INDEX

Explanations

parts of compound words

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

-1.38

 вместо

-1.35

 değil

-1.33

Mə

-1.27

 eenige

-1.27

np

-1.26

miyor

-1.24

</code>

-1.22

:'#

-1.22

 сообщили

-1.21

POSITIVE LOGITS

 when

1.73

 whose

1.48

 lorsqu

1.37

内置

1.37

好啊

1.37

 each

1.27

 later

1.27

 from

1.27

 ロゴ

1.27

 then

1.26

Activations Density 0.170%