INDEX

Explanations

object type checking

This neuron responds to Markdown image‐insertion markers (e.g. the “![” that begins an embedded image).

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

in

-1.34

-1.30

 Their

-1.27

 betyd

-1.27

 följande

-1.24

 those

-1.23

("")]

-1.22

 ciudadana

-1.20

 encontrarás

-1.19

ätz

-1.16

POSITIVE LOGITS

of

1.54

 мы

1.43

เรา

1.38

ές

1.34

ዣ

1.31

fassen

1.27

 सकते

1.27

何时

1.25

 mientras

1.23

 pueden

1.22

Activations Density 0.016%