|
{ |
|
"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": 0.597741935483871, |
|
"eval_loss": 0.212155282497406, |
|
"eval_runtime": 2.0005, |
|
"eval_samples_per_second": 1549.601, |
|
"eval_steps_per_second": 32.492, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"grad_norm": 0.5458779335021973, |
|
"learning_rate": 1.4758909853249476e-05, |
|
"loss": 0.3364, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8183870967741935, |
|
"eval_loss": 0.10624609142541885, |
|
"eval_runtime": 2.1529, |
|
"eval_samples_per_second": 1439.925, |
|
"eval_steps_per_second": 30.192, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8690322580645161, |
|
"eval_loss": 0.07340904325246811, |
|
"eval_runtime": 2.122, |
|
"eval_samples_per_second": 1460.903, |
|
"eval_steps_per_second": 30.632, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"grad_norm": 0.45546621084213257, |
|
"learning_rate": 9.517819706498952e-06, |
|
"loss": 0.1227, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.8870967741935484, |
|
"eval_loss": 0.05869447812438011, |
|
"eval_runtime": 2.0818, |
|
"eval_samples_per_second": 1489.069, |
|
"eval_steps_per_second": 31.222, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"grad_norm": 0.35866251587867737, |
|
"learning_rate": 4.276729559748428e-06, |
|
"loss": 0.0827, |
|
"step": 1500 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 1908, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 6, |
|
"save_steps": 500, |
|
"total_flos": 391368939443328.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.4627351834503085, |
|
"num_train_epochs": 6, |
|
"temperature": 8 |
|
} |
|
} |
|
|