INDEX

Explanations

listing items or examples

This neuron detects “etc.”-type list closers—words or phrases like the Chinese “等” or English “and the like.”

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

 ofic

-1.36

 Allemagne

-1.31

姈

-1.30

⚣

-1.30

 momentan

-1.30

慫

-1.28

 trebui

-1.27

beispiel

-1.25

Ꭳ

-1.25

踯

-1.24

POSITIVE LOGITS

are

1.69

 have

1.64

“

1.38

 things

1.29

accor

1.25

\&

1.23

 highly

1.23

1.21

 aspects

1.20

 relatively

1.17

Activations Density 0.008%