| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.4341020584106445, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.3817, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9515033947623667, | |
| "eval_f1": 0.793193717277487, | |
| "eval_loss": 0.14420253038406372, | |
| "eval_precision": 0.7266187050359713, | |
| "eval_recall": 0.8731988472622478, | |
| "eval_runtime": 0.9468, | |
| "eval_samples_per_second": 197.515, | |
| "eval_steps_per_second": 3.169, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.4988839626312256, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.1266, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9551406401551892, | |
| "eval_f1": 0.8083441981747066, | |
| "eval_loss": 0.13850077986717224, | |
| "eval_precision": 0.7380952380952381, | |
| "eval_recall": 0.8933717579250721, | |
| "eval_runtime": 0.9888, | |
| "eval_samples_per_second": 189.116, | |
| "eval_steps_per_second": 3.034, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.6240944862365723, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.087, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9568380213385063, | |
| "eval_f1": 0.8100929614873836, | |
| "eval_loss": 0.13670620322227478, | |
| "eval_precision": 0.7512315270935961, | |
| "eval_recall": 0.8789625360230547, | |
| "eval_runtime": 0.9262, | |
| "eval_samples_per_second": 201.901, | |
| "eval_steps_per_second": 3.239, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.2754727900028229, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0528, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9595053346265762, | |
| "eval_f1": 0.8163265306122449, | |
| "eval_loss": 0.14675553143024445, | |
| "eval_precision": 0.7731958762886598, | |
| "eval_recall": 0.8645533141210374, | |
| "eval_runtime": 0.9413, | |
| "eval_samples_per_second": 198.665, | |
| "eval_steps_per_second": 3.187, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.4952049255371094, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.0424, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.960717749757517, | |
| "eval_f1": 0.8215767634854771, | |
| "eval_loss": 0.1664169430732727, | |
| "eval_precision": 0.7898936170212766, | |
| "eval_recall": 0.8559077809798271, | |
| "eval_runtime": 1.041, | |
| "eval_samples_per_second": 179.63, | |
| "eval_steps_per_second": 2.882, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.5230388641357422, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.0275, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9582929194956353, | |
| "eval_f1": 0.8241610738255033, | |
| "eval_loss": 0.20444069802761078, | |
| "eval_precision": 0.7713567839195979, | |
| "eval_recall": 0.8847262247838616, | |
| "eval_runtime": 0.9522, | |
| "eval_samples_per_second": 196.387, | |
| "eval_steps_per_second": 3.151, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.4965060949325562, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.019, | |
| "step": 742 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9553831231813773, | |
| "eval_f1": 0.8089005235602094, | |
| "eval_loss": 0.23766779899597168, | |
| "eval_precision": 0.7410071942446043, | |
| "eval_recall": 0.8904899135446686, | |
| "eval_runtime": 0.9461, | |
| "eval_samples_per_second": 197.656, | |
| "eval_steps_per_second": 3.171, | |
| "step": 742 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.0428667068481445, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0145, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9587778855480117, | |
| "eval_f1": 0.8279569892473119, | |
| "eval_loss": 0.24319829046726227, | |
| "eval_precision": 0.7758186397984886, | |
| "eval_recall": 0.8876080691642652, | |
| "eval_runtime": 0.9448, | |
| "eval_samples_per_second": 197.924, | |
| "eval_steps_per_second": 3.175, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.5034522414207458, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.0102, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9641125121241513, | |
| "eval_f1": 0.854024556616644, | |
| "eval_loss": 0.22865384817123413, | |
| "eval_precision": 0.810880829015544, | |
| "eval_recall": 0.9020172910662824, | |
| "eval_runtime": 0.9532, | |
| "eval_samples_per_second": 196.191, | |
| "eval_steps_per_second": 3.147, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.0857200101017952, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0067, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9616876818622696, | |
| "eval_f1": 0.8442622950819673, | |
| "eval_loss": 0.24303482472896576, | |
| "eval_precision": 0.8025974025974026, | |
| "eval_recall": 0.8904899135446686, | |
| "eval_runtime": 0.9473, | |
| "eval_samples_per_second": 197.406, | |
| "eval_steps_per_second": 3.167, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.07605205476284027, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.0064, | |
| "step": 1166 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9602327837051406, | |
| "eval_f1": 0.8396739130434782, | |
