INDEX

Explanations

so ... so, too ... too

The neuron activates on the intensifier “so” when used to emphasize (e.g., “so unique,” “so easily,” “so slow”).

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

welli

-1.37

 gomez

-1.33

 sebastian

-1.32

-1.31

лары

-1.27

 afbeelding

-1.27

 kardashian

-1.27

鋮

-1.27

 jenner

-1.27

 maki

-1.26

POSITIVE LOGITS

*/

1.69

全ての

1.55

ﾏ

1.47

 jedynie

1.46

—

1.43

Also

1.41

ofluor

1.40

 не

1.40

できるだけ

1.35

 they

1.34

Activations Density 0.015%