all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.2331926822662354,
    "eval_logits/rejected": -2.2283315658569336,
    "eval_logps/chosen": -34.061973571777344,
    "eval_logps/rejected": -37.566593170166016,
    "eval_loss": 0.7597638368606567,
    "eval_rewards/accuracies": 0.5307309031486511,
    "eval_rewards/chosen": -0.016451681032776833,
    "eval_rewards/margins": 0.013532036915421486,
    "eval_rewards/rejected": -0.029983719810843468,
    "eval_runtime": 145.5395,
    "eval_samples": 343,
    "eval_samples_per_second": 2.357,
    "eval_steps_per_second": 0.295,
    "train_loss": 0.5965218513042896,
    "train_runtime": 3252.2803,
    "train_samples": 3079,
    "train_samples_per_second": 0.947,
    "train_steps_per_second": 0.118
}