INDEX

Explanations

mathematical matrices and structures

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

if

-1.62

“

-1.58

蛧

-1.51

 each

-1.27

だらけ

-1.23

</h1>

-1.20

 Gebäude

-1.20

 Если

-1.16

in

-1.16

by

-1.16

POSITIVE LOGITS

妧

1.69

ㄓ

1.63

ዣ

1.59

많

1.55

嵓

1.53

こない

1.52

 ewigen

1.49

 limitaciones

1.48

</em>

1.47

鹪

1.43

Activations Density 0.005%