INDEX

Explanations

reporting instructions or speech

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

ulus

-0.09

Cin

-0.09

 Mustang

-0.08

 Minute

-0.08

strup

-0.08

Ð¾ÑĢÐ°Ð·

-0.08

engu

-0.08

imin

-0.08

aways

-0.08

POSITIVE LOGITS

 missing

0.11

 said

0.11

 saying

0.11

 çľģ

0.11

.say

0.11

say

0.10

plu

0.10

 originally

0.10

 says

0.10

 told

0.10

Activations Density 0.001%