all_results.json

{
    "epoch": 0.96,
    "eval_logits/chosen": -0.6184229850769043,
    "eval_logits/rejected": -0.4482918381690979,
    "eval_logps/chosen": -0.7622865438461304,
    "eval_logps/rejected": -0.8116828799247742,
    "eval_loss": 4.593369960784912,
    "eval_rewards/accuracies": 0.5859375,
    "eval_rewards/chosen": -7.622865200042725,
    "eval_rewards/margins": 0.49396347999572754,
    "eval_rewards/rejected": -8.116828918457031,
    "eval_runtime": 1.1825,
    "eval_samples_per_second": 105.71,
    "eval_steps_per_second": 3.383,
    "total_flos": 17535837601792.0,
    "train_loss": 4.453756756252712,
    "train_runtime": 178.4889,
    "train_samples_per_second": 13.306,
    "train_steps_per_second": 0.101
}