INDEX

Explanations

in order of

The neuron detects references to ordering or sequence, especially phrases centered on the word “order” (e.g. “in order,” “order of precedence,” “listed in no particular order”).

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

atedral

-1.52

 dazzling

-1.45

 mammoth

-1.37

過ごす

-1.34

żna

-1.34

-1.32

riwal

-1.31

 marvellous

-1.31

żd

-1.28

↵↵

-1.27

POSITIVE LOGITS

 ſou

1.67

谞

1.63

 eſſ

1.53

 consulte

1.53

er

1.49

 pido

1.48

ala

1.42

 überprü

1.41

 ſol

1.39

 juſ

1.39

Activations Density 0.025%