INDEX

Explanations

locations and specific nouns

core principles message conveyance

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

 Journal

0.41

 Operator

0.37

 journal

0.37

one

0.37

 Scotland

0.37

0.36

who

0.36

 Kitts

0.35

 Hernández

0.35

 Animals

0.35

POSITIVE LOGITS

mobili

0.57

 jogo

0.56

 thorium

0.52

 gleich

0.51

łoż

0.50

łoży

0.50

 transcoding

0.50

ంటీ

0.50

 erricht

0.49

 enkel

0.48

Activations Density 0.000%