INDEX

Explanations

being

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

 being

-1.01

being

-0.75

Being

-0.65

and

-0.60

 Being

-0.60

 doing

-0.58

 having

-0.56

 making

-0.56

↵↵

-0.55

 getting

-0.53

POSITIVE LOGITS

ValueStyle

1.24

 nahilalakip

1.08

脚注の使い方

1.06

 Signalez

1.03

+#+#

1.02

ंदीखरीदारी

1.00

OGND

0.97

 ligiloj

0.97

出版年

0.95

 <=",

0.94

Activations Density 0.056%