# Disentangling Dense Embeddings with Sparse Autoencoders

arXiv: https://arxiv.org/abs/2408.00657 · HTML: https://mlsanity.com/paper/2408.00657/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Placed in SAE features and their applications, one of 13 directions in Mechanistic interpretability.
- Builds on 26 of its 52 references that are in the map.
- First posted on arXiv 1 Aug 2024.

## Placed in

- Mechanistic interpretability / [SAE features and their applications](https://mlsanity.com/mechinterp/sae-features-and-their-applications/)

## Builds on (most cited first, 26 of 26)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Efficient Estimation of Word Representations in Vector Space](https://mlsanity.com/paper/1301.3781/) (2013)
- [Distributed Representations of Words and Phrases and their Compositionality](https://mlsanity.com/paper/1310.4546/) (2013)
- [Deep contextualized word representations](https://mlsanity.com/paper/1802.05365/) (2018)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Towards A Rigorous Science of Interpretable Machine Learning](https://mlsanity.com/paper/1702.08608/) (2017)
- [Text Embeddings by Weakly-Supervised Contrastive Pre-training](https://mlsanity.com/paper/2212.03533/) (2022)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [MTEB: Massive Text Embedding Benchmark](https://mlsanity.com/paper/2210.07316/) (2022)
- [Precise Zero-Shot Dense Retrieval without Relevance Labels](https://mlsanity.com/paper/2212.10496/) (2022)
- [Linguistic Knowledge and Transferability of Contextual Representations](https://mlsanity.com/paper/1903.08855/) (2019)
- [Finding Neurons in a Haystack: Case Studies with Sparse Probing](https://mlsanity.com/paper/2305.01610/) (2023)
- [Evaluating Gender Bias in Machine Translation](https://mlsanity.com/paper/1906.00591/) (2019)
- [k-Sparse Autoencoders](https://mlsanity.com/paper/1312.5663/) (2013)
- [Not All Language Model Features Are One-Dimensionally Linear](https://mlsanity.com/paper/2405.14860/) (2024)
- [Improving Dictionary Learning with Gated Sparse Autoencoders](https://mlsanity.com/paper/2404.16014/) (2024)
- [Towards Principled Evaluations of Sparse Autoencoders for Interpretability and Control](https://mlsanity.com/paper/2405.08366/) (2024)
- [The Curse of Dense Low-Dimensional Information Retrieval for Large Index Sizes](https://mlsanity.com/paper/2012.14210/) (2020)
- [Neuron to Graph: Interpreting Language Model Neurons at Scale](https://mlsanity.com/paper/2305.19911/) (2023)
- [Learning and Evaluating Sparse Interpretable Sentence Embeddings](https://mlsanity.com/paper/1809.08621/) (2018)
- [Knowledge Graph in Astronomical Research with Large Language Models: Quantifying Driving Forces in Interdisciplinary Scientific Discovery](https://mlsanity.com/paper/2406.01391/) (2024)
- [The Unified Astronomy Thesaurus: Semantic Metadata for Astronomy and Astrophysics](https://mlsanity.com/paper/1801.01021/) (2018)

## Built on by (most cited first, 23 of 23)

- [Steering Language Model Refusal with Sparse Autoencoders](https://mlsanity.com/paper/2411.11296/) (2024)
- [Sparse Autoencoders for Hypothesis Generation](https://mlsanity.com/paper/2502.04382/) (2025)
- [What's In My Human Feedback? Learning Interpretable Descriptions of Preference Data](https://mlsanity.com/paper/2510.26202/) (2025)
- [Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms](https://mlsanity.com/paper/2505.20322/) (2025)
- [Position: Use Sparse Autoencoders to Discover Unknowns](https://mlsanity.com/paper/2506.23845/) (2025)
- [Position: Text Embeddings Should Capture Implicit Semantics, Not Just Surface Meaning](https://mlsanity.com/paper/2506.08354/) (2025)
- [AudioSAE: Towards Understanding of Audio-Processing Models with Sparse AutoEncoders](https://mlsanity.com/paper/2602.05027/) (2026)
- [Sparse Autoencoders, Again?](https://mlsanity.com/paper/2506.04859/) (2025)
- [Resurrecting the Salmon: Rethinking Mechanistic Interpretability with Domain-Specific Sparse Autoencoders](https://mlsanity.com/paper/2508.09363/) (2025)
- [Re-envisioning Euclid Galaxy Morphology: Identifying and Interpreting Features with Sparse Autoencoders](https://mlsanity.com/paper/2510.23749/) (2025)
- [Quartet of Diffusions: Structure-Aware Point Cloud Generation through Part and Symmetry Guidance](https://mlsanity.com/paper/2601.20425/) (2026)
- [Interactionless Inverse Reinforcement Learning: A Data-Centric Framework for Durable Alignment](https://mlsanity.com/paper/2602.14844/) (2026)
- [Automated Concept Discovery for LLM-as-a-Judge Preference Analysis](https://mlsanity.com/paper/2603.03319/) (2026)
- [In your own words: computationally identifying interpretable themes in free-text survey data](https://mlsanity.com/paper/2603.26930/) (2026)
- [HistLens: Mapping Idea Change across Concepts and Corpora](https://mlsanity.com/paper/2604.11749/) (2026)
- [A Geometric View for Understanding Concept Learning and Neuron Interpretation in Sparse Autoencoders](https://mlsanity.com/paper/2606.07007/) (2026)
- [Sparse Readout Prism: Explaining Logit-Lens Scores in Features Instead of Tokens](https://mlsanity.com/paper/2609.01936/) (2026)
- [Disentangling Steering Vectors](https://mlsanity.com/paper/2609.07037/) (2026)
- [Navigating the Concept Space of Language Models](https://mlsanity.com/paper/2603.23524/) (2026)
- [Disentanglement Beyond Generative Models with Riemannian ICA](https://mlsanity.com/paper/2605.22531/) (2026)
- [Doc2LoRA Provides Decodable Representations of Scientific Ideas](https://mlsanity.com/paper/2609.38374/) (2026)
- [Sparse Autoencoders Are a Shared Coordinate System for Symbolic Matching](https://mlsanity.com/paper/2610.19636/) (2026)
- [Sparse Autoencoders for Mechanistic Interpretability in NLP: A Critical Survey of Recoverability, Interpretability, and Utility](https://mlsanity.com/paper/2610.20139/) (2026)

## Abstract (arXiv)

Sparse autoencoders (SAEs) have shown promise in extracting interpretable features from complex neural networks. We present one of the first applications of SAEs to dense text embeddings from large language models, demonstrating their effectiveness in disentangling semantic concepts. By training SAEs on embeddings of over 420,000 scientific paper abstracts from computer science and astronomy, we show that the resulting sparse representations maintain semantic fidelity while offering interpretability. We analyse these learned features, exploring their behaviour across different model capacities and introducing a novel method for identifying ``feature families'' that represent related concepts at varying levels of abstraction. To demonstrate the practical utility of our approach, we show how these interpretable features can be used to precisely steer semantic search, allowing for fine-grained control over query semantics. This work bridges the gap between the semantic richness of dense embeddings and the interpretability of sparse representations. We open source our embeddings, trained sparse autoencoders, and interpreted features, as well as a web app for exploring them.