INDEX

Explanations

the presence of the word "to"

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

otch

-0.07

 Ð·Ð°Ð½

-0.07

ickerView

-0.06

aisal

-0.06

á»Ľ

-0.06

olle

-0.06

ÎµÎ¹Î±

-0.06

 rais

-0.06

 invol

-0.06

.workspace

-0.06

POSITIVE LOGITS

 Benedict

0.07

 McInt

0.07

exact

0.07

 Slot

0.07

/repos

0.06

.vs

0.06

heck

0.06

demand

0.06

 Bran

0.06

Activations Density 0.000%