INDEX

Explanations

academic papers and conferences

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ƴ

-0.76

行事

-0.75

épisode

-0.72

 läs

-0.70

维生素

-0.68

žky

-0.68

着一

-0.68

Foi

-0.67

εύ

-0.66

蒼

-0.66

POSITIVE LOGITS

 workshop

1.76

 workshops

1.65

 invited

1.46

 Workshop

1.38

 tutorial

1.32

workshop

1.29

 tutorials

1.28

 poster

1.23

Workshop

1.21

 Invited

1.20

Activations Density 0.028%