INDEX

Explanations

uncommon words

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

 conqueror

-0.73

 commerciales

-0.68

 scolaires

-0.68

 conquerors

-0.68

 élevés

-0.68

 normaux

-0.67

 clínicas

-0.66

 microorganisms

-0.66

 élevées

-0.64

 suivantes

-0.64

POSITIVE LOGITS

ratulations

0.57

 live

0.56

lccn

0.53

par

0.53

 working

0.52

 balance

0.52

hom

0.52

 Balance

0.51

yet

0.50

car

0.50

Activations Density 0.739%