| { | |
| "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": 5.144454002380371, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5655, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7243107769423559, | |
| "eval_f1": 0.6548442906574394, | |
| "eval_loss": 0.5178584456443787, | |
| "eval_precision": 0.6622605615324062, | |
| "eval_recall": 0.6499363520640116, | |
| "eval_runtime": 5.1729, | |
| "eval_samples_per_second": 77.132, | |
| "eval_steps_per_second": 9.666, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.57472562789917, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5048, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7518796992481203, | |
| "eval_f1": 0.7147221319467014, | |
| "eval_loss": 0.49262356758117676, | |
| "eval_precision": 0.707896051974013, | |
| "eval_recall": 0.7269503546099291, | |
| "eval_runtime": 5.0736, | |
| "eval_samples_per_second": 78.643, | |
| "eval_steps_per_second": 9.855, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.7757699489593506, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4529, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7994987468671679, | |
| "eval_f1": 0.7593340973520719, | |
| "eval_loss": 0.430127888917923, | |
| "eval_precision": 0.7580882352941176, | |
| "eval_recall": 0.7606382978723404, | |
| "eval_runtime": 5.0997, | |
| "eval_samples_per_second": 78.24, | |
| "eval_steps_per_second": 9.805, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.7683963775634766, | |
| "learning_rate": 4e-05, | |
| "loss": 0.393, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8220551378446115, | |
| "eval_f1": 0.781440706080221, | |
| "eval_loss": 0.3863331973552704, | |
| "eval_precision": 0.787071406102548, | |
| "eval_recall": 0.776595744680851, | |
| "eval_runtime": 5.1099, | |
| "eval_samples_per_second": 78.084, | |
| "eval_steps_per_second": 9.785, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.220449924468994, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3754, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.800328853302831, | |
| "eval_loss": 0.38681742548942566, | |
| "eval_precision": 0.7891990821678322, | |
| "eval_recall": 0.8208765230041826, | |
| "eval_runtime": 5.0464, | |
| "eval_samples_per_second": 79.067, | |
| "eval_steps_per_second": 9.908, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.7665016651153564, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3455, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8125568285142754, | |
| "eval_loss": 0.36048319935798645, | |
| "eval_precision": 0.8125568285142754, | |
| "eval_recall": 0.8125568285142754, | |
| "eval_runtime": 5.0922, | |
| "eval_samples_per_second": 78.355, | |
| "eval_steps_per_second": 9.819, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.0729649066925049, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3344, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8228567054500919, | |
| "eval_loss": 0.33959081768989563, | |
| "eval_precision": 0.8263351692555232, | |
| "eval_recall": 0.8196490270958356, | |
| "eval_runtime": 5.0467, | |
| "eval_samples_per_second": 79.061, | |
| "eval_steps_per_second": 9.907, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.983180522918701, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3157, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8368501045387564, | |
| "eval_loss": 0.3319353461265564, | |
| "eval_precision": 0.8435514312676942, | |
| "eval_recall": 0.8310147299509002, | |
| "eval_runtime": 5.047, | |
| "eval_samples_per_second": 79.057, | |
| "eval_steps_per_second": 9.907, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 8.283629417419434, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3076, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8209821152299028, | |
| "eval_loss": 0.32728075981140137, | |
| "eval_precision": 0.8284245491932933, | |
| "eval_recall": 0.8146481178396072, | |
| "eval_runtime": 5.0574, | |
| "eval_samples_per_second": 78.895, | |
| "eval_steps_per_second": 9.887, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.000823020935059, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2948, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8448388501742161, | |
| "eval_loss": 0.3238257169723511, | |
| "eval_precision": 0.8551721930610677, | |
| "eval_recall": 0.8363338788870704, | |
| "eval_runtime": 5.0668, | |
| "eval_samples_per_second": 78.749, | |
| "eval_steps_per_second": 9.868, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.291066884994507, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2737, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8395012067578439, | |
| "eval_loss": 0.31985798478126526, | |
| "eval_precision": 0.8473584308763049, | |
| "eval_recall": 0.8327877795962902, | |
| "eval_runtime": 5.0591, | |
| "eval_samples_per_second": 78.868, | |
| "eval_steps_per_second": 9.883, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 9.703360557556152, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2741, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8315338681464504, | |
| "eval_loss": 0.31903842091560364, | |
| "eval_precision": 0.8299369747899159, | |
| "eval_recall": 0.8331969448990726, | |
| "eval_runtime": 5.1101, | |
| "eval_samples_per_second": 78.08, | |
| "eval_steps_per_second": 9.785, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.6231120824813843, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.275, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8458135188208294, | |
| "eval_loss": 0.3146449625492096, | |
| "eval_precision": 0.8627946127946129, | |
| "eval_recall": 0.8331060192762321, | |
| "eval_runtime": 5.0572, | |
| "eval_samples_per_second": 78.898, | |
| "eval_steps_per_second": 9.887, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 7.953589916229248, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2736, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8403508771929824, | |
| "eval_loss": 0.3103943169116974, | |
| "eval_precision": 0.8460491741741742, | |
| "eval_recall": 0.8352882342244045, | |
| "eval_runtime": 5.0659, | |
| "eval_samples_per_second": 78.763, | |
| "eval_steps_per_second": 9.87, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.1334174871444702, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.263, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8401647707947546, | |
| "eval_loss": 0.3111773729324341, | |
| "eval_precision": 0.8393298751432535, | |
| "eval_recall": 0.8410165484633569, | |
| "eval_runtime": 5.0523, | |
| "eval_samples_per_second": 78.975, | |
| "eval_steps_per_second": 9.897, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.466490745544434, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2583, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8461962888779714, | |
| "eval_loss": 0.3085690140724182, | |
| "eval_precision": 0.8453465227094517, | |
| "eval_recall": 0.8470631023822512, | |
| "eval_runtime": 5.0531, | |
| "eval_samples_per_second": 78.961, | |
| "eval_steps_per_second": 9.895, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 7.930349826812744, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2544, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8421640488656195, | |
| "eval_loss": 0.30654484033584595, | |
| "eval_precision": 0.8512313860252005, | |
| "eval_recall": 0.8345608292416803, | |
| "eval_runtime": 5.0492, | |
| "eval_samples_per_second": 79.023, | |
| "eval_steps_per_second": 9.903, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 6.3030524253845215, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2594, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8411818738518064, | |
| "eval_loss": 0.30556628108024597, | |
| "eval_precision": 0.8448542607834644, | |
| "eval_recall": 0.8377886888525186, | |
| "eval_runtime": 5.0901, | |
| "eval_samples_per_second": 78.387, | |
| "eval_steps_per_second": 9.823, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.6230645179748535, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.256, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8421640488656195, | |
| "eval_loss": 0.3042590618133545, | |
| "eval_precision": 0.8512313860252005, | |
| "eval_recall": 0.8345608292416803, | |
| "eval_runtime": 5.0767, | |
| "eval_samples_per_second": 78.595, | |
| "eval_steps_per_second": 9.849, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 8.38916301727295, | |
| "learning_rate": 0.0, | |
| "loss": 0.2515, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8464912280701754, | |
| "eval_loss": 0.3035424053668976, | |
| "eval_precision": 0.8522897897897899, | |
| "eval_recall": 0.8413347881432988, | |
| "eval_runtime": 5.122, | |
| "eval_samples_per_second": 77.9, | |
| "eval_steps_per_second": 9.762, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7609911792720000.0, | |
| "train_loss": 0.32642708059217107, | |
| "train_runtime": 1954.6341, | |
| "train_samples_per_second": 37.224, | |
| "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 | |
| } | |