byt5-base-es_maz / trainer_state.json
mekjr1's picture
End of training
d349f61
raw
history blame
1.45 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 10.0,
"global_step": 3930,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.27,
"learning_rate": 4.3638676844783716e-05,
"loss": 2.4977,
"step": 500
},
{
"epoch": 2.54,
"learning_rate": 3.727735368956743e-05,
"loss": 1.1354,
"step": 1000
},
{
"epoch": 3.82,
"learning_rate": 3.091603053435115e-05,
"loss": 1.0127,
"step": 1500
},
{
"epoch": 5.09,
"learning_rate": 2.455470737913486e-05,
"loss": 0.9531,
"step": 2000
},
{
"epoch": 6.36,
"learning_rate": 1.8193384223918574e-05,
"loss": 0.9053,
"step": 2500
},
{
"epoch": 7.63,
"learning_rate": 1.1832061068702292e-05,
"loss": 0.8799,
"step": 3000
},
{
"epoch": 8.91,
"learning_rate": 5.470737913486006e-06,
"loss": 0.8521,
"step": 3500
},
{
"epoch": 10.0,
"step": 3930,
"total_flos": 2189095450214400.0,
"train_loss": 1.140592199427481,
"train_runtime": 504.5513,
"train_samples_per_second": 124.447,
"train_steps_per_second": 7.789
}
],
"max_steps": 3930,
"num_train_epochs": 10,
"total_flos": 2189095450214400.0,
"trial_name": null,
"trial_params": null
}