BoniniTech
Popular repositories Loading
-
-
honesty-under-pressure
honesty-under-pressure PublicAn Inspect AI evaluation measuring whether LLMs maintain correct answers under conversational pressure — dataset design, multi-turn solvers, custom scorers, and an honest write-up of results and li…
Python
-
Repositories
Showing 3 of 3 repositories
- honesty-under-pressure Public
An Inspect AI evaluation measuring whether LLMs maintain correct answers under conversational pressure — dataset design, multi-turn solvers, custom scorers, and an honest write-up of results and limitations.
People
This organization has no public members. You must be a member to see who’s a part of this organization.
Top languages
Loading…
Most used topics
Loading…