INDEX

Explanations

And

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

afone

-0.60

forChild

-0.59

="@+

-0.59

 TextAppearance

-0.56

 NUKAT

-0.56

phones

-0.55

elemField

-0.55

InputTagHelper

-0.55

GEBURTS

-0.55

 "..\..\

-0.54

POSITIVE LOGITS

 possibility

0.56

 ambition

0.55

 suspicion

0.53

fjspx

0.52

以上就是

0.52

 controversy

0.52

 authenticity

0.52

 option

0.51

 success

0.50

 uncertainty

0.50

Activations Density 0.004%