all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -3.170811176300049,
    "eval_logits/rejected": -3.164327621459961,
    "eval_logps/chosen": -401.41021728515625,
    "eval_logps/rejected": -417.67462158203125,
    "eval_loss": 0.682475209236145,
    "eval_rewards/accuracies": 0.60546875,
    "eval_rewards/chosen": -0.23294878005981445,
    "eval_rewards/margins": 0.03622151538729668,
    "eval_rewards/rejected": -0.26917028427124023,
    "eval_runtime": 135.9607,
    "eval_samples": 994,
    "eval_samples_per_second": 7.355,
    "eval_steps_per_second": 0.235,
    "train_loss": 0.6714850996549313,
    "train_runtime": 5456.9379,
    "train_samples": 19766,
    "train_samples_per_second": 3.664,
    "train_steps_per_second": 0.057
}