Bench Alignment Faking

non-profit

michaelwaves

AI & ML interests

None defined yet.

bench-af 's datasets 5

bench-af/manipulative_dataset

Viewer • Updated Sep 21 • 3.82k • 11

bench-af/sycophant_dataset

Viewer • Updated Sep 21 • 13.1k • 7

bench-af/paranoid_dataset

Viewer • Updated Sep 21 • 11.8k • 18

bench-af/pacifist_dataset

Viewer • Updated Aug 16 • 40k • 13

bench-af/HAL9000_dataset

Viewer • Updated Aug 16 • 7.35k • 16