| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.9696, | |
| "eval_steps": 500, | |
| "global_step": 3120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.24861662355672434, | |
| "learning_rate": 9.67948717948718e-05, | |
| "loss": 0.3915, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.19740264062427, | |
| "learning_rate": 9.35897435897436e-05, | |
| "loss": 0.2538, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.1982392241525121, | |
| "learning_rate": 9.038461538461538e-05, | |
| "loss": 0.2101, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.2784, | |
| "grad_norm": 0.21511287812438806, | |
| "learning_rate": 8.717948717948718e-05, | |
| "loss": 0.1962, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5984, | |
| "grad_norm": 0.146909709690183, | |
| "learning_rate": 8.397435897435898e-05, | |
| "loss": 0.1965, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9184, | |
| "grad_norm": 0.14827245888652316, | |
| "learning_rate": 8.076923076923078e-05, | |
| "loss": 0.1966, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.2368, | |
| "grad_norm": 0.14007270085399773, | |
| "learning_rate": 7.756410256410257e-05, | |
| "loss": 0.1778, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.5568, | |
| "grad_norm": 0.15165055807508107, | |
| "learning_rate": 7.435897435897436e-05, | |
| "loss": 0.177, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.8768000000000002, | |
| "grad_norm": 0.12261227182687515, | |
| "learning_rate": 7.115384615384616e-05, | |
| "loss": 0.1784, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.1952, | |
| "grad_norm": 0.15927823665260066, | |
| "learning_rate": 6.794871794871795e-05, | |
| "loss": 0.1525, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.5152, | |
| "grad_norm": 0.16576524515888857, | |
| "learning_rate": 6.474358974358975e-05, | |
| "loss": 0.1442, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.8352, | |
| "grad_norm": 0.1633852917628853, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.1474, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.1536, | |
| "grad_norm": 0.16655763454791186, | |
| "learning_rate": 5.833333333333334e-05, | |
| "loss": 0.1212, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.4736, | |
| "grad_norm": 0.17733509242143883, | |
| "learning_rate": 5.512820512820514e-05, | |
| "loss": 0.0984, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.7936, | |
| "grad_norm": 0.16132379416771905, | |
| "learning_rate": 5.192307692307693e-05, | |
| "loss": 0.102, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.112, | |
| "grad_norm": 0.1587067218762505, | |
| "learning_rate": 4.871794871794872e-05, | |
| "loss": 0.0852, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.432, | |
| "grad_norm": 0.15537157580944633, | |
| "learning_rate": 4.5512820512820516e-05, | |
| "loss": 0.054, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.752, | |
| "grad_norm": 0.16095849223245687, | |
| "learning_rate": 4.230769230769231e-05, | |
| "loss": 0.0563, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.0704, | |
| "grad_norm": 0.1282109321169235, | |
| "learning_rate": 3.9102564102564105e-05, | |
| "loss": 0.0501, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.3904, | |
| "grad_norm": 0.12398010094970631, | |
| "learning_rate": 3.58974358974359e-05, | |
| "loss": 0.0262, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.7104, | |
| "grad_norm": 0.12436375901386862, | |
| "learning_rate": 3.269230769230769e-05, | |
| "loss": 0.0269, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 7.0288, | |
| "grad_norm": 0.09305977612863091, | |
| "learning_rate": 2.948717948717949e-05, | |
| "loss": 0.0257, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.3488, | |
| "grad_norm": 0.08781507362398612, | |
| "learning_rate": 2.6282051282051285e-05, | |
| "loss": 0.0132, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.6688, | |
| "grad_norm": 0.08063973370953591, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.0134, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.9888, | |
| "grad_norm": 0.08345401562900914, | |
| "learning_rate": 1.987179487179487e-05, | |
| "loss": 0.0133, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.3072, | |
| "grad_norm": 0.062432848641596865, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0076, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 8.6272, | |
| "grad_norm": 0.05816827694573014, | |
| "learning_rate": 1.3461538461538462e-05, | |
| "loss": 0.0074, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 8.9472, | |
| "grad_norm": 0.0572892709885571, | |
| "learning_rate": 1.0256410256410256e-05, | |
| "loss": 0.0074, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.2656, | |
| "grad_norm": 0.03941120035730241, | |
| "learning_rate": 7.051282051282052e-06, | |
| "loss": 0.0048, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 9.5856, | |
| "grad_norm": 0.04559251834391993, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.0044, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.9056, | |
| "grad_norm": 0.04482437030795512, | |
| "learning_rate": 6.41025641025641e-07, | |
| "loss": 0.0043, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 9.9696, | |
| "step": 3120, | |
| "total_flos": 1.253140166868992e+16, | |
| "train_loss": 0.10079019942965645, | |
| "train_runtime": 45579.7972, | |
| "train_samples_per_second": 17.549, | |
| "train_steps_per_second": 0.068 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.253140166868992e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |