INDEX

Explanations

de la bataille de

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

 nested

-0.10

 mood

-0.09

 âĢĲ

-0.09

 hitch

-0.09

unge

-0.09

 rall

-0.09

 wartime

-0.09

meyi

-0.09

idot

-0.08

uries

-0.08

POSITIVE LOGITS

 Battle

0.17

 Waterloo

0.16

 Marathon

0.15

 Hastings

0.14

Battle

0.14

 battle

0.13

 engagements

0.12

battle

0.12

 Therm

0.12

 slag

0.11

Activations Density 0.064%