INDEX

Explanations

so are, so is, so lacks

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

üz

-1.36

 parfüm

-1.23

 parfü

-1.20

 pertes

-1.20

 Drž

-1.17

Ursula

-1.17

 nyel

-1.16

られた

-1.16

 khỏ

-1.15

zeß

-1.15

POSITIVE LOGITS

 other

1.74

 others

1.66

Ι

1.62

 autre

1.35

 else

1.35

其他

1.32

prende

1.31

 they

1.30

 다른

1.29

vs

1.27

Activations Density 0.060%