INDEX

Explanations

Korean polite question endings

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

cha

1.17

 staying

1.16

(~

1.16

(),

1.15

 respected

1.15

 years

1.15

 supposed

1.14

 chances

1.12

born

1.11

jahr

1.11

POSITIVE LOGITS

。

1.73

と

1.58

และ

1.52

।

1.46

此外

1.42

然而

1.39

とその

1.36

）。

1.36

)。

1.34

 valamint

1.33

Activations Density 0.005%