step": 1720 }, { "epoch": 10.421686746987952, "grad_norm": 12.192248344421387, "learning_rate": 9.38656045268611e-05, "loss": 3.6482, "step": 1730 }, { "epoch": 10.481927710843374, "grad_norm": 18.287357330322266, "learning_rate": 9.351836328526563e-05, "loss": 3.3905, "step": 1740 }, { "epoch": 10.542168674698795, "grad_norm": 267.9325866699219, "learning_rate": 9.316984700024612e-05, "loss": 3.7533, "step": 1750 }, { "epoch": 10.602409638554217, "grad_norm": 542.4744262695312, "learning_rate": 9.2820069997942e-05, "loss": 3.8873, "step": 1760 }, { "epoch": 10.662650602409638, "grad_norm": 16.155866622924805, "learning_rate": 9.246904665631588e-05, "loss": 3.6396, "step": 1770 }, { "epoch": 10.72289156626506, "grad_norm": 7662.2783203125, "learning_rate": 9.211679140456242e-05, "loss": 3.1209, "step": 1780 }, { "epoch": 10.783132530120483, "grad_norm": 40.10587692260742, "learning_rate": 9.176331872251536e-05, "loss": 3.4137, "step": 1790 }, { "epoch": 10.843373493975903, "grad_norm": 12.35087776184082, "learning_rate": 9.140864314005222e-05, "loss": 3.6861, "step": 1800 }, { "epoch": 10.843373493975903, "eval_loss": 3.865417957305908, "eval_runtime": 3.8175, "eval_samples_per_second": 9.692, "eval_steps_per_second": 9.692, "step": 1800 }, { "epoch": 10.903614457831326, "grad_norm": 36.35682678222656, "learning_rate": 9.105277923649698e-05, "loss": 3.4116, "step": 1810 }, { "epoch": 10.963855421686747, "grad_norm": 9.996143341064453, "learning_rate": 9.06957416400209e-05, "loss": 3.5382, "step": 1820 }, { "epoch": 11.024096385542169, "grad_norm": 12.074798583984375, "learning_rate": 9.03375450270412e-05, "loss": 3.8299, "step": 1830 }, { "epoch": 11.08433734939759, "grad_norm": 14.15644359588623, "learning_rate": 8.997820412161764e-05, "loss": 3.4951, "step": 1840 }, { "epoch": 11.144578313253012, "grad_norm": 48.67642593383789, "learning_rate": 8.961773369484738e-05, "loss": 3.4306, "step": 1850 }, { "epoch": 11.204819277108435, "grad_norm": 18.517230987548828, "learning_rate": 8.925614856425786e-05, "loss": 3.451, "step": 1860 }, { "epoch": 11.265060240963855, "grad_norm": 11.237759590148926, "learning_rate": 8.88934635931975e-05, "loss": 3.4937, "step": 1870 }, { "epoch": 11.325301204819278, "grad_norm": 14.1367826461792, "learning_rate": 8.852969369022494e-05, "loss": 3.6886, "step": 1880 }, { "epoch": 11.385542168674698, "grad_norm": 9.110208511352539, "learning_rate": 8.816485380849613e-05, "loss": 3.3208, "step": 1890 }, { "epoch": 11.44578313253012, "grad_norm": 18.82927703857422, "learning_rate": 8.779895894514961e-05, "loss": 3.499, "step": 1900 }, { "epoch": 11.506024096385541, "grad_norm": 156.17605590820312, "learning_rate": 8.743202414069011e-05, "loss": 3.2002, "step": 1910 }, { "epoch": 11.566265060240964, "grad_norm": 14.175848960876465, "learning_rate": 8.706406447837023e-05, "loss": 3.7181, "step": 1920 }, { "epoch": 11.626506024096386, "grad_norm": 13.365341186523438, "learning_rate": 8.669509508357052e-05, "loss": 3.3462, "step": 1930 }, { "epoch": 11.686746987951807, "grad_norm": 20.62955093383789, "learning_rate": 8.632513112317761e-05, "loss": 3.4835, "step": 1940 }, { "epoch": 11.74698795180723, "grad_norm": 21.724872589111328, "learning_rate": 8.59541878049609e-05, "loss": 3.4933, "step": 1950 }, { "epoch": 11.74698795180723, "eval_loss": 3.6030540466308594, "eval_runtime": 3.8181, "eval_samples_per_second": 9.691, "eval_steps_per_second": 9.691, "step": 1950 }, { "epoch": 11.80722891566265, "grad_norm": 14.012041091918945, "learning_rate": 8.558228037694728e-05, "loss": 3.035, "step": 1960 }, { "epoch": 11.867469879518072, "grad_norm": 8.956927299499512, "learning_rate": 8.520942412679447e-05, "loss": 3.3299, "step": 1970 }, { "epoch": 11.927710843373493, "grad_norm": 8.330857276916504, "learning_rate": 8.483563438116257e-05, "loss": 3.1257, "step": 1980 }, { "epoch": 11.987951807228916, "grad_norm": 10.927474021911621, "learning_rate": 8.446092650508393e-05, "loss": 4.1444, "step": 1990 }, { "epoch": 12.048192771084338, "grad_norm": 10.727124214172363, "learning_rate": 8.408531590133172e-05, "loss": 3.3155, "step": 2000 }, { "epoch": 12.108433734939759, "grad_norm": 10.256277084350586, "learning_rate": 8.370881800978673e-05, "loss": 3.0151, "step": 2010 }, { "epoch": 12.168674698795181, "grad_norm": 61.829872131347656, "learning_rate": 8.333144830680262e-05, "loss": 3.1291, "step": 2020 }, { "epoch": 12.228915662650602, "grad_norm": 17.653553009033203, "learning_rate": 8.29532223045698e-05, "loss": 3.2804, "step": 2030 }, { "epoch": 12.289156626506024, "grad_norm": 13.582658767700195, "learning_rate": 8.257415555047785e-05, "loss": 3.2628, "step": 2040 }, { "epoch": 12.349397590361447, "grad_norm": 9.667671203613281, "learning_rate": 8.21942636264763e-05, "loss": 3.4144, "step": 2050 }, { "epoch": 12.409638554216867, "grad_norm": 3135.18798828125, "learning_rate": 8.181356214843422e-05, "loss": 3.3164, "step": 2060 }, { "epoch": 12.46987951807229, "grad_norm": 17.966447830200195, "learning_rate": 8.143206676549826e-05, "loss": 3.2951, "step": 2070 }, { "epoch": 12.53012048192771, "grad_norm": 10.217341423034668, "learning_rate": 8.10497931594494e-05, "loss": 3.2733, "step": 2080 }, { "epoch": 12.590361445783133, "grad_norm": 16.07942771911621, "learning_rate": 8.066675704405836e-05, "loss": 3.311, "step": 2090 }, { "epoch": 12.650602409638553, "grad_norm": 771.7515869140625, "learning_rate": 8.028297416443952e-05, "loss": 3.3552, "step": 2100 }, { "epoch": 12.650602409638553, "eval_loss": 3.570127010345459, "eval_runtime": 3.7827, "eval_samples_per_second": 9.781, "eval_steps_per_second": 9.781, "step": 2100 }, { "epoch": 