INDEX

Explanations

collections and libraries

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

もしくは

-1.44

 preferably

-1.31

wiście

-1.30

頂き

-1.30

actéristi

-1.28

整个

-1.27

 they

-1.27

 despond

-1.26

或是

-1.23

泐

-1.20

POSITIVE LOGITS

of

1.36

tvo

1.30

聪

1.24

⅘

1.21

 Pamph

1.21

 diego

1.20

红色的

1.19

들은

1.15

챌

1.15

诟

1.13

Activations Density 0.035%