INDEX

Explanations

common words followed by articles

grams/phrases

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

পো

0.54

 jaunâtre

0.52

łki

0.50

wą

0.49

 tomto

0.47

 włos

0.47

 fonksiyon

0.47

ion

0.47

odne

0.46

جاد

0.45

POSITIVE LOGITS

務

0.46

率先

0.45

но

0.44

」(

0.43

 пациента

0.41

繩

0.41

ارض

0.40

스가

0.40

''}

0.40

羯

0.40

Activations Density 0.001%