{ "epoch": 2.999098751126561, "eval_kl/chosen": -0.00022398879809770733, "eval_kl/rejected": 0.00018860936688724905, "eval_logits/chosen": -1.28125, "eval_logits/rejected": -1.25, "eval_logps/chosen": -324.0, "eval_logps/rejected": -300.0, "eval_loss": 1.9966460466384888, "eval_nll_loss": 1.0625, "eval_ref_probs/chosen": 0.0038271306548267603, "eval_ref_probs/rejected": 0.0034569373819977045, "eval_rejected_term/max": 0.00018860936688724905, "eval_rejected_term/min": 0.00018860936688724905, "eval_rewards/accuracies": 0.3880000114440918, "eval_rewards/chosen": -0.3203125, "eval_rewards/margins": -0.0390625, "eval_rewards/rejected": -0.28125, "eval_runtime": 25.512, "eval_samples_per_second": 39.197, "eval_steps_per_second": 9.799 }