INDEX

Explanations

class and object definitions

The neuron specifically detects occurrences of class-declaration keywords (e.g. “class,” “case class,” “data class,” “enum class”) in code.

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

🕗

-1.69

when

-1.62

-1.60

 terrorismo

-1.59

 humeur

-1.57

 fantasma

-1.56

 parlamento

-1.55

 tortura

-1.55

 luminoso

-1.52

 gigantes

-1.48

POSITIVE LOGITS

1.75

niosek

1.71

to

1.59

ukraine

1.58

 implementación

1.57

᯾

1.55

 treinamento

1.51

 aims

1.50

 imó

1.49

蚡

1.48

Activations Density 0.008%