INDEX

Explanations

recently or lately

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

hled

-0.09

 decades

-0.09

èĲ¬

-0.09

//{{

-0.09

 dÃ¨s

-0.09

oras

-0.09

older

-0.08

elik

-0.08

 ÑĦÑĥÐ½Ð´Ð°

-0.08

eus

-0.08

POSITIVE LOGITS

 lately

0.49

 recently

0.42

 recent

0.42

æľĢè¿ĳ

0.38

recent

0.38

Recently

0.32

 Recently

0.31

 Recent

0.30

ately

0.29

Recent

0.28

Activations Density 0.191%