INDEX

Explanations

Japanese continuations

various scripts and encoded text

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 자신

0.27

脸

0.26

 تصور

0.25

对待

0.25

 TreeSet

0.25

*)

0.24

涤

0.24

’;

0.24

💭

0.24

 实现

0.24

POSITIVE LOGITS

ନ୍ଦ

0.33

いた

0.31

いる

0.30

 frecuente

0.28

ری

0.28

kannya

0.27

дық

0.27

ونکی

0.27

 infrequently

0.27

Activations Density 0.001%