INDEX

Explanations

features, components, or accessories

The neuron detects words that call out parts, features, or accessories included with a product (e.g. “included,” “removable,” “customizable”).

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

⟣

-1.63

ኗ

-1.59

 dépenses

-1.56

崀

-1.56

蛏

-1.55

iamo

-1.55

桫

-1.53

 dégâts

-1.52

皷

-1.52

 famí

-1.52

POSITIVE LOGITS

all

2.08

 both

1.81

 using

1.70

 from

1.68

 multifaceted

1.63

 every

1.60

 taking

1.60

 everything

1.59

 EVERY

1.55

 three

1.54

Activations Density 0.077%