INDEX

Explanations

section headings or document structure

The neuron fires on structural enumeration markers—i.e. numbered section or list labels like “1.”, “2.”, “一、”, “二、” and similar heading/list delimiters.

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

<bos>

-1.69

 нього

-1.57

 '-':

-1.52

-1.46

 }));

-1.42

 ")";

-1.42

 tylko

-1.41

 imediatamente

-1.41

んでいた

-1.41

 suelen

-1.40

POSITIVE LOGITS

 ۱۶

1.53

０

1.48

contentInset

1.44

 ۱۵

1.41

ícias

1.40

 Bagaimana

1.38

 ۱۸

1.37

1.36

new

1.32

Πηγές

1.30

Activations Density 0.039%