INDEX

Explanations

code structure delimiters

The neuron fires specifically on single‐quote characters (apostrophes) in the text.

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

—

-1.52

we

-1.52

Your

-1.49

 time

-1.48

 unserer

-1.46

 χρόν

-1.46

-1.45

اما

-1.37

 connection

-1.34

-1.31

POSITIVE LOGITS

are

1.86

 Infór

1.67

鹛

1.63

 vacía

1.55



1.52

 However

1.51

Of

1.50

偸

1.45

лій

1.41

icester

1.40

Activations Density 0.035%