INDEX

Explanations

functions and operations

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

Negative Logits

wax

-0.11

orsk

-0.10

 ãĢĢ

-0.10

 Craw

-0.10

ast

-0.09

Lif

-0.09

lif

-0.09

overn

-0.08

Â¨Â¨

-0.08

_PWR

-0.08

POSITIVE LOGITS

 simply

0.17

 return

0.16

è¿ĶåĽŀ

0.16

 simplement

0.16

 returning

0.14

return

0.14

 Ð¿ÑĢÐ¾ÑģÑĤÐ¾

0.13

 è¿ĶåĽŀ

0.13

 einfach

0.13

 Return

0.12

Activations Density 0.169%

No Known Activations