Want to create an interactive transcript for this episode?
Podcast: Machine Learning Street Talk (MLST)
Episode: Neel Nanda - Mechanistic Interpretability
Description: In this wide-ranging conversation, Tim Scarfe interviews Neel Nanda, a researcher at DeepMind working on mechanistic interpretability, which aims to understand the algorithms and representations learned by machine learning models. Neel discusses how models can represent their thoughts using motifs, circuits, and linear directional features which are often communicated via a "residual stream", an information highway models use to pass information between layers.
Neel argues that "superposition", the ability for models to represent more features than they have neurons, is one of the biggest open problems in interpretability. This is because superposition thwarts our ability to understand...