INDEX

Explanations

for the next

list items and punctuation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

стера

0.55

 дві

0.54

有两个

0.51

 Пан

0.51

ຈັດສົ່ງ

0.50

ціона

0.50

 вулкан

0.50

 Президента

0.49

 нагрузки

0.48

ឫ

0.48

POSITIVE LOGITS

<h1>

0.58

as

0.57

One

0.51

in

0.50

 possession

0.50

Wid

0.50

but

0.48

 absence

0.48

and

0.47

isation

0.47

Activations Density 0.000%