INDEX

Explanations

Make sure

"with" followed by explanations

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

 کوشش

0.50

 intento

0.48

ମ

0.48

 nhưng

0.47

 nhiệm

0.46

্ে

0.46

 fárm

0.46

 tentativo

0.46

 opportunità

0.46

 אך

0.45

POSITIVE LOGITS

Мы

0.53

Если

0.46

 Мы

0.43

 Бі

0.42

 ராம

0.41

﹉

0.40

Hux

0.40

quete

0.40

背

0.40

fahrer

0.40

Activations Density 0.001%