| { | |
| "best_metric": 0.3162367641925812, | |
| "best_model_checkpoint": "vit-weldclassifyv4/checkpoint-1000", | |
| "epoch": 15.0, | |
| "eval_steps": 100, | |
| "global_step": 2340, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0641025641025641, | |
| "grad_norm": 1.9132781028747559, | |
| "learning_rate": 0.00019914529914529915, | |
| "loss": 1.2054, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1282051282051282, | |
| "grad_norm": 1.633124589920044, | |
| "learning_rate": 0.0001982905982905983, | |
| "loss": 1.0748, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19230769230769232, | |
| "grad_norm": 2.4126267433166504, | |
| "learning_rate": 0.00019743589743589744, | |
| "loss": 1.0973, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2564102564102564, | |
| "grad_norm": 3.881457567214966, | |
| "learning_rate": 0.00019658119658119659, | |
| "loss": 1.0609, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.32051282051282054, | |
| "grad_norm": 3.1995434761047363, | |
| "learning_rate": 0.00019572649572649573, | |
| "loss": 1.0024, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 2.410505533218384, | |
| "learning_rate": 0.00019487179487179487, | |
| "loss": 0.8658, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.44871794871794873, | |
| "grad_norm": 2.05910325050354, | |
| "learning_rate": 0.00019401709401709402, | |
| "loss": 0.9616, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "grad_norm": 4.032101154327393, | |
| "learning_rate": 0.00019316239316239316, | |
| "loss": 0.9391, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5769230769230769, | |
| "grad_norm": 2.779008150100708, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 0.9206, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6410256410256411, | |
| "grad_norm": 2.771672010421753, | |
| "learning_rate": 0.00019145299145299148, | |
| "loss": 0.8146, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6410256410256411, | |
| "eval_accuracy": 0.6834532374100719, | |
| "eval_loss": 0.7348725199699402, | |
| "eval_runtime": 2.6163, | |
| "eval_samples_per_second": 106.259, | |
| "eval_steps_per_second": 13.378, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7051282051282052, | |
| "grad_norm": 2.4781739711761475, | |
| "learning_rate": 0.0001905982905982906, | |
| "loss": 0.7542, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.8731400966644287, | |
| "learning_rate": 0.00018974358974358974, | |
| "loss": 0.7726, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 2.480015277862549, | |
| "learning_rate": 0.00018888888888888888, | |
| "loss": 0.6705, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.8974358974358975, | |
| "grad_norm": 3.7568867206573486, | |
| "learning_rate": 0.00018803418803418803, | |
| "loss": 0.6004, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 2.720820903778076, | |
| "learning_rate": 0.0001871794871794872, | |
| "loss": 0.8144, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0256410256410255, | |
| "grad_norm": 1.9295154809951782, | |
| "learning_rate": 0.00018632478632478634, | |
| "loss": 0.6479, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0897435897435896, | |
| "grad_norm": 3.0400049686431885, | |
| "learning_rate": 0.0001854700854700855, | |
| "loss": 0.522, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 3.7371773719787598, | |
| "learning_rate": 0.00018461538461538463, | |
| "loss": 0.5998, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.217948717948718, | |
| "grad_norm": 2.992065668106079, | |
| "learning_rate": 0.00018376068376068375, | |
| "loss": 0.6268, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.282051282051282, | |
| "grad_norm": 2.213074207305908, | |
| "learning_rate": 0.00018290598290598292, | |
| "loss": 0.6048, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.282051282051282, | |
| "eval_accuracy": 0.697841726618705, | |
| "eval_loss": 0.6820898056030273, | |
| "eval_runtime": 2.7768, | |
| "eval_samples_per_second": 100.114, | |
| "eval_steps_per_second": 12.604, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3461538461538463, | |
| "grad_norm": 3.2302353382110596, | |
| "learning_rate": 0.00018205128205128207, | |
| "loss": 0.6054, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.4102564102564101, | |
| "grad_norm": 3.9419608116149902, | |
| "learning_rate": 0.0001811965811965812, | |
| "loss": 0.5863, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.4743589743589745, | |
| "grad_norm": 2.5351428985595703, | |
| "learning_rate": 0.00018034188034188035, | |
| "loss": 0.4328, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 2.677548885345459, | |
| "learning_rate": 0.0001794871794871795, | |
| "loss": 0.4744, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6025641025641026, | |
| "grad_norm": 2.3627212047576904, | |
| "learning_rate": 0.00017863247863247864, | |
| "loss": 0.3523, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 6.175805568695068, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 0.5934, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.7307692307692308, | |
| "grad_norm": 2.920872211456299, | |
| "learning_rate": 0.00017692307692307693, | |
| "loss": 0.5758, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.7948717948717947, | |
| "grad_norm": 2.1116819381713867, | |
| "learning_rate": 0.00017606837606837607, | |
| "loss": 0.5086, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.858974358974359, | |
| "grad_norm": 3.4553894996643066, | |
| "learning_rate": 0.00017521367521367522, | |
| "loss": 0.4474, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 2.4064671993255615, | |
| "learning_rate": 0.00017435897435897436, | |
| "loss": 0.4796, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "eval_accuracy": 0.8129496402877698, | |
| "eval_loss": 0.48327746987342834, | |
| "eval_runtime": 2.565, | |
