INDEX

Explanations

a range of

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

ä¸Ģç§į

-0.16

ä¸ĢäºĽ

-0.12

à¤¹à¤¨

-0.09

à¸Ħà¸ĵà¸°

-0.09

 varieties

-0.09

Ð¿ÑĢÐ¸Ð¼ÐµÑĢ

-0.09

lyph

-0.09

_rng

-0.09

ities

-0.08

ushima

-0.08

POSITIVE LOGITS

 different

0.18

ä¸įåĲĮçļĦ

0.16

different

0.14

ä¸įåĲĮ

0.13

 Different

0.13

 diferentes

0.13

Different

0.12

 diferente

0.12

iw

0.11

 emotions

0.10

Activations Density 0.021%