INDEX

Explanations

before making health decisions

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

통

0.96

 başta

0.96

회

0.91

 Prop

0.90

 Pizza

0.89

 consistent

0.89

 Jerry

0.88

Lac

0.87

多多

0.87

통

0.86

POSITIVE LOGITS

講的是

1.20

icip

1.18

ძლ

1.17

iyordu

1.16

ąć

1.16

榖

1.15

ędzie

1.13

roduction

1.12

bought

1.10

omass

1.10

Activations Density 0.053%