INDEX

Explanations

invention relates to

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

 steadily

-1.31

 undeniably

-1.21

 epitom

-1.20

🦝

-1.16

 supremely

-1.15

 exquisitely

-1.15

 unrelenting

-1.15

 multifaceted

-1.14

foreignKey

-1.14

 cleverly

-1.14

POSITIVE LOGITS

 molte

1.36

 extremely

1.30

不一定

1.26

 Dinas

1.23

 diversos

1.23

 exceptionally

1.22

 largely

1.20

often

1.19

 neues

1.19

is

1.16

Activations Density 0.001%