INDEX

Explanations

German locations/architecture/coding

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

 Ruhr

-0.71

 AssemblyTitle

-0.62

 Chriftian

-0.51

انيف

-0.50

IUrlHelper

-0.49

")));

-0.49

 fevere

-0.49

rawDesc

-0.48

 BoxFit

-0.47

 charset

-0.47

POSITIVE LOGITS

bart

0.61

исленность

0.60

po

0.59

scope

0.58

PO

0.57

sch

0.56

bara

0.56

orter

0.55

nloa

0.54

pom

0.54

Activations Density 0.001%