INDEX

Explanations

questions, weight, facilitated, religious

The neuron is triggered by TeX environment openings, i.e. the “\begin{…}” sequence.

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

ַּ

-1.80

 overall

-1.77

 tiež

-1.73

-1.69

 موبایل

-1.67

 también

-1.64

-1.59

ֶּ

-1.58

 that

-1.55

and

-1.54

POSITIVE LOGITS

ׇ

2.14

 naturen

1.73

 chrétien

1.70

 évêque

1.65

animity

1.60

 चुनें

1.60

 Apakah

1.57

 Siapa

1.54

getSystemService

1.54

 religieux

1.54

Activations Density 0.018%