12.710843373493976, "grad_norm": 14.326518058776855, "learning_rate": 7.989846029640397e-05, "loss": 3.6309, "step": 2110 }, { "epoch": 12.771084337349398, "grad_norm": 19.646242141723633, "learning_rate": 7.951323124581069e-05, "loss": 3.0986, "step": 2120 }, { "epoch": 12.831325301204819, "grad_norm": 49.34778594970703, "learning_rate": 7.91273028479172e-05, "loss": 3.486, "step": 2130 }, { "epoch": 12.891566265060241, "grad_norm": 21.21525001525879, "learning_rate": 7.874069096672831e-05, "loss": 3.5709, "step": 2140 }, { "epoch": 12.951807228915662, "grad_norm": 12.75536060333252, "learning_rate": 7.83534114943442e-05, "loss": 3.5097, "step": 2150 }, { "epoch": 13.012048192771084, "grad_norm": 11.8561372756958, "learning_rate": 7.796548035030715e-05, "loss": 3.7147, "step": 2160 }, { "epoch": 13.072289156626505, "grad_norm": 7.640058994293213, "learning_rate": 7.757691348094703e-05, "loss": 3.2838, "step": 2170 }, { "epoch": 13.132530120481928, "grad_norm": 1143.57177734375, "learning_rate": 7.718772685872595e-05, "loss": 3.5913, "step": 2180 }, { "epoch": 13.19277108433735, "grad_norm": 96.58565521240234, "learning_rate": 7.679793648158159e-05, "loss": 3.0851, "step": 2190 }, { "epoch": 13.25301204819277, "grad_norm": 35.802677154541016, "learning_rate": 7.640755837226965e-05, "loss": 3.1956, "step": 2200 }, { "epoch": 13.313253012048193, "grad_norm": 665.9241333007812, "learning_rate": 7.601660857770522e-05, "loss": 3.1286, "step": 2210 }, { "epoch": 13.373493975903614, "grad_norm": 40.26634979248047, "learning_rate": 7.562510316830308e-05, "loss": 3.2252, "step": 2220 }, { "epoch": 13.433734939759036, "grad_norm": 34.76996994018555, "learning_rate": 7.523305823731723e-05, "loss": 3.152, "step": 2230 }, { "epoch": 13.493975903614459, "grad_norm": 308.2239990234375, "learning_rate": 7.484048990017919e-05, "loss": 3.2297, "step": 2240 }, { "epoch": 13.55421686746988, "grad_norm": 39.217891693115234, "learning_rate": 7.444741429383578e-05, "loss": 3.4528, "step": 2250 }, { "epoch": 13.55421686746988, "eval_loss": 3.5550057888031006, "eval_runtime": 3.8001, "eval_samples_per_second": 9.737, "eval_steps_per_second": 9.737, "step": 2250 }, { "epoch": 13.614457831325302, "grad_norm": 11.670994758605957, "learning_rate": 7.405384757608555e-05, "loss": 3.4062, "step": 2260 }, { "epoch": 13.674698795180722, "grad_norm": 15.571825981140137, "learning_rate": 7.36598059249148e-05, "loss": 3.3256, "step": 2270 }, { "epoch": 13.734939759036145, "grad_norm": 21.08329963684082, "learning_rate": 7.326530553783243e-05, "loss": 3.4009, "step": 2280 }, { "epoch": 13.795180722891565, "grad_norm": 957.3087768554688, "learning_rate": 7.287036263120425e-05, "loss": 3.3748, "step": 2290 }, { "epoch": 13.855421686746988, "grad_norm": 20.058975219726562, "learning_rate": 7.247499343958621e-05, "loss": 3.5959, "step": 2300 }, { "epoch": 13.91566265060241, "grad_norm": 8.038925170898438, "learning_rate": 7.207921421505724e-05, "loss": 3.4845, "step": 2310 }, { "epoch": 13.975903614457831, "grad_norm": 12.520662307739258, "learning_rate": 7.168304122655113e-05, "loss": 3.0425, "step": 2320 }, { "epoch": 14.036144578313253, "grad_norm": 15.789307594299316, "learning_rate": 7.128649075918768e-05, "loss": 3.3566, "step": 2330 }, { "epoch": 14.096385542168674, "grad_norm": 14.215753555297852, "learning_rate": 7.088957911360347e-05, "loss": 3.3938, "step": 2340 }, { "epoch": 14.156626506024097, "grad_norm": 51.611045837402344, "learning_rate": 7.049232260528163e-05, "loss": 3.3809, "step": 2350 }, { "epoch": 14.216867469879517, "grad_norm": 202.7542266845703, "learning_rate": 7.009473756388128e-05, "loss": 3.3729, "step": 2360 }, { "epoch": 14.27710843373494, "grad_norm": 10.259781837463379, "learning_rate": 6.969684033256622e-05, "loss": 4.1525, "step": 2370 }, { "epoch": 14.337349397590362, "grad_norm": 23.243528366088867, "learning_rate": 6.92986472673332e-05, "loss": 3.5824, "step": 2380 }, { "epoch": 14.397590361445783, "grad_norm": 9.191352844238281, "learning_rate": 6.890017473633946e-05, "loss": 3.3591, "step": 2390 }, { "epoch": 14.457831325301205, "grad_norm": 16.033109664916992, "learning_rate": 6.850143911923011e-05, "loss": 3.6075, "step": 2400 }, { "epoch": 14.457831325301205, "eval_loss": 3.5646708011627197, "eval_runtime": 3.7918, "eval_samples_per_second": 9.758, "eval_steps_per_second": 9.758, "step": 2400 }, { "epoch": 14.518072289156626, "grad_norm": 10.289210319519043, "learning_rate": 6.81024568064646e-05, "loss": 3.2997, "step": 2410 }, { "epoch": 14.578313253012048, "grad_norm": 10.703330039978027, "learning_rate": 6.770324419864309e-05, "loss": 3.1972, "step": 2420 }, { "epoch": 14.638554216867469, "grad_norm": 277.9293212890625, "learning_rate": 6.73038177058323e-05, "loss": 3.6841, "step": 2430 }, { "epoch": 14.698795180722891, "grad_norm": 108.26596069335938, "learning_rate": 6.690419374689087e-05, "loss": 3.3921, "step": 2440 }, { "epoch": 14.759036144578314, "grad_norm": 201.3561248779297, "learning_rate": 6.650438874879456e-05, "loss": 3.4155, "step": 2450 }, { "epoch": 14.819277108433734, "grad_norm": 68.6462631225586, "learning_rate": 6.61044191459609e-05, "loss": 3.2985, "step": 2460 }, { "epoch": 14.879518072289157, "grad_norm": 68.2948989868164, "learning_rate": 6.57043013795737e-05, "loss": 3.0743, "step": 2470 }, { "epoch": 14.939759036144578, "grad_norm": 47.222633361816406, "learning_rate": 6.530405189690719e-05, "loss": 3.1893, "step": 2480 }, { "epoch": 15.0, "grad_norm": 15.256728172302246, "learning_rate": 6.49036871506499e-05, "loss": 3.3537, "step": 2490 }, { "epoch": 15.060240963855422, "grad_norm": 18.81123161315918, "learning_rate": 6.450322359822846e-05, "loss": 