all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -0.8710112571716309,
    "eval_logits/rejected": -1.522813320159912,
    "eval_logps/chosen": -40.126407623291016,
    "eval_logps/rejected": -60.11854934692383,
    "eval_loss": 0.6829047799110413,
    "eval_rewards/accuracies": 0.6299999952316284,
    "eval_rewards/chosen": -0.050214797258377075,
    "eval_rewards/margins": 0.02983289770781994,
    "eval_rewards/rejected": -0.08004770427942276,
    "eval_runtime": 65.4744,
    "eval_samples": 785,
    "eval_samples_per_second": 11.989,
    "eval_steps_per_second": 0.382,
    "train_loss": 0.1692200431117305,
    "train_runtime": 798.9647,
    "train_samples": 17264,
    "train_samples_per_second": 21.608,
    "train_steps_per_second": 0.338
}