# Identity as Attractor: Geometric Evidence for Persistent Agent Architecture in LLM Activation Space

arXiv: https://arxiv.org/abs/2604.12016 · HTML: https://mlsanity.com/paper/2604.12016/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 5 of its 7 references that are in the map.
- First posted on arXiv 13 Apr 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 5 of 5)

- [The Platonic Representation Hypothesis](https://mlsanity.com/paper/2405.07987/) (2024)
- [The Assistant Axis: Situating and Stabilizing the Default Persona of Language Models](https://mlsanity.com/paper/2601.10387/) (2026)
- [Transformer Dynamics: A neuroscientific approach to interpretability of large language models](https://mlsanity.com/paper/2502.12131/) (2025)
- [Your Language Model Secretly Contains Personality Subnetworks](https://mlsanity.com/paper/2602.07164/) (2026)
- [Concept Attractors in LLMs and their Applications](https://mlsanity.com/paper/2601.11575/) (2025)

## Built on by (most cited first, 2 of 2)

- [Relational Positioning as a Measurable Risk Object: History-Carried Lock-in and Self-Confabulation in Multi-Turn Human-AI Dialogue](https://mlsanity.com/paper/2607.11437/) (2026)
- [Identity Is More Than Recall: A Benchmark for Persistent Identity in Deployed AI Agents](https://mlsanity.com/paper/2609.13637/) (2026)

## Abstract (arXiv)

Large language models map semantically related prompts to similar internal representations -- a phenomenon interpretable as attractor-like dynamics. We ask whether the identity document of a persistent cognitive agent (its cognitive_core) exhibits analogous attractor-like behavior. We present a controlled experiment on Llama 3.1 8B Instruct, comparing hidden states of an original cognitive_core (Condition A), seven paraphrases (Condition B), and seven structurally matched controls (Condition C). Mean-pooled states at layers 8, 16, and 24 show that paraphrases converge to a tighter cluster than controls (Cohen's d > 1.88, p < 10^{-27}, Bonferroni-corrected). Replication on Gemma 2 9B confirms cross-architecture generalizability. Ablations suggest the effect is primarily semantic rather than structural, and that structural completeness appears necessary to reach the attractor region. An exploratory experiment shows that reading a scientific description of the agent shifts internal state toward the attractor -- closer than a sham preprint -- distinguishing knowing about an identity from operating as that identity. These results provide representational evidence that agent identity documents induce attractor-like geometry in LLM activation space.