| { | |
| "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": 4.705079078674316, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5593, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7268170426065163, | |
| "eval_f1": 0.6589395923680764, | |
| "eval_loss": 0.5025668144226074, | |
| "eval_precision": 0.6658409387222947, | |
| "eval_recall": 0.6542098563375159, | |
| "eval_runtime": 5.1146, | |
| "eval_samples_per_second": 78.012, | |
| "eval_steps_per_second": 9.776, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.789808988571167, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4995, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7543859649122807, | |
| "eval_f1": 0.7226336397684712, | |
| "eval_loss": 0.47967690229415894, | |
| "eval_precision": 0.7148526077097506, | |
| "eval_recall": 0.7412256773958902, | |
| "eval_runtime": 5.0589, | |
| "eval_samples_per_second": 78.87, | |
| "eval_steps_per_second": 9.883, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.145122051239014, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4612, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7644110275689223, | |
| "eval_f1": 0.7262335766423358, | |
| "eval_loss": 0.4281724691390991, | |
| "eval_precision": 0.7199248120300752, | |
| "eval_recall": 0.7358156028368794, | |
| "eval_runtime": 5.0857, | |
| "eval_samples_per_second": 78.455, | |
| "eval_steps_per_second": 9.831, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.555657386779785, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4019, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7933776044839771, | |
| "eval_loss": 0.3933873772621155, | |
| "eval_precision": 0.7949020208205757, | |
| "eval_recall": 0.7919167121294781, | |
| "eval_runtime": 5.0576, | |
| "eval_samples_per_second": 78.891, | |
| "eval_steps_per_second": 9.886, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.5041733980178833, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3665, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7969924812030075, | |
| "eval_f1": 0.772043420300895, | |
| "eval_loss": 0.42343708872795105, | |
| "eval_precision": 0.7618072289156627, | |
| "eval_recall": 0.7963720676486634, | |
| "eval_runtime": 5.0851, | |
| "eval_samples_per_second": 78.464, | |
| "eval_steps_per_second": 9.833, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.9622180461883545, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.334, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_f1": 0.788441145281018, | |
| "eval_loss": 0.3723360300064087, | |
| "eval_precision": 0.7816537467700257, | |
| "eval_recall": 0.7973267866884888, | |
| "eval_runtime": 5.0586, | |
| "eval_samples_per_second": 78.875, | |
| "eval_steps_per_second": 9.884, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.7456966042518616, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3263, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8086080586080586, | |
| "eval_loss": 0.37042734026908875, | |
| "eval_precision": 0.7989898989898989, | |
| "eval_recall": 0.8229678123295144, | |
| "eval_runtime": 5.0772, | |
| "eval_samples_per_second": 78.586, | |
| "eval_steps_per_second": 9.848, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.445054531097412, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3076, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8160386984618873, | |
| "eval_loss": 0.352139413356781, | |
| "eval_precision": 0.8152632848784607, | |
| "eval_recall": 0.8168303327877796, | |
| "eval_runtime": 5.0512, | |
| "eval_samples_per_second": 78.991, | |
| "eval_steps_per_second": 9.899, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.338367462158203, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.298, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8186863532409097, | |
| "eval_loss": 0.35223379731178284, | |
| "eval_precision": 0.8138123167155425, | |
| "eval_recall": 0.8243316966721222, | |
| "eval_runtime": 5.0482, | |
| "eval_samples_per_second": 79.038, | |
| "eval_steps_per_second": 9.904, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.702072620391846, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2923, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8263588263588264, | |
| "eval_loss": 0.3374755382537842, | |
| "eval_precision": 0.8289473684210527, | |
| "eval_recall": 0.8239225313693399, | |
| "eval_runtime": 5.0805, | |
| "eval_samples_per_second": 78.536, | |
| "eval_steps_per_second": 9.842, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 6.563529968261719, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2689, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8357422474382676, | |
| "eval_loss": 0.3392201364040375, | |
| "eval_precision": 0.8319228265372551, | |
| "eval_recall": 0.8399709038006911, | |
| "eval_runtime": 5.0499, | |
| "eval_samples_per_second": 79.012, | |
| "eval_steps_per_second": 9.901, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 14.090389251708984, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2686, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8372817261706151, | |
| "eval_loss": 0.34843868017196655, | |
| "eval_precision": 0.8308913308913308, | |
| "eval_recall": 0.8449718130569195, | |
| "eval_runtime": 5.0635, | |
| "eval_samples_per_second": 78.799, | |
| "eval_steps_per_second": 9.875, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.997685432434082, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2726, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8298403801632752, | |
| "eval_loss": 0.3257535398006439, | |
| "eval_precision": 0.8315523576240049, | |
| "eval_recall": 0.8281960356428442, | |
| "eval_runtime": 5.0532, | |
| "eval_samples_per_second": 78.96, | |
| "eval_steps_per_second": 9.895, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 10.811493873596191, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2713, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8341332527115377, | |
| "eval_loss": 0.324627548456192, | |
| "eval_precision": 0.8333132275770553, | |
| "eval_recall": 0.8349699945444626, | |
| "eval_runtime": 5.0517, | |
| "eval_samples_per_second": 78.983, | |
| "eval_steps_per_second": 9.898, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.337926983833313, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2577, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8323529411764705, | |
| "eval_loss": 0.3306790590286255, | |
| "eval_precision": 0.8292704679231822, | |
| "eval_recall": 0.8356973995271868, | |
| "eval_runtime": 5.089, | |
| "eval_samples_per_second": 78.404, | |
| "eval_steps_per_second": 9.825, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.205774784088135, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2519, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8365204824303285, | |
| "eval_loss": 0.3305480182170868, | |
| "eval_precision": 0.8313636363636363, | |
| "eval_recall": 0.8424713584288053, | |
| "eval_runtime": 5.0659, | |
| "eval_samples_per_second": 78.761, | |
| "eval_steps_per_second": 9.87, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 7.098486423492432, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2488, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8246499363520641, | |
| "eval_loss": 0.3233925998210907, | |
| "eval_precision": 0.8246499363520641, | |
| "eval_recall": 0.8246499363520641, | |
| "eval_runtime": 5.0647, | |
| "eval_samples_per_second": 78.781, | |
| "eval_steps_per_second": 9.872, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.508778095245361, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2546, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8391129032258065, | |
| "eval_loss": 0.32470664381980896, | |
| "eval_precision": 0.8345705196182396, | |
| "eval_recall": 0.8442444080741953, | |
| "eval_runtime": 5.0482, | |
| "eval_samples_per_second": 79.038, | |
| "eval_steps_per_second": 9.904, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.409043073654175, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2463, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8306964902709584, | |
| "eval_loss": 0.3204318881034851, | |
| "eval_precision": 0.8306964902709584, | |
| "eval_recall": 0.8306964902709584, | |
| "eval_runtime": 5.0518, | |
| "eval_samples_per_second": 78.982, | |
| "eval_steps_per_second": 9.897, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 8.363502502441406, | |
| "learning_rate": 0.0, | |
| "loss": 0.2458, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8349466368826011, | |
| "eval_loss": 0.32170674204826355, | |
| "eval_precision": 0.8325716845878136, | |
| "eval_recall": 0.8374704491725768, | |
| "eval_runtime": 5.054, | |
| "eval_samples_per_second": 78.947, | |
| "eval_steps_per_second": 9.893, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7609911792720000.0, | |
| "train_loss": 0.3216621422376789, | |
| "train_runtime": 1955.2061, | |
| "train_samples_per_second": 37.213, | |
| "train_steps_per_second": 1.248 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7609911792720000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |