INDEX

Explanations

possessive 's'

The neuron detects apostrophe tokens—especially the possessive marker “’s.”

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

 both

-1.95

 their

-1.83

 those

-1.60

 such

-1.52

 each

-1.40

 that

-1.36

 every

-1.35

 although

-1.31

may

-1.30

can

-1.29

POSITIVE LOGITS

}{\

1.40



1.36

 fenêtres

1.34

）（

1.33

）、

1.30

':'

1.27



1.27

 weitgehend

1.27

}{$\

1.25

 Sverige

1.24

Activations Density 0.088%