all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.3130569458007812,
    "eval_logits/rejected": -2.4918906688690186,
    "eval_logps/chosen": -2.289309024810791,
    "eval_logps/rejected": -2.4444260597229004,
    "eval_loss": 3.0870368480682373,
    "eval_rewards/accuracies": 0.6299999952316284,
    "eval_rewards/chosen": -22.893089294433594,
    "eval_rewards/margins": 1.5511717796325684,
    "eval_rewards/rejected": -24.444263458251953,
    "eval_runtime": 12.8093,
    "eval_samples": 100,
    "eval_samples_per_second": 7.807,
    "eval_steps_per_second": 1.952,
    "total_flos": 0.0,
    "train_loss": 2.970746964952526,
    "train_runtime": 37264.6051,
    "train_samples": 59338,
    "train_samples_per_second": 1.592,
    "train_steps_per_second": 0.796
}