INDEX

Explanations

the color black

The neuron selectively activates on mentions of the color “black” (i.e. adjectival uses of “black” describing clothing, objects, etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

-1.46

 персонал

-1.41

ヌード

-1.29

槩

-1.29

 штат

-1.29

approved

-1.28

カップル

-1.27

形で

-1.27

佧

-1.27

 oeste

-1.24

POSITIVE LOGITS

黑

1.59

 black

1.45

 zwart

1.36

 fekete

1.35

 basé

1.27

black

1.27

🆁

1.24

 Black

1.21

 enseñanza

1.20

 чор

1.20

Activations Density 0.065%