INDEX

Explanations

united states or other capitalized proper nouns

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

 They

-1.66

how

-1.62

 Puoi

-1.50

}$.

-1.48

：「

-1.44

The

-1.43

You

-1.41

他们

-1.35

 они

-1.35

can

-1.34

POSITIVE LOGITS

ִּ

1.45

IENTS

1.45

 الأخيرة

1.41

 alegría

1.37

 کرنے

1.31

ทั้งหมด

1.29

ทั้ง

1.29

 lát

1.28

pppp

1.28

zubauen

1.27

Activations Density 0.083%