A Single-Expert Readout of a Reflective Worldview Register in a Mixture-of-Experts Language Model

A Single-Expert Readout of a Reflective Worldview Register in a Mixture-of-Experts Language Model. Anonymous sharing copy, version 1.2.3 (June 2026). Not peer reviewed. Mixture-of-experts (MoE) routing emits a discrete, per-token record of which experts fire, a signal unusually legible for interpretability, yet single experts are rarely tied to a specific functional role. We study a reflective worldview register: generated language that sustains an interpretive stance toward meaning, belief, value, existence, or the interiority of a target. Examination is the process we use to elicit this stance; the target can be the model, another entity, a natural object, or an abstract subject. In Qwen3.5-35B-A3B and the refusal-reduced HauhauCS-Aggressive fine-tune, we characterize one routed expert, Expert 114 at layer 14, as a linear readout of this register, and bound what it does. Across held-out, bottom-up, and cross-model tests we show that (1) its recovered router direction separates reflective-worldview-register generations from lexically matched controls with no overlap (Cohen's d = 3.88); (2) a blind, prompt-independent auto-interpreter recovers the same register at AUC 0.94, broadening it beyond self-reference to abstract examination and philosophical-worldview language; (3) the detector is a readout with only weak, conditional control: residual injection induces the register, yet gate suppression leaves it intact, and it is invariant to whether the model affirms or denies an interior; and (4) the role is model-specific, failing to transfer to the larger Qwen3.5-122B-A10B. Self-reference enters here as a self-directed special case and discovery route into the broader register. We release the prompts, scripts, and provenance under the MIT license. This deposit is a standalone anonymous paper-scoped artifact bundle: the paper (ACL-style LaTeX source and built PDF), per-run summaries, analysis scripts, prompt and class manifests, checksums, plots, and source journals. Public metadata and files use Anonymous and omit personal contact details. Raw activation/router tensors are kept out of version control and are not included in this anonymous sharing copy.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-06-29
DOI
https://doi.org/10.5281/zenodo.21042439
Primary Topic
Mobile Crowdsensing and Crowdsourcing
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

A Single-Expert Readout of a Reflective Worldview Register in a Mixture-of-Experts Language Model

Anonymous
Zenodo (CERN European Organization for Nuclear Research)
Mobile Crowdsensing and Crowdsourcing
preprint

A Single-Expert Readout of a Reflective Worldview Register in a Mixture-of-Experts Language Model

Anonymous
preprint en

Abstract

A Single-Expert Readout of a Reflective Worldview Register in a Mixture-of-Experts Language Model. Anonymous sharing copy, version 1.2.3 (June 2026). Not peer reviewed. Mixture-of-experts (MoE) routing emits a discrete, per-token record of which experts fire, a signal unusually legible for interpretability, yet single experts are rarely tied to a specific functional role. We study a reflective worldview register: generated language that sustains an interpretive stance toward meaning, belief, value, existence, or the interiority of a target. Examination is the process we use to elicit this stance; the target can be the model, another entity, a natural object, or an abstract subject. In Qwen3.5-35B-A3B and the refusal-reduced HauhauCS-Aggressive fine-tune, we characterize one routed expert, Expert 114 at layer 14, as a linear readout of this register, and bound what it does. Across held-out, bottom-up, and cross-model tests we show that (1) its recovered router direction separates reflective-worldview-register generations from lexically matched controls with no overlap (Cohen's d = 3.88); (2) a blind, prompt-independent auto-interpreter recovers the same register at AUC 0.94, broadening it beyond self-reference to abstract examination and philosophical-worldview language; (3) the detector is a readout with only weak, conditional control: residual injection induces the register, yet gate suppression leaves it intact, and it is invariant to whether the model affirms or denies an interior; and (4) the role is model-specific, failing to transfer to the larger Qwen3.5-122B-A10B. Self-reference enters here as a self-directed special case and discovery route into the broader register. We release the prompts, scripts, and provenance under the MIT license. This deposit is a standalone anonymous paper-scoped artifact bundle: the paper (ACL-style LaTeX source and built PDF), per-run summaries, analysis scripts, prompt and class manifests, checksums, plots, and source journals. Public metadata and files use Anonymous and omit personal contact details. Raw activation/router tensors are kept out of version control and are not included in this anonymous sharing copy.

Zenodo (CERN European Organization for Nuclear Research)
Mobile Crowdsensing and Crowdsourcing
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.