INDEX

Explanations

names and specific instructions

illegal or foreign content

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

riend

0.52

 absolv

0.50

crackers

0.49

maz

0.48

ag

0.46

bal

0.46

 words

0.45

 experiment

0.45

mane

0.45

POSITIVE LOGITS

 ক্যাথোডে

0.53

 ಇದಕ್ಕೆ

0.51

 ಸ್ಪ

0.49

颶

0.49

 ಕಡಿಮೆ

0.48

 রুপ

0.48

 ಇದರ

0.48

 अवैध

0.47

 સો

0.46

 незакон

0.46

Activations Density 0.000%