English
Related papers

Related papers: Polysemanticity and Capacity in Neural Networks

200 papers

Neural networks often pack many unrelated concepts into a single neuron - a puzzling phenomenon known as 'polysemanticity' which makes interpretability much more challenging. This paper provides a toy model where polysemanticity can be…

Despite substantial efforts, neural network interpretability remains an elusive goal, with previous research failing to provide succinct explanations of most single neurons' impact on the network output. This limitation is due to the…

Machine Learning · Computer Science 2024-02-01 Simon C. Marshall , Jan H. Kirchner

We present neuron embeddings, a representation that can be used to tackle polysemanticity by identifying the distinct semantic behaviours in a neuron's characteristic dataset examples, making downstream manual or automatic interpretation…

Machine Learning · Computer Science 2024-11-14 Alex Foote

In some neural networks, individual neurons correspond to natural ``features'' in the input. Such \emph{monosemantic} neurons are of great help in interpretability studies, as they can be cleanly understood. In this work we report…

Machine Learning · Computer Science 2022-11-18 Adam S. Jermyn , Nicholas Schiefer , Evan Hubinger

Polysemantic neurons -- neurons that activate for a set of unrelated features -- have been seen as a significant obstacle towards interpretability of task-optimized deep networks, with implications for AI safety. The classic origin story of…

Machine Learning · Computer Science 2024-02-14 Victor Lecomte , Kushal Thaman , Rylan Schaeffer , Naomi Bashkansky , Trevor Chow , Sanmi Koyejo

Mechanistic interpretability aims to explain what a neural network has learned at a nuts-and-bolts level. What are the fundamental primitives of neural network representations? Previous mechanistic descriptions have used individual neurons…

One of the roadblocks to a better understanding of neural networks' internals is \textit{polysemanticity}, where neurons appear to activate in multiple, semantically distinct contexts. Polysemanticity prevents us from identifying concise,…

Machine Learning · Computer Science 2023-10-05 Hoagy Cunningham , Aidan Ewart , Logan Riggs , Robert Huben , Lee Sharkey

Deep learning models often suffer from a lack of interpretability due to polysemanticity, where individual neurons are activated by multiple unrelated semantics, resulting in unclear attributions of model behavior. Recent advances in…

Machine Learning · Computer Science 2024-10-30 Qi Zhang , Yifei Wang , Jingyi Cui , Xiang Pan , Qi Lei , Stefanie Jegelka , Yisen Wang

Mechanistic interpretability aims to understand how models store representations by breaking down neural networks into interpretable units. However, the occurrence of polysemantic neurons, or neurons that respond to multiple unrelated…

Computer Vision and Pattern Recognition · Computer Science 2023-04-20 Laura O'Mahony , Vincent Andrearczyk , Henning Muller , Mara Graziani

To better interpret the intrinsic mechanism of large language models (LLMs), recent studies focus on monosemanticity on its basic units. A monosemantic neuron is dedicated to a single and specific concept, which forms a one-to-one…

Computation and Language · Computer Science 2024-10-17 Hanqi Yan , Yanzheng Xiang , Guangyi Chen , Yifei Wang , Lin Gui , Yulan He

During the last decade, entity embeddings have become ubiquitous in Artificial Intelligence. Such embeddings essentially serve as compact but semantically meaningful representations of the entities of interest. In most approaches, vectors…

Artificial Intelligence · Computer Science 2021-09-15 Steven Schockaert

Designing neural network architectures is a task that lies somewhere between science and art. For a given task, some architectures are eventually preferred over others, based on a mix of intuition, experience, experimentation and luck. For…

Machine Learning · Computer Science 2019-02-13 Jonathan Donier

Multi-modal Large Language Models (MLLMs) have demonstrated remarkable capabilities in understanding and generating content across various modalities, such as images and text. However, their interpretability remains a challenge, hindering…

Computer Vision and Pattern Recognition · Computer Science 2024-05-29 Loris Giulivi , Giacomo Boracchi

The polysemantic nature of synthetic neurons in artificial intelligence language models is currently understood as the result of a necessary superposition of distributed features within the latent space. We propose an alternative approach,…

High quality explanations of neural networks (NNs) should exhibit two key properties. Completeness ensures that they accurately reflect a network's function and interpretability makes them understandable to humans. Many existing methods…

Machine Learning · Computer Science 2025-03-20 Nolan Dey , Eric Taylor , Alexander Wong , Bryan Tripp , Graham W. Taylor

Set representation has become ubiquitous in deep learning for modeling the inductive bias of neural networks that are insensitive to the input order. DeepSets is the most widely used neural network architecture for set representation. It…

Machine Learning · Computer Science 2024-03-08 Peihao Wang , Shenghao Yang , Shu Li , Zhangyang Wang , Pan Li

Being able to interpret, or explain, the predictions made by a machine learning model is of fundamental importance. This is especially true when there is interest in deploying data-driven models to make high-stakes decisions, e.g. in…

Machine Learning · Computer Science 2019-10-01 An-phi Nguyen , María Rodríguez Martínez

Mechanistic interpretability is concerned with analyzing individual components in a (convolutional) neural network (CNN) and how they form larger circuits representing decision mechanisms. These investigations are challenging since CNNs…

Computer Vision and Pattern Recognition · Computer Science 2025-04-18 Robin Hesse , Jonas Fischer , Simone Schaub-Meyer , Stefan Roth

Convolutional neural networks use pooling and other downscaling operations to maintain translational invariance for detection of features, but in their architecture they do not explicitly maintain a representation of the locations of the…

Computer Vision and Pattern Recognition · Computer Science 2021-03-16 Prem Nair , Rohan Doshi , Stefan Keselj

Interpretability benefits the theoretical understanding of representations. Existing word embeddings are generally dense representations. Hence, the meaning of latent dimensions is difficult to interpret. This makes word embeddings like a…

Computation and Language · Computer Science 2023-06-27 Minxue Xia , Hao Zhu
‹ Prev 1 2 3 10 Next ›