0.0001_idpo_same_3iters_iter_1 / train_results.json
ShenaoZ's picture
Model save
47420b0 verified
{
"epoch": 1.0,
"train_loss": 0.6025606791178385,
"train_runtime": 4000.6914,
"train_samples": 20378,
"train_samples_per_second": 5.094,
"train_steps_per_second": 0.04
}