titles of authority

This neuron detects words naming official roles or professional 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

 producenta

-1.61

atother

-1.60

 evrops

-1.55

 utilizamos

-1.53

 embalaje

-1.52

 seleccionados

-1.50

 adaptable

-1.49

 explorar

-1.48

 buscan

-1.46

Visión

-1.46

POSITIVE LOGITS

of

1.94

1.69

1.60

1.59

("

1.59

1.55

1.45

 this

1.43

Activations Density 0.004%