INDEX

Explanations

assess states or qualities

degree of quality or intensity

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

uşt

0.43

did

0.40

>>=

0.40

yle

0.39

 даст

0.39

Ϳ

0.39

 inhabit

0.37

defines

0.37

anted

0.37

vej

0.36

POSITIVE LOGITS

 بوده

1.53

だった

1.47

 gewesen

1.41

이었

1.40

 आहे

1.36

 ہے

1.34

であった

1.33

 geweest

1.33

 olduğuna

1.32

である

1.30

Activations Density 0.053%