INDEX

Explanations

punctuation and conjunctions

code structure

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

</iframe>

0.44

 سابق

0.41

صاف

0.40

 acetate

0.37

게

0.36

 Bestand

0.36

 الأصفر

0.36

Cet

0.36

 Brill

0.36

 അറിയ

0.35

POSITIVE LOGITS

ᱞ

0.50

}^{+}=

0.40

ᐳ

0.40

rme

0.39



0.39

ৃহ

0.39

bak

0.39

['-

0.39

κού

0.38

राह

0.37

Activations Density 0.001%