INDEX

Explanations

Order of

This neuron detects mentions of “Order” (including forms like “Ordensstaat”) in contexts naming chivalric, military, or religious orders and related state entities.

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

but

-1.95

-1.93

䛗

-1.64

𞤤

-1.59

abyrinth

-1.55

縢

-1.55

螣

-1.52

躂

-1.51

 avevano

-1.48

and

-1.47

POSITIVE LOGITS

大概

1.91

 چقدر

1.80

 dichas

1.77

selbe

1.76

 nokt

1.73

нование

1.72

 burbu

1.69

 notare

1.68

 malzeme

1.67

BOYS

1.67

Activations Density 0.011%