INDEX

Explanations

current officer roles

The main thing this neuron does is detect formal ranks and official job or military titles.

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

 former

-1.51

Former

-1.25

 Former

-1.19

former

-1.02

 ehemaligen

-0.99

 býval

-0.87

 verkauft

-0.84

 начинает

-0.83

如同

-0.81

 دوران

-0.80

POSITIVE LOGITS

 currently

1.89

 current

1.73

currently

1.70

current

1.48

 현재

1.44

Currently

1.41

当前

1.38

 actualmente

1.37

Actualmente

1.33

 aktuell

1.27

Activations Density 0.142%