INDEX

Explanations

the

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

Autoritní

-0.73

 aught

-0.73

ItemBackground

-0.65

 secco

-0.59

 inſ

-0.58

OrBuilder

-0.58

CORBA

-0.58

 informée

-0.57

 uſed

-0.57

 optically

-0.57

POSITIVE LOGITS

CloseOperation

0.71

 ones

0.58

 pinulongan

0.52

StoryboardSegue

0.51

ESIS

0.48

WebKit

0.48

 defStyleAttr

0.48

GEBURTS

0.47

 sekali

0.47

new

0.47

Activations Density 0.017%