INDEX

Explanations

few from

without exchanging data

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

取り付け

0.52

 カーボン

0.46

 rubles

0.45

↵↵↵↵↵

0.45

放送

0.44

 الواحده

0.43

酊

0.43

َی

0.42

ровой

0.42

रुण

0.42

POSITIVE LOGITS

laşı

0.45

Tara

0.43

ዬ

0.42

 disag

0.42

arct

0.42

紵

0.42

setRoi

0.40

 ذر

0.39

eta

0.39

 нерав

0.38

Activations Density 0.000%