INDEX

Explanations

Proper nouns and brand names

This neuron detects proper names and titles—that is, named entities such as people, organizations, products, publications, or other specific titles.

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

 that

-1.89

 This

-1.74

 هذا

-1.66

 после

-1.55

 Preferencias

-1.52

迅速

-1.49

 também

-1.47

 this

-1.46

 sebuah

-1.41

 рассказать

-1.41

POSITIVE LOGITS

to

2.06

 bezpeč

1.56

at

1.55

 through

1.48

 černé

1.42

靦

1.40

 hlavy

1.36

Юлия

1.32

dáms

1.30

女兒

1.30

Activations Density 1.357%