|
{ |
|
"best_metric": 0.6491575241088867, |
|
"best_model_checkpoint": "sa_BERT_24_qnli/checkpoint-2184", |
|
"epoch": 7.0, |
|
"global_step": 7644, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 3.9200000000000004e-05, |
|
"loss": 0.6745, |
|
"step": 1092 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.5817316492769541, |
|
"eval_loss": 0.6685570478439331, |
|
"eval_runtime": 11.3958, |
|
"eval_samples_per_second": 479.388, |
|
"eval_steps_per_second": 5.002, |
|
"step": 1092 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 3.8400000000000005e-05, |
|
"loss": 0.6338, |
|
"step": 2184 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.6218195130880468, |
|
"eval_loss": 0.6491575241088867, |
|
"eval_runtime": 11.4075, |
|
"eval_samples_per_second": 478.897, |
|
"eval_steps_per_second": 4.997, |
|
"step": 2184 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 3.76e-05, |
|
"loss": 0.5909, |
|
"step": 3276 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.6251144060040271, |
|
"eval_loss": 0.6560329794883728, |
|
"eval_runtime": 11.3972, |
|
"eval_samples_per_second": 479.329, |
|
"eval_steps_per_second": 5.001, |
|
"step": 3276 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 3.680000000000001e-05, |
|
"loss": 0.5407, |
|
"step": 4368 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.6269449020684605, |
|
"eval_loss": 0.724582850933075, |
|
"eval_runtime": 11.3947, |
|
"eval_samples_per_second": 479.433, |
|
"eval_steps_per_second": 5.002, |
|
"step": 4368 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 3.6e-05, |
|
"loss": 0.4732, |
|
"step": 5460 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.6421380194032583, |
|
"eval_loss": 0.6612172722816467, |
|
"eval_runtime": 11.3971, |
|
"eval_samples_per_second": 479.333, |
|
"eval_steps_per_second": 5.001, |
|
"step": 5460 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"learning_rate": 3.52e-05, |
|
"loss": 0.3999, |
|
"step": 6552 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.6410397217645982, |
|
"eval_loss": 0.7505705952644348, |
|
"eval_runtime": 11.393, |
|
"eval_samples_per_second": 479.506, |
|
"eval_steps_per_second": 5.003, |
|
"step": 6552 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"learning_rate": 3.44e-05, |
|
"loss": 0.3203, |
|
"step": 7644 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.6306058941973275, |
|
"eval_loss": 0.9162063598632812, |
|
"eval_runtime": 11.3864, |
|
"eval_samples_per_second": 479.784, |
|
"eval_steps_per_second": 5.006, |
|
"step": 7644 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"step": 7644, |
|
"total_flos": 9.645664719024947e+16, |
|
"train_loss": 0.5190431563907865, |
|
"train_runtime": 4816.429, |
|
"train_samples_per_second": 1087.351, |
|
"train_steps_per_second": 11.336 |
|
} |
|
], |
|
"max_steps": 54600, |
|
"num_train_epochs": 50, |
|
"total_flos": 9.645664719024947e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|