INDEX

Explanations

code attribute access

The neuron detects programming identifiers used as module or library names when they appear as the left‐hand part of a dotted reference (e.g. “db.”, “sys.”, “os.path”, “xsl:”, “re.”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

or

-2.16

if

-1.89

may

-1.88

 that

-1.83

any

-1.80

its

-1.80

 those

-1.69

can

-1.64

 when

-1.53

 before

-1.52

POSITIVE LOGITS

xvii

1.98

 récemment

1.66

≃

1.66

xiii

1.66

1.65

ciutto

1.63

 geprüft

1.61

 összefoglaló

1.55

 μας

1.54

hermoso

1.52

Activations Density 0.018%