0.001_idpo_4iters_iter_1 / train_results.json
ShenaoZhang's picture
Model save
f6d6776 verified
{
"epoch": 1.0,
"train_loss": 0.6248922909007353,
"train_runtime": 1974.7606,
"train_samples": 15283,
"train_samples_per_second": 7.739,
"train_steps_per_second": 0.06
}