INDEX

Explanations

court of appeals decisions

The neuron fires on tokens in headings that name appellate courts—most strongly when “Court” appears in the phrase “Court of Appeals (for …)”.

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

was

-2.11

 необходимо

-2.05

 ‘‘

-1.96

穑

-1.91

’

-1.90

 tendrá

-1.86

了出去

-1.84

 cuando

-1.81

该如何

-1.81

thschild

-1.75

POSITIVE LOGITS

葙

1.91

to

1.83

or

1.74

='

1.73

ther

1.72

»).

1.72

[#

1.71

Nº

1.70

ल

1.67

🌙

1.66

Activations Density 0.038%