all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -1.9709315299987793,
    "eval_logits/rejected": -1.8150750398635864,
    "eval_logps/chosen": -266.7596130371094,
    "eval_logps/rejected": -254.18397521972656,
    "eval_loss": 0.057394202798604965,
    "eval_rewards/accuracies": 0.3765000104904175,
    "eval_rewards/chosen": -0.0003101456386502832,
    "eval_rewards/margins": -0.004089393652975559,
    "eval_rewards/rejected": 0.003779248334467411,
    "eval_runtime": 709.1028,
    "eval_samples": 2000,
    "eval_samples_per_second": 2.82,
    "eval_steps_per_second": 1.41,
    "train_loss": 0.0016641309279948472,
    "train_runtime": 2302.1859,
    "train_samples": 61135,
    "train_samples_per_second": 0.434,
    "train_steps_per_second": 0.109
}