INDEX

Explanations

United States District Court

The neuron selectively activates on tokens that appear in U.S. federal court captions—especially “District” and “Court” (and their abbreviations) in case headings and citations.

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

 this

-1.52

 their

-1.37

by

-1.34

and

-1.32

 each

-1.24

any

-1.16

 that

-1.13

 both

-1.10

 attore

-1.09

 they

-1.08

POSITIVE LOGITS

мся

1.22

Мария

1.13

╭

1.08

🟢

1.08

 организма

1.02

⅙

1.00

 禁止

0.99

⅘

0.96

 模板

0.95

рги

0.95

Activations Density 0.062%