TIL Inspect
An open-source framework for large language model evaluations
Looks like a nice piece of open source kit from the UK government’s AI Security Institute
A big part of the day job is LLM evaluation so this is definitely of interest.
© 2008-2024 C. Ross Jam. Built using Pelican. Theme based upon Giulio Fidente’s original svbhack, and slightly modified by crossjam.