AINeutralarXiv – CS AI · 10h ago6/10
🧠
What Cohort INRs Encode and Where to Freeze Them
Researchers demonstrate that early layers of cohort-trained Implicit Neural Representations (INRs) encode transferable features for signal fitting, identifying optimal freezing points through weight stable rank analysis. Using sparse autoencoders for mechanistic interpretability, they reveal that SIREN and Fourier-feature MLPs learn fundamentally different dictionary representations despite comparable performance, with implications for designing more generalizable neural architectures.