all_results.json

{
    "epoch": 0.9996020692399522,
    "eval_logits/chosen": -2.402043581008911,
    "eval_logits/rejected": -2.4006083011627197,
    "eval_logps/chosen": -211.9842987060547,
    "eval_logps/rejected": -214.5289306640625,
    "eval_loss": 0.673168957233429,
    "eval_rewards/accuracies": 0.5578358173370361,
    "eval_rewards/chosen": -0.5129190683364868,
    "eval_rewards/margins": 0.052563026547431946,
    "eval_rewards/rejected": -0.5654820799827576,
    "eval_runtime": 167.0868,
    "eval_samples": 8552,
    "eval_samples_per_second": 51.183,
    "eval_steps_per_second": 0.802,
    "total_flos": 0.0,
    "train_loss": 0.678918091354856,
    "train_runtime": 7383.0158,
    "train_samples": 160800,
    "train_samples_per_second": 21.78,
    "train_steps_per_second": 0.043
}