INDEX

Explanations

body part actions

The neuron fires on mentions of body parts (e.g. “feet,” “eyes,” “hands”).

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

-1.74

of

-1.67

嫆

-1.66

 unangemessene

-1.65

柠

-1.63

ьем

-1.60

чный

-1.52

ᐕ

-1.50

Pgina

-1.48

 Avez

-1.48

POSITIVE LOGITS

 sowohl

1.51

见状

1.49

 those

1.44

1.41

 такими

1.37

The

1.36

 軽量

1.35

 первые

1.35

namn

1.34

 obtend

1.34

Activations Density 0.036%