INDEX

Explanations

setting window titles or properties

The neuron strongly activates on common English function words (articles, auxiliaries, prepositions, conjunctions) that structure ordinary sentences.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

protos

0.54

Apk

0.51

Gren

0.50

グリ

0.50

 חו

0.50

Ping

0.48

ارد

0.47

Gr

0.47

 限定

0.47

文本

0.46

POSITIVE LOGITS

statusbar

0.88

statusBar

0.82

StatusBar

0.76

 actionBar

0.74

 statusBar

0.73

 navbar

0.73

NavigationBar

0.73

 StatusBar

0.71

 setContentView

0.68

ActionBar

0.68

Activations Density 0.010%