| [ | |
| { | |
| "loss": 13.2026, | |
| "grad_norm": 6.155358791351318, | |
| "learning_rate": 1.739130434782609e-05, | |
| "epoch": 0.034782608695652174, | |
| "step": 20 | |
| }, | |
| { | |
| "loss": 13.1252, | |
| "grad_norm": 5.816741943359375, | |
| "learning_rate": 3.478260869565218e-05, | |
| "epoch": 0.06956521739130435, | |
| "step": 40 | |
| }, | |
| { | |
| "loss": 13.0001, | |
| "grad_norm": 5.273156642913818, | |
| "learning_rate": 5.2173913043478256e-05, | |
| "epoch": 0.10434782608695652, | |
| "step": 60 | |
| }, | |
| { | |
| "loss": 12.8639, | |
| "grad_norm": 4.86655330657959, | |
| "learning_rate": 6.956521739130436e-05, | |
| "epoch": 0.1391304347826087, | |
| "step": 80 | |
| }, | |
| { | |
| "loss": 12.7376, | |
| "grad_norm": 4.438321113586426, | |
| "learning_rate": 8.695652173913044e-05, | |
| "epoch": 0.17391304347826086, | |
| "step": 100 | |
| }, | |
| { | |
| "loss": 12.5722, | |
| "grad_norm": 4.164404392242432, | |
| "learning_rate": 0.00010434782608695651, | |
| "epoch": 0.20869565217391303, | |
| "step": 120 | |
| }, | |
| { | |
| "loss": 12.4229, | |
| "grad_norm": 3.858990430831909, | |
| "learning_rate": 0.00012173913043478261, | |
| "epoch": 0.24347826086956523, | |
| "step": 140 | |
| }, | |
| { | |
| "loss": 12.2581, | |
| "grad_norm": 3.6574394702911377, | |
| "learning_rate": 0.0001391304347826087, | |
| "epoch": 0.2782608695652174, | |
| "step": 160 | |
| }, | |
| { | |
| "loss": 12.0753, | |
| "grad_norm": 3.3787951469421387, | |
| "learning_rate": 0.0001565217391304348, | |
| "epoch": 0.3130434782608696, | |
| "step": 180 | |
| }, | |
| { | |
| "loss": 11.9261, | |
| "grad_norm": 3.323820114135742, | |
| "learning_rate": 0.00017391304347826088, | |
| "epoch": 0.34782608695652173, | |
| "step": 200 | |
| }, | |
| { | |
| "loss": 11.7417, | |
| "grad_norm": 3.247619152069092, | |
| "learning_rate": 0.00019130434782608697, | |
| "epoch": 0.3826086956521739, | |
| "step": 220 | |
| }, | |
| { | |
| "loss": 11.5771, | |
| "grad_norm": 3.2254152297973633, | |
| "learning_rate": 0.00020869565217391303, | |
| "epoch": 0.41739130434782606, | |
| "step": 240 | |
| }, | |
| { | |
| "loss": 11.3969, | |
| "grad_norm": 3.1803464889526367, | |
| "learning_rate": 0.00022608695652173914, | |
| "epoch": 0.45217391304347826, | |
| "step": 260 | |
| }, | |
| { | |
| "loss": 11.2684, | |
| "grad_norm": 3.41034197807312, | |
| "learning_rate": 0.00024347826086956522, | |
| "epoch": 0.48695652173913045, | |
| "step": 280 | |
| }, | |
| { | |
| "loss": 11.0744, | |
| "grad_norm": 3.246403217315674, | |
| "learning_rate": 0.0002608695652173913, | |
| "epoch": 0.5217391304347826, | |
| "step": 300 | |
| }, | |
| { | |
| "loss": 10.8929, | |
| "grad_norm": 3.202021360397339, | |
| "learning_rate": 0.0002782608695652174, | |
| "epoch": 0.5565217391304348, | |
| "step": 320 | |
| }, | |
| { | |
| "loss": 10.7468, | |
| "grad_norm": 3.1231367588043213, | |
| "learning_rate": 0.0002956521739130435, | |
| "epoch": 0.591304347826087, | |
| "step": 340 | |
| }, | |
| { | |
| "loss": 10.606, | |
| "grad_norm": 3.1820390224456787, | |
| "learning_rate": 0.0003130434782608696, | |
| "epoch": 0.6260869565217392, | |
| "step": 360 | |
| }, | |
| { | |
| "loss": 10.4871, | |
| "grad_norm": 3.2470555305480957, | |
| "learning_rate": 0.0003304347826086956, | |
| "epoch": 0.6608695652173913, | |
| "step": 380 | |
| }, | |
| { | |
| "loss": 10.2836, | |
| "grad_norm": 3.2452709674835205, | |
| "learning_rate": 0.00034782608695652176, | |
| "epoch": 0.6956521739130435, | |
| "step": 400 | |
| }, | |
| { | |
| "loss": 10.1154, | |
| "grad_norm": 3.203894853591919, | |
| "learning_rate": 0.00036521739130434785, | |
| "epoch": 0.7304347826086957, | |
| "step": 420 | |
| }, | |
| { | |
| "loss": 9.9283, | |
| "grad_norm": 3.269970178604126, | |
| "learning_rate": 0.00038260869565217393, | |
| "epoch": 0.7652173913043478, | |
| "step": 440 | |
| }, | |
| { | |
| "loss": 9.8674, | |
| "grad_norm": 3.261357545852661, | |
| "learning_rate": 0.0004, | |
| "epoch": 0.8, | |
| "step": 460 | |
| }, | |
| { | |
| "loss": 9.6224, | |
| "grad_norm": 3.393953323364258, | |
| "learning_rate": 0.00041739130434782605, | |
| "epoch": 0.8347826086956521, | |
| "step": 480 | |
| }, | |
| { | |
| "loss": 9.524, | |
| "grad_norm": 3.321411609649658, | |
| "learning_rate": 0.0004347826086956522, | |
| "epoch": 0.8695652173913043, | |
| "step": 500 | |
| }, | |
| { | |
| "loss": 9.384, | |
| "grad_norm": 3.3886823654174805, | |
| "learning_rate": 0.0004521739130434783, | |
| "epoch": 0.9043478260869565, | |
| "step": 520 | |
| }, | |
| { | |
| "loss": 9.1767, | |
| "grad_norm": 3.4735491275787354, | |
| "learning_rate": 0.00046956521739130436, | |
| "epoch": 0.9391304347826087, | |
| "step": 540 | |
| }, | |
| { | |
| "loss": 9.047, | |
| "grad_norm": 3.416966676712036, | |
| "learning_rate": 0.00048695652173913045, | |
| "epoch": 0.9739130434782609, | |
| "step": 560 | |
| }, | |
| { | |
| "eval_loss": 8.366157531738281, | |
| "eval_accuracy": 0.43039677202420984, | |
| "eval_runtime": 42.3364, | |
| "eval_samples_per_second": 35.123, | |
| "eval_steps_per_second": 35.123, | |
| "epoch": 1.0, | |
| "step": 575 | |
| }, | |
| { | |
| "loss": 8.8835, | |
| "grad_norm": 3.446899890899658, | |
| "learning_rate": 0.0004995169082125604, | |
| "epoch": 1.008695652173913, | |
| "step": 580 | |
| }, | |
| { | |
| "loss": 8.6436, | |
| "grad_norm": 3.5842247009277344, | |
| "learning_rate": 0.0004975845410628019, | |
| "epoch": 1.0434782608695652, | |
| "step": 600 | |
| }, | |
| { | |
| "loss": 8.4775, | |
| "grad_norm": 3.5029306411743164, | |
| "learning_rate": 0.0004956521739130435, | |
| "epoch": 1.0782608695652174, | |
| "step": 620 | |
| }, | |
| { | |
| "loss": 8.322, | |
| "grad_norm": 3.5451033115386963, | |
| "learning_rate": 0.0004937198067632851, | |
| "epoch": 1.1130434782608696, | |
| "step": 640 | |
| }, | |
| { | |
| "loss": 8.1264, | |
