Relative Content

Tag Archive for pythonartificial-intelligencelarge-language-modelllamaollama

How to efficiently chunk/embed json? (ollama, chromadb, gp4allembeddings/langchain)

I’ve been working on a simple chatbot, it responds to inquiries in intercom and in telegram. It makes a database of information to pull from based on current support articles in Intercom. It worked pretty well with 150 articles, but as I’ve added more and its up to almost 400 it seems to completely miss easy questions now, and I feel like i’m not ingesting all that information in the most efficient way possible.