INDEX

Explanations

sums to a quantity or consequence

This neuron detects phrases expressing totals or quantities—specifically verbs like “add,” “amount,” or “translate” followed by “to” and a numeric value.

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

-1.77

沒有

-1.48

huk

-1.46

the

-1.45

能夠

-1.44

 waschen

-1.42

 ragazzo

-1.41

 хозяйства

-1.40

一切都

-1.39

all

-1.37

POSITIVE LOGITS

1.81

꙰

1.72

呖

1.59

イイ

1.57

𓐍

1.56

 monté

1.52

 falando

1.46

 répon

1.46

 talvez

1.45

嫫

1.45

Activations Density 0.012%