| "grad_norm": 3.5502634048461914, | |
| "learning_rate": 0.0004917874396135266, | |
| "epoch": 1.1478260869565218, | |
| "step": 660 | |
| }, | |
| { | |
| "loss": 7.9905, | |
| "grad_norm": 3.607395648956299, | |
| "learning_rate": 0.0004898550724637681, | |
| "epoch": 1.182608695652174, | |
| "step": 680 | |
| }, | |
| { | |
| "loss": 7.8252, | |
| "grad_norm": 3.6438565254211426, | |
| "learning_rate": 0.0004879227053140097, | |
| "epoch": 1.2173913043478262, | |
| "step": 700 | |
| }, | |
| { | |
| "loss": 7.7737, | |
| "grad_norm": 3.656705141067505, | |
| "learning_rate": 0.0004859903381642512, | |
| "epoch": 1.2521739130434781, | |
| "step": 720 | |
| }, | |
| { | |
| "loss": 7.5822, | |
| "grad_norm": 3.7424328327178955, | |
| "learning_rate": 0.0004840579710144928, | |
| "epoch": 1.2869565217391306, | |
| "step": 740 | |
| }, | |
| { | |
| "loss": 7.4563, | |
| "grad_norm": 3.673156261444092, | |
| "learning_rate": 0.0004821256038647343, | |
| "epoch": 1.3217391304347825, | |
| "step": 760 | |
| }, | |
| { | |
| "loss": 7.3379, | |
| "grad_norm": 3.6774067878723145, | |
| "learning_rate": 0.0004801932367149758, | |
| "epoch": 1.3565217391304347, | |
| "step": 780 | |
| }, | |
| { | |
| "loss": 7.1559, | |
| "grad_norm": 3.811283826828003, | |
| "learning_rate": 0.0004782608695652174, | |
| "epoch": 1.391304347826087, | |
| "step": 800 | |
| }, | |
| { | |
| "loss": 7.0834, | |
| "grad_norm": 3.7899839878082275, | |
| "learning_rate": 0.00047632850241545894, | |
| "epoch": 1.4260869565217391, | |
| "step": 820 | |
| }, | |
| { | |
| "loss": 6.9172, | |
| "grad_norm": 3.583247423171997, | |
| "learning_rate": 0.00047439613526570047, | |
| "epoch": 1.4608695652173913, | |
| "step": 840 | |
| }, | |
| { | |
| "loss": 6.7251, | |
| "grad_norm": 3.8192331790924072, | |
| "learning_rate": 0.00047246376811594206, | |
| "epoch": 1.4956521739130435, | |
| "step": 860 | |
| }, | |
| { | |
| "loss": 6.7871, | |
| "grad_norm": 3.8098299503326416, | |
| "learning_rate": 0.0004705314009661836, | |
| "epoch": 1.5304347826086957, | |
| "step": 880 | |
| }, | |
| { | |
| "loss": 6.6103, | |
| "grad_norm": 3.7341325283050537, | |
| "learning_rate": 0.0004685990338164252, | |
| "epoch": 1.5652173913043477, | |
| "step": 900 | |
| }, | |
| { | |
| "loss": 6.4507, | |
| "grad_norm": 3.9190495014190674, | |
| "learning_rate": 0.00046666666666666666, | |
| "epoch": 1.6, | |
| "step": 920 | |
| }, | |
| { | |
| "loss": 6.3619, | |
| "grad_norm": 3.9456422328948975, | |
| "learning_rate": 0.0004647342995169082, | |
| "epoch": 1.634782608695652, | |
| "step": 940 | |
| }, | |
| { | |
| "loss": 6.2957, | |
| "grad_norm": 3.899134874343872, | |
| "learning_rate": 0.0004628019323671498, | |
| "epoch": 1.6695652173913045, | |
| "step": 960 | |
| }, | |
| { | |
| "loss": 6.1362, | |
| "grad_norm": 3.878810167312622, | |
| "learning_rate": 0.0004608695652173913, | |
| "epoch": 1.7043478260869565, | |
| "step": 980 | |
| }, | |
| { | |
| "loss": 5.9814, | |
| "grad_norm": 3.9270784854888916, | |
| "learning_rate": 0.00045893719806763285, | |
| "epoch": 1.7391304347826086, | |
| "step": 1000 | |
| }, | |
| { | |
| "loss": 5.9095, | |
| "grad_norm": 3.8247644901275635, | |
| "learning_rate": 0.00045700483091787444, | |
| "epoch": 1.7739130434782608, | |
| "step": 1020 | |
| }, | |
| { | |
| "loss": 5.7793, | |
| "grad_norm": 3.8870134353637695, | |
| "learning_rate": 0.000455072463768116, | |
| "epoch": 1.808695652173913, | |
| "step": 1040 | |
| }, | |
| { | |
| "loss": 5.7754, | |
| "grad_norm": 3.9533441066741943, | |
| "learning_rate": 0.00045314009661835745, | |
| "epoch": 1.8434782608695652, | |
| "step": 1060 | |
| }, | |
| { | |
| "loss": 5.5886, | |
| "grad_norm": 3.9928998947143555, | |
| "learning_rate": 0.00045120772946859904, | |
| "epoch": 1.8782608695652174, | |
| "step": 1080 | |
| }, | |
| { | |
| "loss": 5.5482, | |
| "grad_norm": 4.030064582824707, | |
| "learning_rate": 0.0004492753623188406, | |
| "epoch": 1.9130434782608696, | |
| "step": 1100 | |
| }, | |
| { | |
| "loss": 5.4807, | |
| "grad_norm": 3.961806297302246, | |
| "learning_rate": 0.0004473429951690821, | |
| "epoch": 1.9478260869565216, | |
| "step": 1120 | |
| }, | |
| { | |
| "loss": 5.3508, | |
| "grad_norm": 4.003119945526123, | |
| "learning_rate": 0.0004454106280193237, | |
| "epoch": 1.982608695652174, | |
| "step": 1140 | |
| }, | |
| { | |
| "eval_loss": 4.025164604187012, | |
| "eval_accuracy": 0.8190988567585743, | |
| "eval_runtime": 42.7144, | |
| "eval_samples_per_second": 34.813, | |
| "eval_steps_per_second": 34.813, | |
| "epoch": 2.0, | |
| "step": 1150 | |
| }, | |
| { | |
| "loss": 5.1229, | |
| "grad_norm": 3.958116292953491, | |
| "learning_rate": 0.00044347826086956523, | |
| "epoch": 2.017391304347826, | |
| "step": 1160 | |
| }, | |
| { | |
| "loss": 4.8146, | |
| "grad_norm": 3.864279270172119, | |
| "learning_rate": 0.00044154589371980677, | |
| "epoch": 2.0521739130434784, | |
| "step": 1180 | |
| }, | |
| { | |
| "loss": 4.8843, | |
| "grad_norm": 4.045077323913574, | |
| "learning_rate": 0.0004396135265700483, | |
| "epoch": 2.0869565217391304, | |
| "step": 1200 | |
| }, | |
| { | |
| "loss": 4.8078, | |
| "grad_norm": 4.061978816986084, | |
| "learning_rate": 0.00043768115942028983, | |
| "epoch": 2.121739130434783, | |
| "step": 1220 | |
| }, | |
| { | |
| "loss": 4.6812, | |
| "grad_norm": 4.040159225463867, | |
| "learning_rate": 0.0004357487922705314, | |
| "epoch": 2.1565217391304348, | |
| "step": 1240 | |
| }, | |
| { | |
| "loss": 4.6701, | |
| "grad_norm": 4.234623908996582, | |
| "learning_rate": 0.00043381642512077296, | |
| "epoch": 2.1913043478260867, | |
| "step": 1260 | |
| }, | |
| { | |
| "loss": 4.6221, | |
| "grad_norm": 4.030038356781006, | |
| "learning_rate": 0.0004318840579710145, | |
| "epoch": 2.226086956521739, | |
| "step": 1280 | |
| }, | |
| { | |
| "loss": 4.5647, | |
| "grad_norm": 3.9954497814178467, | |
