llama-3-8b-base-new-dpo-hh-helpful-s_star1.0-4xh200-batch-64-20260421-233802/train_results.json

{
    "epoch": 1.0,
    "total_flos": 0.0,
    "train_loss": 0.846195821282447,
    "train_runtime": 1845.654,
    "train_samples": 43598,
    "train_samples_per_second": 23.622,
    "train_steps_per_second": 0.369
}