INDEX

Explanations

posted on category names

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

 where

-2.19

 Despite

-1.87

 using

-1.86

Many

-1.79

 despite

-1.74

to

-1.70

 While

-1.69

 Often

-1.59

Despite

-1.59

Following

-1.56

POSITIVE LOGITS

represents

1.81

begins

1.73

continues

1.62

夘

1.59

probably

1.59

clearly

1.59

 петербург

1.59

indicates

1.55

becomes

1.54

mainly

1.53

Activations Density 0.025%