INDEX

Explanations

signal processing and mathematics

The neuron fires on numeric values and numerical constants (whole numbers and decimals) in the text.

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

<em>

-1.69

 idea

-1.41

<i>

-1.39

-1.38

de

-1.38

-1.33

 change

-1.30

⅛

-1.25

 presence

-1.25

to

-1.25

POSITIVE LOGITS

från

1.98

🩴

1.76

妪

1.74

när

1.70

 ensalada

1.59

nın

1.59

 उपकरण

1.59

 Moscú

1.57

 alberg

1.55

 rasg

1.54

Activations Density 0.205%