INDEX

Explanations

instantly share code

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

 timely

-0.16

 eventual

-0.11

 dáº§n

-0.11

 eventually

-0.10

 gradual

-0.10

thinkable

-0.10

 Morr

-0.10

abo

-0.10

isty

-0.09

 Ð¿Ð¾ÑģÑĤÐµÐ¿

-0.09

POSITIVE LOGITS

 grat

0.13

 chÃ³ng

0.12

 Grat

0.11

aneous

0.10

aneously

0.10

 Sullivan

0.10

 prot

0.09

 Cher

0.09

-before

0.09

avel

0.09

Activations Density 0.114%