| { | |
| "best_metric": 0.7241379310344828, | |
| "best_model_checkpoint": "swinv2-base-patch4-window8-256/checkpoint-28", | |
| "epoch": 29.734513274336283, | |
| "eval_steps": 500, | |
| "global_step": 840, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.35398230088495575, | |
| "grad_norm": 9.937736511230469, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.6341, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7079646017699115, | |
| "grad_norm": 7.19003438949585, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.5428, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9911504424778761, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6211036443710327, | |
| "eval_runtime": 1.0465, | |
| "eval_samples_per_second": 27.71, | |
| "eval_steps_per_second": 27.71, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0619469026548674, | |
| "grad_norm": 9.382635116577148, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.5783, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.415929203539823, | |
| "grad_norm": 6.914736270904541, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.5546, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.7699115044247788, | |
| "grad_norm": 5.816603183746338, | |
| "learning_rate": 5.9523809523809524e-05, | |
| "loss": 0.6494, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9823008849557522, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6130104064941406, | |
| "eval_runtime": 1.0448, | |
| "eval_samples_per_second": 27.757, | |
| "eval_steps_per_second": 27.757, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.1238938053097347, | |
| "grad_norm": 1.15414559841156, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 0.6304, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.47787610619469, | |
| "grad_norm": 3.671969175338745, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.6236, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.831858407079646, | |
| "grad_norm": 7.640573501586914, | |
| "learning_rate": 9.523809523809524e-05, | |
| "loss": 0.5752, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.9734513274336285, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6845608949661255, | |
| "eval_runtime": 1.0508, | |
| "eval_samples_per_second": 27.597, | |
| "eval_steps_per_second": 27.597, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.185840707964602, | |
| "grad_norm": 1.3297613859176636, | |
| "learning_rate": 9.920634920634922e-05, | |
| "loss": 0.6231, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5398230088495577, | |
| "grad_norm": 1.8182591199874878, | |
| "learning_rate": 9.78835978835979e-05, | |
| "loss": 0.6603, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.893805309734513, | |
| "grad_norm": 0.6220849752426147, | |
| "learning_rate": 9.656084656084657e-05, | |
| "loss": 0.7165, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.964207112789154, | |
| "eval_runtime": 2.4229, | |
| "eval_samples_per_second": 11.969, | |
| "eval_steps_per_second": 11.969, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.247787610619469, | |
| "grad_norm": 0.8275606036186218, | |
| "learning_rate": 9.523809523809524e-05, | |
| "loss": 0.6544, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.601769911504425, | |
| "grad_norm": 5.596276760101318, | |
| "learning_rate": 9.391534391534393e-05, | |
| "loss": 0.6842, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.95575221238938, | |
| "grad_norm": 0.6999258995056152, | |
| "learning_rate": 9.25925925925926e-05, | |
| "loss": 0.5699, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.991150442477876, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6072261929512024, | |
| "eval_runtime": 1.0484, | |
| "eval_samples_per_second": 27.662, | |
| "eval_steps_per_second": 27.662, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 5.3097345132743365, | |
| "grad_norm": 5.042980194091797, | |
| "learning_rate": 9.126984126984128e-05, | |
| "loss": 0.6453, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.663716814159292, | |
| "grad_norm": 2.5722317695617676, | |
| "learning_rate": 8.994708994708995e-05, | |
| "loss": 0.5517, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.982300884955752, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6231208443641663, | |
| "eval_runtime": 1.0298, | |
| "eval_samples_per_second": 28.161, | |
| "eval_steps_per_second": 28.161, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 6.017699115044247, | |
| "grad_norm": 0.5366746783256531, | |
| "learning_rate": 8.862433862433864e-05, | |
| "loss": 0.5733, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.371681415929204, | |
| "grad_norm": 0.9455431699752808, | |
| "learning_rate": 8.730158730158731e-05, | |
| "loss": 0.6246, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.725663716814159, | |
| "grad_norm": 1.3559240102767944, | |
| "learning_rate": 8.597883597883598e-05, | |
| "loss": 0.5268, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.9734513274336285, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6098366379737854, | |
| "eval_runtime": 1.0741, | |
| "eval_samples_per_second": 27.0, | |
| "eval_steps_per_second": 27.0, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 7.079646017699115, | |
| "grad_norm": 0.570971667766571, | |
| "learning_rate": 8.465608465608466e-05, | |
| "loss": 0.5397, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.433628318584071, | |
| "grad_norm": 1.1077288389205933, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.4539, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.787610619469026, | |
