INDEX

Explanations

your own, oneself, yourself

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

isha

-0.11

ashi

-0.10

onth

-0.09

igs

-0.09

 Himself

-0.09

elts

-0.09

 IOCTL

-0.09

omain

-0.09

barang

-0.08

olta

-0.08

POSITIVE LOGITS

 oneself

0.51

 ones

0.38

 Ones

0.36

ones

0.28

 your

0.27

ä½łçļĦ

0.21

your

0.21

 yourself

0.21

ONES

0.21

own

0.19

Activations Density 0.371%