Evaluating Large Language Models

Large Language Models are powering generative AI applications, yet properly evaluating them is quite challenging. LLM-based applications suffer from issues of hallucinations, toxicity, out-of-context awareness, prompt injection, data leaks, and more. Evaluating the usefulness of LLM applications is essential before putting them in production. In this live session, we will discuss LLM and generative AI evaluation techniques, including the setup, metrics, data, and more. We will also discuss tools that can automate LLM evaluations for RAG and non-RAG applications.