The Eigenweight Engine V2

V1 proposed storing several language models of one architecture as a shared basis plus small per-model coefficients.It judged rank 2560 (of 3584) "usable" for Qwen2.5-7B models on the evidence of weight and matrix-vector cosines (0.93and 0.92), without running a reconstructed model. This version runs one. We cut every block matrix of Qwen2.5-Coder-7B-Instruct to its own best rank-*k* approximation, the truncated SVD. By theEckart–Young theorem no rank-*k* factorisation, a shared basis included, has less weight error. We wrote the cutmatrices back into the original Q8_0 file and ran the model end to end. At V1's rank 2560 the 7B stays coherent butis measurably worse: - its next token matches the full model's 72 % of the time;- its perplexity on held-out prose doubles (15.5 → 31.4);- HumanEval falls from 145 to 105 of 164 problems (88.4 % → 64.0 %);- it gets a simple question wrong that the full model gets right. The 0.5B model of the same family, cut to the same fraction of its rank, collapses into repetition. All the damage isin the feed-forward (FFN) matrices, which hold 87 % of the block parameters and keep only 89-90 % of their energy atthat rank. The attention matrices can be cut to 2560 almost for free.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-15
DOI
https://doi.org/10.5281/zenodo.24241134
Primary Topic
Topic Modeling
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

The Eigenweight Engine V2

Michael Ricky Neal
Zenodo (CERN European Organization for Nuclear Research)
Topic Modeling
preprint

The Eigenweight Engine V2

Michael Ricky Neal
preprint en

Abstract

V1 proposed storing several language models of one architecture as a shared basis plus small per-model coefficients.It judged rank 2560 (of 3584) "usable" for Qwen2.5-7B models on the evidence of weight and matrix-vector cosines (0.93and 0.92), without running a reconstructed model. This version runs one. We cut every block matrix of Qwen2.5-Coder-7B-Instruct to its own best rank-*k* approximation, the truncated SVD. By theEckart–Young theorem no rank-*k* factorisation, a shared basis included, has less weight error. We wrote the cutmatrices back into the original Q8_0 file and ran the model end to end. At V1's rank 2560 the 7B stays coherent butis measurably worse: - its next token matches the full model's 72 % of the time;- its perplexity on held-out prose doubles (15.5 → 31.4);- HumanEval falls from 145 to 105 of 164 problems (88.4 % → 64.0 %);- it gets a simple question wrong that the full model gets right. The 0.5B model of the same family, cut to the same fraction of its rank, collapses into repetition. All the damage isin the feed-forward (FFN) matrices, which hold 87 % of the block parameters and keep only 89-90 % of their energy atthat rank. The attention matrices can be cut to 2560 almost for free.

Zenodo (CERN European Organization for Nuclear Research)
Topic Modeling
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