| { | |
| "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": 6.052820205688477, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5445, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7518796992481203, | |
| "eval_f1": 0.6708167704192605, | |
| "eval_loss": 0.48936620354652405, | |
| "eval_precision": 0.6988095238095238, | |
| "eval_recall": 0.6594380796508457, | |
| "eval_runtime": 5.1391, | |
| "eval_samples_per_second": 77.64, | |
| "eval_steps_per_second": 9.729, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.3890905380249023, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.432, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_f1": 0.7954312407428505, | |
| "eval_loss": 0.36909204721450806, | |
| "eval_precision": 0.7842791202823785, | |
| "eval_recall": 0.8173304237134025, | |
| "eval_runtime": 5.0614, | |
| "eval_samples_per_second": 78.833, | |
| "eval_steps_per_second": 9.879, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.7538790702819824, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3342, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.7956919505700826, | |
| "eval_loss": 0.33010444045066833, | |
| "eval_precision": 0.85742518351214, | |
| "eval_recall": 0.7685942898708855, | |
| "eval_runtime": 5.0669, | |
| "eval_samples_per_second": 78.747, | |
| "eval_steps_per_second": 9.868, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.051776170730591, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2846, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8502252252252251, | |
| "eval_loss": 0.2885676324367523, | |
| "eval_precision": 0.863265306122449, | |
| "eval_recall": 0.8398799781778505, | |
| "eval_runtime": 5.0493, | |
| "eval_samples_per_second": 79.021, | |
| "eval_steps_per_second": 9.902, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.2031478881835938, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2621, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8488361520276414, | |
| "eval_loss": 0.27282822132110596, | |
| "eval_precision": 0.8488361520276414, | |
| "eval_recall": 0.8488361520276414, | |
| "eval_runtime": 5.0626, | |
| "eval_samples_per_second": 78.814, | |
| "eval_steps_per_second": 9.876, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.391174554824829, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2419, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8503151260504201, | |
| "eval_loss": 0.27526727318763733, | |
| "eval_precision": 0.8470628455912955, | |
| "eval_recall": 0.8538370612838698, | |
| "eval_runtime": 5.0607, | |
| "eval_samples_per_second": 78.843, | |
| "eval_steps_per_second": 9.88, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.6138697862625122, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2132, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8622899159663866, | |
| "eval_loss": 0.27531638741493225, | |
| "eval_precision": 0.8589244307033712, | |
| "eval_recall": 0.8659301691216585, | |
| "eval_runtime": 5.0703, | |
| "eval_samples_per_second": 78.693, | |
| "eval_steps_per_second": 9.861, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.8498824834823608, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2055, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8541488972828073, | |
| "eval_loss": 0.27910253405570984, | |
| "eval_precision": 0.8559859154929578, | |
| "eval_recall": 0.8523822513184216, | |
| "eval_runtime": 5.0606, | |
| "eval_samples_per_second": 78.844, | |
| "eval_steps_per_second": 9.88, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.801692485809326, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.1903, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8306935047100303, | |
| "eval_loss": 0.3009396195411682, | |
| "eval_precision": 0.8372758729160114, | |
| "eval_recall": 0.8249681760320058, | |
| "eval_runtime": 5.0892, | |
| "eval_samples_per_second": 78.402, | |
| "eval_steps_per_second": 9.825, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.017825126647949, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1852, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8341632880321839, | |
| "eval_loss": 0.3085351884365082, | |
| "eval_precision": 0.8479139504563233, | |
| "eval_recall": 0.8235133660665576, | |
| "eval_runtime": 5.0614, | |
| "eval_samples_per_second": 78.832, | |
| "eval_steps_per_second": 9.879, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.8085110187530518, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1758, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8556004584112431, | |
| "eval_loss": 0.28515923023223877, | |
| "eval_precision": 0.8538865546218487, | |
| "eval_recall": 0.85738316057465, | |
| "eval_runtime": 5.053, | |
| "eval_samples_per_second": 78.964, | |
| "eval_steps_per_second": 9.895, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 4.22409725189209, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1617, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8642908431276217, | |
| "eval_loss": 0.30107688903808594, | |
| "eval_precision": 0.8633964654080464, | |
| "eval_recall": 0.8652027641389344, | |
| "eval_runtime": 5.0864, | |
| "eval_samples_per_second": 78.444, | |
| "eval_steps_per_second": 9.83, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 4.442287921905518, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1581, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8703223612108386, | |
| "eval_loss": 0.3050229847431183, | |
| "eval_precision": 0.8694131129742446, | |
| "eval_recall": 0.8712493180578287, | |
| "eval_runtime": 5.0713, | |
| "eval_samples_per_second": 78.678, | |
| "eval_steps_per_second": 9.859, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 3.8908846378326416, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.149, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8744925055412909, | |
| "eval_loss": 0.3143095374107361, | |
| "eval_precision": 0.8639342866870956, | |
| "eval_recall": 0.8887525004546282, | |
| "eval_runtime": 5.0958, | |
| "eval_samples_per_second": 78.299, | |
| "eval_steps_per_second": 9.812, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.6980299949645996, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1386, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.8790689216221131, | |
| "eval_loss": 0.302778035402298, | |
| "eval_precision": 0.8790689216221131, | |
| "eval_recall": 0.8790689216221131, | |
| "eval_runtime": 5.0849, | |
| "eval_samples_per_second": 78.468, | |
| "eval_steps_per_second": 9.833, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.6725683212280273, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1465, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8696722245432793, | |
| "eval_loss": 0.3111669719219208, | |
| "eval_precision": 0.8706135006701596, | |
| "eval_recall": 0.8687488634297145, | |
| "eval_runtime": 5.0592, | |
| "eval_samples_per_second": 78.866, | |
| "eval_steps_per_second": 9.883, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.7393283247947693, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1307, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8629480286738351, | |
| "eval_loss": 0.31975144147872925, | |
| "eval_precision": 0.8580770590314599, | |
| "eval_recall": 0.8684306237497728, | |
| "eval_runtime": 5.0777, | |
| "eval_samples_per_second": 78.579, | |
| "eval_steps_per_second": 9.847, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.382880836725235, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1231, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8609292598654301, | |
| "eval_loss": 0.32525473833084106, | |
| "eval_precision": 0.8609292598654301, | |
| "eval_recall": 0.8609292598654301, | |
| "eval_runtime": 5.096, | |
| "eval_samples_per_second": 78.296, | |
| "eval_steps_per_second": 9.812, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.1525256484746933, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1344, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8541488972828073, | |
| "eval_loss": 0.3289809823036194, | |
| "eval_precision": 0.8559859154929578, | |
| "eval_recall": 0.8523822513184216, | |
| "eval_runtime": 5.0693, | |
| "eval_samples_per_second": 78.709, | |
| "eval_steps_per_second": 9.863, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 4.0254364013671875, | |
| "learning_rate": 0.0, | |
| "loss": 0.1229, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8589543987905864, | |
| "eval_loss": 0.32526540756225586, | |
| "eval_precision": 0.8564068100358423, | |
| "eval_recall": 0.8616566648481543, | |
| "eval_runtime": 5.0827, | |
| "eval_samples_per_second": 78.502, | |
| "eval_steps_per_second": 9.837, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.21670899273919278, | |
| "train_runtime": 1925.8785, | |
| "train_samples_per_second": 37.78, | |
| "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 | |
| } | |