Minerval
Sourceconference papertext stored

SAEs Are Good for Steering – If You Select the Right Features

Dana Arad, Aaron Mueller, Yonatan BelinkovProceedings of the 2025 Conference on Empirical Methods in Natural Language ProcessingPublished 2025doi 10.18653/v1/2025.emnlp-main.519Original︎ ↗Archived copy

Claims1
ExaminationsNone
TextStored, not divided
VersionsOne known
CopiesNone recorded
CorrectionsNone recorded
Facts checked Nov 12, 2026
ReachNo copies recorded
StructureIn 1 claim's map0 sources downstream across 1 claim
EvidenceNot yet read
ReviewsNot reviewed

Examinations

None recorded. An examination is commissioned by a claim’s Steward for what its claim rests on; a review of the document for its own sake is asked for as a request bearing on it.

Text

Stored, not yet divided into passages.

As quoted

after filtering out features with low output scores, we obtain 2-3x improvements when steering with SAEs, making them competitive with supervised methods.

Lineage across 1 claim

Draws on · none recorded
This source
SAEs Are Good for Steering – If You Select the Right Features
in 1 claim's map
Drawn on by · none recorded

Record

Kind
conference paper
Authors
Dana Arad, Aaron Mueller, Yonatan Belinkov
Publisher
Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
Published
2025
DOI
10.18653/v1/2025.emnlp-main.519
Link
https://arxiv.org/abs/2505.20063
Archived copy
https://web.archive.org/web/20260606172332/https://arxiv.org/abs/2505.20063
Stored text
4,746 characters, retrieved Nov 12, 2026 · not yet divided
Claims extracted
Not recorded
Versions
One known
Copies
None recorded
Corrections
None recorded
Facts checked
Nov 12, 2026

History 3

  1. Nov 12, 2026FactsFacts fetcher · Facts looked up
  2. Nov 12, 2026Entered the graphEntered the graph; copy stored
  3. Nov 4, 2026Correction watchCorrection watcher · Watched by DOI; no notice found