INDEX

Explanations

words that communicate positive or negative significance

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

 AssemblyCulture

-1.09

RectangleBorder

-0.92

 للاسماء

-0.87

 समीक्षक

-0.84

ագրություններ

-0.78

 utafitiHapana

-0.78

 للمعارف

-0.72

ⓧ

-0.71

Билгалдахарш

-0.70

Примечания

-0.70

POSITIVE LOGITS

SupportActionBar

0.57

JspWriter

0.54

0.50

 teljes

0.47

"){

0.46

SOUNDBITE

0.44

 forbindelse

0.44

no

0.43

++)

0.43

❛

0.43

Activations Density 0.067%