| { | |
| "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": 2.183218240737915, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5437, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7368421052631579, | |
| "eval_f1": 0.5754673962608299, | |
| "eval_loss": 0.4910251796245575, | |
| "eval_precision": 0.7052341597796143, | |
| "eval_recall": 0.5812875068194218, | |
| "eval_runtime": 5.0888, | |
| "eval_samples_per_second": 78.407, | |
| "eval_steps_per_second": 9.825, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.327057838439941, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4527, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.7979441442122369, | |
| "eval_loss": 0.3824920356273651, | |
| "eval_precision": 0.7972133421798662, | |
| "eval_recall": 0.7986906710310966, | |
| "eval_runtime": 5.1156, | |
| "eval_samples_per_second": 77.996, | |
| "eval_steps_per_second": 9.774, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.68674236536026, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3443, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8306935047100303, | |
| "eval_loss": 0.32941704988479614, | |
| "eval_precision": 0.8372758729160114, | |
| "eval_recall": 0.8249681760320058, | |
| "eval_runtime": 5.0642, | |
| "eval_samples_per_second": 78.789, | |
| "eval_steps_per_second": 9.873, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.137700080871582, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3026, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8446368446368446, | |
| "eval_loss": 0.3160928785800934, | |
| "eval_precision": 0.8473684210526315, | |
| "eval_recall": 0.8420621931260229, | |
| "eval_runtime": 5.0473, | |
| "eval_samples_per_second": 79.052, | |
| "eval_steps_per_second": 9.906, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 7.94999885559082, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2848, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8406078002012192, | |
| "eval_loss": 0.3151061534881592, | |
| "eval_precision": 0.8335745715609078, | |
| "eval_recall": 0.8492453173304237, | |
| "eval_runtime": 5.0502, | |
| "eval_samples_per_second": 79.006, | |
| "eval_steps_per_second": 9.901, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.067786455154419, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2605, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8514869535493182, | |
| "eval_loss": 0.28789207339286804, | |
| "eval_precision": 0.8523821128305106, | |
| "eval_recall": 0.8506092016730314, | |
| "eval_runtime": 5.0562, | |
| "eval_samples_per_second": 78.913, | |
| "eval_steps_per_second": 9.889, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.847172975540161, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2292, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8541488972828073, | |
| "eval_loss": 0.2645197808742523, | |
| "eval_precision": 0.8559859154929578, | |
| "eval_recall": 0.8523822513184216, | |
| "eval_runtime": 5.193, | |
| "eval_samples_per_second": 76.835, | |
| "eval_steps_per_second": 9.628, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.1587507724761963, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2196, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8569798318958896, | |
| "eval_loss": 0.2868158221244812, | |
| "eval_precision": 0.8469526727953693, | |
| "eval_recall": 0.870612838697945, | |
| "eval_runtime": 5.0649, | |
| "eval_samples_per_second": 78.778, | |
| "eval_steps_per_second": 9.872, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 2.913994789123535, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2076, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8602043795620438, | |
| "eval_loss": 0.2794610559940338, | |
| "eval_precision": 0.849624060150376, | |
| "eval_recall": 0.8748863429714493, | |
| "eval_runtime": 5.0987, | |
| "eval_samples_per_second": 78.255, | |
| "eval_steps_per_second": 9.806, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.650660991668701, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.201, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8582079268956014, | |
| "eval_loss": 0.30101457238197327, | |
| "eval_precision": 0.8464828897338403, | |
| "eval_recall": 0.8756137479541735, | |
| "eval_runtime": 5.0731, | |
| "eval_samples_per_second": 78.65, | |
| "eval_steps_per_second": 9.856, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.5577316284179688, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1919, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8556189207966651, | |
| "eval_loss": 0.29280906915664673, | |
| "eval_precision": 0.8436368195241544, | |
| "eval_recall": 0.8738406983087834, | |
| "eval_runtime": 5.1133, | |
| "eval_samples_per_second": 78.032, | |
| "eval_steps_per_second": 9.778, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.802311420440674, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1866, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8739355018846853, | |
| "eval_loss": 0.26604750752449036, | |
| "eval_precision": 0.864426651415499, | |
| "eval_recall": 0.886252045826514, | |
| "eval_runtime": 5.0575, | |
| "eval_samples_per_second": 78.893, | |
| "eval_steps_per_second": 9.886, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.0932844877243042, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1721, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8727838950061173, | |
| "eval_loss": 0.25936877727508545, | |
| "eval_precision": 0.8656898656898657, | |
| "eval_recall": 0.8812511365702855, | |
| "eval_runtime": 5.0873, | |
| "eval_samples_per_second": 78.431, | |
| "eval_steps_per_second": 9.828, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 7.470301151275635, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1734, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8629480286738351, | |
| "eval_loss": 0.24869810044765472, | |
| "eval_precision": 0.8580770590314599, | |
| "eval_recall": 0.8684306237497728, | |
| "eval_runtime": 5.0448, | |
| "eval_samples_per_second": 79.092, | |
| "eval_steps_per_second": 9.911, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.3498764038085938, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1601, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8666056166056166, | |
| "eval_loss": 0.29580533504486084, | |
| "eval_precision": 0.8549663299663299, | |
| "eval_recall": 0.8834333515184578, | |
| "eval_runtime": 5.0517, | |
| "eval_samples_per_second": 78.983, | |
| "eval_steps_per_second": 9.898, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.15194371342658997, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1586, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8602993213495533, | |
| "eval_loss": 0.2718944847583771, | |
| "eval_precision": 0.8547653958944281, | |
| "eval_recall": 0.8666575741043827, | |
| "eval_runtime": 5.0745, | |
| "eval_samples_per_second": 78.628, | |
| "eval_steps_per_second": 9.853, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.3160400986671448, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1486, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8543546116197471, | |
| "eval_loss": 0.27369293570518494, | |
| "eval_precision": 0.8489149560117302, | |
| "eval_recall": 0.8606110201854882, | |
| "eval_runtime": 5.0819, | |
| "eval_samples_per_second": 78.513, | |
| "eval_steps_per_second": 9.839, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.23276424407958984, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1474, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8642908431276217, | |
| "eval_loss": 0.26779675483703613, | |
| "eval_precision": 0.8633964654080464, | |
| "eval_recall": 0.8652027641389344, | |
| "eval_runtime": 5.0647, | |
| "eval_samples_per_second": 78.781, | |
| "eval_steps_per_second": 9.872, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.8074817657470703, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1511, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8583180446233059, | |
| "eval_loss": 0.2719312310218811, | |
| "eval_precision": 0.850900301065308, | |
| "eval_recall": 0.8673849790871068, | |
| "eval_runtime": 5.05, | |
| "eval_samples_per_second": 79.01, | |
| "eval_steps_per_second": 9.901, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.2888840138912201, | |
| "learning_rate": 0.0, | |
| "loss": 0.1296, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8589607635206786, | |
| "eval_loss": 0.28081414103507996, | |
| "eval_precision": 0.8503875968992248, | |
| "eval_recall": 0.8698854337152209, | |
| "eval_runtime": 5.0678, | |
| "eval_samples_per_second": 78.732, | |
| "eval_steps_per_second": 9.866, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.2332671552407937, | |
| "train_runtime": 1925.9874, | |
| "train_samples_per_second": 37.778, | |
| "train_steps_per_second": 1.267 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8444128359504000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |