|
{ |
|
"best_metric": 0.7603706972639012, |
|
"best_model_checkpoint": "./outputs/deberta-large-output/vua_metaphor_8_0.01_0.00002_03-21-23_22-02/checkpoint-3612", |
|
"epoch": 3.999584889995849, |
|
"global_step": 4816, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 1.595758780649437e-05, |
|
"loss": 0.1306, |
|
"step": 1204 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.9674111647440617, |
|
"eval_f1": 0.7417333045169656, |
|
"eval_loss": 0.08462577313184738, |
|
"eval_precision": 0.7609756097560976, |
|
"eval_recall": 0.7234401349072512, |
|
"eval_runtime": 13.8793, |
|
"eval_samples_per_second": 173.64, |
|
"eval_steps_per_second": 21.759, |
|
"step": 1204 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 1.0638391870996245e-05, |
|
"loss": 0.0527, |
|
"step": 2408 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.9700564509531212, |
|
"eval_f1": 0.7538116591928251, |
|
"eval_loss": 0.08208993822336197, |
|
"eval_precision": 0.8050766283524904, |
|
"eval_recall": 0.7086846543001686, |
|
"eval_runtime": 13.6308, |
|
"eval_samples_per_second": 176.806, |
|
"eval_steps_per_second": 22.156, |
|
"step": 2408 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 5.319195935498122e-06, |
|
"loss": 0.0189, |
|
"step": 3612 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.9703837028552729, |
|
"eval_f1": 0.7603706972639012, |
|
"eval_loss": 0.11565817892551422, |
|
"eval_precision": 0.7976851851851852, |
|
"eval_recall": 0.7263912310286678, |
|
"eval_runtime": 15.1081, |
|
"eval_samples_per_second": 159.517, |
|
"eval_steps_per_second": 19.989, |
|
"step": 3612 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.006, |
|
"step": 4816 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9703564318634269, |
|
"eval_f1": 0.7563326608383771, |
|
"eval_loss": 0.15718454122543335, |
|
"eval_precision": 0.8075634274772618, |
|
"eval_recall": 0.7112141652613828, |
|
"eval_runtime": 13.5443, |
|
"eval_samples_per_second": 177.934, |
|
"eval_steps_per_second": 22.297, |
|
"step": 4816 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 4816, |
|
"total_flos": 2366884617110520.0, |
|
"train_loss": 0.052028106791632514, |
|
"train_runtime": 1649.8443, |
|
"train_samples_per_second": 23.355, |
|
"train_steps_per_second": 2.919 |
|
} |
|
], |
|
"max_steps": 4816, |
|
"num_train_epochs": 4, |
|
"total_flos": 2366884617110520.0, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|