INDEX

Explanations

negative numbers and symbols

negative and command flags

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

াহরণ

0.39

 হইয়৷

0.39

চৈতন্যের

0.37

ருங்கள்

0.37

জেল

0.37

 ख़ु

0.36

 freshmen

0.36

NewUrlParser

0.36

 publicly

0.35

কানিক্যাল

0.35

POSITIVE LOGITS

(-

3.50

(-

3.14

$(-

3.02

[-

2.91

 (−

2.86

 negative

2.69

=-

2.69

=-

2.69

$-

2.64

}(-

2.64

Activations Density 0.190%