INDEX

Explanations

words and phrases related to thematic contrasts and evaluations

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

éĥ

-0.06

emek

-0.06

robat

-0.06

cki

-0.06

 dwarf

-0.06

enge

-0.06

jt

-0.06

ona

-0.06

 OMIT

-0.06

ì¦Ī

-0.06

POSITIVE LOGITS

ãĥ«ãĥĪ

0.06

ReadWrite

0.06

Ø¬Ø±

0.06

Ø³Ø±

0.06

Î½Î·ÏĤ

0.06

 thÃ´i

0.06

DataStream

0.06

Î²Î±

0.06

angu

0.06

æ²ĸ

0.06

Activations Density 0.170%