| { | |
| "best_metric": 0.13913680613040924, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned_v2024-7-24-frost/checkpoint-1000", | |
| "epoch": 17.0, | |
| "eval_steps": 100, | |
| "global_step": 1088, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15625, | |
| "grad_norm": 0.388886958360672, | |
| "learning_rate": 1.834862385321101e-05, | |
| "loss": 0.6876, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 0.43717288970947266, | |
| "learning_rate": 3.669724770642202e-05, | |
| "loss": 0.6556, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 0.41058769822120667, | |
| "learning_rate": 5.504587155963303e-05, | |
| "loss": 0.5865, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.37728703022003174, | |
| "learning_rate": 7.339449541284404e-05, | |
| "loss": 0.5101, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "grad_norm": 0.3225114643573761, | |
| "learning_rate": 9.174311926605506e-05, | |
| "loss": 0.4621, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 0.27789732813835144, | |
| "learning_rate": 0.00011009174311926606, | |
| "loss": 0.4339, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.09375, | |
| "grad_norm": 0.3035929501056671, | |
| "learning_rate": 0.00012844036697247707, | |
| "loss": 0.3999, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.29363128542900085, | |
| "learning_rate": 0.0001467889908256881, | |
| "loss": 0.3808, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.40625, | |
| "grad_norm": 0.2852456867694855, | |
| "learning_rate": 0.0001651376146788991, | |
| "loss": 0.356, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 0.2548869550228119, | |
| "learning_rate": 0.00018348623853211012, | |
| "loss": 0.3281, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "eval_accuracy": 0.9008849557522124, | |
| "eval_f1": 0.6956521739130436, | |
| "eval_loss": 0.31767982244491577, | |
| "eval_precision": 0.8767123287671232, | |
| "eval_recall": 0.5765765765765766, | |
| "eval_runtime": 0.9054, | |
| "eval_samples_per_second": 124.801, | |
| "eval_steps_per_second": 16.566, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.71875, | |
| "grad_norm": 0.34990283846855164, | |
| "learning_rate": 0.00019979570990806946, | |
| "loss": 0.3169, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.4318304657936096, | |
| "learning_rate": 0.00019775280898876404, | |
| "loss": 0.307, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.03125, | |
| "grad_norm": 0.4618605077266693, | |
| "learning_rate": 0.00019570990806945865, | |
| "loss": 0.2988, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.1875, | |
| "grad_norm": 0.35013139247894287, | |
| "learning_rate": 0.00019366700715015323, | |
| "loss": 0.2962, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.34375, | |
| "grad_norm": 0.2871937155723572, | |
| "learning_rate": 0.00019162410623084782, | |
| "loss": 0.2718, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.5972371101379395, | |
| "learning_rate": 0.0001895812053115424, | |
| "loss": 0.2724, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.65625, | |
| "grad_norm": 0.49075496196746826, | |
| "learning_rate": 0.00018753830439223698, | |
| "loss": 0.2681, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.8125, | |
| "grad_norm": 0.23726806044578552, | |
| "learning_rate": 0.00018549540347293156, | |
| "loss": 0.2527, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.96875, | |
| "grad_norm": 0.4829351305961609, | |
| "learning_rate": 0.00018345250255362615, | |
| "loss": 0.2484, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 0.3746371567249298, | |
| "learning_rate": 0.00018140960163432076, | |
| "loss": 0.2532, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "eval_accuracy": 0.9176991150442478, | |
| "eval_f1": 0.7832167832167832, | |
| "eval_loss": 0.24235816299915314, | |
| "eval_precision": 0.8115942028985508, | |
| "eval_recall": 0.7567567567567568, | |
| "eval_runtime": 0.8872, | |
| "eval_samples_per_second": 127.371, | |
| "eval_steps_per_second": 16.908, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.28125, | |
| "grad_norm": 0.5272416472434998, | |
| "learning_rate": 0.00017936670071501534, | |
| "loss": 0.2438, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.4375, | |
| "grad_norm": 0.2519098222255707, | |
| "learning_rate": 0.00017732379979570992, | |
| "loss": 0.2241, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.59375, | |
