| { |
| "best_global_step": 60, |
| "best_metric": 0.5672379732131958, |
| "best_model_checkpoint": "Mistral-7B-v0.1/r4/checkpoint-60", |
| "epoch": 0.25069637883008355, |
| "eval_steps": 60, |
| "global_step": 60, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 2.3212984272175365, |
| "epoch": 0.25069637883008355, |
| "grad_norm": 4.5548248291015625, |
| "learning_rate": 7.638888888888889e-05, |
| "loss": 2.8492, |
| "mean_token_accuracy": 0.6425296268943284, |
| "num_tokens": 1474560.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.25069637883008355, |
| "eval_entropy": 0.6633218830669081, |
| "eval_loss": 0.5672379732131958, |
| "eval_mean_token_accuracy": 0.8630788239178719, |
| "eval_num_tokens": 1474560.0, |
| "eval_runtime": 1043.9462, |
| "eval_samples_per_second": 2.357, |
| "eval_steps_per_second": 0.295, |
| "step": 60 |
| } |
| ], |
| "logging_steps": 60, |
| "max_steps": 720, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 60, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 5, |
| "early_stopping_threshold": 0.01 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 0 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 6.293144324800512e+16, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
