INDEX

Explanations

conjunctions and relational terms

after punctuation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 applicants

0.29

 alphanumeric

0.27

 conforman

0.26

CLUDING

0.26

 strateg

0.25

 exclude

0.25

 adherence

0.25

soc

0.25

 classifications

0.25

 également

0.25

POSITIVE LOGITS

 Когда

0.37

 Якщо

0.36

 Это

0.35

 Тогда

0.34

 यही

0.33

 ಅದನ್ನು

0.33

 Çünkü

0.33

 そして

0.33

 знаю

0.33

 ಏಕೆಂದರೆ

0.33

Activations Density 0.326%