| { | |
| "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": 4.279876708984375, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5457, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7192982456140351, | |
| "eval_f1": 0.6013986013986015, | |
| "eval_loss": 0.4752861261367798, | |
| "eval_precision": 0.6464762782128051, | |
| "eval_recall": 0.5963811602109474, | |
| "eval_runtime": 5.2249, | |
| "eval_samples_per_second": 76.365, | |
| "eval_steps_per_second": 9.57, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.457512855529785, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4518, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7969924812030075, | |
| "eval_f1": 0.7684901970616256, | |
| "eval_loss": 0.4070126414299011, | |
| "eval_precision": 0.7589055735189347, | |
| "eval_recall": 0.7863702491362066, | |
| "eval_runtime": 5.1546, | |
| "eval_samples_per_second": 77.407, | |
| "eval_steps_per_second": 9.7, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.400643825531006, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3461, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.7970482088214634, | |
| "eval_loss": 0.3412257730960846, | |
| "eval_precision": 0.8231252179979072, | |
| "eval_recall": 0.7807783233315149, | |
| "eval_runtime": 5.1898, | |
| "eval_samples_per_second": 76.882, | |
| "eval_steps_per_second": 9.634, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.3028647899627686, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2958, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8228567054500919, | |
| "eval_loss": 0.3252820372581482, | |
| "eval_precision": 0.8263351692555232, | |
| "eval_recall": 0.8196490270958356, | |
| "eval_runtime": 5.1759, | |
| "eval_samples_per_second": 77.088, | |
| "eval_steps_per_second": 9.66, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.7779637575149536, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2659, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8560793854229822, | |
| "eval_loss": 0.2941116690635681, | |
| "eval_precision": 0.8609538327526132, | |
| "eval_recall": 0.8516548463356974, | |
| "eval_runtime": 5.1453, | |
| "eval_samples_per_second": 77.546, | |
| "eval_steps_per_second": 9.718, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.210702896118164, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2482, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8563451067988272, | |
| "eval_loss": 0.2965351343154907, | |
| "eval_precision": 0.8473119816985988, | |
| "eval_recall": 0.8681123840698308, | |
| "eval_runtime": 5.1607, | |
| "eval_samples_per_second": 77.315, | |
| "eval_steps_per_second": 9.689, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.5390880703926086, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2264, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8530841286673736, | |
| "eval_loss": 0.286903977394104, | |
| "eval_precision": 0.8446597760551249, | |
| "eval_recall": 0.8638388797963266, | |
| "eval_runtime": 5.4309, | |
| "eval_samples_per_second": 73.468, | |
| "eval_steps_per_second": 9.207, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.480511426925659, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2218, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.8730431462390225, | |
| "eval_loss": 0.2795129120349884, | |
| "eval_precision": 0.8961111111111111, | |
| "eval_recall": 0.8565648299690853, | |
| "eval_runtime": 5.1715, | |
| "eval_samples_per_second": 77.153, | |
| "eval_steps_per_second": 9.668, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.716876029968262, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2106, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8715803025426456, | |
| "eval_loss": 0.27050527930259705, | |
| "eval_precision": 0.8673433153814287, | |
| "eval_recall": 0.8762502273140571, | |
| "eval_runtime": 5.1606, | |
| "eval_samples_per_second": 77.317, | |
| "eval_steps_per_second": 9.689, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.200309991836548, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1981, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9072681704260651, | |
| "eval_f1": 0.8867007927797945, | |
| "eval_loss": 0.275096595287323, | |
| "eval_precision": 0.89198606271777, | |
| "eval_recall": 0.8818876159301692, | |
| "eval_runtime": 5.1671, | |
| "eval_samples_per_second": 77.22, | |
| "eval_steps_per_second": 9.677, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.14013409614563, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1802, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_f1": 0.8862394957983193, | |
