INDEX

Explanations

bi-equivalency

recognizing different variations

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

鼠标

0.47

掩

0.47

 ойно

0.47

 ойнотуу

0.46

缠

0.44

由

0.43

вь

0.43

\%(

0.42

 entanto

0.42

 sendo

0.42

POSITIVE LOGITS

 Cleveland

0.52

 Carnegie

0.51

 सूजन

0.49

agin

0.47

APA

0.47

 Pittsburgh

0.46

 United

0.46

 federal

0.46

cogn

0.45

federal

0.45

Activations Density 0.001%