all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.344653844833374,
    "eval_logits/rejected": -2.157512664794922,
    "eval_logps/chosen": -243.46983337402344,
    "eval_logps/rejected": -238.70326232910156,
    "eval_loss": 1755.65673828125,
    "eval_rewards/accuracies": 0.6635000109672546,
    "eval_rewards/chosen": -0.11464886367321014,
    "eval_rewards/margins": 0.1562647968530655,
    "eval_rewards/rejected": -0.27091366052627563,
    "eval_runtime": 719.9551,
    "eval_samples": 2000,
    "eval_samples_per_second": 2.778,
    "eval_steps_per_second": 1.389,
    "train_loss": 1830.215161976219,
    "train_runtime": 182270.758,
    "train_samples": 61135,
    "train_samples_per_second": 0.335,
    "train_steps_per_second": 0.084
}