all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.2310752868652344,
    "eval_logits/rejected": -2.226254940032959,
    "eval_logps/chosen": -33.75144958496094,
    "eval_logps/rejected": -37.262733459472656,
    "eval_loss": 0.49370110034942627,
    "eval_rewards/accuracies": 0.5274086594581604,
    "eval_rewards/chosen": 0.25479215383529663,
    "eval_rewards/margins": 0.02629559487104416,
    "eval_rewards/rejected": 0.22849655151367188,
    "eval_runtime": 145.6018,
    "eval_samples": 343,
    "eval_samples_per_second": 2.356,
    "eval_steps_per_second": 0.295,
    "train_loss": 0.447351616079157,
    "train_runtime": 3253.4458,
    "train_samples": 3079,
    "train_samples_per_second": 0.946,
    "train_steps_per_second": 0.118
}