INDEX

Explanations

gibberish text

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

 Wicidata

-0.57

NameInMap

-0.52

 τοῦ

-0.51

ch

-0.50

 chi̍t

-0.49

 Schia

-0.47

 dobro

-0.47

 defesa

-0.47

Sprintf

-0.45

 subscriptions

-0.45

POSITIVE LOGITS

oredCriteria

0.57

Manbalar

0.55

 برانيه

0.54

ineno

0.50

ГЛА

0.50

jgl

0.49

RTGC

0.48

 eventdata

0.47

legi

0.47

InjectAttribute

0.46

Activations Density 0.001%