Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 2012 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.1111. Min Act: -0.0001

Data Index: 725399 (C4 (Web Text))

Max Activating Token Index: 1015

Text #1

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0922. Min Act: -0.0001

Data Index: 189971 (C4 (Web Text))

Max Activating Token Index: 918

Text #2

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0786. Min Act: -0.0001

Data Index: 753782 (C4 (Web Text))

Max Activating Token Index: 884

Text #3

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0798. Min Act: -0.0001

Data Index: 856587 (C4 (Web Text))

Max Activating Token Index: 850

Text #4

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0779. Min Act: -0.0001

Data Index: 48222 (C4 (Web Text))

Max Activating Token Index: 244

Text #5

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0799. Min Act: -0.0001

Data Index: 926496 (C4 (Web Text))

Max Activating Token Index: 73

Text #6

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0798. Min Act: -0.0001

Data Index: 1591240 (Python Code)

Max Activating Token Index: 519

Text #7

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0762. Min Act: -0.0001

Data Index: 233356 (C4 (Web Text))

Max Activating Token Index: 156

Text #8

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0690. Min Act: -0.0001

Data Index: 944497 (C4 (Web Text))

Max Activating Token Index: 823

Text #9

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0768. Min Act: -0.0001

Data Index: 1408551 (Python Code)

Max Activating Token Index: 403

Text #10

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0738. Min Act: -0.0001

Data Index: 1572536 (Python Code)

Max Activating Token Index: 969

Text #11

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0687. Min Act: -0.0001

Data Index: 352707 (C4 (Web Text))

Max Activating Token Index: 1016

Text #12

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0740. Min Act: -0.0001

Data Index: 772949 (C4 (Web Text))

Max Activating Token Index: 454

Text #13

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0670. Min Act: -0.0001

Data Index: 413447 (C4 (Web Text))

Max Activating Token Index: 1007

Text #14

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0720. Min Act: -0.0001

Data Index: 7081 (C4 (Web Text))

Max Activating Token Index: 429

Text #15

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0737. Min Act: -0.0001

Data Index: 988352 (C4 (Web Text))

Max Activating Token Index: 187

Text #16

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0674. Min Act: -0.0001

Data Index: 1429592 (Python Code)

Max Activating Token Index: 260

Text #17

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0641. Min Act: -0.0001

Data Index: 264005 (C4 (Web Text))

Max Activating Token Index: 491

Text #18

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0650. Min Act: -0.0001

Data Index: 17614 (C4 (Web Text))

Max Activating Token Index: 725

Text #19

Max Range: 0.1111. Min Range: -0.1111

Max Act: 0.0686. Min Act: -0.0001

Data Index: 1444869 (Python Code)

Max Activating Token Index: 415