leap-laboratories/PIZZA
An attribution library for LLMs
This project helps anyone working with Large Language Models (LLMs) understand exactly which words or phrases in their prompt are most influential in shaping the model's response. You provide your prompt and an LLM's generated output, and it shows you a detailed breakdown of how each input token contributed to the generated response. This is ideal for AI product managers, researchers, or anyone debugging LLM behavior.
No commits in the last 6 months.
Use this if you need to understand the 'why' behind an LLM's output by dissecting the impact of individual prompt elements.
Not ideal if you are looking for a tool to train LLMs or optimize their performance without needing to interpret their internal workings.
Stars
46
Forks
6
Language
Python
License
—
Category
Last pushed
Sep 17, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/leap-laboratories/PIZZA"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
MadryLab/context-cite
Attribute (or cite) statements generated by LLMs back to in-context information.
microsoft/augmented-interpretable-models
Interpretable and efficient predictors using pre-trained language models. Scikit-learn compatible.
Trustworthy-ML-Lab/CB-LLMs
[ICLR 25] A novel framework for building intrinsically interpretable LLMs with...
poloclub/LLM-Attributor
LLM Attributor: Attribute LLM's Generated Text to Training Data
THUDM/LongCite
LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA