INDEX

Explanations

Organizations and abbreviations

The neuron fires on tokens that are part of organization or institution names—especially capitalized acronyms and official body titles.

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

隴

-1.40

 "'";

-1.37

ÈRE

-1.36

-1.34

rostis

-1.34

👙

-1.34

 "'");

-1.32

 リスト

-1.29

 "','

-1.29

\&

-1.28

POSITIVE LOGITS

...."

1.52

 "...

1.48

<tr>

1.34

("

1.32

..."

1.27

''

1.24

 such

1.23

 other

1.23

etc

1.23

↵↵

1.19

Activations Density 0.044%