INDEX

Explanations

score

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 score

-0.67

 Score

-0.66

 SCORE

-0.60

 scores

-0.57

routeProvider

-0.56

WriteAttribute

-0.55

 Waterways

-0.52

MENAFN

-0.50

Matty

-0.49

SCORE

-0.49

POSITIVE LOGITS

ThemeOverlay

0.63

 препратки

0.61

ustimmung

0.60

oredCriteria

0.59

 riduzione

0.57

bootstrapcdn

0.56

MLLoader

0.54

marshaller

0.52

card

0.50

usammen

0.50

Activations Density 0.021%

No Known Activations