INDEX

Explanations

forms of 'to be' and 'will'

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

and

-2.16

 also

-1.62

of

-1.56

or

-1.52

 will

-1.34

in

-1.32

 this

-1.29

 when

-1.26

altro

-1.20

...»

-1.19

POSITIVE LOGITS

 mindestens

1.49

楽しめます

1.46

ATIVA

1.42

 ถ้า

1.38

 Ausführungen

1.31

 новых

1.30

 indiqué

1.28

鈺

1.27

݂

1.26

 Wirkungen

1.25

Activations Density 0.121%