INDEX

Explanations

quantity per unit

The neuron fires on “per” when it’s used to express a rate or quantity per unit (e.g. cost per day, pounds per person, admissions per insured person).

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

 других

-1.34

があり

-1.27

や

-1.27

อื่น

-1.23

と

-1.21

つまり

-1.18

がなく

-1.16

或者

-1.14

 herido

-1.13

 другим

-1.13

POSITIVE LOGITS

 each

2.39

 annum

1.71

 every

1.64

Each

1.44

 month

1.30

 Each

1.29

每

1.27

 mỗi

1.27

Chaque

1.22

んですよね

1.19

Activations Density 0.022%