INDEX

Explanations

booked all, collected all, filled out

ethnicity and race

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

 prvo

0.43

 flexDirection

0.40

 Associated

0.38

 productName

0.38

ทรง

0.37

釵

0.37

 concentration

0.37

 tabular

0.37

xml

0.36

鶚

0.36

POSITIVE LOGITS

 Koreans

0.47

 natomiast

0.47

ják

0.46

 வரைய

0.45

 malah

0.45

adorable

0.45

నుంది

0.44

 shrewd

0.44

善良

0.44

 वाय

0.44

Activations Density 0.001%