INDEX

Explanations

phrases expressing skepticism or uncertainty

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

ä½ķãģĭ

-0.07

æŁĲ

-0.07

ä¸ĢåĮº

-0.07

ä¸Ģç§į

-0.07

ä¸Ģå®ļ

-0.07

>[]

-0.07

æŃ£åľ¨

-0.06

utzer

-0.06

ãģĤãĤĭ

-0.06

esome

-0.06

POSITIVE LOGITS

few

0.26

 very

0.22

 hardly

0.20

few

0.19

 none

0.19

Few

0.18

 fewer

0.18

Few

0.18

 scarcely

0.18

 little

0.17

Activations Density 0.137%