INDEX

Explanations

width and height

The neuron fires on numeric literals (e.g. dimensions, thresholds, sizes) 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

but

-1.91

 have

-1.66

 muitas

-1.45

—

-1.43

项目的

-1.42

enguaje

-1.38

ate

-1.38

retario

-1.37

就不能

-1.34

POSITIVE LOGITS

婓

1.49

嫫

1.44

 heilige

1.44

ஂ

1.43

瑒

1.42

Soccer

1.41

嫱

1.39

ଃ

1.38

SuspendLayout

1.38

wendi

1.37

Activations Density 0.002%