Linear probes tell you where quantization will hurt

Epistemic status: I have only tested one encoder family (BERT-base and its relatives) and one decoder LLM (Qwen2.5-3B), one seed, token-level tasks, and post-training weight quantization. I trust the results because I am not inventing anything; I am just connecting quantization with a very general idea from mech interp: the model does the easy syntax work first, the semantic work in later layers, and the prompt-relevant task work in the latest layers. That's the main idea I lean on in this proje...