INDEX

Explanations

chapter numbers, dates, figures, rankings

The neuron responds to standalone numeric tokens (numbers, years, counts, chapter/episode numbers, etc.).

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

as

-1.63

 without

-1.53

or

-1.45

 with

-1.41

 what

-1.38

 dirilis

-1.34

but

-1.34

jzen

-1.34

 видит

-1.33

 deciso

-1.28

POSITIVE LOGITS

↵

2.11

 usando

1.41

 både

1.38



1.34

 indicando

1.34

 другие

1.32

 ワイド

1.31

 таких

1.28

：“……”

1.27

 также

1.24

Activations Density 0.062%