Chris Olah

"Neural network parameters can be thought of as compiled computer programs. Somehow, they encode sophisticated algorithms, capable of things no human knows how to write a computer program to do. Mechanistic interpretability seeks to reverse engineer neural networks into human understandable algorithms. Previous work has tended to focus on vision models; this talk will explore how we might reverse engineer transformer language models.  In particular, we'll focus on what we call ""induction head circuits"", a mechanism that appears to be significantly responsible for in-context learning. Using a pair of attention heads, these circuits allow models to repeat text from earlier in the context, translate text seen earlier, mimic functions from examples earlier in the context, and much more. The discovery of induction heads in the learning process appears to drive a sharp phase change, creating a bump in the loss curve, pivoting models learning trajectories, and greatly increasing their capacity for in-context learning, in the span of just a few hundred training steps." Chris Olah is a co-founder of Anthropic, an AI company focused on the safety of large models, where he leads Anthropic's interpretability efforts. Previously, Chris led OpenAI's interpretability team, and was a researcher at Google Brain. Chris' work includes the Circuits project, his blog (especially his tutorial on LSTMs), the Distill journal, and DeepDream. View the entire CS25 Transformers United playlist: https://www.youtube.com/playlist?list=PLoROMvodv4rNiJRchCzutFw5ItR_Z27CM