INDEX

Explanations

references to marginalized identities and social justice themes

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤ¸

-0.06

 submarine

-0.06

 manned

-0.06

 íĨ

-0.06

emens

-0.06

EXTERN

-0.06

ĻĤ

-0.06

Geh

-0.06

 submarines

-0.05

é¼

-0.05

POSITIVE LOGITS

 trans

0.19

 transgender

0.18

 Trans

0.18

Trans

0.15

trans

0.14

/trans

0.14

 TRANS

0.13

-trans

0.13

.trans

0.13

 tranny

0.13

Activations Density 0.009%