INDEX

Explanations

references to health-related organizations and incidents

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

³

-0.08

elho

-0.07

éĳĳ

-0.07

ÙĳÙĲ

-0.07

 Stim

-0.07

rieg

-0.07

Ø§ØŃÛĮ

-0.07

euillez

-0.07

ourg

-0.07

 withStyles

-0.07

POSITIVE LOGITS

 health

0.12

 Health

0.11

-health

0.10

 Epidemi

0.10

health

0.10

Health

0.10

 epidemi

0.09

CDC

0.09

 Surveillance

0.09

 surveillance

0.09

Activations Density 0.069%