INDEX

Explanations

punctuation followed by capital letters

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

to

-1.69

 vitales

-1.42

 tačiau

-1.41

</strong>

-1.31

 kartą

-1.23

ists

-1.23

 problemer

-1.22

咔

-1.22

にほんブログ村

-1.21

-1.20

POSITIVE LOGITS

鈊

1.49

퐿

1.45

 दोनों

1.32

 تضيف

1.31

 Ours

1.27

鋇

1.27

瓛

1.27

triques

1.22

 μόνο

1.22

)$,

1.22

Activations Density 0.013%