INDEX

Explanations

verbs in multiple languages

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<h3>

-1.52

and

-1.50

in

-1.41

-1.38

http

-1.38

他还

-1.36

<h4>

-1.36

or

-1.35

https

-1.30

 источники

-1.28

POSITIVE LOGITS

 bytu

1.69

 znamená

1.49

 Economía

1.47

 obrazu

1.46

 podoba

1.42

ஂ

1.42

کد

1.41

 okna

1.39

 kancel

1.34

 blessés

1.34

Activations Density 0.007%