INDEX

Explanations

glad to be involved

The neuron flags words that signal upcoming or ongoing opportunities and involvement (e.g. “opportunity,” “will,” “attending,” “support,” “encouraged,” “involved,” “going to”).

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

-1.57

or

-1.45

by

-1.43

of

-1.30

 entweder

-1.26

 либо

-1.22

、

-1.17

 either

-1.13

It

-1.12

 chyba

-1.12

POSITIVE LOGITS

鵜

1.52

如此

1.45

そも

1.41

能让

1.39

метра

1.37

nehå

1.31

thumbsup

1.31

又能

1.26

神情

1.26

了出来

1.26

Activations Density 0.070%