INDEX

Explanations

divisionSBSubmittedZuni

The neuron fires on numeric literals—especially decimal numbers (floating-point values) embedded 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

to

-2.17

in

-1.92

by

-1.57

so

-1.55

or

-1.49

on

-1.45

 berdua

-1.38

 then

-1.36

–

-1.34

 would

-1.30

POSITIVE LOGITS

ſſen

1.48

ował

1.39

 gruppen

1.38

 stoff

1.38

 bolj

1.38

our

1.34

 anciennes

1.34

1.33

 diferentes

1.33

Activations Density 0.000%