INDEX

Explanations

decrementing by amounts

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

ሷ

-1.42

盖

-1.30

ફ

-1.26

indelijk

-1.23

节省

-1.22

 λα

-1.20

渽

-1.20

 selben

-1.18

ксана

-1.16

 θέ

-1.16

POSITIVE LOGITS

but

1.33

1.29

1.23

 teníamos

1.10

だよ

1.10

н

1.08

を使った

1.03

 pensent

1.02

 querían

0.98

:「

0.98

Activations Density 0.001%