INDEX

Explanations

but not limited to

The neuron fires on occurrences of “but not” (as in “including, but not limited to”) in 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

የ

-1.50

 kelahiran

-1.47

还真

-1.43

чкой

-1.39

?」

-1.38

)),

-1.35

，

-1.34

 this

-1.33

 disconcer

-1.33

 pemer

-1.31

POSITIVE LOGITS

翃

1.53

 include

1.41

 wszystkich

1.40

那個

1.40

1.38

opheles

1.36

 miljoner

1.32

 almeno

1.30

ཎ

1.30

 overras

1.29

Activations Density 0.019%