INDEX

Explanations

author or speaker introductions

The neuron fires on section headings introducing metadata—especially “About the Author” (and similar “About the …” headers like “Bio:” or “About the Book”).

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

 also

-1.47

 always

-1.41

 discussed

-1.35

老祖

-1.28

 asked

-1.27

 attacked

-1.27

 poderosa

-1.27

 valioso

-1.25

 thought

-1.25

を考える

-1.23

POSITIVE LOGITS

1.45

😜

1.40

1.39

 kunde

1.38

1.35

1.34

мся

1.30

卓

1.28

 selepas

1.27

 dikutip

1.27

Activations Density 0.015%