all_results.json

{
    "epoch": 0.9997382884061764,
    "eval_logits/chosen": -0.41565001010894775,
    "eval_logits/rejected": -0.3842959403991699,
    "eval_logps/chosen": -292.3936462402344,
    "eval_logps/rejected": -284.65582275390625,
    "eval_loss": 0.5193021297454834,
    "eval_rewards/accuracies": 0.7279999852180481,
    "eval_rewards/chosen": 0.015352733433246613,
    "eval_rewards/margins": 0.813278079032898,
    "eval_rewards/rejected": -0.7979253530502319,
    "eval_runtime": 347.9856,
    "eval_samples": 2000,
    "eval_samples_per_second": 5.747,
    "eval_steps_per_second": 0.359,
    "total_flos": 0.0,
    "train_loss": 0.5431846207973221,
    "train_runtime": 19104.9346,
    "train_samples": 61135,
    "train_samples_per_second": 3.2,
    "train_steps_per_second": 0.05
}