INDEX

Explanations

asking for preferences

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

='{{

0.99

words

0.97

 säga

0.97

χής

0.90

 tash

0.89

<?=$

0.89

 বাড়ানোর

0.88

 freepik

0.87

संख्यक

0.87

্যো

0.86

POSITIVE LOGITS

⓪

1.42

iii

1.33

III

1.26

VIII

1.24

ii

1.22

II

1.16

ix

1.16

VII

1.14

ip

1.13

viii

1.11

Activations Density 0.054%