eval_results.json

{
    "epoch": 0.999607072691552,
    "eval_log_odds_chosen": 0.31633657217025757,
    "eval_log_odds_ratio": -0.6614669561386108,
    "eval_logits/chosen": -0.2274637073278427,
    "eval_logits/rejected": -0.2321418672800064,
    "eval_logps/chosen": -0.8470349907875061,
    "eval_logps/rejected": -1.0492140054702759,
    "eval_loss": 0.5733794569969177,
    "eval_nll_loss": 0.5669150352478027,
    "eval_rewards/accuracies": 0.6069999933242798,
    "eval_rewards/chosen": -0.00847034901380539,
    "eval_rewards/margins": 0.0020217914134263992,
    "eval_rewards/rejected": -0.010492141358554363,
    "eval_runtime": 285.2409,
    "eval_samples": 1999,
    "eval_samples_per_second": 7.008,
    "eval_steps_per_second": 0.438
}