INDEX

Explanations

publisher names and books

The neuron detects publisher imprints and related bibliographic metadata (e.g. publisher names and associated price/publication details).

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

镲

-1.56

bym

-1.53

 nowa

-1.53

麃

-1.47

opropyl

-1.45

 startet

-1.45

isn

-1.41

quite

-1.41

weren

-1.41

 prace

-1.41

POSITIVE LOGITS

 meeste

1.32

 will

1.30

 should

1.29

⅝

1.28

ukkig

1.26

Nuorodos

1.26

⇰

1.26

 From

1.23

 would

1.22

 Some

1.21

Activations Density 0.016%