INDEX

Explanations

ard

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ARD

-0.50

modelBuilder

-0.50

detectChanges

-0.50

ReferenceEquals

-0.49

addGap

-0.48

CODES

-0.48

 Duke

-0.47

 respect

-0.46

POSITIVE LOGITS

 виправивши

0.75

 مشين

0.69

LookAnd

0.66

vician

0.63

GraphicsUnit

0.59

canestro

0.58

 Paglinawan

0.56

fvar

0.56

NPC

0.54

osoba

0.54

Activations Density 0.013%