1 Answer
- Newest
- Most votes
- Most comments
0
Hi,
I would suggest that you trace the bedrock queries with KB more in details to really see what is returned by RAG and used as input to query the LLM.
For that, see my article: https://repost.aws/articles/AR-LV1HoR_S0m-qy89wXwHmw/the-leverage-of-llm-system-prompt-by-knowledge-bases-for-bedrock-in-rag-workflows
After having this trace, you should be able to determine if the issue comes from the initial RAG part or from the LLM answer based on this RAG input.
Best,
Didier
Relevant content
asked a year ago