| "learning_rate": 0.0004299516908212561, | |
| "epoch": 2.260869565217391, | |
| "step": 1300 | |
| }, | |
| { | |
| "loss": 4.4502, | |
| "grad_norm": 4.188636779785156, | |
| "learning_rate": 0.0004280193236714976, | |
| "epoch": 2.2956521739130435, | |
| "step": 1320 | |
| }, | |
| { | |
| "loss": 4.359, | |
| "grad_norm": 4.185456275939941, | |
| "learning_rate": 0.00042608695652173915, | |
| "epoch": 2.3304347826086955, | |
| "step": 1340 | |
| }, | |
| { | |
| "loss": 4.2863, | |
| "grad_norm": 4.123263359069824, | |
| "learning_rate": 0.0004241545893719807, | |
| "epoch": 2.365217391304348, | |
| "step": 1360 | |
| }, | |
| { | |
| "loss": 4.3354, | |
| "grad_norm": 4.194387435913086, | |
| "learning_rate": 0.0004222222222222222, | |
| "epoch": 2.4, | |
| "step": 1380 | |
| }, | |
| { | |
| "loss": 4.2176, | |
| "grad_norm": 4.065763473510742, | |
| "learning_rate": 0.00042028985507246375, | |
| "epoch": 2.4347826086956523, | |
| "step": 1400 | |
| }, | |
| { | |
| "loss": 4.0597, | |
| "grad_norm": 4.120363712310791, | |
| "learning_rate": 0.00041835748792270534, | |
| "epoch": 2.4695652173913043, | |
| "step": 1420 | |
| }, | |
| { | |
| "loss": 4.028, | |
| "grad_norm": 4.3197174072265625, | |
| "learning_rate": 0.00041642512077294687, | |
| "epoch": 2.5043478260869563, | |
| "step": 1440 | |
| }, | |
| { | |
| "loss": 3.9833, | |
| "grad_norm": 4.2683610916137695, | |
| "learning_rate": 0.0004144927536231884, | |
| "epoch": 2.5391304347826087, | |
| "step": 1460 | |
| }, | |
| { | |
| "loss": 4.0065, | |
| "grad_norm": 4.15448522567749, | |
| "learning_rate": 0.00041256038647343, | |
| "epoch": 2.573913043478261, | |
| "step": 1480 | |
| }, | |
| { | |
| "loss": 3.8134, | |
| "grad_norm": 4.348177433013916, | |
| "learning_rate": 0.0004106280193236715, | |
| "epoch": 2.608695652173913, | |
| "step": 1500 | |
| }, | |
| { | |
| "loss": 3.8548, | |
| "grad_norm": 4.100021839141846, | |
| "learning_rate": 0.00040869565217391306, | |
| "epoch": 2.643478260869565, | |
| "step": 1520 | |
| }, | |
| { | |
| "loss": 3.7814, | |
| "grad_norm": 4.344174385070801, | |
| "learning_rate": 0.0004067632850241546, | |
| "epoch": 2.6782608695652175, | |
| "step": 1540 | |
| }, | |
| { | |
| "loss": 3.7578, | |
| "grad_norm": 4.240079402923584, | |
| "learning_rate": 0.00040483091787439613, | |
| "epoch": 2.7130434782608694, | |
| "step": 1560 | |
| }, | |
| { | |
| "loss": 3.7331, | |
| "grad_norm": 4.468689918518066, | |
| "learning_rate": 0.0004028985507246377, | |
| "epoch": 2.747826086956522, | |
| "step": 1580 | |
| }, | |
| { | |
| "loss": 3.6396, | |
| "grad_norm": 4.28464937210083, | |
| "learning_rate": 0.00040096618357487925, | |
| "epoch": 2.782608695652174, | |
| "step": 1600 | |
| }, | |
| { | |
| "loss": 3.5799, | |
| "grad_norm": 4.166805744171143, | |
| "learning_rate": 0.0003990338164251208, | |
| "epoch": 2.8173913043478263, | |
| "step": 1620 | |
| }, | |
| { | |
| "loss": 3.4734, | |
| "grad_norm": 4.237683296203613, | |
| "learning_rate": 0.0003971014492753624, | |
| "epoch": 2.8521739130434782, | |
| "step": 1640 | |
| }, | |
| { | |
| "loss": 3.5183, | |
| "grad_norm": 4.153097152709961, | |
| "learning_rate": 0.00039516908212560385, | |
| "epoch": 2.8869565217391306, | |
| "step": 1660 | |
| }, | |
| { | |
| "loss": 3.3963, | |
| "grad_norm": 4.2313947677612305, | |
| "learning_rate": 0.0003932367149758454, | |
| "epoch": 2.9217391304347826, | |
| "step": 1680 | |
| }, | |
| { | |
| "loss": 3.3081, | |
| "grad_norm": 3.992475748062134, | |
| "learning_rate": 0.000391304347826087, | |
| "epoch": 2.9565217391304346, | |
| "step": 1700 | |
| }, | |
| { | |
| "loss": 3.3124, | |
| "grad_norm": 4.4731059074401855, | |
| "learning_rate": 0.0003893719806763285, | |
| "epoch": 2.991304347826087, | |
| "step": 1720 | |
| }, | |
| { | |
| "eval_loss": 2.1082653999328613, | |
| "eval_accuracy": 0.9260255548083389, | |
| "eval_runtime": 22.1676, | |
| "eval_samples_per_second": 67.08, | |
| "eval_steps_per_second": 67.08, | |
| "epoch": 3.0, | |
| "step": 1725 | |
| }, | |
| { | |
| "loss": 3.1247, | |
| "grad_norm": 4.272000312805176, | |
| "learning_rate": 0.00038743961352657004, | |
| "epoch": 3.026086956521739, | |
| "step": 1740 | |
| }, | |
| { | |
| "loss": 3.1064, | |
| "grad_norm": 4.102330207824707, | |
| "learning_rate": 0.00038550724637681163, | |
| "epoch": 3.0608695652173914, | |
| "step": 1760 | |
| }, | |
| { | |
| "loss": 2.9371, | |
| "grad_norm": 4.381846904754639, | |
| "learning_rate": 0.00038357487922705317, | |
| "epoch": 3.0956521739130434, | |
| "step": 1780 | |
| }, | |
| { | |
| "loss": 2.9355, | |
| "grad_norm": 4.1588921546936035, | |
| "learning_rate": 0.00038164251207729465, | |
| "epoch": 3.130434782608696, | |
| "step": 1800 | |
| }, | |
| { | |
| "loss": 2.8545, | |
| "grad_norm": 4.279609203338623, | |
| "learning_rate": 0.00037971014492753623, | |
| "epoch": 3.1652173913043478, | |
| "step": 1820 | |
| }, | |
| { | |
| "loss": 2.8096, | |
| "grad_norm": 4.240756988525391, | |
| "learning_rate": 0.00037777777777777777, | |
| "epoch": 3.2, | |
| "step": 1840 | |
| }, | |
| { | |
| "loss": 2.8138, | |
| "grad_norm": 4.11091947555542, | |
| "learning_rate": 0.00037584541062801936, | |
| "epoch": 3.234782608695652, | |
| "step": 1860 | |
| }, | |
| { | |
| "loss": 2.7417, | |
| "grad_norm": 4.078794479370117, | |
| "learning_rate": 0.0003739130434782609, | |
| "epoch": 3.269565217391304, | |
| "step": 1880 | |
| }, | |
| { | |
| "loss": 2.7937, | |
| "grad_norm": 4.368116855621338, | |
| "learning_rate": 0.0003719806763285024, | |
| "epoch": 3.3043478260869565, | |
| "step": 1900 | |
| }, | |
| { | |
| "loss": 2.7361, | |
| "grad_norm": 4.044319152832031, | |
| "learning_rate": 0.000370048309178744, | |
| "epoch": 3.3391304347826085, | |
| "step": 1920 | |
| }, | |
| { | |
| "loss": 2.7054, | |
| "grad_norm": 4.314040184020996, | |
| "learning_rate": 0.0003681159420289855, | |
