all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.2439348697662354,
    "eval_logits/rejected": -2.2390594482421875,
    "eval_logps/chosen": -34.016693115234375,
    "eval_logps/rejected": -37.499900817871094,
    "eval_loss": 0.9991074204444885,
    "eval_rewards/accuracies": 0.5365448594093323,
    "eval_rewards/chosen": 0.01607290655374527,
    "eval_rewards/margins": 0.0010251295752823353,
    "eval_rewards/rejected": 0.015047776512801647,
    "eval_runtime": 146.1502,
    "eval_samples": 343,
    "eval_samples_per_second": 2.347,
    "eval_steps_per_second": 0.294,
    "train_loss": 0.21873634759481853,
    "train_runtime": 627.4596,
    "train_samples": 3079,
    "train_samples_per_second": 4.907,
    "train_steps_per_second": 0.614
}