| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.16, |
| "eval_steps": 500, |
| "global_step": 50, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "alignment_loss": 0.957948, |
| "epoch": 0.032, |
| "grad_norm": 13216.34765625, |
| "learning_rate": 9e-06, |
| "loss": 604.8516, |
| "mean_token_accuracy": 0.6210215302184224, |
| "num_tokens": 988428.0, |
| "step": 10, |
| "teacher_ce_loss": 39.105175 |
| }, |
| { |
| "alignment_loss": 0.946621, |
| "epoch": 0.064, |
| "grad_norm": 3091.22509765625, |
| "learning_rate": 9.853896103896104e-06, |
| "loss": 171.2203, |
| "mean_token_accuracy": 0.7685627818107605, |
| "num_tokens": 1972507.0, |
| "step": 20, |
| "teacher_ce_loss": 10.471479 |
| }, |
| { |
| "alignment_loss": 0.94924, |
| "epoch": 0.096, |
| "grad_norm": 3218.343017578125, |
| "learning_rate": 9.691558441558442e-06, |
| "loss": 115.0068, |
| "mean_token_accuracy": 0.7847720004618168, |
| "num_tokens": 2964436.0, |
| "step": 30, |
| "teacher_ce_loss": 7.006755 |
| }, |
| { |
| "alignment_loss": 0.938964, |
| "epoch": 0.128, |
| "grad_norm": 2620.491455078125, |
| "learning_rate": 9.52922077922078e-06, |
| "loss": 83.4615, |
| "mean_token_accuracy": 0.8658171582967042, |
| "num_tokens": 3939005.0, |
| "step": 40, |
| "teacher_ce_loss": 5.536875 |
| }, |
| { |
| "alignment_loss": 0.9248, |
| "epoch": 0.16, |
| "grad_norm": 1182.9658203125, |
| "learning_rate": 9.366883116883117e-06, |
| "loss": 56.0003, |
| "mean_token_accuracy": 0.8695644401013851, |
| "num_tokens": 4920499.0, |
| "step": 50, |
| "teacher_ce_loss": 3.622461 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 626, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 50, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.2867793587837338e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
