INDEX

Explanations

food and diet

desire for family or security

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

非常的

0.47

ittäin

0.45

่าว

0.44

first

0.41

ificados

0.41

bete

0.41

irli

0.40

ar

0.40

stup

0.40

Anda

0.40

POSITIVE LOGITS

 wanted

0.52

 vegetarian

0.49

 indiquant

0.46

 souhait

0.43

 gering

0.43

 calorie

0.43

 yogurt

0.42

 preven

0.42

 wyniku

0.42

 vegan

0.42

Activations Density 0.004%