| "eval_loss": 0.27452006936073303, | |
| "eval_precision": 0.8826476009275225, | |
| "eval_recall": 0.8901163847972358, | |
| "eval_runtime": 5.1559, | |
| "eval_samples_per_second": 77.387, | |
| "eval_steps_per_second": 9.698, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 6.242837905883789, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1828, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9072681704260651, | |
| "eval_f1": 0.8854915648632926, | |
| "eval_loss": 0.279909610748291, | |
| "eval_precision": 0.8956662848415425, | |
| "eval_recall": 0.8768867066739408, | |
| "eval_runtime": 5.1353, | |
| "eval_samples_per_second": 77.698, | |
| "eval_steps_per_second": 9.737, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.8258763551712036, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1707, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9097744360902256, | |
| "eval_f1": 0.8894736842105263, | |
| "eval_loss": 0.27386558055877686, | |
| "eval_precision": 0.8959740990990992, | |
| "eval_recall": 0.8836606655755592, | |
| "eval_runtime": 5.1775, | |
| "eval_samples_per_second": 77.065, | |
| "eval_steps_per_second": 9.657, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 3.6909942626953125, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1606, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9072681704260651, | |
| "eval_f1": 0.888964101175568, | |
| "eval_loss": 0.28680744767189026, | |
| "eval_precision": 0.8862007168458781, | |
| "eval_recall": 0.8918894344426259, | |
| "eval_runtime": 5.1734, | |
| "eval_samples_per_second": 77.126, | |
| "eval_steps_per_second": 9.665, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.49116143584251404, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1499, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.8817957385392532, | |
| "eval_loss": 0.2929786145687103, | |
| "eval_precision": 0.8827677592299257, | |
| "eval_recall": 0.8808419712675032, | |
| "eval_runtime": 5.1562, | |
| "eval_samples_per_second": 77.383, | |
| "eval_steps_per_second": 9.697, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.3892961740493774, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1555, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8760282890453928, | |
| "eval_loss": 0.3041006028652191, | |
| "eval_precision": 0.8682260305697083, | |
| "eval_recall": 0.8855246408437898, | |
| "eval_runtime": 5.1466, | |
| "eval_samples_per_second": 77.527, | |
| "eval_steps_per_second": 9.715, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.9728598594665527, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1396, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.8823853973772722, | |
| "eval_loss": 0.28764259815216064, | |
| "eval_precision": 0.8814464081066409, | |
| "eval_recall": 0.8833424258956174, | |
| "eval_runtime": 5.1398, | |
| "eval_samples_per_second": 77.63, | |
| "eval_steps_per_second": 9.728, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.8131844997406006, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1477, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_f1": 0.8845345436822225, | |
| "eval_loss": 0.28996890783309937, | |
| "eval_precision": 0.8865278628291489, | |
| "eval_recall": 0.8826150209128933, | |
| "eval_runtime": 5.3979, | |
| "eval_samples_per_second": 73.917, | |
| "eval_steps_per_second": 9.263, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.289058208465576, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1434, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_f1": 0.8851154755410074, | |
| "eval_loss": 0.2917177975177765, | |
| "eval_precision": 0.8851154755410074, | |
| "eval_recall": 0.8851154755410074, | |
| "eval_runtime": 5.1418, | |
| "eval_samples_per_second": 77.6, | |
| "eval_steps_per_second": 9.724, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.46004581451416, | |
| "learning_rate": 0.0, | |
| "loss": 0.1386, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_f1": 0.8851154755410074, | |
| "eval_loss": 0.2913039028644562, | |
| "eval_precision": 0.8851154755410074, | |
| "eval_recall": 0.8851154755410074, | |
| "eval_runtime": 5.1424, | |
| "eval_samples_per_second": 77.591, | |
| "eval_steps_per_second": 9.723, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.22896970373685244, | |
| "train_runtime": 1968.7629, | |
| "train_samples_per_second": 36.957, | |
| "train_steps_per_second": 1.239 | |
| } | |
| ], | |
| "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 | |
| } | |