all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.7815444469451904,
    "eval_logits/rejected": -2.673560619354248,
    "eval_logps/chosen": -230.3512725830078,
    "eval_logps/rejected": -263.2083740234375,
    "eval_loss": 0.13737119734287262,
    "eval_rewards/accuracies": 0.8670212626457214,
    "eval_rewards/chosen": -4.81953239440918,
    "eval_rewards/margins": 6.819788932800293,
    "eval_rewards/rejected": -11.639321327209473,
    "eval_runtime": 240.1885,
    "eval_samples": 1495,
    "eval_samples_per_second": 6.224,
    "eval_steps_per_second": 0.196,
    "train_loss": 0.18541636852003773,
    "train_runtime": 4261.3779,
    "train_samples": 12879,
    "train_samples_per_second": 3.022,
    "train_steps_per_second": 0.189
}