|
{ |
|
"best_metric": 0.90448, |
|
"best_model_checkpoint": "checkpoint/roberta-base/imdb_42/checkpoint-7820", |
|
"epoch": 10.0, |
|
"global_step": 7820, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.89504, |
|
"eval_loss": 0.2561816871166229, |
|
"eval_runtime": 15.2562, |
|
"eval_samples_per_second": 1638.677, |
|
"eval_steps_per_second": 6.424, |
|
"step": 782 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"learning_rate": 4.360613810741688e-05, |
|
"loss": 0.31, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.88456, |
|
"eval_loss": 0.27308961749076843, |
|
"eval_runtime": 14.8397, |
|
"eval_samples_per_second": 1684.673, |
|
"eval_steps_per_second": 6.604, |
|
"step": 1564 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"learning_rate": 3.721227621483376e-05, |
|
"loss": 0.1868, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.88904, |
|
"eval_loss": 0.3969782292842865, |
|
"eval_runtime": 14.8251, |
|
"eval_samples_per_second": 1686.329, |
|
"eval_steps_per_second": 6.61, |
|
"step": 2346 |
|
}, |
|
{ |
|
"epoch": 3.84, |
|
"learning_rate": 3.081841432225064e-05, |
|
"loss": 0.1164, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.89612, |
|
"eval_loss": 0.41772395372390747, |
|
"eval_runtime": 14.7947, |
|
"eval_samples_per_second": 1689.796, |
|
"eval_steps_per_second": 6.624, |
|
"step": 3128 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.90036, |
|
"eval_loss": 0.41006845235824585, |
|
"eval_runtime": 14.7809, |
|
"eval_samples_per_second": 1691.378, |
|
"eval_steps_per_second": 6.63, |
|
"step": 3910 |
|
}, |
|
{ |
|
"epoch": 5.12, |
|
"learning_rate": 2.442455242966752e-05, |
|
"loss": 0.0645, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.90156, |
|
"eval_loss": 0.4374634325504303, |
|
"eval_runtime": 14.7992, |
|
"eval_samples_per_second": 1689.277, |
|
"eval_steps_per_second": 6.622, |
|
"step": 4692 |
|
}, |
|
{ |
|
"epoch": 6.39, |
|
"learning_rate": 1.80306905370844e-05, |
|
"loss": 0.041, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.90168, |
|
"eval_loss": 0.5321515202522278, |
|
"eval_runtime": 14.7993, |
|
"eval_samples_per_second": 1689.271, |
|
"eval_steps_per_second": 6.622, |
|
"step": 5474 |
|
}, |
|
{ |
|
"epoch": 7.67, |
|
"learning_rate": 1.163682864450128e-05, |
|
"loss": 0.0283, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.90372, |
|
"eval_loss": 0.5915437936782837, |
|
"eval_runtime": 14.8252, |
|
"eval_samples_per_second": 1686.323, |
|
"eval_steps_per_second": 6.61, |
|
"step": 6256 |
|
}, |
|
{ |
|
"epoch": 8.95, |
|
"learning_rate": 5.242966751918159e-06, |
|
"loss": 0.0167, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.90312, |
|
"eval_loss": 0.6018744707107544, |
|
"eval_runtime": 14.825, |
|
"eval_samples_per_second": 1686.339, |
|
"eval_steps_per_second": 6.61, |
|
"step": 7038 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.90448, |
|
"eval_loss": 0.5969827175140381, |
|
"eval_runtime": 14.9088, |
|
"eval_samples_per_second": 1676.864, |
|
"eval_steps_per_second": 6.573, |
|
"step": 7820 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"step": 7820, |
|
"total_flos": 1.644444096e+16, |
|
"train_loss": 0.09933292054764145, |
|
"train_runtime": 786.2572, |
|
"train_samples_per_second": 317.962, |
|
"train_steps_per_second": 9.946 |
|
} |
|
], |
|
"max_steps": 7820, |
|
"num_train_epochs": 10, |
|
"total_flos": 1.644444096e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|