INDEX

Explanations

div with class

This neuron fires on markup element tokens—i.e. the `<…>` tags (HTML/JSX component or element names) in code.

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

 ediciones

-0.96

centos

-0.95

 Öffentlich

-0.93

 since

-0.91

 dejar

-0.91

 năm

-0.88

 appuntamento

-0.88

listbox

-0.85

 February

-0.80

}/>

-0.79

POSITIVE LOGITS

 className

1.50

className

1.33

 style

1.09

 styles

0.86

style

0.85

чних

0.85

 høj

0.85

 klasse

0.83

 ailes

0.81

 styled

0.81

Activations Density 0.010%