INDEX

Explanations

Titles of articles, posts, or publications

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

あぁ

-1.28

 something

-1.24

つづく

-1.23

うわ

-1.22

為に

-1.21

 tersebut

-1.20

 your

-1.15

ほら

-1.12

 such

-1.05

an

-1.02

POSITIVE LOGITS

<bos>

1.41

 contextLoads

1.34

 parroquia

1.21

ebly

1.19

RIOS

1.16

がありますが

1.15

アルバム

1.14

といいます

1.14

🫁

1.13

 impecable

1.12

Activations Density 0.076%