INDEX

Explanations

source and destination

The neuron detects variable or parameter names that refer to “source”–“target” (and their abbreviations like src, dest) identifiers 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

</strong>

-1.97

</u>

-1.83

</em>

-1.81

…”

-1.80

</h4>

-1.74

</b>

-1.49

</code>

-1.42

):

-1.36

 следующие

-1.35

 …”

-1.34

POSITIVE LOGITS

始めて

1.61

 taşlı

1.55

 tensión

1.52

ྜ

1.51

࿙

1.48

 razu

1.48

 producent

1.47

 okna

1.45

錞

1.45

 daje

1.44

Activations Density 0.039%