|
{ |
|
"best_metric": 0.9036697247706422, |
|
"best_model_checkpoint": "distilbert-base-uncased-finetuned-sst2/run-0/checkpoint-4210", |
|
"epoch": 2.0, |
|
"eval_steps": 500, |
|
"global_step": 8420, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.1187648456057007, |
|
"grad_norm": 9.763530731201172, |
|
"learning_rate": 2.4130303407723998e-05, |
|
"loss": 0.3459, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.2375296912114014, |
|
"grad_norm": 7.90731954574585, |
|
"learning_rate": 2.3543191402669885e-05, |
|
"loss": 0.2557, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.35629453681710216, |
|
"grad_norm": 7.188111305236816, |
|
"learning_rate": 2.2956079397615776e-05, |
|
"loss": 0.2242, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.4750593824228028, |
|
"grad_norm": 9.416199684143066, |
|
"learning_rate": 2.2368967392561663e-05, |
|
"loss": 0.2281, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.5938242280285035, |
|
"grad_norm": 6.982799053192139, |
|
"learning_rate": 2.1781855387507554e-05, |
|
"loss": 0.2196, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.7125890736342043, |
|
"grad_norm": 3.475909948348999, |
|
"learning_rate": 2.1194743382453438e-05, |
|
"loss": 0.199, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.831353919239905, |
|
"grad_norm": 18.930234909057617, |
|
"learning_rate": 2.0607631377399328e-05, |
|
"loss": 0.1988, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 0.9501187648456056, |
|
"grad_norm": 21.44248390197754, |
|
"learning_rate": 2.0020519372345215e-05, |
|
"loss": 0.1964, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.9036697247706422, |
|
"eval_loss": 0.279630184173584, |
|
"eval_runtime": 1.1981, |
|
"eval_samples_per_second": 727.831, |
|
"eval_steps_per_second": 45.907, |
|
"step": 4210 |
|
}, |
|
{ |
|
"epoch": 1.0688836104513064, |
|
"grad_norm": 6.991998672485352, |
|
"learning_rate": 1.9433407367291103e-05, |
|
"loss": 0.1449, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 1.187648456057007, |
|
"grad_norm": 1.0682291984558105, |
|
"learning_rate": 1.8846295362236993e-05, |
|
"loss": 0.1307, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.3064133016627077, |
|
"grad_norm": 1.4475390911102295, |
|
"learning_rate": 1.825918335718288e-05, |
|
"loss": 0.1248, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 1.4251781472684084, |
|
"grad_norm": 0.4122050702571869, |
|
"learning_rate": 1.767207135212877e-05, |
|
"loss": 0.1288, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 1.5439429928741093, |
|
"grad_norm": 0.35570672154426575, |
|
"learning_rate": 1.7084959347074655e-05, |
|
"loss": 0.1353, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 1.66270783847981, |
|
"grad_norm": 0.9978006482124329, |
|
"learning_rate": 1.6497847342020545e-05, |
|
"loss": 0.1316, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 1.7814726840855108, |
|
"grad_norm": 7.568444728851318, |
|
"learning_rate": 1.5910735336966433e-05, |
|
"loss": 0.1363, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 1.9002375296912115, |
|
"grad_norm": 0.1252310872077942, |
|
"learning_rate": 1.532362333191232e-05, |
|
"loss": 0.1271, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8887614678899083, |
|
"eval_loss": 0.37789419293403625, |
|
"eval_runtime": 1.2, |
|
"eval_samples_per_second": 726.651, |
|
"eval_steps_per_second": 45.832, |
|
"step": 8420 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 21050, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 5, |
|
"save_steps": 500, |
|
"total_flos": 1165689305437104.0, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": { |
|
"learning_rate": 2.471741541277811e-05, |
|
"num_train_epochs": 5, |
|
"per_device_train_batch_size": 16, |
|
"seed": 27 |
|
} |
|
} |
|
|