知识卡片

普通读书笔记卡 · 6640.g · 生成模型:GPT-6

提示和解码属于评估条件

理解等级:中级

同一模型在不同提示或解码设置下可能表现不同。应保存输入模板、示例、预算、参数及输出解析;是否使用分步提示依目标场景而定,不能把没用某一种提示就一律判为低估能力。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      sensitivity of the prompts and the probabilistic nature of generative models. For example, I often see papers claiming that “GPT-4 does not have reasoning capabilities,” while not using any prompting techniques during evaluation. In many of these cases, it turns out that the model can in fact perform the task if prompted with CoT prompting. While evaluation prompts need not be heavily engineered, using rudimentary

    参考来源

    - 位置:《Designing Large Language Model Applications》第05章《按任务适配模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch05.html) - 结论依据:摘录中关于提示和解码属于评估条件的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:sensitivity of the prompts and the probabilistic nature of generative models. For example, I often see papers claiming that “GPT-4 does not have reasoning capabilities,” while not using any prompting techniques during evaluation. In many of these cases, it turns out that the model can in fact perform the task if prompted with CoT prompting. While evaluation prompts need not be heavily engineered, using rudimentary