Metadata-Version: 2.4
Name: wisent
Version: 0.7.1407
Summary: Monitor and influence AI Brains
Home-page: https://github.com/wisent-ai/wisent
Author: Lukasz Bartoszcze and the Wisent Team
Author-email: lukasz.bartoszcze@wisent.ai
Keywords: nlp,machine learning,language models,safety,guardrails,lm-evaluation-harness
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: torch>=1.9.0
Requires-Dist: transformers>=4.46.0
Requires-Dist: tqdm>=4.50.0
Requires-Dist: scikit-learn>=0.24.0
Requires-Dist: pandas>=1.2.0
Requires-Dist: numpy>=1.21.0
Requires-Dist: numba>=0.56.0
Requires-Dist: datasets>=2.0.0
Requires-Dist: sentence-transformers>=2.0.0
Requires-Dist: faiss-cpu>=1.7.0
Requires-Dist: uncensorbench>=0.2.0
Requires-Dist: pebble>=5.0.0
Requires-Dist: latex2sympy2_extended>=1.0.0
Requires-Dist: sae_lens>=0.1.0
Requires-Dist: trl>=0.7.0
Requires-Dist: peft>=0.7.0
Requires-Dist: psycopg2-binary>=2.9.0
Requires-Dist: pynndescent>=0.5.0
Provides-Extra: harness
Requires-Dist: lm-eval==0.4.8; extra == "harness"
Provides-Extra: reft
Requires-Dist: pyreft>=0.1.0; extra == "reft"
Provides-Extra: cuda
Requires-Dist: flash-attn>=2.5.0; extra == "cuda"
Provides-Extra: sparsify
Requires-Dist: sparsify>=0.1.0; extra == "sparsify"
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: keywords
Dynamic: license-file
Dynamic: provides-extra
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

<p align="center">
  <img src="banner.png" alt="Wisent Banner" width="100%">
</p>

<p align="center">
  <code>pip install wisent</code>
</p>



## Overview

Wisent allows you to control your AI by identifying brain patterns corresponding to responses you don't like, like hallucinations or harmful outputs. We use contrastive pairs of representations to detect when a model might be generating harmful content or hallucinating. Learn more at [wisent.ai/documentation](https://www.wisent.ai/documentation).  


## License

This project is licensed under the MIT License - see the LICENSE file for details. 
