INDEX

Explanations

CSS units and positioning

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

to

-2.30

for

-1.48

也很

-1.47

 vulgares

-1.47

 perdida

-1.46

Здра

-1.44

 fresa

-1.42

 silnika

-1.42

 than

-1.41

erobic

-1.41

POSITIVE LOGITS

ׇ

1.64

öra

1.48

hlten

1.40

 pestañas

1.40

 mesmas

1.37

 који

1.34

ској

1.30

fourths

1.28

 која

1.26

НЕ

1.24

Activations Density 0.063%