INDEX

Explanations

on/with

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

-0.47

ktar

-0.44

 popular

-0.44

twimg

-0.44

everything

-0.43

لكتر

-0.43

léans

-0.43

ционное

-0.43

ogni

-0.43

devise

-0.43

POSITIVE LOGITS

<bos>

1.46

AnimationsModule

0.69

TargetException

0.66

<?>>

0.63

postIndex

0.62

MethodManager

0.60

WaitGroup

0.59

ParallelGroup

0.58

Sprintf

0.57

PhysRevLett

0.56

Activations Density 0.012%