INDEX

Explanations

articles indicating the introduction of new entities or concepts

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

 sayÄ±da

-0.08

vla

-0.07

(er

-0.06

âĢĮØ§ÙĨ

-0.06

datable

-0.06

bÃŃr

-0.06

ç´¹

-0.06

Äįem

-0.06

heimer

-0.06

ibbon

-0.06

POSITIVE LOGITS

.k

0.13

woke

0.08

/k

0.08

 which

0.07

which

0.07

 feat

0.07

 Miles

0.07

 frequent

0.06

cui

0.06

 whose

0.06

Activations Density 0.034%