| "epoch": 3.373913043478261, | |
| "step": 1940 | |
| }, | |
| { | |
| "loss": 2.6682, | |
| "grad_norm": 4.185855388641357, | |
| "learning_rate": 0.000366183574879227, | |
| "epoch": 3.408695652173913, | |
| "step": 1960 | |
| }, | |
| { | |
| "loss": 2.6644, | |
| "grad_norm": 4.433622360229492, | |
| "learning_rate": 0.0003642512077294686, | |
| "epoch": 3.4434782608695653, | |
| "step": 1980 | |
| }, | |
| { | |
| "loss": 2.618, | |
| "grad_norm": 4.048947811126709, | |
| "learning_rate": 0.00036231884057971015, | |
| "epoch": 3.4782608695652173, | |
| "step": 2000 | |
| }, | |
| { | |
| "loss": 2.5982, | |
| "grad_norm": 4.145406246185303, | |
| "learning_rate": 0.0003603864734299517, | |
| "epoch": 3.5130434782608697, | |
| "step": 2020 | |
| }, | |
| { | |
| "loss": 2.6138, | |
| "grad_norm": 4.2812910079956055, | |
| "learning_rate": 0.00035845410628019327, | |
| "epoch": 3.5478260869565217, | |
| "step": 2040 | |
| }, | |
| { | |
| "loss": 2.5039, | |
| "grad_norm": 4.400162220001221, | |
| "learning_rate": 0.0003565217391304348, | |
| "epoch": 3.5826086956521737, | |
| "step": 2060 | |
| }, | |
| { | |
| "loss": 2.5249, | |
| "grad_norm": 4.217800617218018, | |
| "learning_rate": 0.0003545893719806763, | |
| "epoch": 3.617391304347826, | |
| "step": 2080 | |
| }, | |
| { | |
| "loss": 2.4547, | |
| "grad_norm": 4.076215744018555, | |
| "learning_rate": 0.0003526570048309179, | |
| "epoch": 3.6521739130434785, | |
| "step": 2100 | |
| }, | |
| { | |
| "loss": 2.4315, | |
| "grad_norm": 4.139514446258545, | |
| "learning_rate": 0.0003507246376811594, | |
| "epoch": 3.6869565217391305, | |
| "step": 2120 | |
| }, | |
| { | |
| "loss": 2.3836, | |
| "grad_norm": 4.118022918701172, | |
| "learning_rate": 0.00034879227053140094, | |
| "epoch": 3.7217391304347824, | |
| "step": 2140 | |
| }, | |
| { | |
| "loss": 2.3284, | |
| "grad_norm": 4.137601852416992, | |
| "learning_rate": 0.00034685990338164253, | |
| "epoch": 3.756521739130435, | |
| "step": 2160 | |
| }, | |
| { | |
| "loss": 2.3095, | |
| "grad_norm": 4.023979663848877, | |
| "learning_rate": 0.00034492753623188406, | |
| "epoch": 3.791304347826087, | |
| "step": 2180 | |
| }, | |
| { | |
| "loss": 2.305, | |
| "grad_norm": 4.042725086212158, | |
| "learning_rate": 0.00034299516908212565, | |
| "epoch": 3.8260869565217392, | |
| "step": 2200 | |
| }, | |
| { | |
| "loss": 2.3237, | |
| "grad_norm": 4.265875339508057, | |
| "learning_rate": 0.0003410628019323672, | |
| "epoch": 3.860869565217391, | |
| "step": 2220 | |
| }, | |
| { | |
| "loss": 2.335, | |
| "grad_norm": 4.205041408538818, | |
| "learning_rate": 0.00033913043478260867, | |
| "epoch": 3.8956521739130436, | |
| "step": 2240 | |
| }, | |
| { | |
| "loss": 2.2341, | |
| "grad_norm": 4.1344709396362305, | |
| "learning_rate": 0.00033719806763285025, | |
| "epoch": 3.9304347826086956, | |
| "step": 2260 | |
| }, | |
| { | |
| "loss": 2.251, | |
| "grad_norm": 4.247790813446045, | |
| "learning_rate": 0.0003352657004830918, | |
| "epoch": 3.965217391304348, | |
| "step": 2280 | |
| }, | |
| { | |
| "loss": 2.3212, | |
| "grad_norm": 4.859626770019531, | |
| "learning_rate": 0.0003333333333333333, | |
| "epoch": 4.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "eval_loss": 1.2223739624023438, | |
| "eval_accuracy": 0.9435104236718225, | |
| "eval_runtime": 14.8513, | |
| "eval_samples_per_second": 100.126, | |
| "eval_steps_per_second": 100.126, | |
| "epoch": 4.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "loss": 1.9133, | |
| "grad_norm": 4.098020553588867, | |
| "learning_rate": 0.0003314009661835749, | |
| "epoch": 4.034782608695652, | |
| "step": 2320 | |
| }, | |
| { | |
| "loss": 1.9814, | |
| "grad_norm": 4.198029041290283, | |
| "learning_rate": 0.00032946859903381644, | |
| "epoch": 4.069565217391304, | |
| "step": 2340 | |
| }, | |
| { | |
| "loss": 1.9505, | |
| "grad_norm": 3.960844039916992, | |
| "learning_rate": 0.000327536231884058, | |
| "epoch": 4.104347826086957, | |
| "step": 2360 | |
| }, | |
| { | |
| "loss": 1.8815, | |
| "grad_norm": 4.0190300941467285, | |
| "learning_rate": 0.0003256038647342995, | |
| "epoch": 4.139130434782609, | |
| "step": 2380 | |
| }, | |
| { | |
| "loss": 1.8365, | |
| "grad_norm": 4.040708541870117, | |
| "learning_rate": 0.00032367149758454105, | |
| "epoch": 4.173913043478261, | |
| "step": 2400 | |
| }, | |
| { | |
| "loss": 1.84, | |
| "grad_norm": 4.077364444732666, | |
| "learning_rate": 0.0003217391304347826, | |
| "epoch": 4.208695652173913, | |
| "step": 2420 | |
| }, | |
| { | |
| "loss": 1.8864, | |
| "grad_norm": 4.267309188842773, | |
| "learning_rate": 0.0003199033816425121, | |
| "epoch": 4.243478260869566, | |
| "step": 2440 | |
| }, | |
| { | |
| "loss": 1.9015, | |
| "grad_norm": 3.978663921356201, | |
| "learning_rate": 0.00031797101449275363, | |
| "epoch": 4.278260869565218, | |
| "step": 2460 | |
| }, | |
| { | |
| "loss": 1.8388, | |
| "grad_norm": 4.089256763458252, | |
| "learning_rate": 0.0003160386473429952, | |
| "epoch": 4.3130434782608695, | |
| "step": 2480 | |
| }, | |
| { | |
| "loss": 1.7845, | |
| "grad_norm": 3.9317057132720947, | |
| "learning_rate": 0.0003141062801932367, | |
| "epoch": 4.3478260869565215, | |
| "step": 2500 | |
| }, | |
| { | |
| "loss": 1.7725, | |
| "grad_norm": 3.9738080501556396, | |
| "learning_rate": 0.00031217391304347823, | |
| "epoch": 4.3826086956521735, | |
| "step": 2520 | |
| }, | |
| { | |
| "loss": 1.852, | |
| "grad_norm": 4.232215881347656, | |
| "learning_rate": 0.0003102415458937198, | |
| "epoch": 4.417391304347826, | |
| "step": 2540 | |
| }, | |
| { | |
| "loss": 1.8234, | |
| "grad_norm": 4.050131797790527, | |
| "learning_rate": 0.00030830917874396136, | |
| "epoch": 4.452173913043478, | |
| "step": 2560 | |
| }, | |
| { | |
| "loss": 1.8148, | |
| "grad_norm": 4.217935085296631, | |
| "learning_rate": 0.0003063768115942029, | |
| "epoch": 4.48695652173913, | |
