INDEX

Explanations

door/camera/specific nouns

comparison or superlative

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 businessmen

0.48

'});

0.47

ଜ

0.47

liono

0.46

appers

0.45

igere

0.45

協議

0.45

 考え

0.43

 netizens

0.42

越來越

0.42

POSITIVE LOGITS

ंटर

0.42

 изо

0.41

吵

0.41

 камера

0.41

들이

0.40

 दरवाजा

0.40

\}$.

0.39

 எரி

0.39

 rostrum

0.38

 двер

0.38

Activations Density 0.001%