INDEX

Explanations

computer code

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+#+#

-1.01

SharedCtor

-0.76

styleType

-0.75

RectangleBorder

-0.71

AndEndTag

-0.68

PhysRevD

-0.68

 gunpowder

-0.66

IntoConstraints

-0.65

twimg

-0.65

 '\\;'

-0.63

POSITIVE LOGITS

WriteBarrier

0.54

 changed

0.45

en

0.42

asts

0.42

 turned

0.41

 cursed

0.39

ased

0.39

 Changed

0.38

成

0.38

Activations Density 0.006%