all_results.json

{
    "epoch": 2.888888888888889,
    "eval_logits/chosen": -2.3964450359344482,
    "eval_logits/rejected": -2.363112688064575,
    "eval_logps/chosen": -258.9571838378906,
    "eval_logps/rejected": -248.72129821777344,
    "eval_loss": 0.6194320917129517,
    "eval_rewards/accuracies": 0.7916666865348816,
    "eval_rewards/chosen": 2.725722551345825,
    "eval_rewards/margins": 2.437701940536499,
    "eval_rewards/rejected": 0.28802117705345154,
    "eval_runtime": 24.6153,
    "eval_samples_per_second": 15.438,
    "eval_steps_per_second": 0.244,
    "total_flos": 459455468666880.0,
    "train_loss": 0.3703473011652629,
    "train_runtime": 1882.4229,
    "train_samples_per_second": 5.449,
    "train_steps_per_second": 0.021
}