all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -1.2132024765014648,
    "eval_logits/rejected": -1.2924575805664062,
    "eval_logps/chosen": -396.52020263671875,
    "eval_logps/rejected": -409.31964111328125,
    "eval_loss": 0.4216844439506531,
    "eval_rewards/accuracies": 0.8134920597076416,
    "eval_rewards/chosen": -0.1933162659406662,
    "eval_rewards/margins": 2.1035311222076416,
    "eval_rewards/rejected": -2.296847343444824,
    "eval_runtime": 398.9749,
    "eval_samples": 2000,
    "eval_samples_per_second": 5.013,
    "eval_steps_per_second": 0.158,
    "train_loss": 0.4461688995361328,
    "train_runtime": 44067.2139,
    "train_samples": 58917,
    "train_samples_per_second": 1.337,
    "train_steps_per_second": 0.021
}