byt5_en_zul_news / trainer_state.json
Davlan's picture
add MT model
1740c75
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 10.0,
"global_step": 3500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.43,
"learning_rate": 4.2857142857142856e-05,
"loss": 1.4078,
"step": 500
},
{
"epoch": 2.86,
"learning_rate": 3.571428571428572e-05,
"loss": 1.0467,
"step": 1000
},
{
"epoch": 4.29,
"learning_rate": 2.857142857142857e-05,
"loss": 0.9334,
"step": 1500
},
{
"epoch": 5.71,
"learning_rate": 2.1428571428571428e-05,
"loss": 0.8692,
"step": 2000
},
{
"epoch": 7.14,
"learning_rate": 1.4285714285714285e-05,
"loss": 0.8261,
"step": 2500
},
{
"epoch": 8.57,
"learning_rate": 7.142857142857143e-06,
"loss": 0.7967,
"step": 3000
},
{
"epoch": 10.0,
"learning_rate": 0.0,
"loss": 0.782,
"step": 3500
},
{
"epoch": 10.0,
"step": 3500,
"total_flos": 4.868588025962496e+16,
"train_loss": 0.9517042933872768,
"train_runtime": 2753.4267,
"train_samples_per_second": 12.711,
"train_steps_per_second": 1.271
}
],
"max_steps": 3500,
"num_train_epochs": 10,
"total_flos": 4.868588025962496e+16,
"trial_name": null,
"trial_params": null
}