cuenb-mnli / trainer_state.json
Thinh Truong
first model version
b55bd7b
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"global_step": 36816,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.41,
"learning_rate": 2.592568448500652e-05,
"loss": 0.5569,
"step": 5000
},
{
"epoch": 0.41,
"eval_accuracy": 0.8273051451859399,
"eval_loss": 0.4415350556373596,
"eval_runtime": 67.2256,
"eval_samples_per_second": 146.001,
"eval_steps_per_second": 9.133,
"step": 5000
},
{
"epoch": 0.81,
"learning_rate": 2.185136897001304e-05,
"loss": 0.4598,
"step": 10000
},
{
"epoch": 0.81,
"eval_accuracy": 0.8424859908303617,
"eval_loss": 0.4233930706977844,
"eval_runtime": 67.0843,
"eval_samples_per_second": 146.308,
"eval_steps_per_second": 9.153,
"step": 10000
},
{
"epoch": 1.22,
"learning_rate": 1.7777053455019558e-05,
"loss": 0.3832,
"step": 15000
},
{
"epoch": 1.22,
"eval_accuracy": 0.8474783494651045,
"eval_loss": 0.4397834539413452,
"eval_runtime": 67.4964,
"eval_samples_per_second": 145.415,
"eval_steps_per_second": 9.097,
"step": 15000
},
{
"epoch": 1.63,
"learning_rate": 1.3702737940026075e-05,
"loss": 0.3314,
"step": 20000
},
{
"epoch": 1.63,
"eval_accuracy": 0.8494141619969434,
"eval_loss": 0.41370636224746704,
"eval_runtime": 64.9501,
"eval_samples_per_second": 151.116,
"eval_steps_per_second": 9.453,
"step": 20000
},
{
"epoch": 2.04,
"learning_rate": 9.628422425032594e-06,
"loss": 0.3158,
"step": 25000
},
{
"epoch": 2.04,
"eval_accuracy": 0.8526744778400408,
"eval_loss": 0.448415607213974,
"eval_runtime": 64.7745,
"eval_samples_per_second": 151.526,
"eval_steps_per_second": 9.479,
"step": 25000
},
{
"epoch": 2.44,
"learning_rate": 5.554106910039114e-06,
"loss": 0.2294,
"step": 30000
},
{
"epoch": 2.44,
"eval_accuracy": 0.8552215995924605,
"eval_loss": 0.4471375644207001,
"eval_runtime": 64.5567,
"eval_samples_per_second": 152.037,
"eval_steps_per_second": 9.511,
"step": 30000
},
{
"epoch": 2.85,
"learning_rate": 1.4797913950456324e-06,
"loss": 0.2283,
"step": 35000
},
{
"epoch": 2.85,
"eval_accuracy": 0.8557310239429444,
"eval_loss": 0.4541068375110626,
"eval_runtime": 64.9188,
"eval_samples_per_second": 151.189,
"eval_steps_per_second": 9.458,
"step": 35000
},
{
"epoch": 3.0,
"step": 36816,
"total_flos": 7.74938740264658e+16,
"train_loss": 0.35126533325938025,
"train_runtime": 16355.0125,
"train_samples_per_second": 72.033,
"train_steps_per_second": 2.251
}
],
"max_steps": 36816,
"num_train_epochs": 3,
"total_flos": 7.74938740264658e+16,
"trial_name": null,
"trial_params": null
}