INDEX

Explanations

knowledge

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 knows

-1.15

 understands

-0.74

 automatic

-0.72

knows

-0.64

 wants

-0.63

 knew

-0.60

 hears

-0.60

 знает

-0.59

 automatically

-0.58

 BoxDecoration

-0.57

POSITIVE LOGITS

 Roskov

0.74

Vidite

0.69

neté

0.68

ernalia

0.65

 EconPapers

0.64

adə

0.63

 Harbor

0.62

 Harbour

0.60

acité

0.60

browserify

0.60

Activations Density 0.075%