all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.2439303398132324,
    "eval_logits/rejected": -2.2390527725219727,
    "eval_logps/chosen": -34.02507400512695,
    "eval_logps/rejected": -37.508995056152344,
    "eval_loss": 0.6933426260948181,
    "eval_rewards/accuracies": 0.5132890343666077,
    "eval_rewards/chosen": 0.0028422565665096045,
    "eval_rewards/margins": 0.0005549592897295952,
    "eval_rewards/rejected": 0.002287296811118722,
    "eval_runtime": 145.8223,
    "eval_samples": 343,
    "eval_samples_per_second": 2.352,
    "eval_steps_per_second": 0.295,
    "train_loss": 0.691356255791404,
    "train_runtime": 3254.7157,
    "train_samples": 3079,
    "train_samples_per_second": 0.946,
    "train_steps_per_second": 0.118
}