pith. sign in

arxiv: 2411.08166 · v1 · pith:LCDHIVXOnew · submitted 2024-11-12 · 💻 cs.LG

Tackling Polysemanticity with Neuron Embeddings

classification 💻 cs.LG
keywords neuronembeddingspolysemanticitymakingmodelusedactualagnostic
0
0 comments X
read the original abstract

We present neuron embeddings, a representation that can be used to tackle polysemanticity by identifying the distinct semantic behaviours in a neuron's characteristic dataset examples, making downstream manual or automatic interpretation much easier. We apply our method to GPT2-small, and provide a UI for exploring the results. Neuron embeddings are computed using a model's internal representations and weights, making them domain and architecture agnostic and removing the risk of introducing external structure which may not reflect a model's actual computation. We describe how neuron embeddings can be used to measure neuron polysemanticity, which could be applied to better evaluate the efficacy of Sparse Auto-Encoders (SAEs).

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.