INDEX

Explanations

instructional verbs

rewrite the following

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

Spacer

0.40

咩

0.40

………………………………

0.39

――――

0.37

――

0.36

HERE

0.35

ForThe

0.35

的那

0.35

तरह

0.34

何

0.34

POSITIVE LOGITS

 निम्नलिखित

0.71

 each

0.64

下列

0.64

∫

0.63

 folgenden

0.62

 siguientes

0.59

 নিম্নলিখিত

0.59

 følgende

0.58

 folgender

0.58

以下の

0.57

Activations Density 0.025%