| "eval_samples_per_second": 108.383, | |
| "eval_steps_per_second": 13.645, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.9871794871794872, | |
| "grad_norm": 3.8495571613311768, | |
| "learning_rate": 0.0001735042735042735, | |
| "loss": 0.4925, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "grad_norm": 2.036381244659424, | |
| "learning_rate": 0.00017264957264957268, | |
| "loss": 0.3936, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.1153846153846154, | |
| "grad_norm": 3.381953001022339, | |
| "learning_rate": 0.0001717948717948718, | |
| "loss": 0.3416, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.1794871794871793, | |
| "grad_norm": 5.715399265289307, | |
| "learning_rate": 0.00017094017094017094, | |
| "loss": 0.4081, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.2435897435897436, | |
| "grad_norm": 2.237466335296631, | |
| "learning_rate": 0.00017008547008547008, | |
| "loss": 0.3313, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 5.658877372741699, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 0.3926, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.371794871794872, | |
| "grad_norm": 3.633448362350464, | |
| "learning_rate": 0.00016837606837606837, | |
| "loss": 0.4664, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.435897435897436, | |
| "grad_norm": 4.460226058959961, | |
| "learning_rate": 0.00016752136752136754, | |
| "loss": 0.4206, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.1033204793930054, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 0.2678, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "grad_norm": 6.69362735748291, | |
| "learning_rate": 0.00016581196581196583, | |
| "loss": 0.4532, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "eval_accuracy": 0.802158273381295, | |
| "eval_loss": 0.5380275249481201, | |
| "eval_runtime": 3.354, | |
| "eval_samples_per_second": 82.885, | |
| "eval_steps_per_second": 10.435, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.628205128205128, | |
| "grad_norm": 2.1671810150146484, | |
| "learning_rate": 0.00016495726495726495, | |
| "loss": 0.2966, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 4.671816825866699, | |
| "learning_rate": 0.0001641025641025641, | |
| "loss": 0.3559, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.7564102564102564, | |
| "grad_norm": 2.6795239448547363, | |
| "learning_rate": 0.00016324786324786327, | |
| "loss": 0.2111, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.8205128205128203, | |
| "grad_norm": 1.8368570804595947, | |
| "learning_rate": 0.0001623931623931624, | |
| "loss": 0.2227, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 0.7519993185997009, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 0.2185, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.948717948717949, | |
| "grad_norm": 4.014621734619141, | |
| "learning_rate": 0.0001606837606837607, | |
| "loss": 0.2038, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.0128205128205128, | |
| "grad_norm": 2.412414073944092, | |
| "learning_rate": 0.00015982905982905984, | |
| "loss": 0.1874, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 3.7715134620666504, | |
| "learning_rate": 0.00015897435897435896, | |
| "loss": 0.1546, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.141025641025641, | |
| "grad_norm": 1.5307694673538208, | |
| "learning_rate": 0.00015811965811965813, | |
| "loss": 0.1115, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.2051282051282053, | |
| "grad_norm": 2.7572405338287354, | |
| "learning_rate": 0.00015726495726495727, | |
| "loss": 0.1242, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.2051282051282053, | |
| "eval_accuracy": 0.8741007194244604, | |
| "eval_loss": 0.3899326026439667, | |
| "eval_runtime": 2.8072, | |
| "eval_samples_per_second": 99.033, | |
| "eval_steps_per_second": 12.468, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.269230769230769, | |
| "grad_norm": 4.4104390144348145, | |
| "learning_rate": 0.00015641025641025642, | |
| "loss": 0.1986, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.8930767774581909, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 0.0582, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.3974358974358974, | |
| "grad_norm": 8.353619575500488, | |
| "learning_rate": 0.0001547008547008547, | |
| "loss": 0.2485, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 0.09837932884693146, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 0.1007, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.5256410256410255, | |
| "grad_norm": 3.90265154838562, | |
| "learning_rate": 0.000152991452991453, | |
| "loss": 0.2279, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.58974358974359, | |
| "grad_norm": 6.65275764465332, | |
| "learning_rate": 0.00015213675213675214, | |
| "loss": 0.1781, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.6538461538461537, | |
| "grad_norm": 3.493739604949951, | |
| "learning_rate": 0.00015128205128205128, | |
| "loss": 0.2098, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.717948717948718, | |
| "grad_norm": 4.4887614250183105, | |
| "learning_rate": 0.00015042735042735043, | |
| "loss": 0.1899, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.782051282051282, | |
| "grad_norm": 1.8387681245803833, | |
| "learning_rate": 0.00014957264957264957, | |
| "loss": 0.1831, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 5.91892671585083, | |
| "learning_rate": 0.00014871794871794872, | |
| "loss": 0.124, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "eval_accuracy": 0.8273381294964028, | |
| "eval_loss": 0.523663341999054, | |
| "eval_runtime": 2.6143, | |
| "eval_samples_per_second": 106.339, | |
| "eval_steps_per_second": 13.388, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.91025641025641, | |
| "grad_norm": 0.5099517703056335, | |
| "learning_rate": 0.0001478632478632479, | |
| "loss": 0.1566, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.9743589743589745, | |
| "grad_norm": 0.4991530179977417, | |
| "learning_rate": 0.00014700854700854703, | |
| "loss": 0.1168, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.038461538461538, | |
| "grad_norm": 4.448193550109863, | |
| "learning_rate": 0.00014615384615384615, | |
| "loss": 0.1044, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.102564102564102, | |
| "grad_norm": 0.5183725357055664, | |
| "learning_rate": 0.0001452991452991453, | |
| "loss": 0.1078, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 0.9525802135467529, | |
| "learning_rate": 0.00014444444444444444, | |
| "loss": 0.0967, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 0.5207259058952332, | |
| "learning_rate": 0.0001435897435897436, | |
| "loss": 0.0973, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.294871794871795, | |
| "grad_norm": 1.3160842657089233, | |
| "learning_rate": 0.00014273504273504275, | |
| "loss": 0.1256, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.358974358974359, | |
| "grad_norm": 2.892195463180542, | |
| "learning_rate": 0.0001418803418803419, | |
| "loss": 0.1178, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.423076923076923, | |
| "grad_norm": 3.8142576217651367, | |
| "learning_rate": 0.00014102564102564104, | |
| "loss": 0.0594, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.487179487179487, | |
| "grad_norm": 0.11079952865839005, | |
| "learning_rate": 0.00014017094017094016, | |
| "loss": 0.1239, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.487179487179487, | |
| "eval_accuracy": 0.8848920863309353, | |
| "eval_loss": 0.4221162796020508, | |
| "eval_runtime": 2.7273, | |
| "eval_samples_per_second": 101.932, | |
| "eval_steps_per_second": 12.833, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.551282051282051, | |
| "grad_norm": 9.544878959655762, | |
| "learning_rate": 0.0001393162393162393, | |
| "loss": 0.1185, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.06085001304745674, | |
| "learning_rate": 0.00013846153846153847, | |
| "loss": 0.0499, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.67948717948718, | |
| "grad_norm": 12.285767555236816, | |
| "learning_rate": 0.00013760683760683762, | |
| "loss": 0.1623, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 4.743589743589744, | |
| "grad_norm": 1.4333381652832031, | |
| "learning_rate": 0.00013675213675213676, | |
| "loss": 0.0903, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 4.8076923076923075, | |
| "grad_norm": 0.37026920914649963, | |
| "learning_rate": 0.0001358974358974359, | |
| "loss": 0.082, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.871794871794872, | |
| "grad_norm": 7.013845443725586, | |
| "learning_rate": 0.00013504273504273505, | |
| "loss": 0.0443, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 4.935897435897436, | |
| "grad_norm": 0.3148520588874817, | |
| "learning_rate": 0.0001341880341880342, | |
| "loss": 0.1237, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.5136359930038452, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.0592, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.064102564102564, | |
| "grad_norm": 7.698183536529541, | |
| "learning_rate": 0.00013247863247863248, | |
| "loss": 0.0742, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.128205128205128, | |
| "grad_norm": 0.05358889326453209, | |
| "learning_rate": 0.00013162393162393163, | |
| "loss": 0.0785, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.128205128205128, | |
| "eval_accuracy": 0.9136690647482014, | |
| "eval_loss": 0.36830753087997437, | |
| "eval_runtime": 3.1493, | |
| "eval_samples_per_second": 88.273, | |
| "eval_steps_per_second": 11.114, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.1923076923076925, | |
| "grad_norm": 0.045300308614969254, | |
| "learning_rate": 0.00013076923076923077, | |
| "loss": 0.1987, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.256410256410256, | |
| "grad_norm": 6.118052959442139, | |
| "learning_rate": 0.00012991452991452992, | |
| "loss": 0.0708, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.32051282051282, | |
| "grad_norm": 0.36830875277519226, | |
| "learning_rate": 0.00012905982905982906, | |
| "loss": 0.0329, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 0.5043929219245911, | |
| "learning_rate": 0.00012820512820512823, | |
| "loss": 0.0546, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.448717948717949, | |
| "grad_norm": 5.8541035652160645, | |
| "learning_rate": 0.00012735042735042735, | |
| "loss": 0.0589, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.512820512820513, | |
| "grad_norm": 0.09965494275093079, | |
| "learning_rate": 0.0001264957264957265, | |
| "loss": 0.0257, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.576923076923077, | |
| "grad_norm": 0.03202090039849281, | |
| "learning_rate": 0.00012564102564102564, | |
| "loss": 0.0349, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 5.641025641025641, | |
| "grad_norm": 7.21024751663208, | |
| "learning_rate": 0.00012478632478632478, | |
| "loss": 0.081, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 5.705128205128205, | |
| "grad_norm": 0.03198171406984329, | |
| "learning_rate": 0.00012393162393162395, | |
| "loss": 0.037, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 5.769230769230769, | |
| "grad_norm": 1.1413763761520386, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 0.093, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.769230769230769, | |
| "eval_accuracy": 0.8597122302158273, | |
| "eval_loss": 0.6375630497932434, | |
| "eval_runtime": 2.5134, | |
| "eval_samples_per_second": 110.607, | |
| "eval_steps_per_second": 13.925, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.833333333333333, | |
