all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -0.786741316318512,
    "eval_logits/rejected": -0.567284107208252,
    "eval_logps/chosen": -312.4064025878906,
    "eval_logps/rejected": -771.6371459960938,
    "eval_loss": 0.27778080105781555,
    "eval_rewards/accuracies": 0.949999988079071,
    "eval_rewards/chosen": -0.10158518701791763,
    "eval_rewards/margins": 2.0500597953796387,
    "eval_rewards/rejected": -2.1516449451446533,
    "eval_runtime": 15.9542,
    "eval_samples": 152,
    "eval_samples_per_second": 9.527,
    "eval_steps_per_second": 0.313,
    "train_loss": 0.3946254751575527,
    "train_runtime": 1910.8543,
    "train_samples": 8578,
    "train_samples_per_second": 4.489,
    "train_steps_per_second": 0.07
}