| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 10.745368957519531, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.3971, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.851257564821498, | |
| "eval_loss": 0.27043774724006653, | |
| "eval_precision": 0.8714882943143812, | |
| "eval_recall": 0.8366521185670122, | |
| "eval_runtime": 4.9491, | |
| "eval_samples_per_second": 80.621, | |
| "eval_steps_per_second": 10.103, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 7.182548999786377, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.214, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9197994987468672, | |
| "eval_f1": 0.9032551372976905, | |
| "eval_loss": 0.23814059793949127, | |
| "eval_precision": 0.9032551372976905, | |
| "eval_recall": 0.9032551372976905, | |
| "eval_runtime": 4.9784, | |
| "eval_samples_per_second": 80.146, | |
| "eval_steps_per_second": 10.043, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 93.81620025634766, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.1301, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8439374185136896, | |
| "eval_loss": 0.38320210576057434, | |
| "eval_precision": 0.8794955044955045, | |
| "eval_recall": 0.822376795781051, | |
| "eval_runtime": 5.0811, | |
| "eval_samples_per_second": 78.526, | |
| "eval_steps_per_second": 9.84, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.22968685626983643, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0904, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8782051282051282, | |
| "eval_loss": 0.3873072564601898, | |
| "eval_precision": 0.8661616161616161, | |
| "eval_recall": 0.8955264593562466, | |
| "eval_runtime": 4.9496, | |
| "eval_samples_per_second": 80.613, | |
| "eval_steps_per_second": 10.102, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.057639382779598236, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.0587, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9147869674185464, | |
| "eval_f1": 0.8956140350877193, | |
| "eval_loss": 0.4033690392971039, | |
| "eval_precision": 0.9022147147147147, | |
| "eval_recall": 0.8897072194944535, | |
| "eval_runtime": 4.9468, | |
| "eval_samples_per_second": 80.659, | |
| "eval_steps_per_second": 10.108, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.12450123578310013, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.0496, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8755702215614461, | |
| "eval_loss": 0.5248555541038513, | |
| "eval_precision": 0.8632157235517781, | |
| "eval_recall": 0.8937534097108566, | |
| "eval_runtime": 4.9627, | |
| "eval_samples_per_second": 80.399, | |
| "eval_steps_per_second": 10.075, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.029339170083403587, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.0362, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9072681704260651, | |
| "eval_f1": 0.8842272010790718, | |
| "eval_loss": 0.533008873462677, | |
| "eval_precision": 0.8999348109517601, | |
| "eval_recall": 0.8718857974177123, | |
| "eval_runtime": 4.956, | |
| "eval_samples_per_second": 80.508, | |
| "eval_steps_per_second": 10.089, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.012251710519194603, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0223, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8769602202215754, | |
| "eval_loss": 0.6168325543403625, | |
| "eval_precision": 0.8742831541218639, | |
| "eval_recall": 0.8797963266048372, | |
| "eval_runtime": 4.9435, | |
| "eval_samples_per_second": 80.712, | |
| "eval_steps_per_second": 10.114, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.027291299775242805, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.0191, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8674628282189181, | |
| "eval_loss": 0.7103565335273743, | |
| "eval_precision": 0.8869858462356303, | |
| "eval_recall": 0.8530187306783051, | |
| "eval_runtime": 4.9624, | |
| "eval_samples_per_second": 80.405, | |
| "eval_steps_per_second": 10.076, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.26321402192115784, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0147, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8775533117267087, | |
| "eval_loss": 0.6125138998031616, | |
| "eval_precision": 0.873246730188791, | |
| "eval_recall": 0.8822967812329514, | |
| "eval_runtime": 4.95, | |
| "eval_samples_per_second": 80.606, | |
| "eval_steps_per_second": 10.101, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.00528654595836997, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.0121, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_f1": 0.888343937787204, | |