| "grad_norm": 0.030478307977318764, | |
| "learning_rate": 0.00012222222222222224, | |
| "loss": 0.1015, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 5.897435897435898, | |
| "grad_norm": 7.971870422363281, | |
| "learning_rate": 0.00012136752136752136, | |
| "loss": 0.0421, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 5.961538461538462, | |
| "grad_norm": 0.7655214667320251, | |
| "learning_rate": 0.00012051282051282052, | |
| "loss": 0.0315, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 6.0256410256410255, | |
| "grad_norm": 0.10178599506616592, | |
| "learning_rate": 0.00011965811965811966, | |
| "loss": 0.008, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 6.089743589743589, | |
| "grad_norm": 0.024569841101765633, | |
| "learning_rate": 0.0001188034188034188, | |
| "loss": 0.0054, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 0.36783352494239807, | |
| "learning_rate": 0.00011794871794871796, | |
| "loss": 0.0752, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.217948717948718, | |
| "grad_norm": 0.04280726611614227, | |
| "learning_rate": 0.00011709401709401711, | |
| "loss": 0.0286, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 6.282051282051282, | |
| "grad_norm": 14.323381423950195, | |
| "learning_rate": 0.00011623931623931625, | |
| "loss": 0.0559, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 6.346153846153846, | |
| "grad_norm": 0.025405921041965485, | |
| "learning_rate": 0.00011538461538461538, | |
| "loss": 0.0238, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "grad_norm": 0.02457793429493904, | |
| "learning_rate": 0.00011452991452991453, | |
| "loss": 0.0056, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "eval_accuracy": 0.9244604316546763, | |
| "eval_loss": 0.3162367641925812, | |
| "eval_runtime": 3.2283, | |
| "eval_samples_per_second": 86.114, | |
| "eval_steps_per_second": 10.842, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.4743589743589745, | |
| "grad_norm": 0.020618008449673653, | |
| "learning_rate": 0.00011367521367521367, | |
| "loss": 0.0113, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 6.538461538461538, | |
| "grad_norm": 0.021217485889792442, | |
| "learning_rate": 0.00011282051282051283, | |
| "loss": 0.0115, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 6.602564102564102, | |
| "grad_norm": 0.028808822855353355, | |
| "learning_rate": 0.00011196581196581197, | |
| "loss": 0.022, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 3.5672314167022705, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 0.0706, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 6.730769230769231, | |
| "grad_norm": 0.25913771986961365, | |
| "learning_rate": 0.00011025641025641027, | |
| "loss": 0.021, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 6.794871794871795, | |
| "grad_norm": 12.088153839111328, | |
| "learning_rate": 0.00010940170940170942, | |
| "loss": 0.0914, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 6.858974358974359, | |
| "grad_norm": 0.7027952671051025, | |
| "learning_rate": 0.00010854700854700855, | |
| "loss": 0.0766, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 6.911967754364014, | |
| "learning_rate": 0.0001076923076923077, | |
| "loss": 0.0881, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 6.987179487179487, | |
| "grad_norm": 0.03289846330881119, | |
| "learning_rate": 0.00010683760683760684, | |
| "loss": 0.0125, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 7.051282051282051, | |
| "grad_norm": 0.025492513552308083, | |
| "learning_rate": 0.000105982905982906, | |
| "loss": 0.0472, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.051282051282051, | |
| "eval_accuracy": 0.8884892086330936, | |
| "eval_loss": 0.5225415825843811, | |
| "eval_runtime": 3.7495, | |
| "eval_samples_per_second": 74.144, | |
| "eval_steps_per_second": 9.335, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.115384615384615, | |
| "grad_norm": 0.022507918998599052, | |
| "learning_rate": 0.00010512820512820514, | |
| "loss": 0.0386, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 7.17948717948718, | |
| "grad_norm": 0.020967524498701096, | |
| "learning_rate": 0.00010427350427350428, | |
| "loss": 0.0289, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 7.243589743589744, | |
| "grad_norm": 0.5489076972007751, | |
| "learning_rate": 0.00010341880341880343, | |
| "loss": 0.0041, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 7.3076923076923075, | |
| "grad_norm": 0.12584710121154785, | |
| "learning_rate": 0.00010256410256410256, | |
| "loss": 0.0697, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 7.371794871794872, | |
| "grad_norm": 0.022198162972927094, | |
| "learning_rate": 0.0001017094017094017, | |
| "loss": 0.0039, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 7.435897435897436, | |
| "grad_norm": 11.968843460083008, | |
| "learning_rate": 0.00010085470085470086, | |
| "loss": 0.0855, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.779564380645752, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0585, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 7.564102564102564, | |
| "grad_norm": 0.23576153814792633, | |
| "learning_rate": 9.914529914529915e-05, | |
| "loss": 0.0559, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 7.628205128205128, | |
| "grad_norm": 0.020965000614523888, | |
| "learning_rate": 9.829059829059829e-05, | |
| "loss": 0.0785, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 6.914454936981201, | |
| "learning_rate": 9.743589743589744e-05, | |
| "loss": 0.0234, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "eval_accuracy": 0.8597122302158273, | |
| "eval_loss": 0.6095559000968933, | |
| "eval_runtime": 2.5288, | |
| "eval_samples_per_second": 109.933, | |