| "step": 2580 | |
| }, | |
| { | |
| "loss": 1.7134, | |
| "grad_norm": 3.9807074069976807, | |
| "learning_rate": 0.0003044444444444445, | |
| "epoch": 4.521739130434782, | |
| "step": 2600 | |
| }, | |
| { | |
| "loss": 1.6752, | |
| "grad_norm": 4.05940580368042, | |
| "learning_rate": 0.000302512077294686, | |
| "epoch": 4.556521739130435, | |
| "step": 2620 | |
| }, | |
| { | |
| "loss": 1.8413, | |
| "grad_norm": 4.454566955566406, | |
| "learning_rate": 0.00030057971014492755, | |
| "epoch": 4.591304347826087, | |
| "step": 2640 | |
| }, | |
| { | |
| "loss": 1.7948, | |
| "grad_norm": 4.144088268280029, | |
| "learning_rate": 0.0002986473429951691, | |
| "epoch": 4.626086956521739, | |
| "step": 2660 | |
| }, | |
| { | |
| "loss": 1.7468, | |
| "grad_norm": 3.940176010131836, | |
| "learning_rate": 0.0002967149758454106, | |
| "epoch": 4.660869565217391, | |
| "step": 2680 | |
| }, | |
| { | |
| "loss": 1.709, | |
| "grad_norm": 4.198675632476807, | |
| "learning_rate": 0.0002948792270531401, | |
| "epoch": 4.695652173913043, | |
| "step": 2700 | |
| }, | |
| { | |
| "loss": 1.6506, | |
| "grad_norm": 3.976001501083374, | |
| "learning_rate": 0.00029294685990338167, | |
| "epoch": 4.730434782608696, | |
| "step": 2720 | |
| }, | |
| { | |
| "loss": 1.7042, | |
| "grad_norm": 4.033059120178223, | |
| "learning_rate": 0.0002910144927536232, | |
| "epoch": 4.765217391304348, | |
| "step": 2740 | |
| }, | |
| { | |
| "loss": 1.6795, | |
| "grad_norm": 4.062041759490967, | |
| "learning_rate": 0.0002890821256038648, | |
| "epoch": 4.8, | |
| "step": 2760 | |
| }, | |
| { | |
| "loss": 1.7029, | |
| "grad_norm": 3.988589286804199, | |
| "learning_rate": 0.00028714975845410627, | |
| "epoch": 4.834782608695652, | |
| "step": 2780 | |
| }, | |
| { | |
| "loss": 1.6641, | |
| "grad_norm": 4.16325044631958, | |
| "learning_rate": 0.0002852173913043478, | |
| "epoch": 4.869565217391305, | |
| "step": 2800 | |
| }, | |
| { | |
| "loss": 1.6953, | |
| "grad_norm": 4.323537349700928, | |
| "learning_rate": 0.0002832850241545894, | |
| "epoch": 4.904347826086957, | |
| "step": 2820 | |
| }, | |
| { | |
| "loss": 1.5863, | |
| "grad_norm": 3.8293144702911377, | |
| "learning_rate": 0.0002813526570048309, | |
| "epoch": 4.939130434782609, | |
| "step": 2840 | |
| }, | |
| { | |
| "loss": 1.6276, | |
| "grad_norm": 3.8955535888671875, | |
| "learning_rate": 0.00027942028985507246, | |
| "epoch": 4.973913043478261, | |
| "step": 2860 | |
| }, | |
| { | |
| "eval_loss": 0.8229038715362549, | |
| "eval_accuracy": 0.9677202420981843, | |
| "eval_runtime": 88.6744, | |
| "eval_samples_per_second": 16.769, | |
| "eval_steps_per_second": 16.769, | |
| "epoch": 5.0, | |
| "step": 2875 | |
| }, | |
| { | |
| "loss": 1.5701, | |
| "grad_norm": 3.8480091094970703, | |
| "learning_rate": 0.00027748792270531405, | |
| "epoch": 5.008695652173913, | |
| "step": 2880 | |
| }, | |
| { | |
| "loss": 1.3786, | |
| "grad_norm": 3.679872512817383, | |
| "learning_rate": 0.0002755555555555556, | |
| "epoch": 5.043478260869565, | |
| "step": 2900 | |
| }, | |
| { | |
| "loss": 1.3563, | |
| "grad_norm": 4.13381290435791, | |
| "learning_rate": 0.00027362318840579706, | |
| "epoch": 5.078260869565217, | |
| "step": 2920 | |
| }, | |
| { | |
| "loss": 1.3588, | |
| "grad_norm": 3.7467329502105713, | |
| "learning_rate": 0.00027169082125603865, | |
| "epoch": 5.113043478260869, | |
| "step": 2940 | |
| }, | |
| { | |
| "loss": 1.3782, | |
| "grad_norm": 3.5837419033050537, | |
| "learning_rate": 0.0002698550724637681, | |
| "epoch": 5.147826086956521, | |
| "step": 2960 | |
| }, | |
| { | |
| "loss": 1.3969, | |
| "grad_norm": 4.077097415924072, | |
| "learning_rate": 0.00026792270531400964, | |
| "epoch": 5.182608695652174, | |
| "step": 2980 | |
| }, | |
| { | |
| "loss": 1.3346, | |
| "grad_norm": 3.5995211601257324, | |
| "learning_rate": 0.00026599033816425123, | |
| "epoch": 5.217391304347826, | |
| "step": 3000 | |
| }, | |
| { | |
| "loss": 1.3772, | |
| "grad_norm": 3.714010000228882, | |
| "learning_rate": 0.00026405797101449277, | |
| "epoch": 5.252173913043478, | |
| "step": 3020 | |
| }, | |
| { | |
| "loss": 1.3452, | |
| "grad_norm": 3.807094097137451, | |
| "learning_rate": 0.00026231884057971016, | |
| "epoch": 5.28695652173913, | |
| "step": 3040 | |
| }, | |
| { | |
| "loss": 1.3161, | |
| "grad_norm": 4.012477397918701, | |
| "learning_rate": 0.0002603864734299517, | |
| "epoch": 5.321739130434783, | |
| "step": 3060 | |
| }, | |
| { | |
| "loss": 1.3146, | |
| "grad_norm": 3.850520372390747, | |
| "learning_rate": 0.0002584541062801932, | |
| "epoch": 5.356521739130435, | |
| "step": 3080 | |
| }, | |
| { | |
| "loss": 1.3057, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00025661835748792274, | |
| "epoch": 5.391304347826087, | |
| "step": 3100 | |
| }, | |
| { | |
| "loss": 1.2619, | |
| "grad_norm": 3.697744607925415, | |
| "learning_rate": 0.0002546859903381643, | |
| "epoch": 5.426086956521739, | |
| "step": 3120 | |
| }, | |
| { | |
| "loss": 1.3436, | |
| "grad_norm": 4.125018119812012, | |
| "learning_rate": 0.00025275362318840576, | |
| "epoch": 5.460869565217392, | |
| "step": 3140 | |
| }, | |
| { | |
| "loss": 1.3289, | |
| "grad_norm": 4.1491899490356445, | |
| "learning_rate": 0.00025082125603864735, | |
| "epoch": 5.495652173913044, | |
| "step": 3160 | |
| }, | |
| { | |
| "loss": 1.218, | |
| "grad_norm": 3.9294846057891846, | |
| "learning_rate": 0.0002488888888888889, | |
| "epoch": 5.530434782608696, | |
| "step": 3180 | |
| }, | |
| { | |
| "loss": 1.3219, | |
| "grad_norm": 3.9030706882476807, | |
| "learning_rate": 0.00024695652173913047, | |
| "epoch": 5.565217391304348, | |
| "step": 3200 | |
| }, | |
| { | |
| "loss": 1.2694, | |
| "grad_norm": 4.124849319458008, | |
| "learning_rate": 0.000245024154589372, | |
| "epoch": 5.6, | |
| "step": 3220 | |
| }, | |
| { | |
| "loss": 1.2379, | |
