simple_online_epoch_2_dpo_iter_6 / train_results.json
YYYYYYibo's picture
Model save
64526f6 verified
{
"epoch": 1.0,
"train_loss": 0.6911558170377472,
"train_runtime": 23474.6733,
"train_samples": 21135,
"train_samples_per_second": 0.9,
"train_steps_per_second": 0.007
}