roberta-large_boolq / trainer_state.json
nfliu's picture
End of training
dfd0426
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"eval_steps": 250,
"global_step": 1475,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.85,
"eval_accuracy": 0.8024464831804281,
"eval_loss": 0.45075055956840515,
"eval_runtime": 33.5258,
"eval_samples_per_second": 97.537,
"eval_steps_per_second": 12.2,
"step": 250
},
{
"epoch": 1.69,
"learning_rate": 6.637288135593221e-06,
"loss": 0.5086,
"step": 500
},
{
"epoch": 1.69,
"eval_accuracy": 0.8501529051987767,
"eval_loss": 0.3660411536693573,
"eval_runtime": 33.4804,
"eval_samples_per_second": 97.669,
"eval_steps_per_second": 12.216,
"step": 500
},
{
"epoch": 2.54,
"eval_accuracy": 0.8507645259938837,
"eval_loss": 0.40917864441871643,
"eval_runtime": 33.4889,
"eval_samples_per_second": 97.644,
"eval_steps_per_second": 12.213,
"step": 750
},
{
"epoch": 3.39,
"learning_rate": 3.2610169491525428e-06,
"loss": 0.2387,
"step": 1000
},
{
"epoch": 3.39,
"eval_accuracy": 0.8553516819571866,
"eval_loss": 0.49750199913978577,
"eval_runtime": 33.4928,
"eval_samples_per_second": 97.633,
"eval_steps_per_second": 12.212,
"step": 1000
},
{
"epoch": 4.24,
"eval_accuracy": 0.8525993883792049,
"eval_loss": 0.5576593279838562,
"eval_runtime": 33.4993,
"eval_samples_per_second": 97.614,
"eval_steps_per_second": 12.209,
"step": 1250
},
{
"epoch": 5.0,
"step": 1475,
"total_flos": 4.392658481046528e+16,
"train_loss": 0.2942116430250265,
"train_runtime": 1876.9877,
"train_samples_per_second": 25.112,
"train_steps_per_second": 0.786
}
],
"logging_steps": 500,
"max_steps": 1475,
"num_train_epochs": 5,
"save_steps": 250,
"total_flos": 4.392658481046528e+16,
"trial_name": null,
"trial_params": null
}