INDEX

Explanations

which of the following questions

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

umb

1.37

by

1.36

to

1.34

pet

1.33

laş

1.28

bere

1.27

bys

1.27

laştır

1.25

ের

1.23

oly

1.20

POSITIVE LOGITS

COURSE

1.66

 course

1.62

 aquello

1.61

course

1.56

>*</

1.52

 }^{*

1.49

りの

1.48

 COURSE

1.48

recer

1.46

நுட்ப

1.45

Activations Density 1.130%