INDEX

Explanations

cities

This neuron activates on named geographic locations—especially city (and sometimes country) names—in running text.

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

Nebraska

-1.09

⩾

-0.99

uit

-0.97

 pseudo

-0.92

 computing

-0.92

 bulunan

-0.92

ccccc

-0.88

isesi

-0.88

 tiet

-0.88

 helper

-0.87

POSITIVE LOGITS

 Tokyo

1.91

 London

1.91

 Moscow

1.66

 Seoul

1.60

 Sydney

1.60

 Berlin

1.55

 Paris

1.53

 Beijing

1.51

 Cairo

1.48

Tokyo

1.44

Activations Density 0.091%