{ "epoch": 0.9997382884061764, "eval_logits/chosen": -0.6658732891082764, "eval_logits/rejected": -0.6216869950294495, "eval_logps/chosen": -282.2928161621094, "eval_logps/rejected": -265.7748107910156, "eval_loss": 0.9743963479995728, "eval_rewards/accuracies": 0.6919999718666077, "eval_rewards/chosen": -1.00694739818573, "eval_rewards/margins": 0.7492353320121765, "eval_rewards/rejected": -1.7561826705932617, "eval_runtime": 140.7369, "eval_samples": 2000, "eval_samples_per_second": 14.211, "eval_steps_per_second": 1.776 }