INDEX

Explanations

prediction

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

 يتيمه

-0.75

 MainAxisSize

-0.63

CloseOperation

-0.62

 engraçadas

-0.60

^(@)

-0.59

 déploy

-0.58

 Schmel

-0.57

vician

-0.57

 transfé

-0.57

endaft

-0.56

POSITIVE LOGITS

 pres

0.93

 prop

0.91

 fore

0.88

 port

0.82

prop

0.81

fore

0.76

pre

0.71

 correctly

0.69

port

0.67

 accurately

0.65

Activations Density 0.002%