English
Related papers

Related papers: Sparse Autoencoders Find Highly Interpretable Feat…

200 papers

Sparse autoencoders (SAEs) are a useful tool for uncovering human-interpretable features in the activations of large language models (LLMs). While some expect SAEs to find the true underlying features used by a model, our research shows…

Machine Learning · Computer Science 2025-01-31 Gonçalo Paulo , Nora Belrose

Neural networks, specifically deep convolutional neural networks, have achieved unprecedented performance in various computer vision tasks, but the rationale for the computations and structures of successful neural networks is not fully…

Computer Vision and Pattern Recognition · Computer Science 2021-08-19 Joshua Bowren

The classical sparse coding model represents visual stimuli as a linear combination of a handful of learned basis functions that are Gabor-like when trained on natural image data. However, the Gabor-like filters learned by classical sparse…

Artificial Intelligence · Computer Science 2023-02-23 Jonathan Huml , Abiy Tasissa , Demba Ba

Sparse autoencoders (SAEs) are a promising technique for decomposing language model activations into interpretable linear features. However, current SAEs fall short of completely explaining model performance, resulting in "dark matter":…

Machine Learning · Computer Science 2025-03-26 Joshua Engels , Logan Riggs , Max Tegmark

Sparse autoencoders (SAEs) are a recent technique for decomposing neural network activations into human-interpretable features. However, in order for SAEs to identify all features represented in frontier models, it will be necessary to…

Machine Learning · Computer Science 2025-06-04 Anish Mudide , Joshua Engels , Eric J. Michaud , Max Tegmark , Christian Schroeder de Witt

Interpretability research often adopts a neuron-centric lens, treating individual neurons as the fundamental units of explanation. However, neuron-level explanations can be undermined by superposition, where single units respond to mixtures…

Computer Vision and Pattern Recognition · Computer Science 2026-03-17 Julien Colin , Lore Goetschalckx , Thomas Fel , Victor Boutin , Thomas Serre , Nuria Oliver

Understanding how different AI models encode the same high-level concepts, such as objects or attributes, remains challenging because each model typically produces its own isolated representation. Existing interpretability methods like…

Computer Vision and Pattern Recognition · Computer Science 2026-03-09 Ali Nasiri-Sarvi , Hassan Rivaz , Mahdi S. Hosseini

It has been long known that sparsity is an effective inductive bias for learning efficient representation of data in vectors with fixed dimensionality, and it has been explored in many areas of representation learning. Of particular…

Computation and Language · Computer Science 2021-04-20 Victor Prokhorov , Yingzhen Li , Ehsan Shareghi , Nigel Collier

We investigate whether sparse autoencoders (SAEs) can be used to remove knowledge from language models. We use the biology subset of the Weapons of Mass Destruction Proxy dataset and test on the gemma-2b-it and gemma-2-2b-it language…

Machine Learning · Computer Science 2024-11-05 Eoin Farrell , Yeu-Tong Lau , Arthur Conmy

A new line of research for feature selection based on neural networks has recently emerged. Despite its superiority to classical methods, it requires many training iterations to converge and detect informative features. The computational…

Machine Learning · Computer Science 2022-11-29 Ghada Sokar , Zahra Atashgahi , Mykola Pechenizkiy , Decebal Constantin Mocanu

Sparse autoencoders (SAEs) have recently emerged as pivotal tools for introspection into large language models. SAEs can uncover high-quality, interpretable features at different levels of granularity and enable targeted steering of the…

Information Retrieval · Computer Science 2026-01-19 Martin Spišák , Ladislav Peška , Petr Škoda , Vojtěch Vančura , Rodrigo Alves

Interpretable Multi-Task Learning can be expressed as learning a sparse graph of the task relationship based on the prediction performance of the learned models. Since many natural phenomenon exhibit sparse structures, enforcing sparsity on…

Machine Learning · Computer Science 2020-09-14 Francesco Alesiani , Shujian Yu , Ammar Shaker , Wenzhe Yin

A new method for the unsupervised learning of sparse representations using autoencoders is proposed and implemented by ordering the output of the hidden units by their activation value and progressively reconstructing the input in this…

Machine Learning · Computer Science 2016-05-09 Paul Bertens

In complex visual recognition tasks it is typical to adopt multiple descriptors, that describe different aspects of the images, for obtaining an improved recognition performance. Descriptors that have diverse forms can be fused into a…

Computer Vision and Pattern Recognition · Computer Science 2015-06-15 Jayaraman J. Thiagarajan , Karthikeyan Natesan Ramamurthy , Andreas Spanias

Deep neural networks have emerged as powerful tools for learning operators defined over infinite-dimensional function spaces. However, existing theories frequently encounter difficulties related to dimensionality and limited…

Machine Learning · Computer Science 2026-05-12 Jianfei Li , Shuo Huang , Han Feng , Ding-Xuan Zhou , Gitta Kutyniok

Dense embeddings deliver strong retrieval performance but often lack interpretability and controllability. This paper introduces a novel approach using sparse autoencoders (SAE) to interpret and control dense embeddings via the learned…

Information Retrieval · Computer Science 2025-02-25 Hao Kang , Tevin Wang , Chenyan Xiong

The fidelity with which neural networks can now generate content such as music presents a scientific opportunity: these systems appear to have learned implicit theories of such content's structure through statistical learning alone. This…

Sound · Computer Science 2026-03-03 Nikhil Singh , Manuel Cherep , Pattie Maes

Artificial intelligence in healthcare requires models that are accurate and interpretable. We advance mechanistic interpretability in medical vision by applying Medical Sparse Autoencoders (MedSAEs) to the latent space of MedCLIP, a…

Artificial Intelligence · Computer Science 2026-05-25 Riccardo Renzulli , Colas Lepoutre , Enrico Cassano , Marco Grangetto

Sparse autoencoders (SAEs) are designed to extract interpretable features from language models by enforcing a sparsity constraint. Ideally, training an SAE would yield latents that are both sparse and semantically meaningful. However, many…

Machine Learning · Computer Science 2025-11-06 Xiaoqing Sun , Alessandro Stolfo , Joshua Engels , Ben Wu , Senthooran Rajamanoharan , Mrinmaya Sachan , Max Tegmark

Word embeddings have found their way into a wide range of natural language processing tasks including those in the biomedical domain. While these vector representations successfully capture semantic and syntactic word relations, hidden…