| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 347, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1440922190201729, | |
| "grad_norm": 0.8574005961418152, | |
| "learning_rate": 9.98886824135084e-05, | |
| "loss": 1.7489, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2881844380403458, | |
| "grad_norm": 0.41624724864959717, | |
| "learning_rate": 9.769082245289474e-05, | |
| "loss": 0.2055, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4322766570605187, | |
| "grad_norm": 0.34527522325515747, | |
| "learning_rate": 9.279617620528497e-05, | |
| "loss": 0.1596, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5763688760806917, | |
| "grad_norm": 0.48223987221717834, | |
| "learning_rate": 8.548152260216614e-05, | |
| "loss": 0.1554, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7204610951008645, | |
| "grad_norm": 0.2791622281074524, | |
| "learning_rate": 7.61604854083727e-05, | |
| "loss": 0.1334, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8645533141210374, | |
| "grad_norm": 0.36355826258659363, | |
| "learning_rate": 6.536014392327366e-05, | |
| "loss": 0.1359, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 694, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5841775439212544e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |