INDEX

Explanations

delimiters within text

The neuron detects punctuation used as example or option delimiters—especially slashes ( / ) and parentheses.

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

 kakao

-1.91

yng

-1.80

 elever

-1.77

 feil

-1.70

 takk

-1.69

鲧

-1.67

 usta

-1.66

 skak

-1.66

 unſ

-1.63

zio

-1.62

POSITIVE LOGITS

in

1.63

 dramatic

1.46

bosky

1.45

 umfang

1.42

for

1.38

yym

1.36

 diferentes

1.35

praia

1.33

 einzigartigen

1.31

壺

1.31

Activations Density 0.009%