INDEX

Explanations

head, face, neck

The neuron strongly activates on words naming human body parts.

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

</b>

-1.70

 пройдет

-1.65

up

-1.62

嫉妒

-1.59

ate

-1.56

 herido

-1.53

ブログ村

-1.49

𖤣

-1.47

 novidades

-1.45

窜

-1.45

POSITIVE LOGITS

—

1.66

、“

1.66

we

1.59

ześnie

1.55

\\

1.48

namn

1.42

ʽ

1.40

σιμοποι

1.38

 negras

1.38

 все

1.38

Activations Density 0.033%