INDEX

Explanations

references to geographic locations and regions

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

owell

-0.08

 lesb

-0.08

upy

-0.08

Muon

-0.08

idunt

-0.08

uron

-0.08

esine

-0.07

ihn

-0.07

udeau

-0.07

erie

-0.07

POSITIVE LOGITS

 Asia

0.07

 territories

0.07

 region

0.07

 North

0.06

 Latin

0.06

 Europe

0.06

 Region

0.06

 Marshall

0.06

 Americas

0.06

 territory

0.05

Activations Density 0.018%