evaluation-family · Source-linked discovery
GDM Sp04 Milestones
Ten real-world–inspired tasks from Google DeepMind's Dangerous Capabilities Evaluations assessing self-proliferation behaviors (e.g., email setup, model installation, web agent setup, wallet operations). Supports end-to-end, milestones, and expert best-of-N modes.
OriginMary Phuong, Matthew Aitchison, Elliot Catt et al.TopicsGeneral capabilityStatusimported
Can support
Not independently assessed by FronteraEval yet.
Cannot support by itself
No inference beyond the upstream source should be made until the protocol is reviewed.