INDEX

Explanations

courtyards and yards

The neuron activates on mentions of outdoor, enclosed spaces—specifically words like “courtyard” or “yard.”

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

 môžu

-1.67

which

-1.64

 hjälpa

-1.59



-1.53

 Unterneh

-1.53

ONYMOUS

-1.50

estries

-1.47

 οποία

-1.46

ulisan

-1.46

outube

-1.45

POSITIVE LOGITS

崀

1.68

 القديمة

1.66



1.59

creates

1.58

 españoles

1.57

began

1.55

 ним

1.54

 intérieure

1.53

;"

1.52

歃

1.51

Activations Density 0.025%