INDEX

Explanations

preorder

This neuron detects mentions of ordering or pre-ordering (e.g., “pre-order,” “order,” “orders”) in product or promotional contexts.

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

 сначала

-1.36

ousel

-1.34

 recordar

-1.33

 zuerst

-1.33

呿

-1.32

 kontaktu

-1.31

 wszystkie

-1.30

 في

-1.30

 vecchio

-1.30

 précédente

-1.29

POSITIVE LOGITS

for

1.88

ции

1.37

the

1.30

eynman

1.26

Это

1.22

 some

1.21

an

1.20

Linker

1.20

 серпня

1.16

Сред

1.16

Activations Density 0.028%