INDEX

Explanations

entertainment stars and performers

This neuron primarily detects the word “star” when it’s used to label or refer to a celebrity.

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

隄

-1.52

шные

-1.41

They

-1.35

جموع

-1.34

 fubject

-1.34

)=\

-1.34

CARACTERIST

-1.32

 smarter

-1.31

 gentil

-1.30

])){

-1.30

POSITIVE LOGITS

1.62

1.55

 peut

1.39

 puede

1.28

1.24

 kunnen

1.23

 vitales

1.23

1.20

1.18

 może

1.18

Activations Density 0.053%