INDEX

Explanations

legal and social contexts

study independently without tutorial

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

Brow

0.48

Mon

0.48

นี

0.47

Meu

0.47

针对

0.46

恽

0.46

前景

0.45

DoubleClick

0.45

 outstanding

0.45

专注

0.45

POSITIVE LOGITS

 alcohols

0.45

 plasmids

0.43

 vessels

0.42

 tubes

0.42

 intoxicated

0.42

 jurisdictions

0.41

 sentidos

0.41

 seasonings

0.41

)=>{

0.40

 ハンドバッグ

0.40

Activations Density 0.001%