use embeddings stored in vector db to reduce work for LLM generating response题意:使用存储在向量数据库中的嵌入来表示,以减少大型语言模型(LLM)生成响应的工作量。I'm trying to understand what the correct strategy is for storing and using embeddings in a vector database, to be used with an LLM. If my goal is to reduce the amount of work the LLM has to do when generating a