4 citations · 6 across the 14 of their papers we have counts for
14 papers
Think Before You Lie: How Reasoning Leads to Honesty
Ann Yuan, Asma Ghandeharioun, Carter Blum +6
While existing evaluations of large language models (LLMs) measure deception rates, the underlying conditions that give rise to deceptive behavior are poorly understood. We investi…
Decomposing Query-Key Feature Interactions Using Contrastive Covariances
Andrew Lee, Yonatan Belinkov, Fernanda Viégas +1
Despite the central role of attention heads in Transformers, we lack tools to understand why a model attends to a particular token. To address this, we study the query-key (QK) spa…
Why Can't Transformers Learn Multiplication? Reverse-Engineering Reveals Long-Range Dependency Pitfalls
Xiaoyan Bai, Itamar Pres, Yuntian Deng +5
Language models are increasingly capable, yet still fail at a seemingly simple task of multi-digit multiplication. In this work, we study why, by reverse-engineering a model that s…
Chronotome: Real-Time Topic Modeling for Streaming Embedding Spaces
Matte Lim, Catherine Yeh, Martin Wattenberg +2
Many real-world datasets -- from an artist's body of work to a person's social media history -- exhibit meaningful semantic changes over time that are difficult to capture with exi…
Story Ribbons: Reimagining Storyline Visualizations with Large Language Models
Catherine Yeh, Tara Menon, Robin Singh Arya +4
Analyzing literature involves tracking interactions between characters, locations, and themes. Visualization has the potential to facilitate the mapping and analysis of these compl…
What Does it Mean for a Neural Network to Learn a "World Model"?
Kenneth Li, Fernanda Viégas, Martin Wattenberg
We propose a set of precise criteria for saying a neural net learns and uses a "world model." The goal is to give an operational meaning to terms that are often used informally, in…
The Geometry of Self-Verification in a Task-Specific Reasoning Model
Andrew Lee, Lihao Sun, Chris Wendler +2
How do reasoning models verify their own answers? We study this question by training a model using DeepSeek R1's recipe on the CountDown task. We leverage the fact that preference…
When Bad Data Leads to Good Models
Kenneth Li, Yida Chen, Fernanda Viégas +1
In large language model (LLM) pretraining, data quality is believed to determine model quality. In this paper, we re-examine the notion of "quality" from the perspective of pre- an…
Shared Global and Local Geometry of Language Model Embeddings
Andrew Lee, Melanie Weber, Fernanda Viégas +1
Researchers have recently suggested that models share common representations. In our work, we find numerous geometric similarities across the token embeddings of large language mod…
Relational Composition in Neural Networks: A Survey and Call to Action
Martin Wattenberg, Fernanda B. Viégas
Many neural nets appear to represent data as linear combinations of "feature vectors." Algorithms for discovering these vectors have seen impressive recent success. However, we arg…
Dialogue Action Tokens: Steering Language Models in Goal-Directed Dialogue with a Multi-Turn Planner
Kenneth Li, Yiming Wang, Fernanda Viégas +1
We present an approach called Dialogue Action Tokens (DAT) that adapts language model agents to plan goal-directed dialogues. The core idea is to treat each utterance as an action,…
Designing a Dashboard for Transparency and Control of Conversational AI
Yida Chen, Aoyu Wu, Trevor DePodesta +9
Conversational LLMs function as black box systems, leaving users guessing about why they see the output they do. This lack of transparency is potentially problematic, especially gi…
Q-Probe: A Lightweight Approach to Reward Maximization for Language Models
Kenneth Li, Samy Jelassi, Hugh Zhang +3
We present an approach called Q-probing to adapt a pre-trained language model to maximize a task-specific reward function. At a high level, Q-probing sits between heavier approache…
Measuring and Controlling Instruction (In)Stability in Language Model Dialogs
Kenneth Li, Tianle Liu, Naomi Bashkansky +4
System-prompting is a standard tool for customizing language-model chatbots, enabling them to follow a specific instruction. An implicit assumption in the use of system prompts is…