Find answers from the community

Updated 8 months ago

Metadta

Hey @Logan M, I have a bunch of English explanations of our code base and instead of storing the embeddings of the code base itself, I store the English explanation of the code and embed that. At the same time I store the code snippets as metadata in the Document under the metadata key called “code”. My question: how do I retrieve the corresponding code snippets in that metadata along with the explanation text during retrieval and context setting during synthesizing LLM prompt?
Thanks !

7 comments

LLogan M

The metadata is already included with the text, unless you explicitly disabled that

ccmosguy

Thanks @Logan M I’m not seeing it for some reason. How would you debug and print out all the calls ? The global set handler “simple” is not outputting anything for me.

LLogan M

It really should be outputting lol

But you can debug by doing

Plain Text

retriever = index.as_retriever(...)

nodes = retriever.retrieve("query")

for node in nodes:
    print(node.node.get_content(metadata_mode="llm"))

LLogan M

Whatever gets printed there is what goes into the prompt template sent to the llm

LLogan M

The query engine runs basically exactly that under the hood, before passing it to the response synthesizer

ccmosguy

OK thanks, let me run this and I’ll let you know

ccmosguy

@Logan M thanks, you were right about this, I had to set some instrument observers to check this.

Add a reply