Do Localization Methods Actually Localize Memorized Data in LLMs? A Tale of Two Benchmarks

arXiv:2311.09060v2 Announce Type: replace
Abstract: The concept of localization in LLMs is often mentioned in prior work; however, methods for localization have never been systematically and directly evaluated. We propose two complementary benchmarks that evaluate the ability of localization methods to pinpoint LLM components responsible for memorized data. In our INJ benchmark, we actively inject a piece of new information into a small subset of LLM weights, enabling us to directly evaluate whether localization methods can identify these "ground truth" …

abstract arxiv benchmarks components concept cs.cl data however llm llms localization prior responsible type work