| "eval_loss": 0.2674606740474701, | |
| "eval_precision": 0.794344473007712, | |
| "eval_recall": 0.8904899135446686, | |
| "eval_runtime": 0.9552, | |
| "eval_samples_per_second": 195.768, | |
| "eval_steps_per_second": 3.141, | |
| "step": 1166 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.8096215724945068, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0046, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9619301648884578, | |
| "eval_f1": 0.8344549125168236, | |
| "eval_loss": 0.2742844820022583, | |
| "eval_precision": 0.7828282828282829, | |
| "eval_recall": 0.8933717579250721, | |
| "eval_runtime": 1.0274, | |
| "eval_samples_per_second": 182.02, | |
| "eval_steps_per_second": 2.92, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.9064086675643921, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.0034, | |
| "step": 1378 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9619301648884578, | |
| "eval_f1": 0.8451086956521738, | |
| "eval_loss": 0.2666186988353729, | |
| "eval_precision": 0.7994858611825193, | |
| "eval_recall": 0.8962536023054755, | |
| "eval_runtime": 0.9543, | |
| "eval_samples_per_second": 195.964, | |
| "eval_steps_per_second": 3.144, | |
| "step": 1378 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.7048726081848145, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0036, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9633850630455868, | |
| "eval_f1": 0.8453038674033149, | |
| "eval_loss": 0.2606286108493805, | |
| "eval_precision": 0.8116710875331565, | |
| "eval_recall": 0.8818443804034583, | |
| "eval_runtime": 0.9565, | |
| "eval_samples_per_second": 195.509, | |
| "eval_steps_per_second": 3.137, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.10259877145290375, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0027, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9626576139670223, | |
| "eval_f1": 0.8405405405405406, | |
| "eval_loss": 0.2861556112766266, | |
| "eval_precision": 0.7913486005089059, | |
| "eval_recall": 0.8962536023054755, | |
| "eval_runtime": 0.9627, | |
| "eval_samples_per_second": 194.249, | |
| "eval_steps_per_second": 3.116, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.023264136165380478, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0016, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9629000969932104, | |
| "eval_f1": 0.8426812585499316, | |
| "eval_loss": 0.2792609930038452, | |
| "eval_precision": 0.8020833333333334, | |
| "eval_recall": 0.8876080691642652, | |
| "eval_runtime": 0.9674, | |
| "eval_samples_per_second": 193.295, | |
| "eval_steps_per_second": 3.101, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 2.410804271697998, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.0012, | |
| "step": 1802 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.962172647914646, | |
| "eval_f1": 0.841248303934871, | |
| "eval_loss": 0.29505419731140137, | |
| "eval_precision": 0.7948717948717948, | |
| "eval_recall": 0.8933717579250721, | |
| "eval_runtime": 0.9612, | |
| "eval_samples_per_second": 194.551, | |
| "eval_steps_per_second": 3.121, | |
| "step": 1802 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.004150555469095707, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0012, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9616876818622696, | |
| "eval_f1": 0.8380952380952381, | |
| "eval_loss": 0.29299911856651306, | |
| "eval_precision": 0.7938144329896907, | |
| "eval_recall": 0.8876080691642652, | |
| "eval_runtime": 0.9455, | |
| "eval_samples_per_second": 197.787, | |
| "eval_steps_per_second": 3.173, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.0056025926023721695, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0014, | |
| "step": 2014 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9612027158098934, | |
| "eval_f1": 0.8353741496598639, | |
| "eval_loss": 0.29529693722724915, | |
| "eval_precision": 0.7912371134020618, | |
| "eval_recall": 0.8847262247838616, | |
| "eval_runtime": 0.9556, | |
| "eval_samples_per_second": 195.689, | |
| "eval_steps_per_second": 3.139, | |
| "step": 2014 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.012953029945492744, | |
| "learning_rate": 0.0, | |
| "loss": 0.0007, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9619301648884578, | |
| "eval_f1": 0.8387978142076504, | |
| "eval_loss": 0.29178306460380554, | |
| "eval_precision": 0.7974025974025974, | |
| "eval_recall": 0.8847262247838616, | |
| "eval_runtime": 0.9529, | |
| "eval_samples_per_second": 196.248, | |
| "eval_steps_per_second": 3.148, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2120, | |
| "total_flos": 918687991011936.0, | |
| "train_loss": 0.03976648653734405, | |
| "train_runtime": 515.0309, | |
| "train_samples_per_second": 65.549, | |
| "train_steps_per_second": 4.116 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 918687991011936.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |