INDEX

Explanations

separating clauses and titles

The neuron fires on mentions of corporate entity names—specifically company names with organization suffixes (e.g. “& Sons,” “Inc.”) in business‐title contexts.

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

 coarser

-1.48

-1.47

šiť

-1.43

 möjligt

-1.37

("")]

-1.35

',

-1.31

icose

-1.31



-1.27

갇

-1.27

的其他

-1.26

POSITIVE LOGITS

 печень

1.59

 kämp

1.39

 Pandemie

1.23

◆◆

1.20

 eivät

1.17

 også

1.16



1.16

 bús

1.15

 Songtext

1.15

Activations Density 0.073%