INDEX

Explanations

prepositions followed by nouns or articles

This neuron picks out words that signal topic or content focus—i.e. discourse markers and topic‐introducing terms like “on,” “over,” “re,” “about,” and nouns such as “post,” “story,” “questions,” “information.”

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

and

-2.48

by

-1.48

 with

-1.29

in

-1.28

ējās

-1.22

 diensten

-1.21

 into

-1.17

\&

-1.16

-1.13

𝖋

-1.13

POSITIVE LOGITS

がなく

1.56

how

1.55

colgante

1.46

 aikaa

1.45

 recomiendo

1.39

 mitään

1.38

çou

1.32

谖

1.30

traueranzeige

1.30

 kentang

1.29

Activations Density 0.217%