INDEX

Explanations

rocket, missile, propellant, oxidizer

The neuron detects dense technical jargon and specialized terms—especially long acronyms, chemical names, and other highly domain-specific vocabulary.

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

czych

-0.75

ೕ

-0.73

inesis

-0.71

MCG

-0.71

aig

-0.71

등학교

-0.69

 sledge

-0.69

Зри

-0.68

Fug

-0.68

setHorizontal

-0.68

POSITIVE LOGITS

 rocket

1.43

 ракет

1.31

 Rocket

1.16

Rocket

1.09

 rockets

1.01

 propellant

1.00

rocket

0.94

 missile

0.93

ellant

0.91

 liquid

0.90

Activations Density 0.050%