INDEX

Explanations

those who do something

The neuron fires on the first content word (verb or modifier) immediately following the demonstrative “Those,” i.e. it detects “Those …” clause openings.

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

ynı

-1.76

鲥

-1.75

 現貨

-1.66

/"

-1.61



-1.55

 itſelf

-1.53

呲

-1.53

Both

-1.51

When

-1.50

Its

-1.49

POSITIVE LOGITS

 themselves

2.02

to

1.94

 their

1.64

 する

1.40

“

1.39

gaat

1.38

gráficos

1.37

ৢ

1.35

 have

1.34

Ưu

1.34

Activations Density 0.058%