INDEX

Explanations

path, breadcrumbs, hierarchy

The neuron activates on terms that denote hierarchical or navigational structures—e.g. “path,” “breadcrumb,” or “tree.”

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

↵↵

-1.76

);

-1.31

erialized

-1.28

ciąż

-1.25

aſ

-1.23

告诉

-1.20

estries

-1.20

}^{-

-1.16

letoe

-1.16

bz

-1.15

POSITIVE LOGITS

1.28

 なら

1.25

includes

1.21

 lesquels

1.20

溌

1.16

 joga

1.16

誰でも

1.16

┉

1.16

 your

1.15

 mentes

1.14

Activations Density 0.078%