INDEX

Explanations

Âł separated words

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

 intelligent

-0.12

ifest

-0.10

 Verde

-0.10

iani

-0.09

 Intelligent

-0.09

inka

-0.09

 Brill

-0.08

 hearty

-0.08

esson

-0.08

 SharedModule

-0.08

POSITIVE LOGITS

 experience

0.16

 Experience

0.14

Experience

0.13

experience

0.12

æĵ

0.11

 leadership

0.11

 Projects

0.11

ç»ıéªĮ

0.11

çĨŁ

0.11

 skills

0.11

Activations Density 0.110%