INDEX

Explanations

news categories or sections

The neuron detects news‐site section or category labels (e.g. “World,” “Nation,” “Local,” “Lifestyle,” “Politics,” etc.).

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

 prodot

-1.40

 desideri

-1.27

COHOL

-1.24

crito

-1.23

 ricev

-1.22

܃

-1.21

 conosci

-1.21

 particolarmente

-1.20

 prevalent

-1.20

 When

-1.17

POSITIVE LOGITS

 إحدى

1.27

/*

1.16

 insi

1.15

 thậm

1.14

larg

1.14

1.13

 indescri

1.13

 fraudul

1.12

 行っ

1.11

 schoenen

1.11

Activations Density 0.055%