| "grad_norm": 0.22354206442832947, | |
| "learning_rate": 0.0001752808988764045, | |
| "loss": 0.223, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.32517552375793457, | |
| "learning_rate": 0.0001732379979570991, | |
| "loss": 0.2414, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.90625, | |
| "grad_norm": 0.32100722193717957, | |
| "learning_rate": 0.00017119509703779367, | |
| "loss": 0.2188, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0625, | |
| "grad_norm": 0.5668182969093323, | |
| "learning_rate": 0.00016915219611848828, | |
| "loss": 0.1933, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.21875, | |
| "grad_norm": 0.42494627833366394, | |
| "learning_rate": 0.00016710929519918286, | |
| "loss": 0.2244, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 0.4030027687549591, | |
| "learning_rate": 0.00016506639427987742, | |
| "loss": 0.1939, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.53125, | |
| "grad_norm": 0.3662806749343872, | |
| "learning_rate": 0.000163023493360572, | |
| "loss": 0.2117, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.6875, | |
| "grad_norm": 0.23639629781246185, | |
| "learning_rate": 0.0001609805924412666, | |
| "loss": 0.1762, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.6875, | |
| "eval_accuracy": 0.9407079646017699, | |
| "eval_f1": 0.8452655889145496, | |
| "eval_loss": 0.18491290509700775, | |
| "eval_precision": 0.8672985781990521, | |
| "eval_recall": 0.8243243243243243, | |
| "eval_runtime": 1.2006, | |
| "eval_samples_per_second": 94.116, | |
| "eval_steps_per_second": 12.493, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.84375, | |
| "grad_norm": 0.26584741473197937, | |
| "learning_rate": 0.0001589376915219612, | |
| "loss": 0.1844, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.44927456974983215, | |
| "learning_rate": 0.00015689479060265578, | |
| "loss": 0.2022, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.15625, | |
| "grad_norm": 0.24379728734493256, | |
| "learning_rate": 0.00015485188968335036, | |
| "loss": 0.185, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.3125, | |
| "grad_norm": 0.34985587000846863, | |
| "learning_rate": 0.00015280898876404494, | |
| "loss": 0.1567, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.46875, | |
| "grad_norm": 0.325273334980011, | |
| "learning_rate": 0.00015076608784473953, | |
| "loss": 0.1658, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.625, | |
| "grad_norm": 0.26209816336631775, | |
| "learning_rate": 0.00014872318692543413, | |
| "loss": 0.1739, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.78125, | |
| "grad_norm": 0.32775962352752686, | |
| "learning_rate": 0.00014668028600612872, | |
| "loss": 0.1755, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.9375, | |
| "grad_norm": 0.2825620174407959, | |
| "learning_rate": 0.0001446373850868233, | |
| "loss": 0.1735, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.09375, | |
| "grad_norm": 0.18162801861763, | |
| "learning_rate": 0.00014259448416751788, | |
| "loss": 0.1694, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.3598109483718872, | |
| "learning_rate": 0.00014055158324821247, | |
| "loss": 0.1525, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_accuracy": 0.9256637168141593, | |
| "eval_f1": 0.8055555555555555, | |
| "eval_loss": 0.18338988721370697, | |
| "eval_precision": 0.8285714285714286, | |
| "eval_recall": 0.7837837837837838, | |
| "eval_runtime": 0.8607, | |
| "eval_samples_per_second": 131.29, | |
| "eval_steps_per_second": 17.428, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.40625, | |
| "grad_norm": 0.592587947845459, | |
| "learning_rate": 0.00013850868232890705, | |
| "loss": 0.1518, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.5625, | |
| "grad_norm": 0.1838444322347641, | |
| "learning_rate": 0.00013646578140960163, | |
| "loss": 0.161, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.71875, | |
| "grad_norm": 0.326659232378006, | |
| "learning_rate": 0.00013442288049029624, | |
| "loss": 0.1603, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.875, | |
| "grad_norm": 0.45708832144737244, | |
| "learning_rate": 0.00013237997957099082, | |
| "loss": 0.1609, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 7.03125, | |
| "grad_norm": 0.17103692889213562, | |
| "learning_rate": 0.0001303370786516854, | |
