INDEX

Explanations

everyday places and activities

The neuron fires most strongly on words naming leisure trips or everyday outing destinations (e.g. “vacation,” “visit,” “grandma,” “grocery store,” “hair salon,” “date”).

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

츨

-1.55

蝋

-1.53

Тел

-1.48

Адрес

-1.45

܊

-1.44

颤抖

-1.41

 tulum

-1.40

くなり

-1.34

 ZEALAND

-1.34

 Главная

-1.32

POSITIVE LOGITS

or

2.48

if

2.13

we

1.76

 tenemos

1.59

 your

1.48

one

1.48

 tienen

1.47

 например

1.44

 there

1.44

 ricord

1.40

Activations Density 0.139%