Back to Course
LLM Engineering: Transformers & RAG
Module 7 of 12
7. Evaluation (RAGAS)
1. How to grade a Chatbot?
Accuracy is subjective. We use LLM-as-a-Judge. Ask GPT-4 to grade Llama-3's answer.
2. RAGAS Metrics
- Faithfulness: Did the answer come from the context?
- Answer Relevance: Did it answer the user's question?