INDEX

Explanations

Anti-racism and diversity

New Auto-Interp

Configuration

Dataset (Dashboard)

Various

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bahn

-0.10

 Hurry

-0.08

 Pisa

-0.08

 Hardware

-0.08

 Ingeniería

-0.08

 bankrupt

-0.08

 patent

-0.07

aon

-0.07

compressed

-0.07

 financed

-0.07

POSITIVE LOGITS

 feminist

0.11

 nuanced

0.11

 perpetrators

0.11

 queer

0.10

 Inclus

0.10

 inclus

0.10

 sexist

0.10

 sexism

0.10

 respectful

0.10

 activism

0.09

Activations Density 0.064%