INDEX

Explanations

advice or sensitive questions

New Auto-Interp

Top Features by Cosine Similarity

Configuration

Prompts (Dashboard)

10,000 prompts, 128 tokens each

Dataset (Dashboard)

lmsys/lmsys-chat-1m

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ï¼ģãĢį\n\n

-0.09

 Gree

-0.08

ernal

-0.08

 known

-0.08

 Attributes

-0.08

 fich

-0.08

 Status

-0.08

icial

-0.08

atus

-0.08

""

-0.08

POSITIVE LOGITS

 advice

0.15

Advice

0.12

 Advice

0.12

 approach

0.10

 approaches

0.10

 question

0.10

 sensitive

0.10

 Sanat

0.09

 tips

0.09

appro

0.09

Activations Density 0.342%