2 citations · 5 across the 8 of their papers we have counts for
8 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…
Beyond the Rosetta Stone: Unification Forces in Generalization Dynamics
Carter Blum, Katja Filippova, Ann Yuan +8
Large language models (LLMs) struggle with cross-lingual knowledge transfer: they sometimes hallucinate when asked in one language about facts expressed in a different language dur…
Towards Unifying Interpretability and Control: Evaluation via Intervention
Usha Bhalla, Suraj Srinivas, Asma Ghandeharioun +1
With the growing complexity and capability of large language models, a need to understand model reasoning has emerged, often motivated by an underlying goal of controlling and alig…
Racing Thoughts: Explaining Contextualization Errors in Large Language Models
Michael A. Lepori, Michael C. Mozer, Asma Ghandeharioun
The profound success of transformer-based language models can largely be attributed to their ability to integrate relevant contextual information from an input sequence in order to…
When Can Transformers Count to n?
Gilad Yehudai, Haim Kaplan, Guy Dar +4
Large language models based on the transformer architecture can solve highly complex tasks, yet their fundamental limitations on simple algorithmic problems remain poorly understoo…
Who's asking? User personas and the mechanics of latent misalignment
Asma Ghandeharioun, Ann Yuan, Marius Guerard +3
Despite investments in improving model safety, studies show that misaligned capabilities remain latent in safety-tuned models. In this work, we shed light on the mechanics of this…
Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models
Asma Ghandeharioun, Avi Caciularu, Adam Pearce +2
Understanding the internal representations of large language models (LLMs) can help explain models' behavior and verify their alignment with human values. Given the capabilities of…
Interpretability Illusions in the Generalization of Simplified Models
Dan Friedman, Andrew Lampinen, Lucas Dixon +2
A common method to study deep learning systems is to use simplified model representations--for example, using singular value decomposition to visualize the model's hidden states in…