INDEX

Explanations

numerical ranges and durations

The neuron selectively activates on numeric tokens—especially decimal or percentage figures—highlighting measurements or statistics in the text.

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

 know

-1.92

</em>

-1.83

</h2>

-1.80

-1.78

—

-1.75

“

-1.73

ным

-1.66

—

-1.63

咙

-1.63

POSITIVE LOGITS

 yelek

1.78

of

1.71

Можно

1.63

Какие

1.63

 Salón

1.61

استفاده

1.60

 parís

1.59

Даже

1.57



1.56

продам

1.49

Activations Density 0.122%