all_results.json

{
    "epoch": 3.0,
    "eval_logits/chosen": -2.065433979034424,
    "eval_logits/rejected": -1.9405803680419922,
    "eval_logps/chosen": -266.1706848144531,
    "eval_logps/rejected": -228.30780029296875,
    "eval_loss": 0.5255534052848816,
    "eval_rewards/accuracies": 0.7419999837875366,
    "eval_rewards/chosen": -0.15385985374450684,
    "eval_rewards/margins": 0.7486297488212585,
    "eval_rewards/rejected": -0.9024895429611206,
    "eval_runtime": 601.6531,
    "eval_samples": 2000,
    "eval_samples_per_second": 3.324,
    "eval_steps_per_second": 0.208,
    "train_loss": 0.5642068754707158,
    "train_runtime": 89225.6094,
    "train_samples": 61966,
    "train_samples_per_second": 2.083,
    "train_steps_per_second": 0.033
}