INDEX

Explanations

possessive pronouns

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

 ModelRenderer

-0.54

rewsbury

-0.51

 teach

-0.51

 Reap

-0.49

Rule

-0.49

 endif

-0.49

 JMenuItem

-0.47

ge

-0.47

cyklopedia

-0.47

iální

-0.47

POSITIVE LOGITS

+#+#

0.68

GeneratedCode

0.61

 EconPapers

0.60

extAlignment

0.57

 craft

0.56

craft

0.55

 HasFactory

0.55

CloseOperation

0.54

 nonatomic

0.53

 Craft

0.52

Activations Density 0.001%