|
{ |
|
"best_metric": 0.786692759295499, |
|
"best_model_checkpoint": "tiny-bert-sst2-distilled/run-21/checkpoint-192", |
|
"epoch": 1.0, |
|
"eval_steps": 500, |
|
"global_step": 192, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 5.036266803741455, |
|
"learning_rate": 7.741005485591573e-05, |
|
"loss": 0.556, |
|
"step": 192 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.786692759295499, |
|
"eval_f1": 0.8080985915492958, |
|
"eval_loss": 0.4501095712184906, |
|
"eval_mcc": 0.5882099387037968, |
|
"eval_precision": 0.7344, |
|
"eval_recall": 0.898238747553816, |
|
"eval_runtime": 67.0331, |
|
"eval_samples_per_second": 15.246, |
|
"eval_steps_per_second": 0.477, |
|
"step": 192 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 960, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 5, |
|
"save_steps": 500, |
|
"total_flos": 235695197280.0, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.8653499588119579, |
|
"learning_rate": 9.676256856989465e-05, |
|
"num_train_epochs": 5, |
|
"per_device_train_batch_size": 16, |
|
"temperature": 45 |
|
} |
|
} |
|
|