INDEX

Explanations

conjunctions and contrasting statements

and, or, including

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 오히려

0.48

反而

0.45

 또한

0.45

 inoltre

0.44

 Alas

0.44

 свои

0.43

 Tats

0.43

ेद

0.42

 이는

0.41

 azonban

0.41

POSITIVE LOGITS

 आणि

0.74

 sorta

0.71

And

0.68

 และ

0.64

And

0.62

Which

0.61

Và

0.61

 včetně

0.61

 그러니까

0.60

Including

0.60

Activations Density 0.472%