INDEX

Explanations

places following of

The neuron fires on the “of” in phrases naming a place—e.g. “town of X,” “city of Y,” “island of Z,” etc.—i.e. geographic “… of …” constructions.

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

Karakteristik

-1.80

Kelebihan

-1.59

ョン

-1.53

Ukuran

-1.48

MITTED

-1.47

Pendidikan

-1.45

Kenapa

-1.44

その他

-1.42

ITIONS

-1.41

 possibility

-1.41

POSITIVE LOGITS

to

2.17

 sebagai

1.69

 zelfs

1.53

ЧА

1.49

%);

1.48

 tienen

1.47

但

1.46

 でも

1.45

蛍

1.43

 sogar

1.42

Activations Density 0.049%