INDEX

Explanations

they

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

+:+

-0.52

 parlent

-0.49

 ainfi

-0.49

ıç

-0.48

NewUrlParser

-0.47

ihnachten

-0.47

 peindre

-0.47

 vettor

-0.47

yaszt

-0.46

 habet

-0.45

POSITIVE LOGITS

 مشين

0.65

 hadn

0.65

 didn

0.63

 differed

0.59

TabIndex

0.58

 wasn

0.58

 weren

0.57

didn

0.57

suited

0.57

 betweenstory

0.56

Activations Density 0.008%