| "loss": 0.1263, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.1875, | |
| "grad_norm": 0.40236711502075195, | |
| "learning_rate": 0.00012829417773238, | |
| "loss": 0.1386, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.34375, | |
| "grad_norm": 0.5935817360877991, | |
| "learning_rate": 0.00012625127681307457, | |
| "loss": 0.1419, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.42183375358581543, | |
| "learning_rate": 0.00012420837589376916, | |
| "loss": 0.1296, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.65625, | |
| "grad_norm": 0.28589242696762085, | |
| "learning_rate": 0.00012216547497446374, | |
| "loss": 0.1298, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.8125, | |
| "grad_norm": 0.14183469116687775, | |
| "learning_rate": 0.00012012257405515832, | |
| "loss": 0.1447, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.8125, | |
| "eval_accuracy": 0.9415929203539823, | |
| "eval_f1": 0.8472222222222222, | |
| "eval_loss": 0.16118969023227692, | |
| "eval_precision": 0.8714285714285714, | |
| "eval_recall": 0.8243243243243243, | |
| "eval_runtime": 0.9028, | |
| "eval_samples_per_second": 125.167, | |
| "eval_steps_per_second": 16.615, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.96875, | |
| "grad_norm": 0.4977594017982483, | |
| "learning_rate": 0.0001180796731358529, | |
| "loss": 0.1385, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.125, | |
| "grad_norm": 0.6004766225814819, | |
| "learning_rate": 0.0001160367722165475, | |
| "loss": 0.1532, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.28125, | |
| "grad_norm": 0.20785477757453918, | |
| "learning_rate": 0.00011399387129724208, | |
| "loss": 0.1329, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.4375, | |
| "grad_norm": 0.30308064818382263, | |
| "learning_rate": 0.00011195097037793667, | |
| "loss": 0.1351, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.59375, | |
| "grad_norm": 0.40658825635910034, | |
| "learning_rate": 0.00010990806945863126, | |
| "loss": 0.1289, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 0.15297789871692657, | |
| "learning_rate": 0.00010786516853932584, | |
| "loss": 0.1353, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.90625, | |
| "grad_norm": 0.2919471859931946, | |
| "learning_rate": 0.00010582226762002043, | |
| "loss": 0.1215, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 9.0625, | |
| "grad_norm": 0.23703481256961823, | |
| "learning_rate": 0.00010377936670071502, | |
| "loss": 0.1107, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 9.21875, | |
| "grad_norm": 0.15953165292739868, | |
| "learning_rate": 0.0001017364657814096, | |
| "loss": 0.1349, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "grad_norm": 0.2888895869255066, | |
| "learning_rate": 9.969356486210419e-05, | |
| "loss": 0.1114, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "eval_accuracy": 0.9433628318584071, | |
| "eval_f1": 0.8545454545454546, | |
| "eval_loss": 0.15215742588043213, | |
| "eval_precision": 0.8623853211009175, | |
| "eval_recall": 0.8468468468468469, | |
| "eval_runtime": 0.8895, | |
| "eval_samples_per_second": 127.032, | |
| "eval_steps_per_second": 16.863, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.53125, | |
| "grad_norm": 0.23866873979568481, | |
| "learning_rate": 9.765066394279879e-05, | |
| "loss": 0.1097, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.6875, | |
| "grad_norm": 0.2734706997871399, | |
| "learning_rate": 9.560776302349337e-05, | |
| "loss": 0.1096, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.84375, | |
| "grad_norm": 0.5498181581497192, | |
| "learning_rate": 9.356486210418795e-05, | |
| "loss": 0.1039, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.17813219130039215, | |
| "learning_rate": 9.152196118488255e-05, | |
| "loss": 0.1002, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 10.15625, | |
| "grad_norm": 0.34687340259552, | |
| "learning_rate": 8.947906026557712e-05, | |
| "loss": 0.1164, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 10.3125, | |
| "grad_norm": 0.18163448572158813, | |
| "learning_rate": 8.743615934627171e-05, | |
| "loss": 0.1038, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 10.46875, | |
| "grad_norm": 0.2275228351354599, | |
| "learning_rate": 8.53932584269663e-05, | |
| "loss": 0.1165, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.625, | |
| "grad_norm": 0.3568095862865448, | |
| "learning_rate": 8.335035750766088e-05, | |
| "loss": 0.0986, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 10.78125, | |
| "grad_norm": 0.2227988988161087, | |
| "learning_rate": 8.130745658835548e-05, | |
| "loss": 0.1139, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 10.9375, | |
| "grad_norm": 0.3007115423679352, | |
| "learning_rate": 7.926455566905006e-05, | |
| "loss": 0.1004, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 10.9375, | |
| "eval_accuracy": 0.9451327433628318, | |
| "eval_f1": 0.8571428571428571, | |
| "eval_loss": 0.15245945751667023, | |
| "eval_precision": 0.8773584905660378, | |
| "eval_recall": 0.8378378378378378, | |
| "eval_runtime": 1.1963, | |
| "eval_samples_per_second": 94.461, | |
| "eval_steps_per_second": 12.539, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 11.09375, | |
| "grad_norm": 0.2825964689254761, | |
| "learning_rate": 7.722165474974464e-05, | |
| "loss": 0.1071, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "grad_norm": 0.2866518795490265, | |
| "learning_rate": 7.517875383043922e-05, | |
| "loss": 0.0797, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 11.40625, | |
| "grad_norm": 0.40209805965423584, | |
| "learning_rate": 7.313585291113382e-05, | |
| "loss": 0.0919, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 11.5625, | |
| "grad_norm": 0.22357816994190216, | |
| "learning_rate": 7.10929519918284e-05, | |
| "loss": 0.0862, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 11.71875, | |
| "grad_norm": 0.21823176741600037, | |
| "learning_rate": 6.905005107252299e-05, | |
| "loss": 0.0921, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 11.875, | |
| "grad_norm": 0.37442561984062195, | |
| "learning_rate": 6.700715015321757e-05, | |
| "loss": 0.0926, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 12.03125, | |
| "grad_norm": 0.10888390988111496, | |
| "learning_rate": 6.496424923391215e-05, | |
| "loss": 0.0758, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 12.1875, | |
| "grad_norm": 0.30619969964027405, | |
| "learning_rate": 6.292134831460675e-05, | |
| "loss": 0.0824, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 12.34375, | |
| "grad_norm": 0.2507191598415375, | |
| "learning_rate": 6.087844739530133e-05, | |
| "loss": 0.0815, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 0.1795010268688202, | |
| "learning_rate": 5.883554647599592e-05, | |
| "loss": 0.0831, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_accuracy": 0.9513274336283186, | |
| "eval_f1": 0.8741418764302059, | |
| "eval_loss": 0.14417380094528198, | |
| "eval_precision": 0.8883720930232558, | |
| "eval_recall": 0.8603603603603603, | |
| "eval_runtime": 0.9097, | |
| "eval_samples_per_second": 124.22, | |
| "eval_steps_per_second": 16.489, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 12.65625, | |
| "grad_norm": 0.3208577334880829, | |
| "learning_rate": 5.67926455566905e-05, | |
| "loss": 0.0759, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 12.8125, | |
| "grad_norm": 0.13223229348659515, | |
| "learning_rate": 5.474974463738509e-05, | |
| "loss": 0.087, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 12.96875, | |
| "grad_norm": 0.25973808765411377, | |
| "learning_rate": 5.270684371807968e-05, | |
| "loss": 0.0919, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 13.125, | |
| "grad_norm": 0.2930574417114258, | |
| "learning_rate": 5.0663942798774264e-05, | |
| "loss": 0.0826, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 13.28125, | |
| "grad_norm": 0.28331613540649414, | |
| "learning_rate": 4.862104187946885e-05, | |
| "loss": 0.0703, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 13.4375, | |
| "grad_norm": 0.5403579473495483, | |
| "learning_rate": 4.657814096016344e-05, | |
| "loss": 0.0877, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 13.59375, | |
| "grad_norm": 0.24780268967151642, | |
| "learning_rate": 4.453524004085802e-05, | |
| "loss": 0.0783, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "grad_norm": 0.43411558866500854, | |
| "learning_rate": 4.24923391215526e-05, | |
| "loss": 0.0809, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 13.90625, | |
| "grad_norm": 0.33748045563697815, | |
| "learning_rate": 4.044943820224719e-05, | |
| "loss": 0.0875, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 14.0625, | |
| "grad_norm": 0.18760047852993011, | |
| "learning_rate": 3.840653728294178e-05, | |
| "loss": 0.0654, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 14.0625, | |
| "eval_accuracy": 0.9495575221238938, | |
| "eval_f1": 0.8689655172413793, | |
| "eval_loss": 0.13779258728027344, | |
| "eval_precision": 0.8873239436619719, | |
| "eval_recall": 0.8513513513513513, | |
| "eval_runtime": 0.8895, | |
| "eval_samples_per_second": 127.038, | |
| "eval_steps_per_second": 16.863, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 14.21875, | |
| "grad_norm": 0.15740624070167542, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.0729, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 14.375, | |
| "grad_norm": 0.21466964483261108, | |
| "learning_rate": 3.4320735444330954e-05, | |
| "loss": 0.0613, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 14.53125, | |
| "grad_norm": 0.32680827379226685, | |
| "learning_rate": 3.2277834525025536e-05, | |
| "loss": 0.0636, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 14.6875, | |
| "grad_norm": 0.14597666263580322, | |
| "learning_rate": 3.0234933605720123e-05, | |
| "loss": 0.0758, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 14.84375, | |
| "grad_norm": 0.1691855937242508, | |
| "learning_rate": 2.819203268641471e-05, | |
| "loss": 0.0788, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.6366788148880005, | |
| "learning_rate": 2.61491317671093e-05, | |
| "loss": 0.0775, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 15.15625, | |
| "grad_norm": 0.4018935561180115, | |
| "learning_rate": 2.410623084780388e-05, | |
| "loss": 0.0616, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 15.3125, | |
| "grad_norm": 0.2861553728580475, | |
| "learning_rate": 2.2063329928498467e-05, | |
| "loss": 0.0648, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 15.46875, | |
| "grad_norm": 0.14629952609539032, | |
| "learning_rate": 2.0020429009193057e-05, | |
| "loss": 0.0687, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 15.625, | |
| "grad_norm": 0.18827380239963531, | |
| "learning_rate": 1.797752808988764e-05, | |
| "loss": 0.0583, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 15.625, | |
| "eval_accuracy": 0.9530973451327434, | |
| "eval_f1": 0.8798185941043084, | |
| "eval_loss": 0.13913680613040924, | |
| "eval_precision": 0.8858447488584474, | |
| "eval_recall": 0.8738738738738738, | |
| "eval_runtime": 1.128, | |
| "eval_samples_per_second": 100.175, | |
| "eval_steps_per_second": 13.298, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 15.78125, | |
| "grad_norm": 0.4585842788219452, | |
| "learning_rate": 1.593462717058223e-05, | |
| "loss": 0.061, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 15.9375, | |
| "grad_norm": 0.20938844978809357, | |
| "learning_rate": 1.3891726251276812e-05, | |
| "loss": 0.0682, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 16.09375, | |
| "grad_norm": 0.1632617712020874, | |
| "learning_rate": 1.18488253319714e-05, | |
| "loss": 0.065, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "grad_norm": 0.25956493616104126, | |
| "learning_rate": 9.805924412665986e-06, | |
| "loss": 0.0761, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 16.40625, | |
| "grad_norm": 0.17571082711219788, | |
| "learning_rate": 7.763023493360572e-06, | |
| "loss": 0.0676, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 16.5625, | |
| "grad_norm": 0.1871979534626007, | |
| "learning_rate": 5.720122574055159e-06, | |
| "loss": 0.0663, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 16.71875, | |
| "grad_norm": 0.18988612294197083, | |
| "learning_rate": 3.677221654749745e-06, | |
| "loss": 0.0608, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 16.875, | |
| "grad_norm": 0.18478451669216156, | |
| "learning_rate": 1.6343207354443311e-06, | |
| "loss": 0.0511, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "step": 1088, | |
| "total_flos": 1.333267779734618e+18, | |
| "train_loss": 0.16752693287151701, | |
| "train_runtime": 275.2566, | |
| "train_samples_per_second": 62.502, | |
| "train_steps_per_second": 3.953 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1088, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 17, | |
| "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": 1.333267779734618e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |