INDEX

Explanations

number ranges

The neuron activates on standalone numeric tokens—especially numbers denoting counts, ranges, or other quantitative values in structured headings and lists.

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

 referrerpolicy

-1.75

—–

-1.63

 antaranya

-1.59

땃

-1.54

()}>

-1.52

xvii

-1.52

綃

-1.52

ᚔ

-1.51

 Jérusalem

-1.48

 ($("#

-1.47

POSITIVE LOGITS

 поддержка

1.67

but

1.66

 для

1.63

 setiap

1.62

and

1.60

 also

1.58

1.55

 этого

1.46

 disesuaikan

1.45

 подоб

1.45

Activations Density 0.044%