| { |
| "best_metric": 1.7990472316741943, |
| "best_model_checkpoint": "./news/context/checkpoint-500", |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 1431, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0482180293501049, |
| "grad_norm": 3.671875, |
| "learning_rate": 3.639130983978101e-05, |
| "loss": 0.8185, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0482180293501049, |
| "eval_accuracy": 0.7391231816361937, |
| "eval_loss": 1.7990472316741943, |
| "eval_runtime": 3.533, |
| "eval_samples_per_second": 66.799, |
| "eval_steps_per_second": 1.698, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.0964360587002098, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.0380955189271334e-05, |
| "loss": 0.2771, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.0964360587002098, |
| "eval_accuracy": 0.748465234218604, |
| "eval_loss": 1.8269299268722534, |
| "eval_runtime": 1.671, |
| "eval_samples_per_second": 141.234, |
| "eval_steps_per_second": 3.591, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 1431, |
| "total_flos": 3.4357905384800256e+16, |
| "train_loss": 0.42274059239673417, |
| "train_runtime": 2538.8478, |
| "train_samples_per_second": 2.255, |
| "train_steps_per_second": 0.564 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 1431, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 3, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 0 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.4357905384800256e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
