INDEX

Explanations

Instead of passive voice

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 fundamental

0.41

 professionally

0.40

0.39

 apro

0.37

 Austrian

0.37

 undisclosed

0.37

 musically

0.37

انیه

0.36

ვილი

0.36

 fundamentals

0.36

POSITIVE LOGITS

 ಭೂ

0.52

 তারা

0.50

 उनके

0.47

enfants

0.46

 他们

0.45

敵人

0.44

他們

0.43

उनके

0.43

頑

0.42

उनकी

0.42

Activations Density 0.000%

This feature has no known activations.