|
{ |
|
"best_metric": 0.6385659575462341, |
|
"best_model_checkpoint": "hBERTv2_new_no_pretrain_qnli/checkpoint-2457", |
|
"epoch": 8.0, |
|
"global_step": 6552, |
|
"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": 0.6866, |
|
"step": 819 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.5570199524071023, |
|
"eval_loss": 0.6769956350326538, |
|
"eval_runtime": 24.1164, |
|
"eval_samples_per_second": 226.526, |
|
"eval_steps_per_second": 1.783, |
|
"step": 819 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 3.8400000000000005e-05, |
|
"loss": 0.652, |
|
"step": 1638 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.6263957532491305, |
|
"eval_loss": 0.6468318104743958, |
|
"eval_runtime": 23.7137, |
|
"eval_samples_per_second": 230.373, |
|
"eval_steps_per_second": 1.813, |
|
"step": 1638 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 3.76e-05, |
|
"loss": 0.5986, |
|
"step": 2457 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.6351821343584111, |
|
"eval_loss": 0.6385659575462341, |
|
"eval_runtime": 23.8251, |
|
"eval_samples_per_second": 229.296, |
|
"eval_steps_per_second": 1.805, |
|
"step": 2457 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 3.680000000000001e-05, |
|
"loss": 0.5144, |
|
"step": 3276 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.6589785831960461, |
|
"eval_loss": 0.693040132522583, |
|
"eval_runtime": 23.8147, |
|
"eval_samples_per_second": 229.396, |
|
"eval_steps_per_second": 1.806, |
|
"step": 3276 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 3.6e-05, |
|
"loss": 0.4272, |
|
"step": 4095 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.6547684422478491, |
|
"eval_loss": 0.7034492492675781, |
|
"eval_runtime": 23.82, |
|
"eval_samples_per_second": 229.345, |
|
"eval_steps_per_second": 1.805, |
|
"step": 4095 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"learning_rate": 3.52e-05, |
|
"loss": 0.3558, |
|
"step": 4914 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.6637378729635731, |
|
"eval_loss": 0.8170884847640991, |
|
"eval_runtime": 24.2778, |
|
"eval_samples_per_second": 225.02, |
|
"eval_steps_per_second": 1.771, |
|
"step": 4914 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"learning_rate": 3.44e-05, |
|
"loss": 0.2874, |
|
"step": 5733 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.6600768808347062, |
|
"eval_loss": 0.9057153463363647, |
|
"eval_runtime": 23.8916, |
|
"eval_samples_per_second": 228.657, |
|
"eval_steps_per_second": 1.8, |
|
"step": 5733 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"learning_rate": 3.3600000000000004e-05, |
|
"loss": 0.2391, |
|
"step": 6552 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.6445176642870217, |
|
"eval_loss": 1.0089880228042603, |
|
"eval_runtime": 24.1599, |
|
"eval_samples_per_second": 226.119, |
|
"eval_steps_per_second": 1.78, |
|
"step": 6552 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"step": 6552, |
|
"total_flos": 1.2262403072013107e+17, |
|
"train_loss": 0.4701447958474631, |
|
"train_runtime": 11467.4572, |
|
"train_samples_per_second": 456.697, |
|
"train_steps_per_second": 3.571 |
|
} |
|
], |
|
"max_steps": 40950, |
|
"num_train_epochs": 50, |
|
"total_flos": 1.2262403072013107e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|