INDEX

Explanations

CSS properties

The neuron activates on numeric literals (numbers and measurements like pixel values) in the CSS.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 désolés

-1.60

and

-1.55

—

-1.55

)][

-1.50

iocene

-1.45

Of

-1.43

 most

-1.41

or

-1.41

IZACIÓN

-1.38

him

-1.36

POSITIVE LOGITS

趼

1.81

 incessantly

1.66

receta

1.64

Sekarang

1.63

景点

1.60

絎

1.59

schuhe

1.56

ﺢ

1.55

鱨

1.55

 کردم

1.54

Activations Density 0.013%