INDEX

Explanations

proper names of people and organizations

This neuron detects capitalized name fragments—parts of proper names of people, places, or organizations.

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

我们将

-1.66

伱

-1.57

 cotidiana

-1.56

 includes

-1.56

禵

-1.55

 мы

-1.55

 those

-1.53

ujemy

-1.52

 milieux

-1.52

푆

-1.48

POSITIVE LOGITS

2.34

of

1.41

1.37

’

1.34

and

1.22

for

1.21

,"

1.20

 plus

1.20

 خدمة

1.19

”،

1.17

Activations Density 0.093%