INDEX

Explanations

programming code snippets

This neuron fires on numeric tokens—particularly version numbers, exit codes, build IDs, and other standalone digit sequences in code or config files.

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

Cy

-0.93

xF

-0.89

＝

-0.88

encers

-0.84

CY

-0.83

 latter

-0.83

Cy

-0.81

-0.79

٪

-0.79

เกี่ยวกับ

-0.78

POSITIVE LOGITS

){}

0.85

 annulus

0.82

有難

0.79

などが

0.78

 currentValue

0.77

נצ

0.75

(){}

0.73

”!

0.69

備

0.69

varnothing

0.69

Activations Density 0.223%