0.001_optm_idpo_4iters_iter_1 / all_results.json
ShenaoZhang's picture
Model save
acddb6b verified
{
"epoch": 1.0,
"train_loss": 0.6038635838933352,
"train_runtime": 3119.4458,
"train_samples": 15283,
"train_samples_per_second": 4.899,
"train_steps_per_second": 0.038
}