INDEX

Explanations

only possibility or condition

only ... can

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

 encarg

0.48

ड़क

0.46

ờ

0.46

Wnd

0.45

ׁ

0.45

 ossia

0.44

ളി

0.43

 기본적인

0.43

 основных

0.41

 argumento

0.40

POSITIVE LOGITS

 Only

0.68

 Только

0.67

 only

0.58

Only

0.58

才能

0.56

could

0.55

 толькі

0.54

 could

0.54

เท่านั้น

0.54

only

0.53

Activations Density 0.090%