INDEX

Explanations

any query or question

The neuron fires on numeric tokens—especially decimal numbers or standalone numbers—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

-1.80

and

-1.54

-1.45

—

-1.44

 will

-1.37

by

-1.35

</b>

-1.23

各大

-1.20

has

-1.20

of

-1.20

POSITIVE LOGITS

 this

1.63

 orice

1.56

 miniaturka

1.50

 această

1.48

nieuw

1.47

aneity

1.47

 cualquier

1.46



1.46

 qualquer

1.45

 sprüche

1.45

Activations Density 0.134%