INDEX

Explanations

the word "modify" and related words

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DockStyle

-0.81

reportError

-0.80

Kjelder

-0.76

InputBorder

-0.76

setViewportView

-0.75

 betweenstory

-0.75

 ProtoMessage

-0.73

 imageNamed

-0.72

RenderAtEndOf

-0.71

kmäler

-0.70

POSITIVE LOGITS

 modification

2.66

 modified

2.50

 modify

2.48

 modifications

2.45

 modifying

2.45

 Modification

2.39

 modifies

2.31

 Modify

2.27

 Modifications

2.25

 Modif

2.23

Activations Density 0.380%