| { | |
| "best_metric": 0.0380592942237854, | |
| "best_model_checkpoint": "doc-topic-model_eval-02_train-01/checkpoint-14000", | |
| "epoch": 9.368836291913215, | |
| "eval_steps": 1000, | |
| "global_step": 19000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2465483234714004, | |
| "grad_norm": 0.32817578315734863, | |
| "learning_rate": 1.9950690335305722e-05, | |
| "loss": 0.1659, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4930966469428008, | |
| "grad_norm": 0.34106990694999695, | |
| "learning_rate": 1.9901380670611442e-05, | |
| "loss": 0.0941, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4930966469428008, | |
| "eval_accuracy": 0.9813654208653996, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.09066470712423325, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 12.1347, | |
| "eval_samples_per_second": 668.329, | |
| "eval_steps_per_second": 2.637, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7396449704142012, | |
| "grad_norm": 0.4004117548465729, | |
| "learning_rate": 1.9852071005917162e-05, | |
| "loss": 0.088, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9861932938856016, | |
| "grad_norm": 0.37588435411453247, | |
| "learning_rate": 1.980276134122288e-05, | |
| "loss": 0.0787, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9861932938856016, | |
| "eval_accuracy": 0.9813654208653996, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.07067307829856873, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 12.2547, | |
| "eval_samples_per_second": 661.789, | |
| "eval_steps_per_second": 2.611, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.232741617357002, | |
| "grad_norm": 0.43946701288223267, | |
| "learning_rate": 1.9753451676528602e-05, | |
| "loss": 0.0696, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4792899408284024, | |
| "grad_norm": 0.4252438545227051, | |
| "learning_rate": 1.9704142011834322e-05, | |
| "loss": 0.0628, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4792899408284024, | |
| "eval_accuracy": 0.9822313873889196, | |
| "eval_f1": 0.1224889759921607, | |
| "eval_loss": 0.057456616312265396, | |
| "eval_precision": 0.7682177348551361, | |
| "eval_recall": 0.066550045634317, | |
| "eval_runtime": 14.2849, | |
| "eval_samples_per_second": 567.732, | |
| "eval_steps_per_second": 2.24, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.725838264299803, | |
| "grad_norm": 0.3535268306732178, | |
| "learning_rate": 1.965483234714004e-05, | |
| "loss": 0.0584, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.972386587771203, | |
| "grad_norm": 0.3976692855358124, | |
| "learning_rate": 1.9605522682445763e-05, | |
| "loss": 0.0537, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.972386587771203, | |
| "eval_accuracy": 0.9842042603852205, | |
| "eval_f1": 0.3201366436893796, | |
| "eval_loss": 0.050263650715351105, | |
| "eval_precision": 0.8086286594761171, | |
| "eval_recall": 0.19957407970794036, | |
| "eval_runtime": 14.3334, | |
| "eval_samples_per_second": 565.811, | |
| "eval_steps_per_second": 2.233, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.2189349112426036, | |
| "grad_norm": 0.24940212070941925, | |
| "learning_rate": 1.9556213017751483e-05, | |
| "loss": 0.0499, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.465483234714004, | |
| "grad_norm": 0.40094295144081116, | |
| "learning_rate": 1.95069033530572e-05, | |
| "loss": 0.0478, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.465483234714004, | |
| "eval_accuracy": 0.9851467607749763, | |
| "eval_f1": 0.4262564327165225, | |
| "eval_loss": 0.04696211963891983, | |
| "eval_precision": 0.7606486908948809, | |
| "eval_recall": 0.2960906601764527, | |
| "eval_runtime": 14.3295, | |
| "eval_samples_per_second": 565.965, | |
| "eval_steps_per_second": 2.233, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.712031558185404, | |
| "grad_norm": 0.42049625515937805, | |
| "learning_rate": 1.9457593688362923e-05, | |
| "loss": 0.0463, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.9585798816568047, | |
| "grad_norm": 0.41673073172569275, | |
| "learning_rate": 1.940828402366864e-05, | |
| "loss": 0.0453, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.9585798816568047, | |
| "eval_accuracy": 0.9857760392306929, | |
| "eval_f1": 0.4983003399320136, | |
| "eval_loss": 0.044437214732170105, | |
| "eval_precision": 0.7269544924154026, | |
| "eval_recall": 0.37906905993306966, | |
| "eval_runtime": 14.5887, | |
| "eval_samples_per_second": 555.911, | |
| "eval_steps_per_second": 2.193, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.2051282051282053, | |
| "grad_norm": 0.3240315914154053, | |
| "learning_rate": 1.935897435897436e-05, | |
| "loss": 0.0424, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.4516765285996054, | |
| "grad_norm": 0.39667147397994995, | |
| "learning_rate": 1.930966469428008e-05, | |
| "loss": 0.0389, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.4516765285996054, | |
| "eval_accuracy": 0.9864237425060589, | |
| "eval_f1": 0.5409058231488139, | |
| "eval_loss": 0.04192586615681648, | |
| "eval_precision": 0.7312427108980174, | |
| "eval_recall": 0.4291907514450867, | |
| "eval_runtime": 14.4636, | |
| "eval_samples_per_second": 560.719, | |
| "eval_steps_per_second": 2.212, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.698224852071006, | |
| "grad_norm": 0.39966270327568054, | |
| "learning_rate": 1.92603550295858e-05, | |
| "loss": 0.0403, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.9447731755424065, | |
| "grad_norm": 0.42601555585861206, | |
| "learning_rate": 1.921104536489152e-05, | |
| "loss": 0.0393, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.9447731755424065, | |
| "eval_accuracy": 0.9863302011139929, | |
| "eval_f1": 0.5480106846618867, | |
| "eval_loss": 0.04111693799495697, | |
| "eval_precision": 0.7138322549139299, | |
| "eval_recall": 0.4447064192272589, | |
| "eval_runtime": 14.5558, | |
| "eval_samples_per_second": 557.167, | |
| "eval_steps_per_second": 2.198, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.191321499013807, | |
| "grad_norm": 0.41485270857810974, | |
| "learning_rate": 1.916173570019724e-05, | |
| "loss": 0.0361, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.437869822485207, | |
| "grad_norm": 0.4282112717628479, | |
| "learning_rate": 1.911242603550296e-05, | |
| "loss": 0.0349, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.437869822485207, | |
| "eval_accuracy": 0.9868149156001531, | |
| "eval_f1": 0.57472, | |
| "eval_loss": 0.039900876581668854, | |
| "eval_precision": 0.7202933425002864, | |
| "eval_recall": 0.4780955278369334, | |
| "eval_runtime": 14.4367, | |
| "eval_samples_per_second": 561.763, | |
| "eval_steps_per_second": 2.217, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.684418145956608, | |
| "grad_norm": 0.5737874507904053, | |
| "learning_rate": 1.906311637080868e-05, | |
| "loss": 0.0358, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.930966469428008, | |
| "grad_norm": 0.42196542024612427, | |
| "learning_rate": 1.90138067061144e-05, | |
| "loss": 0.0344, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.930966469428008, | |
| "eval_accuracy": 0.9870388480235838, | |
| "eval_f1": 0.575815204786864, | |
| "eval_loss": 0.03913598880171776, | |
| "eval_precision": 0.7379621923671382, | |
| "eval_recall": 0.47208700943109216, | |
| "eval_runtime": 14.3745, | |
| "eval_samples_per_second": 564.193, | |
| "eval_steps_per_second": 2.226, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.177514792899408, | |
| "grad_norm": 0.2073846012353897, | |
| "learning_rate": 1.896459566074951e-05, | |
| "loss": 0.0315, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.424063116370808, | |
| "grad_norm": 0.3816153109073639, | |
| "learning_rate": 1.891528599605523e-05, | |
| "loss": 0.0302, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.424063116370808, | |
| "eval_accuracy": 0.9871295548280113, | |
| "eval_f1": 0.5903739455997113, | |
| "eval_loss": 0.03845709562301636, | |
| "eval_precision": 0.7254184680190666, | |
| "eval_recall": 0.49771828414968056, | |
| "eval_runtime": 14.5373, | |
| "eval_samples_per_second": 557.876, | |
| "eval_steps_per_second": 2.201, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.670611439842209, | |
| "grad_norm": 0.4880998730659485, | |
| "learning_rate": 1.886597633136095e-05, | |
| "loss": 0.0321, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.9171597633136095, | |
| "grad_norm": 0.31077542901039124, | |
| "learning_rate": 1.881666666666667e-05, | |
| "loss": 0.0305, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.9171597633136095, | |
| "eval_accuracy": 0.9871040435392662, | |
| "eval_f1": 0.5966218912089374, | |
| "eval_loss": 0.038683824241161346, | |
| "eval_precision": 0.7151663301094696, | |
| "eval_recall": 0.5117888652266505, | |
| "eval_runtime": 14.4549, | |
| "eval_samples_per_second": 561.056, | |
| "eval_steps_per_second": 2.214, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.16370808678501, | |
| "grad_norm": 0.4404219686985016, | |
| "learning_rate": 1.876735700197239e-05, | |
| "loss": 0.0281, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "grad_norm": 0.39860355854034424, | |
| "learning_rate": 1.8718145956607497e-05, | |
| "loss": 0.027, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "eval_accuracy": 0.9874441940558697, | |
| "eval_f1": 0.605653238370799, | |
| "eval_loss": 0.0383986160159111, | |
| "eval_precision": 0.7301706557904905, | |
| "eval_recall": 0.5174170976574384, | |
| "eval_runtime": 14.2815, | |
| "eval_samples_per_second": 567.867, | |
| "eval_steps_per_second": 2.241, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.65680473372781, | |
| "grad_norm": 0.3162294924259186, | |
| "learning_rate": 1.8668836291913217e-05, | |
| "loss": 0.0267, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.903353057199211, | |
| "grad_norm": 0.3698256313800812, | |
| "learning_rate": 1.8619526627218937e-05, | |
| "loss": 0.0282, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.903353057199211, | |
| "eval_accuracy": 0.9875349008602974, | |
| "eval_f1": 0.6079436544376589, | |
| "eval_loss": 0.0380592942237854, | |
| "eval_precision": 0.7344103392568659, | |
| "eval_recall": 0.5186340127776088, | |
| "eval_runtime": 14.3349, | |
| "eval_samples_per_second": 565.751, | |
| "eval_steps_per_second": 2.232, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.149901380670611, | |
| "grad_norm": 0.42440325021743774, | |
| "learning_rate": 1.8570216962524657e-05, | |
| "loss": 0.0252, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.396449704142012, | |
| "grad_norm": 0.5403586626052856, | |
| "learning_rate": 1.8521005917159765e-05, | |
| "loss": 0.0235, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.396449704142012, | |
| "eval_accuracy": 0.9874016752412943, | |
| "eval_f1": 0.6180880773361976, | |
| "eval_loss": 0.038525186479091644, | |
| "eval_precision": 0.7102794509726473, | |
| "eval_recall": 0.5470794037115911, | |
| "eval_runtime": 14.711, | |
| "eval_samples_per_second": 551.288, | |
| "eval_steps_per_second": 2.175, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.642998027613412, | |
| "grad_norm": 0.3601667582988739, | |
| "learning_rate": 1.8471696252465485e-05, | |
| "loss": 0.0243, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 7.889546351084813, | |
| "grad_norm": 0.4192611575126648, | |
| "learning_rate": 1.8422386587771205e-05, | |
| "loss": 0.0255, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 7.889546351084813, | |
| "eval_accuracy": 0.9876312768400017, | |
| "eval_f1": 0.6257237208903376, | |
| "eval_loss": 0.038173139095306396, | |
| "eval_precision": 0.7173763398564263, | |
| "eval_recall": 0.5548372376026772, | |
| "eval_runtime": 14.4228, | |
| "eval_samples_per_second": 562.305, | |
| "eval_steps_per_second": 2.219, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.136094674556213, | |
| "grad_norm": 0.13463503122329712, | |
| "learning_rate": 1.8373076923076926e-05, | |
| "loss": 0.0224, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 8.382642998027613, | |
| "grad_norm": 0.25073572993278503, | |
| "learning_rate": 1.8323865877712033e-05, | |
| "loss": 0.0214, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.382642998027613, | |
| "eval_accuracy": 0.9877319047011636, | |
| "eval_f1": 0.6352911435072048, | |
| "eval_loss": 0.0381532646715641, | |
| "eval_precision": 0.7121670130360854, | |
| "eval_recall": 0.5733951931852753, | |
| "eval_runtime": 14.3056, | |
| "eval_samples_per_second": 566.91, | |
| "eval_steps_per_second": 2.237, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.629191321499015, | |
| "grad_norm": 0.39663228392601013, | |
| "learning_rate": 1.8274556213017754e-05, | |
| "loss": 0.0215, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 8.875739644970414, | |
| "grad_norm": 0.7587671279907227, | |
| "learning_rate": 1.822524654832347e-05, | |
| "loss": 0.0222, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 8.875739644970414, | |
| "eval_accuracy": 0.9876114347265331, | |
| "eval_f1": 0.628153316033522, | |
| "eval_loss": 0.038794856518507004, | |
| "eval_precision": 0.7127135823921228, | |
| "eval_recall": 0.5615302707636143, | |
| "eval_runtime": 14.4829, | |
| "eval_samples_per_second": 559.969, | |
| "eval_steps_per_second": 2.209, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.122287968441814, | |
| "grad_norm": 0.14514634013175964, | |
| "learning_rate": 1.8175936883629194e-05, | |
| "loss": 0.0198, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 9.368836291913215, | |
| "grad_norm": 0.44190242886543274, | |
| "learning_rate": 1.8126725838264302e-05, | |
| "loss": 0.0192, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 9.368836291913215, | |
| "eval_accuracy": 0.9874668707569766, | |
| "eval_f1": 0.632078219263574, | |
| "eval_loss": 0.039640188217163086, | |
| "eval_precision": 0.6977128685588316, | |
| "eval_recall": 0.5777304533008822, | |
| "eval_runtime": 14.7066, | |
| "eval_samples_per_second": 551.452, | |
| "eval_steps_per_second": 2.176, | |
| "step": 19000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 202800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 429748313255712.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |