INDEX

Explanations

has_attribute OR has_many

The neuron flags identifier tokens that begin with “has” (e.g. methods or properties like hasValue, hasAirport, has_content, hasInstance).

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

ſam

-1.91

رفع

-1.84

ſem

-1.67



-1.63

댔

-1.63

 vårt

-1.62

 שלנו

-1.62

 feroit

-1.61

 весен

-1.60

丄

-1.60

POSITIVE LOGITS

无论

1.93

 Each

1.83

即使

1.78

1.67

 multitude

1.64

AND

1.63

無論

1.63

дца

1.63

 Allowing

1.60

 cuyo

1.59

Activations Density 0.020%