| "grad_norm": 1.5305235385894775, | |
| "learning_rate": 8.201058201058202e-05, | |
| "loss": 0.672, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5890592932701111, | |
| "eval_runtime": 1.1176, | |
| "eval_samples_per_second": 25.948, | |
| "eval_steps_per_second": 25.948, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 8.141592920353983, | |
| "grad_norm": 1.5520473718643188, | |
| "learning_rate": 8.068783068783069e-05, | |
| "loss": 0.5567, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 8.495575221238939, | |
| "grad_norm": 1.5060292482376099, | |
| "learning_rate": 7.936507936507937e-05, | |
| "loss": 0.6923, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.849557522123893, | |
| "grad_norm": 0.2911463677883148, | |
| "learning_rate": 7.804232804232805e-05, | |
| "loss": 0.5448, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.991150442477876, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6022756695747375, | |
| "eval_runtime": 1.0313, | |
| "eval_samples_per_second": 28.12, | |
| "eval_steps_per_second": 28.12, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 9.20353982300885, | |
| "grad_norm": 0.38871487975120544, | |
| "learning_rate": 7.671957671957673e-05, | |
| "loss": 0.4555, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.557522123893806, | |
| "grad_norm": 1.8789817094802856, | |
| "learning_rate": 7.53968253968254e-05, | |
| "loss": 0.7061, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 9.91150442477876, | |
| "grad_norm": 1.5270930528640747, | |
| "learning_rate": 7.407407407407407e-05, | |
| "loss": 0.555, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.982300884955752, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5916692614555359, | |
| "eval_runtime": 1.0352, | |
| "eval_samples_per_second": 28.013, | |
| "eval_steps_per_second": 28.013, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 10.265486725663717, | |
| "grad_norm": 1.4813481569290161, | |
| "learning_rate": 7.275132275132276e-05, | |
| "loss": 0.5704, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 10.619469026548673, | |
| "grad_norm": 1.4199880361557007, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 0.5333, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 10.973451327433628, | |
| "grad_norm": 1.5007575750350952, | |
| "learning_rate": 7.010582010582011e-05, | |
| "loss": 0.5818, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 10.973451327433628, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5940178632736206, | |
| "eval_runtime": 1.0438, | |
| "eval_samples_per_second": 27.784, | |
| "eval_steps_per_second": 27.784, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 11.327433628318584, | |
| "grad_norm": 0.39495939016342163, | |
| "learning_rate": 6.878306878306878e-05, | |
| "loss": 0.5062, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 11.68141592920354, | |
| "grad_norm": 0.27960801124572754, | |
| "learning_rate": 6.746031746031747e-05, | |
| "loss": 0.6556, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5965593457221985, | |
| "eval_runtime": 1.0472, | |
| "eval_samples_per_second": 27.693, | |
| "eval_steps_per_second": 27.693, | |
| "step": 339 | |
| }, | |
| { | |
| "epoch": 12.035398230088495, | |
| "grad_norm": 1.464916467666626, | |
| "learning_rate": 6.613756613756614e-05, | |
| "loss": 0.4407, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 12.389380530973451, | |
| "grad_norm": 0.8306865096092224, | |
| "learning_rate": 6.481481481481482e-05, | |
| "loss": 0.5427, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 12.743362831858407, | |
| "grad_norm": 1.798261284828186, | |
| "learning_rate": 6.349206349206349e-05, | |
| "loss": 0.716, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 12.991150442477876, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5903679132461548, | |
| "eval_runtime": 1.0433, | |
| "eval_samples_per_second": 27.795, | |
| "eval_steps_per_second": 27.795, | |
| "step": 367 | |
| }, | |
| { | |
| "epoch": 13.097345132743364, | |
| "grad_norm": 0.575097918510437, | |
| "learning_rate": 6.216931216931218e-05, | |
| "loss": 0.4514, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 13.451327433628318, | |
| "grad_norm": 1.393051028251648, | |
| "learning_rate": 6.084656084656085e-05, | |
| "loss": 0.5387, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 13.805309734513274, | |
| "grad_norm": 2.5438356399536133, | |
| "learning_rate": 5.9523809523809524e-05, | |
| "loss": 0.6104, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 13.982300884955752, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5938333868980408, | |
| "eval_runtime": 1.0518, | |
| "eval_samples_per_second": 27.571, | |
| "eval_steps_per_second": 27.571, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 14.15929203539823, | |
| "grad_norm": 1.7508420944213867, | |
| "learning_rate": 5.82010582010582e-05, | |
| "loss": 0.5812, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 14.513274336283185, | |
| "grad_norm": 0.19824433326721191, | |
| "learning_rate": 5.6878306878306885e-05, | |
| "loss": 0.5678, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 14.867256637168142, | |
| "grad_norm": 0.09461919218301773, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.5046, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 14.973451327433628, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5921294689178467, | |
| "eval_runtime": 1.0718, | |
| "eval_samples_per_second": 27.057, | |
| "eval_steps_per_second": 27.057, | |
| "step": 423 | |
| }, | |
| { | |
| "epoch": 15.221238938053098, | |
