Towards Monosemanticity: Decomposing Language Models Into Understandable Components

This week, we're discussing "Decomposing Language Models Into Understandable Components", which addresses the challenge of understanding the inner workings of neural networks, drawing parallels with the complexity of human brain function. It explores the concept of "features," (patterns of neuron activations) providing a more interpretable way to dissect neural networks. By decomposing a layer of neurons into thousands of features, this approach uncovers hidden model properties that are not evident when examining individual neurons. These features are demonstrated to be more interpretable and consistent, offering the potential to steer model behavior and improve AI safety.

Read the transcript and more on the blog: https://arize.com/blog/towards-monose...

Link to paper: https://transformer-circuits.pub/2023...