# CytoSAE: Interpretable Cell Embeddings for Hematology

arXiv: https://arxiv.org/abs/2507.12464 · HTML: https://mlsanity.com/paper/2507.12464/

## Key facts

- Cited by 8 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 30 references that are in the map.
- First posted on arXiv 16 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)

- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Quantifying Attention Flow in Transformers](https://mlsanity.com/paper/2005.00928/) (2020)
- [Scaling and evaluating sparse autoencoders](https://mlsanity.com/paper/2406.04093/) (2024)
- [Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models](https://mlsanity.com/paper/2403.19647/) (2024)
- [Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2](https://mlsanity.com/paper/2408.05147/) (2024)
- [k-Sparse Autoencoders](https://mlsanity.com/paper/1312.5663/) (2013)
- [Improving Dictionary Learning with Gated Sparse Autoencoders](https://mlsanity.com/paper/2404.16014/) (2024)
- [Open Problems in Mechanistic Interpretability](https://mlsanity.com/paper/2501.16496/) (2025)
- [SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders](https://mlsanity.com/paper/2501.18052/) (2025)
- [Sparse autoencoders reveal selective remapping of visual concepts during adaptation](https://mlsanity.com/paper/2412.05276/) (2024)
- [Explainable Medical Imaging AI Needs Human-Centered Design: Guidelines and Evidence from a Systematic Review](https://mlsanity.com/paper/2112.12596/) (2021)
- [DinoBloom: A Foundation Model for Generalizable Cell Embeddings in Hematology](https://mlsanity.com/paper/2404.05022/) (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)
- [Learning biologically relevant features in a pathology foundation model using sparse autoencoders](https://mlsanity.com/paper/2407.10785/) (2024)

## Built on by (most cited first, 8 of 8)

- [ConceptScope: Characterizing Dataset Bias via Disentangled Visual Concepts](https://mlsanity.com/paper/2510.26186/) (2025)
- [VisualScratchpad: Inference-time Visual Concepts Analysis in Vision Language Models](https://mlsanity.com/paper/2603.07335/) (2026)
- [GeoSAE: Geometric Prior-Guided Layer-Wise Sparse Autoencoder Annotation of Brain MRI Foundation Models](https://mlsanity.com/paper/2605.01829/) (2026)
- [Equivariant Sparse Autoencoders: Mechanistic Interpretability of Neural Networks on Symmetric Data](https://mlsanity.com/paper/2511.09432/) (2025)
- [MedConcept: Unsupervised Concept Discovery for Interpretability in Medical VLMs](https://mlsanity.com/paper/2604.11868/) (2026)
- [EXPOSE: Explainable and Domain-Robust Embeddings from Pathology Vision Foundation Models using Sparse Autoencoders](https://mlsanity.com/paper/2608.28191/) (2026)
- [Medical AI Encodes a "Feeling of Error": Verifying Cancer Segmentation via Internal Concepts](https://mlsanity.com/paper/2609.08879/) (2026)
- [HemaHier: Chain-Conditioned Ordinal Hierarchies for Lineage-Aware Bone-Marrow Cytology](https://mlsanity.com/paper/2609.12835/) (2026)

## Abstract (arXiv)

Sparse autoencoders (SAEs) emerged as a promising tool for mechanistic interpretability of transformer-based foundation models. Very recently, SAEs were also adopted for the visual domain, enabling the discovery of visual concepts and their patch-wise attribution to tokens in the transformer model. While a growing number of foundation models emerged for medical imaging, tools for explaining their inferences are still lacking. In this work, we show the applicability of SAEs for hematology. We propose CytoSAE, a sparse autoencoder which is trained on over 40,000 peripheral blood single-cell images. CytoSAE generalizes to diverse and out-of-domain datasets, including bone marrow cytology, where it identifies morphologically relevant concepts which we validated with medical experts. Furthermore, we demonstrate scenarios in which CytoSAE can generate patient-specific and disease-specific concepts, enabling the detection of pathognomonic cells and localized cellular abnormalities at the patch level. We quantified the effect of concepts on a patient-level AML subtype classification task and show that CytoSAE concepts reach performance comparable to the state-of-the-art, while offering explainability on the sub-cellular level. Source code and model weights are available at https://github.com/dynamical-inference/cytosae.