Has any one else noticed that the similarity score is often quite high (>0.7) even for chunks with content not directly related to the user query ? I would assume it; would be much lower. Is there any tips, resources or pointers from the community on this issue ?

Any one mind sharing code for successful use of embedding using AzureOpenAi ?

Does anyone else find that responses obtained through LLM+openAI API are too short ? Is it because most tokens are used by context ?

@kapa.ai How do I ask the same query to all documents in the docstore ?

issues with persist storage : this code : storage_context.persist(persist_dir="./indexes") results in this error : "FileNotFoundError: [Errno 2] No such file or directory: 'c:/Users/Someone/OneDrive/Documents/AI/indexes/docstore.json'. feels silly but I'm unable to find the root cause and solution

We notice a change in behavior that seems to be also identified by others, see https://stackoverflow.com/questions/76259712/llama-index-with-llm-doing-out-of-context-answering