INDEX

Explanations

factual consistency of summary

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

onn

-0.10

ÂĢÂĢ

-0.09

ãģĬãĤĬ

-0.09

tober

-0.09

---</

-0.09

ascript

-0.09

åł¡

-0.08

 SPELL

-0.08

íĺĳ

-0.08

idl

-0.08

POSITIVE LOGITS

 below

0.11

 statement

0.10

 accurately

0.10

(summary

0.10

 Statement

0.09

yy

0.09

-sum

0.09

=sum

0.09

ç®Ĺ

0.09

 Below

0.09

Activations Density 0.031%