INDEX

Explanations

angel followed by names

The neuron strongly activates on occurrences of the substring “Angel” (in names like Angel, Angelo, Angelica, etc.) within the 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

-2.63

杮

-2.19

是有

-2.19

情報を

-2.08

ak

-2.08

 infuse

-2.08

 afic

-2.06

 renowned

-2.06

 cubiertos

-2.03

end

-2.00

POSITIVE LOGITS

“

2.86

of

2.77

2.59

&#

2.38

istocene

2.28

 ﬁrst

2.25

—

2.19

蟜

2.13

2.11

ían

2.09

Activations Density 0.017%