INDEX

Explanations

language related to policy and security issues

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

Carrier

-0.07

 Carrier

-0.07

DL

-0.07

 Delta

-0.06

urb

-0.06

irut

-0.06

 Marilyn

-0.06

á»ĳc

-0.06

 Couch

-0.06

 Daytona

-0.06

POSITIVE LOGITS

 Royals

0.22

 Reading

0.21

Reading

0.18

READING

0.14

 reading

0.14

Roy

0.13

 Royal

0.13

ROY

0.13

 Berkshire

0.12

Roy

0.12

Activations Density 0.002%