INDEX

Explanations

price figures

The neuron fires strongly on pricing and sale information—labels like “Sale,” “Price,” “Regular,” dollar signs, and the associated numbers or discounts.

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

おはよう

-1.55

 кроссовки

-1.45

 soldado

-1.41

ₒ

-1.40

勇者

-1.40

—

-1.38

 Русский

-1.36

⎯

-1.33

Ошибка

-1.33

—

-1.32

POSITIVE LOGITS

 again

2.13

our

2.11

 only

2.02

not

1.85

 also

1.58

all

1.55

 ponownie

1.52

 like

1.49

也不是

1.43

Who

1.41

Activations Density 0.006%