INDEX

Explanations

whether or not

The neuron activates on numeric tokens (especially numbers, including decimals).

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

and

-2.47

in

-1.91

 untuk

-1.82

！”

-1.75

 With

-1.70

 because

-1.69

इटम

-1.64

)$,

-1.62

 dengan

-1.54

ètes

-1.50

POSITIVE LOGITS

 they

1.91

鱚

1.77

 caratteri

1.72

 utilisés

1.68

れました

1.67

秾

1.67

那种

1.63

 présenté

1.57

を採用

1.56

 précie

1.56

Activations Density 0.025%