INDEX

Explanations

question structure (: or)

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

会う

-1.80

楽しめます

-1.49

corbic

-1.48

"$_

-1.48

 говорили

-1.45

 cérebro

-1.42

觊

-1.42

 dollari

-1.42

感じた

-1.40

 Их

-1.39

POSITIVE LOGITS

if

1.70

 then

1.66

 services

1.50

 τοῦ

1.38

now

1.34

 घर

1.31

괄

1.24

您

1.23

を使

1.23

you

1.23

Activations Density 0.168%