INDEX

Explanations

production and live use

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

 Lever

-0.10

mae

-0.09

iers

-0.09

uther

-0.09

uen

-0.09

 Rath

-0.09

ãĢĢãĤ¤

-0.09

ogene

-0.08

 Offline

-0.08

POSITIVE LOGITS

 live

0.37

 production

0.34

live

0.29

 LIVE

0.29

 Production

0.28

production

0.28

Production

0.26

 Live

0.25

Live

0.24

-live

0.24

Activations Density 0.071%

No Known Activations