SolidGoldMagikarp (plus, prompt generation)

MATS Fellow:

Jessica Cooper (Rumbelow), Matthew Watkins

Authors:

Jessica Cooper, Matthew Watkins

Citations

18 Citations

Abstract:

Anomalous tokens: a mysterious failure mode for GPT (which reliably insulted Matthew)

  • We have found a set of anomalous tokens which result in a previously undocumented failure mode for GPT-2 and GPT-3 models. (The 'instruct' models “are particularly deranged” in this context, as janus has observed.)
  • Many of these tokens reliably break determinism in the OpenAI GPT-3 playground at temperature 0 (which theoretically shouldn't happen).

Recent research

You Are What You Read: Misalignment via In-Context Persona Induction

Authors:

Kyuhee Kim, Benjamin Berczi

Date:

September 6, 2026

Citations:

Non-Great-Power Conflict and AI Risk

Authors:

Kristina Kempkey

Date:

August 26, 2026

Citations:

Frequently asked questions

What is the MATS Program?
How long does the program last?