| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 10560, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.21040940284729, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.79, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8595349299413694, | |
| "eval_f1": 0.13053613053613053, | |
| "eval_loss": 0.4638139605522156, | |
| "eval_precision": 0.330188679245283, | |
| "eval_recall": 0.08134805345729228, | |
| "eval_runtime": 4.441, | |
| "eval_samples_per_second": 210.54, | |
| "eval_steps_per_second": 3.378, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.8728626370429993, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.3919, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9275067077412302, | |
| "eval_f1": 0.6317512274959083, | |
| "eval_loss": 0.2518700361251831, | |
| "eval_precision": 0.5953727506426735, | |
| "eval_recall": 0.6728646135967461, | |
| "eval_runtime": 4.584, | |
| "eval_samples_per_second": 203.97, | |
| "eval_steps_per_second": 3.272, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.2418427467346191, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.2386, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9381894067375535, | |
| "eval_f1": 0.7159389794844818, | |
| "eval_loss": 0.1926584243774414, | |
| "eval_precision": 0.6540124939932724, | |
| "eval_recall": 0.7908192911098199, | |
| "eval_runtime": 4.5313, | |
| "eval_samples_per_second": 206.344, | |
| "eval_steps_per_second": 3.31, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.416585087776184, | |
| "learning_rate": 4e-05, | |
| "loss": 0.193, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.944797774023651, | |
| "eval_f1": 0.7463787200421385, | |
| "eval_loss": 0.16769319772720337, | |
| "eval_precision": 0.6825626204238922, | |
| "eval_recall": 0.8233585124927367, | |
| "eval_runtime": 4.5613, | |
| "eval_samples_per_second": 204.984, | |
| "eval_steps_per_second": 3.289, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.0125393867492676, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.1712, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9475802444599026, | |
| "eval_f1": 0.7629299028616435, | |
| "eval_loss": 0.1593998372554779, | |
| "eval_precision": 0.6958812260536399, | |
| "eval_recall": 0.8442765833817548, | |
| "eval_runtime": 4.6306, | |
| "eval_samples_per_second": 201.917, | |
| "eval_steps_per_second": 3.239, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.202476143836975, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.1596, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9497664712312431, | |
| "eval_f1": 0.7750591949486978, | |
| "eval_loss": 0.1543859839439392, | |
| "eval_precision": 0.708173076923077, | |
| "eval_recall": 0.8558977338756537, | |
| "eval_runtime": 4.6085, | |
| "eval_samples_per_second": 202.887, | |
| "eval_steps_per_second": 3.255, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.2051475048065186, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.1524, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.950561462784458, | |
| "eval_f1": 0.7727628489433862, | |
| "eval_loss": 0.15189574658870697, | |
| "eval_precision": 0.7012310606060606, | |
| "eval_recall": 0.8605461940732132, | |
| "eval_runtime": 4.9569, | |
| "eval_samples_per_second": 188.627, | |
| "eval_steps_per_second": 3.026, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.5321469306945801, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1452, | |
| "step": 4224 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9521514458908874, | |
| "eval_f1": 0.7842202806460153, | |
| "eval_loss": 0.14606598019599915, | |
| "eval_precision": 0.7203307392996109, | |
| "eval_recall": 0.8605461940732132, | |
| "eval_runtime": 4.8452, | |
| "eval_samples_per_second": 192.976, | |
| "eval_steps_per_second": 3.096, | |
| "step": 4224 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.277453064918518, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.1397, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9535426811090132, | |
| "eval_f1": 0.7858095492131236, | |
| "eval_loss": 0.14319637417793274, | |
| "eval_precision": 0.7263313609467456, | |
| "eval_recall": 0.8558977338756537, | |
| "eval_runtime": 4.5883, | |
| "eval_samples_per_second": 203.778, | |
| "eval_steps_per_second": 3.269, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.3887606859207153, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1369, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9539401768856206, | |
| "eval_f1": 0.784512683578104, | |
| "eval_loss": 0.13940192759037018, | |
| "eval_precision": 0.7257905138339921, | |
| "eval_recall": 0.8535735037768739, | |
| "eval_runtime": 4.5465, | |
| "eval_samples_per_second": 205.653, | |
| "eval_steps_per_second": 3.299, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.879906415939331, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1336, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.954287985690152, | |
| "eval_f1": 0.7872111767866737, | |
| "eval_loss": 0.13752727210521698, | |
