INDEX

Explanations

twitter and facebook mentions

This neuron spotlights references to social media platforms and subscription or follow links.

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

 medži

-1.60

 dėl

-1.54

 your

-1.42

 todėl

-1.41

łosi

-1.39

bolistas

-1.38

 snabbt

-1.35

 jälkeen

-1.34

 devido

-1.34

WriteLiteral

-1.34

POSITIVE LOGITS

but

1.85

and

1.81

AND

1.48

although

1.35

是個

1.32

 और

1.30

though

1.28

 บาง

1.27

 radiant

1.27

只不过

1.24

Activations Density 0.017%