INDEX

Explanations

the word "links," especially when preceded by "External"

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

</caption>

-0.69

 initComponents

-0.68

 MainAxisSize

-0.66

peka

-0.66

 Eocene

-0.63

BoxShadow

-0.63

 ModelExpression

-0.62

 whiteboard

-0.60

帖最后由

-0.60

 Mucha

-0.59

POSITIVE LOGITS

 links

3.38

 link

3.22

 Link

3.00

 Links

2.98

links

2.91

link

2.83

Link

2.78

 LINK

2.70

Links

2.67

 LINKS

2.61

Activations Density 0.255%

No Known Activations