all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.192172050476074,
    "eval_logits/rejected": -2.095205545425415,
    "eval_logps/chosen": -147.71951293945312,
    "eval_logps/rejected": -158.54696655273438,
    "eval_loss": 0.5194684863090515,
    "eval_rewards/accuracies": 0.7358490824699402,
    "eval_rewards/chosen": -1.382088541984558,
    "eval_rewards/margins": 1.068930983543396,
    "eval_rewards/rejected": -2.451019525527954,
    "eval_runtime": 422.7886,
    "eval_samples": 418,
    "eval_samples_per_second": 0.989,
    "eval_steps_per_second": 0.125,
    "train_loss": 0.5536013367607837,
    "train_runtime": 17040.3461,
    "train_samples": 8130,
    "train_samples_per_second": 0.477,
    "train_steps_per_second": 0.06
}