INDEX

Explanations

class definitions, methods, properties

This neuron detects high-level code structure markers—e.g. interpreter directives (shebang), import statements, class or function definitions, and decorators.

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

<bos>

-2.34

 épreuve

-1.44

を

-1.42

 économie

-1.41

だが

-1.39

 activité

-1.37

などを

-1.34

UCIÓN

-1.33

 when

-1.28

 identité

-1.28

POSITIVE LOGITS

 leſs

2.09

えっ

2.06

ほら

1.99

つづく

1.97

What

1.89

あぁ

1.77

さぁ

1.75

ようです

1.72

ことも

1.71

感じで

1.65

Activations Density 52.043%