INDEX

Explanations

opening and closing businesses

This neuron detects mentions of how long an establishment has been open—that is, “open for…month(s)”-style duration phrases.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 小学校

-1.38

niña

-1.35

 metus

-1.34

usepackage

-1.33

 either

-1.32

ύ

-1.30

ležité

-1.23

 ristorante

-1.22

च्च

-1.20

 yearly

-1.17

POSITIVE LOGITS

ﻻ

1.73

 ویتامین

1.50

 النبات

1.48

ׇ

1.37

ﻷ

1.34

 концеп

1.33

 tilby

1.30

ästi

1.28

camis

1.27

 alianza

1.27

Activations Density 0.073%