all_results.json

{
    "epoch": 0.9993060374739764,
    "eval_logits/chosen": 0.6849595308303833,
    "eval_logits/rejected": 1.771243691444397,
    "eval_logps/chosen": -536.7197875976562,
    "eval_logps/rejected": -805.5457763671875,
    "eval_loss": 0.28813818097114563,
    "eval_rewards/accuracies": 0.8815789222717285,
    "eval_rewards/chosen": -1.5213572978973389,
    "eval_rewards/margins": 2.4328768253326416,
    "eval_rewards/rejected": -3.9542336463928223,
    "eval_runtime": 387.4996,
    "eval_samples": 8491,
    "eval_samples_per_second": 21.912,
    "eval_steps_per_second": 0.343,
    "total_flos": 0.0,
    "train_loss": 0.1523281472424666,
    "train_runtime": 20936.5354,
    "train_samples": 184443,
    "train_samples_per_second": 8.81,
    "train_steps_per_second": 0.034
}