AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Semantic Regexes: Auto-Interpreting LLM Features with a Structured Language

Angie Boggust, Donghao Ren, Yannick Assogba, Dominik Moritz, Arvind Satyanarayan, Fred Hohman

ICLR 2026 · Large language models › Language, safety and interpretability · approach: Linguistic encoding in LLM representations · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace GPT-4o-mini with GPT-4o as explainer/evaluator (ablation)semantic-regex · GPT-2-RES-25k (520 features)purity: 0.68 ± 0.32 -> 0.74 ± 0.30No clear effect
replace GPT-4o-mini with GPT-4o as explainer/evaluator (ablation)semantic-regex · GPT-2-RES-25k (520 features)clarity: 0.69 ± 0.37 -> 0.68 ± 0.36No clear effect
replace GPT-4o-mini with GPT-4o as explainer/evaluator (ablation)token-act-pair · GPT-2-RES-25k (520 features)clarity: 0.45 ± 0.37 -> 0.59 ± 0.37Improved
replace GPT-4o-mini with GPT-4o as explainer/evaluator (ablation)token-act-pair · GPT-2-RES-25k (520 features)purity: 0.73 ± 0.26 -> 0.84 ± 0.21Improved
use semantic regex instead of token-act-pair descriptionstoken-act-pair · GPT-2-RES-25kclarity: 0.45 ± 0.36 -> 0.68 ± 0.36Improved
use semantic regex instead of max-acts descriptionsmax-acts · GPT-2-RES-25kfaithfulness: 0.52 ± 0.46 -> 0.37 ± 0.45Hurt
use semantic regex instead of max-acts descriptionsmax-acts · GPT-2-RES-25kpurity: 0.78 ± 0.26 -> 0.66 ± 0.32Hurt
use semantic regex instead of max-acts descriptionsmax-acts · GPT-2-RES-25kclarity: 0.70 ± 0.35 -> 0.68 ± 0.36No clear effect
use semantic regex instead of max-acts descriptionsmax-acts · Gemma-2-2B-RES-65k (100 features/layer)clarity: 0.45 ± 0.40 -> 0.47 ± 0.42No clear effect
give participants semantic regex instead of natural language descriptionsmax-acts · user study (24 participants, 12 features)decision boundary understanding: as well or higher on 9 of 12 featuresMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy