INDEX

Explanations

menu items and navigation

The neuron activates most strongly on occurrences of the term “menu” (and related UI‐menu tokens like “Menus,” “menuItem,” etc.) in code or configuration contexts.

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

\,

-0.83

 Lewandowski

-0.81

 ومع

-0.77

eah

-0.76

首都

-0.75

KeyDown

-0.74

 landscapes

-0.74

岬

-0.73

 обратить

-0.72

 profesjonal

-0.72

POSITIVE LOGITS

⣶

0.84

⢤

0.78

ɵɵelementStart

0.78

てて

0.77

íteni

0.75

 ২০

0.74

 vroe

0.72

ട്ട

0.71

 برگ

0.71

oben

0.71

Activations Density 0.021%