INDEX

Explanations

a + noun/adjective-noun

qualities and inclinations

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

 prilikom

0.33

 Dempsey

0.31

 получаем

0.30

 حصول

0.29

เป็นการ

0.29

 Gilmore

0.28

 utilizing

0.27

 باستخدام

0.26

 olisi

0.26

当時

0.26

POSITIVE LOGITS

 penchant

0.48

 pretensions

0.45

 tendencies

0.41

 tendency

0.40

 knack

0.39

 inclinations

0.38

 reputations

0.38

 ambitions

0.37

 affinity

0.36

 probl

0.35

Activations Density 0.518%