INDEX

Explanations

US cities and states

The neuron activates on mentions of administrative-region names (U.S. states or similar subdivisions, and Canadian provinces).

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

 about

-1.55

 späteren

-1.35

 verranno

-1.25

来店

-1.15

 quoque

-1.13

 avrà

-1.13

úsqueda

-1.13

梘

-1.12

trze

-1.11

irá

-1.09

POSITIVE LOGITS

},

1.41

 الرغم

1.39

și

1.38

rité

1.24

จาก

1.18

vü

1.16

殒

1.15

虬

1.15

telijke

1.15

 Dana

1.14

Activations Density 0.044%