INDEX

Explanations

objectification and degradation of individuals

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 stuff

0.45

PRI

0.40

 Stuff

0.40

 everything

0.40

PRE

0.39

 answers

0.39

devi

0.38

 zaji

0.38

 Needs

0.38

DEL

0.37

POSITIVE LOGITS

 личности

0.67

 dignity

0.60

 женщин

0.58

 अपमान

0.56

 достоин

0.56

 dignidad

0.54

 уважа

0.51

 dignit

0.51

 insulted

0.50

 ด่า

0.50

Activations Density 0.052%