| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 5940, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.2936731278896332, | |
| "eval_runtime": 114.2823, | |
| "eval_samples_per_second": 41.502, | |
| "eval_steps_per_second": 0.656, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 4.57912457912458e-05, | |
| "loss": 0.5604, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.21221154928207397, | |
| "eval_runtime": 121.2766, | |
| "eval_samples_per_second": 39.109, | |
| "eval_steps_per_second": 0.618, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.05269444361329079, | |
| "eval_runtime": 119.2093, | |
| "eval_samples_per_second": 39.787, | |
| "eval_steps_per_second": 0.629, | |
| "step": 891 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 4.158249158249159e-05, | |
| "loss": 0.1318, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.014130596071481705, | |
| "eval_runtime": 183.4661, | |
| "eval_samples_per_second": 25.852, | |
| "eval_steps_per_second": 0.409, | |
| "step": 1188 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.00778605230152607, | |
| "eval_runtime": 130.1216, | |
| "eval_samples_per_second": 36.451, | |
| "eval_steps_per_second": 0.576, | |
| "step": 1485 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 3.7373737373737376e-05, | |
| "loss": 0.0552, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.004267835058271885, | |
| "eval_runtime": 130.782, | |
| "eval_samples_per_second": 36.266, | |
| "eval_steps_per_second": 0.573, | |
| "step": 1782 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 3.3164983164983165e-05, | |
| "loss": 0.0197, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.002914531622081995, | |
| "eval_runtime": 126.9043, | |
| "eval_samples_per_second": 37.375, | |
| "eval_steps_per_second": 0.591, | |
| "step": 2079 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.0024584291968494654, | |
| "eval_runtime": 129.977, | |
| "eval_samples_per_second": 36.491, | |
| "eval_steps_per_second": 0.577, | |
| "step": 2376 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 2.8956228956228958e-05, | |
| "loss": 0.0104, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.0008079797262325883, | |
| "eval_runtime": 124.2007, | |
| "eval_samples_per_second": 38.188, | |
| "eval_steps_per_second": 0.604, | |
| "step": 2673 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.0014460551319643855, | |
| "eval_runtime": 124.7091, | |
| "eval_samples_per_second": 38.033, | |
| "eval_steps_per_second": 0.601, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 10.1, | |
| "learning_rate": 2.474747474747475e-05, | |
| "loss": 0.0062, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.0016354549443349242, | |
| "eval_runtime": 126.5389, | |
| "eval_samples_per_second": 37.483, | |
| "eval_steps_per_second": 0.593, | |
| "step": 3267 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 2.0538720538720542e-05, | |
| "loss": 0.0045, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.000995765090920031, | |
| "eval_runtime": 131.3803, | |
| "eval_samples_per_second": 36.101, | |
| "eval_steps_per_second": 0.571, | |
| "step": 3564 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.00040818448178470135, | |
| "eval_runtime": 136.6658, | |
| "eval_samples_per_second": 34.705, | |
| "eval_steps_per_second": 0.549, | |
| "step": 3861 | |
| }, | |
| { | |
| "epoch": 13.47, | |
| "learning_rate": 1.632996632996633e-05, | |
| "loss": 0.0031, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.0001601761905476451, | |
| "eval_runtime": 128.2083, | |
| "eval_samples_per_second": 36.994, | |
| "eval_steps_per_second": 0.585, | |
| "step": 4158 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.0001224653678946197, | |
| "eval_runtime": 104.2251, | |
| "eval_samples_per_second": 45.507, | |
| "eval_steps_per_second": 0.72, | |
| "step": 4455 | |
| }, | |
| { | |
| "epoch": 15.15, | |
| "learning_rate": 1.2121212121212122e-05, | |
| "loss": 0.0017, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.00017290345567744225, | |
| "eval_runtime": 131.2825, | |
| "eval_samples_per_second": 36.128, | |
| "eval_steps_per_second": 0.571, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "learning_rate": 7.912457912457913e-06, | |
| "loss": 0.0012, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 4.626844383892603e-05, | |
| "eval_runtime": 98.0039, | |
| "eval_samples_per_second": 48.396, | |
| "eval_steps_per_second": 0.765, | |
| "step": 5049 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 7.748394273221493e-05, | |
| "eval_runtime": 114.2754, | |
| "eval_samples_per_second": 41.505, | |
| "eval_steps_per_second": 0.656, | |
| "step": 5346 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0009, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 3.669047146104276e-05, | |
| "eval_runtime": 103.8038, | |
| "eval_samples_per_second": 45.692, | |
| "eval_steps_per_second": 0.723, | |
| "step": 5643 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 3.5578090319177136e-05, | |
| "eval_runtime": 99.0675, | |
| "eval_samples_per_second": 47.876, | |
| "eval_steps_per_second": 0.757, | |
| "step": 5940 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5940, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 785921954204160.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |