INDEX

Explanations

back

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 back

-2.42

 Back

-1.98

back

-1.95

Back

-1.95

 BACK

-1.68

BACK

-1.60

バック

-1.27

 backs

-1.22

 tilbake

-1.21

 tillbaka

-1.20

POSITIVE LOGITS

EndContext

0.68

 ashore

0.62

 createContext

0.60

+#+#

0.56

to

0.55

melons

0.53

 تضيفلها

0.53

robes

0.52

 Meksiku

0.51

 spas

0.50

Activations Density 0.050%