| { | |
| "best_metric": 0.7952755905511811, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-nv5m4p1b/checkpoint-800", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 800, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 18.084890365600586, | |
| "learning_rate": 5.399134210269773e-05, | |
| "loss": 0.556, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 6.5466694831848145, | |
| "learning_rate": 5.35376333455322e-05, | |
| "loss": 0.428, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 49.454490661621094, | |
| "learning_rate": 5.308392458836667e-05, | |
| "loss": 0.6545, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 7.895252227783203, | |
| "learning_rate": 5.263021583120115e-05, | |
| "loss": 1.0023, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 4.6973557472229, | |
| "learning_rate": 5.217650707403562e-05, | |
| "loss": 0.5786, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 3.350486993789673, | |
| "learning_rate": 5.172279831687009e-05, | |
| "loss": 0.6606, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 3.6134517192840576, | |
| "learning_rate": 5.126908955970456e-05, | |
| "loss": 0.6316, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.9971961379051208, | |
| "learning_rate": 5.081538080253904e-05, | |
| "loss": 0.4201, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 8.022686958312988, | |
| "learning_rate": 5.0361672045373515e-05, | |
| "loss": 0.62, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.8167064189910889, | |
| "learning_rate": 4.9907963288207984e-05, | |
| "loss": 0.5476, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "grad_norm": 5.9797797203063965, | |
| "learning_rate": 4.9454254531042454e-05, | |
| "loss": 0.4258, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 6.271321773529053, | |
| "learning_rate": 4.900054577387693e-05, | |
| "loss": 0.516, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "grad_norm": 4.605900764465332, | |
| "learning_rate": 4.8546837016711406e-05, | |
| "loss": 0.59, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 1.0727519989013672, | |
| "learning_rate": 4.8093128259545876e-05, | |
| "loss": 0.4789, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 4.704715728759766, | |
| "learning_rate": 4.7639419502380346e-05, | |
| "loss": 0.6051, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.037107229232788, | |
| "learning_rate": 4.718571074521482e-05, | |
| "loss": 0.5743, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.425, | |
| "grad_norm": 3.19757080078125, | |
| "learning_rate": 4.673200198804929e-05, | |
| "loss": 0.4151, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 2.1952502727508545, | |
| "learning_rate": 4.627829323088377e-05, | |
| "loss": 0.3531, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.475, | |
| "grad_norm": 14.308310508728027, | |
| "learning_rate": 4.582458447371824e-05, | |
| "loss": 0.8135, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 33.65572738647461, | |
| "learning_rate": 4.5370875716552714e-05, | |
| "loss": 0.4655, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.525, | |
| "grad_norm": 6.842982769012451, | |
| "learning_rate": 4.491716695938718e-05, | |
| "loss": 0.3649, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 15.931615829467773, | |
| "learning_rate": 4.446345820222166e-05, | |
| "loss": 0.4005, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.575, | |
| "grad_norm": 2.720935583114624, | |
| "learning_rate": 4.400974944505613e-05, | |
| "loss": 0.523, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 9.985756874084473, | |
| "learning_rate": 4.3556040687890605e-05, | |
| "loss": 0.5802, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.3308340609073639, | |
| "learning_rate": 4.3102331930725075e-05, | |
| "loss": 0.2914, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 10.186966896057129, | |
| "learning_rate": 4.264862317355955e-05, | |
| "loss": 0.2921, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.675, | |
| "grad_norm": 0.18446782231330872, | |
| "learning_rate": 4.219491441639402e-05, | |
| "loss": 0.4803, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.7676132917404175, | |
| "learning_rate": 4.17412056592285e-05, | |
| "loss": 0.4423, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.725, | |
| "grad_norm": 10.771402359008789, | |
| "learning_rate": 4.1287496902062967e-05, | |
| "loss": 0.5073, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 4.975055694580078, | |
| "learning_rate": 4.083378814489744e-05, | |
| "loss": 0.6165, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.775, | |
| "grad_norm": 4.036560535430908, | |
| "learning_rate": 4.038007938773191e-05, | |
| "loss": 0.4933, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.0244158506393433, | |
| "learning_rate": 3.992637063056638e-05, | |
| "loss": 0.4126, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.825, | |
| "grad_norm": 11.22476577758789, | |
| "learning_rate": 3.947266187340086e-05, | |
| "loss": 0.2672, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 4.474503993988037, | |
| "learning_rate": 3.9018953116235335e-05, | |
| "loss": 0.4559, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 4.21657133102417, | |
| "learning_rate": 3.856524435906981e-05, | |
| "loss": 0.5414, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 10.542683601379395, | |
| "learning_rate": 3.8111535601904274e-05, | |
| "loss": 0.3255, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.925, | |
| "grad_norm": 0.3486887216567993, | |
| "learning_rate": 3.765782684473875e-05, | |
| "loss": 0.2273, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.1581709086894989, | |
| "learning_rate": 3.7204118087573226e-05, | |
| "loss": 0.4178, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.975, | |
| "grad_norm": 0.31422266364097595, | |
| "learning_rate": 3.67504093304077e-05, | |
| "loss": 0.4512, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 6.026626110076904, | |
| "learning_rate": 3.6296700573242165e-05, | |
| "loss": 0.4803, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8325, | |
| "eval_f1": 0.7649122807017544, | |
| "eval_loss": 0.44036921858787537, | |
| "eval_precision": 0.6728395061728395, | |
| "eval_recall": 0.8861788617886179, | |
| "eval_runtime": 1.5152, | |
| "eval_samples_per_second": 263.984, | |
| "eval_steps_per_second": 16.499, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.025, | |
| "grad_norm": 0.3346502184867859, | |
