INDEX

Explanations

name attribute or property

The neuron fires on code tokens that name parameters or properties—e.g. attribute/key identifiers like “name,” “title,” or “type.”

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

 Füßen

-1.34

 أنّ

-1.27

ụp

-1.20

 democracies

-1.20

Nevertheless

-1.20

 sicheren

-1.16

 Ihrem

-1.16

 شناسی

-1.14

 vollen

-1.13

joni

-1.12

POSITIVE LOGITS

1.52

࿚

1.46

︲

1.38

1.37

佧

1.34

໊

1.33

Ẵ

1.33

垕

1.31

mability

1.30

تیجه

1.30

Activations Density 0.012%