INDEX

Explanations

brand new

new concepts and introductions

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

倨

0.36

चुरल

0.36

утбу

0.35

 großer

0.34

ஞ்சு

0.33

}_{+}

0.33

BlackElo

0.33

 hoher

0.32

скохозяй

0.32

க்கமாக

0.32

POSITIVE LOGITS

new

4.91

 নতুন

4.44

new

4.38

 nieuwe

4.34

 nuevos

4.25

 новых

4.25

 nuovi

4.25

新的

4.22

 neuen

4.22

 новые

4.22

Activations Density 1.739%