INDEX

Explanations

object near location

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

Negative Logits

 overhead

-0.11

 preceded

-0.10

 à¤ĩà¤¸à¤®

-0.09

ä¸Ńãģ«

-0.09

ograd

-0.09

criptor

-0.09

 Underground

-0.09

anson

-0.09

|R

-0.08

åħ¶ä¸Ń

-0.08

POSITIVE LOGITS

 near

0.50

near

0.34

 Near

0.33

 outside

0.30

 Ð±ÑĸÐ»Ñı

0.28

Near

0.27

 prÃ¨s

0.25

_near

0.24

 gáº§n

0.24

è¿ĳ

0.24

Activations Density 0.149%

No Known Activations