3.3208, "step": 2500 }, { "epoch": 15.120481927710843, "grad_norm": 11.176192283630371, "learning_rate": 6.410267770113098e-05, "loss": 3.1341, "step": 2510 }, { "epoch": 15.180722891566266, "grad_norm": 39.81761932373047, "learning_rate": 6.370206592423045e-05, "loss": 2.8517, "step": 2520 }, { "epoch": 15.240963855421686, "grad_norm": 11.802648544311523, "learning_rate": 6.330140473510796e-05, "loss": 3.1888, "step": 2530 }, { "epoch": 15.301204819277109, "grad_norm": 51.57283401489258, "learning_rate": 6.29007106033757e-05, "loss": 3.4263, "step": 2540 }, { "epoch": 15.36144578313253, "grad_norm": 27.341453552246094, "learning_rate": 6.25e-05, "loss": 3.5018, "step": 2550 }, { "epoch": 15.36144578313253, "eval_loss": 3.581845283508301, "eval_runtime": 3.7845, "eval_samples_per_second": 9.777, "eval_steps_per_second": 9.777, "step": 2550 }, { "epoch": 15.421686746987952, "grad_norm": 34.628868103027344, "learning_rate": 6.20992893966243e-05, "loss": 3.4468, "step": 2560 }, { "epoch": 15.481927710843374, "grad_norm": 51.63805389404297, "learning_rate": 6.169859526489204e-05, "loss": 3.2905, "step": 2570 }, { "epoch": 15.542168674698795, "grad_norm": 1207.8072509765625, "learning_rate": 6.129793407576955e-05, "loss": 3.7135, "step": 2580 }, { "epoch": 15.602409638554217, "grad_norm": 88.09565734863281, "learning_rate": 6.089732229886904e-05, "loss": 3.9768, "step": 2590 }, { "epoch": 15.662650602409638, "grad_norm": 43.92035675048828, "learning_rate": 6.049677640177155e-05, "loss": 3.7418, "step": 2600 }, { "epoch": 15.72289156626506, "grad_norm": 158.6419219970703, "learning_rate": 6.00963128493501e-05, "loss": 3.327, "step": 2610 }, { "epoch": 15.783132530120483, "grad_norm": 271.9347229003906, "learning_rate": 5.969594810309284e-05, "loss": 3.4304, "step": 2620 }, { "epoch": 15.843373493975903, "grad_norm": 83.00497436523438, "learning_rate": 5.929569862042631e-05, "loss": 3.822, "step": 2630 }, { "epoch": 15.903614457831326, "grad_norm": 201.45443725585938, "learning_rate": 5.889558085403911e-05, "loss": 3.6166, "step": 2640 }, { "epoch": 15.963855421686747, "grad_norm": 378.865478515625, "learning_rate": 5.849561125120545e-05, "loss": 3.4505, "step": 2650 }, { "epoch": 16.02409638554217, "grad_norm": 24.07649040222168, "learning_rate": 5.809580625310912e-05, "loss": 3.3913, "step": 2660 }, { "epoch": 16.08433734939759, "grad_norm": 261.75494384765625, "learning_rate": 5.769618229416773e-05, "loss": 3.1857, "step": 2670 }, { "epoch": 16.14457831325301, "grad_norm": 362.3284912109375, "learning_rate": 5.7296755801356926e-05, "loss": 3.448, "step": 2680 }, { "epoch": 16.204819277108435, "grad_norm": 675.8724975585938, "learning_rate": 5.6897543193535414e-05, "loss": 3.5349, "step": 2690 }, { "epoch": 16.265060240963855, "grad_norm": 88.61731719970703, "learning_rate": 5.649856088076989e-05, "loss": 3.8733, "step": 2700 }, { "epoch": 16.265060240963855, "eval_loss": 3.6278464794158936, "eval_runtime": 3.8328, "eval_samples_per_second": 9.653, "eval_steps_per_second": 9.653, "step": 2700 }, { "epoch": 16.325301204819276, "grad_norm": 96.55177307128906, "learning_rate": 5.609982526366054e-05, "loss": 3.2823, "step": 2710 }, { "epoch": 16.3855421686747, "grad_norm": 44.386287689208984, "learning_rate": 5.570135273266683e-05, "loss": 3.2381, "step": 2720 }, { "epoch": 16.44578313253012, "grad_norm": 57.28308868408203, "learning_rate": 5.53031596674338e-05, "loss": 3.1898, "step": 2730 }, { "epoch": 16.50602409638554, "grad_norm": 5471.37109375, "learning_rate": 5.490526243611873e-05, "loss": 3.1008, "step": 2740 }, { "epoch": 16.566265060240966, "grad_norm": 73.38491821289062, "learning_rate": 5.450767739471837e-05, "loss": 3.4751, "step": 2750 }, { "epoch": 16.626506024096386, "grad_norm": 94.40169525146484, "learning_rate": 5.411042088639655e-05, "loss": 3.2401, "step": 2760 }, { "epoch": 16.686746987951807, "grad_norm": 31.844371795654297, "learning_rate": 5.371350924081234e-05, "loss": 3.5455, "step": 2770 }, { "epoch": 16.746987951807228, "grad_norm": 58.31970977783203, "learning_rate": 5.331695877344888e-05, "loss": 3.489, "step": 2780 }, { "epoch": 16.80722891566265, "grad_norm": 47.033939361572266, "learning_rate": 5.292078578494275e-05, "loss": 3.007, "step": 2790 }, { "epoch": 16.867469879518072, "grad_norm": 117.59281921386719, "learning_rate": 5.2525006560413816e-05, "loss": 3.6114, "step": 2800 }, { "epoch": 16.927710843373493, "grad_norm": 160.5786590576172, "learning_rate": 5.212963736879578e-05, "loss": 3.3447, "step": 2810 }, { "epoch": 16.987951807228917, "grad_norm": 375.9039611816406, "learning_rate": 5.173469446216757e-05, "loss": 3.5652, "step": 2820 }, { "epoch": 17.048192771084338, "grad_norm": 368.878662109375, "learning_rate": 5.134019407508521e-05, "loss": 3.4137, "step": 2830 }, { "epoch": 17.10843373493976, "grad_norm": 20.2479305267334, "learning_rate": 5.0946152423914456e-05, "loss": 3.4078, "step": 2840 }, { "epoch": 17.16867469879518, "grad_norm": 300.1862487792969, "learning_rate": 5.0552585706164246e-05, "loss": 3.1655, "step": 2850 }, { "epoch": 17.16867469879518, "eval_loss": 3.585510730743408, "eval_runtime": 3.839, "eval_samples_per_second": 9.638, "eval_steps_per_second": 9.638, "step": 2850 }, { "epoch": 17.228915662650603, "grad_norm": 170.2447052001953, "learning_rate": 5.015951009982081e-05, "loss": 3.5166, "step": 2860 }, { "epoch": 17.289156626506024, "grad_norm": 696.30712890625, "learning_rate": 4.976694176268278e-05, "loss": 3.5968, "step": 2870 }, { "epoch": 17.349397590361445, "grad_norm": 68.88721466064453, "learning_rate": 4.937489683169692e-05, "loss": 3.0614, "step": 2880 }, { "epoch": 17.40963855421687, "grad_norm": 