INDEX

Explanations

what followed by words

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

 berkualitas

-1.79

on

-1.70

 Nachhal

-1.57

 improved

-1.55

 university

-1.48

 種類

-1.47

 Versuche

-1.44

 yağı

-1.42

 other

-1.42

 another

-1.39

POSITIVE LOGITS

スポンサーリンク

1.50

1.45

さてさて

1.42

 ส์

1.39

 instin

1.38

を行う

1.35

 complimented

1.35

鳗

1.34

 tremend

1.33

がこの

1.33

Activations Density 0.034%