INDEX

Explanations

episodes, songs, chapters, matches

The neuron fires strongly on words naming or labeling media/content formats (e.g. essay, book, film, episode, podcast).

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

</h2>

-1.83

 also

-1.41

<i>

-1.36

isn

-1.27

她们

-1.23

</h1>

-1.18

-1.17

Как

-1.17

is

-1.16

よい

-1.16

POSITIVE LOGITS

midler

1.73

lenia

1.66



1.56

vermectin

1.45

banez

1.43

rfan

1.41

zzie

1.38

spects

1.38

ḿ

1.37

 kanskje

1.36

Activations Density 0.372%