INDEX

Explanations

not followed by a word

evolving, initially hostile, not disappearing

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

 છો

0.38

koi

0.37

不稳定

0.37

这是一个

0.37

không

0.37

}[!

0.36

 অ্যান

0.36

বে

0.35

ắn

0.35

\"{

0.35

POSITIVE LOGITS

orious

0.51

 permitted

0.49

 contradictory

0.45

 hemos

0.44

sal

0.44

و

0.43

有關

0.42

 conç

0.42

ופה

0.42

 brigades

0.42

Activations Density 0.000%