Concept Subspaces Compute Beyond the Logit Lens: A Weights-Only Test for Locating Representations Upstream of Readout

arXiv:2609.39263 · cs.CL, cs.LG · Submitted 2026-09-30 · Read on arXiv

cs.CL, cs.LG

Submitted: 2026-09-30

Updated: 2026-09-30

Terminology

• Ablation — 53× in this paper · explained in 1 episode
• Accuracy — 40× in this paper · explained in 1 episode
• Alignment — 34× in this paper · explained in 1 episode
• Benchmark — 20× in this paper · explained in 2 episodes
• Trace — 19× in this paper · explained in 2 episodes
• AUC — 16× in this paper · explained in 2 episodes
• Steering — 16× in this paper · explained in 2 episodes
• AGN — 15× in this paper · explained in 1 episode
• Transformer — 11× in this paper · explained in 2 episodes
• Generalization — 10× in this paper · explained in 13 episodes
• Activation Patching — 7× in this paper · explained in 2 episodes
• Interpretability — 6× in this paper · explained in 15 episodes
• Generative Model — 5× in this paper · explained in 2 episodes
• Generative Models — 5× in this paper · explained in 2 episodes
• Abstraction — 4× in this paper · explained in 1 episode
• Instruction Tuning — 4× in this paper · explained in 2 episodes
• Structural Control — 4× in this paper · explained in 2 episodes
• Adam — 3× in this paper · explained in 2 episodes
• Provenance — 3× in this paper · explained in 3 episodes
• Calibration — 2× in this paper · explained in 7 episodes
• Decode — 2× in this paper · explained in 2 episodes
• Format-Agnostic Reasoning Subspace — 2× in this paper
• Adversarial Attacks — 1× in this paper · explained in 1 episode
• Ambient Dimension — 1× in this paper · explained in 1 episode
• Causal Reasoning — 1× in this paper · explained in 2 episodes
• Computational Biology — 1× in this paper · explained in 2 episodes
• Core — 1× in this paper · explained in 2 episodes
• Cosine Similarity — 1× in this paper · explained in 3 episodes
• Cross-lingual Transfer — 1× in this paper · explained in 2 episodes
• Data Provenance — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Frozen model — 1× in this paper · explained in 2 episodes
• Intrinsic Dimension — 1× in this paper · explained in 2 episodes
• Linear probe — 1× in this paper · explained in 2 episodes
• Perplexity — 1× in this paper · explained in 3 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Sparsity — 1× in this paper · explained in 4 episodes
• Tokenization — 1× in this paper · explained in 2 episodes
• Verification — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• Training Verifiers to Solve Math Word Problems
• Universal and Transferable Adversarial Attacks on Aligned Language Models
• Prototype-Based Dynamic Steering for Large Language Models
• Steering Language Models With Activation Engineering
• Eliciting Latent Predictions from Transformers with the Tuned Lens
• Steerable but Not Decodable: Function Vectors Operate Beyond the Logit Lens
• Predicting Where Steering Vectors Succeed
• Hidden Error Awareness in Chain-of-Thought Reasoning: The Signal Is Diagnostic, Not Causal

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