# Uncovering Branch specialization in InceptionV1 using k sparse autoencoders

arXiv: https://arxiv.org/abs/2504.11489 · HTML: https://mlsanity.com/paper/2504.11489/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Sparse autoencoder training and evaluation, one of 13 directions in Mechanistic interpretability.
- Builds on 4 of its 15 references that are in the map.
- First posted on arXiv 14 Apr 2025.

## Placed in

- Mechanistic interpretability / [Sparse autoencoder training and evaluation](https://mlsanity.com/mechinterp/sparse-autoencoder-training-and-evaluation/)

## Builds on (most cited first, 4 of 4)

- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [A Multimodal Automated Interpretability Agent](https://mlsanity.com/paper/2404.14394/) (2024)
- [The Missing Curve Detectors of InceptionV1: Applying Sparse Autoencoders to InceptionV1 Early Vision](https://mlsanity.com/paper/2406.03662/) (2024)

## Built on by (most cited first, 1 of 1)

- [Characterising the Inductive Biases of Neural Networks on Boolean Data](https://mlsanity.com/paper/2505.24060/) (2025)

## Abstract (arXiv)

Sparse Autoencoders (SAEs) have shown to find interpretable features in neural networks from polysemantic neurons caused by superposition. Previous work has shown SAEs are an effective tool to extract interpretable features from the early layers of InceptionV1. Since then, there have been many improvements to SAEs but branch specialization is still an enigma in the later layers of InceptionV1. We show various examples of branch specialization occuring in each layer of the mixed4a-4e branch, in the 5x5 branch and in one 1x1 branch. We also provide evidence to claim that branch specialization seems to be consistent across layers, similar features across the model will be localized in the same convolution size branches in their respective layer.