INDEX

Explanations

programming language keywords and syntax

The neuron strongly activates on C/C++ preprocessor macro names (the identifiers introduced by “#define”).

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

and

-1.67

佐々木

-1.48

your

-1.47

 autorytatywna

-1.45

-1.44

-1.43

 Также

-1.38

 जिसमें

-1.36

an

-1.34

れました

-1.32

POSITIVE LOGITS

 like

1.50

Después

1.42

♥

1.41

..\..\

1.38

Más

1.33

说起

1.30

是因为

1.28

</b>

1.25

 verdades

1.25

Según

1.24

Activations Density 0.004%