INDEX

Explanations

common word following

GH doesn't, positions for, follow the, Why it

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

 UNIT

0.45

 Wort

0.45

Goa

0.44

 كلمه

0.43

unit

0.42

*{

0.41

 நிமிட

0.41

 ملازمت

0.41

 لفظ

0.41

 الليل

0.40

POSITIVE LOGITS

ଫ

0.52

йня

0.49

ल

0.49

 auparavant

0.48

 immigrants

0.48

ל

0.48

 আগে

0.47

мию

0.47

ह

0.47

이스

0.46

Activations Density 0.001%