INDEX

Explanations

to

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

 sanitaires

-0.71

 giudi

-0.67

 médicaux

-0.65

 chrétien

-0.63

 nødvendig

-0.63

 aveug

-0.62

 scolas

-0.61

 scolaires

-0.60

 beginning

-0.60

 nationaux

-0.59

POSITIVE LOGITS

abestanden

0.88

aarrggbb

0.77

 autorytatywna

0.77

setupUi

0.76

Datuak

0.75

mappedBy

0.73

AsUp

0.72

 emerge

0.70

XmlAccessType

0.69

 argue

0.69

Activations Density 0.020%