INDEX

Explanations

foreign words

The neuron fires on words naming groups or individuals as information sources or speakers (e.g. officials, experts, residents, Democrats, say/said).

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

There

-1.51

).

-1.45

 there

-1.34

 erwäh

-1.33

so

-1.27

the

-1.25

had

-1.24

These

-1.21

 "...

-1.19

 nahm

-1.19

POSITIVE LOGITS

 mijne

1.63

 allerlei

1.62

 quiser

1.52

 geda

1.46

 zijne

1.39

 بیشتری

1.39

 blauw

1.38

 bloemen

1.37

 området

1.36

 blauwe

1.34

Activations Density 0.087%