INDEX

Explanations

many people

The neuron fires on words and phrases that express quantity or amount (e.g. many, half, all, every, few, most, thousands).

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

the

-1.60

if

-1.40

 from

-1.19

便利です

-1.18

 хуже

-1.18

and

-1.13

Boas

-1.11

or

-1.08

 what

-1.07

his

-1.06

POSITIVE LOGITS

 people

2.52

 folks

1.59

 everyone

1.52

 árbitro

1.51

 individuals

1.48

odziel

1.46

 mojej

1.40

 kleid

1.38

 моего

1.37

you

1.36

Activations Density 0.057%