43.65079116821289, "learning_rate": 4.8983391422294786e-05, "loss": 3.2525, "step": 2890 }, { "epoch": 17.46987951807229, "grad_norm": 19.629438400268555, "learning_rate": 4.8592441627730355e-05, "loss": 3.1851, "step": 2900 }, { "epoch": 17.53012048192771, "grad_norm": 176.78993225097656, "learning_rate": 4.820206351841842e-05, "loss": 3.1973, "step": 2910 }, { "epoch": 17.59036144578313, "grad_norm": 122.77618408203125, "learning_rate": 4.781227314127405e-05, "loss": 3.1958, "step": 2920 }, { "epoch": 17.650602409638555, "grad_norm": 128.48361206054688, "learning_rate": 4.7423086519052966e-05, "loss": 3.2149, "step": 2930 }, { "epoch": 17.710843373493976, "grad_norm": 757.3467407226562, "learning_rate": 4.703451964969287e-05, "loss": 3.136, "step": 2940 }, { "epoch": 17.771084337349397, "grad_norm": 31.354238510131836, "learning_rate": 4.66465885056558e-05, "loss": 3.4783, "step": 2950 }, { "epoch": 17.83132530120482, "grad_norm": 38.50337600708008, "learning_rate": 4.62593090332717e-05, "loss": 3.7484, "step": 2960 }, { "epoch": 17.89156626506024, "grad_norm": 25.400075912475586, "learning_rate": 4.587269715208281e-05, "loss": 3.228, "step": 2970 }, { "epoch": 17.951807228915662, "grad_norm": 33.36684036254883, "learning_rate": 4.5486768754189305e-05, "loss": 3.1605, "step": 2980 }, { "epoch": 18.012048192771083, "grad_norm": 55.69667434692383, "learning_rate": 4.510153970359606e-05, "loss": 3.2666, "step": 2990 }, { "epoch": 18.072289156626507, "grad_norm": 34.30318069458008, "learning_rate": 4.4717025835560476e-05, "loss": 3.7476, "step": 3000 }, { "epoch": 18.072289156626507, "eval_loss": 3.4825525283813477, "eval_runtime": 3.828, "eval_samples_per_second": 9.666, "eval_steps_per_second": 9.666, "step": 3000 }, { "epoch": 18.132530120481928, "grad_norm": 50.15314865112305, "learning_rate": 4.433324295594166e-05, "loss": 3.2661, "step": 3010 }, { "epoch": 18.19277108433735, "grad_norm": 246.28211975097656, "learning_rate": 4.3950206840550585e-05, "loss": 3.3131, "step": 3020 }, { "epoch": 18.253012048192772, "grad_norm": 21.07671546936035, "learning_rate": 4.3567933234501746e-05, "loss": 3.6945, "step": 3030 }, { "epoch": 18.313253012048193, "grad_norm": 45.0478401184082, "learning_rate": 4.318643785156579e-05, "loss": 3.2986, "step": 3040 }, { "epoch": 18.373493975903614, "grad_norm": 26.682987213134766, "learning_rate": 4.280573637352371e-05, "loss": 3.2984, "step": 3050 }, { "epoch": 18.433734939759034, "grad_norm": 38.422367095947266, "learning_rate": 4.242584444952216e-05, "loss": 3.2741, "step": 3060 }, { "epoch": 18.49397590361446, "grad_norm": 36.31643295288086, "learning_rate": 4.204677769543019e-05, "loss": 3.4015, "step": 3070 }, { "epoch": 18.55421686746988, "grad_norm": 103.95494842529297, "learning_rate": 4.16685516931974e-05, "loss": 3.4717, "step": 3080 }, { "epoch": 18.6144578313253, "grad_norm": 2644.721435546875, "learning_rate": 4.1291181990213286e-05, "loss": 3.1046, "step": 3090 }, { "epoch": 18.674698795180724, "grad_norm": 1700.5968017578125, "learning_rate": 4.0914684098668286e-05, "loss": 3.0716, "step": 3100 }, { "epoch": 18.734939759036145, "grad_norm": 34.94669723510742, "learning_rate": 4.053907349491608e-05, "loss": 3.1178, "step": 3110 }, { "epoch": 18.795180722891565, "grad_norm": 93.29631042480469, "learning_rate": 4.016436561883746e-05, "loss": 3.1197, "step": 3120 }, { "epoch": 18.855421686746986, "grad_norm": 104.23704528808594, "learning_rate": 3.979057587320554e-05, "loss": 3.6272, "step": 3130 }, { "epoch": 18.91566265060241, "grad_norm": 27.856496810913086, "learning_rate": 3.941771962305274e-05, "loss": 3.1739, "step": 3140 }, { "epoch": 18.97590361445783, "grad_norm": 102.71600341796875, "learning_rate": 3.9045812195039125e-05, "loss": 3.2773, "step": 3150 }, { "epoch": 18.97590361445783, "eval_loss": 3.497561454772949, "eval_runtime": 3.8383, "eval_samples_per_second": 9.64, "eval_steps_per_second": 9.64, "step": 3150 }, { "epoch": 19.03614457831325, "grad_norm": 8.53116226196289, "learning_rate": 3.8674868876822395e-05, "loss": 2.9576, "step": 3160 }, { "epoch": 19.096385542168676, "grad_norm": 53.600685119628906, "learning_rate": 3.83049049164295e-05, "loss": 3.1658, "step": 3170 }, { "epoch": 19.156626506024097, "grad_norm": 42.970123291015625, "learning_rate": 3.793593552162978e-05, "loss": 3.075, "step": 3180 }, { "epoch": 19.216867469879517, "grad_norm": 11.737751960754395, "learning_rate": 3.75679758593099e-05, "loss": 3.3461, "step": 3190 }, { "epoch": 19.27710843373494, "grad_norm": 36.907981872558594, "learning_rate": 3.720104105485039e-05, "loss": 3.2005, "step": 3200 }, { "epoch": 19.337349397590362, "grad_norm": 155.0701141357422, "learning_rate": 3.6835146191503885e-05, "loss": 3.1341, "step": 3210 }, { "epoch": 19.397590361445783, "grad_norm": 32.42080307006836, "learning_rate": 3.647030630977508e-05, "loss": 3.41, "step": 3220 }, { "epoch": 19.457831325301203, "grad_norm": 61.9431266784668, "learning_rate": 3.6106536406802524e-05, "loss": 3.0682, "step": 3230 }, { "epoch": 19.518072289156628, "grad_norm": 218.18533325195312, "learning_rate": 3.5743851435742176e-05, "loss": 3.3301, "step": 3240 }, { "epoch": 19.57831325301205, "grad_norm": 34.84703826904297, "learning_rate": 3.538226630515262e-05, "loss": 3.1957, "step": 3250 }, { "epoch": 19.63855421686747, "grad_norm": 75.88587188720703, "learning_rate": 3.502179587838238e-05, "loss": 3.1191, "step": 3260 }, { "epoch": 19.698795180722893, "grad_norm": 133.42489624023438, "learning_rate": 3.46624549729588e-05, "loss": 3.3155, "step": 3270 }, { "epoch": 19.759036144578314, "grad_norm": 13.925497055053711, "learning_rate": 3.430425835997908e-05, "loss": 