|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 4.0, |
|
"eval_steps": 500, |
|
"global_step": 16, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.25, |
|
"grad_norm": 6.53125, |
|
"learning_rate": 0.0001, |
|
"loss": 6.4013, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"grad_norm": 3.84375, |
|
"learning_rate": 0.000178183148246803, |
|
"loss": 6.2911, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 2.5, |
|
"grad_norm": 2.765625, |
|
"learning_rate": 7.774790660436858e-05, |
|
"loss": 6.0613, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 3.75, |
|
"grad_norm": 2.5625, |
|
"learning_rate": 2.5072087818176382e-06, |
|
"loss": 5.9747, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 16, |
|
"total_flos": 535126081536000.0, |
|
"train_loss": 6.083620756864548, |
|
"train_runtime": 28.0153, |
|
"train_samples_per_second": 35.838, |
|
"train_steps_per_second": 0.571 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 16, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 4, |
|
"save_steps": 100, |
|
"total_flos": 535126081536000.0, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|