INDEX

Explanations

end of sentence or question

words ending sentences

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

 zatem

0.39

 kursi

0.38

larda

0.37

tea

0.36

 keng

0.36

re

0.36

Dense

0.36

go

0.35

ेस

0.35

vang

0.35

POSITIVE LOGITS

湎

0.36

性价比

0.35

拨款

0.35

 professionally

0.34

 ಹೆಚ್ಚ

0.33

整体

0.30

 রাষ্ট্রের

0.30

预算

0.30

 Personally

0.30

 Honestly

0.30

Activations Density 0.006%