# The Assistant Axis: Situating and Stabilizing the Default Persona of Language Models

arXiv: https://arxiv.org/abs/2601.10387 · HTML: https://mlsanity.com/paper/2601.10387/

## Key facts

- Cited by 90 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Jan 2026.

## Built on by (most cited first, 30 of 90)

- [Verbalizable Representations Form a Global Workspace in Language Models](https://mlsanity.com/paper/2607.15495/) (2026)
- [Emotion Concepts and their Function in a Large Language Model](https://mlsanity.com/paper/2604.07729/) (2026)
- [Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models](https://mlsanity.com/paper/2601.14004/) (2026)
- [Subliminal Effects in Your Data: A General Mechanism via Log-Linearity](https://mlsanity.com/paper/2602.04863/) (2026)
- [Exploring Silicon-Based Societies: An Early Study of the Moltbook Agent Community](https://mlsanity.com/paper/2602.02613/) (2026)
- [Linear representations in language models can change dramatically over a conversation](https://mlsanity.com/paper/2601.20834/) (2026)
- [Where is the Mind? Persona Vectors and LLM Individuation](https://mlsanity.com/paper/2604.17031/) (2026)
- [When Agents Commit Too Soon: Diagnosing Premature Commitment in LLM Agents](https://mlsanity.com/paper/2606.22936/) (2026)
- [Steered LLM Activations are Non-Surjective](https://mlsanity.com/paper/2604.09839/) (2026)
- [Evaluating Alignment of Behavioral Dispositions in LLMs](https://mlsanity.com/paper/2602.11328/) (2026)
- [Tracing Persona Vectors Through LLM Pretraining](https://mlsanity.com/paper/2605.13329/) (2026)
- [TACT: Mitigating Overthinking and Overacting in Coding Agents via Activation Steering](https://mlsanity.com/paper/2605.05980/) (2026)
- [Negation Neglect: When models fail to learn negations in training](https://mlsanity.com/paper/2605.13829/) (2026)
- [Attractor States Emerge in Multi-Turn LLM Conversations](https://mlsanity.com/paper/2606.30571/) (2026)
- [Personalization Increases Affective Alignment but Has Role-Dependent Effects on Epistemic Independence in LLMs](https://mlsanity.com/paper/2603.00024/) (2026)
- [The Consciousness Cluster: Emergent preferences of Models that Claim to be Conscious](https://mlsanity.com/paper/2604.13051/) (2026)
- [ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions](https://mlsanity.com/paper/2605.24279/) (2026)
- [Alignment midtraining for animals](https://mlsanity.com/paper/2604.13076/) (2026)
- [Playing Devil's Advocate: Off-the-Shelf Persona Vectors Rival Targeted Steering for Sycophancy](https://mlsanity.com/paper/2605.21006/) (2026)
- [Dissociative Identity: Language Model Agents Lack Grounding for Reputation Mechanisms](https://mlsanity.com/paper/2605.30169/) (2026)
- [Persona Cartography: Charting Language Model Personality Traits in Weight Space](https://mlsanity.com/paper/2607.07916/) (2026)
- [Mind Viruses: Self-Propagating Ideas in Multi-Agent LLM Systems](https://mlsanity.com/paper/2608.10218/) (2026)
- [No Reliable Evidence of Self-Reported Sentience in Small Large Language Models](https://mlsanity.com/paper/2601.15334/) (2026)
- [Probing Persona-Dependent Preferences in Language Models](https://mlsanity.com/paper/2605.13339/) (2026)
- [Reinforcement Learning Can Amplify Emergent Misalignment from Harmless Rewards](https://mlsanity.com/paper/2605.31328/) (2026)
- [Emergent Misalignment Recruits a Pre-existing Persona Subspace](https://mlsanity.com/paper/2607.21356/) (2026)
- [Affective Context Amplifies Sycophancy in LLM Responses](https://mlsanity.com/paper/2608.21242/) (2026)
- [MOSAIC: Unveiling the Moral, Social and Individual Dimensions of Large Language Models](https://mlsanity.com/paper/2603.00048/) (2026)
- [Identity as Attractor: Geometric Evidence for Persistent Agent Architecture in LLM Activation Space](https://mlsanity.com/paper/2604.12016/) (2026)
- [Low-Resource Safety Failures Are Action Failures, Not Representation Failures](https://mlsanity.com/paper/2606.01196/) (2026)

## Abstract (arXiv)

Large language models can represent a variety of personas but typically default to a helpful Assistant identity cultivated during post-training. We investigate the structure of the space of model personas by extracting activation directions corresponding to diverse character archetypes. Across several different models, we find that the leading component of this persona space is an "Assistant Axis," which captures the extent to which a model is operating in its default Assistant mode. Steering towards the Assistant direction reinforces helpful and harmless behavior; steering away increases the model's tendency to identify as other entities. Moreover, steering away with more extreme values often induces a mystical, theatrical speaking style. We find this axis is also present in pre-trained models, where it primarily promotes helpful human archetypes like consultants and coaches and inhibits spiritual ones. Measuring deviations along the Assistant Axis predicts "persona drift," a phenomenon where models slip into exhibiting harmful or bizarre behaviors that are uncharacteristic of their typical persona. We find that persona drift is often driven by conversations demanding meta-reflection on the model's processes or featuring emotionally vulnerable users. We show that restricting activations to a fixed region along the Assistant Axis can stabilize model behavior in these scenarios -- and also in the face of adversarial persona-based jailbreaks. Our results suggest that post-training steers models toward a particular region of persona space but only loosely tethers them to it, motivating work on training and steering strategies that more deeply anchor models to a coherent persona.