INDEX

Explanations

first to

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

eseorang

1.42

 percobaan

1.38

 iemand

1.34

いくつかの

1.33

Lucky

1.29

 alguém

1.28

 seorang

1.26

 seseorang

1.25

buah

1.23

another

1.22

POSITIVE LOGITS

 calculate

1.15

 spacing

1.08

 determine

1.05

 facilitate

1.05

max

1.02

 Фор

1.02

 ка

1.02

 require

1.00

can

1.00

 inch

1.00

Activations Density 0.001%