INDEX

Explanations

targeting specific audiences

The neuron fires on phrases and numbers that specify intended audiences or targeting metrics (e.g. “targeted at,” “aimed at,” “target audience,” and related numeric values).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 about

-1.32

on

-1.16

↵

-1.10

 start

-1.09

它是

-1.05

 also

-1.02

闵

-1.01

 affects

-1.00

Sometimes

-0.99

 change

-0.97

POSITIVE LOGITS

BoxShadow

1.31

 géographie

1.30

]--;

1.24

createCell

1.22

EndInit

1.19

 девочек

1.16

 bezpeč

1.15

盉

1.14

 encontraron

1.14

 peduli

1.13

Activations Density 0.078%