WGWes Gurnee

Wes Gurnee

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Wes Gurnee has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Combating Gerrymandering with Ranked Choice Voting: An Experimental Analysis of Multimember Districts in the United States

    by Nikhil Garg, Wes Gurnee, David Rothschild, David B. Shmoys

    Blunting Gerrymandering with Multimember Districts How much can election rules blunt the power of partisan mapmakers? In the Operations Research paper “Combating Gerrymandering with Ranked Choice Voting: An Experimental Analysis of Multimember Districts in the United States,” Nikhil Garg, Wes Gurnee, David Rothschild, and David B. Shmoys study this question by combining computational redistricting with models of multiwinner voting. The authors compare single-member districts with multimember districts elected under rules such as single transferable vote (ranked choice voting), using algorithmically generated maps for the U.S. House of Representatives. Their results suggest that even modest three-member districts can help independent commissions reach proportional outcomes, sharply reducing the range of outcomes available to partisan gerrymanderers. The study also tests robustness under voter crossover and explores trade-offs involving geographic cohesion and within-party diversity. The paper points to election-system design as an operations research problem with direct democratic consequences.


  • Open Access

    Language Models Represent Space and Time

    by Wes Gurnee, Max Tegmark

    The capabilities of large language models (LLMs) have sparked debate over whether such systems just learn an enormous collection of superficial statistics or a set of more coherent and grounded representations that reflect the real world. We find evidence for the latter by analyzing the learned representations of three spatial datasets (world, US, NYC places) and three temporal datasets (historical figures, artworks, news headlines) in the Llama-2 family of models. We discover that LLMs learn linear representations of space and time across multiple scales. These representations are robust to prompting variations and unified across different entity types (e.g. cities and landmarks). In addition, we identify individual "space neurons" and "time neurons" that reliably encode spatial and temporal coordinates. While further investigation is needed, our results suggest modern LLMs learn rich spatiotemporal representations of the real world and possess basic ingredients of a world model.


  • Open Access

    Finding Neurons in a Haystack: Case Studies with Sparse Probing

    by Wes Gurnee, Neel Nanda, Matthew Pauly, Katherine Harvey, Dmitrii Troitskii, Dimitris Bertsimas

    Despite rapid adoption and deployment of large language models (LLMs), the internal computations of these models remain opaque and poorly understood. In this work, we seek to understand how high-level human-interpretable features are represented within the internal neuron activations of LLMs. We train $k$-sparse linear classifiers (probes) on these internal activations to predict the presence of features in the input; by varying the value of $k$ we study the sparsity of learned representations and how this varies with model scale. With $k=1$, we localize individual neurons which are highly relevant for a particular feature, and perform a number of case studies to illustrate general properties of LLMs. In particular, we show that early layers make use of sparse combinations of neurons to represent many features in superposition, that middle layers have seemingly dedicated neurons to represent higher-level contextual features, and that increasing scale causes representational sparsity to increase on average, but there are multiple types of scaling dynamics. In all, we probe for over 100 unique features comprising 10 different categories in 7 different models spanning 70 million to 6.9 billion parameters.


  • Top Cited Papers
    Open Access

    Learning sparse nonlinear dynamics via mixed-integer optimization

    by Dimitris Bertsimas, Wes Gurnee

    Discovering governing equations of complex dynamical systems directly from data is a central problem in scientific machine learning. In recent years, the sparse identification of nonlinear dynamics (SINDy) framework, powered by heuristic sparse regression methods, has become a dominant tool for learning parsimonious models. We propose an exact formulation of the SINDy problem using mixed-integer optimization (MIO-SINDy) to solve the sparsity constrained regression problem to provable optimality in seconds. On a large number of canonical ordinary and partial differential equations, we illustrate the dramatic improvement in our approach in accurate model discovery while being more sample efficient, robust to noise, and flexible in accommodating physical constraints.