INDEX

Explanations

mathematical fractions and derivatives

This neuron is sensitive to LaTeX math‐mode markup—it fires strongly on TeX commands (e.g. \frac, \mathbb, \mathcal, \mathfrak) and other math‐format tokens.

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>

-1.63

了嗎

-1.59

-1.58

xiii

-1.47

 befes

-1.46

 capucha

-1.44

xii

-1.41

–

-1.41

xvii

-1.41

 camiones

-1.40

POSITIVE LOGITS

of

1.88

to

1.86

1.59

 pembunuhan

1.45

ViewChild

1.41

 kecelakaan

1.36

淡淡的

1.34

inazione

1.32

になります

1.31

Activations Density 0.033%