| { | |
| "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": 4.803397178649902, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5411, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7368421052631579, | |
| "eval_f1": 0.6508662716567915, | |
| "eval_loss": 0.49393221735954285, | |
| "eval_precision": 0.6761904761904762, | |
| "eval_recall": 0.6412984178941625, | |
| "eval_runtime": 1.7881, | |
| "eval_samples_per_second": 223.142, | |
| "eval_steps_per_second": 27.963, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.498361587524414, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4231, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.7995262704565029, | |
| "eval_loss": 0.3851858377456665, | |
| "eval_precision": 0.7887596899224806, | |
| "eval_recall": 0.8183760683760684, | |
| "eval_runtime": 1.7893, | |
| "eval_samples_per_second": 222.988, | |
| "eval_steps_per_second": 27.943, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.0573930740356445, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3331, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8080535642463387, | |
| "eval_loss": 0.33130019903182983, | |
| "eval_precision": 0.8232818379877204, | |
| "eval_recall": 0.796826695762866, | |
| "eval_runtime": 1.7976, | |
| "eval_samples_per_second": 221.96, | |
| "eval_steps_per_second": 27.815, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.250720500946045, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2924, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8560793854229822, | |
| "eval_loss": 0.30566585063934326, | |
| "eval_precision": 0.8609538327526132, | |
| "eval_recall": 0.8516548463356974, | |
| "eval_runtime": 1.7957, | |
| "eval_samples_per_second": 222.195, | |
| "eval_steps_per_second": 27.844, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.8374593257904053, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2705, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8422176526415692, | |
| "eval_loss": 0.3068975508213043, | |
| "eval_precision": 0.8604724566416373, | |
| "eval_recall": 0.8288325150027278, | |
| "eval_runtime": 1.7956, | |
| "eval_samples_per_second": 222.215, | |
| "eval_steps_per_second": 27.846, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.305149078369141, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2461, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8562182887453875, | |
| "eval_loss": 0.31193241477012634, | |
| "eval_precision": 0.8435805201992252, | |
| "eval_recall": 0.8763411529368976, | |
| "eval_runtime": 1.7944, | |
| "eval_samples_per_second": 222.356, | |
| "eval_steps_per_second": 27.864, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.4272942841053009, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2313, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8662440310793597, | |
| "eval_loss": 0.28799474239349365, | |
| "eval_precision": 0.8606158357771261, | |
| "eval_recall": 0.872704128023277, | |
| "eval_runtime": 1.8006, | |
| "eval_samples_per_second": 221.595, | |
| "eval_steps_per_second": 27.769, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.898254871368408, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2183, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8676331036823873, | |
| "eval_loss": 0.27734559774398804, | |
| "eval_precision": 0.8749292230261088, | |
| "eval_recall": 0.8612474995453718, | |
| "eval_runtime": 1.7976, | |
| "eval_samples_per_second": 221.957, | |
| "eval_steps_per_second": 27.814, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.726850509643555, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2093, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8587719298245614, | |
| "eval_loss": 0.28041473031044006, | |
| "eval_precision": 0.864771021021021, | |
| "eval_recall": 0.8534278959810875, | |
| "eval_runtime": 1.7999, | |
| "eval_samples_per_second": 221.684, | |
| "eval_steps_per_second": 27.78, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.7861063480377197, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1986, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8654532336864889, | |
| "eval_loss": 0.28901827335357666, | |
| "eval_precision": 0.8804269882659713, | |
| "eval_recall": 0.8537461356610292, | |
| "eval_runtime": 1.7942, | |
| "eval_samples_per_second": 222.384, | |
| "eval_steps_per_second": 27.868, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.18681001663208, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1881, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8629148629148629, | |
| "eval_loss": 0.29107582569122314, | |
| "eval_precision": 0.8657894736842104, | |
| "eval_recall": 0.860201854882706, | |
| "eval_runtime": 1.8131, | |
| "eval_samples_per_second": 220.068, | |
| "eval_steps_per_second": 27.577, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 3.137617588043213, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1802, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8568221901555235, | |
| "eval_loss": 0.28662246465682983, | |
| "eval_precision": 0.8596491228070176, | |
| "eval_recall": 0.8541553009638116, | |
| "eval_runtime": 1.8223, | |
| "eval_samples_per_second": 218.951, | |
| "eval_steps_per_second": 27.437, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.8551347851753235, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.169, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8564658408408408, | |
| "eval_loss": 0.2963576018810272, | |
| "eval_precision": 0.8697278911564625, | |
| "eval_recall": 0.8459265320967448, | |
| "eval_runtime": 1.8327, | |
| "eval_samples_per_second": 217.71, | |
| "eval_steps_per_second": 27.282, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 5.66387414932251, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1709, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8629148629148629, | |
| "eval_loss": 0.29438090324401855, | |
| "eval_precision": 0.8657894736842104, | |
| "eval_recall": 0.860201854882706, | |
| "eval_runtime": 1.8299, | |
| "eval_samples_per_second": 218.043, | |
| "eval_steps_per_second": 27.324, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.363886833190918, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1492, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8636104675452922, | |
| "eval_loss": 0.28655046224594116, | |
| "eval_precision": 0.8645363713902765, | |
| "eval_recall": 0.8627023095108202, | |
| "eval_runtime": 1.8286, | |
| "eval_samples_per_second": 218.201, | |
| "eval_steps_per_second": 27.344, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.6481318473815918, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1493, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8742647058823529, | |
| "eval_loss": 0.2950553596019745, | |
| "eval_precision": 0.8707860158154468, | |
| "eval_recall": 0.8780232769594472, | |
| "eval_runtime": 1.8286, | |
| "eval_samples_per_second": 218.205, | |
| "eval_steps_per_second": 27.344, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.7666211724281311, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1425, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8710526315789473, | |
| "eval_loss": 0.304831326007843, | |
| "eval_precision": 0.8772522522522522, | |
| "eval_recall": 0.8655210038188761, | |
| "eval_runtime": 1.7983, | |
| "eval_samples_per_second": 221.876, | |
| "eval_steps_per_second": 27.804, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.819899797439575, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1375, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.8790689216221131, | |
| "eval_loss": 0.298705130815506, | |
| "eval_precision": 0.8790689216221131, | |
| "eval_recall": 0.8790689216221131, | |
| "eval_runtime": 1.8038, | |
| "eval_samples_per_second": 221.2, | |
| "eval_steps_per_second": 27.719, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 1.7430284023284912, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1326, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.8778322106552358, | |
| "eval_loss": 0.30734923481941223, | |
| "eval_precision": 0.8818924438393465, | |
| "eval_recall": 0.8740680123658847, | |
| "eval_runtime": 1.8064, | |
| "eval_samples_per_second": 220.886, | |
| "eval_steps_per_second": 27.68, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.931983709335327, | |
| "learning_rate": 0.0, | |
| "loss": 0.1365, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.8784574144023395, | |
| "eval_loss": 0.29978305101394653, | |
| "eval_precision": 0.8804194733619106, | |
| "eval_recall": 0.8765684669939988, | |
| "eval_runtime": 1.8111, | |
| "eval_samples_per_second": 220.313, | |
| "eval_steps_per_second": 27.608, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.2259816083751741, | |
| "train_runtime": 638.5228, | |
| "train_samples_per_second": 113.951, | |
| "train_steps_per_second": 3.821 | |
| } | |
| ], | |
| "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 | |
| } | |