| "eval_loss": 0.6882844567298889, | |
| "eval_precision": 0.8790267011197244, | |
| "eval_recall": 0.9001182033096926, | |
| "eval_runtime": 4.9822, | |
| "eval_samples_per_second": 80.085, | |
| "eval_steps_per_second": 10.036, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.007998102344572544, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0118, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8781334505389722, | |
| "eval_loss": 0.6760030388832092, | |
| "eval_precision": 0.872316715542522, | |
| "eval_recall": 0.8847972358610656, | |
| "eval_runtime": 4.9668, | |
| "eval_samples_per_second": 80.334, | |
| "eval_steps_per_second": 10.067, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.002160990610718727, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.0034, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_f1": 0.8851154755410074, | |
| "eval_loss": 0.7163070440292358, | |
| "eval_precision": 0.8851154755410074, | |
| "eval_recall": 0.8851154755410074, | |
| "eval_runtime": 4.9658, | |
| "eval_samples_per_second": 80.349, | |
| "eval_steps_per_second": 10.069, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.09897086024284363, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0064, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9072681704260651, | |
| "eval_f1": 0.8884169154604891, | |
| "eval_loss": 0.7180221080780029, | |
| "eval_precision": 0.8874630556728391, | |
| "eval_recall": 0.8893889798145117, | |
| "eval_runtime": 4.9417, | |
| "eval_samples_per_second": 80.741, | |
| "eval_steps_per_second": 10.118, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.004853605292737484, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.008, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_f1": 0.8807643913180245, | |
| "eval_loss": 0.6914934515953064, | |
| "eval_precision": 0.8978286538966151, | |
| "eval_recall": 0.8676122931442081, | |
| "eval_runtime": 4.9615, | |
| "eval_samples_per_second": 80.42, | |
| "eval_steps_per_second": 10.078, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.0016086915275081992, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0052, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9072681704260651, | |
| "eval_f1": 0.8884169154604891, | |
| "eval_loss": 0.6778000593185425, | |
| "eval_precision": 0.8874630556728391, | |
| "eval_recall": 0.8893889798145117, | |
| "eval_runtime": 4.9627, | |
| "eval_samples_per_second": 80.4, | |
| "eval_steps_per_second": 10.075, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.0021244632080197334, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.0066, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9072681704260651, | |
| "eval_f1": 0.8884169154604891, | |
| "eval_loss": 0.6992811560630798, | |
| "eval_precision": 0.8874630556728391, | |
| "eval_recall": 0.8893889798145117, | |
| "eval_runtime": 4.9649, | |
| "eval_samples_per_second": 80.364, | |
| "eval_steps_per_second": 10.071, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.0017945035360753536, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0053, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_f1": 0.8814283033033032, | |
| "eval_loss": 0.6966382265090942, | |
| "eval_precision": 0.895578231292517, | |
| "eval_recall": 0.8701127477723223, | |
| "eval_runtime": 4.9631, | |
| "eval_samples_per_second": 80.393, | |
| "eval_steps_per_second": 10.074, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.0012101498432457447, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0022, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.8765393898137261, | |
| "eval_loss": 0.7112175822257996, | |
| "eval_precision": 0.8852261942423283, | |
| "eval_recall": 0.8690671031096563, | |
| "eval_runtime": 4.9613, | |
| "eval_samples_per_second": 80.422, | |
| "eval_steps_per_second": 10.078, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.009784560650587082, | |
| "learning_rate": 0.0, | |
| "loss": 0.0042, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.8793019197207679, | |
| "eval_loss": 0.7049560546875, | |
| "eval_precision": 0.8893184421534936, | |
| "eval_recall": 0.8708401527550463, | |
| "eval_runtime": 4.9243, | |
| "eval_samples_per_second": 81.027, | |
| "eval_steps_per_second": 10.154, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7584162436176000.0, | |
| "train_loss": 0.05485985055565834, | |
| "train_runtime": 2706.9642, | |
| "train_samples_per_second": 26.879, | |
| "train_steps_per_second": 0.901 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7584162436176000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |