all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.2312498092651367,
    "eval_logits/rejected": -2.2264044284820557,
    "eval_logps/chosen": -34.088890075683594,
    "eval_logps/rejected": -37.596412658691406,
    "eval_loss": 1.625149130821228,
    "eval_rewards/accuracies": 0.5078904032707214,
    "eval_rewards/chosen": -0.02173524722456932,
    "eval_rewards/margins": 0.010181299410760403,
    "eval_rewards/rejected": -0.0319165475666523,
    "eval_runtime": 145.6511,
    "eval_samples": 343,
    "eval_samples_per_second": 2.355,
    "eval_steps_per_second": 0.295,
    "train_loss": 1.3100324989913346,
    "train_runtime": 3250.9757,
    "train_samples": 3079,
    "train_samples_per_second": 0.947,
    "train_steps_per_second": 0.118
}