INDEX

Explanations

sequences of numbers and delimiters

3

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

Negative Logits

 defaultstate

-0.52

 tủ

-0.50

 Mérimée

-0.45

urah

-0.45

,:]

-0.45

GraphicsUnit

-0.45

Curios

-0.44

 noté

-0.43

дельник

-0.43

\{\\

-0.43

POSITIVE LOGITS

__":

0.59

 GENERATED

0.58

 Jefus

0.56

 дописавши

0.56

 propOrder

0.54

 GenerationType

0.53

__':

0.51

 تضيفلها

0.51

issional

0.51

ſelf

0.51

Activations Density 2.008%