响应预置
可以在脚本中提供LLM响应的开头部分(assistant
消息)。
这允许将LLM的答案引导至特定的语法或格式。
使用assistant
函数来提供助手文本。
$`List 5 colors. Answer with a JSON array. Do not emit the enclosing markdown.`
// help the LLM by starting the JSON array syntax// in the assistant responseassistant(`[`)
👤 user
List 5 colors. Answer with a JSON array. Do not emit the enclosing markdown.
🤖 assistant
[
🤖 assistant
"red","blue","green","yellow","purple"]
它是如何工作的?
在内部调用LLM时,会向查询添加一条额外消息,就好像LLM生成了这个内容一样。
{ "messages": [ ..., { "role": "assistant", "content": "[\n" } ]}