INDEX

Explanations

phrases related to requesting more information or assistance

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

ÑĢÐ¾Ð¶

-0.07

ZERO

-0.07

 Ø§Ø±Ø²

-0.07

 Cumhur

-0.06

 Phonetic

-0.06

emailer

-0.06

_MPI

-0.06

Æ°á»Ľi

-0.06

 Ballard

-0.06

íĻľ

-0.06

POSITIVE LOGITS

 information

0.10

 details

0.10

 info

0.07

_information

0.07

 informaciÃ³n

0.07

details

0.07

information

0.07

berman

0.06

-information

0.06

unate

0.06

Activations Density 0.004%