| "grad_norm": 4.1668500900268555, | |
| "learning_rate": 0.0002432850241545894, | |
| "epoch": 5.6347826086956525, | |
| "step": 3240 | |
| }, | |
| { | |
| "loss": 1.2892, | |
| "grad_norm": 4.098198890686035, | |
| "learning_rate": 0.00024135265700483093, | |
| "epoch": 5.6695652173913045, | |
| "step": 3260 | |
| }, | |
| { | |
| "loss": 1.2742, | |
| "grad_norm": 3.690241813659668, | |
| "learning_rate": 0.00023942028985507246, | |
| "epoch": 5.7043478260869565, | |
| "step": 3280 | |
| }, | |
| { | |
| "loss": 1.1755, | |
| "grad_norm": 3.978963613510132, | |
| "learning_rate": 0.00023748792270531402, | |
| "epoch": 5.739130434782608, | |
| "step": 3300 | |
| }, | |
| { | |
| "loss": 1.2256, | |
| "grad_norm": 3.7397215366363525, | |
| "learning_rate": 0.00023574879227053139, | |
| "epoch": 5.773913043478261, | |
| "step": 3320 | |
| }, | |
| { | |
| "loss": 1.238, | |
| "grad_norm": 3.9201064109802246, | |
| "learning_rate": 0.00023391304347826088, | |
| "epoch": 5.808695652173913, | |
| "step": 3340 | |
| }, | |
| { | |
| "loss": 1.1706, | |
| "grad_norm": 3.725389242172241, | |
| "learning_rate": 0.0002319806763285024, | |
| "epoch": 5.843478260869565, | |
| "step": 3360 | |
| }, | |
| { | |
| "loss": 1.1644, | |
| "grad_norm": 3.5844123363494873, | |
| "learning_rate": 0.00023004830917874397, | |
| "epoch": 5.878260869565217, | |
| "step": 3380 | |
| }, | |
| { | |
| "loss": 1.2256, | |
| "grad_norm": 3.79936146736145, | |
| "learning_rate": 0.00022821256038647343, | |
| "epoch": 5.913043478260869, | |
| "step": 3400 | |
| }, | |
| { | |
| "loss": 1.2488, | |
| "grad_norm": 3.5947725772857666, | |
| "learning_rate": 0.00022628019323671497, | |
| "epoch": 5.947826086956522, | |
| "step": 3420 | |
| }, | |
| { | |
| "loss": 1.1418, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00022444444444444446, | |
| "epoch": 5.982608695652174, | |
| "step": 3440 | |
| }, | |
| { | |
| "eval_loss": 0.5840117335319519, | |
| "eval_accuracy": 0.9757901815736382, | |
| "eval_runtime": 97.2696, | |
| "eval_samples_per_second": 15.287, | |
| "eval_steps_per_second": 15.287, | |
| "epoch": 6.0, | |
| "step": 3450 | |
| }, | |
| { | |
| "loss": 1.1254, | |
| "grad_norm": 3.5959298610687256, | |
| "learning_rate": 0.00022260869565217392, | |
| "epoch": 6.017391304347826, | |
| "step": 3460 | |
| }, | |
| { | |
| "loss": 1.0343, | |
| "grad_norm": 3.9623775482177734, | |
| "learning_rate": 0.00022067632850241545, | |
| "epoch": 6.052173913043478, | |
| "step": 3480 | |
| }, | |
| { | |
| "loss": 1.0348, | |
| "grad_norm": 3.735102415084839, | |
| "learning_rate": 0.00021874396135265702, | |
| "epoch": 6.086956521739131, | |
| "step": 3500 | |
| }, | |
| { | |
| "loss": 0.9796, | |
| "grad_norm": 3.4255013465881348, | |
| "learning_rate": 0.00021681159420289855, | |
| "epoch": 6.121739130434783, | |
| "step": 3520 | |
| }, | |
| { | |
| "loss": 0.9865, | |
| "grad_norm": 3.981841564178467, | |
| "learning_rate": 0.00021497584541062804, | |
| "epoch": 6.156521739130435, | |
| "step": 3540 | |
| }, | |
| { | |
| "loss": 1.0054, | |
| "grad_norm": 3.9057116508483887, | |
| "learning_rate": 0.00021314009661835748, | |
| "epoch": 6.191304347826087, | |
| "step": 3560 | |
| }, | |
| { | |
| "loss": 1.0012, | |
| "grad_norm": 3.626560688018799, | |
| "learning_rate": 0.00021120772946859904, | |
| "epoch": 6.226086956521739, | |
| "step": 3580 | |
| }, | |
| { | |
| "loss": 1.0129, | |
| "grad_norm": 3.687683582305908, | |
| "learning_rate": 0.0002093719806763285, | |
| "epoch": 6.260869565217392, | |
| "step": 3600 | |
| }, | |
| { | |
| "loss": 0.9333, | |
| "grad_norm": 3.8632826805114746, | |
| "learning_rate": 0.00020763285024154592, | |
| "epoch": 6.2956521739130435, | |
| "step": 3620 | |
| }, | |
| { | |
| "loss": 1.0259, | |
| "grad_norm": 4.089422702789307, | |
| "learning_rate": 0.0002058937198067633, | |
| "epoch": 6.3304347826086955, | |
| "step": 3640 | |
| }, | |
| { | |
| "loss": 1.0184, | |
| "grad_norm": 4.261268615722656, | |
| "learning_rate": 0.00020415458937198067, | |
| "epoch": 6.3652173913043475, | |
| "step": 3660 | |
| }, | |
| { | |
| "loss": 1.0293, | |
| "grad_norm": 2.3901586532592773, | |
| "learning_rate": 0.0002026086956521739, | |
| "epoch": 6.4, | |
| "step": 3680 | |
| }, | |
| { | |
| "loss": 1.0026, | |
| "grad_norm": 2.233633518218994, | |
| "learning_rate": 0.00020067632850241546, | |
| "epoch": 6.434782608695652, | |
| "step": 3700 | |
| }, | |
| { | |
| "loss": 1.0426, | |
| "grad_norm": 2.049773693084717, | |
| "learning_rate": 0.00019893719806763285, | |
| "epoch": 6.469565217391304, | |
| "step": 3720 | |
| }, | |
| { | |
| "loss": 1.0324, | |
| "grad_norm": 2.21939754486084, | |
| "learning_rate": 0.0001970048309178744, | |
| "epoch": 6.504347826086956, | |
| "step": 3740 | |
| }, | |
| { | |
| "loss": 1.0666, | |
| "grad_norm": 2.2138895988464355, | |
| "learning_rate": 0.00019516908212560387, | |
| "epoch": 6.539130434782608, | |
| "step": 3760 | |
| }, | |
| { | |
| "loss": 1.0724, | |
| "grad_norm": 1.9186855554580688, | |
| "learning_rate": 0.0001932367149758454, | |
| "epoch": 6.573913043478261, | |
| "step": 3780 | |
| }, | |
| { | |
| "loss": 1.0867, | |
| "grad_norm": 1.302451729774475, | |
| "learning_rate": 0.00019159420289855073, | |
| "epoch": 6.608695652173913, | |
| "step": 3800 | |
| }, | |
| { | |
| "loss": 1.0659, | |
| "grad_norm": 1.1770459413528442, | |
| "learning_rate": 0.00018975845410628022, | |
| "epoch": 6.643478260869565, | |
| "step": 3820 | |
| }, | |
| { | |
| "loss": 1.0494, | |
| "grad_norm": 0.2651650309562683, | |
| "learning_rate": 0.0001881159420289855, | |
| "epoch": 6.678260869565217, | |
| "step": 3840 | |
| }, | |
| { | |
| "loss": 1.0464, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0001867632850241546, | |
| "epoch": 6.71304347826087, | |
| "step": 3860 | |
| }, | |
| { | |
| "loss": 1.0457, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.000185024154589372, | |