| "grad_norm": 0.46888694167137146, | |
| "learning_rate": 5.423280423280423e-05, | |
| "loss": 0.5344, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 15.575221238938052, | |
| "grad_norm": 2.1239452362060547, | |
| "learning_rate": 5.291005291005291e-05, | |
| "loss": 0.4711, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 15.929203539823009, | |
| "grad_norm": 2.082629442214966, | |
| "learning_rate": 5.158730158730159e-05, | |
| "loss": 0.5871, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6027411818504333, | |
| "eval_runtime": 1.0332, | |
| "eval_samples_per_second": 28.069, | |
| "eval_steps_per_second": 28.069, | |
| "step": 452 | |
| }, | |
| { | |
| "epoch": 16.283185840707965, | |
| "grad_norm": 3.216160535812378, | |
| "learning_rate": 5.026455026455027e-05, | |
| "loss": 0.7275, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 16.63716814159292, | |
| "grad_norm": 0.5754966139793396, | |
| "learning_rate": 4.894179894179895e-05, | |
| "loss": 0.503, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 16.991150442477878, | |
| "grad_norm": 1.4574097394943237, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.5222, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 16.991150442477878, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5921464562416077, | |
| "eval_runtime": 1.0314, | |
| "eval_samples_per_second": 28.118, | |
| "eval_steps_per_second": 28.118, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 17.345132743362832, | |
| "grad_norm": 0.36778560280799866, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.5548, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 17.699115044247787, | |
| "grad_norm": 1.3566818237304688, | |
| "learning_rate": 4.4973544973544974e-05, | |
| "loss": 0.5511, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 17.98230088495575, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5947627425193787, | |
| "eval_runtime": 1.0877, | |
| "eval_samples_per_second": 26.661, | |
| "eval_steps_per_second": 26.661, | |
| "step": 508 | |
| }, | |
| { | |
| "epoch": 18.053097345132745, | |
| "grad_norm": 3.3262369632720947, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.6192, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 18.4070796460177, | |
| "grad_norm": 0.4368630051612854, | |
| "learning_rate": 4.232804232804233e-05, | |
| "loss": 0.4436, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 18.761061946902654, | |
| "grad_norm": 0.6899360418319702, | |
| "learning_rate": 4.100529100529101e-05, | |
| "loss": 0.6394, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 18.97345132743363, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5969327688217163, | |
| "eval_runtime": 1.0482, | |
| "eval_samples_per_second": 27.667, | |
| "eval_steps_per_second": 27.667, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 19.115044247787612, | |
| "grad_norm": 0.5160775780677795, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.5015, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 19.469026548672566, | |
| "grad_norm": 1.3984384536743164, | |
| "learning_rate": 3.835978835978836e-05, | |
| "loss": 0.6334, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 19.82300884955752, | |
| "grad_norm": 0.24300755560398102, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.566, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6005056500434875, | |
| "eval_runtime": 1.0644, | |
| "eval_samples_per_second": 27.247, | |
| "eval_steps_per_second": 27.247, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 20.17699115044248, | |
| "grad_norm": 1.3175139427185059, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.4341, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 20.530973451327434, | |
| "grad_norm": 1.367430567741394, | |
| "learning_rate": 3.439153439153439e-05, | |
| "loss": 0.5373, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 20.884955752212388, | |
| "grad_norm": 0.3279801905155182, | |
| "learning_rate": 3.306878306878307e-05, | |
| "loss": 0.6032, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 20.991150442477878, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5968325138092041, | |
| "eval_runtime": 1.0696, | |
| "eval_samples_per_second": 27.113, | |
| "eval_steps_per_second": 27.113, | |
| "step": 593 | |
| }, | |
| { | |
| "epoch": 21.238938053097346, | |
| "grad_norm": 3.318119764328003, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.5679, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 21.5929203539823, | |
| "grad_norm": 1.7342548370361328, | |
| "learning_rate": 3.0423280423280425e-05, | |
| "loss": 0.6877, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 21.946902654867255, | |
| "grad_norm": 0.20895572006702423, | |
| "learning_rate": 2.91005291005291e-05, | |
| "loss": 0.4824, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 21.98230088495575, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5933610200881958, | |
| "eval_runtime": 1.0375, | |
| "eval_samples_per_second": 27.951, | |
| "eval_steps_per_second": 27.951, | |
| "step": 621 | |
| }, | |
| { | |
| "epoch": 22.300884955752213, | |
| "grad_norm": 0.5702241659164429, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.5076, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 22.654867256637168, | |
| "grad_norm": 1.405441164970398, | |
| "learning_rate": 2.6455026455026456e-05, | |
| "loss": 0.4975, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 22.97345132743363, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5979170799255371, | |
| "eval_runtime": 1.0599, | |
| "eval_samples_per_second": 27.362, | |