2.8935, "step": 3280 }, { "epoch": 19.819277108433734, "grad_norm": 31.686290740966797, "learning_rate": 3.394722076350302e-05, "loss": 3.5147, "step": 3290 }, { "epoch": 19.879518072289155, "grad_norm": 26.684673309326172, "learning_rate": 3.359135685994781e-05, "loss": 3.3202, "step": 3300 }, { "epoch": 19.879518072289155, "eval_loss": 3.471327543258667, "eval_runtime": 3.787, "eval_samples_per_second": 9.77, "eval_steps_per_second": 9.77, "step": 3300 }, { "epoch": 19.93975903614458, "grad_norm": 32.881187438964844, "learning_rate": 3.3236681277484654e-05, "loss": 2.8533, "step": 3310 }, { "epoch": 20.0, "grad_norm": 11.029044151306152, "learning_rate": 3.2883208595437584e-05, "loss": 3.1859, "step": 3320 }, { "epoch": 20.06024096385542, "grad_norm": 341.1537780761719, "learning_rate": 3.2530953343684136e-05, "loss": 3.1306, "step": 3330 }, { "epoch": 20.120481927710845, "grad_norm": 20.780797958374023, "learning_rate": 3.217993000205799e-05, "loss": 3.2068, "step": 3340 }, { "epoch": 20.180722891566266, "grad_norm": 5419.01123046875, "learning_rate": 3.1830152999753903e-05, "loss": 3.1233, "step": 3350 }, { "epoch": 20.240963855421686, "grad_norm": 29.549694061279297, "learning_rate": 3.148163671473439e-05, "loss": 3.4924, "step": 3360 }, { "epoch": 20.301204819277107, "grad_norm": 13.60914134979248, "learning_rate": 3.113439547313892e-05, "loss": 3.244, "step": 3370 }, { "epoch": 20.36144578313253, "grad_norm": 34.05803680419922, "learning_rate": 3.0788443548694874e-05, "loss": 3.0187, "step": 3380 }, { "epoch": 20.42168674698795, "grad_norm": 7.142261028289795, "learning_rate": 3.0443795162130876e-05, "loss": 3.1763, "step": 3390 }, { "epoch": 20.481927710843372, "grad_norm": 10.070602416992188, "learning_rate": 3.0100464480592185e-05, "loss": 3.2928, "step": 3400 }, { "epoch": 20.542168674698797, "grad_norm": 19.883346557617188, "learning_rate": 2.9758465617058404e-05, "loss": 2.9957, "step": 3410 }, { "epoch": 20.602409638554217, "grad_norm": 14.350828170776367, "learning_rate": 2.9417812629763285e-05, "loss": 3.2999, "step": 3420 }, { "epoch": 20.662650602409638, "grad_norm": 8.832179069519043, "learning_rate": 2.9078519521616894e-05, "loss": 3.115, "step": 3430 }, { "epoch": 20.72289156626506, "grad_norm": 326.385498046875, "learning_rate": 2.8740600239630002e-05, "loss": 3.3029, "step": 3440 }, { "epoch": 20.783132530120483, "grad_norm": 13.17289924621582, "learning_rate": 2.8404068674340714e-05, "loss": 3.0413, "step": 3450 }, { "epoch": 20.783132530120483, "eval_loss": 3.4084925651550293, "eval_runtime": 3.8274, "eval_samples_per_second": 9.667, "eval_steps_per_second": 9.667, "step": 3450 }, { "epoch": 20.843373493975903, "grad_norm": 9.489799499511719, "learning_rate": 2.80689386592436e-05, "loss": 3.1958, "step": 3460 }, { "epoch": 20.903614457831324, "grad_norm": 11.09453296661377, "learning_rate": 2.7735223970220955e-05, "loss": 3.0116, "step": 3470 }, { "epoch": 20.96385542168675, "grad_norm": 285.0513916015625, "learning_rate": 2.7402938324976576e-05, "loss": 3.1415, "step": 3480 }, { "epoch": 21.02409638554217, "grad_norm": 8.894499778747559, "learning_rate": 2.70720953824719e-05, "loss": 2.8106, "step": 3490 }, { "epoch": 21.08433734939759, "grad_norm": 8.07825756072998, "learning_rate": 2.674270874236441e-05, "loss": 2.8643, "step": 3500 }, { "epoch": 21.14457831325301, "grad_norm": 9.730963706970215, "learning_rate": 2.64147919444488e-05, "loss": 2.9905, "step": 3510 }, { "epoch": 21.204819277108435, "grad_norm": 11.237829208374023, "learning_rate": 2.6088358468100247e-05, "loss": 3.6566, "step": 3520 }, { "epoch": 21.265060240963855, "grad_norm": 10.315103530883789, "learning_rate": 2.5763421731720435e-05, "loss": 2.9594, "step": 3530 }, { "epoch": 21.325301204819276, "grad_norm": 21.455280303955078, "learning_rate": 2.5439995092185892e-05, "loss": 3.1741, "step": 3540 }, { "epoch": 21.3855421686747, "grad_norm": 275.41705322265625, "learning_rate": 2.5118091844299e-05, "loss": 3.0803, "step": 3550 }, { "epoch": 21.44578313253012, "grad_norm": 12.709835052490234, "learning_rate": 2.479772522024147e-05, "loss": 2.9853, "step": 3560 }, { "epoch": 21.50602409638554, "grad_norm": 27.6678466796875, "learning_rate": 2.4478908389030427e-05, "loss": 3.0706, "step": 3570 }, { "epoch": 21.566265060240966, "grad_norm": 13.315126419067383, "learning_rate": 2.41616544559771e-05, "loss": 2.9522, "step": 3580 }, { "epoch": 21.626506024096386, "grad_norm": 9.57907485961914, "learning_rate": 2.3845976462148033e-05, "loss": 3.0357, "step": 3590 }, { "epoch": 21.686746987951807, "grad_norm": 8.764331817626953, "learning_rate": 2.3531887383829157e-05, "loss": 2.9487, "step": 3600 }, { "epoch": 21.686746987951807, "eval_loss": 3.37250018119812, "eval_runtime": 3.782, "eval_samples_per_second": 9.783, "eval_steps_per_second": 9.783, "step": 3600 }, { "epoch": 21.746987951807228, "grad_norm": 11.81778621673584, "learning_rate": 2.3219400131992273e-05, "loss": 3.2796, "step": 3610 }, { "epoch": 21.80722891566265, "grad_norm": 9.797389030456543, "learning_rate": 2.2908527551764404e-05, "loss": 3.2562, "step": 3620 }, { "epoch": 21.867469879518072, "grad_norm": 10.367379188537598, "learning_rate": 2.259928242189966e-05, "loss": 2.996, "step": 3630 }, { "epoch": 21.927710843373493, "grad_norm": 8.591730117797852, "learning_rate": 2.2291677454254136e-05, "loss": 3.1695, "step": 3640 }, { "epoch": 21.987951807228917, "grad_norm": 84.82453155517578, "learning_rate": 2.1985725293263237e-05, "loss": 3.1198, "step": 3650 }, { "epoch": 22.048192771084338, "grad_norm": 10.37941837310791, "learning_rate": 2.1681438515421953e-05, "loss": 3.0494, "step": 3660 }, { "epoch": 22.10843373493976, "grad_norm": 167.1722869873047, "learning_rate": 2.1378829628767965e-05, "loss": 3.2033, "step": 3670 }, { "epoch": 22.16867469879518, "grad_norm": 245.55747985839844, "learning_rate": 2.1077911072367317e-05, "loss": 3.3328, "step": 3680 }, { "epoch": 22.228915662650603, "grad_norm": 12.856267929077148, "learning_rate": 2.077869521580325e-05, "loss": 2.7443, "step": 3690 }, { "epoch": 22.289156626506024, "grad_norm": 8.544069290161133, "learning_rate": 2.0481194358667695e-05, "loss": 2.9283, "step": 3700 }, { "epoch": 22.349397590361445, "grad_norm": 11.428658485412598, "learning_rate": 2.018542073005567e-05, "loss": 3.1091, "step": 3710 }, { "epoch": 22.40963855421687, "grad_norm": 18.596616744995117, "learning_rate": 1.9891386488062538e-05, "loss": 2.944, "step": 3720 }, { "epoch": 22.46987951807229, "grad_norm": 17.378009796142578, "learning_rate": 1.959910371928436e-05, "loss": 3.2485, "step": 3730 }, { "epoch": 22.53012048192771, "grad_norm": 7.827014923095703, "learning_rate": 1.930858443832096e-05, "loss": 3.0034, "step": 3740 }, { "epoch": 22.59036144578313, "grad_norm": 31.390344619750977, "learning_rate": 1.90198405872821e-05, "loss": 3.1377, "step": 3750 }, { "epoch": 22.59036144578313, "eval_loss": 3.3621747493743896, "eval_runtime": 3.7897, "eval_samples_per_second": 9.763, "eval_steps_per_second": 9.763, "step": 3750 }, { "epoch": 22.650602409638555, "grad_norm": 19.31273651123047, "learning_rate": 1.8732884035296582e-05, "loss": 3.1508, "step": 3760 }, { "epoch": 22.710843373493976, "grad_norm": 9.601163864135742, "learning_rate": 1.844772657802428e-05, "loss": 2.8004, "step": 3770 }, { "epoch": 22.771084337349397, "grad_norm": 27.831907272338867, "learning_rate": 1.8164379937171382e-05, "loss": 2.8072, "step": 3780 }, { "epoch": 22.83132530120482, "grad_norm": 15.658758163452148, "learning_rate": 1.7882855760008547e-05, "loss": 3.1902, "step": 3790 }, { "epoch": 22.89156626506024, "grad_norm": 22.477460861206055, "learning_rate": 1.760316561889203e-05, "loss": 3.3437, "step": 3800 }, { "epoch": 22.951807228915662, "grad_norm": 29.21653175354004, "learning_rate": 1.7325321010788034e-05, "loss": 3.0924, "step": 3810 }, { "epoch": 23.012048192771083, "grad_norm": 19.53176498413086, "learning_rate": 1.7049333356800167e-05, "loss": 3.0771, "step": 3820 }, { "epoch": 23.072289156626507, "grad_norm": 19.212852478027344, "learning_rate": 1.6775214001699914e-05, "loss": 2.7683, "step": 3830 }, { "epoch": 23.132530120481928, "grad_norm": 34.5575065612793, "learning_rate": 1.6502974213460316e-05, "loss": 3.0162, "step": 3840 }, { "epoch": 23.19277108433735, "grad_norm": 13.085418701171875, "learning_rate": 1.623262518279279e-05, "loss": 3.1395, "step": 3850 }, { "epoch": 23.253012048192772, "grad_norm": 10.661376953125, "learning_rate": 1.596417802268707e-05, "loss": 3.0344, "step": 3860 }, { "epoch": 23.313253012048193, "grad_norm": 15.066116333007812, "learning_rate": 1.5697643767954488e-05, "loss": 3.0326, "step": 3870 }, { "epoch": 23.373493975903614, "grad_norm": 21.323190689086914, "learning_rate": 1.543303337477432e-05, "loss": 3.0845, "step": 3880 }, { "epoch": 23.433734939759034, "grad_norm": 17.696609497070312, "learning_rate": 1.517035772024343e-05, "loss": 2.9169, "step": 3890 }, { "epoch": 23.49397590361446, "grad_norm": 121.86737060546875, "learning_rate": 1.49096276019291e-05, "loss": 3.411, "step": 3900 }, { "epoch": 23.49397590361446, "eval_loss": 3.3336081504821777, "eval_runtime": 3.8199, "eval_samples_per_second": 9.686, "eval_steps_per_second": 9.686, "step": 3900 }, { "epoch": 23.55421686746988, "grad_norm": 25.903724670410156, "learning_rate": 1.4650853737425327e-05, "loss": 2.6079, "step": 3910 }, { "epoch": 23.6144578313253, "grad_norm": 303.8630065917969, "learning_rate": 1.4394046763912122e-05, "loss": 3.0013, "step": 3920 }, { "epoch": 23.674698795180724, "grad_norm": 96.3546371459961, "learning_rate": 1.413921723771832e-05, "loss": 3.2609, "step": 3930 }, { "epoch": 23.734939759036145, "grad_norm": 51.00669479370117, "learning_rate": 1.3886375633887665e-05, "loss": 3.2636, "step": 3940 }, { "epoch": 23.795180722891565, "grad_norm": 21.779207229614258, "learning_rate": 1.3635532345748137e-05, "loss": 3.1889, "step": 3950 }, { "epoch": 23.855421686746986, "grad_norm": 483.495361328125, "learning_rate": 1.3386697684484853e-05, "loss": 2.8758, "step": 3960 }, { "epoch": 23.91566265060241, "grad_norm": 38.69334411621094, "learning_rate": 1.3139881878716107e-05, "loss": 2.7819, "step": 3970 }, { "epoch": 23.97590361445783, "grad_norm": 40.966644287109375, "learning_rate": 1.2895095074072986e-05, "loss": 2.9545, "step": 3980 }, { "epoch": 24.03614457831325, "grad_norm": 15.748128890991211, "learning_rate": 1.2652347332782227e-05, "loss": 2.9577, "step": 3990 }, { "epoch": 24.096385542168676, "grad_norm": 89.11782836914062, "learning_rate": 1.2411648633252719e-05, "loss": 3.2177, "step": 4000 }, { "epoch": 24.156626506024097, "grad_norm": 21.81198501586914, "learning_rate": 1.2173008869665241e-05, "loss": 2.9163, "step": 4010 }, { "epoch": 24.216867469879517, "grad_norm": 11.634493827819824, "learning_rate": 1.1936437851565791e-05, "loss": 2.9567, "step": 4020 }, { "epoch": 24.27710843373494, "grad_norm": 13.151474952697754, "learning_rate": 1.1701945303462337e-05, "loss": 2.9042, "step": 4030 }, { "epoch": 24.337349397590362, "grad_norm": 9.143220901489258, "learning_rate": 1.146954086442508e-05, "loss": 3.1724, "step": 4040 }, { "epoch": 24.397590361445783, "grad_norm": 10.673160552978516, "learning_rate": 1.1239234087690252e-05, "loss": 3.0989, "step": 4050 }, { "epoch": 24.397590361445783, "eval_loss": 