INDEX

Explanations

obstacles and buildings

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

-0.10

©

-0.09

owers

-0.09

 Ð³ÑĢÐ¸

-0.09

arl

-0.09

er

-0.09

 Arbitrary

-0.08

 Ludwig

-0.08

ushed

-0.08

Hab

-0.08

POSITIVE LOGITS

 objects

0.27

 obstacles

0.23

éļľ

0.22

ob

0.20

 Objects

0.20

objects

0.18

 Ð¿ÑĢÐµÐ¿

0.17

Objects

0.17

 obstacle

0.16

 obstruction

0.16

Activations Density 0.082%