How Pruning Reshapes Features: Sparse Autoencoder Analysis of Weight-Pruned Language Models

How Pruning Reshapes Features: Sparse Autoencoder Analysis of Weight-Pruned Language Models explores This research analyzes how weight pruning in language models affects learned features, revealing that rare features are surprisingly resilient and suggesting pruning acts as implicit feature selection.. Commercial viability score: 4/10 in LLM Interpretability.