INDEX

Explanations

German, English, or Spanish words

The neuron is essentially a “rarity” detector—it spikes on low‐frequency, content‐bearing tokens (e.g. foreign‐language words or infrequent modals) rather than on common function words.

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

to

-1.98

-1.50

épaisseur

-1.49

has

-1.48

 möjlighet

-1.46

use

-1.41

is

-1.38

เช่น

-1.38

teni

-1.37

 ベクトル

-1.37

POSITIVE LOGITS

you

2.66

of

2.33

it

2.30

we

2.27

 they

1.93

 this

1.88

 podemos

1.70

 ledig

1.68

klart

1.66

llten

1.63

Activations Density 0.012%