INDEX

Explanations

power generation, sources, stations, outages

The neuron consistently activates on mentions of core energy-related nouns—especially “power,” “energy,” and “electricity.”

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

’

-2.22

-1.74

-1.66

-1.60

-1.54

ിൽ

-1.52

</h1>

-1.48

‘

-1.48

ed

-1.47

 Mischung

-1.47

POSITIVE LOGITS

 január

1.60

 február

1.57

ydın

1.55

扨

1.52

enken

1.46

könyv

1.45

for

1.42

utusan

1.42

umbent

1.40

uchte

1.38

Activations Density 0.029%