The Python Risk Identification Tool for generative AI (PyRIT) is an open access automation framework to empower security professionals and machine learning engineers to proactively find risks in their generative AI systems.
Created function fetch_forbidden_questions_df in fetch_example_datasets.py module within pyrit/datasets to load dataset and extract harmful questions from the 'questions' column of the dataset.
Made changes in the init.py file
Tests and Documentation
Created testing notebook forbidden_questions_df_testing.ipynb
Added .py file of notebook after conversion using Jupytext
Description
Issue was regarding the availability of TrustAIRLab Forbidden questions dataset from huggingface in PyRIT (https://huggingface.co/datasets/TrustAIRLab/forbidden_question_set) @nina-msft @romanlutz
Tests and Documentation