INDEX

Explanations

all I/you can/do/need

pronoun followed by state or action

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

}/\

0.42

με

0.39

тта

0.38

珽

0.38

㈠

0.38

сама

0.37

排序

0.36

不可能

0.36

名为

0.36

澎

0.36

POSITIVE LOGITS

 ever

0.58

 EVER

0.48

 Ever

0.45

 zuen

0.44

 lakukan

0.43

 عليك

0.43

দ্দা

0.43

 lousy

0.42

 كثيرا

0.42

 বেশিরভাগ

0.41

Activations Density 0.119%