INDEX

Explanations

preventing negative outcomes

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

 masculinity

1.31

 inaccur

1.21

ेच्छा

1.20

xious

1.16

 Utara

1.14

 incorrectly

1.13

constit

1.13

ाइ

1.13

 affirmat

1.13

 retrieved

1.13

POSITIVE LOGITS

 ngừa

1.62

Prevent

1.56

 prevent

1.44

 Prevent

1.43

 preventing

1.39

 Preventing

1.35

 prevents

1.32

prevent

1.31

jau

1.29

 prevention

1.24

Activations Density 0.235%