INDEX

Explanations

periods followed by words

end of sections

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

 superbly

0.46

 prachtige

0.44

ഷ്ട

0.42

 swiftly

0.42

clim

0.39

 buvo

0.39

ранже

0.39

充满了

0.38

惊喜

0.38

ttemberg

0.37

POSITIVE LOGITS

 eventuali

0.50

 GOOD

0.47

 אם

0.46

DAV

0.45

 आमतौर

0.45

 Fuck

0.43

FYI

0.41

 ಇದರಿಂದ

0.41

AND

0.40

 Congrats

0.40

Activations Density 0.185%