INDEX

Explanations

semicolon statements

The neuron is strongly triggered by numeric literal tokens (e.g. integers, floats, “0”, “1”, “2”, etc.) in code.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 automóviles

-1.66

 numerosos

-1.65

 многих

-1.63

 decenas

-1.63

referred

-1.59

 secado

-1.55

číta

-1.53

nearly

-1.53

 geöffnet

-1.52

 acoged

-1.52

POSITIVE LOGITS

 this

1.82

 that

1.78

 their

1.75

 with

1.71

any

1.56

for

1.38

 they

1.37

who

1.33

()

1.30

 idiota

1.28

Activations Density 0.026%