INDEX

Explanations

codes and web structures

The neuron is strongly activated by HTML markup tags (especially structural and closing tags like `<body>`, `</head>`, `</div>`, `</p>`, etc.).

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

in

-2.25

-1.89

so

-1.86

but

-1.86

-1.85

-1.80

-1.77

-1.72

-1.71

POSITIVE LOGITS

↵

3.48

쐈

1.93

Mvh

1.90

 verschillende

1.89

 trám

1.80

mvh

1.76

dónde

1.67

 предлагает

1.66

 evaluaciones

1.66

冧

1.66

Activations Density 0.062%