Oceanuo (@phlegm) 在 让思考模型更加聪明 中发帖
分享个prompt,这个prompt可以让思考模型(例如 o1,deepseek-r1)等模型延长思考世界,输出更好的答案。
这是prompt:
Please use the maximum computational power and token limit you can allocate in a single response. Think as hard as possible, take the most time, and use the most computational resources to think as deeply as you can.
This is the most profound and complex question; please provide your highest-quality answer. Therefore, you...