Universal Activation Verbalizer: A Unified Framework for Cross-Model Activation Explanation

arXiv:2605.25903 · cs.CL, cs.LG · Submitted 2026-05-25 · Read on arXiv

cs.CL, cs.LG

Submitted: 2026-05-25

Updated: 2026-08-31

Code: https://github.com/allenai/pes2o

Terminology

• Alignment — 28× in this paper · explained in 1 episode
• Adaptation — 18× in this paper · explained in 1 episode
• Ablation — 15× in this paper · explained in 1 episode
• Embeddings — 6× in this paper · explained in 2 episodes
• Activation Oracle — 4× in this paper
• Adam — 4× in this paper · explained in 2 episodes
• Fine-Tuning — 4× in this paper · explained in 5 episodes
• Semantic Similarity — 4× in this paper · explained in 2 episodes
• Transformer — 4× in this paper · explained in 2 episodes
• Universal Activation Verbalizer — 4× in this paper
• Actor — 3× in this paper · explained in 3 episodes
• Decode — 2× in this paper · explained in 2 episodes
• Instruction Tuning — 2× in this paper · explained in 2 episodes
• Redundancy — 2× in this paper · explained in 2 episodes
• Transferability — 2× in this paper · explained in 4 episodes
• Unified Framework — 2× in this paper · explained in 5 episodes
• Ablation study — 1× in this paper · explained in 1 episode
• Activation Addition — 1× in this paper · explained in 1 episode
• Cross-Model Transfer — 1× in this paper · explained in 2 episodes
• Internal Revenue Service — 1× in this paper
• Interpretability — 1× in this paper · explained in 15 episodes
• Model-Agnostic — 1× in this paper · explained in 2 episodes
• Perplexity — 1× in this paper · explained in 3 episodes
• Representation Learning — 1× in this paper · explained in 2 episodes
• Sparse Autoencoders (SAEs) — 1× in this paper · explained in 3 episodes
• Steering — 1× in this paper · explained in 2 episodes
• Two-stage training — 1× in this paper · explained in 2 episodes

Sources

• Sparse Autoencoders Find Highly Interpretable Features in Language Models
• GPT-4 Technical Report
• Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers
• DeepSeek-V3 Technical Report
• Gemini: A Family of Highly Capable Multimodal Models

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering