INDEX

Explanations

husband and wife

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

 Girlfriend

-0.19

 girlfriend

-0.17

 unmarried

-0.15

 Dating

-0.14

 celib

-0.13

 girlfriends

-0.13

 dating

-0.13

Dating

-0.13

 boyfriend

-0.13

æģĭ

-0.12

POSITIVE LOGITS

 husband

0.56

 wife

0.50

 Husband

0.43

hus

0.41

å¤«

0.41

å¦»

0.41

wife

0.40

ä¸Īå¤«

0.40

 Wife

0.39

-wife

0.36

Activations Density 0.132%