INDEX

Explanations

conform

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

 Numerade

-0.57

 Dise

-0.55

ليه

-0.55

 Theſe

-0.53

asable

-0.53

Dise

-0.52

 للاسماء

-0.52

 Assim

-0.52

دار

-0.52

POSITIVE LOGITS

isContained

0.59

 الرياضيه

0.55

AddHtmlAttribute

0.54

nloa

0.54

verwijspagina

0.53

 تضيفلها

0.53

sidemargin

0.52

pecker

0.48

rodalom

0.48

transQ

0.48

Activations Density 0.002%