all_results.json

{
    "epoch": 2.0,
    "eval_logits/chosen": -13.3125,
    "eval_logits/rejected": -13.0625,
    "eval_logps/chosen": -556.0,
    "eval_logps/rejected": -564.0,
    "eval_loss": 0.7793359160423279,
    "eval_rewards/accuracies": 0.72265625,
    "eval_rewards/chosen": -11.75,
    "eval_rewards/margins": 1.9140625,
    "eval_rewards/rejected": -13.6875,
    "eval_runtime": 46.6612,
    "eval_samples": 2000,
    "eval_samples_per_second": 42.862,
    "eval_steps_per_second": 0.686,
    "total_flos": 0.0,
    "train_loss": 0.41369995401518184,
    "train_runtime": 7266.4225,
    "train_samples": 61134,
    "train_samples_per_second": 16.826,
    "train_steps_per_second": 0.526
}