INDEX

Explanations

various professional roles

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

 their

-1.50

 that

-1.48

now

-1.44

 they

-1.38

が

-1.31

 їх

-1.18

who

-1.17

 today

-1.12

 που

-1.09

 each

-1.07

POSITIVE LOGITS

 مختلف

1.27

 различни

1.24

various

1.20

 différente

1.16

corrhi

1.14

塀

1.14

ต่างๆ

1.13

larda

1.13

before

1.12

 diverses

1.10

Activations Density 0.041%