INDEX

Explanations

copyright, licensed, created

This neuron fires on the leading “*” markers used in code comment blocks (e.g. Javadoc‐style or other starred header comments).

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

 about

-2.03

でした

-1.55

 hará

-1.50

 Bemerkungen

-1.46

 Beiträge

-1.44

經常

-1.43

𖡼

-1.40

 angefangen

-1.38

之間

-1.37

 tessuto

-1.36

POSITIVE LOGITS

1.65

 militaire

1.63

1.52

 accessoire

1.46

金の

1.45

–

1.43

 frambo

1.41

1.39

pü

1.36

 регионе

1.36

Activations Density 0.033%