o3/o4-mini
OpenAI
partial detail 8 method fields published
Evaluation
In-context Scheming Reasoning (Apollo, o3/o4-mini)
In-context scheming reasoning (Covert Subversion, Deferred Subversion, sandbagging)
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.