| "learning_rate": 3.584299181607664e-05, | |
| "loss": 0.3594, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.33411934971809387, | |
| "learning_rate": 3.538928305891112e-05, | |
| "loss": 0.2471, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.075, | |
| "grad_norm": 0.25553181767463684, | |
| "learning_rate": 3.4935574301745594e-05, | |
| "loss": 0.2113, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.14269419014453888, | |
| "learning_rate": 3.448186554458006e-05, | |
| "loss": 0.2325, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 0.24317187070846558, | |
| "learning_rate": 3.4028156787414533e-05, | |
| "loss": 0.2899, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 2.1619760990142822, | |
| "learning_rate": 3.357444803024901e-05, | |
| "loss": 0.2133, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.175, | |
| "grad_norm": 0.12860938906669617, | |
| "learning_rate": 3.312073927308348e-05, | |
| "loss": 0.3433, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 56.71578598022461, | |
| "learning_rate": 3.266703051591795e-05, | |
| "loss": 1.18, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.225, | |
| "grad_norm": 1.0443360805511475, | |
| "learning_rate": 3.2213321758752425e-05, | |
| "loss": 0.1778, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 5.665371417999268, | |
| "learning_rate": 3.17596130015869e-05, | |
| "loss": 0.2249, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.275, | |
| "grad_norm": 7.218950271606445, | |
| "learning_rate": 3.130590424442137e-05, | |
| "loss": 0.5477, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 11.46741008758545, | |
| "learning_rate": 3.085219548725584e-05, | |
| "loss": 0.4139, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.325, | |
| "grad_norm": 18.249954223632812, | |
| "learning_rate": 3.0398486730090317e-05, | |
| "loss": 0.3268, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.25124236941337585, | |
| "learning_rate": 2.9944777972924793e-05, | |
| "loss": 0.4642, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 29.628658294677734, | |
| "learning_rate": 2.949106921575926e-05, | |
| "loss": 0.3392, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 5.741694450378418, | |
| "learning_rate": 2.9037360458593736e-05, | |
| "loss": 0.1288, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.425, | |
| "grad_norm": 18.95865821838379, | |
| "learning_rate": 2.858365170142821e-05, | |
| "loss": 0.481, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.11765194684267044, | |
| "learning_rate": 2.8129942944262685e-05, | |
| "loss": 0.5451, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.475, | |
| "grad_norm": 0.4500724673271179, | |
| "learning_rate": 2.7676234187097155e-05, | |
| "loss": 0.5212, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.705097496509552, | |
| "learning_rate": 2.7222525429931627e-05, | |
| "loss": 0.3214, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.525, | |
| "grad_norm": 0.7480908632278442, | |
| "learning_rate": 2.67688166727661e-05, | |
| "loss": 0.0825, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.7471107244491577, | |
| "learning_rate": 2.6315107915600573e-05, | |
| "loss": 0.7354, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.575, | |
| "grad_norm": 21.289236068725586, | |
| "learning_rate": 2.5861399158435046e-05, | |
| "loss": 0.3518, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.137455016374588, | |
| "learning_rate": 2.540769040126952e-05, | |
| "loss": 0.1983, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.19616670906543732, | |
| "learning_rate": 2.4953981644103992e-05, | |
| "loss": 0.6471, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 0.26698774099349976, | |
| "learning_rate": 2.4500272886938465e-05, | |
| "loss": 0.0821, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.675, | |
| "grad_norm": 5.839506149291992, | |
| "learning_rate": 2.4046564129772938e-05, | |
| "loss": 0.3946, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 4.699467182159424, | |
| "learning_rate": 2.359285537260741e-05, | |
| "loss": 0.5219, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.725, | |
| "grad_norm": 7.885561466217041, | |
| "learning_rate": 2.3139146615441884e-05, | |
| "loss": 0.7581, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.13073155283927917, | |
| "learning_rate": 2.2685437858276357e-05, | |
| "loss": 0.2281, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.775, | |
| "grad_norm": 14.315596580505371, | |
| "learning_rate": 2.223172910111083e-05, | |
| "loss": 0.2418, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 7.1058669090271, | |
| "learning_rate": 2.1778020343945303e-05, | |
| "loss": 0.3002, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.825, | |
| "grad_norm": 0.06962228566408157, | |
| "learning_rate": 2.1324311586779776e-05, | |
| "loss": 0.3782, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 5.521971225738525, | |
| "learning_rate": 2.087060282961425e-05, | |
| "loss": 0.3967, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.1677718609571457, | |
| "learning_rate": 2.041689407244872e-05, | |
| "loss": 0.1929, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 64.3675308227539, | |
| "learning_rate": 1.996318531528319e-05, | |
| "loss": 0.1186, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.925, | |
| "grad_norm": 0.12294362485408783, | |
| "learning_rate": 1.9509476558117667e-05, | |
| "loss": 0.07, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 6.7742390632629395, | |
| "learning_rate": 1.9055767800952137e-05, | |
| "loss": 0.8638, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.975, | |
| "grad_norm": 0.38380953669548035, | |
| "learning_rate": 1.8602059043786613e-05, | |
| "loss": 0.6582, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.6939935684204102, | |
| "learning_rate": 1.8148350286621083e-05, | |
| "loss": 0.2934, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.87, | |
| "eval_f1": 0.7952755905511811, | |
| "eval_loss": 0.42544516921043396, | |
| "eval_precision": 0.7709923664122137, | |
| "eval_recall": 0.8211382113821138, | |
| "eval_runtime": 1.5207, | |
| "eval_samples_per_second": 263.039, | |
| "eval_steps_per_second": 16.44, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 423630740901888.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 5.4445050859863255e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 4, | |
| "seed": 8 | |
| } | |
| } | |