0.001_idpo_4iters_iter_2 / train_results.json
ShenaoZhang's picture
Model save
7d4ee2d verified
raw
history blame
192 Bytes
{
"epoch": 1.0,
"train_loss": 0.672575786334126,
"train_runtime": 1976.2133,
"train_samples": 15283,
"train_samples_per_second": 7.733,
"train_steps_per_second": 0.06
}