INDEX

Explanations

compound words and their prefixes related to maintenance and quality

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

æĮī

-0.06

elight

-0.06

onta

-0.06

lik

-0.06

 Norm

-0.06

 Pazar

-0.06

 exact

-0.06

dw

-0.06

yay

-0.06

 Ð¶Ðµ

-0.05

POSITIVE LOGITS

ington

0.08

 enough

0.08

venida

0.07

ãĥ¼ãĥĨãĤ£

0.07

essen

0.07

çĥ

0.07

 Enough

0.07

 considering

0.07

burg

0.07

Activations Density 0.016%