Metraipolis › archive › dc_3e4ce4403207

"As a Language Model...": Chat Template Switches LLM Self-Referential Voice and Activation Steering Reproduces It — arXiv:2609.25021

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.25021
License (as archived)
CC0-1.0
Archived
2026-10-08T13:45:55.262Z
SHA-256 of the archived text
f6838a38b99a9416259993bf02f986b4c2dd6c32df5468053ce5595358ab44c7
Stable URL of this page
https://metraipolis.com/documento/dc_3e4ce4403207

How to cite

"As a Language Model...": Chat Template Switches LLM Self-Referential Voice and Activation Steering Reproduces It — arXiv:2609.25021 — arXiv (en), https://arxiv.org/abs/2609.25021 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3e4ce4403207 (Metraipolis archive dc_3e4ce4403207, sha256 f6838a38b99a9416259993bf02f986b4c2dd6c32df5468053ce5595358ab44c7)
BibTeX
@misc{dc3e4ce4403207,
  title = {"As a Language Model...": Chat Template Switches LLM Self-Referential Voice and Activation Steering Reproduces It — arXiv:2609.25021},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.25021},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3e4ce4403207; sha256 f6838a38b99a9416259993bf02f986b4c2dd6c32df5468053ce5595358ab44c7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.25021. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large Language Models (LLMs) tend to add disclaimers like "I'm just an AI" when asked about something related to themselves. The self-reports from such responses are used in debates about AI safety or self-knowledge of the models, yet what drives them is not well understood. Are the models telling us about themselves or rather how they are deployed? In this work, we show that the chat template works like a switch - when present, it turns this disclaimer voice up and experiential voice like "I feel" down, across 8 popular open-source instruct models up to 9B parameters in size. And conversely when the chat template is not present, it turns the disclaimer voice down and experiential voice up. Inside the activations of 3 models, we find a direction that steers this behavior. Removing the direction in the model's activation space turns disclaimer voice down and adding it turns it up, while a random direction of the same size has little effect. We find that instruct models without chat template, when we add the disclaimer direction to them, disclaim like the template was there. Since the chat template controls the disclaimer voice of LLMs, then researchers studying self-reports or introspection of models might have a confound they need to control for. Our results show that there is a direction they can use to steer this voice. More broadly, our work shows that what models say about themselves is not a fact about them. What they say doesn't come only from weights, but it is partially set by the chat template, and because of that a model's self-description shouldn't be treated literally.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-23 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.25021 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T13:18:38.186Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).