INDEX

Explanations

first retrospective exhibition

The neuron strongly activates on emphasis or formatting markup (e.g. <em>…</em> tags or underscores used to italicize text).

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

not

-1.39

 other

-1.38

on

-1.37

 most

-1.34

 first

-1.32

at

-1.26

 last

-1.25

all

-1.21

 only

-1.21

 just

-1.20

POSITIVE LOGITS

</em>

1.83

 eenig

1.49

なのだ

1.48

</sup>

1.47

</strong>

1.42

↵

1.40

 participer

1.39

 eenen

1.36

 accél

1.35

 ſol

1.30

Activations Density 0.000%