eval_results.json

{
    "epoch": 4.988190836088805,
    "eval_dpo_loss": 1.3281691074371338,
    "eval_logits": -0.6171583533287048,
    "eval_logps": -80.06896209716797,
    "eval_loss": 2.2556095123291016,
    "eval_objective": 2.241896152496338,
    "eval_ranking_idealized": 0.5247933864593506,
    "eval_ranking_idealized_expo": 0.5092975497245789,
    "eval_ranking_simple": 0.5134297609329224,
    "eval_regularize": 2.241896152496338,
    "eval_runtime": 259.631,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.301,
    "eval_steps_per_second": 0.932
}