INDEX

Explanations

the word "remember"

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 note

-0.86

 Note

-0.77

 remarque

-0.72

 remarqué

-0.71

note

-0.68

 noting

-0.67

ContentAlignment

-0.66

 noté

-0.66

Note

-0.65

 notare

-0.65

POSITIVE LOGITS

antwo

0.60

formik

0.60

enumi

0.59

ktop

0.57

 screenWidth

0.57

essy

0.57

 McCulloch

0.56

VIAF

0.55

ActionButton

0.54

edicated

0.54

Activations Density 0.015%