INDEX

Explanations

about oneself, love, it

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

 Ð·Ð²Ð¸ÑĩÐ°Ð¹

-0.19

 Ð²Ð¸ÑıÐ²

-0.17

 Ð¸Ð¼Ñĥ

-0.15

 Ð·Ð°ÑıÐ²

-0.15

 zÃ¡stup

-0.14

 ÐµÐ»ÐµÐºÑĤÑĢÐ¾Ð½

-0.14

 Ð¸Ð½ÑĦÐ¾ÑĢÐ¼Ð°

-0.13

 ÑģÐ»ÑĥÑĩÐ°ÐµÐ²

-0.13

 Ð¼ÐµÑģÑĤÐ°

-0.12

 Ð¾Ð³ÑĢÐ°

-0.12

POSITIVE LOGITS

 ÑĤÐ¾Ð¼

0.16

ÐµÐ½Ð¸Ð¸

0.15

tom

0.15

èº«ä¸Ĭ

0.14

Ð»ÐµÐ½Ð¸Ð¸

0.13

ÑĤÐ¾Ð¼

0.11

 tome

0.11

Ð°Ð»ÑĮÐ½Ð¾Ð¼

0.11

 &#8203;&#8203;

0.11

Ð°Ð½Ð¸Ð¸

0.11

Activations Density 0.063%