{ "epoch": 2.999098751126561, "eval_kl/chosen": 0.00677796546369791, "eval_kl/rejected": 0.0004128168220631778, "eval_logits/chosen": -1.171875, "eval_logits/rejected": -1.140625, "eval_logps/chosen": -328.0, "eval_logps/rejected": -300.0, "eval_loss": 2.1474804878234863, "eval_nll_loss": 1.0625, "eval_ref_probs/chosen": 0.0038271306548267603, "eval_ref_probs/rejected": 0.0034569373819977045, "eval_rejected_term/max": 0.0004128168220631778, "eval_rejected_term/min": 0.0004128168220631778, "eval_rewards/accuracies": 0.35600000619888306, "eval_rewards/chosen": -0.5859375, "eval_rewards/margins": -0.2421875, "eval_rewards/rejected": -0.341796875, "eval_runtime": 25.3611, "eval_samples_per_second": 39.43, "eval_steps_per_second": 9.858 }