question words and modals

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

ÂĢÂĢ

-0.14

******č\n

-0.14

¶Į

-0.14

.Formatter

-0.12

įng

-0.12

.Currency

-0.11

ĥ½

-0.10

%č\n

-0.10

ıa

-0.10

¨ë¶Ģ

-0.10

POSITIVE LOGITS

 however

0.10

0.08

 therefore

0.08

 roughly

0.08

 then

0.08

 enim

0.08

Activations Density 1.036%