INDEX

Explanations

question starters

The neuron detects the leading, title-case tokens at the start of a question heading (e.g., “How Do I…”, “How Would I…”) in a Q&A document.

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

as

-1.76

of

-1.54

 such

-1.50

Comme

-1.44

 while

-1.37

pielen

-1.29

 vienas

-1.27

 there

-1.26

 each

-1.25

 arşiv

-1.23

POSITIVE LOGITS

 hicimos

1.48

 staten

1.48

his

1.45

ícias

1.45

ologico

1.41

梦幻

1.41

 schmeckt

1.41

owią

1.41

ÍTULO

1.40

綣

1.40

Activations Density 1.405%