INDEX

Explanations

multi-language conjunctions

text from multiple languages

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

恚

0.30

圯

0.27

قتص

0.25

맏

0.25

ήμε

0.25

 Índice

0.25

 Entonces

0.24

toolStrip

0.24

بتدي

0.24

뭡

0.24

POSITIVE LOGITS

and

0.48

or

0.46

in

0.45

на

0.42

and

0.41

 এবং

0.38

ing

0.38

et

0.38

 ಹಾಗೂ

0.38

ও

0.37

Activations Density 0.065%