INDEX

Explanations

susceptibility

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

Caution

-0.93

caution

-0.78

 Caution

-0.76

 warning

-0.70

 cautioned

-0.68

CAUTION

-0.67

 cautions

-0.66

 warned

-0.62

 CAUTION

-0.62

Precautions

-0.59

POSITIVE LOGITS

 myſelf

0.76

givers

0.73

takers

0.72

orship

0.71

 Monfieur

0.70

 becauſe

0.69

 estekak

0.69

veness

0.68

GEBURTSDATUM

0.68

Subviews

0.68

Activations Density 0.091%