{ "epoch": 1.0, "eval_logits/chosen": -1.4170825481414795, "eval_logits/rejected": 0.8718620538711548, "eval_logps/chosen": -337.4761962890625, "eval_logps/rejected": -356.8194885253906, "eval_loss": 0.6016259789466858, "eval_rewards/accuracies": 0.7108209133148193, "eval_rewards/chosen": -0.9077931046485901, "eval_rewards/margins": 0.7775766849517822, "eval_rewards/rejected": -1.685369849205017, "eval_runtime": 220.0694, "eval_samples": 2142, "eval_samples_per_second": 9.733, "eval_steps_per_second": 0.304 }