INDEX

Explanations

touch and voice commands

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

çľĭçĿĢ

-0.11

 looked

-0.10

 glanced

-0.10

 Theft

-0.09

look

-0.09

 pollen

-0.09

 nhÃ¬n

-0.09

_look

-0.09

 gaze

-0.09

 glance

-0.09

POSITIVE LOGITS

 blind

0.22

 Blind

0.22

 blindness

0.21

blind

0.19

æĳ¸

0.18

 touch

0.17

 cane

0.17

Br

0.17

 bump

0.16

 sight

0.16

Activations Density 0.047%