|
{ |
|
"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": 0.6438709677419355, |
|
"eval_loss": 0.24634820222854614, |
|
"eval_runtime": 1.9852, |
|
"eval_samples_per_second": 1561.588, |
|
"eval_steps_per_second": 32.743, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"grad_norm": 0.6095134615898132, |
|
"learning_rate": 1.685534591194969e-05, |
|
"loss": 0.396, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8406451612903226, |
|
"eval_loss": 0.107815682888031, |
|
"eval_runtime": 2.02, |
|
"eval_samples_per_second": 1534.621, |
|
"eval_steps_per_second": 32.178, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.886774193548387, |
|
"eval_loss": 0.065730981528759, |
|
"eval_runtime": 2.1527, |
|
"eval_samples_per_second": 1440.085, |
|
"eval_steps_per_second": 30.195, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"grad_norm": 0.4446919560432434, |
|
"learning_rate": 1.371069182389937e-05, |
|
"loss": 0.1256, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9067741935483871, |
|
"eval_loss": 0.047372784465551376, |
|
"eval_runtime": 2.1197, |
|
"eval_samples_per_second": 1462.502, |
|
"eval_steps_per_second": 30.665, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"grad_norm": 0.3825507164001465, |
|
"learning_rate": 1.0566037735849058e-05, |
|
"loss": 0.0734, |
|
"step": 1500 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 3180, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"total_flos": 391368939443328.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.7524264974224779, |
|
"num_train_epochs": 10, |
|
"temperature": 4 |
|
} |
|
} |
|
|