INDEX

Explanations

Mixed topics/random 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

<bos>

-0.74

 estimés

-0.70

 bezeichneter

-0.69

 snippetHide

-0.67

archiviato

-0.67

ंदीखरीदारी

-0.65

 insuffisamment

-0.63

الحياه

-0.63

Бахар

-0.62

Hauptartikel

-0.62

POSITIVE LOGITS

APORE

0.48

깥

0.46

flage

0.45

Filmographie

0.42

cionados

0.42

фель

0.42

IGENCE

0.42

FLP

0.42

Ivoire

0.41

konferenz

0.40

Activations Density 0.013%