{ "epoch": 2.999098751126561, "eval_kl/chosen": 0.0006784984725527465, "eval_kl/rejected": -0.00011524503497639671, "eval_logits/chosen": -1.2890625, "eval_logits/rejected": -1.25, "eval_logps/chosen": -324.0, "eval_logps/rejected": -300.0, "eval_loss": 0.8628925681114197, "eval_nll_loss": 1.0625, "eval_ref_probs/chosen": 0.0038271306548267603, "eval_ref_probs/rejected": 0.0034569373819977045, "eval_rejected_term/max": -0.00011524503497639671, "eval_rejected_term/min": -0.00011524503497639671, "eval_rewards/accuracies": 0.41999998688697815, "eval_rewards/chosen": -0.3125, "eval_rewards/margins": -0.056396484375, "eval_rewards/rejected": -0.255859375, "eval_runtime": 25.3686, "eval_samples_per_second": 39.419, "eval_steps_per_second": 9.855 }