INDEX

Explanations

color and brush attributes

This neuron detects style property names related to colors or brushes (e.g. “background,” “color,” “BorderBrush,” “GridLinesBrush,” “sliderColor,” “theme_color”).

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

-1.60

as

-1.49

may

-1.47

In

-1.47

 There

-1.40

If

-1.35

-1.33

 take

-1.30

it

-1.27

 Only

-1.27

POSITIVE LOGITS

 ДЕ

1.47

źródło

1.38

 ЛИ

1.37

 kancel

1.35

kita

1.30

 parachoque

1.30

selben

1.29

 ПРЕ

1.29

 стоковые

1.26

ейчас

1.25

Activations Density 0.003%