INDEX

Explanations

your liking, advantage, or convenience

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

garian

1.30

话题

1.29

bla

1.27

 politician

1.26

okine

1.26

waa

1.25

項

1.24

 scathing

1.22

ionine

1.21

 hoon

1.20

POSITIVE LOGITS

 consente

1.55

Knowing

1.53

 prerogative

1.53

consent

1.52

而言

1.49

getNum

1.47

 Determines

1.47

Determ

1.45

允许

1.45

 Knowing

1.44

Activations Density 0.125%