INDEX

Explanations

king Abhaya

The neuron consistently fires on multi‐token proper nouns and specialized topic terms—especially names (of deities, kings, sages, etc.) and key mythological or historical concepts.

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

 octobre

-1.41

GIH

-1.26

 πρ

-1.22

ícias

-1.21

 Dicapai

-1.19

 septembre

-1.17

تصویر

-1.16

 コート

-1.16

 сентября

-1.15

 фаса

-1.14

POSITIVE LOGITS

 сахара

1.61

 coloro

1.45

 regarding

1.39

离婚

1.31

 красивая

1.27

 according

1.26

Regarding

1.23

of

1.23

 inalámb

1.23

 lievito

1.23

Activations Density 0.003%