Sparse Feature Analysis of Deep Layer Expansion: A Mechanistic Interpretation via SAE

Zhao (2026) demonstrated that expert-level prompts induce "Deep Layer Expansion"—a 60-100% increase in Effective Intrinsic Dimension (EID) at deep layers. However, EID is a global metric that does not reveal which semantic features are activated. In this paper, we apply Sparse Autoencoder (SAE) analysis to decompose the activation differences between prompt styles. Using Goodfire's Llama-3.3-70B SAE (Layer 50, 65,536 features), we find that: (1) "Explain to a novice" activates 17% more features than "explain to an expert" (132.4 vs 113.1 on average); (2) 369 features are exclusively activated by novice prompts vs 208 for expert prompts; (3) 10 features show perfect separation (100% activation in one condition, 0% in the other). These findings provide mechanistic evidence that prompt-induced EID differences reflect distinct sparse feature activation patterns, not merely statistical noise.