INDEX

Explanations

departments and ministries

The neuron fires on mentions of “Department,” i.e. references to organizational or governmental departments.

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

ãs

-1.55

:"-"`

-1.55

omock

-1.48

狳

-1.39

actéristi

-1.38

ܐ

-1.38

ibouti

-1.34

imentos

-1.34

тся

-1.32

땃

-1.31

POSITIVE LOGITS

of

2.14

is

1.74

↵↵

1.45

 verbess

1.34

 reportedly

1.31

 menyimpan

1.27

 ホテル

1.27

<bos>

1.27

 شیائومی

1.26

 escuelas

1.25

Activations Density 0.006%