INDEX

Explanations

went missing or gone rogue

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

ioni

-0.11

go

-0.11

 farther

-0.10

oref

-0.10

etro

-0.10

GO

-0.09

Ø³ØªÙĩ

-0.09

_GO

-0.09

 Rowe

-0.08

INED

-0.08

POSITIVE LOGITS

 missing

0.15

ad

0.15

aded

0.14

à¸£à¸©

0.13

 Missing

0.13

ading

0.13

missing

0.12

 rounds

0.12

 viral

0.12

worth

0.11

Activations Density 0.032%