INDEX

Explanations

the phrase "used to" when describing how something was used in a technical context.

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

Efq

-0.75

 myſelf

-0.75

 Monfieur

-0.74

 Shakspeare

-0.72

 ProtoMessage

-0.71

lichter

-0.69

*)&

-0.68

 Jefus

-0.68

ThroughAttribute

-0.68

 whoſe

-0.67

POSITIVE LOGITS

Vorlage

0.56

endpush

0.46

льше

0.45

ształ

0.44

ready

0.42

 ready

0.42

 inmedi

0.41

 mismas

0.40

setcounter

0.40

 aprove

0.39

Activations Density 0.056%