all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.2444307804107666,
    "eval_logits/rejected": -2.2395598888397217,
    "eval_logps/chosen": -34.009063720703125,
    "eval_logps/rejected": -37.49898147583008,
    "eval_loss": 0.49849173426628113,
    "eval_rewards/accuracies": 0.531146228313446,
    "eval_rewards/chosen": 0.020389538258314133,
    "eval_rewards/margins": 0.00627789506688714,
    "eval_rewards/rejected": 0.014111644588410854,
    "eval_runtime": 145.7275,
    "eval_samples": 343,
    "eval_samples_per_second": 2.354,
    "eval_steps_per_second": 0.295,
    "train_loss": 0.10985468208015739,
    "train_runtime": 628.1848,
    "train_samples": 3079,
    "train_samples_per_second": 4.901,
    "train_steps_per_second": 0.613
}