INDEX

Explanations

references to illness or medical conditions, particularly cancer and other serious diseases

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

ault

-0.08

 æİ

-0.07

Î¿Îº

-0.07

vla

-0.07

alama

-0.07

ench

-0.07

ulk

-0.07

abei

-0.06

igham

-0.06

achu

-0.06

POSITIVE LOGITS

gressive

0.06

undi

0.06

ä¸¥

0.06

.TYPE

0.06

 rare

0.06

ful

0.06

 Browns

0.06

imenti

0.06

Beh

0.06

chron

0.06

Activations Density 0.021%