INDEX

Explanations

phrases that encourage innovative thinking and creativity

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

#af

-0.07

 é¢¨

-0.07

bern

-0.07

lop

-0.07

'gc

-0.07

une

-0.07

enth

-0.06

Ú¯Ø§Ø¨

-0.06

lep

-0.06

è©±

-0.06

POSITIVE LOGITS

 traditional

0.07

 Traditional

0.07

 bounds

0.07

limits

0.06

 conventional

0.06

 boundaries

0.06

Traditional

0.06

traditional

0.06

bounds

0.06

 linear

0.06

Activations Density 0.002%