INDEX

Explanations

I believe, or correctly

explaining capabilities/attributes with uncertainty

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

丝毫

0.49

充满

0.43

 iniziamo

0.43

 시작하겠습니다

0.42

充滿

0.40

 идеально

0.40

精心

0.40

 રહેશે

0.39

 نئے

0.38

five

0.37

POSITIVE LOGITS

(?)

1.00

(?)

0.98

好像

0.97

 somewhere

0.95

確か

0.93

 নাকি

0.93

 chyba

0.91

 reportedly

0.89

 alguma

0.84

 nějak

0.82

Activations Density 0.078%