| "eval_precision": 0.7321339330334833, | |
| "eval_recall": 0.8512492736780941, | |
| "eval_runtime": 4.5285, | |
| "eval_samples_per_second": 206.468, | |
| "eval_steps_per_second": 3.312, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.3187131881713867, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1305, | |
| "step": 6336 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9547351684388353, | |
| "eval_f1": 0.7895726955119592, | |
| "eval_loss": 0.13747519254684448, | |
| "eval_precision": 0.7345, | |
| "eval_recall": 0.8535735037768739, | |
| "eval_runtime": 4.6283, | |
| "eval_samples_per_second": 202.018, | |
| "eval_steps_per_second": 3.241, | |
| "step": 6336 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.600463628768921, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1281, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9547351684388353, | |
| "eval_f1": 0.7887248322147652, | |
| "eval_loss": 0.13505811989307404, | |
| "eval_precision": 0.7330339321357285, | |
| "eval_recall": 0.8535735037768739, | |
| "eval_runtime": 4.594, | |
| "eval_samples_per_second": 203.525, | |
| "eval_steps_per_second": 3.265, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.121996283531189, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1252, | |
| "step": 7392 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9553314121037464, | |
| "eval_f1": 0.794345158708989, | |
| "eval_loss": 0.13602255284786224, | |
| "eval_precision": 0.7342209072978304, | |
| "eval_recall": 0.8651946542707728, | |
| "eval_runtime": 4.5938, | |
| "eval_samples_per_second": 203.536, | |
| "eval_steps_per_second": 3.265, | |
| "step": 7392 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.0593712329864502, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.124, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9541389247739243, | |
| "eval_f1": 0.7874899759422614, | |
| "eval_loss": 0.13636602461338043, | |
| "eval_precision": 0.7292079207920792, | |
| "eval_recall": 0.8558977338756537, | |
| "eval_runtime": 4.563, | |
| "eval_samples_per_second": 204.907, | |
| "eval_steps_per_second": 3.287, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.4150090217590332, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1234, | |
| "step": 8448 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.954933916327139, | |
| "eval_f1": 0.7875565009306035, | |
| "eval_loss": 0.13505925238132477, | |
| "eval_precision": 0.7259803921568627, | |
| "eval_recall": 0.8605461940732132, | |
| "eval_runtime": 4.502, | |
| "eval_samples_per_second": 207.687, | |
| "eval_steps_per_second": 3.332, | |
| "step": 8448 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.8472805619239807, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1224, | |
| "step": 8976 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.954933916327139, | |
| "eval_f1": 0.7918106886466365, | |
| "eval_loss": 0.13572688400745392, | |
| "eval_precision": 0.7299019607843137, | |
| "eval_recall": 0.8651946542707728, | |
| "eval_runtime": 4.5475, | |
| "eval_samples_per_second": 205.606, | |
| "eval_steps_per_second": 3.298, | |
| "step": 8976 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.521100401878357, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1208, | |
| "step": 9504 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9553314121037464, | |
| "eval_f1": 0.7947830715996806, | |
| "eval_loss": 0.1359829157590866, | |
| "eval_precision": 0.7333005893909627, | |
| "eval_recall": 0.8675188843695526, | |
| "eval_runtime": 4.5351, | |
| "eval_samples_per_second": 206.171, | |
| "eval_steps_per_second": 3.308, | |
| "step": 9504 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.9698516726493835, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1201, | |
| "step": 10032 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.95553015999205, | |
| "eval_f1": 0.7956301625366373, | |
| "eval_loss": 0.13499899208545685, | |
| "eval_precision": 0.734744094488189, | |
| "eval_recall": 0.8675188843695526, | |
| "eval_runtime": 4.5154, | |
| "eval_samples_per_second": 207.071, | |
| "eval_steps_per_second": 3.322, | |
| "step": 10032 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.61683189868927, | |
| "learning_rate": 0.0, | |
| "loss": 0.1205, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.95553015999205, | |
| "eval_f1": 0.794345158708989, | |
| "eval_loss": 0.1346217691898346, | |
| "eval_precision": 0.7342209072978304, | |
| "eval_recall": 0.8651946542707728, | |
| "eval_runtime": 4.496, | |
| "eval_samples_per_second": 207.961, | |
| "eval_steps_per_second": 3.336, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 10560, | |
| "total_flos": 4541164131293502.0, | |
| "train_loss": 0.18834613236514006, | |
| "train_runtime": 1254.4553, | |
| "train_samples_per_second": 134.513, | |
| "train_steps_per_second": 8.418 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10560, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 4541164131293502.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |