|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 4.716981132075472, |
|
"eval_steps": 500, |
|
"global_step": 1500, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.5741935483870968, |
|
"eval_loss": 0.20105397701263428, |
|
"eval_runtime": 2.0388, |
|
"eval_samples_per_second": 1520.515, |
|
"eval_steps_per_second": 31.882, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"grad_norm": 0.5318933129310608, |
|
"learning_rate": 1.371069182389937e-05, |
|
"loss": 0.3164, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.7974193548387096, |
|
"eval_loss": 0.10539479553699493, |
|
"eval_runtime": 2.2286, |
|
"eval_samples_per_second": 1390.984, |
|
"eval_steps_per_second": 29.166, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8561290322580645, |
|
"eval_loss": 0.07662773132324219, |
|
"eval_runtime": 2.1158, |
|
"eval_samples_per_second": 1465.159, |
|
"eval_steps_per_second": 30.721, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"grad_norm": 0.46274250745773315, |
|
"learning_rate": 7.421383647798742e-06, |
|
"loss": 0.1216, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.8745161290322581, |
|
"eval_loss": 0.0644095242023468, |
|
"eval_runtime": 2.1756, |
|
"eval_samples_per_second": 1424.872, |
|
"eval_steps_per_second": 29.876, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"grad_norm": 0.3605389893054962, |
|
"learning_rate": 1.1320754716981133e-06, |
|
"loss": 0.0876, |
|
"step": 1500 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 1590, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 5, |
|
"save_steps": 500, |
|
"total_flos": 391368939443328.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.3048460649231045, |
|
"num_train_epochs": 5, |
|
"temperature": 15 |
|
} |
|
} |
|
|