all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -3.4721531867980957,
    "eval_logits/rejected": -3.3256866931915283,
    "eval_logps/chosen": -254.42779541015625,
    "eval_logps/rejected": -714.5497436523438,
    "eval_loss": 0.00404787715524435,
    "eval_rewards/accuracies": 0.9983165264129639,
    "eval_rewards/chosen": 0.13777638971805573,
    "eval_rewards/margins": 29.16945457458496,
    "eval_rewards/rejected": -29.03167724609375,
    "eval_runtime": 460.1921,
    "eval_samples": 9500,
    "eval_samples_per_second": 20.644,
    "eval_steps_per_second": 0.645,
    "train_loss": 0.0552501158515984,
    "train_runtime": 31216.715,
    "train_samples": 188284,
    "train_samples_per_second": 6.032,
    "train_steps_per_second": 0.094
}