INDEX

Explanations

a mix of short words such as numbers, single letters, abbreviations, and parts of longer words and brand names, but there is no common theme between these

seemingly random tokens in running text

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

MLLoader

-0.90

 itſelf

-0.83

 myſelf

-0.81

 Monfieur

-0.76

featureID

-0.76

 houſe

-0.73

GenerationType

-0.73

AddTagHelper

-0.71

 purpoſe

-0.70

 againſt

-0.69

POSITIVE LOGITS

 former

0.52

 another

0.47

 popular

0.44

Subview

0.44

ays

0.44

JAH

0.42

ney

0.41

 mede

0.41

NEY

0.41

former

0.40

Activations Density 2.789%