INDEX

Explanations

list titles or enumerated items

The neuron primarily activates on period characters (dots).

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

甥

-1.34

 élections

-1.34

琉

-1.34

 apartament

-1.25

 hôtels

-1.24

 originalmente

-1.23

ubahan

-1.21

 musik

-1.16

EdgeInsets

-1.16

"....

-1.16

POSITIVE LOGITS

1.44

1.29

のではない

1.23

ことがある

1.20

1.18

行って

1.16

をします

1.16

 verkligen

1.15

Activations Density 0.072%