| "eval_steps_per_second": 13.84, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.756410256410256, | |
| "grad_norm": 1.3177701234817505, | |
| "learning_rate": 9.658119658119658e-05, | |
| "loss": 0.0768, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 7.82051282051282, | |
| "grad_norm": 4.212278842926025, | |
| "learning_rate": 9.572649572649574e-05, | |
| "loss": 0.0914, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 7.884615384615385, | |
| "grad_norm": 0.02418905310332775, | |
| "learning_rate": 9.487179487179487e-05, | |
| "loss": 0.0348, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 7.948717948717949, | |
| "grad_norm": 3.008629322052002, | |
| "learning_rate": 9.401709401709401e-05, | |
| "loss": 0.0624, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 8.012820512820513, | |
| "grad_norm": 0.052931949496269226, | |
| "learning_rate": 9.316239316239317e-05, | |
| "loss": 0.0076, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.076923076923077, | |
| "grad_norm": 7.994688034057617, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.0326, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 8.14102564102564, | |
| "grad_norm": 0.026721293106675148, | |
| "learning_rate": 9.145299145299146e-05, | |
| "loss": 0.0324, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 8.205128205128204, | |
| "grad_norm": 0.049855004996061325, | |
| "learning_rate": 9.05982905982906e-05, | |
| "loss": 0.0057, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 8.26923076923077, | |
| "grad_norm": 0.014473488554358482, | |
| "learning_rate": 8.974358974358975e-05, | |
| "loss": 0.0237, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 0.03150290250778198, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.0354, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "eval_accuracy": 0.8776978417266187, | |
| "eval_loss": 0.5520122647285461, | |
| "eval_runtime": 3.3151, | |
| "eval_samples_per_second": 83.858, | |
| "eval_steps_per_second": 10.558, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.397435897435898, | |
| "grad_norm": 0.013332781381905079, | |
| "learning_rate": 8.803418803418804e-05, | |
| "loss": 0.0049, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 0.013341937214136124, | |
| "learning_rate": 8.717948717948718e-05, | |
| "loss": 0.0026, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 8.525641025641026, | |
| "grad_norm": 0.012689488008618355, | |
| "learning_rate": 8.632478632478634e-05, | |
| "loss": 0.0051, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 8.58974358974359, | |
| "grad_norm": 0.014231017790734768, | |
| "learning_rate": 8.547008547008547e-05, | |
| "loss": 0.0027, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 8.653846153846153, | |
| "grad_norm": 0.052165694534778595, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.0036, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 8.717948717948717, | |
| "grad_norm": 0.11598876118659973, | |
| "learning_rate": 8.376068376068377e-05, | |
| "loss": 0.0281, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 8.782051282051283, | |
| "grad_norm": 0.01339508593082428, | |
| "learning_rate": 8.290598290598292e-05, | |
| "loss": 0.0026, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 8.846153846153847, | |
| "grad_norm": 0.011919701471924782, | |
| "learning_rate": 8.205128205128205e-05, | |
| "loss": 0.0025, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 8.91025641025641, | |
| "grad_norm": 3.217728614807129, | |
| "learning_rate": 8.11965811965812e-05, | |
| "loss": 0.0379, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 8.974358974358974, | |
| "grad_norm": 0.01331857219338417, | |
| "learning_rate": 8.034188034188035e-05, | |
| "loss": 0.026, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.974358974358974, | |
| "eval_accuracy": 0.8992805755395683, | |
| "eval_loss": 0.49377354979515076, | |
| "eval_runtime": 2.8453, | |
| "eval_samples_per_second": 97.704, | |
| "eval_steps_per_second": 12.301, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.038461538461538, | |
| "grad_norm": 0.1426580250263214, | |
| "learning_rate": 7.948717948717948e-05, | |
| "loss": 0.0072, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 9.102564102564102, | |
| "grad_norm": 0.021561838686466217, | |
| "learning_rate": 7.863247863247864e-05, | |
| "loss": 0.0025, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 9.166666666666666, | |
| "grad_norm": 0.010494213551282883, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 0.0025, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 0.015301249921321869, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.0051, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 9.294871794871796, | |
| "grad_norm": 0.013643044047057629, | |
| "learning_rate": 7.606837606837607e-05, | |
| "loss": 0.0136, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 9.35897435897436, | |
| "grad_norm": 0.02054368518292904, | |
| "learning_rate": 7.521367521367521e-05, | |
| "loss": 0.0025, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 9.423076923076923, | |
| "grad_norm": 0.011097296141088009, | |
| "learning_rate": 7.435897435897436e-05, | |
| "loss": 0.0025, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 9.487179487179487, | |
| "grad_norm": 0.5705698132514954, | |
| "learning_rate": 7.350427350427352e-05, | |
| "loss": 0.0048, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 9.551282051282051, | |
| "grad_norm": 0.009772556833922863, | |
| "learning_rate": 7.264957264957265e-05, | |
| "loss": 0.002, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 9.615384615384615, | |
| "grad_norm": 0.011127009056508541, | |
| "learning_rate": 7.17948717948718e-05, | |
| "loss": 0.002, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.615384615384615, | |
| "eval_accuracy": 0.9172661870503597, | |
