INDEX

Explanations

the presence of specific code-embedded structured data or script-related content

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

+#+#

-0.71

 arşivlendi

-0.68

?',

-0.68

?";

-0.60

 مواليد

-0.59

########.

-0.57

 gọn

-0.57

nonatomic

-0.56

odorant

-0.56

 సౌకర్య

-0.54

POSITIVE LOGITS

-{

2.47

.{

2.02

+{

1.36

}-{

1.02

:{

0.83

-${

0.77

/{

0.68

 noDo

0.65

{}{}

0.61

<{

0.59

Activations Density 0.001%