INDEX

Explanations

and introducing other items

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

-0.09

æĢĿãģĦ

-0.08

å°¤

-0.08

 Nigel

-0.08

-ÑĤÐ°ÐºÐ¸

-0.08

IDER

-0.08

arto

-0.08

ellipse

-0.07

aid

-0.07

POSITIVE LOGITS

 others

0.21

 other

0.21

/or

0.18

 various

0.18

 several

0.15

 numerous

0.14

other

0.13

 autres

0.13

others

0.13

 Ð´ÑĢÑĥÐ³Ð¸Ðµ

0.13

Activations Density 0.120%