INDEX

Explanations

GNU General Public License

The neuron fires on specialized technical content words found in documentation or code comments (e.g. terms like “cycle,” “triangles,” “modify,” etc.).

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

Ré

-0.82

 hundred

-0.78

 hỗ

-0.77

fe

-0.70

sab

-0.70

Xue

-0.69

RH

-0.68

رياض

-0.68

 Trom

-0.67

দ

-0.66

POSITIVE LOGITS

Gail

0.82

 Unite

0.74

प

0.67

 Laufe

0.66

namespace

0.66

പ

0.66

 Norbert

0.66

ía

0.66

Philippe

0.66

 ドラゴン

0.66

Activations Density 0.467%