INDEX

Explanations

The neuron fires on numeric tokens—especially decimal numbers or other multi‐digit numeric values.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

but

0.69

izes

0.69

acres

0.69

槨

0.64

berries

0.64

goers

0.64

 ecos

0.63

 reun

0.63

 spinoff

0.63

</h3>

0.61

POSITIVE LOGITS

 استفاده

0.77

 استخدام

0.72

ﺵ

0.70

ل

0.69

 використовувати

0.68

 mettere

0.68

自動

0.67

ٹنگ

0.67

旋转

0.67

atória

0.66

Activations Density 0.002%