| "eval_loss": 0.43497270345687866, | |
| "eval_runtime": 2.5545, | |
| "eval_samples_per_second": 108.826, | |
| "eval_steps_per_second": 13.701, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.679487179487179, | |
| "grad_norm": 0.013090673834085464, | |
| "learning_rate": 7.094017094017095e-05, | |
| "loss": 0.002, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 9.743589743589745, | |
| "grad_norm": 0.00843009538948536, | |
| "learning_rate": 7.008547008547008e-05, | |
| "loss": 0.0021, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 9.807692307692308, | |
| "grad_norm": 0.00981289241462946, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.0018, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 9.871794871794872, | |
| "grad_norm": 0.01045091450214386, | |
| "learning_rate": 6.837606837606838e-05, | |
| "loss": 0.0047, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 9.935897435897436, | |
| "grad_norm": 0.07456047832965851, | |
| "learning_rate": 6.752136752136753e-05, | |
| "loss": 0.0021, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.010238745249807835, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.0268, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 10.064102564102564, | |
| "grad_norm": 0.009372313506901264, | |
| "learning_rate": 6.581196581196581e-05, | |
| "loss": 0.0018, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 10.128205128205128, | |
| "grad_norm": 0.009544081054627895, | |
| "learning_rate": 6.495726495726496e-05, | |
| "loss": 0.0017, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 10.192307692307692, | |
| "grad_norm": 0.014180944301187992, | |
| "learning_rate": 6.410256410256412e-05, | |
| "loss": 0.0055, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 10.256410256410255, | |
| "grad_norm": 0.19402286410331726, | |
| "learning_rate": 6.324786324786325e-05, | |
| "loss": 0.0021, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.256410256410255, | |
| "eval_accuracy": 0.9172661870503597, | |
| "eval_loss": 0.4223933219909668, | |
| "eval_runtime": 3.538, | |
| "eval_samples_per_second": 78.575, | |
| "eval_steps_per_second": 9.893, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.320512820512821, | |
| "grad_norm": 0.008879870176315308, | |
| "learning_rate": 6.239316239316239e-05, | |
| "loss": 0.0018, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 10.384615384615385, | |
| "grad_norm": 0.008063827641308308, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.002, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 10.448717948717949, | |
| "grad_norm": 0.008835590444505215, | |
| "learning_rate": 6.068376068376068e-05, | |
| "loss": 0.0018, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 10.512820512820513, | |
| "grad_norm": 0.008632234297692776, | |
| "learning_rate": 5.982905982905983e-05, | |
| "loss": 0.0017, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 10.576923076923077, | |
| "grad_norm": 0.00828844029456377, | |
| "learning_rate": 5.897435897435898e-05, | |
| "loss": 0.0016, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 10.64102564102564, | |
| "grad_norm": 0.0323554202914238, | |
| "learning_rate": 5.8119658119658126e-05, | |
| "loss": 0.0016, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 10.705128205128204, | |
| "grad_norm": 0.008372778072953224, | |
| "learning_rate": 5.726495726495726e-05, | |
| "loss": 0.0016, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 10.76923076923077, | |
| "grad_norm": 0.007286165375262499, | |
| "learning_rate": 5.6410256410256414e-05, | |
| "loss": 0.0016, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 10.833333333333334, | |
| "grad_norm": 0.012557004578411579, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.0017, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 10.897435897435898, | |
| "grad_norm": 0.007122470065951347, | |
| "learning_rate": 5.470085470085471e-05, | |
| "loss": 0.0016, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 10.897435897435898, | |
| "eval_accuracy": 0.9280575539568345, | |
| "eval_loss": 0.38381046056747437, | |
| "eval_runtime": 2.8652, | |
| "eval_samples_per_second": 97.028, | |
| "eval_steps_per_second": 12.216, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 10.961538461538462, | |
| "grad_norm": 0.013124003075063229, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.0015, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 11.025641025641026, | |
| "grad_norm": 0.007307114545255899, | |
| "learning_rate": 5.2991452991453e-05, | |
| "loss": 0.0015, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 11.08974358974359, | |
| "grad_norm": 0.007222812157124281, | |
| "learning_rate": 5.213675213675214e-05, | |
| "loss": 0.0014, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 11.153846153846153, | |
| "grad_norm": 0.007087068632245064, | |
| "learning_rate": 5.128205128205128e-05, | |
| "loss": 0.0015, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 11.217948717948717, | |
| "grad_norm": 0.010747412219643593, | |
| "learning_rate": 5.042735042735043e-05, | |
| "loss": 0.0016, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 11.282051282051283, | |
| "grad_norm": 0.007549288682639599, | |
| "learning_rate": 4.9572649572649575e-05, | |
| "loss": 0.0014, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 11.346153846153847, | |
| "grad_norm": 0.007861124351620674, | |
| "learning_rate": 4.871794871794872e-05, | |
| "loss": 0.0015, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 11.41025641025641, | |
| "grad_norm": 0.008711726404726505, | |
| "learning_rate": 4.786324786324787e-05, | |
| "loss": 0.0014, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 11.474358974358974, | |
| "grad_norm": 0.006650915369391441, | |
| "learning_rate": 4.700854700854701e-05, | |
