INDEX

Explanations

sub-prefixes

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

 overheating

1.45

рана

1.39

 gebo

1.34

 μπορού

1.33

ləşdirilib

1.33

癜

1.29

퀀

1.28

oramic

1.27

砝

1.26

 jooks

1.25

POSITIVE LOGITS

stantially

1.41

sequent

1.32

dued

1.31

tle

1.29

lime

1.26

wesen

1.25

stant

1.24

stantial

1.23

consciously

1.22

oxone

1.17

Activations Density 0.314%