all_results.json

{
    "epoch": 0.9997167941093175,
    "eval_logits/chosen": -0.2702828347682953,
    "eval_logits/rejected": -0.28983867168426514,
    "eval_logps/chosen": -703.930419921875,
    "eval_logps/rejected": -742.163330078125,
    "eval_loss": 0.5928736925125122,
    "eval_rewards/accuracies": 0.5694444179534912,
    "eval_rewards/chosen": -2.3028335571289062,
    "eval_rewards/margins": 0.34491339325904846,
    "eval_rewards/rejected": -2.647747039794922,
    "eval_runtime": 67.4846,
    "eval_samples": 571,
    "eval_samples_per_second": 8.461,
    "eval_steps_per_second": 0.533,
    "total_flos": 0.0,
    "train_loss": 0.6297533516167919,
    "train_runtime": 10521.5341,
    "train_samples": 56489,
    "train_samples_per_second": 5.369,
    "train_steps_per_second": 0.168
}