INDEX

Explanations

then followed by next step

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

Hence

1.45

所以

1.39

Therefore

1.38

那么

1.38

Yet

1.33

Thus

1.30

ység

1.29

那麼

1.26

 pertanto

1.25

そのため

1.23

POSITIVE LOGITS

 proceeded

2.71

 proceed

2.54

ardier

2.32

 алып

2.26

ค่อย

2.24

bam

2.21

forth

2.08

ynchronous

2.07

 onwards

2.05

 lastly

2.03

Activations Density 0.168%