4 papers
Abstractive Red-Teaming of Language Model Character
Nate Rahn, Allison Qi, Avery Griffin +3
We want language model assistants to conform to a character specification, which asserts how the model should act across diverse user interactions. While models typically follow th…
Eliciting Harmful Capabilities by Fine-Tuning On Safeguarded Outputs
Jackson Kaunismaa, Avery Griffin, John Hughes +3
Model developers implement safeguards in frontier models to prevent misuse, for example, by employing classifiers to filter dangerous outputs. In this work, we demonstrate that eve…
Optimizing AI Agent Attacks With Synthetic Data
Chloe Loughridge, Paul Colognese, Avery Griffin +3
As AI deployments become more complex and high-stakes, it becomes increasingly important to be able to estimate their risk. AI control is one framework for doing so. However, good…
The Local Interaction Basis: Identifying Computationally-Relevant and Sparsely Interacting Features in Neural Networks
Lucius Bushnaq, Stefan Heimersheim, Nicholas Goldowsky-Dill +7
Mechanistic interpretability aims to understand the behavior of neural networks by reverse-engineering their internal computations. However, current methods struggle to find clear…