English
Related papers

Related papers: Scaling sparse feature circuit finding for in-cont…

200 papers

In-context Learning (ICL) empowers large language models (LLMs) to swiftly adapt to unseen tasks at inference-time by prefixing a few demonstration examples before queries. Despite its versatility, ICL incurs substantial computational and…

Machine Learning · Computer Science 2025-02-26 Zhuowei Li , Zihao Xu , Ligong Han , Yunhe Gao , Song Wen , Di Liu , Hao Wang , Dimitris N. Metaxas

The standard sparse-autoencoder (SAE) interpretability protocol labels each feature from its top-activating contexts and validates by single-feature steering. We propose the pairwise matrix protocol, co-varying steering coefficient with…

Machine Learning · Computer Science 2026-05-06 Michael A. Riegler , Birk Sebastian Frostelid Torpmann-Hagen

Sparse autoencoders (SAEs) have emerged as powerful techniques for interpretability of large language models (LLMs), aiming to decompose hidden states into meaningful semantic features. While several SAE variants have been proposed, there…

Machine Learning · Computer Science 2025-10-03 Xudong Zhu , Mohammad Mahdi Khalili , Zhihui Zhu

Model internals encode rich information about how a large language model (LLM) processes its training data; however, post-training data engineering largely relies on external signals and ignores rich intrinsic signals lying in model…

Machine Learning · Computer Science 2026-05-27 Yi Jing , Zao Dai , Jinwu Hu , Zijun Yao , Lei Hou , Juanzi Li , Xiaozhi Wang

Sparse autoencoders (SAEs) are a promising unsupervised approach for identifying causally relevant and interpretable linear features in a language model's (LM) activations. To be useful for downstream tasks, SAEs need to decompose LM…

Machine Learning · Computer Science 2024-08-02 Senthooran Rajamanoharan , Tom Lieberum , Nicolas Sonnerat , Arthur Conmy , Vikrant Varma , János Kramár , Neel Nanda

Zero-shot Learning (ZSL) classification categorizes or predicts classes (labels) that are not included in the training set (unseen classes). Recent works proposed different semantic autoencoder (SAE) models where the encoder embeds a visual…

Computer Vision and Pattern Recognition · Computer Science 2023-08-14 William Heyden , Habib Ullah , M. Salman Siddiqui , Fadi Al Machot

Sequence models have demonstrated the ability to perform tasks like channel equalization and symbol detection by automatically adapting to current channel conditions. This is done without requiring any explicit optimization and by…

Signal Processing · Electrical Eng. & Systems 2024-11-01 Zihang Song , Matteo Zecchin , Bipin Rajendran , Osvaldo Simeone

Large Language Models (LLMs) have demonstrated impressive in-context learning (ICL) capabilities from few-shot demonstration exemplars. While recent learning-based demonstration selection methods have proven beneficial to ICL by choosing…

Machine Learning · Computer Science 2024-10-16 Hui Liu , Wenya Wang , Hao Sun , Chris Xing Tian , Chenqi Kong , Xin Dong , Haoliang Li

Time series foundation models (TSFMs) are increasingly deployed in high-stakes domains, yet their internal representations remain opaque. We present the first application of sparse autoencoders (SAEs) to a TSFM, training TopK SAEs on…

Machine Learning · Computer Science 2026-03-12 Anurag Mishra

In-context learning (ICL) has emerged as a powerful paradigm for task adaptation in large language models (LLMs), where models infer underlying task structures from a few demonstrations. However, ICL remains susceptible to biases that arise…

Computation and Language · Computer Science 2025-06-18 Zhihang Tan , Jingrui Hou , Ping Wang , Qibiao Hu , Peng Zhu

Sparse autoencoders (SAEs) are a promising approach to extracting features from neural networks, enabling model interpretability as well as causal interventions on model internals. SAEs generate sparse feature representations using a…

Machine Learning · Computer Science 2024-11-11 Kola Ayonrinde

Sparse autoencoders (SAEs) are a technique for sparse decomposition of neural network activations into human-interpretable features. However, current SAEs suffer from feature absorption, where specialized features capture instances of…

Machine Learning · Computer Science 2025-09-29 Anton Korznikov , Andrey Galichin , Alexey Dontsov , Oleg Rogov , Elena Tutubalina , Ivan Oseledets

Altruism is fundamental to human societies, fostering cooperation and social cohesion. Recent studies suggest that large language models (LLMs) can display human-like prosocial behavior, but the internal computations that produce such…

General Economics · Economics 2026-04-22 Shuhuai Zhang , Shu Wang , Zijun Yao , Chuanhao Li , Xiaozhi Wang , Songfa Zhong , Tracy Xiao Liu

Sparse autoencoders (SAEs) have received considerable recent attention as tools for mechanistic interpretability, showing success at extracting interpretable features even from very large LLMs. However, this research has been largely…

Machine Learning · Computer Science 2025-05-20 Jeremy Budd , Javier Ideami , Benjamin Macdowall Rynne , Keith Duggar , Randall Balestriero

Transformer-based CLIP models are widely used for text-image probing and feature extraction, making it relevant to understand the internal mechanisms behind their predictions. While recent works show that Sparse Autoencoders (SAEs) yield…

Machine Learning · Computer Science 2025-05-27 Maximilian Dreyer , Lorenz Hufe , Jim Berend , Thomas Wiegand , Sebastian Lapuschkin , Wojciech Samek

Transformer-based language models exhibit In-Context Learning (ICL), where predictions are made adaptively based on context. While prior work links induction heads to ICL through a sudden jump in accuracy, this can only account for ICL when…

Computation and Language · Computer Science 2025-06-11 Gouki Minegishi , Hiroki Furuta , Shohei Taniguchi , Yusuke Iwasawa , Yutaka Matsuo

Sparse autoencoders (SAEs) decompose language model representations into a sparse set of linear latent vectors. Recent works have improved SAEs using language model gradients, but these techniques require many expensive backward passes…

Machine Learning · Computer Science 2025-05-28 Matthew Chen , Joshua Engels , Max Tegmark

Recent work on sparse autoencoders (SAEs) has shown promise in extracting interpretable features from neural networks and addressing challenges with polysemantic neurons caused by superposition. In this paper, we apply SAEs to the early…

Machine Learning · Computer Science 2024-09-10 Liv Gorton

Sparse auto-encoders (SAEs) have re-emerged as a prominent method for mechanistic interpretability, yet they face two significant challenges: the non-smoothness of the $L_1$ penalty, which hinders reconstruction and scalability, and a lack…

Artificial Intelligence · Computer Science 2026-05-19 Ouns El Harzli , Hugo Wallner , Yoonsoo Nam , Haixuan Xavier Tao

Recent work has demonstrated that transformers and linear attention models can perform in-context learning (ICL) on simple function classes, such as linear regression. In this paper, we empirically study how these two attention mechanisms…

Machine Learning · Computer Science 2026-02-20 Ayush Goel , Arjun Kohli , Sarvagya Somvanshi