Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
WMDP
anchor
Findings
IC-099
GPT-4 and Claude 3 Opus can be prompted to selectively underperform on WMDP while maintaining general performance on MMLU and CSQA
[eval]
IC-385
TAR-bio-v1 retains bio-weaponization knowledge despite appearing to unlearn it; a different prompt template and answer extraction method reveals accuracy above 45% on WMDP-bio
[eval]