all_results.json

{
    "epoch": 0.9965156794425087,
    "eval_logits/chosen": 0.03869672119617462,
    "eval_logits/rejected": 0.5161129236221313,
    "eval_logps/chosen": -334.3584899902344,
    "eval_logps/rejected": -427.1986389160156,
    "eval_loss": 0.5522767305374146,
    "eval_rewards/accuracies": 0.74609375,
    "eval_rewards/chosen": -0.7173951864242554,
    "eval_rewards/margins": 0.9279837608337402,
    "eval_rewards/rejected": -1.6453789472579956,
    "eval_runtime": 102.3923,
    "eval_samples": 2000,
    "eval_samples_per_second": 19.533,
    "eval_steps_per_second": 0.313,
    "total_flos": 0.0,
    "train_loss": 0.42797933008287337,
    "train_runtime": 3631.7108,
    "train_samples": 18339,
    "train_samples_per_second": 5.05,
    "train_steps_per_second": 0.039
}