| "epoch": 6.747826086956522, | |
| "step": 3880 | |
| }, | |
| { | |
| "loss": 0.9815, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00018328502415458937, | |
| "epoch": 6.782608695652174, | |
| "step": 3900 | |
| }, | |
| { | |
| "loss": 1.0094, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0001816425120772947, | |
| "epoch": 6.817391304347826, | |
| "step": 3920 | |
| }, | |
| { | |
| "loss": 1.0023, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018028985507246377, | |
| "epoch": 6.852173913043478, | |
| "step": 3940 | |
| }, | |
| { | |
| "loss": 1.0278, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00017893719806763288, | |
| "epoch": 6.886956521739131, | |
| "step": 3960 | |
| }, | |
| { | |
| "loss": 1.0123, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0001771014492753623, | |
| "epoch": 6.921739130434783, | |
| "step": 3980 | |
| }, | |
| { | |
| "loss": 1.0774, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00017565217391304346, | |
| "epoch": 6.956521739130435, | |
| "step": 4000 | |
| }, | |
| { | |
| "loss": 1.0484, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00017391304347826088, | |
| "epoch": 6.9913043478260875, | |
| "step": 4020 | |
| }, | |
| { | |
| "eval_loss": 0.5780686736106873, | |
| "eval_accuracy": 0.9737726967047747, | |
| "eval_runtime": 118.8154, | |
| "eval_samples_per_second": 12.515, | |
| "eval_steps_per_second": 12.515, | |
| "epoch": 7.0, | |
| "step": 4025 | |
| }, | |
| { | |
| "loss": 0.9799, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0001723671497584541, | |
| "epoch": 7.026086956521739, | |
| "step": 4040 | |
| }, | |
| { | |
| "loss": 0.9588, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00017091787439613525, | |
| "epoch": 7.060869565217391, | |
| "step": 4060 | |
| }, | |
| { | |
| "loss": 0.9421, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016966183574879226, | |
| "epoch": 7.095652173913043, | |
| "step": 4080 | |
| }, | |
| { | |
| "loss": 0.9551, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00016782608695652175, | |
| "epoch": 7.130434782608695, | |
| "step": 4100 | |
| }, | |
| { | |
| "loss": 0.9622, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00016618357487922704, | |
| "epoch": 7.165217391304348, | |
| "step": 4120 | |
| }, | |
| { | |
| "loss": 0.9712, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00016444444444444446, | |
| "epoch": 7.2, | |
| "step": 4140 | |
| }, | |
| { | |
| "loss": 0.9834, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00016299516908212561, | |
| "epoch": 7.234782608695652, | |
| "step": 4160 | |
| }, | |
| { | |
| "loss": 0.9968, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016135265700483093, | |
| "epoch": 7.269565217391304, | |
| "step": 4180 | |
| }, | |
| { | |
| "loss": 0.956, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00015961352657004833, | |
| "epoch": 7.304347826086957, | |
| "step": 4200 | |
| }, | |
| { | |
| "loss": 0.8981, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00015806763285024155, | |
| "epoch": 7.339130434782609, | |
| "step": 4220 | |
| }, | |
| { | |
| "loss": 0.9515, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00015642512077294684, | |
| "epoch": 7.373913043478261, | |
| "step": 4240 | |
| }, | |
| { | |
| "loss": 0.9535, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0001548792270531401, | |
| "epoch": 7.408695652173913, | |
| "step": 4260 | |
| }, | |
| { | |
| "loss": 0.9646, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015333333333333334, | |
| "epoch": 7.443478260869565, | |
| "step": 4280 | |
| }, | |
| { | |
| "loss": 0.9821, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00015140096618357487, | |
| "epoch": 7.478260869565218, | |
| "step": 4300 | |
| }, | |
| { | |
| "loss": 0.9259, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00015014492753623188, | |
| "epoch": 7.51304347826087, | |
| "step": 4320 | |
| }, | |
| { | |
| "loss": 0.9494, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00014869565217391303, | |
| "epoch": 7.547826086956522, | |
| "step": 4340 | |
| }, | |
| { | |
| "loss": 0.9305, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00014714975845410628, | |
| "epoch": 7.582608695652174, | |
| "step": 4360 | |
| }, | |
| { | |
| "loss": 0.8889, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0001455072463768116, | |
| "epoch": 7.6173913043478265, | |
| "step": 4380 | |
| }, | |
| { | |
| "loss": 0.9524, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00014396135265700482, | |
| "epoch": 7.6521739130434785, | |
| "step": 4400 | |
| }, | |
| { | |
| "loss": 0.9065, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00014231884057971014, | |
| "epoch": 7.6869565217391305, | |
| "step": 4420 | |
| }, | |
| { | |
| "loss": 0.9153, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.00014048309178743963, | |
| "epoch": 7.721739130434782, | |
| "step": 4440 | |
| }, | |
| { | |
| "loss": 0.6675, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 7.756521739130434, | |
| "step": 4460 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 7.791304347826087, | |
| "step": 4480 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 7.826086956521739, | |
| "step": 4500 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 7.860869565217391, | |
| "step": 4520 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 7.895652173913043, | |
| "step": 4540 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 7.930434782608696, | |
| "step": 4560 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 7.965217391304348, | |
| "step": 4580 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.0, | |
| "step": 4600 | |
| }, | |
| { | |
| "eval_loss": NaN, | |
| "eval_accuracy": 0.0006724949562878278, | |
| "eval_runtime": 129.6238, | |
| "eval_samples_per_second": 11.472, | |
| "eval_steps_per_second": 11.472, | |
| "epoch": 8.0, | |
