INDEX

Explanations

universal control

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Negative Logits

 kidding

1.16

 convince

1.07

 convinces

1.01

 convincing

0.99

 deceive

0.96

 self

0.90

 fooling

0.89

 convain

0.88

 convinced

0.86

 Self

0.85

POSITIVE LOGITS

 universal

0.44

 Universal

0.38

Universal

0.38

 универса

0.38

ropa

0.37

 codes

0.37

 evaporated

0.37

普遍

0.36

MOD

0.35

codes

0.35

Activations Density 0.007%

No Known Activations