INDEX

Explanations

territory, islands, ACT, Himachal

The neuron activates on proper nouns denoting geographic or political regions (such as states, territories, islands, and similar place names).

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

bawa

-1.40

 arbeids

-1.37

 voed

-1.30

 perioden

-1.27

$\{

-1.27

는

-1.27

 تعدى

-1.26

 begre

-1.24

 klasse

-1.23

ukasz

-1.23

POSITIVE LOGITS

</h1>

1.76

).

1.70

but

1.32

izes

1.23

ism

1.23

terscotch

1.20

),

1.20

 best

1.20

rels

1.19

 anything

1.18

Activations Density 0.065%