Jannik Brinkmann

I am a Ph.D. student interested in the structure and interpretation of natural and artificial systems. I am advised by Christian Bartelt and affiliated with the Interpretable Neural Networks group, led by David Bau.

Google Scholar GitHub

Selected Publications

ICLR 2026 Jailbreak Transferability Emerges From Shared Representations
Computational Linguistics 2025 The Quest for the Right Mediator: A History, Survey, and Theoretical Grounding of Causal Interpretability
NAACL 2025 · Oral Large Language Models Share Representations of Latent Grammatical Concepts across Typologically Diverse Languages
ICLR 2025 NNsight and NDIF: Democratizing Access to Open-Weight Foundation Model Internals
ACL 2024 A Mechanistic Analysis of a Transformer Trained on a Symbolic Multi-Step Reasoning Task

Positions

Mar 2026
OpenAI
Red-Teaming Contractor
Jun 2025 – Oct 2025
J.P. Morgan Research
Research Intern
Nov 2024 – Mar 2025
New York University
Visiting Researcher
Apr 2024 – Aug 2024
Northeastern University
Visiting Researcher