INDEX

Explanations

ausüben, durchführen, Trick namens

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

(proc

-0.10

assel

-0.09

FromBody

-0.09

 âĻ

-0.09

 Concurrent

-0.09

 ulaÅŁ

-0.08

 Lights

-0.08

ãģĮåĩº

-0.08

 cigaret

-0.08

POSITIVE LOGITS

 Ø§ÙĨØ¬Ø§Ùħ

0.21

 performed

0.20

 perform

0.18

è¿Ľè¡Į

0.17

 conduct

0.16

 conducted

0.16

 carried

0.16

 performs

0.15

è¡Į

0.14

 ìĪĺíĸī

0.14

Activations Density 0.115%