all_results.json

{
    "epoch": 2.992914501653283,
    "eval_dpo_loss": 0.6748924255371094,
    "eval_logits": -1.294857382774353,
    "eval_logps": -92.335205078125,
    "eval_loss": 0.7416455745697021,
    "eval_objective": 0.7355117797851562,
    "eval_ranking_simple": 0.5502070188522339,
    "eval_runtime": 372.5292,
    "eval_samples": 5790,
    "eval_samples_per_second": 15.542,
    "eval_steps_per_second": 1.297,
    "total_flos": 0.0,
    "train_loss": 0.4763158837220434,
    "train_runtime": 34658.3644,
    "train_samples": 50802,
    "train_samples_per_second": 4.397,
    "train_steps_per_second": 0.03
}