INDEX

Explanations

This neuron detects words that introduce places or settings—prepositions and verbs like “in,” “of,” “found,” “located,” “based,” and “takes place in” that signal location or origin.

locations following "in"

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

 other

-1.44

 others

-1.41

 there

-1.40

and

-1.27

if

-1.27

as

-1.27

 each

-1.23

our

-1.22

see

-1.21

''

-1.16

POSITIVE LOGITS

 vermel

1.55

 estadounid

1.49

 egenskaper

1.48

 siè

1.46

 chô

1.45

 constater

1.45

 vœux

1.35

 horloge

1.35

 chers

1.34

 provincie

1.34

Activations Density 0.127%