3.318880081176758, "eval_runtime": 3.7938, "eval_samples_per_second": 9.753, "eval_steps_per_second": 9.753, "step": 4050 }, { "epoch": 24.457831325301203, "grad_norm": 22.973133087158203, "learning_rate": 1.1011034440267395e-05, "loss": 3.1335, "step": 4060 }, { "epoch": 24.518072289156628, "grad_norm": 57.02553939819336, "learning_rate": 1.078495130255023e-05, "loss": 2.6128, "step": 4070 }, { "epoch": 24.57831325301205, "grad_norm": 109.61993408203125, "learning_rate": 1.0560993967931004e-05, "loss": 3.1313, "step": 4080 }, { "epoch": 24.63855421686747, "grad_norm": 10.217292785644531, "learning_rate": 1.0339171642418585e-05, "loss": 3.0932, "step": 4090 }, { "epoch": 24.698795180722893, "grad_norm": 39.34014892578125, "learning_rate": 1.0119493444259963e-05, "loss": 3.0317, "step": 4100 }, { "epoch": 24.759036144578314, "grad_norm": 27.805980682373047, "learning_rate": 9.901968403565428e-06, "loss": 2.6487, "step": 4110 }, { "epoch": 24.819277108433734, "grad_norm": 13.394316673278809, "learning_rate": 9.686605461937441e-06, "loss": 2.9028, "step": 4120 }, { "epoch": 24.879518072289155, "grad_norm": 9.390585899353027, "learning_rate": 9.473413472102982e-06, "loss": 3.1307, "step": 4130 }, { "epoch": 24.93975903614458, "grad_norm": 50.32940673828125, "learning_rate": 9.262401197549744e-06, "loss": 2.7928, "step": 4140 }, { "epoch": 25.0, "grad_norm": 24.728273391723633, "learning_rate": 9.05357731216587e-06, "loss": 3.0874, "step": 4150 }, { "epoch": 25.06024096385542, "grad_norm": 17.07667350769043, "learning_rate": 8.846950399883368e-06, "loss": 3.1649, "step": 4160 }, { "epoch": 25.120481927710845, "grad_norm": 22.225461959838867, "learning_rate": 8.64252895432531e-06, "loss": 2.9153, "step": 4170 }, { "epoch": 25.180722891566266, "grad_norm": 9.772006034851074, "learning_rate": 8.440321378456656e-06, "loss": 2.7644, "step": 4180 }, { "epoch": 25.240963855421686, "grad_norm": 12.998729705810547, "learning_rate": 8.240335984238844e-06, "loss": 2.8009, "step": 4190 }, { "epoch": 25.301204819277107, "grad_norm": 11.086814880371094, "learning_rate": 8.042580992288163e-06, "loss": 2.9233, "step": 4200 }, { "epoch": 25.301204819277107, "eval_loss": 3.316469192504883, "eval_runtime": 3.8267, "eval_samples_per_second": 9.669, "eval_steps_per_second": 9.669, "step": 4200 }, { "epoch": 25.36144578313253, "grad_norm": 22.682485580444336, "learning_rate": 7.847064531537774e-06, "loss": 2.6035, "step": 4210 }, { "epoch": 25.42168674698795, "grad_norm": 42.95654296875, "learning_rate": 7.653794638903574e-06, "loss": 3.1161, "step": 4220 }, { "epoch": 25.481927710843372, "grad_norm": 55.98487091064453, "learning_rate": 7.462779258953875e-06, "loss": 2.7615, "step": 4230 }, { "epoch": 25.542168674698797, "grad_norm": 10.97606372833252, "learning_rate": 7.274026243582796e-06, "loss": 3.0223, "step": 4240 }, { "epoch": 25.602409638554217, "grad_norm": 37.118247985839844, "learning_rate": 7.087543351687493e-06, "loss": 2.9984, "step": 4250 }, { "epoch": 25.662650602409638, "grad_norm": 20.084211349487305, "learning_rate": 6.903338248849269e-06, "loss": 3.1486, "step": 4260 }, { "epoch": 25.72289156626506, "grad_norm": 14.11424446105957, "learning_rate": 6.7214185070183925e-06, "loss": 2.8441, "step": 4270 }, { "epoch": 25.783132530120483, "grad_norm": 9.799378395080566, "learning_rate": 6.541791604202936e-06, "loss": 3.033, "step": 4280 }, { "epoch": 25.843373493975903, "grad_norm": 34.30189514160156, "learning_rate": 6.364464924161311e-06, "loss": 3.1671, "step": 4290 }, { "epoch": 25.903614457831324, "grad_norm": 213.951904296875, "learning_rate": 6.1894457560988106e-06, "loss": 2.9741, "step": 4300 }, { "epoch": 25.96385542168675, "grad_norm": 186.6619415283203, "learning_rate": 6.016741294367911e-06, "loss": 3.1161, "step": 4310 }, { "epoch": 26.02409638554217, "grad_norm": 12.259818077087402, "learning_rate": 5.846358638172615e-06, "loss": 2.8821, "step": 4320 }, { "epoch": 26.08433734939759, "grad_norm": 11.736944198608398, "learning_rate": 5.678304791276567e-06, "loss": 2.7778, "step": 4330 }, { "epoch": 26.14457831325301, "grad_norm": 59.023399353027344, "learning_rate": 5.51258666171519e-06, "loss": 3.1458, "step": 4340 }, { "epoch": 26.204819277108435, "grad_norm": 11.082910537719727, "learning_rate": 5.349211061511726e-06, "loss": 2.8066, "step": 4350 }, { "epoch": 26.204819277108435, "eval_loss": 3.3215532302856445, "eval_runtime": 3.834, "eval_samples_per_second": 9.65, "eval_steps_per_second": 9.65, "step": 4350 }, { "epoch": 26.265060240963855, "grad_norm": 9.732200622558594, "learning_rate": 5.188184706397182e-06, "loss": 3.1099, "step": 4360 }, { "epoch": 26.325301204819276, "grad_norm": 52.13785171508789, "learning_rate": 5.029514215534339e-06, "loss": 2.9703, "step": 4370 }, { "epoch": 26.3855421686747, "grad_norm": 19.796157836914062, "learning_rate": 4.873206111245594e-06, "loss": 3.1718, "step": 4380 }, { "epoch": 26.44578313253012, "grad_norm": 224.9999237060547, "learning_rate": 4.719266818744912e-06, "loss": 3.0249, "step": 4390 }, { "epoch": 26.50602409638554, "grad_norm": 9.841102600097656, "learning_rate": 4.567702665873648e-06, "loss": 3.0305, "step": 4400 }, { "epoch": 26.566265060240966, "grad_norm": 14.756407737731934, "learning_rate": 4.418519882840505e-06, "loss": 2.7736, "step": 4410 }, { "epoch": 26.626506024096386, "grad_norm": 25.35453987121582, "learning_rate": 4.271724601965371e-06, "loss": 2.5669, "step": 4420 }, { "epoch": 26.686746987951807, "grad_norm": 272.5387268066406, "learning_rate": 4.127322857427306e-06, "loss": 3.1571, "step": 4430 }, { "epoch": 26.746987951807228, "grad_norm": 46.450382232666016, "learning_rate": 3.985320585016425e-06, "loss": 