Introduction to AI Evals

Join our interactive AI Evaluations Workshop to learn practical skills in evaluating large language models (LLMs) and AI agents! This is a free workshop open to the public!

We’ll cover key concepts including using LLM APIs, understanding jailbreaks, why evaluations (evals) are essential yet challenging, and insights from the alignment faking paper. The session features live coding exercises based on the ARENA materials (section 3.1), using Python and Jupyter notebooks.

Prior Python experience is required. Ideal for AI practitioners, researchers, or enthusiasts looking to deepen their understanding of AI safety and evaluations.