Jacobian Scopes: A Unified Geometric Framework for Token-Level LLM Attribution
arXiv:2601.16407v5 Announce Type: replace-cross Abstract: Large language models (LLMs) make next-token predictions based on clues present in their context, such as semantic descriptions and in-context examples. Yet, elucidating which prior tokens most strongly influence a given prediction remains challenging due to the proliferation of layers and attention heads in modern architectures. We propose Jacobian Scopes, a suite of gradient-based, token level causal attribution methods for interpretin…