INDEX

Explanations

understanding certain languages

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

 almost

-0.13

 Almost

-0.11

Almost

-0.11

 both

-0.11

 Various

-0.10

 mostly

-0.10

 rather

-0.09

 bÃ¥de

-0.09

igar

-0.09

 casi

-0.09

POSITIVE LOGITS

 certain

0.40

 certains

0.29

 Certain

0.29

Certain

0.27

æŁĲ

0.24

 bestimm

0.22

ertain

0.21

 some

0.19

æľīäºĽ

0.18

 older

0.17

Activations Density 0.100%