INDEX

Explanations

phrases indicating relationships or associations represented by the word "of"

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

him

-0.07

 PMID

-0.07

uf

-0.07

.scalablytyped

-0.07

ãģĭãģ«

-0.07

ä½³

-0.06

 VerfÃ¼g

-0.06

uc

-0.06

istine

-0.06

 jinÃ½

-0.06

POSITIVE LOGITS

few

0.10

 several

0.10

 many

0.09

strup

0.08

many

0.07

Several

0.07

veral

0.07

 Several

0.07

cle

0.06

_many

0.06

Activations Density 0.026%