INDEX

Explanations

names and organizations

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

for

-1.59

by

-1.46

They

-1.45

 first

-1.44

homonymie

-1.39

 like

-1.33

 while

-1.29

 without

-1.26

君は

-1.23

 using

-1.22

POSITIVE LOGITS

 prípade

1.29

junctive

1.29

particularly

1.26

</tfoot>

1.26

 transmettre

1.23

citement

1.23

特别是

1.22

tainable

1.22

stander

1.20

rium

1.20

Activations Density 0.225%