INDEX

Explanations

women

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 women

-1.58

 Women

-1.13

women

-1.13

Women

-1.09

 WOMEN

-1.07

 vrouwen

-1.07

 woman

-1.04

 womens

-0.98

 kvinnor

-0.96

 ladies

-0.96

POSITIVE LOGITS

’

0.81

0.79

folk

0.73

who

0.66

0.62

and

0.57

0.54

 employed

0.51

´

0.50

in

0.47

Activations Density 0.022%