INDEX

Explanations

the followed by common nouns

specific phrases after "the"

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

 begrij

0.43

 behalf

0.42

𝒐

0.39

 следующий

0.39

 организм

0.37

 ব্যস্ত

0.37

кою

0.37

лів

0.36

稣

0.36

 яко

0.36

POSITIVE LOGITS

 Indies

0.39

 これは

0.38

ysty

0.37

這是

0.35

_",

0.35

($

0.35

astore

0.34

 %@",

0.34

究

0.34

 adı

0.33

Activations Density 0.197%