INDEX

Explanations

heart 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

 deduct

-0.10

Ë

-0.09

EMPLARY

-0.09

 bunch

-0.08

.myapplication

-0.08

 Î£Ïį

-0.08

unal

-0.08

 bumped

-0.08

pH

-0.08

 Ð²ÐºÐ»Ð°Ð´

-0.08

POSITIVE LOGITS

 heart

0.31

heart

0.21

-heart

0.19

å¿ĥ

0.19

 Heart

0.18

Heart

0.17

 hart

0.16

 hearts

0.15

 å¿ĥ

0.15

 ÑģÐµÑĢÐ´

0.15

Activations Density 0.079%