INDEX

Explanations

aerospace, military, or specific entities

The neuron strongly activates on aviation terms—especially specific aircraft names, model designations, and related words like “Yak-1,” “liaison aircraft,” “monoplane,” “Dakotas,” or “Spitfire.”

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

-1.77

 был

-1.44

 nowych

-1.36

 telefono

-1.33

 myös

-1.25

mpagne

-1.25

 employee

-1.22

chairs

-1.20

 também

-1.20

 была

-1.20

POSITIVE LOGITS

ección

1.28

vação

1.21

íticos

1.20

⇡

1.19

 parto

1.11

 contribue

1.09

سجيل

1.08

 lẻ

1.07

avoir

1.05

Bip

1.05

Activations Density 0.030%