INDEX

Explanations

family and relationships

The neuron strongly activates on terms describing personal relationship milestones or status updates—engagements, splits, pregnancies, and similar family announcements.

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

</h3>

-1.31

 هذا

-1.23

 Bagdad

-1.21

⭒

-1.20

γή

-1.16

 kebaya

-1.16

 influential

-1.14

 motiva

-1.13

дии

-1.12

 musíte

-1.11

POSITIVE LOGITS

1.43

unjung

1.21

urgical

1.18

’

1.17

 then

1.16

則是

1.13

 วัน

1.12

を描く

1.11

 sudah

1.10

ømme

1.08

Activations Density 0.046%