知识卡片
提示是带目标约束的上下文
自回归模型的下一步受输入提示和已生成内容共同影响。设计提示应明确任务、输入边界、允许输出与判据,并通过样本验证效果;自然语言仍可能含糊,不能把一句角色描述当作可靠接口契约。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 大语言模型入门
原书出处 · 1 处
- · 查看原书依据
interacting with an LLM, you are prompting them and not chatting with them as you would with a human being. Remember that LLMs are next-word predictors. This means that the text they generate is heavily dependent on the text they are fed, which includes the input (called the prompt) and the output tokens generated so far by the model. This is collectively called the context.
参考来源
- 位置:《Designing Large Language Model Applications》第01章《大语言模型入门》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch01.html)
- 结论依据:摘录中关于提示是带目标约束的上下文的定义、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:interacting with an LLM, you are prompting them and not chatting with them as you would with a human being. Remember that LLMs are next-word predictors. This means that the text they generate is heavily dependent on the text they are fed, which includes the input (called the prompt) and the output tokens generated so far by the model. This is collectively called the context.