# Safe\_Mode = \[True, False\]

**URL:** <https://discuss.huggingface.co/t/safe-mode-true-false/67206>\
**Category:** Community Calls\
**Created:** [December 27, 2023, 9:18pm UTC](https://discuss.huggingface.co/t/safe-mode-true-false/67206 "2023-12-27T21:18:57Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![SocraGPT](https://avatars.discourse-cdn.com/v4/letter/s/47e85d/32.png) [@SocraGPT](https://discuss.huggingface.co/u/SocraGPT)\
**Post date:** [December 27, 2023, 9:18pm UTC](https://discuss.huggingface.co/t/safe-mode-true-false/67206/1 "2023-12-27T21:18:57Z")

</div>

I’m looking for alignment and safety pros to collaborate on a framework for evaluating open-source LLMs. I have a hunch the open-source community can deliver responsible AI faster than the proprietary shops. Is anyone else working on this?

I’m starting the process with an evaluation of [Mistral](https://docs.google.com/document/d/1lSxpIbJss4fkON6HHmZ0rzs4KtielouJ6Uj8cH06XYs/edit#heading=h.1qdyraff1cvp). I welcome your [feedback](https://github.com/socragpt/GRC/blob/main/Assessments/EARI.md)!