INDEX

Explanations

always there for the player

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

 vÅ¯bec

-0.12

eya

-0.11

 altogether

-0.11

 ëª¨ëĳĲ

-0.11

 Ð²Ð¾Ð¾Ð±ÑīÐµ

-0.11

 Madd

-0.09

cba

-0.09

LAB

-0.09

 AGAIN

-0.09

çļĨ

-0.09

POSITIVE LOGITS

 whenever

0.18

 never

0.17

enever

0.17

 whether

0.17

 except

0.16

 dÃ¹

0.16

 unless

0.15

Never

0.15

 Never

0.14

 Ð½Ð¸ÐºÐ¾Ð³Ð´Ð°

0.14

Activations Density 0.127%

No Known Activations