| "eval_steps_per_second": 27.362, | |
| "step": 649 | |
| }, | |
| { | |
| "epoch": 23.008849557522122, | |
| "grad_norm": 1.4034713506698608, | |
| "learning_rate": 2.5132275132275137e-05, | |
| "loss": 0.5977, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 23.36283185840708, | |
| "grad_norm": 1.3884029388427734, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.4975, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 23.716814159292035, | |
| "grad_norm": 0.6450229287147522, | |
| "learning_rate": 2.2486772486772487e-05, | |
| "loss": 0.4976, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.603380560874939, | |
| "eval_runtime": 1.0349, | |
| "eval_samples_per_second": 28.023, | |
| "eval_steps_per_second": 28.023, | |
| "step": 678 | |
| }, | |
| { | |
| "epoch": 24.07079646017699, | |
| "grad_norm": 0.3577538728713989, | |
| "learning_rate": 2.1164021164021164e-05, | |
| "loss": 0.7906, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 24.424778761061948, | |
| "grad_norm": 0.5209062099456787, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.4288, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 24.778761061946902, | |
| "grad_norm": 1.9989219903945923, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.5355, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 24.991150442477878, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6033223867416382, | |
| "eval_runtime": 1.0457, | |
| "eval_samples_per_second": 27.733, | |
| "eval_steps_per_second": 27.733, | |
| "step": 706 | |
| }, | |
| { | |
| "epoch": 25.13274336283186, | |
| "grad_norm": 1.9387425184249878, | |
| "learning_rate": 1.7195767195767195e-05, | |
| "loss": 0.6319, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 25.486725663716815, | |
| "grad_norm": 0.4894324541091919, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.5932, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 25.84070796460177, | |
| "grad_norm": 0.26004230976104736, | |
| "learning_rate": 1.455026455026455e-05, | |
| "loss": 0.4323, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 25.98230088495575, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.601476788520813, | |
| "eval_runtime": 1.0397, | |
| "eval_samples_per_second": 27.892, | |
| "eval_steps_per_second": 27.892, | |
| "step": 734 | |
| }, | |
| { | |
| "epoch": 26.194690265486727, | |
| "grad_norm": 1.3441038131713867, | |
| "learning_rate": 1.3227513227513228e-05, | |
| "loss": 0.6053, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 26.548672566371682, | |
| "grad_norm": 1.3334237337112427, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.5, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 26.902654867256636, | |
| "grad_norm": 0.3579448461532593, | |
| "learning_rate": 1.0582010582010582e-05, | |
| "loss": 0.5579, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 26.97345132743363, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6042701601982117, | |
| "eval_runtime": 1.0854, | |
| "eval_samples_per_second": 26.718, | |
| "eval_steps_per_second": 26.718, | |
| "step": 762 | |
| }, | |
| { | |
| "epoch": 27.256637168141594, | |
| "grad_norm": 1.311233639717102, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.5794, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 27.61061946902655, | |
| "grad_norm": 1.3221209049224854, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.494, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 27.964601769911503, | |
| "grad_norm": 0.31315815448760986, | |
| "learning_rate": 6.613756613756614e-06, | |
| "loss": 0.5639, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6023350358009338, | |
| "eval_runtime": 1.0409, | |
| "eval_samples_per_second": 27.86, | |
| "eval_steps_per_second": 27.86, | |
| "step": 791 | |
| }, | |
| { | |
| "epoch": 28.31858407079646, | |
| "grad_norm": 0.7090272307395935, | |
| "learning_rate": 5.291005291005291e-06, | |
| "loss": 0.5711, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 28.672566371681416, | |
| "grad_norm": 1.377335548400879, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.5595, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 28.991150442477878, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.5995790958404541, | |
| "eval_runtime": 1.0359, | |
| "eval_samples_per_second": 27.995, | |
| "eval_steps_per_second": 27.995, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 29.02654867256637, | |
| "grad_norm": 1.294968605041504, | |
| "learning_rate": 2.6455026455026455e-06, | |
| "loss": 0.5339, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 29.38053097345133, | |
| "grad_norm": 1.297398328781128, | |
| "learning_rate": 1.3227513227513228e-06, | |
| "loss": 0.6959, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 29.734513274336283, | |
| "grad_norm": 1.3724257946014404, | |
| "learning_rate": 0.0, | |
| "loss": 0.4372, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 29.734513274336283, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.599529504776001, | |
| "eval_runtime": 1.0824, | |
| "eval_samples_per_second": 26.793, | |
| "eval_steps_per_second": 26.793, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 29.734513274336283, | |
| "step": 840, | |
| "total_flos": 3.444219560381645e+17, | |
| "train_loss": 0.5698859515644256, | |
| "train_runtime": 500.495, | |
| "train_samples_per_second": 6.773, | |
| "train_steps_per_second": 1.678 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 840, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.444219560381645e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |