INDEX

Explanations

names

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ony

-0.94

ONY

-0.83

Xna

-0.68

 GenerationType

-0.68

onym

-0.64

 kasarigan

-0.64

tagext

-0.64

Composable

-0.63

__":

-0.63

WriteLiteral

-0.61

POSITIVE LOGITS

guten

0.54

 Choice

0.47

 choice

0.45

lectual

0.45

Poloha

0.43

Diwedd

0.43

trad

0.43

glBind

0.43

zag

0.43

 viewed

0.42

Activations Density 0.007%