INDEX

Explanations

numbers related to three

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

amu

-0.10

onya

-0.09

rem

-0.09

 Hakk

-0.09

dib

-0.09

 pair

-0.08

?,?,?,?,

-0.08

chap

-0.08

agas

-0.08

Psy

-0.08

POSITIVE LOGITS

 trio

0.14

 three

0.12

 ä¸ī

0.12

ä¸ī

0.12

ä¸īä¸ª

0.12

 Trio

0.11

 Three

0.11

 trilogy

0.11

 Ã¼Ã§

0.11

three

0.11

Activations Density 0.176%