# Binary Sparse Coding for Interpretability

arXiv: https://arxiv.org/abs/2509.25596 · HTML: https://mlsanity.com/paper/2509.25596/

## Key facts

- Cited by 3 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 19 of its 29 references that are in the map.
- First posted on arXiv 29 Sep 2025.

## Placed in

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

## Builds on (most cited first, 19 of 19)

- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [signSGD: Compressed Optimisation for Non-Convex Problems](https://mlsanity.com/paper/1802.04434/) (2018)
- [Scaling and evaluating sparse autoencoders](https://mlsanity.com/paper/2406.04093/) (2024)
- [Finding Neurons in a Haystack: Case Studies with Sparse Probing](https://mlsanity.com/paper/2305.01610/) (2023)
- [Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoders](https://mlsanity.com/paper/2407.14435/) (2024)
- [Transcoders Find Interpretable LLM Feature Circuits](https://mlsanity.com/paper/2406.11944/) (2024)
- [Linear Algebraic Structure of Word Senses, with Applications to Polysemy](https://mlsanity.com/paper/1601.03764/) (2016)
- [The Road Less Scheduled](https://mlsanity.com/paper/2405.15682/) (2024)
- [BatchTopK Sparse Autoencoders](https://mlsanity.com/paper/2412.06410/) (2024)
- [Automatically Interpreting Millions of Features in Large Language Models](https://mlsanity.com/paper/2410.13928/) (2024)
- [Universal Neurons in GPT2 Language Models](https://mlsanity.com/paper/2401.12181/) (2024)
- [Codebook Features: Sparse and Discrete Interpretability for Neural Networks](https://mlsanity.com/paper/2310.17230/) (2023)
- [Transcoders Beat Sparse Autoencoders for Interpretability](https://mlsanity.com/paper/2501.18823/) (2025)
- [Interpretability as Compression: Reconsidering SAE Explanations of Neural Activations with MDL-SAEs](https://mlsanity.com/paper/2410.11179/) (2024)
- [Sparse Autoencoder Features for Classifications and Transferability](https://mlsanity.com/paper/2502.11367/) (2025)
- [Revisiting End-To-End Sparse Autoencoder Training: A Short Finetune Is All You Need](https://mlsanity.com/paper/2503.17272/) (2025)

## Built on by (most cited first, 3 of 3)

- [Exploratory Causal Inference in SAEnce](https://mlsanity.com/paper/2510.14073/) (2025)
- [Safe Transformer: An Explicit Safety Bit For Interpretable And Controllable Alignment](https://mlsanity.com/paper/2603.06727/) (2026)
- [Investigating Permutation-Invariant Discrete Representation Learning for Spatially Aligned Images](https://mlsanity.com/paper/2604.01843/) (2026)

## Abstract (arXiv)

Sparse autoencoders (SAEs) are used to decompose neural network activations into sparsely activating features, but many SAE features are only interpretable at high activation strengths. To address this issue we propose to use binary sparse autoencoders (BAEs) and binary transcoders (BTCs), which constrain all activations to be zero or one. We find that binarisation significantly improves the interpretability and monosemanticity of the discovered features, while increasing reconstruction error. By eliminating the distinction between high and low activation strengths, we prevent uninterpretable information from being smuggled in through the continuous variation in feature activations. However, we also find that binarisation increases the number of uninterpretable ultra-high frequency features, and when interpretability scores are frequency-adjusted, the scores for continuous sparse coders are slightly better than those of binary ones. This suggests that polysemanticity may be an ineliminable property of neural activations.