INDEX

Explanations

parentheses and their usage in numerical contexts

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

otify

-0.07

ÐµÑĢÐ¶

-0.07

cw

-0.07

gere

-0.06

roi

-0.06

etti

-0.06

eden

-0.06

ãģ»ãģĨ

-0.06

hare

-0.06

ãĥ»ãĥŀ

-0.06

POSITIVE LOGITS

éĻ©

0.07

hma

0.07

Ð±ÑĥÐ´ÑĮ

0.06

 height

0.06

elu

0.06

 afar

0.06

izz

0.06

eltas

0.06

height

0.06

æ³ī

0.06

Activations Density 0.004%