all_results.json

{
    "epoch": 2.998953427524856,
    "eval_logits/chosen": -8.875,
    "eval_logits/rejected": -7.5625,
    "eval_logps/chosen": -364.0,
    "eval_logps/rejected": -444.0,
    "eval_loss": 2.1908750534057617,
    "eval_nll_loss": 1.171875,
    "eval_rewards/accuracies": 0.59375,
    "eval_rewards/chosen": -3.640625,
    "eval_rewards/margins": 0.80078125,
    "eval_rewards/rejected": -4.4375,
    "eval_runtime": 23.576,
    "eval_samples": 2000,
    "eval_samples_per_second": 84.832,
    "eval_steps_per_second": 1.357,
    "total_flos": 0.0,
    "train_loss": 0.0,
    "train_runtime": 0.0505,
    "train_samples": 61134,
    "train_samples_per_second": 3630511.85,
    "train_steps_per_second": 56713.757
}