INDEX

Explanations

in which

explaining concepts and why

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

 γιατί

0.50

 جهت

0.47

 Vieira

0.43

 абсолю

0.42

 jurisdictional

0.42

 យើង

0.42

峯

0.41

 वेगवेग

0.41

 ஏனெ

0.41

 Núñez

0.40

POSITIVE LOGITS

0.51

 which

0.48

ري

0.44

 supports

0.44

 prohibited

0.43

0.40

 Which

0.40

CONNECTED

0.40

 kojoj

0.39

Activations Density 0.001%