INDEX

Explanations

foreign place names and origins

The neuron is specializing in spotting proper nouns that are geographic place names (cities, regions, countries, etc.).

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

 itself

-1.62

 these

-1.55

🪛

-1.43

 servidores

-1.40

SizedBox

-1.40

cestry

-1.37

 AssemblyCompany

-1.28

 zoude

-1.26

 imó

-1.24

but

-1.23

POSITIVE LOGITS

んでいる

1.50

 encontré

1.32

➡

1.32

щено

1.31

 conozco

1.30

 that

1.28

೨

1.28

 wenigstens

1.27

 estoy

1.27

EXAMPLE

1.27

Activations Density 0.465%