INDEX

Explanations

adharnos, engines Policing, Food Lab, ATTORNEY GENERAL, Jet Video

The neuron responds to capitalized named entities—proper nouns such as people’s names, place names, organization names, and 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

it

-1.88

any

-1.67

or

-1.63

and

-1.55

roscopy

-1.34

 saca

-1.27

 Zwar

-1.26

人で

-1.20

 find

-1.17

 much

-1.16

POSITIVE LOGITS

訁

1.34

 таких

1.33

:”

1.27

quarelle

1.25

。」

1.23

 hygge

1.23

 スリム

1.23

Estas

1.22

 relig

1.22

 Estas

1.20

Activations Density 0.106%