eval_results.json

{
    "epoch": 1.995276334435522,
    "eval_dpo_loss": 0.7840022444725037,
    "eval_logits": -1.9736981391906738,
    "eval_logps": -120.04292297363281,
    "eval_loss": 0.7555916905403137,
    "eval_objective": 0.7840022444725037,
    "eval_ranking_idealized": 0.5888429880142212,
    "eval_ranking_idealized_expo": 0.5103305578231812,
    "eval_ranking_simple": 0.5402892827987671,
    "eval_regularize": 0.7840022444725037,
    "eval_runtime": 259.1615,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.341,
    "eval_steps_per_second": 0.934
}