INDEX

Explanations

so/such followed by adjective

so/such indicating degree

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

更容易

0.46

っき

0.40

 শিব

0.40

decyd

0.39

大きな

0.39

ል።

0.39

 rectángulo

0.38

ッキリ

0.38

స్తోంది

0.37

ล็อต

0.37

POSITIVE LOGITS

 настолько

1.31

 এতটাই

1.04

 tellement

0.95

 sufficiently

0.89

 بحيث

0.89

 enough

0.83

 suffisamment

0.78

so

0.76

Enough

0.74

 इतना

0.73

Activations Density 0.073%