Continuous Interpretive Steering and the GraSD dataset reveal that LLMs encode graded sensitivity to scalar diversity in their internal representations, recoverable via controlled activation interventions.
Grains: Gradient-based attribution for inference-time steering of llms and vlms.CoRR, abs/2507.18043, 2025a
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
dataset 1
citation-polarity summary
years
2026 3verdicts
UNVERDICTED 3roles
dataset 1polarities
use dataset 1representative citing papers
The Master Key Hypothesis states that capabilities are low-dimensional directions transferable across models through linear subspace alignment, with UNLOCK demonstrating gains such as 12.1% accuracy improvement on MATH when transferring CoT from 14B to 7B models.
The survey organizes mechanistic interpretability techniques into a Locate-Steer-Improve framework to enable actionable improvements in LLM alignment, capability, and efficiency.
citing papers explorer
-
The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment
The Master Key Hypothesis states that capabilities are low-dimensional directions transferable across models through linear subspace alignment, with UNLOCK demonstrating gains such as 12.1% accuracy improvement on MATH when transferring CoT from 14B to 7B models.