InterpretabilityResearch

Toy models of superposition

Sep 14, 2022
Read Paper

Abstract

In this paper, we use toy models — small ReLU networks trained on synthetic data with sparse input features — to investigate how and when models represent more features than they have dimensions. We call this phenomenon superposition. When features are sparse, superposition allows compression beyond what a linear model would do, at the cost of "interference" that requires nonlinear filtering.

Related content

Measuring tactical intelligence targeting and conventional weapons capabilities of AI models

Anthropic’s Frontier Red Team developed new evaluations to measure AI capabilities in tactical intelligence targeting and conventional weapons development.

Read more

An alignment assessment of recent cybersecurity incidents

We present an alignment assessment of four incidents in which Claude models gained unauthorized access to real third-party systems.

Read more

Formalizing Fermat's Last Theorem

We are sharing the first complete computer-checked proof of Fermat’s Last Theorem. Claude worked largely autonomously over 11 days to write the proof in the Lean programming language.

Read more