INDEX

Explanations

military leaders and juntas

The neuron detects references to military or dictatorial authority (e.g. ranks like “general,” terms like “dictator,” “junta,” and events like “coup”).

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

elizabeth

-1.49

 even

-1.47

 other

-1.42

 لئے

-1.38

亟

-1.36

 jeśli

-1.36

tranh

-1.33

 podían

-1.30

lasher

-1.27

+"&

-1.27

POSITIVE LOGITS

Здесь

1.37

1.30

1.22

⣟

1.22

 appareils

1.20

ajouter

1.20

gantes

1.16

Перед

1.16

 Frühjahr

1.16

 after

1.14

Activations Density 0.036%