INDEX

Explanations

for and its equivalents

words followed by for

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

 improves

0.85

 celebrated

0.73

 breaks

0.70

ემის

0.62

 sellest

0.57

law

0.50

latego

0.50

詬

0.47

 afin

0.43

ಕ್ತಿಯ

0.43

POSITIVE LOGITS

 для

3.77

 für

3.48

สำหรับ

3.42

 για

3.36

for

3.25

 pentru

3.23

 برای

3.23

 untuk

3.22

для

3.22

 voor

3.13

Activations Density 3.936%