| "loss": 0.0014, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 11.538461538461538, | |
| "grad_norm": 0.009336930699646473, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.0014, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 11.538461538461538, | |
| "eval_accuracy": 0.9280575539568345, | |
| "eval_loss": 0.3943016529083252, | |
| "eval_runtime": 2.5492, | |
| "eval_samples_per_second": 109.052, | |
| "eval_steps_per_second": 13.73, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 11.602564102564102, | |
| "grad_norm": 0.0071232253685593605, | |
| "learning_rate": 4.52991452991453e-05, | |
| "loss": 0.0015, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 11.666666666666666, | |
| "grad_norm": 0.0070044491440057755, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0014, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 11.73076923076923, | |
| "grad_norm": 0.00735941668972373, | |
| "learning_rate": 4.358974358974359e-05, | |
| "loss": 0.0014, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 11.794871794871796, | |
| "grad_norm": 0.006876462604850531, | |
| "learning_rate": 4.2735042735042735e-05, | |
| "loss": 0.0013, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 11.85897435897436, | |
| "grad_norm": 0.008532642386853695, | |
| "learning_rate": 4.1880341880341886e-05, | |
| "loss": 0.0014, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 11.923076923076923, | |
| "grad_norm": 0.007169618736952543, | |
| "learning_rate": 4.1025641025641023e-05, | |
| "loss": 0.0013, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 11.987179487179487, | |
| "grad_norm": 0.006206741090863943, | |
| "learning_rate": 4.0170940170940174e-05, | |
| "loss": 0.0012, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 12.051282051282051, | |
| "grad_norm": 0.006877180654555559, | |
| "learning_rate": 3.931623931623932e-05, | |
| "loss": 0.0013, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 12.115384615384615, | |
| "grad_norm": 0.006645900197327137, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.0013, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 12.179487179487179, | |
| "grad_norm": 0.007376631256192923, | |
| "learning_rate": 3.760683760683761e-05, | |
| "loss": 0.0013, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 12.179487179487179, | |
| "eval_accuracy": 0.9280575539568345, | |
| "eval_loss": 0.40119558572769165, | |
| "eval_runtime": 2.73, | |
| "eval_samples_per_second": 101.833, | |
| "eval_steps_per_second": 12.821, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 12.243589743589743, | |
| "grad_norm": 0.007013231050223112, | |
| "learning_rate": 3.675213675213676e-05, | |
| "loss": 0.0013, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 12.307692307692308, | |
| "grad_norm": 0.006970913149416447, | |
| "learning_rate": 3.58974358974359e-05, | |
| "loss": 0.0013, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 12.371794871794872, | |
| "grad_norm": 0.006338655948638916, | |
| "learning_rate": 3.504273504273504e-05, | |
| "loss": 0.0013, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 12.435897435897436, | |
| "grad_norm": 0.007881653495132923, | |
| "learning_rate": 3.418803418803419e-05, | |
| "loss": 0.0013, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 0.005947918631136417, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0012, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 12.564102564102564, | |
| "grad_norm": 0.005899305455386639, | |
| "learning_rate": 3.247863247863248e-05, | |
| "loss": 0.0012, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 12.628205128205128, | |
| "grad_norm": 0.0061206454411149025, | |
| "learning_rate": 3.162393162393162e-05, | |
| "loss": 0.0013, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 12.692307692307692, | |
| "grad_norm": 0.00656491843983531, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.0012, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 12.756410256410255, | |
| "grad_norm": 0.006318471394479275, | |
| "learning_rate": 2.9914529914529915e-05, | |
| "loss": 0.0012, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 12.820512820512821, | |
| "grad_norm": 0.00670122355222702, | |
| "learning_rate": 2.9059829059829063e-05, | |
| "loss": 0.0012, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.820512820512821, | |
| "eval_accuracy": 0.9280575539568345, | |
| "eval_loss": 0.4066712558269501, | |
| "eval_runtime": 3.6428, | |
| "eval_samples_per_second": 76.315, | |
| "eval_steps_per_second": 9.608, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.884615384615385, | |
| "grad_norm": 0.005691882688552141, | |
| "learning_rate": 2.8205128205128207e-05, | |
| "loss": 0.0012, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 12.948717948717949, | |
| "grad_norm": 0.005753946490585804, | |
| "learning_rate": 2.7350427350427355e-05, | |
| "loss": 0.0012, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 13.012820512820513, | |
| "grad_norm": 0.006568002514541149, | |
| "learning_rate": 2.64957264957265e-05, | |
| "loss": 0.0013, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 13.076923076923077, | |
| "grad_norm": 0.005731898359954357, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 0.0012, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 13.14102564102564, | |
| "grad_norm": 0.005868157371878624, | |
| "learning_rate": 2.4786324786324787e-05, | |
| "loss": 0.0011, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 13.205128205128204, | |
| "grad_norm": 0.006337730213999748, | |
| "learning_rate": 2.3931623931623935e-05, | |
| "loss": 0.0012, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 13.26923076923077, | |
| "grad_norm": 0.006973997224122286, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.0011, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 0.00554188247770071, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0011, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 13.397435897435898, | |
