知识卡片

普通读书笔记卡 · 8691.c · 生成模型:GPT-6

调用协议须区分等待与结束

理解等级:高级

工具调用、工具返回和最终回答应由明确消息角色或结束标记区分。协议与模型模板相关,书中Llama特殊token是具体版本示例;运行时不能将任何普通文本中出现的函数名都当待执行调用。

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

      the 2024 Summer Olympics?")<|eom_id|> The <|python_tag|> token is a special token generated by Llama 3.1 to indicate that it is entering tool-calling mode. The <|eom_id|> special token indicates that the model has not ended its turn yet and will wait to be fed with the results of the tool invocation. You can also provide your own tools in the prompt: using JSON is recommended. Tip

    参考来源

    - 位置:《Designing Large Language Model Applications》第10章《外部工具与智能体》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch10.html) - 结论依据:摘录中关于调用协议须区分等待与结束的定义、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:the 2024 Summer Olympics?")<|eom_id|> The <|python_tag|> token is a special token generated by Llama 3.1 to indicate that it is entering tool-calling mode. The <|eom_id|> special token indicates that the model has not ended its turn yet and will wait to be fed with the results of the tool invocation. You can also provide your own tools in the prompt: using JSON is recommended. Tip