INDEX

Explanations

world and global concepts

This neuron detects the key content words in multiword proper names or events that follow “World” (e.g. “Suicide Prevention,” “Series,” “time clock,” “Small Animal Veterinary Association”).

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

 than

-1.38

šče

-1.34

なども

-1.29

 nhung

-1.27

חן

-1.24

 också

-1.20

ſi

-1.19

 myös

-1.19

），

-1.17

 dacă

-1.16

POSITIVE LOGITS

🌎

1.69

 world

1.36

🌏

1.28

 ihrer

1.23

 warme

1.16

Their

1.14

 heutigen

1.12

it

1.11

 worldwide

1.10

Activations Density 0.200%