INDEX

Explanations

place names

The neuron activates on named geographic locations (cities, counties, states, etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 régionales

-1.44

nefis

-1.40

 bestimmten

-1.38

 désert

-1.38

 internasional

-1.38

甞

-1.37

硐

-1.35

짭

-1.34

 telepon

-1.34

+)$

-1.33

POSITIVE LOGITS

’

1.35

 praktijk

1.20

 colorear

1.19

 hehe

1.16

 ativos

1.16

 albeit

1.16

 なる

1.16

 toepassing

1.16

no

1.10

 prób

1.10

Activations Density 0.068%