INDEX

Explanations

expressions related to surprise or astonishment

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

æ²¢

-0.06

itch

-0.06

ksen

-0.06

/misc

-0.06

umba

-0.06

 territorial

-0.06

odyn

-0.06

amus

-0.06

.force

-0.06

 misc

-0.06

POSITIVE LOGITS

 mind

0.13

 blown

0.12

 blow

0.12

 away

0.12

 socks

0.12

 Away

0.11

 Mind

0.11

 minds

0.11

 blew

0.11

mind

0.11

Activations Density 0.007%