Achyuta Rajaram — accepted AI-conference papers

2026

Persona Features Control Emergent Misalignment

Miles Wang, Tom Dupre la Tour, Olivia Watkins, Aleksandar Makelov, Ryan Andrew Chi, Samuel Miserendino +5

ICLR 2026poster

Understanding how language models generalize behaviors from their training to a broader deployment distribution is an important problem in AI safety. Betley et al. discovered that fine-tuning GPT-4o on intentionally insecure code causes "emergent misalignment," where models give stereotypically mali…

Cited by 0 Source ↗code ↗Scholar ↗

2026

Weight-sparse transformers have interpretable circuits

Leo Gao, Achyuta Rajaram, Jacob Coxon, Soham Govande, Bowen Baker, Daniel Mossing

ICML 2026poster

Finding human-understandable circuits in language models is a central goal of the field of mechanistic interpretability. We train models to have more understandable circuits by constraining most of their weights to be zeros, so that each neuron only has a few connections. To recover fine-grained cir…

Cited by 0 Source ↗Scholar ↗

2024

A Multimodal Automated Interpretability Agent

Tamar Rott Shaham, Sarah Schwettmann, Franklin Wang, Achyuta Rajaram, Evan Hernandez, Jacob Andreas +1

ICML 2024poster

This paper describes MAIA, a Multimodal Automated Interpretability Agent. MAIA is a system that uses neural models to automate neural model understanding tasks like feature interpretation and failure mode discovery. It equips a pre-trained vision-language model with a set of tools that support itera…

Cited by 67 Source ↗PDF ↗Scholar ↗