English
Related papers

Related papers: Sanity Checks for Sparse Autoencoders: Do SAEs Bea…

200 papers

Sparse Autoencoders (SAEs) have recently gained attention as a means to improve the interpretability and steerability of Large Language Models (LLMs), both of which are essential for AI safety. In this work, we extend the application of…

Computer Vision and Pattern Recognition · Computer Science 2025-12-01 Mateusz Pach , Shyamgopal Karthik , Quentin Bouniot , Serge Belongie , Zeynep Akata

Large Language Models (LLMs) encode factual knowledge within hidden parametric spaces that are difficult to inspect or control. While Sparse Autoencoders (SAEs) can decompose hidden activations into more fine-grained, interpretable…

Machine Learning · Computer Science 2026-01-14 Minglai Yang , Xinyu Guo , Zhengliang Shi , Jinhe Bi , Steven Bethard , Mihai Surdeanu , Liangming Pan

Sparse autoencoders (SAEs) are useful for detecting and steering interpretable features in neural networks, with particular potential for understanding complex multimodal representations. Given their ability to uncover interpretable…

Computer Vision and Pattern Recognition · Computer Science 2025-05-29 Vladimir Zaigrajew , Hubert Baniecki , Przemyslaw Biecek

Text-to-image diffusion models generate images through an iterative denoising process, so internal neural layers produce trajectories of activations rather than single static representations. Sparse autoencoders (SAEs) have recently been…

Computer Vision and Pattern Recognition · Computer Science 2026-05-28 Calvin Yeung , Prathyush Poduval , Ali Zakeri , Zhuowen Zou , Mohsen Imani

Artificial neural networks that can recover latent dynamics from recorded neural activity may provide a powerful avenue for identifying and interpreting the dynamical motifs underlying biological computation. Given that neural variance…

Neurons and Cognition · Quantitative Biology 2023-07-03 Andrew R. Sedler , Christopher Versteeg , Chethan Pandarinath

Sparse autoencoders (SAEs) have emerged as a powerful tool for interpreting neural networks by extracting the concepts represented in their activations. However, choosing the size of the SAE dictionary (i.e. number of learned concepts)…

Machine Learning · Computer Science 2025-03-25 Bart Bussmann , Noa Nabeshima , Adam Karvonen , Neel Nanda

LLMs increasingly require surgical model editing to enhance domain-specific capabilities without incurring the computational cost or catastrophic forgetting associated with full fine-tuning. Sparse Autoencoders (SAEs) have emerged as a…

Machine Learning · Computer Science 2026-05-28 Li Lei , Madalina Ciobanu , Qingqing Mao , Ritankar Das

Sparse Autoencoders (SAEs) have been successfully used to probe Large Language Models (LLMs) and extract interpretable concepts from their internal representations. These concepts are linear combinations of neuron activations that…

Computation and Language · Computer Science 2026-02-23 Mathis Le Bail , Jérémie Dentan , Davide Buscaldi , Sonia Vanier

Sparse autoencoders (SAEs) have emerged as a promising approach for learning interpretable features from neural network activations. However, the optimization landscape for SAE training can be challenging due to correlations in the input…

Machine Learning · Computer Science 2025-11-19 Ashwin Saraswatula , David Klindt

Sparse autoencoders (SAEs) are used to disentangle the dense, polysemantic internal representations of large language models (LLMs) into interpretable, monosemantic concepts. However, standard $\ell_1$-regularized SAEs suffer from feature…

Machine Learning · Computer Science 2026-05-08 Faris Chaudhry , Keisuke Yano , Anthea Monod

Sparse autoencoders provide a promising unsupervised approach for extracting interpretable features from a language model by reconstructing activations from a sparse bottleneck layer. Since language models learn many concepts, autoencoders…

Machine Learning · Computer Science 2024-06-07 Leo Gao , Tom Dupré la Tour , Henk Tillman , Gabriel Goh , Rajan Troll , Alec Radford , Ilya Sutskever , Jan Leike , Jeffrey Wu

Understanding what sparse auto-encoder (SAE) features in vision transformers truly represent is usually done by inspecting the patches where a feature's activation is highest. However, self-attention mixes information across the entire…

Computer Vision and Pattern Recognition · Computer Science 2025-09-03 Sangyu Han , Yearim Kim , Nojun Kwak

Sparse autoencoders (SAEs) enable interpretability research by decomposing entangled model activations into monosemantic features. However, under what circumstances SAEs derive most fine-grained latent features for safety, a low-frequency…

Machine Learning · Computer Science 2026-04-15 Jiaqi Weng , Han Zheng , Hanyu Zhang , Ej Zhou , Qinqin He , Jialing Tao , Hui Xue , Zhixuan Chu , Xiting Wang

Large Language Models (LLMs) have achieved strong complex reasoning capabilities through Chain-of-Thought (CoT) reasoning. However, their reasoning patterns remain too complicated to analyze. While Sparse Autoencoders (SAEs) have emerged as…

Machine Learning · Computer Science 2026-03-04 Xuan Yang , Jiayu Liu , Yuhang Lai , Hao Xu , Zhenya Huang , Ning Miao

Sparse autoencoders (SAEs) are an unsupervised method for learning a sparse decomposition of a neural network's latent representations into seemingly interpretable features. Despite recent excitement about their potential, research…

Transformer models have become state-of-the-art in decoding stimuli and behavior from neural activity, significantly advancing neuroscience research. Yet greater transparency in their decision-making processes would substantially enhance…

Quantitative Methods · Quantitative Biology 2025-06-18 Laurence Freeman , Philip Shamash , Vinam Arora , Caswell Barry , Tiago Branco , Eva Dyer

We present Universal Sparse Autoencoders (USAEs), a framework for uncovering and aligning interpretable concepts spanning multiple pretrained deep neural networks. Unlike existing concept-based interpretability methods, which focus on a…

Computer Vision and Pattern Recognition · Computer Science 2026-03-20 Harrish Thasarathan , Julian Forsyth , Thomas Fel , Matthew Kowal , Konstantinos G. Derpanis

Sparse autoencoders (SAEs) decompose neural activations into interpretable features. A widely adopted variant, the TopK SAE, reconstructs each token from its K most active latents. However, this approach is inefficient, as some tokens carry…

Machine Learning · Computer Science 2025-09-01 Narmeen Oozeer , Nirmalendu Prakash , Michael Lan , Alice Rigg , Amirali Abdullah

The high-level concepts that a neural network uses to perform computation need not be aligned to individual neurons (Smolensky, 1986). Language model interpretability research has thus turned to techniques such as \textit{sparse…

Computation and Language · Computer Science 2026-02-02 Aryaman Arora , Zhengxuan Wu , Jacob Steinhardt , Sarah Schwettmann

Sensitive directions experiments attempt to understand the computational features of Language Models (LMs) by measuring how much the next token prediction probabilities change by perturbing activations along specific directions. We extend…

Machine Learning · Computer Science 2024-11-19 Daniel J. Lee , Stefan Heimersheim