|
{ |
|
"best_metric": 0.8646788990825688, |
|
"best_model_checkpoint": "bert-base-uncased-finetuned-sst2/run-1/checkpoint-842", |
|
"epoch": 1.0, |
|
"eval_steps": 500, |
|
"global_step": 842, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.59, |
|
"grad_norm": 5.992064476013184, |
|
"learning_rate": 2.846915930651389e-05, |
|
"loss": 0.4119, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.8646788990825688, |
|
"eval_loss": 0.40675637125968933, |
|
"eval_runtime": 2.35, |
|
"eval_samples_per_second": 371.063, |
|
"eval_steps_per_second": 23.404, |
|
"step": 842 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 1684, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 2, |
|
"save_steps": 500, |
|
"total_flos": 62016097970400.0, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": { |
|
"learning_rate": 4.0491608337980904e-05, |
|
"num_train_epochs": 2, |
|
"per_device_train_batch_size": 8, |
|
"seed": 26 |
|
} |
|
} |
|
|