INDEX

Explanations

show you round, suggest turning

being measured

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

ır

0.50

分岐

0.47

쯤

0.43

zle

0.42

etta

0.41

 পি

0.41

名的

0.41

ırken

0.41

스

0.41

attiv

0.39

POSITIVE LOGITS

ระยะ

0.48

 الأن

0.47

Código

0.46

 soulful

0.45

 الحركه

0.44

 Código

0.44

 gracious

0.44

ที่เป็น

0.44

jov

0.43

 कमरे

0.42

Activations Density 0.001%