INDEX

Explanations

business

The neuron detects mentions of business or industry contexts (e.g., “business,” “industry,” “in this business,” etc.).

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

 anledning

-1.70

 those

-1.66

郇

-1.63

Ꮚ

-1.62

ﹶ

-1.60

 navnet

-1.60

 noticeable

-1.57



-1.56

受講

-1.56

 voks

-1.56

POSITIVE LOGITS

 likes

1.87

of

1.79

↵

1.79

.”

1.69

ttal

1.50

pios

1.48

也

1.46

如同

1.45

於

1.43

Activations Density 0.012%