INDEX

Explanations

this choice, purchase

The neuron detects promotional, positively-framed statements about how a product looks or adds appeal (e.g. “will look great in your home,” “offers a delicate touch of beauty”).

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

着一个

-2.05

了一個

-1.70

一辆

-1.63

一个小

-1.55

création

-1.50

着自己的

-1.47

pantalon

-1.45

一阵

-1.43

 estranho

-1.38

一颗

-1.34

POSITIVE LOGITS

and

1.66

or

1.66

1.45

 einverstanden

1.38

but

1.34

is

1.31

you

1.30

hasMany

1.29

 atau

1.28

 With

1.27

Activations Density 0.461%