INDEX

Explanations

special characters before descriptions

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

បង្កើត

1.42

득

1.33

书记

1.32

গুলো

1.32

Broker

1.32

ત્મ

1.32

付

1.32

﨑

1.32

brook

1.31

वाहिक

1.30

POSITIVE LOGITS

1.27

and

1.26

1.24

 after

1.23

an

1.21

 early

1.17

1.15

in

1.15

 before

1.14

 without

1.14

Activations Density 0.476%