all_results.json

{
    "epoch": 3.0,
    "eval_logits/chosen": -2.1214964389801025,
    "eval_logits/rejected": -2.000164031982422,
    "eval_logps/chosen": -265.42364501953125,
    "eval_logps/rejected": -224.6269989013672,
    "eval_loss": 0.5655443072319031,
    "eval_rewards/accuracies": 0.699999988079071,
    "eval_rewards/chosen": -0.07451467216014862,
    "eval_rewards/margins": 0.4583480656147003,
    "eval_rewards/rejected": -0.5328627228736877,
    "eval_runtime": 238.5316,
    "eval_samples": 2000,
    "eval_samples_per_second": 8.385,
    "eval_steps_per_second": 0.524,
    "train_loss": 0.6037390495627379,
    "train_runtime": 33041.7964,
    "train_samples": 61966,
    "train_samples_per_second": 5.626,
    "train_steps_per_second": 0.022
}