|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.0, |
|
"eval_steps": 500, |
|
"global_step": 250, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 0.48472875356674194, |
|
"learning_rate": 0.0003, |
|
"loss": 2.3219, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 0.4583953619003296, |
|
"learning_rate": 0.0003, |
|
"loss": 2.2487, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.5168253968253969, |
|
"eval_loss": 2.311781167984009, |
|
"eval_runtime": 5.4425, |
|
"eval_samples_per_second": 91.87, |
|
"eval_steps_per_second": 11.576, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_exact_match": 18.6, |
|
"eval_f1": 26.999121042356347, |
|
"step": 250 |
|
} |
|
], |
|
"logging_steps": 100, |
|
"max_steps": 12500, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 50, |
|
"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": 6467693003669504.0, |
|
"train_batch_size": 1, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|