INDEX

Explanations

counting length or size

The neuron is detecting numeric literals (numbers) in code or 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

᠂

-1.31

"=

-1.29

-1.27

あった

-1.24

されていた

-1.23

izacja

-1.23

 detalhes

-1.19

%")

-1.19

 exemplos

-1.18

POSITIVE LOGITS

any

1.40

its

1.38

can

1.38

 when

1.38

侮

1.23

علق

1.21

ность

1.20

選択肢

1.18

 which

1.16

 each

1.15

Activations Density 0.015%