INDEX

Explanations

The neuron fires on the PHP keyword “function,” i.e. it detects PHP function or method declarations.

function definitions

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

 parfü

-2.20

 parfüm

-1.98

 abiye

-1.83

 sokak

-1.80

 blusa

-1.74

 triko

-1.70

끫

-1.69

 OBITUARY

-1.66

 galeri

-1.66

 komik

-1.64

POSITIVE LOGITS

</h2>

2.88

</em>

2.34

</strong>

2.16

2.08

</b>

1.99

1.97

<em>

1.91

1.89

1.84

Activations Density 0.002%