Toward transparent LLM: test, observe and evaluate your RAG assisted chatbot Similarity score = 0.23 More