INDEX

Explanations

self-compassion and other concepts

explaining code and fostering responsible behavior

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 Prosecutor

0.48

 Gesetz

0.48

 Usted

0.47

κέ

0.47

 Kultur

0.45

 Está

0.45

 Junto

0.45

 Estoy

0.44

 Ihrem

0.44

 ಕಾನೂ

0.44

POSITIVE LOGITS

Breaking

0.51

Punch

0.49

Walls

0.48

Scott

0.46

ForItem

0.46

นาน

0.46

Shake

0.45

Coming

0.45

Dean

0.44

 countDown

0.44

Activations Density 0.002%