INDEX

Explanations

url followed by parentheses

The main thing this neuron does is spot CSS function calls with numeric or URL arguments (e.g. url(), translateZ(), linear-gradient()).

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

 comprendere

-1.02

 arşivlendi

-0.98

±

-0.87

/.

-0.86

rab

-0.83

 defire

-0.81

urale

-0.81

ـ

-0.81

ー

-0.79

__["

-0.78

POSITIVE LOGITS

2.02

1.98

('

1.60

("

1.59

(\

1.48

($

1.45

(&

1.39

$(\

1.37

}(\

1.32

()

1.31

Activations Density 0.040%