all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.90625,
    "eval_logits/rejected": -3.171875,
    "eval_logps/chosen": -684.0,
    "eval_logps/rejected": -1088.0,
    "eval_loss": 0.22805513441562653,
    "eval_rewards/accuracies": 0.8979725241661072,
    "eval_rewards/chosen": -4.96875,
    "eval_rewards/margins": 4.3125,
    "eval_rewards/rejected": -9.25,
    "eval_runtime": 2936.8516,
    "eval_samples": 104759,
    "eval_samples_per_second": 33.312,
    "eval_steps_per_second": 0.521,
    "total_flos": 0.0,
    "train_loss": 0.3429703987491938,
    "train_runtime": 175730.8191,
    "train_samples": 1849702,
    "train_samples_per_second": 10.526,
    "train_steps_per_second": 0.082
}