| "step": 4600 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.034782608695652, | |
| "step": 4620 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.069565217391304, | |
| "step": 4640 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.104347826086956, | |
| "step": 4660 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.139130434782608, | |
| "step": 4680 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.173913043478262, | |
| "step": 4700 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.208695652173914, | |
| "step": 4720 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.243478260869566, | |
| "step": 4740 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.278260869565218, | |
| "step": 4760 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.31304347826087, | |
| "step": 4780 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.347826086956522, | |
| "step": 4800 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.382608695652173, | |
| "step": 4820 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.417391304347825, | |
| "step": 4840 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.452173913043477, | |
| "step": 4860 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.486956521739131, | |
| "step": 4880 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.521739130434783, | |
| "step": 4900 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.556521739130435, | |
| "step": 4920 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.591304347826087, | |
| "step": 4940 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.626086956521739, | |
| "step": 4960 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.660869565217391, | |
| "step": 4980 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.695652173913043, | |
| "step": 5000 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.730434782608695, | |
| "step": 5020 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.765217391304347, | |
| "step": 5040 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.8, | |
| "step": 5060 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.834782608695653, | |
| "step": 5080 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.869565217391305, | |
| "step": 5100 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.904347826086957, | |
| "step": 5120 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.939130434782609, | |
| "step": 5140 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 8.97391304347826, | |
| "step": 5160 | |
| }, | |
| { | |
| "eval_loss": NaN, | |
| "eval_accuracy": 0.0006724949562878278, | |
| "eval_runtime": 117.1288, | |
| "eval_samples_per_second": 12.695, | |
| "eval_steps_per_second": 12.695, | |
| "epoch": 9.0, | |
| "step": 5175 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.008695652173913, | |
| "step": 5180 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.043478260869565, | |
| "step": 5200 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.078260869565218, | |
| "step": 5220 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.11304347826087, | |
| "step": 5240 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.147826086956522, | |
| "step": 5260 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.182608695652174, | |
| "step": 5280 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.217391304347826, | |
| "step": 5300 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.252173913043478, | |
| "step": 5320 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.28695652173913, | |
| "step": 5340 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.321739130434782, | |
| "step": 5360 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.356521739130434, | |
| "step": 5380 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.391304347826088, | |
| "step": 5400 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.42608695652174, | |
| "step": 5420 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.460869565217392, | |
| "step": 5440 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.495652173913044, | |
| "step": 5460 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.530434782608696, | |
| "step": 5480 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.565217391304348, | |
| "step": 5500 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.6, | |
| "step": 5520 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.634782608695652, | |
| "step": 5540 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.669565217391304, | |
| "step": 5560 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.704347826086957, | |
| "step": 5580 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.73913043478261, | |
| "step": 5600 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.773913043478261, | |
| "step": 5620 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.808695652173913, | |
| "step": 5640 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.843478260869565, | |
| "step": 5660 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.878260869565217, | |
| "step": 5680 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.91304347826087, | |
| "step": 5700 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.947826086956521, | |
| "step": 5720 | |
| }, | |
| { | |
| "loss": 0.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001403864734299517, | |
| "epoch": 9.982608695652173, | |
| "step": 5740 | |
| }, | |
| { | |
| "eval_loss": NaN, | |
| "eval_accuracy": 0.0006724949562878278, | |
| "eval_runtime": 103.3199, | |
| "eval_samples_per_second": 14.392, | |
| "eval_steps_per_second": 14.392, | |
| "epoch": 10.0, | |
| "step": 5750 | |
| }, | |
| { | |
| "train_runtime": 59857.6179, | |
| "train_samples_per_second": 24.584, | |
| "train_steps_per_second": 0.096, | |
| "total_flos": 2.7398100529152e+18, | |
| "train_loss": 2.9414075751926587, | |
| "epoch": 10.0, | |
| "step": 5750 | |
| } | |
| ] |