Relation-State Continuity in Qwen2.5-7B Base: Evidence for an Attention-Topology Path

I probe Qwen2.5-7B base across eight relation-tracking cases and report four aggregate scores summarising the 29-layer profile: residual carry AUC 0.55, attention anchor mass AUC 0.089, formation peak 0.94, cross-layer circuit score 0.39. The formation profile is bimodal with peaks at layer 1 and layer 28, and no single intermediate layer dominates. The work of routing anchor information to query position is spread across layers 19 through 27, where residual carry rises from 0.60 to 0.81 while attention to anchors stays in the 0.12 to 0.23 range. I label this pattern attention topology path. The classification is observational; causal verification via activation patching is the next experiment.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-05-29
DOI
https://doi.org/10.5281/zenodo.20449372
Primary Topic
Advanced Memory and Neural Computing
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

Relation-State Continuity in Qwen2.5-7B Base: Evidence for an Attention-Topology Path

Ho Yiing Chen
Zenodo (CERN European Organization for Nuclear Research)
Advanced Memory and Neural Computing
preprint

Relation-State Continuity in Qwen2.5-7B Base: Evidence for an Attention-Topology Path

Ho Yiing Chen
preprint en

Abstract

I probe Qwen2.5-7B base across eight relation-tracking cases and report four aggregate scores summarising the 29-layer profile: residual carry AUC 0.55, attention anchor mass AUC 0.089, formation peak 0.94, cross-layer circuit score 0.39. The formation profile is bimodal with peaks at layer 1 and layer 28, and no single intermediate layer dominates. The work of routing anchor information to query position is spread across layers 19 through 27, where residual carry rises from 0.60 to 0.81 while attention to anchors stays in the 0.12 to 0.23 range. I label this pattern attention topology path. The classification is observational; causal verification via activation patching is the next experiment.

Zenodo (CERN European Organization for Nuclear Research)
Advanced Memory and Neural Computing
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.