INDEX

Explanations

being with friends and family

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

 ãĥ½

-0.11

 ï¾ī

-0.10

'Â´

-0.10

ulle

-0.09

ytt

-0.09

rage

-0.09

ILI

-0.09

 teammate

-0.09

ÐµÑģÑı

-0.09

 Parad

-0.08

POSITIVE LOGITS

 friends

0.20

 family

0.15

 loved

0.14

 Friends

0.14

ering

0.13

outh

0.13

olding

0.13

indo

0.12

ought

0.12

 group

0.12

Activations Density 0.065%