INDEX

Explanations

specific quantities or distributions

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

ໆ

1.46

ๆ

1.27

応用

1.26

メカ

1.24

なお

1.19

 natomiast

1.16

 revanche

1.15

Exceptions

1.15

则是

1.13

則是

1.13

POSITIVE LOGITS

ibn

1.42

`"

1.38

ğından

1.36

 empêcher

1.35

ូង

1.33

[`

1.33

 他们

1.32

Someone

1.30

ادل

1.29

 während

1.28

Activations Density 0.075%