0.001_idpo_4iters_iter_4 / train_results.json
ShenaoZhang's picture
Model save
2f15e93 verified
{
"epoch": 1.0,
"train_loss": 0.6447748657034225,
"train_runtime": 1974.0915,
"train_samples": 15283,
"train_samples_per_second": 7.742,
"train_steps_per_second": 0.06
}