INDEX

Explanations

power/rights

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

 étoit

-0.60

 auroit

-0.56

 troppo

-0.56

 sèche

-0.55

 pubblica

-0.54

 sauvages

-0.54

 marié

-0.52

 superiori

-0.51

 avoient

-0.51

 pregunto

-0.51

POSITIVE LOGITS

 area

0.78

 thing

0.74

 team

0.67

 mergeFrom

0.66

 section

0.65

 version

0.63

 original

0.63

 portion

0.63

 الدولى

0.62

 part

0.62

Activations Density 0.083%