all_results.json

{
    "epoch": 2.998430141287284,
    "eval_logits/chosen": -0.58203125,
    "eval_logits/rejected": 1.25,
    "eval_logps/chosen": -113.0,
    "eval_logps/rejected": -171.0,
    "eval_loss": 0.849609375,
    "eval_nll_loss": 0.0,
    "eval_rewards/accuracies": 0.671875,
    "eval_rewards/chosen": -1.1328125,
    "eval_rewards/margins": 0.578125,
    "eval_rewards/rejected": -1.7109375,
    "eval_runtime": 23.4202,
    "eval_samples": 2000,
    "eval_samples_per_second": 85.396,
    "eval_steps_per_second": 1.366,
    "total_flos": 0.0,
    "train_loss": 0.0,
    "train_runtime": 0.0608,
    "train_samples": 61134,
    "train_samples_per_second": 3016192.654,
    "train_steps_per_second": 47117.218
}