INDEX

Explanations

gestures and signals

The neuron is strongly activated by content-bearing words (nouns and verbs conveying key information) rather than by common function words.

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

bold

-0.81

洛

-0.81

acterísticas

-0.80

ulada

-0.79

ctive

-0.76

Mayor

-0.76

eake

-0.75

ached

-0.74

tiert

-0.74

Cette

-0.73

POSITIVE LOGITS

 sign

1.91

 gesture

1.73

 signs

1.59

 gestures

1.59

sign

1.46

 signal

1.46

 Sign

1.40

 fingers

1.36

signs

1.35

 signals

1.33

Activations Density 0.015%