zephyr-8b-dpo-full / train_results.json
li-muyang's picture
Model save
f9518e6 verified
{
"epoch": 0.9994767137624281,
"total_flos": 0.0,
"train_loss": 0.5461677596207064,
"train_runtime": 19976.9989,
"train_samples": 61134,
"train_samples_per_second": 3.06,
"train_steps_per_second": 0.048
}