INDEX

Explanations

proper nouns and places

The neuron fires on capitalized proper names—brand, product, software, or other named-entity 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

 natale

-1.48

 Congresso

-1.30

 ceramica

-1.30

 Formato

-1.27

 responsabili

-1.27

 Londra

-1.24

 castel

-1.24

 cabina

-1.21

 cioc

-1.21

 combatt

-1.20

POSITIVE LOGITS

</h2>

1.88

by

1.70

</h1>

1.50

↵↵

1.43

 including

1.42

and

1.31

 keine

1.23

 그리고

1.22

 using

1.22

on

1.22

Activations Density 0.382%