INDEX

Explanations

cookie cutter

copyright and identify individuals

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

걀

0.48

}\

0.46

壢

0.45

一篇

0.43

蒐

0.40

лім

0.40

 NONCLUSTERED

0.40

\}$

0.40

)\

0.39

化妆

0.39

POSITIVE LOGITS

 gener

0.53

 amigos

0.52

 hutan

0.51

 nokt

0.50

 respekt

0.50

 forest

0.48

 hysteria

0.48

 zakon

0.48

zum

0.47

 mereka

0.46

Activations Density 0.000%