INDEX

Explanations

mentioning or followed by what

stronger emotion or intensity

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ಪ

0.52

 shameful

0.46

ｚ

0.45

確立

0.42

 bosom

0.42

効

0.42

 Asturias

0.41

 vanguard

0.41

 Ее

0.41

POSITIVE LOGITS

Brock

0.49

্তিক

0.48

 cultivés

0.47

Codes

0.46

焕

0.46

 uploads

0.45

speeds

0.45

 derivadas

0.45

Giovanni

0.44

 Utilizing

0.44

Activations Density 0.001%