llama-8b-dpo-full / train_results.json
fenguhao's picture
Model save
b002b79 verified
{
"epoch": 1.0,
"train_loss": 0.6480738864519209,
"train_runtime": 26013.0665,
"train_samples": 61135,
"train_samples_per_second": 2.35,
"train_steps_per_second": 0.073
}