INDEX

Explanations

proper nouns, places, names

The neuron is detecting tokens used in geographic or locational descriptions—prepositions and place‐related words that introduce where something is.

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

 <=",

-1.34

!!!!!!

-1.34



-1.29

!!!!!!!

-1.28

 obat

-1.21

 další

-1.19

 другие

-1.17

 startet

-1.16

 cómoda

-1.16

౩

-1.14

POSITIVE LOGITS

ğlık

1.42

 dicendo

1.30

 trattano

1.26

şehir

1.25

 tertawa

1.23

gæ

1.21

があり

1.20

Zdrav

1.17

layan

1.17

 menangis

1.16

Activations Density 0.109%