INDEX

Explanations

legal condemnation

The neuron fires on digit‐based identifiers in legal headers—e.g. docket numbers, years, dates, and other numeric citations.

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

enei

-0.78

Alf

-0.77

Resume

-0.77

Alphabet

-0.75

doug

-0.75

Captain

-0.74

 Félix

-0.72

-0.71

('--

-0.71

EXP

-0.71

POSITIVE LOGITS

 eminent

2.77

 condemnation

2.58

 condemn

2.33

eminent

2.11

 condemned

1.97

 condemning

1.95

 Condem

1.88

tak

1.80

minent

1.69

 condem

1.61

Activations Density 0.058%