INDEX

Explanations

in / ∈ mathematical contexts

This neuron detects nouns like “element,” “member,” or “subset” when they appear in “X of Y” expressions (e.g. “element of M,” “member of A,” “subset of M”).

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

これを

-1.45

oherty

-1.40

 یا

-1.34

-1.33

 lekker

-1.32

*]

-1.31

啭

-1.31

部長

-1.30

馴

-1.29

POSITIVE LOGITS

也会

1.66

 Moreover

1.64

 other

1.58

But

1.52

 Other

1.50

Otras

1.47

 regresó

1.46

年も

1.45

 Their

1.42

 posee

1.42

Activations Density 0.052%