INDEX

Explanations

phrases that describe group composition and structure

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

istrovstvÃŃ

-0.10

indr

-0.08

AccessorType

-0.08

iesen

-0.08

ormsg

-0.08

gaard

-0.08

 DXGI

-0.08

engu

-0.08

Ð¾Ð»Ð¸ÑĤ

-0.07

 tomu

-0.07

POSITIVE LOGITS

 members

0.11

 representatives

0.10

members

0.08

 member

0.08

all

0.07

 representation

0.07

 equal

0.07

 representative

0.07

 some

0.07

 exclusively

0.07

Activations Density 0.018%