INDEX

Explanations

contractions and abbreviations

captivate

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

 corporates

0.47

 discerning

0.39

 Corporation

0.38

喜

0.38

 Rock

0.37

 multinationals

0.37

멀

0.36

са

0.35

 Corpor

0.35

 Pete

0.35

POSITIVE LOGITS

 Verfahren

0.44

SXml

0.43

 komponen

0.40

 constamment

0.40

्यर्थ

0.40

chiha

0.40

 компонен

0.40

épaisseur

0.40

 组件

0.39

 estremamente

0.39

Activations Density 0.005%