# Insights into a radiology-specialised multimodal large language model with sparse autoencoders

arXiv: https://arxiv.org/abs/2507.12950 · HTML: https://mlsanity.com/paper/2507.12950/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 35 of its 54 references that are in the map.
- First posted on arXiv 17 Jul 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)
- Mechanistic interpretability / [Interpreting biomedical and science models](https://mlsanity.com/mechinterp/interpreting-biomedical-and-science-models/)

## Builds on (most cited first, 30 of 35)

- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison](https://mlsanity.com/paper/1901.07031/) (2019)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [The Linear Representation Hypothesis and the Geometry of Large Language Models](https://mlsanity.com/paper/2311.03658/) (2023)
- [Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2](https://mlsanity.com/paper/2408.05147/) (2024)
- [Towards Generalist Biomedical AI](https://mlsanity.com/paper/2307.14334/) (2023)
- [A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders](https://mlsanity.com/paper/2409.14507/) (2024)
- [The Remarkable Robustness of LLMs: Stages of Inference?](https://mlsanity.com/paper/2406.19384/) (2024)
- [Learning Multi-Level Features with Matryoshka Sparse Autoencoders](https://mlsanity.com/paper/2503.17547/) (2025)
- [Exploring scalable medical image encoders beyond text supervision](https://mlsanity.com/paper/2401.10815/) (2024)
- [BatchTopK Sparse Autoencoders](https://mlsanity.com/paper/2412.06410/) (2024)
- [Dynamic Graph Enhanced Contrastive Learning for Chest X-ray Report Generation](https://mlsanity.com/paper/2303.10323/) (2023)
- [METransformer: Radiology Report Generation by Transformer with Multiple Learnable Expert Tokens](https://mlsanity.com/paper/2304.02211/) (2023)
- [Steering Language Model Refusal with Sparse Autoencoders](https://mlsanity.com/paper/2411.11296/) (2024)
- [Advancing Multimodal Medical Capabilities of Gemini](https://mlsanity.com/paper/2405.03162/) (2024)
- [MAIRA-1: A specialised large multimodal model for radiology report generation](https://mlsanity.com/paper/2311.13668/) (2023)
- [Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models](https://mlsanity.com/paper/2504.02821/) (2025)
- [InterPLM: Discovering Interpretable Features in Protein Language Models via Sparse Autoencoders](https://mlsanity.com/paper/2412.12101/) (2024)
- [A Multimodal Automated Interpretability Agent](https://mlsanity.com/paper/2404.14394/) (2024)
- [Multimodal Healthcare AI: Identifying and Designing Clinically Relevant Vision-Language Applications for Radiology](https://mlsanity.com/paper/2402.14252/) (2024)
- [Enhancing Automated Interpretability with Output-Centric Feature Descriptions](https://mlsanity.com/paper/2501.08319/) (2025)
- [A Concept-Based Explainability Framework for Large Multimodal Models](https://mlsanity.com/paper/2406.08074/) (2024)
- [Efficient Dictionary Learning with Switch Sparse Autoencoders](https://mlsanity.com/paper/2410.08201/) (2024)
- [An X-Ray Is Worth 15 Features: Sparse Autoencoders for Interpretable Radiology Report Generation](https://mlsanity.com/paper/2410.03334/) (2024)
- [Interpretable and Testable Vision Features via Sparse Autoencoders](https://mlsanity.com/paper/2502.06755/) (2025)
- [Relational Composition in Neural Networks: A Survey and Call to Action](https://mlsanity.com/paper/2407.14662/) (2024)
- [Large Multi-modal Models Can Interpret Features in Large Multi-modal Models](https://mlsanity.com/paper/2411.14982/) (2024)

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

- [Can SAEs reveal and mitigate racial biases of LLMs in healthcare?](https://mlsanity.com/paper/2511.00177/) (2025)
- [MI-CXR: A Benchmark for Longitudinal Reasoning over Multi-Interval Chest X-rays](https://mlsanity.com/paper/2605.15574/) (2026)

## Abstract (arXiv)

Interpretability can improve the safety, transparency and trust of AI models, which is especially important in healthcare applications where decisions often carry significant consequences. Mechanistic interpretability, particularly through the use of sparse autoencoders (SAEs), offers a promising approach for uncovering human-interpretable features within large transformer-based models. In this study, we apply Matryoshka-SAE to the radiology-specialised multimodal large language model, MAIRA-2, to interpret its internal representations. Using large-scale automated interpretability of the SAE features, we identify a range of clinically relevant concepts - including medical devices (e.g., line and tube placements, pacemaker presence), pathologies such as pleural effusion and cardiomegaly, longitudinal changes and textual features. We further examine the influence of these features on model behaviour through steering, demonstrating directional control over generations with mixed success. Our results reveal practical and methodological challenges, yet they offer initial insights into the internal concepts learned by MAIRA-2 - marking a step toward deeper mechanistic understanding and interpretability of a radiology-adapted multimodal large language model, and paving the way for improved model transparency. We release the trained SAEs and interpretations: https://huggingface.co/microsoft/maira-2-sae.