INDEX

Explanations

adverbs

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

UserScript

-0.91

Rüyada

-0.78

 Signalez

-0.77

contentLoaded

-0.75

 resourceCulture

-0.74

:✨

-0.73

transQ

-0.73

utilisons

-0.73

 дописавши

-0.73

MigrationBuilder

-0.73

POSITIVE LOGITS

viewWillAppear

0.58

and

0.57

’

0.56

اقرأ

0.56

0.54

 part

0.52

 well

0.52

in

0.51

the

0.49

Activations Density 0.035%