INDEX

Explanations

justice system and corrections

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

flot

-1.38

companyname

-1.37

 prominently

-1.35

indice

-1.34

是可以

-1.31

实力

-1.26

 pisang

-1.24

Slut

-1.24

DIRE

-1.23

Detal

-1.23

POSITIVE LOGITS

$}}

1.55

logous

1.55

enterOuterAlt

1.44

 のか

1.44

seled

1.41

\}}

1.39

 malle

1.38

 rework

1.37

huahua

1.34

twimg

1.33

Activations Density 0.024%