INDEX

Explanations

funicular for views of hills

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

 crossings

-0.11

 Nile

-0.10

berger

-0.10

 upright

-0.10

 overhead

-0.09

pearance

-0.09

 Lake

-0.09

 rank

-0.09

baugh

-0.09

 gorge

-0.09

POSITIVE LOGITS

 hill

0.33

 Hill

0.29

hill

0.24

ä¸ĺ

0.22

 hills

0.20

 Hills

0.14

 summit

0.14

 ÑħÐ¾Ð»

0.14

hil

0.14

 ìĸ¸

0.13

Activations Density 0.061%