all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.5889480113983154,
    "eval_logits/rejected": -2.5375468730926514,
    "eval_logps/chosen": -302.0460205078125,
    "eval_logps/rejected": -314.4243469238281,
    "eval_loss": 0.5835465788841248,
    "eval_rewards/accuracies": 0.7105000019073486,
    "eval_rewards/chosen": -0.1485649198293686,
    "eval_rewards/margins": 0.33676549792289734,
    "eval_rewards/rejected": -0.4853304326534271,
    "eval_runtime": 1163.5942,
    "eval_samples": 2000,
    "eval_samples_per_second": 1.719,
    "eval_steps_per_second": 0.215,
    "train_loss": 0.6024045768410584,
    "train_runtime": 59158.061,
    "train_samples": 61135,
    "train_samples_per_second": 1.033,
    "train_steps_per_second": 0.129
}