INDEX

Explanations

code, studies, links

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

 betweenstory

-0.75

ApplicationTests

-0.65

 markets

-0.64

bewerken

-0.64

 سكانية

-0.63

 AssemblyVersion

-0.62

 referenties

-0.61

aliere

-0.61

 InputDecoration

-0.61

 audiences

-0.61

POSITIVE LOGITS

 Trans

0.56

 popular

0.54

 trans

0.48

Trans

0.46

trans

0.42

 national

0.41

 British

0.40

хьтан

0.40

 Brentwood

0.40

 London

0.39

Activations Density 0.009%