INDEX

Explanations

arrival and welcome

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 hesitant

-0.11

lien

-0.10

 Conserv

-0.10

/art

-0.10

 crawled

-0.10

instanc

-0.10

${(

-0.10

ifact

-0.10

 Passed

-0.10

 Knee

-0.10

POSITIVE LOGITS

 welcoming

0.39

 welcome

0.30

 greet

0.28

 greeting

0.28

 Welcome

0.27

 welcomes

0.27

 awaiting

0.27

 waiting

0.27

 awaited

0.27

Wel

0.25

Activations Density 0.192%