| "grad_norm": 0.006806936115026474, | |
| "learning_rate": 2.1367521367521368e-05, | |
| "loss": 0.0011, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 13.461538461538462, | |
| "grad_norm": 0.005874712951481342, | |
| "learning_rate": 2.0512820512820512e-05, | |
| "loss": 0.0011, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 13.461538461538462, | |
| "eval_accuracy": 0.9280575539568345, | |
| "eval_loss": 0.41012829542160034, | |
| "eval_runtime": 2.5393, | |
| "eval_samples_per_second": 109.477, | |
| "eval_steps_per_second": 13.783, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 13.525641025641026, | |
| "grad_norm": 0.006118403282016516, | |
| "learning_rate": 1.965811965811966e-05, | |
| "loss": 0.0012, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 13.58974358974359, | |
| "grad_norm": 0.005724759306758642, | |
| "learning_rate": 1.8803418803418804e-05, | |
| "loss": 0.0012, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 13.653846153846153, | |
| "grad_norm": 0.005641784518957138, | |
| "learning_rate": 1.794871794871795e-05, | |
| "loss": 0.0011, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 13.717948717948717, | |
| "grad_norm": 0.006412914022803307, | |
| "learning_rate": 1.7094017094017095e-05, | |
| "loss": 0.0012, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 13.782051282051283, | |
| "grad_norm": 0.0061592236161231995, | |
| "learning_rate": 1.623931623931624e-05, | |
| "loss": 0.0012, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 13.846153846153847, | |
| "grad_norm": 0.006390335038304329, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.0012, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 13.91025641025641, | |
| "grad_norm": 0.006186114624142647, | |
| "learning_rate": 1.4529914529914531e-05, | |
| "loss": 0.0012, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 13.974358974358974, | |
| "grad_norm": 0.006987506989389658, | |
| "learning_rate": 1.3675213675213677e-05, | |
| "loss": 0.0013, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 14.038461538461538, | |
| "grad_norm": 0.0060087586753070354, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 0.0011, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 14.102564102564102, | |
| "grad_norm": 0.005536227021366358, | |
| "learning_rate": 1.1965811965811967e-05, | |
| "loss": 0.0011, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 14.102564102564102, | |
| "eval_accuracy": 0.9280575539568345, | |
| "eval_loss": 0.41235998272895813, | |
| "eval_runtime": 3.6965, | |
| "eval_samples_per_second": 75.206, | |
| "eval_steps_per_second": 9.468, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 14.166666666666666, | |
| "grad_norm": 0.00747127179056406, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0011, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 14.23076923076923, | |
| "grad_norm": 0.006075258832424879, | |
| "learning_rate": 1.0256410256410256e-05, | |
| "loss": 0.0012, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 14.294871794871796, | |
| "grad_norm": 0.005355818197131157, | |
| "learning_rate": 9.401709401709402e-06, | |
| "loss": 0.0011, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 14.35897435897436, | |
| "grad_norm": 0.006171481683850288, | |
| "learning_rate": 8.547008547008548e-06, | |
| "loss": 0.0012, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 14.423076923076923, | |
| "grad_norm": 0.006203506141901016, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.0011, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 14.487179487179487, | |
| "grad_norm": 0.0053332289680838585, | |
| "learning_rate": 6.837606837606839e-06, | |
| "loss": 0.0011, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 14.551282051282051, | |
| "grad_norm": 0.006036951672285795, | |
| "learning_rate": 5.982905982905984e-06, | |
| "loss": 0.0012, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 14.615384615384615, | |
| "grad_norm": 0.006114748306572437, | |
| "learning_rate": 5.128205128205128e-06, | |
| "loss": 0.0012, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 14.679487179487179, | |
| "grad_norm": 0.0059860167093575, | |
| "learning_rate": 4.273504273504274e-06, | |
| "loss": 0.0011, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 14.743589743589745, | |
| "grad_norm": 0.005834794137626886, | |
| "learning_rate": 3.4188034188034193e-06, | |
| "loss": 0.0012, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 14.743589743589745, | |
| "eval_accuracy": 0.9280575539568345, | |
| "eval_loss": 0.4135644733905792, | |
| "eval_runtime": 2.5193, | |
| "eval_samples_per_second": 110.348, | |
| "eval_steps_per_second": 13.893, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 14.807692307692308, | |
| "grad_norm": 0.005964505951851606, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 0.0012, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 14.871794871794872, | |
| "grad_norm": 0.005720064975321293, | |
| "learning_rate": 1.7094017094017097e-06, | |
| "loss": 0.0011, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 14.935897435897436, | |
| "grad_norm": 0.005913382861763239, | |
| "learning_rate": 8.547008547008548e-07, | |
| "loss": 0.0011, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.005278407130390406, | |
| "learning_rate": 0.0, | |
| "loss": 0.0011, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 2340, | |
| "total_flos": 2.900189697360077e+18, | |
| "train_loss": 0.14551133991808146, | |
| "train_runtime": 927.2479, | |
| "train_samples_per_second": 40.361, | |
| "train_steps_per_second": 2.524 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.900189697360077e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |