# Mammo-SAE: Interpreting Breast Cancer Concept Learning with Sparse Autoencoders

arXiv: https://arxiv.org/abs/2507.15227 · HTML: https://mlsanity.com/paper/2507.15227/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Interpreting biomedical and science models, one of 13 directions in Mechanistic interpretability.
- Builds on 16 of its 26 references that are in the map.
- First posted on arXiv 21 Jul 2025.

## Placed in

- Mechanistic interpretability / [Interpreting biomedical and science models](https://mlsanity.com/mechinterp/interpreting-biomedical-and-science-models/)

## Builds on (most cited first, 16 of 16)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Gemma: Open Models Based on Gemini Research and Technology](https://mlsanity.com/paper/2403.08295/) (2024)
- [Scaling and evaluating sparse autoencoders](https://mlsanity.com/paper/2406.04093/) (2024)
- [Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2](https://mlsanity.com/paper/2408.05147/) (2024)
- [Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders](https://mlsanity.com/paper/2410.20526/) (2024)
- [Discover-then-Name: Task-Agnostic Concept Bottlenecks via Automated Concept Discovery](https://mlsanity.com/paper/2407.14499/) (2024)
- [Towards Principled Evaluations of Sparse Autoencoders for Interpretability and Control](https://mlsanity.com/paper/2405.08366/) (2024)
- [VinDr-Mammo: A large-scale benchmark dataset for computer-aided diagnosis in full-field digital mammography](https://mlsanity.com/paper/2203.11205/) (2022)
- [Sparse autoencoders reveal selective remapping of visual concepts during adaptation](https://mlsanity.com/paper/2412.05276/) (2024)
- [Mammo-CLIP: A Vision Language Foundation Model to Enhance Data Efficiency and Robustness in Mammography](https://mlsanity.com/paper/2405.12255/) (2024)
- [PURE: Turning Polysemantic Neurons Into Pure Features by Identifying Relevant Circuits](https://mlsanity.com/paper/2404.06453/) (2024)
- [Disentangling Neuron Representations with Concept Vectors](https://mlsanity.com/paper/2304.09707/) (2023)
- [Understanding polysemanticity in neural networks through coding theory](https://mlsanity.com/paper/2401.17975/) (2024)
- [Breast Cancer Classification with Enhanced Interpretability: DALAResNet50 and DT Grad-CAM](https://mlsanity.com/paper/2308.13150/) (2023)
- [Prototype-based Interpretable Breast Cancer Prediction Models: Analysis and Challenges](https://mlsanity.com/paper/2403.20260/) (2024)

## Built on by (most cited first, 2 of 2)

- [ConceptScope: Characterizing Dataset Bias via Disentangled Visual Concepts](https://mlsanity.com/paper/2510.26186/) (2025)
- [Mechanistic Interpretability of EEG Foundation Models via Sparse Autoencoders](https://mlsanity.com/paper/2605.13930/) (2026)

## Abstract (arXiv)

Interpretability is critical in high-stakes domains such as medical imaging, where understanding model decisions is essential for clinical adoption. In this work, we introduce Sparse Autoencoder (SAE)-based interpretability to breast imaging by analyzing Mammo-CLIP, a vision--language foundation model pretrained on large-scale mammogram image--report pairs. We train a patch-level textttMammo-SAE on Mammo-CLIP to identify and probe latent features associated with clinically relevant breast concepts such as textitmass and textitsuspicious calcification. Our findings reveal that top activated class level latent neurons in the SAE latent space often tend to align with ground truth regions, and also uncover several confounding factors influencing the model's decision-making process. Additionally, we analyze which latent neurons the model relies on during downstream finetuning for improving the breast concept prediction. This study highlights the promise of interpretable SAE latent representations in providing deeper insight into the internal workings of foundation models at every layer for breast imaging. The code will be released at https://krishnakanthnakka.github.io/MammoSAE/