INDEX

Explanations

explaining the secret

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ı

-1.36

..."

-1.27

—

-1.19

...'

-1.16

—"

-1.13

」

-1.13

)--

-1.13

REGON

-1.12

-1.10

 situaciones

-1.09

POSITIVE LOGITS

.,

1.58

,”

1.40

viembre

1.22

 juft

1.18

 "-";

1.16

.(

1.15

 jotka

1.13

 richting

1.11

IDADES

1.11

ancer

1.09

Activations Density 0.013%