INDEX

Explanations

technical/instructions

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

Packing

-0.53

nd

-0.53

Accession

-0.50

 tema

-0.50

таки

-0.50

jus

-0.47

 pilot

-0.46

 goda

-0.46

nt

-0.46

やる

-0.45

POSITIVE LOGITS

ThroughAttribute

0.76

ParallelGroup

0.73

EndContext

0.69

 defaultstate

0.67

MLLoader

0.66

OrBuilder

0.65

 Wicidata

0.64

 pinulongan

0.61

 téléphonique

0.60

 مواليد

0.59

Activations Density 0.019%