INDEX

Explanations

gender descriptors in biographical details

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

hörige

-1.38

mentaires

-1.37

ar

-1.33

で使

-1.33

ura

-1.32

 naturligt

-1.30

でしたが

-1.30

ciler

-1.29

 />';

-1.28

dzić

-1.27

POSITIVE LOGITS

 that

2.02

it

1.63

and

1.58

 zapal

1.56

 descub

1.42

 этот

1.35

 الأولى

1.34

borsa

1.34

 meisjes

1.33

往往

1.32

Activations Density 0.042%