INDEX

Explanations

that is

comma followed by a word

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

Nor

0.48

AND

0.40

 Conversely

0.39

Nor

0.39

和一个

0.38

 tanpa

0.37

 ilma

0.37

 lihtsalt

0.36

 lama

0.36

 uite

0.36

POSITIVE LOGITS

 अर्थात्

0.45

 тобто

0.43

сті

0.42

ос

0.41

 அதாவது

0.40

substantial

0.40

也就是

0.39

 অর্থাৎ

0.38

 /**@

0.38

即

0.38

Activations Density 0.276%