INDEX

Explanations

The neuron detects named geopolitical entities and related proper nouns (countries, regions, political/organizational names).

The neuron activates on three-letter all-caps acronyms.

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

0.37

 یک

0.33

0.32

İ

0.32

0.31

the

0.31

ফ

0.31

 hilarious

0.31

 दो

0.31

न

0.30

POSITIVE LOGITS

.[

0.33

.}

0.33

).

0.31

 systems

0.31

izate

0.31

izers

0.31

 alike

0.30

0.29

 अवधारणा

0.29

].

0.28

Activations Density 0.000%