INDEX

Explanations

points and geometric relationships

The neuron primarily responds to numeric tokens (standalone numbers or numeric expressions) 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

甭

-1.71

==""){

-1.55

齁

-1.54

LATIN

-1.54

恚

-1.52

 ratón

-1.47

 they

-1.45

不愧

-1.45

DeclareMath

-1.43

恁

-1.43

POSITIVE LOGITS

說

1.48

 other

1.44

so

1.38

 good

1.34

我覺得

1.27

 excessive

1.25

two

1.24

她

1.23

all

1.23

 paramount

1.21

Activations Density 0.019%