English
Related papers

Related papers: Transcoders Find Interpretable LLM Feature Circuit…

200 papers

Large Language Models (LLMs) have demonstrated unprecedented capabilities across various natural language processing tasks. Their ability to process and generate viable text and code has made them ubiquitous in many fields, while their…

Machine Learning · Computer Science 2025-05-13 Stef De Sabbata , Stefano Mizzaro , Kevin Roitero

Neural network models have achieved high performance on a wide variety of complex tasks, but the algorithms that they implement are notoriously difficult to interpret. It is often necessary to hypothesize intermediate variables involved in…

Computation and Language · Computer Science 2025-02-13 Michael A. Lepori , Thomas Serre , Ellie Pavlick

The remarkable capabilities of modern large language models are rooted in their vast repositories of knowledge encoded within their parameters, enabling them to perceive the world and engage in reasoning. The inner workings of how these…

Computation and Language · Computer Science 2025-01-06 Yunzhi Yao , Ningyu Zhang , Zekun Xi , Mengru Wang , Ziwen Xu , Shumin Deng , Huajun Chen

In recent years, many interpretability methods have been proposed to help interpret the internal states of Transformer-models, at different levels of precision and complexity. Here, to analyze encoder-decoder Transformers, we propose a…

Computation and Language · Computer Science 2024-04-04 Anna Langedijk , Hosein Mohebbi , Gabriele Sarti , Willem Zuidema , Jaap Jumelet

What latent features are encoded in language model (LM) representations? Recent work on training sparse autoencoders (SAEs) to disentangle interpretable features in LM representations has shown significant promise. However, evaluating the…

Transformer-based language models (LMs) can perform a wide range of tasks, and mechanistic interpretability (MI) aims to reverse engineer the components responsible for task completion to understand their behavior. Previous MI research has…

Computation and Language · Computer Science 2025-08-25 Karim Saraipour , Shichang Zhang

Sparse Autoencoders (SAEs) have proven to be powerful tools for interpreting neural networks by decomposing hidden representations into disentangled, interpretable features via sparsity constraints. However, conventional SAEs are…

We study the challenge of achieving theoretically grounded feature recovery using Sparse Autoencoders (SAEs) for the interpretation of Large Language Models. Existing SAE training algorithms often lack rigorous mathematical guarantees and…

Machine Learning · Computer Science 2025-06-18 Siyu Chen , Heejune Sheen , Xuyuan Xiong , Tianhao Wang , Zhuoran Yang

Locating and editing knowledge in large language models (LLMs) is crucial for enhancing their accuracy, safety, and inference rationale. We introduce ``concept editing'', an innovative variation of knowledge editing that uncovers…

Computation and Language · Computer Science 2024-08-23 Nura Aljaafari , Danilo S. Carvalho , André Freitas

Prototypical part learning is emerging as a promising approach for making semantic segmentation interpretable. The model selects real patches seen during training as prototypes and constructs the dense prediction map based on the similarity…

Computer Vision and Pattern Recognition · Computer Science 2025-04-29 Hugo Porta , Emanuele Dalsasso , Diego Marcos , Devis Tuia

Modern LLMs face inference efficiency challenges due to their scale. To address this, many compression methods have been proposed, such as pruning and quantization. However, the effect of compression on a model's interpretability remains…

Machine Learning · Computer Science 2025-07-23 Suchit Gupte , Vishnu Kabir Chhabra , Mohammad Mahdi Khalili

Automated mechanistic interpretation research has attracted great interest due to its potential to scale explanations of neural network internals to large models. Existing automated circuit discovery work relies on activation patching or…

Artificial Intelligence · Computer Science 2025-03-04 Aliyah R. Hsu , Georgia Zhou , Yeshwanth Cherapanamjeri , Yaxuan Huang , Anobel Y. Odisho , Peter R. Carroll , Bin Yu

Sparse autoencoders (SAEs) provide a powerful mechanism for decomposing the dense representations produced by Large Language Models (LLMs) into interpretable latent features. We posit that SAEs constitute a natural foundation for Learned…

Machine Learning · Computer Science 2026-03-17 Thibault Formal , Maxime Louis , Hervé Dejean , Stéphane Clinchant

Large Language Models such as GPTs (Generative Pre-trained Transformers) exhibit remarkable capabilities across a broad spectrum of applications. Nevertheless, due to their intrinsic complexity, these models present substantial challenges…

Machine Learning · Computer Science 2024-10-17 Ashkan Golgoon , Khashayar Filom , Arjun Ravi Kannan

Large neural networks achieve state-of-the-art performance on many tasks, yet their sheer size hinders deployment on resource-constrained devices. Among existing compression approaches, cross-layer parameter sharing remains relatively…

Machine Learning · Computer Science 2026-05-26 Cem Üyük , Mike Lasby , Mohamed Yassin , Utku Evci , Yani Ioannou

Sparse autoencoders (SAEs) have proven useful in disentangling the opaque activations of neural networks, primarily large language models, into sets of interpretable features. However, adapting them to domains beyond language, such as…

Machine Learning · Computer Science 2025-11-13 Ege Erdogan , Ana Lucic

Which components in transformer language models are responsible for discourse understanding? We hypothesize that sparse computational graphs, termed as discursive circuits, control how models process discourse relations. Unlike simpler…

Computation and Language · Computer Science 2025-10-14 Yisong Miao , Min-Yen Kan

Pretrained transformer-based Language Models (LMs) are well-known for their ability to achieve significant improvement on NLP tasks, but their black-box nature, which leads to a lack of interpretability, has been a major concern. My…

Computation and Language · Computer Science 2024-12-06 Ximing Wen

Recent work has found that sparse autoencoders (SAEs) are an effective technique for unsupervised discovery of interpretable features in language models' (LMs) activations, by finding sparse, linear reconstructions of LM activations. We…

Machine Learning · Computer Science 2024-05-01 Senthooran Rajamanoharan , Arthur Conmy , Lewis Smith , Tom Lieberum , Vikrant Varma , János Kramár , Rohin Shah , Neel Nanda

While Large Language Models (LLMs) have achieved strong performance across many NLP tasks, their opaque internal mechanisms hinder trustworthiness and safe deployment. Existing surveys in explainable AI largely focus on post-hoc explanation…

Computation and Language · Computer Science 2026-04-21 Yutong Gao , Qinglin Meng , Yuan Zhou , Liangming Pan