2.7388, "step": 4440 }, { "epoch": 26.80722891566265, "grad_norm": 20.076871871948242, "learning_rate": 3.845723621889973e-06, "loss": 3.16, "step": 4450 }, { "epoch": 26.867469879518072, "grad_norm": 13.591471672058105, "learning_rate": 3.7085377063323447e-06, "loss": 2.9016, "step": 4460 }, { "epoch": 26.927710843373493, "grad_norm": 18.94786834716797, "learning_rate": 3.5737684775191887e-06, "loss": 2.7776, "step": 4470 }, { "epoch": 26.987951807228917, "grad_norm": 22.884519577026367, "learning_rate": 3.441421475285679e-06, "loss": 3.041, "step": 4480 }, { "epoch": 27.048192771084338, "grad_norm": 11.475288391113281, "learning_rate": 3.3115021398986768e-06, "loss": 3.3, "step": 4490 }, { "epoch": 27.10843373493976, "grad_norm": 95.50624084472656, "learning_rate": 3.18401581183321e-06, "loss": 2.9552, "step": 4500 }, { "epoch": 27.10843373493976, "eval_loss": 3.3192508220672607, "eval_runtime": 3.8297, "eval_samples_per_second": 9.661, "eval_steps_per_second": 9.661, "step": 4500 }, { "epoch": 27.16867469879518, "grad_norm": 312.4958190917969, "learning_rate": 3.0589677315529044e-06, "loss": 2.4796, "step": 4510 }, { "epoch": 27.228915662650603, "grad_norm": 35.864219665527344, "learning_rate": 2.9363630392945513e-06, "loss": 3.2925, "step": 4520 }, { "epoch": 27.289156626506024, "grad_norm": 28.503623962402344, "learning_rate": 2.816206774856854e-06, "loss": 2.9739, "step": 4530 }, { "epoch": 27.349397590361445, "grad_norm": 104.54164123535156, "learning_rate": 2.6985038773932046e-06, "loss": 3.0139, "step": 4540 }, { "epoch": 27.40963855421687, "grad_norm": 28.280534744262695, "learning_rate": 2.583259185208714e-06, "loss": 2.952, "step": 4550 }, { "epoch": 27.46987951807229, "grad_norm": 18.73903465270996, "learning_rate": 2.4704774355612943e-06, "loss": 3.0347, "step": 4560 }, { "epoch": 27.53012048192771, "grad_norm": 32.26377868652344, "learning_rate": 2.3601632644669536e-06, "loss": 2.8445, "step": 4570 }, { "epoch": 27.59036144578313, "grad_norm": 8.18028736114502, "learning_rate": 2.2523212065091723e-06, "loss": 2.8002, "step": 4580 }, { "epoch": 27.650602409638555, "grad_norm": 31.256370544433594, "learning_rate": 2.1469556946525706e-06, "loss": 3.0538, "step": 4590 }, { "epoch": 27.710843373493976, "grad_norm": 13.404227256774902, "learning_rate": 2.0440710600606595e-06, "loss": 2.8497, "step": 4600 }, { "epoch": 27.771084337349397, "grad_norm": 48.90232849121094, "learning_rate": 1.9436715319177956e-06, "loss": 2.9533, "step": 4610 }, { "epoch": 27.83132530120482, "grad_norm": 101.96350860595703, "learning_rate": 1.8457612372553348e-06, "loss": 2.9476, "step": 4620 }, { "epoch": 27.89156626506024, "grad_norm": 73.58414459228516, "learning_rate": 1.75034420078201e-06, "loss": 3.0937, "step": 4630 }, { "epoch": 27.951807228915662, "grad_norm": 28.492570877075195, "learning_rate": 1.6574243447184597e-06, "loss": 2.784, "step": 4640 }, { "epoch": 28.012048192771083, "grad_norm": 10.421324729919434, "learning_rate": 1.567005488636024e-06, "loss": 2.9865, "step": 4650 }, { "epoch": 28.012048192771083, "eval_loss": 3.310655355453491, "eval_runtime": 3.8346, "eval_samples_per_second": 9.649, "eval_steps_per_second": 9.649, "step": 4650 }, { "epoch": 28.072289156626507, "grad_norm": 9.195672035217285, "learning_rate": 1.4790913492997438e-06, "loss": 2.6247, "step": 4660 }, { "epoch": 28.132530120481928, "grad_norm": 14.86363697052002, "learning_rate": 1.3936855405155408e-06, "loss": 2.8831, "step": 4670 }, { "epoch": 28.19277108433735, "grad_norm": 11.328372955322266, "learning_rate": 1.3107915729816954e-06, "loss": 2.7178, "step": 4680 }, { "epoch": 28.253012048192772, "grad_norm": 15.23580551147461, "learning_rate": 1.230412854144547e-06, "loss": 3.1991, "step": 4690 }, { "epoch": 28.313253012048193, "grad_norm": 169.89739990234375, "learning_rate": 1.15255268805841e-06, "loss": 3.0329, "step": 4700 }, { "epoch": 28.373493975903614, "grad_norm": 323.2249450683594, "learning_rate": 1.0772142752497604e-06, "loss": 3.1795, "step": 4710 }, { "epoch": 28.433734939759034, "grad_norm": 24.42875099182129, "learning_rate": 1.004400712585646e-06, "loss": 2.8986, "step": 4720 }, { "epoch": 28.49397590361446, "grad_norm": 14.24337387084961, "learning_rate": 9.341149931464537e-07, "loss": 2.7472, "step": 4730 }, { "epoch": 28.55421686746988, "grad_norm": 90.30467224121094, "learning_rate": 8.663600061028162e-07, "loss": 2.9509, "step": 4740 }, { "epoch": 28.6144578313253, "grad_norm": 32.362003326416016, "learning_rate": 8.011385365968641e-07, "loss": 3.0889, "step": 4750 }, { "epoch": 28.674698795180724, "grad_norm": 20.82928466796875, "learning_rate": 7.384532656277698e-07, "loss": 3.0301, "step": 4760 }, { "epoch": 28.734939759036145, "grad_norm": 12.928521156311035, "learning_rate": 6.783067699414891e-07, "loss": 2.8634, "step": 4770 }, { "epoch": 28.795180722891565, "grad_norm": 14.081971168518066, "learning_rate": 6.207015219248866e-07, "loss": 2.6318, "step": 4780 }, { "epoch": 28.855421686746986, "grad_norm": 13.051186561584473, "learning_rate": 5.656398895040813e-07, "loss": 2.7618, "step": 4790 }, { "epoch": 28.91566265060241, "grad_norm": 221.75543212890625, "learning_rate": 5.131241360471217e-07, "loss": 3.0757, "step": 4800 }, { "epoch": 28.91566265060241, "eval_loss": 3.3106162548065186, "eval_runtime": 3.8397, "eval_samples_per_second": 9.636, "eval_steps_per_second": 9.636, "step": 4800 } ], "logging_steps": 10, "max_steps": 5000, "num_input_tokens_seen": 0, "num_train_epochs": 31, "save_steps": 150, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1.5578958526488576e+16, "train_batch_size": 2, "trial_name": null, "trial_params": null }