INDEX

Explanations

introducing conditions or context

The neuron fires on words that signal beginnings or origins—introductory terms like “Originating,” “began,” “happens,” “It all begins,” etc.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 pero

0.49

and

0.47

 които

0.44

Pero

0.42

 utilisant

0.42

但这

0.41

 nhưng

0.41

 poput

0.41

but

0.41

 bunu

0.41

POSITIVE LOGITS

 przede

0.67

 chiefly

0.55

 především

0.52

 כאשר

0.51

IDField

0.51

 clearest

0.51

当中

0.50

当我

0.49

 sobretudo

0.48

 જ્યારે

0.46

Activations Density 0.149%