Documentation#
What it is:
llmSHAP is a Python library for attributing the contributions of different input parts to the output of large language models (LLMs), using Shapley values.
Who it’s for: Researchers and developers working with LLMs who need insight into why a model produced a particular response.
Why it exists:
LLMs are powerful but often opaque. llmSHAP helps make their outputs interpretable by quantifying the impact of each input element on the result.