INDEX

Explanations

digits followed by characters

The neuron activates on standalone numeric tokens, i.e. number literals.

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

 there

-1.45

 while

-1.16

 before

-1.13

as

-1.12

 with

-1.09

for

-1.07

in

-1.07

and

-1.02

 came

-1.02

 without

-1.00

POSITIVE LOGITS

鹩

1.27

 этих

1.12

 ส์

1.09

瘣

1.09

 innych

1.09

aturation

1.06

uride

1.06

ピール

1.06

🫃

1.05

èmes

1.05

Activations Density 0.056%