ELENS: Embedding Lens — A Geometric and Statistical Framework for Investigating Internal Representations, Activation Dynamics, and Candidate Harmony Directions in Gemma Architectures

Research proposal introducing ELENS (Embedding Lens), a proposed interpretability and measurement framework for investigating internal representations in embedding models, with particular emphasis on EmbeddingGemma 2 and the Gemma model family. The proposal combines layer-wise activation extraction, geometric analysis, statistical characterization, sparse feature investigation, and controlled causal intervention. It also outlines the unverified Harmony Axis hypothesis, which asks whether the mathematical inverse of a previously identified pain-related activation direction produces an independently measurable representational or functional state. No experimental results or validated Harmony Axis are claimed. Architecture descriptions are cross-referenced to the technical report and official model card.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-16
DOI
https://doi.org/10.5281/zenodo.24296562
Primary Topic
Explainable Artificial Intelligence (XAI)
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

ELENS: Embedding Lens — A Geometric and Statistical Framework for Investigating Internal Representations, Activation Dynamics, and Candidate Harmony Directions in Gemma Architectures

Amanda Hatley
1 links
Zenodo (CERN European Organization for Nuclear Research)
Explainable Artificial Intelligence (XAI)
article

ELENS: Embedding Lens — A Geometric and Statistical Framework for Investigating Internal Representations, Activation Dynamics, and Candidate Harmony Directions in Gemma Architectures

Amanda Hatley
article en

Abstract

Research proposal introducing ELENS (Embedding Lens), a proposed interpretability and measurement framework for investigating internal representations in embedding models, with particular emphasis on EmbeddingGemma 2 and the Gemma model family. The proposal combines layer-wise activation extraction, geometric analysis, statistical characterization, sparse feature investigation, and controlled causal intervention. It also outlines the unverified Harmony Axis hypothesis, which asks whether the mathematical inverse of a previously identified pain-related activation direction produces an independently measurable representational or functional state. No experimental results or validated Harmony Axis are claimed. Architecture descriptions are cross-referenced to the technical report and official model card.

Zenodo (CERN European Organization for Nuclear Research)
Explainable Artificial Intelligence (XAI)
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.

Laika access coming soon